Professional Documents
Culture Documents
Jean Keeling, Anthony Busuttil Paediatric Forensic Medicine and Pathology
Jean Keeling, Anthony Busuttil Paediatric Forensic Medicine and Pathology
MEDICINE AND
PATHOLOGY
: dited by
Anthony Busuttil
Em eritus Regius Professor of Forensic Medicine, University of Edinburgh; and Medical Director,
Fo rensic Medical Services, NHS Lothian, Edinburgh, UK
Jean W Keeling
_::ormerly Consultant Paediatric Pathologist, Royal Hospital for Sick Children,
::diilburgh, UK
i, ARNOLD
HODDER
?ART OF HACHETTE LIVRE UK
I CONTENTS
I
Contributors xiii
Preface xiv
Acknowledgements xv
ist of abbreviations used xvi
Helen Hammond
Introduction 1
Alerting signs 2
Types of abuse 3
Significant harm 4
Legislation 7
Formulating an opinion 21
References 22
Jacqueline YQ Mok
Introduction 24
Forensic evidence 27
Consistent vocabulalY 29
Perianal findings 36
SummalY 43
References 44
vi I Contents
Maeve McPhillips
Radiological investigations
48
Skeletal injuries
51
Head injury 60
Visceral injuries 68
Soft-tissue injury 69
Differential diagnosis 69
Conclusion 73
References 73
Angela Thomas
Introduction
76
Primary haemostasis 78
Secondary haemostasis
79
Laboratory tests
81
Coagulation defects 96
The neonate
97
Conclusion 101
References 102
5 Biochemical investigations on post-mortem specimens 106
Denis R Benjamin
Introduction 106
General evaluation 107
Hypoxia 109
Inflammation
109
Anaphylaxis 109
Infection
110
Endocrine disorders
112
References 120
Harry Willshaw
Introduction
125
Scope of ocular and adnexal injury 125
,
' .-' -----. .
Contents I vii
References 134
Bruising 140
References 143
Jean W Keeling
Introduction 145
Photography 147
References 164
viii I Contents
Infections 177
References 178
Jean W Keeling
Introduction 180
Definitions 180
References 195
Jean W Keeling
Introduction 198
Epidemiology 201
References 219
Dick Variend
Introduction 225
' -
.
------ ....
.
Contents I ix
Diabetes mellitus
243
Genetic metabolic disorders
244
Other bacterial infections
247
Deaths related to obstetric events and premature birth
248
Miscellaneous causes of sudden natural death
248
Sudden unexplained death in older children
249
Sudden natural death in the early neonatal period
249
Sudden death associated with 'intermediate' pathology
249
References
250
Strangulation 331
Prevention 333
References 334
Anthony Busuttil
Overview of paediatric trauma 336
Falls 339
Prevention 341
Hypersensitivity 342
References 342
John Pearn
Introduction 345
References 359
Jem Berry
Introduction 362
Accidents 372
Suicide 373
,~------~-- .
,
.
Contents I xi
References 377
References 394
Introduction 395
References 414
David Whittaker
Introduction 420
References 432
G Howard Moody
Introduction 435
Comparison 435
xii I Contents
References 444
Anthony Busuttil
Introduction 447
Opinions 450
Conclusion 454
References 455
Index 485
~ - - - - - - - - - - - - - - - - -- -~
.
-------
.
I CONTRIBUTORS I
Cook Children's Medica l Cen ter Royal Hospital for Sick Children
The possibility that a child may have been injured, abused, In this book, some of the top ics covered here are rele
neglected or otherwise ill treated rightly raises the indigna vant specifically to maltreatment in ea rly life, beginning
tion and anxiety of the caring professions and involves law with the examination of an infant or child for whom ab use
enforcement agencies. However, in the interests ofjustice a nd is suspected, incorporating the family environment and set
fairness, a person accused of such injury or negl ect is entitled against criteria for normal deve lopment. The difficult prob
to appropriate legal representatio n at any hearing and is lem of suspected sexual abuse of children is considered
deemed to be innocent until proven guilty. Quite frequently, separately. The extensive clinical experience of the authors
dubiety and uncertainty linger about whether or not, in any of the opening chapters is readily apparent, highlighting,
specific situation, observations made, clinically or pathologi as they do, the pitfalls of incomplete investigation and iIl
cally, can be interpreted solely as a manifestation of inflicted considered interpretation. The ap propri ate level of investi
injury or neglect, or whether there is a possibility that the gation of specific findings, interpretation of investigations
observed findings could have come about in other, non and consideration of differential diagnoses are addressed
criminal, circumstances. These matters necessitate advice in chapters contributed by a paediatric radiologist, a haema
from those with expelience and expertise in this field. tolo gist and a clinical pathologist, respectively. Those areas
The aim of this book is to furnish an authoritative, com requiring specialist clinical expertise and experience - the
prehensive tex t to assist practitioners of medicine and the eyes, mouth a nd central nervous system - are considered
law dealing with such cases in the approp ri ate interpreta by specialists in those fields with ex tensive paedi atric
tion of these matters and to enable clinical and pathological experience. The examination of the scene of death or injury
findings to be presented in an unbiased and dispassio nate is discussed as a backg round to post-mortem examination
manner so that the co urts are able to better evaluate the of the very yo ung. The interpretation of cerebral pathology
specialist evidence put before them. in the newborn, the investigation of sudden or suspicious
The investigation and interpretation of findin gs of perinatal death and sudden death in both infa nts and older
alleged ill treatment of infants and children requires a children are addressed by experienced practitioners. Sepa
multidisciplinary app roach , centred on the child, his or her rate consideration is given to sudden or suspicious deaths
well-being in both the short term and longer term, as well that occur in hospital.
as that of any siblings within the same environment. All In subsequent chapters, more general areas of forensic
of the avai lable information about any in cident must be pathology, including asphyxia and thermal injury, drown
care full y collected, collated and evalu ated. Laboratol), ing, injury to road users and oth er accidents are add ressed
data, both clinical and forensic, the results of radiological from a paediatric viewpoint. A similar approach is evident
investigatio ns and information from the examinat.ion of in the chapters covering toxicological investigation, DNA
the scene where any incident took place sho uld be carefully profiling and dental identification.
sought and evaluated against the clinical findings. A team The book concludes with consideration of the role of th e
approach is essential, with close collaboration of famil y expert witness in criminal judicial cases and the provision
physicians, paediatricians involved in both community of reports in the civil medicolegal context.
and hospital practice, the clinical fo rensic medical examiner Although the majority of contributors to this te xt are
and specialist pa tho logists, together with police and social UK based , the subject matter is presented, as far as possible,
welfare serv ices. No incident should be looked at in isola wit hout national or geographic bias, so that the contents
tion but rather in the context of the child's development have in te rnation al releva nce.
and interaction with his or her family, environment and
peer group. The survivors of inflicted injul)' or neglect in Anthony Busuttil
childhood must be carefully followed up, protected and Jean W Keelin g
their family unit supported. January 2008
. .
I ACKNOWLEDGEMENTS I
We would like to thank our contributing authors for their and expertise in the preparation of the illustrations for her
hard work and for their patience and ready responses in the chapters. We would like to thank the many staff at Hodder
li g ht of requests for updates and answers to specific Arnold with whom we have been involved fo r their expert
quelies. Colleagues in Edinburgh and elsewhere in the UK ise and encouragement ; in particular, Philip Shaw and our
have read the Editors' contribu tions and made useful sug Project Editor Amy Mulick for their most helpful sugges
ges tions. JWK thanks Dr Roger Malcomson for his assistance tions in the late stages of manuscript completion.
I LIST OF ABBREVIATIONS USED I
~ ~----------
.
.
.
List of abbreviations used I xvii
SEXUAL ABUSE
Jacqueline YQ Mok
INTRODUCTION
The widespread occurrence of child sexual abuse has been
known for many years, although the exact prevalence is diffi
Sexual abuse has been defined as the involvement of depend cult to define. Prevalence studies have been canied out by
ent, developmentally immature children and adolescents in interviewing adults about childhood events. In a summary of
sexual activity that they do not fully comprehend and to 19 studies conducted in the USA or Canada between J 980 and
which they are unable to give informed consent or that vio 1994, Finkelhor4 found that the rate of sexual abuse reported
late the social taboos of family roles. l In this situation, there by women varied from 2 per cent to 62 per cent, with an
is an imbalance of power between abuser and abused, and an approximate prevalence of 20 per cent. The ra tes reported by
element of control of the child by a trusted adult. Sexual men varied from 3 per cent to J 6 per cent, with a reasonable
abuse ranges from violent assault to gentle seduction. The estimate of approximately 10 per cent. However, studies based
activities include all forms of digital-genital , oral-genital, on interviews of adults provide limited information owing to
genital-genital contact between abuser and child, as well as differences in study design, the response rate, methods of data
non-contact abuse such as exhibitionism and use of the child collection, definition of sexual abuse, definition of a child, and
in the production of pornographic material. the accuracy of recall of events that might be traumatizing.
A diagnosis of sexual abuse has both civil and criminal Data from studies of reported incidences of abuse pro
implications. The medical profession 's early involvement vide information about the number of children recognized
with child sexual abuse was limited to psychiatrists, who annually, usually at individual centres. It does appear
were interested in the behavioural manifestations following that increasing numbers of children are referred because of
sexual abuse. The subject received scant paediatric recogni suspicions of sexual abuse, owing to either a t ru e increase
tion until publications by Jaffe 2 and Kempe3 changed the in the occurrence or better recognition of the problem.
willingness and ability of paediatricians to recognize and In the USA it is estimated that between 120 and 150 per
deal with the problem. Paediatricians must have a high index 10000 children have been subjected to sex ual abuse.
of suspicion and be willing to consider sexual abuse as a dif More recently, Jones et a1 5 identified a significant reduction
ferential diagnosis when a child presents with behavioural in the incidence of substan tiated child sexual abuse in the
problems or somatic symptoms that suggest that sexual USA, and a population-based study in Australia has also
abuse might have occurred. All physician s should act accord provided evidence of a decline in the underlying rate of
ing to their local child protection procedures, with the welfare child sexual abuse.6 Such findings may indicate the effect
and protection of the child as paramount considerations. iveness of personal safety progra mmes for young children.
An earlier paper from one city in the UK 7 described 51 and case-dependent skills for any paediatrician or forensic
children in 1985 who presented after they disclosed abuse, medical examiner who undertakes such examinations have
a number that rose to 79 in the subsequent year. Thirty been defined by the Royal College of Paediatrics and Child
eight per cent of the children were aged less than 5 years of Health and the Faculty of Forensic and Legal Medicine
age at diagnosis and the mean age was 8 years. Of the (October 2007).12 Core skills include: the ability to commu
alleged abusers, 60 per cent were related to the child, and nicate with children and their carers about sensitive issues;
one-half of these were natural fathers. Because the vast an understanding of the child's developmental, social and
majority of children are abused by someone they know and emotional needs; a knowledge of consent and confidential
trust, force and restraint are not commonly used when ity issues as they relate to children; an understanding of
abusers engage children in sexually inappropriate activities. the range of normal genital and anal anatomy for the age
There is usually little intent to harm the child, and bribes or and gender of the child to be examined; an ability to docu
threats are often used to prevent the child from reporting ment and interpret the clinical findings; competence in the
the abuse. Children are ideal victims for sexual exploitation production of a report; a willingness to communicate and
and abuse, as they are vulnerable and see adults as trusted co-operate with other agencies; and the aptitude to present
and powerful. The 'child sexual abuse accommodation syn the evidence and be cross-examined in civil and criminal
drome's describes the stages of traumatic sexualization, proceedings.
feelings of betrayal, powerlessness and entrapment that
lead to the child's accommodation of the abuse. As a result,
very few children disclose abuse immediately following the History
event, making the retrospective interpretation of healed
trauma a great diagnostic challenge. As with all medical consultations, the starting point is in
the history, taking care to avoid asking leading questions.
Investigative interviews are usually conducted by desig
The Medical Evaluation nated agencies (police and social services) to avoid repeti
tive questioning of the child. This should not preclude
Child sexual abuse should rarely, if ever, be diagnosed on physicians asking relevant questions that are essential to
the basis of physical signs alone. A clear statement from the medical examination. In eliciting the history, the clin
the child is the single most important factor towards mak ician should approach the child in the same manner as he or
ing a diagnosis. The medical evaluation of children sus she would in any other paediatric condition. Details should
pected to have been sexually abused should be part of a be obtained about the child's birth and development, past
multidisciplinary process that spans the investigative, medical history, and family and social background. The
diagnostic and therapeutic needs of the child and family. systematic enquiry should cover symptoms and signs relat
The medical examination should be a comprehensive ing to the genitourinary and gastrointestinal systems.
health assessment that should aim to: Specific questions should be directed to the presence of
• establish any need for immediate treatment; pain, itch, rash, discharge or inflammation 'down below',
• provide background information that mayor may not episodes of bleeding (on pants or on toilet paper), and fre
support the diagnosis; quency and consistency of bowel movement. When appro
• provide information or evidence to sustain criminal priate, girls should be asked the menstrual history, type of
proceedings and/or care plans; sanitary protection used, sexual history and previous
• plan or coordinate ongoing care; gynaecological examinations. The child's terminology for
• reassure the child and family. the various body parts should be documented, and any
statement made by the child recorded verbatim. 13
A minimum of physical examinations should be con The importance of the psychological aspects of sexual
ducted as is necessary, and if several medical specialists are abuse, both in the short and long term, has been high
to be involved it is desirable that they should examine the lighted. 14 ,15 Emotional difficulties include anxiety, sadness,
child together. Joint examinations can be performed by a anger, behaviour problems, school refusal, sleeplessness,
paediatrician and a forensic medical examiner (police sur withdrawal and sexualized behaviour. Somatic complaints
geon) to encompass, in a single examination, the child's that have been reported include: eating disorders, abdom
medical needs with the legal requirements for evidence. 9 inal pain and headaches, as well as loss of bladder and bowel
Although a carefully structured approach to the examin control. The medical evaluation should include an enquiry
ation is required for legal purposes, the medical assessment into the child's emotional status and general well-being, as
should be compassionate and thorough, resulting in a thera the carers may not associate behavioural difficulties with a
peutic experience for the child and family.lO.11 The examin past history of abuse. The carers may even be unaware that
ers must be both familiar and comfortable with normal sexual abuse has occurred. Child sexual abuse is a diagno
childhood behaviour and development, genital and anal sis that, like other paediatric diagnoses, requires a consider
anatomy, as well as physical findings of abuse. Core skills ation of the history, physical examination and supportive
26 I Investigation of suspected sexual abuse
laboratory tests when appropriate. 16 Although the medical Sometimes the hymen can be difficult to visualize
history qualifies as 'hearsay evidence', many states in the because of sticky or redundant folds. If labial traction does
USA permit an exception to the hearsay rule, for medical not cause the hymen to open, douching with warm water
histories obtained by physici ans,u will result in 'floating of the hymen', Further definition of
the anatomy can be obtained with the use of a cotton bud,
applied behind the hymen to 'tease out' the folds, This
Technique of the Examination technique should not be used in prepubertal children, in
whom the hymen is extremely sensitive, unless preceded
Th e exa mination should be carried out in the presence of a by the gentle application of a local anaesthetic gel. In ado
trusted adult, usually the child's mother. The whole child lescent girls, the hymenal edge can also be examined using
should be examined, and this includes measuremen t of the Foley cath eter stretch technique,18 A 14-gauge Foley
hei ght and weight, assessment of the general appearance, catheter is inserted through the hymenal orifice into the
developmental milestones, demeanour and behaviour. As vaginal vault, and infl ated with 40 mL of air. Gentle pulling
part of the physical examination, general signs associ ated toward the hymen results in stretching and displaying of
with trauma should be sought, such as distribution a nd hymenal tissue over the surface of the balloon,19,20 The
pattern of bruises, grip marks, ' love bites', teeth marks and toluidine blue dye test was deve loped by Lauber and
scratches, as well as injuries within the mouth. In most situ Souma 21 to help detect lacerations in adult rape victims.
ations, the disclosure involves past abuse and th e exam Toluidine blue is a nuclear stain that will bind to nuclei in
in at ion can be planned to suit the child and family. Delay the deeper dermis when exposed by lacerations in the skin.
should be minimized in the following circumstances, when: In a study of girls "vho alleged sexual abuse, the applica
• the a buse has occurred within the previous 72 hours; tion of a I per cent aqueous solution of toluidine blue dye
• there is a history of acute trauma; was found to increase the detection rate of posterior
• there is a possibility of pregnancy resulting from the fourchette lacerations from 4 per cent to 28 per cent in
abuse, so that post-coital contraception can be adolescents, and from 16.5 per cent to 33 per cent in the
prescribed. paediatric population .22
The prone knee-chest position is recommended if there
Considerable reassurance will be required and to avoid is difficulty in visualizing the posterior hymenal rim. With
further distress no force or restraint should be used, no mat t he child resting on her knees and elbows, the buttocks are
ter how well intentioned. An explanation ofwhat the exam parted upwards and outwards, using the examiner's palms.
ination entails should be given, with clear emphasis on 'a This technique usually exposes the vestibule, perihymenal
health check' and 'taking a look'. Wi th calm reassurance and structures and hymen. Any suspicious findin gs in the pos
an unhurried approach, most children can be examined terior hymenal rim can be clarified , as the effect of gravity
without the use of any sedation. The recommended position causes the posterior rim to stretch out.
for the female genital examination is the supine 'frog The perianal area is traditionally examined in the left lat
legged' position, with the hips flex ed and abducted, the eral position. With use of the colposcope, some examiners
soles of the feet touching. Very young children can be prefer to continue with the examination in the supine pos
examined on their mothers' laps. Sometimes the adult can ition, by asking the child to 'curl up into a little ball'. This is
assist by sitting astride the examining couch, cradling the achieved by flexion of the child's hips and knees against the
child as she leans back against the adult's body. abdomen. The buttocks are gently separated using the palms
In the supine 'frog-legged' position , the external geni of both hands, and the perianal area inspected for signs of
talia should be inspected for signs of injury. Gentle lateral abuse. The external anal sphincter usually relaxes during
parting of the labia majora with the examiner's middle and this procedure, making it easier to visualize fissures in the
index fingers (labial separation) allows visualization of the anal margin. Sometimes the ana l canal opens to reveal a
posterior fourchette, vestibule, perihymen al regions and clear view of the rectum. A digi ta l examination of the anus
urethra; the hymen may be visible at this stage. In order to is rarely necessary and is unhelpful in assessing anal tone.
visualize th e ma rgins, configuration and size of the There is also no evidence to suggest that a digital examin
hymenal openin g, labial traction is applied . This involves ation provides useful clinical information. If there are suspi
grasping the posterior ends of the labia majora between cions of injuries or pathology higher up the anal canal or
thumb and index finger and pulling gently up an d out rectum, referral should be made to a paedi atlic surgeon for
wards. The vaginal walls may be demonstrated by this pro an examination under anaesthesia.
cedure; however, digital examination of the vagina is Most units have a dedicated suite of rooms where chil
rarely indicated in the prepubertal child. Any sign observed dren can be examined in a child- fr iendly environment. The
should be described and documented, using the clock face minimum requirement is an area that is integrated into a
to denote the location of the finding; in the anatomical children's outpatient department to allow access to labora
position, 12 o'clock is anterior while 6 o'clock denotes the tories a nd other investigative facilities but which offers
posterior position . some privacy. Other important considera tio ns include the
specific needs of adolescents, the gender of the examiner the prospective study by Muram and Elias,25 130 prepubertal
and the availability of follow-up for sexual health coun girls (mean age 5.5 years) were evaluated bo th by an
selling and therapeutic support. unaided examination and by colposcopy. Overall, 92 girls
were found to have abnormal findings, the majority of
which (96 per cent) were observed during the unaided
Examination of Boys examination. Of the four patients in whom findings were
detected initially by colposcopy, these were observed dur
The same sensitive and age-appropriate techniques should ing the repeat unaided examination. In only one patient
be used when examining boys. As with girls, any evidence of were the findings observed only by colposcopic examin
trauma such as bruises, swellings, scratches and bites should ation alone. The authors concluded therefore that una id ed
be sought for and documented . The genital examination examination by an experienced cl ini cian is adequate for
should focus on the groin, penis, urethra, scrotum and testes the evaluation of most victims of sexual abuse.
while the ch ild is supine. The presence of discharge or warts The most obvious advantage of the colposcope is the
should also be noted. "'lhen examining the pelianal region, integral photographic facility. Documentation of all visible
various positions can be used, depending on the age of the findings in abuse is increasingly expected as the standard
boy. The young child may be examined in the supine of good practice. Images produced by the colposcope can
knee-chest position, immediately after examination of the be converted into slides or photographs using a single-lens
penis and scrotum. Older boys may be more comfortable in reflex camera or a Polaroid camera. Modern video technol
the left lateral position, with either the right knee or both ogy allows the image to be recorded onto a videotape,
knees pulled up to the chest. The anal examination usually compact disc (CD) or digital versatile disc (DVD) for imme
involves inspection of the soft tissues only. An assessment of diate viewing, and has the advantage over still photog
the anal to ne can be made, with the buttocks parted. Find raphy as it shows the dynamic variability of anogenital
ings should be documented either by line drawings or photo anatomy as the examination proceeds.
graphs. The colposcope can be used as a source of light and The colposcope is now standard and acceptab le equip
magnification, as well as photographic documentation. ment in the examination of sexua l abuse. With photo doc
umentation, the most obvious benefit to the child is that
there is no need for repeated examinations. Many examin
THE COLPOSCOPE IN THE MEDICAL ers do not position their eyes in the binocular lenses of the
EXAMINATION colposcope; rather they view the image produced on the
monitor at a more comf0l1able position and distance from
The medical examination of children reporting sexual the ch ild . The child is also able to observe the examination
abuse evolved from a search for ways to improve lighting on the monitor, and this helps to achi eve a sense of control
and visualization of the young child's genitalia. Methods and participation in the examination. The instrument is
for recording observations were also important, to assist in usually in trod uced to the child to gain cooperation, and
the interpretation of findings. In the early 1980s when many enjoy playing with it prior to the examination. The
physicians began to examine the genitalia of children, use of high technology is both accepted and expected in
instruments such as a hand-held magnifying glass and an the modern health care system, and is preferable to
auriscope were used to enhance visualization. In 1925, the attempting to peer between a ch ild's legs wit h an auriscope.
colposcope was invented by Hinselman , to assist in exam The ability to capture images, either as slides, photographs,
ination of the female cervix. Teixeira 23 used the colposcope videotapes, CDs or DVDs enhances undergraduate and
to examine victims of alleged sexua l assault. Five years postgraduate physician training. 26 - 3o
later, Woodling and Heger24 promoted the use of the colpo
scope in the assessment of chi ldren who alleged sexual
ab use. The colposcope is a no n- invasive magnifying
instrument with a built-in light source, which allows mag FORENSIC EVIDENCE
nification of the external genitalia from 2 to 25 times. It
contains a binocular system of lenses of varying strengths, Verba l consent is usually adequate for both the examin
co upl ed to an integral light source. The instrument can ation and photo-documentation. This should be obtained
either be mounted on the wall, the examination couch, or from the person with parental responsibility and from the
on a caster system to enable mobility. Light intensity can child of sufficient maturity to understand the nature and
be varied to cope with changes in magnification, and most consequences of the examination. If the child is the subject
instruments incorporate a green filter to improve the visu of legal proceedings, the consent of the court is required.
alization of abnormal vasculature or scars. The person obtainin g consent must record the procedure in
Tei xei ra reported that an additional 11.8 per cent of sus the case notes. Written consent for each component of the
pected cases were corroborated than would have been examination can be documented on a standard form and
when conventional examination techniques were used. In on appropriate forms.
28 I Investigation of suspected sexual abuse
Correct procedures must be followed for the collection and evidence for prosecution in child abuse cases. Tradi
of forensic sampJes and evidential material, recognizing tional systems for photographing abused children using a
the principle of the unbroken chain of evidence. This legal single-lens reflex camera still provides the highest quality
concept requires that the origin and history of any exhibit image. The use of Hi-8 and Super VHS video cameras offers
to be presented in a court of law must be clearly demon significant improvements in image quality and resolution
strated to have followed an unbroken chain from its source, over the traditional videotapes. More recent developments in
through its examination and to the court. A note of the computer imaging technology have resulted in greater ease
persons handling the sample, time, date and place where in transforming analogue images to a digital format for
the sample was obtained, along with the places and condi computer storage and conversion into slides or prints.
tions of storage must be documented. In an acute assault [mages can also be converted into a computer file, which
(within 72 hours), the clothing worn during the assault assures preservation without degradation of the initial
should be coI1ected individually in paper bags and handed image. A video capture card, a scanner and/or a digital cam
to the investigating police officer. The skin should be era, allow a computerized database to be developed. With
inspected for a ny stains, and swabs taken to examine for appropriate software products, images can be transmitted
saliva, lubricant or semen. If judged appropriate, a sample over the Internet, enabling a second opinion to be sought
of blood should be taken for DNA analysis. Stains on cloth and thereby brings telemedicine into remote areas. J 5,J6
ing are best preserved by drying and storing at cool room Photographs, video, CD and DVD recordings must be
temperature. Blood or semen collected on swabs is best properly verified and relevant, they must bear the patient
preserved by frozen storage. The presence of lubricant or identifier, date and time of recording, and must be signed
spermatozoa can be sought from swabs in and around the by both medical and forensic examiners. Images are neces
mouth (taking care to swab in the labial-gingival sulci) and sary to explain or illustrate adequately the complexity of
from saliva. Young children will tolerate external and the injuries, and are therefore relevant for courtroom pro
internal anal a nd vaginal swabs, if the a ppropriate size of duction. However, the production of sensitive images and
swabs is used, moistened with water. Spermatozoa can be photographs of children's genitalia in the courtroom is not
detected for up to 14 hours in saliva J I 3 days in anal sam common practice and should be discouraged. An alterna
ples J 2 and 6 days in vaginal samples. JJ The time limits for tive line diagram can always be used to illustrate the
detection of seminal fluid are 3 hours in the anus and injuries. All recorded material should be available to any
J 2- J 8 hours in the vagina. Improved DNA diagnostic tech medical expert instructed by defence solicitors.
niques allow detection of useful evidence from unwashed
or partially washed clothing, bedding and other items used
during the assault. General guidelines for the collection of
forensic evidence in cases of acute sexual assault a re not SKILLS AND EXPERIENCE REQUIRED
well suited for prepubertal children. In a review of 273
child victims of sexual assault aged less than 10 years, the Although medical findings are not necessary for the legal
majority (64 per cent) of forensic evidence was found on conclusion of sexual abuse, health-care professionals who
linen or clothing. Over 90 per cent of children with positive examine children for signs of sexual abuse are often asked to
forensic evidence were seen within 24 hours of the assault. render an opinion as to whether their examination was nor
After 24 hours, all evidence, with the exception of one mal, non-specific, suggestive or indicative of abuse. Despite
pubic hair on a child, was recovered from clothing or this important role, the literature contains little information
linens. The authors suggested that any collection of foren regarding examiner competence in assessing children's geni
sic evidence from a prepubertal child 's body may not be talia. Undergraduate and postgraduate training on the subject
necessary more than 2 days after the assault. J4 In cases of of child sexual abuse is woefully inadequate. The medical
chronic abuse, when the last episode of contact was more investigation of children in whom abuse and neglect is
than a week previously, collection of evidence consists of a suspected has become a complex and technical specialty.
careful histolY, thorough examination and screening for A new group of physicians has emerged, who are specialists
sexually transmitted infections (STls) . in the wide ramifications of child abuse and neglect. A
Although a highJy trained and experienced examiner is sound knowledge is expected in general, behavioural and
not likely to miss any abnormalities during an unaided developmental paediatrics, as well as in . gynaecology and
examin ation, the colposcope allows a review of photographs infectious diseases. The child abuse specialist also needs to be
and video recordings when findings thought to be suspicious familiar with forensic examinations, civil and criminal laws,
during the examination might be interpreted on review as child advocacy and public policy, and must be comfortable
normal or non-specific. Permanent documentation of visible when testifying in court. For these reasons, structured train
findings protects the examining practitioner by providing ing programmes must be developed to improve the quality of
objective evidence, thus only the interpretation of the find care delivered to abuse children and their families.37
ings can be challenged. Still photographs, videotapes or CDs Brayden et al J8 evaluated the interobserver reliability of
of the injuries often provide key information for assessment clinicians rating colposcopic photographs, and examined
correlates of reliable interpretations. Seventy-one physici ans studies and a consistent application of established terminol
and two nurse practitioners were asked to assess blindly six ogy that can prevent the misinterpretation of non-specific
photographs of prepubertal female genitalia, five of which or congenital findings as post-traumatic changes.
were taken from children who had given histories of sexual
abuse. Experts in the field of child sexual abuse assessment
made significantly more accurate assessments than paedia
NORMAL FEMALE GENITAL ANATOMY
tricians, paediatric and family practice residents, and trainee
physicians. In another study, Adams and Wells 39 sought to
Figure 2.1 is a diagrammatic representation of the prepu
determine how well medical examiners agreed on the
bertal female genitalia, with the parts labelled using recom
significance of certain anogenital findings in children, by
mended terminology. The most consistent landmark is the
showing colposcopic photographs of 16 patients to 170 med
clitoris, which is usually prominent in young girls because
ical examiners who were blinded to the history of each
of the lack of subcutaneous fat in the surrounding tissues.
patient. The agreement between the participants and the
The vestibule is the area enclosed within the labia minora,
experts on the abnormal cases (mean 81 per cent) was signif
and includes the urethral opening which can be pinpoint or
icantly higher than on the normal cases (mean 71 per cent,
patulous. Bilateral lines (Hart's lines) drawn medially from
p = < 0.001). There was also higher agreement on genital
the posterior ends of the labia minora converge at the pos
findings (78 per cent) than on anal findings (63 per cent,
terior fourchette. The introitus refers to the hymen and
p = 0.000). In the participants, higher experience level and
hymenal opening or orifice. Sometimes the vaginal walls
use of a colposcope were associated with higher overall
can be seen through the hymenal opening. The area
agreement with the experts (74 per cent versus 44 per cent,
between the posterior fourchette and inferior edge of the
p = < 0.0001). The history was also found to influence
hymen is referred to as the fossa navicularis. Between the
physicians' interpretation of girls' genital findings. Paradise
posterior fourchette and the anus, deep within subcuta
et al 40 conducted a questionnaire survey of 1387 randomly
neous tissues, is the perineal body.
selected fellows of the American Academy of Pediatrics and
In order to identify the signs of sexual abuse, the clinician
all 802 members of child abuse professional groups, using
must have a sound knowledge of normal anogenital anatomy
seven simulated cases, in 6 of which the histories were
and its variants. Contemporary textbooks provide little
changed at a second mailing 4 months later. The proportion
insight into the characteristics of hymenal anatomy or the
of changed or reversed opinions varied from 0 per cent to 5.6
changes of the hymen in response to oestrogen levels during
per cent amongst experienced examiners; 1.6-19.8 per cent
the childhood years. The appearances vary considerably
amongst moderately experienced examiners and 3.6-27.2
according to the age of the child. In the newborn, the effects
per cent where the examiners were inexperienced. The like
lihood of an interpretation being changed was influenced by
a diagnostic expectation (change in history) and ambiguity
of the photographs. Labium-
majorum
(}L-1--\;\---\---'.-- Ure th ra I
must be taken to avoid terms that can be misinterpreted and opening
--tt
Fossa ----''<---'..--~-
genitalia, as well as interpretation of anogenital findings, so navicularis ...... - _ _~,-'--_ Posterior
P"~,~:
that those who examine children for alleged sex ual abuse
can understand each other's descriptions. 41 - 43 Practice
guidelines are now available and should be widely dissemi "",,h""
nated. 44 - 47 In an excellent review of hymenal morphology
and non-specific findings in girls selected for non-abuse,
Heger et al 48 called for a thorough understanding of normal Figure 2.1 Prepubertal female genitalia.
of maternal oestrogens result in promin ence of the labia and fine submucosal blood vessels that extend onto the edge of the
clitoris, a redundant and fleshy hymen and copious vaginal hymen . The hymen itself is thin and almost translucent. With
secretions. In early infancy, the labia majora gra dually fold the onset of early puberty, there is a reduction in the superfi
across the introitus to form a protective pad. As the oestrogen cial vascular prominence in the vestibule, fossa navicularis
effects wear off, the labia majora appear flattened and the and hymen. As puberty progresses, the labia majora and
labia minora are seen as thin folds. The inner margins of the minora become pigmented. Secretions are abundant in the
labia minora may be set deeply and easily mistaken for vestibule, which becomes darker and more textured in its
the hymenal orifice. appearance. Subcutaneous deposits of adipose tissue may be
Although there are several cross-sectional studies on seen. The hymen becomes fleshy and redundant, making
genital appea rances throughout childhood, a methodo detailed examination of the rim difficult.
logical flaw of most of these studies is the inclusion of girls Labial adhesions, either partial or extensive [Fig. 2.2),
who may have been sexually abused, as the authors have have been reported in 5-39 per cent of girls studied. 52 ,53
relied on parental history alone to exclude abuse. These The presence of labial adhesions does not imply that sexual
cross-sectional studies have included girls over multipl e abuse has occurred, as they could result from poor hygiene,
age groups, making it difficult to eva luate findings by age. inflammation or recurrent infection, especially in a young
Table 2.1 summarizes some studies that have attempted child still in nappies.47
to define normal genital anatomy in non-abused girls. Increased friability (Fig. 2.3) is defined as a smal l dehis
Although cross-sectional studies cannot assess the forma cence of the tissues of the posterior fourchette, with or with
tion of new features such as bumps or notches, they have out bleeding. This is usually seen towards the end of the
been useful in clarifying the range of normal findings, as examination and has been reported in 4.7 per cent of girls
well as helped to define hymenal configuration with age. examined in the cohort reported by McCann. 52 In a minolity
Longitudinal studies have been useful in documenting of girls abnormalities that had previously been documented
anatomical changes over time. Reports of longitudinal in sexually abused girls were noted. These included
studies have mainly come from one centre in the USA 49,50 increased vascularity, mid-line avascular areas, as well as
and are summarized in Table 2.2. urethral dilatation (Fig. 2.4). Although it is likely that sexu
ally abused girls may have been inadvertently included in
these cohorts, these studies highlight the dangers of over
Labial, Posterior Fourchette and interpreting minor anatom ic findings that may lie within the
Vestibular Findings range of normal variation. 54 .55 In a case-control study of
anatomic changes resulting from sexual abuse, no difference
Yordan and Yordan 51 have documented changes in the labia was observed in the percentage of abused versus non-abused
minora with age and onset of puberty [see Table 2.2). In the children with regard to labial adhesions, increased vascular
absence of oestrogens, the labia minora appear small, thin and ity, linea vestibularis, friability, a perineal depression, or a
without pigment. The epidermis of the vestibule is thin and hymenal bump, tag, longitudinal intravaginal ridge, external
smooth. The fossa navicularis is characterized by a network of ridge, band or a superficial notch. Furthermore, the mean
Berenson 1993 49 62 Birth and 1 year Decrease in hymenal tissue seen in 33 out of 57
Hymenal notches seen for the first time in nine girls;
resolved in five
Hyme nal tags seen for first time in four girls; resolved in two
Posterior hymenal notches not observed
Berenson 1995 50 134 <2 months and about Change from annular/fimbriated to crescentic hymen in
3 years; 42 also examined 87 out of 134
at about 1 year External hymenal ridges resolved by 3 years
Tags decreased by 1 year; resolved by 3 years
Longitudinal intravaginal ridges and periurethral bands more
easily seen at 3 years
Mean horizontal and vertical transverse hymenal diameters
increa sed with age
Posterior hymenal rim did not va ry with age; > 2 mm
Racial differences in hymenal anatomy
32 I Investigation of suspected sexual abuse
Hymenal Configuration
Figure 2.6 Crescentic hymen in a 6-year-old girl. Note the Figure 2,7 Bump or mound of tissue at 6 o'clock on a crescentic
absence of hymenal tissue between 11 and 1 o'clock and the hymen.
translucent smooth-edged posterior rim.
Figure 2.9 Symmetrical perihymenal bands in a 4-year-old girl. Figure 2.10 Transection of the hymen in the 5 o'clock position.
The semilunar shaped spaces created by the bands could be The hymen also shows ea rly oestrogen effects, appearing
mistaken for the hymenal opening. thickened and redundant.
the width of the hymenal membrane, or deep, when it is healed laceration, rupture and gap. It is widely agreed that
more than one-half of the width. Notches or clefts are seen a deep/complete cleft or transection located between the
commonly at the 12 o'clock position of annular or redun 4 o'clock and 8 o 'clock positions in the hy men is consistent
dant hymens. They should not be reco rded as pathological with abuse (Fig. 2.10), as these have never been reported in
findings in the fimbriated hy men (because of its fringed non-abused children .47 ,48,56
nature) or between 11 a nd 1 o'clock in a crescentic hymen ,
whe re the re is a natural absence of hymenal tissue. 56 The
sign ificance of a hymenal notch or cleft depends on the Gaping/Narrowing/Attenuation of the Hymen
location and ex tent of the defect. Anterior and latera l
notches, between the 9 and 3 o'clock positions, have been The hymenal orifice has been described as gaping w hen,
observed in 35 per cent of newborn girls 57 are usually with the child in the supine frog-legged position, a view is
caused by asymmetrical insertions of a crescentic hymen, obtained into the vagina, with thigh abduction and without
and should be considered normal. Anterior notches tend to the exam iner manipulating the labia majora (Hobbs, Wynne
widen to form crescentic hymens, whereas some notches and Thomas, 1995).61 This subjective term is best avoided. A
disappear as hymena l tissue becomes less redundant. 49 narrow ing of hymenal tissue is referred to as attenuation of
Although the observation of a latera l notch in the first 3 the hymen. The narrowed rim is usually found at the poster
years of life has been documented as a new finding, none ior edge and should be confirmed with the child in the
of the children followed up longitudinally developed a new knee-chest position. Although usually associated wit h
posterior notch. 50 Care must be taken to distinguish a true chronic abuse, this term is comparative and implies that the
concavity from an appa rent hollo wing in the hymenal rim examiner has prior knowledge of the hy menal anatomy.
caused by an adjacent mound or tag. Until recently it was Unless the child has been examined previously it is not pos
thought that notches in the posterior portion of the hymen sible to determine w hether the hy menal rim is narrowed.
were rarely observed in non-abused children. 53 ,5 7 With a Heger et al 48 defined a narrow hymenal membrane as the
more consistent application of terminology, two groups of w id th of th e membra ne viewed in the coronal plane (i.e.
examiners have established that superficial notches /partial from the edge of the hymen to the muscular portion of the
clefts can be found in the posterior hymen in 5-18 per cent vaginal introitus) of less th an 2 mm . A 'thickened edge' was
of girls who had been screened for non-abuse, and have a term used to describe the relative amount of tissue
proposed that this is a non-specific finding. 48 ,56 A recent between the internal and external surfaces of the hymenal
study 59 suggested that some partial tears of the hymen heal membrane. Using these terms, the hymenal rim was found
to leave shallow notches. Figure 2.3 demonstrates a notch to be thickened in 45.5 per cent; irregular in 51.7 per cent
in the 5 o'clock posi tio n of the hymen, with friability of the and narrowed in 22.4 per cent of 147 prepubertal girls
posterior fourchette. screened for non-abuse. Over three-quarters (78 .7 per cent)
A hymenal transection describes a complete deficiency of girls with a narrowed hym enal rim were found to be
in hymenal tissue that ex tends to the junction between the overweight (> 75th centile for weight).
hymen and vestibule and, when seen in any location, is The posterior hymenal width cannot be measured accu
associated with trauma. 56 ,60 Other terms used to describe rately, especially when the rim is narrow. However, a com
this a ngular concavity include complete notch/cleft, tear, plete absence of the posterior hymen (confirmed in the
Hymenal Measurements
~
Figure 2.14 Same ch ild as in Fig. 2.13, examined under general
Figure 2.11 Labial separation shows a flesh y annular hymen in anaesthesia, using the same technique. Note the dilated
a 5-year-old girl. appearance of th e hymen .
F'
The skin of the anal verge and epithelium of the anal canal
exhibit regular folds when the anus is closed. Midline anterior
or posterior smooth areas in the anal verge are common find
ings, where the defe ct is lined by unbroken skin and repre
sents a gap in the fibres of the superficial part of the external
anal sphincter. The significance of lateral defects in the anal
verge is unclear. The perianal skin may look red and moist,
due to poor hygiene, irritation, inflammation, infections or
infestations. Caution must be exercised when interpreting
perianal hyperpigmentation in non-white children. The
plexus of veins underlying the peri anal skin can be seen to
distend if examination is prolonged, producing a purple-blue
discolouration around the anus (venous poolin g). This is a
normal finding, caused when buttock traction and increased
intra-abdominal pressure occludes venous drainage (Fi g.
2. 15). Haemorrhoids in children are extremely rare.
Intermittent cont ractions of the external a nal sphincter Figure 2.16 Gross 'reflex anal dilatation', allowing a clear view
can sometimes be seen (twitching or winking) , a sign which of the anal canal. The child was examined in the supine
has been observed in normal children. This must be differen knee-chest position.
tiated from reflex anal dilatation, when the anus opens
widely after a brief period of buttock separation. Reflex a nal
dilatation is an incompletely understood phenomenon that
must be interpreted with caution a nd in contex t. It is best
demonstrated by the buttock separation test in the left lateral
position, when the buttocks are parted gently and observa
tion takes place for 30 seconds. A positive result is obtained
when the ex ternal anal sphincter contracts briefly, follow ed
by dramatic rel axation of both the external and intern al
sphincters to reveal a cylindrical opening into the rectum
(Figs 2.16-2.18). Although first described as indicative of
anal abuse,57 the significance of this sign rema ins controver
si al. Dilatation of up to 10 mm has been reported in up to
49 per cent of 267 normal children. 58 Other authors have
reported a prevalence of the sign in 4-14 per cent of non
abused children. 59 ,7o Reflex anal dilatation is observed more
freq uently when the exam ination is prolonged, in the
knee-chest position, and in children who have a history of
constipation. 7I ,72 Chronic straining at stool may lead to Figure 2.17 Same child as in Fig. 2.16. Appearances following
weakness of the anal sphincter, and a 'visibly relaxed ' anus anal dilatation, showing perianal erythema and loss of anal folds.
McCann etal1989 68 161 girls, 106 boys 2 months-11 yea rs Erythema 41 0/0
Hyperpigmentation 30%
Venous engorgement after 2 minutes of examination 52%
Midline wedge-shaped smooth areas 26%
Anal skin tags/folds 11%
Anal dilatation 49%
Berenson et al 1993 74 89 girls Mean 10.7 months Smooth areas 26%
Hyper/hypopigmentation 10%
Erythema 7%
Skin tags 3%
Venous pooling 1%
-
- - -- =-
--------~~
-- ~
38 I Investigation of suspected sexual abuse
been recognized as non-specific findings, present also in children, and two children had slight anal dilatation. Va ri
control subjects who have been carefully selected for non ous degrees of laceration were noted in all the children. By
abuse. It is also important to remember that some forms of the eighth day, acute signs of trauma were present in only
child sexual abuse will leave no physical signs. Other fac one child. Second-degree lacerations healed over 5 weeks,
tors that influence physical findings include the position of leaving na rrow bands of scar tissue; however, deeper lacer
the child in relation to the abuser, the size of the object ations took 12-14 months to hea l. Even deep injuries faded
inserted relative to the size of the orifice, the degree of into surrounding tissues with time, making detection of
force used, the use of lubricant, the frequency and duration periana l injury increasingly difficult. 77
of abuse, as well as the period of time that has elapsed More recently, the healing patterns of anogenital injuries
between the last episode of abuse and the examination. in 94 prepubertal children were described in a longitudinal
Erythema and superficial abrasions may disappear within study over 10 years. 59 The authors confirmed that anogenital
2 days, and bruises fade after a few days. A single episode trauma healed quickly, often without residua. Of the 171 acute
of penetrative anal abuse may result in perianal svvelling irUuries see n, diagnostic changes remained in only 25 (14.6
and a gaping anus, signs that will resolve within 1 week. per cent). All trauma to the labia majora or minora healed
There is now good documentation of norm a l variants in without residua. Injuries to the posterior fourchette and fossa
anoge nital anatomy, but literature on the healing process navicularis (24 out of 47) healed with non-specific residua
of the traumatized anogenital region is scant. In general, (labial fusion, vascular changes and scarring) . Hymena l tran
hea ling is rapid and occurs through rege neration and sections did not heal unless repaired surgically, whereas 5 out
repa ir. Wound healing by regeneration is complete within of 37 partial tears of the hymen healed with residual sha llow
48-72 hours, and normal appearances are restored by 6 notches. Only J out of 31 anal injuries healed with anatomic
weeks. Deeper injuries heal by repair, a process involving changes (tag, scarring and hypeJ1ligmentation).
granulation and scarring that takes approximately 2
months. Because of the frequent delay between the alleged
abuse, disclosure and medical examination, there is a need FEMALE GENITAL FINDINGS IN
to understand the heal ing process, how an injury might SEXUAL ABUSE
affect the genital anatomy of the prepubertal girl and
whether signs of trauma persist after the onset of pubeliy. Over the past decade a great deal of information has been
Finkel 75 reported on the findings of seven children who gathered by medical exam iners who see children who have
experienced acute genital and anal trauma, and were fol been sexually abused, and a wide range of findings have
lowed up until their injuries were healed . All of the chil been reported. It is well documented tha t ab normalities of
dren had acute signs of injury documented after the the genital tract associated with abuse may be minor and
incident, most of which were superficial lacerations. In non-specific, occurrin g a lso as a result of inflammation,
only one child, who had a deep lacerated penetrating infection or poor hygiene. Although Emans et a l58 found
wound, was there any sca r formation that distorted the that a number of genital findings were significantly more
"hymenal and perihymenal tissues. Obvious acute superfi common in sexually ab used children than in asy mptomatic
cial trauma healed without residua, by a process of regen girls, many of the findings were also present in girls who
eration. Healing was rapid, and by the second examination were examined bec ause of symptoms relating to vaginitis,
(3-13 days later) t here was little apparent sign of injury. vulvitis, bleeding or dysuria. Debate continues regarding
McCann et al76 followed the resolu tion of acute genital the frequency of ab normal findings in sexually abused
injuries in three children who had been sexually assaulted. children because of difficulties in reaching an agreement
Initial signs of injury included erythema, oedema, submu on what constitutes an abnormality and the lack of a true
cosal haemorrhages, abrasions, lacerations and hy menal go ld standard diagnostic test for sexual abuse. There may
transections. Acute tissue changes resolved quickly, with ery be no physical findings in more than one-half of the chil
thema and oedema disappearing by 11 days, whereas submu dren examined for suspected sexual abuse. The absence of
cosal haemorrhages resolved within 27 days. The shaJ1l physical signs neither confirms nor negates a diagnosis of
jagged edges of the damaged hymen gradually smoothed out abuse. An earlier study by Muram 78 on 31 girls whose per
as healing took place, and there was relatively little scarring petrators confessed abuse reported that the medical exam
in the hym en or posterior fourchette. With the onset of ination failed to detect any abnormalitY in 29 per cent of
puberty, the hymenal changes in one gi rl were obscured by the girls. In a case review of 236 children whose peJ1letra
hypertrophy of the hymen. This highlights the importance of tor had received conviction for sexual abuse, Adams et al 79
using techniques to enable a thorough examination of the found that the genital examination was normal in 28 per
fimbriated, redundant hymen seen in pubertal girls. cent, non-sp ecific in 49 per cent, suspicious in 9 per cent
The natural history of perianal injuries has also been and abnormal in 14 per cent of cases. Abnormal anal find
studied in four children. Acute signs seen included ery ings were found in only 1 per cent of patients. Using dis
thema and perianal oedema, which caused the folds to criminating ana lYSis, the two factors that significantly
thicken or flatten . Venous congestion was present in three correlated with the presence of abnormal genital findin gs
in girls were (1) the time since the last incident and (2) a 1997 publication has been replaced by an evidence-based
history of bleeding being reported at the time of ab use. review of physica l signs of child sex ual ab use.47
Among the clinical findings that require diagnostic These signs are currently subject to further scrutiny in
interpretations are concave va ri ation s of the hymenal rim. the light of more recent studies that suggest that the geni
These have been referred to as clefts, notches, tears, fis tal examination of the abused child rarely differs from that
sures, transections, healed laceratio ns or hymenal attenua of the non-abused child. Using a case-control study
tions. The cha racteristics and clinical correlates of concave design, 192 prepubertal children who alleged penetratio n
hymena l variations were described in a large cohort of were examined using a colposcope, and the photographic
13 83 prepubertal and adolescent girls who alleged sexual records we re compa red with 200 control subj ects who were
abuse. Concavities in the hymen were found in 174 matched for age and ethnicity, and who were found not to
patients (12 .6 per cent), of whi ch 3 5 per cent were anterior. have been abused after careful screenin g. Abnormal find
Of these, two -thirds were smooth and curved, wh ereas one ings (hymen al transection, perforation or deep notch) were
third were angular and /or irregular. Posterior or lateral seen in only 2.5 per cent of subjects or four children, aLi of
concavi ties were found in another 57 per cent of the girls, whom gave a history of penile or digita l penetration. How
of whom one-third were curved and smooth, and two ever, only 17 ab used children were examined within 1
thirds were angular and/or irregu lar. Multiple hymenal week of the abuse.56 A prospective study of 2384 children
concavities were found in 13 girls. Each hymenal concav referred to a tertiary centre from 1985 to 1990 for possible
ity was categorized by location, contour (curved or angu sexual abuse found that 96.3 per cent of all children had a
lar) and smoothn ess (smooth or irregular). The concavities normal medical exa min ation. Even with a history of vaginal
were fUlther sorted into one of five categories, reflecting or anal penetration, the rate of abnormal medical signs was
the presence or absence of the clinical impression of only 5.5 per cent, seen in 61 out of 444 (6 per cent) girls and
anogenital tra uma: benign, benign with reservations, 2 out of 251 (1 per cent) boys. Most children in this study
uncertain, suggestive of trauma , and trauma. Poste ri or were examined within 1 week of abuse. s1
and/or lateral location of the concavity, angular contour In a study of female sexua l assault victims , 94 per cent
and rim irregu larity of the concavities we re found to be of 200 women were found to have trauma at one or more
significantly associated with trauma .60 Abnormal genital locations - posterior fourchette, labia minora, hymen and
findings are more likely to be documen ted when the girls fo ssa navicularis. Injuries were found to vary by site, with
report bleeding at the time of the assault, or when the tears appearing most often on the posterior fourc hette and
examina tion occurred within 72 hours of the last episod e fossa navicularis, abrasions on the labia, and ecchymosis
of abuse. so The vulvovaginal signs of ab use have been on the hymen. s2 However, the mean age of the cohort was
summarized by a working party of the Royal College of 24 years (range 11-85 years) and the majority were exam
Physicians of London,12 with emphasis on the type of ined within 24 hours of the assa ult, factors that a re atyp
abuse and t he timing of the examination (Table 2.4). The ical of children who disclose sexual abuse. Eighty-four
(88 per cent) women returned for follow-up examinations
within an average of 25 days (range 4- 50 days). All
injuries had resolved , with no evidence of t rauma in 71 (87
Table 2.4 Vulvovaginal signs of abuse' per cent) women. Tears to the hymen and fossa navicularis
did not reunite, and no scarring was observed at follow-up.
Non-specific signs when seen within 72 hours of abuse Kellogg et al s3 conduc ted a retrospective case review of
Erythema 36 pregnant ado lescent gi rls (age range 12.3- 17 .8 yea rs)
Bruising who presented for sexual abuse exa minations. Despite
Superficial laceration/a brasion definitive evidence of sexua l contact (pregnancy), evidence
Oedema of penetrative trauma (clefts in the posterior, one-half of
Signs supportive of abuse which extended to the base of the hymen) was found in
Notch/cleft in posterio r hymena l rim, which may be associated only 2 of the 36 girls, whereas four had suggestive findings
with scarring (deep notches in the poste rior half that did not extend to
Scar in posterior fourchette the base of the hymen, or visible scars) .
Hym enal opening> 15 mm due to attenuat ion of hymen
Signs diagnostic of penetrating injury SIGNS OF ANAL ABUSE
Fresh laceration of hymen
Old tear in hymen that may have healed with scarring and Non-specific signs that may be seen in the acute stages
interruption of hymenal rim include erythema, bruising and oedema, which vary in
Attenuation of hymen/disappearance of hymenal rim, usually exte nt depending on the degree of friction and force
posteriorly involved. Bruising may result in a haematoma formation at
'Adapted from ref. 46. the anal verge, caused by sheari ng of the subcutaneous
40 I Investigation of suspected sexual abuse
venous plexus. Fissures or tears may be single or multiple, Table 2. 5 Perianal signs af abuse'
seen as superficial or deep clefts in the perianal skin or
Non-spe cific acute changes
mucosa, which, in the acute situation, are painful. The fis
Erythema
sure is usually triangular or fan-shaped, with its apex point
Oedema
ing towards the anal canal and the lower edges pulled apart
Fi ssures
by the fibres of the external sphincter. There may be bl eed
Venou s co ngestion
ing or anal spasm associated with the fissure, or on its own.
Bru isi ng
Deep fissures are likely to heal with scarring and there may
be an associated skin tag caused by skin overgrowth during Signs supportive of abuse
healing. Opinions differ as to the significance of anal fissures Anal laxity without other explanation
in relation to abuse, and the presence of a fissure gives no Reproducib le reflex anal dilatation> 15 mm
information on its aetiology. Fissures resulting from consti Chronic changes, i.e., thickening of anal skin verge, increased
pation cannot be distinguished from those caused by anal elasticity and reduced anal sphincter tone
abuse. They have been reported in normal cohorts of infants Bite marks
and children,68,74 and also in children with constipation .lI ,72
Signs diagnostic of blunt force penetrating trauma
Chronic trauma may lead to thickening of the anal canal
Fresh lace ration
folds. There may be squamous metaplasia of the columnar
Healed sca r extending beyond anal margin on to perianal sk in in
epithelium. The perianal tissues look and feel calloused,
the absence of a reasonable alternative explanation
and hypertrophy of the external sp hincter results in a
prominent ring of tissue surrounding the anal orifice. This -Ada pted from ref. 46.
~
Figure 2.21 A 10-year-old girl with a 4-year history of 'vaginal
bleeding'. An oedematous and erythematous 'polyp' is seen
Figure 2.20 Lichen sclerosis. Cutaneous and submucosal anterior to the hymenal opening. The appearances suggest a
haemorrhages seen in the posterior fourchette and vestibule. The prolapse of the urethra.
patient was a 6-year-old girl who presented with pain and
bleeding 'down below:
sexual play between children, and by parents engaged in
custody disputes. The normal, age-appropriate sexual
accidental injury, when the labium minomm is the most exploitative behaviour must be differentiated from disturbed
frequent structure involved. Deeper injuries tend to be behaviour arising from child abuse.96.97
anterior or lateral to the hymen. In a multicentre study of
56 prepubertal girls who sustained unintentional perineal
injuries, the hymen was involved in only one child who fell SCREENING FOR SEXUALLY TRANSMITIED
in a park with her legs abducted. A pinpoint abraded area INFECTIONS
was found on the hymen at the 3 o'clock position. 9o
Congenital abnormalities such as haemangiomas and Debate continues as to whether the screening for STls
failure of fusion of the median raphe raise concerns of should be routine in the evaluation of sexual abuse. The
abuse when they first present. The raphe is a flesh-coloured. presence of a sexually acquired organism can indicate prior
slightly raised, linear structure in the perineum, and when sexual contact in a child, and when accompanied by other
failure of fusion occurs the opposing edges may split and indicators of sexual abuse supports the diagnosis. Non-sex
appear like a traumatic lesion, with bleeding. Midline avas ual transmission of STls is rarely an issue in adults,
cular areas in the posterior part of the vestibule or posterior but when the same diseases are found in children, there is a
fourchette may be seen as a white line (linea vestibularis) or tendency to attribute them to an asexual mode of transmis
a white spot (partial linea vestibularis), and is easily mis sion. 98 The immature anogenital tract is more vulnerable to
taken for scar tissue. They have been observed in 25 per infection, especially if there is a breach of the mucosal lin
cent of newborn girls,91 and can vary in size as well as con ing following traumatic abuse. The low numbers of children
figuration in the first year of life. 92 In contrast with scar tis reported to have acquired STIs from sexual abuse may rep
sue the linea vestibularis is generally in the midline, without resent the lack of systematic screening, or the non-recogni
accompanying disturbance in vascularization. Normal vari tion by clinicians that children can be infected through
ants such as bumps and notches between 3 and 9 o'clock on sexual abuse. However, other routes of transmission that
the hymen can be mistaken as residua of sexual abuse. must be considered, although difficult to exclude are:
Urethral prolapse is a circular eversion of the mucosa that • perinatal acquisition from an infected mother who may
usually occurs at the urethral meatus without accompanying be asymptomatic (Chlamydia trachoma tis, Neisseria
symptoms. Some bleeding may occur following straining, gonorrhoeae, Trichomonas vagina/is, herpes simplex
and assumed by carers to have come from the vagina. 9 ] Care vims, human papillomavirus [HPV], human
ful examination reveals an oedematous area anterior to the immunodeficiency virus [HNJ);
hymen, through which the urethral opening may be identi • non-sexual adult to child contact (HPV, herpes simplex
fied (Fig. 2.21). Crohn's disease can involve any part of the vims);
alimentary system, and one presentation consists of extra • auto-inoculation (HPV, herpes simplex virus).
intestinal manifestations. There may be perianal and vulval
ulceration with oedema, suggesting trauma. 94 ,9 5 Clinicians The risk of a child or young person acquiring an STI
may also have to sort through histories given by adults depends on the prevalence of STis within the local population:
who have misinterpreted nomlal childhood masturbation or maternal STI during pregnancy that might lead to vertical
transmission; the type of sexual contact during abuse ; injuries adult population. Results from studies of children with
to the genital tract; the sexual maturity of the victim; and anogenital warts suggest that the majority of children do
whether a condom was used during abuse. Transmission from not acquire these sexually. Vertical transmission is an
mother to child is welJ documented for most STIs and the important source of infection in young children, and
organism can lie dormant for up to 2 years (and possibly household members of children with anogenital warts
3 years). In dealing with a child who is less than 3 years old, it should be routinely screened for the presence of non-geni
is therefore important to establish if the parents are infected tal as well as anogenital warts. 105.106 At present, there is no
and to screen them if possible. Sexual abuse should be sus reliable method to determine the mode of acquisition in a
pected when an infection is diagnosed after infancy and child with HPV infection.
before sexual activity occurs in the older child. Under these Children who disclose sexual abuse should be screened
circumstances, gonorrhoea and syphilis would be diagnostic for STIs under the following circumstances:
of sexual abuse, whereas infection with Chlamydia, Tri • presence of symptoms and signs at the time of assault
chomonas, herpes simplex virus and HPV would be suggestive or subsequently (bleeding, discha rge, dysuria);
of sexual abuse. 12,44,99,100 • suspicious anogenital findings;
A prospective study of 1538 children who were examined • history of genital-genital or genital-anal contact or
for possible sexual abuse found the overall prevalence of penetration;
STIs to be 6 per cent. The diagnosis of a STI was highly cor • alleged abuser known to have, or at high risk of
related with a history of sex ual contact (alleged by 49 per acquiring, STI;
cent of the children) and the presence of a discharge. l01 The • STI prevalent in the community;
authors had attempted to exclude perinatal acquisition as a • request from child or young person.
source of infection by selecting verbal children who could
understand questions regarding sexual contact. Although The forensic science laboratory will not examine samples
the prevalence of STIs was only 3.7 per cent in a cohort of obtained for screening of STls. For each organism, sampling
159 girls who had been sexuaJly abused, Robinson et al 102 techniques and transport media should be individually tai
also found a significant association of sexually transmitted lored and must meet the criteria for maintaining the chain of
organisms with the presence of a vaginal discharge. The evidence. Swabs must be taken by a doctor with appropriate
authors recommended that the presence of vaginal discharge training and experience. lt is therefore important to establish
in sexuaJly abused girls was an indication for STI screening. good liaison with the diagnostic laboratories, as well as
The use of an algorithm to assess the risk of gonococcal physicians in genitourinary medicine. Fine, wire-mounted
and chlamydial infections was found to reduce the cost and urethral swabs can be passed through the hymen, even in
trauma of unnecessary sampling. Major factors in the risk young children, to enable cultures for STls. 107 A vaginal wash
assessment were genital-genital or genital-anal contact or procedure, using stelile saline introduced through tubing
penetration, suspicious anogenital findings, genital dis from a butterfly needle, has also been found to be acceptable
charge and suspicion of an STI in the perpetrator, although as a diagnostic test for STIs in prepubertal girls.!08 Testing for
vaginitis and referral for suspected abuse in a sibling were infection with HlV involves a blood test for HlV antibody,
designated minor factors. Restricting testing to children obtained at the time of the initial examination and at follow
with at least one major or two minor factors in their study up. Newer, more sophisticated tests for HfV-RNA or proviral
population of 3040, 45 per cent would have been tested and DNA need only be done in a young child to exclude vertical
all known cases of gonorrhoea or chlamydia infection transmission. 109 Serological testing for infection with Tre
would have been identified. Testing would have been ponema pallidum, hepatitis B or hepatitis C virus should be
avoided in 51 per cent of girls and 72 per cent of boys. 103 considered on an individual basis. The child or young person
More recently, the Association of Genitourinary Medicine may require active and passive immunization against hepati
and the Medical Society for the Study of Venereal Disease in tis B infection as well as antiretroviral therapy. Expert opin
the UK produced guidelines for the management of sus ion should be sought from an infectious diseases specialist.
pected STls in children and young people. The recommen In general, there have been relatively few studies where
dations included the need for sta Ff worki ng in genitourinary children with a particular STI have been evaluated for the
medicine to be: alert to the possibility of child abuse and possibility of child sexual abuse. This has resulted in a lim
neglect ; aware of local guidelines; and trained in child pro ited evidence base to determine whether a particular STI is
tection procedures and protocols. Screening should be con a marker for sexual abuse 47
sidered in all young people who may have been sexually
abused or who have been found to have an STI. 104
Human papillomavirus infection is one of the most INTERPRETATION OF CLINICAL AND
common STIs, and is estimated to affect 10 per cent of the LABORATORY FINDINGS
adult population. Subclinical infection is common, and the
increasing incidence of condyloma in children probably Many changes have occurred in the medica l evaluation of
reflects the increased prevalence of HPV disease in the children suspected of having been sexually abused, and
how physical findings are interpreted. A comprehensive list predictive of, conviction. The child's history is the single
of the clinical and laboratory findings seen in abused and most important factor in the accurate diagnosis of most
non-abused children was first drawn up by Adams et al. 43 cases of sexual abuse. III ,11 2 In a review of child sexual abuse
Sometimes known as the Adams Classification System, it criminal court cases, those involving the youngest victims
has been developed over the years and the latest revision 45 were found to have a significantly lower conviction rate.
is the result of examining published data and consensus Successful prosecution, particularly in the youngest chil
amongst 18 experienced physicians. The classification sys dren, depended on the quality of the verbal evidence and the
tem can be summarized as: effectiveness of the child's testimony. II ) The literature
• findings documented in newborns, or commonly seen reports a marked variability of medical findings in children
in non-abused children: examined for sexual abuse and the diagnosis of sexual abuse
- normal variants; cannot be made on the basis of physical findings alone. It
- findings commonly caused by other medical has been suggested that a model could be developed for the
conditions. prediction of anatomic findings, based on an inventory of
• indeterminate findings (owing to insufficient or variables that would include the age of the child, type of
conflicting data): contact, whether penetration occurred, and a history of pain
- physical examination fmdings, for example deep or b1eeding.1l4
notches or clefts in the hymen; smooth, Expert medical testimony may help to interpret the
uninterrupted rim of hymen of < 1 mm wide; presence or absence of physical signs, but the overall effect
immediate dilatation of the anus> 2 cm; on the legal outcome is unknown. Increased communica
- Lesions with indeterminate specificity for sexual tion between prosecutors and paediatricians may improve
transmission: for example, anogenital condyloma the outcome, especially if the paediatrician is told whether
accuminata, anogenital herpes. she is a witness to fact, or an expert witness. The distinc
• fi ndings diagnostic of trauma and/or sexual contact: tion is important to allow adequate preparation and to
- acute trauma to external genital/anal tissues, for avoid an injustice to the child. A medical expelt should be
example fresh laceration or extensive bruising; able to demonstrate relevant training or experience in child
- residual (healing) injuries, for example perianal abuse cases that are similar to ones in which he or she has
scar; been called on to provide expert testimony. Irresponsible
- injuries indicative of blunt force penetrating medical testimony must be avoided; this includes use of
trauma: for example acute laceration, extensive unique theories of causation, unusual interpretations of
bruising, healed hymenal transection; medical findings, alleging non-existent physical signs,
- presence of infection that confirms sexual
deliberate omission of pertinent facts or knowledge and
contact, for exam pIe gonorrhoea;
misquoting of medical literature H5
- diagnostic of sexual contact: for example, Medical , social and legal professionals have relied too
pregnancy or sperm identified in specimens taken heavily on the medical examination in diagnosing sexual
directly from the child's body. abuse. Normal findings are consistent with abuse, and the
examiner must document all signs, positive or negative,
Until a better evidence base is available, this system whenever any child is examined. There is now a consensus
provides a useful tool that both assists physicians in inter on terminology and interpretation of findings in child sex
preting clinical findings and helps to achieve some consist ual abuse, 12,44,45,47,48,56 but these guidelines are only helpful
ency in terminology. if examiners are meticulous in documenting their findings.
There needs to be an agreement on the minimum standards
for the training of medical professionals who perform sex
SUMMARY ual abuse examinations. 9 ,37,116,117 Medical care for children
and young people who disclose abuse has evolved into a
Whenever abuse of a child is suspected the clinician must complex and technical specialty. Expert medical testimony
refer to local multi-agency child protection procedures for can be crucial to the legal outcome of a case of alleged sex
appropriate action. Siblings and close friends of the victim ual abuse. The evaluation of child sexual abuse is not rou
must also be interviewed and examined if appropriate, as it tinely taught in undergraduate or postgraduate training
is likely that they may also have been exposed to the programmes. Specialized training programmes must be
alleged perpetrator. 110 There may be an association between developed to ensure better research as well as the dissemin
different types of abuse, and following a diagnosis of any ation of knowledge and expertise.37, 11 8 In September 2006,
form of abuse a medical assessment for sexual abuse the American Academy of Pediatrics recognized child abuse
should also be considered. as a new pediatric subspecialty. It is hoped that the accred
In the investigation of sexual abuse of children, clinicians itation of trained specialists will benefit child victims of
should remember that the medical examination is only a part sexual abuse, families, health-care professionals and those
of the jigsaw. Physical evidence is neither essential for, nor in social services and law enforcement.
44 I Investigation of s us pected sexual abu s e
REFERENCES 24 Woo dling BA, Heger A. The use of the colpos cope in the
diag nos is o f sex ua l a bu se in th e pedi a tri c age group. Child
Sch echter IVID, Roberge L. Sexual ex ploitatio n. In Hel fe r RE, A buse Et Neglect 1986 ; 10: J11 - 14.
Kempe CH (ed s) Child A buse and Neglect: Th e Family and th e 25 Mura m D, Eli as S. Ch ild sex ual abus e: geni ta l tra ct findin gs in
Community. Ca mbridge, MA: America n Academy o f prep ubertal girls . n. Comp arison of colp oscopi c and unaided
Pedia tri cs, 1978. examinati ons. Am J Obste tr Gy J1 ecol 1989; 160 :3 33 - 5.
2 Ja FFe AC, Dynn eson L, ten Bensel RW. Sex ual a bu se of 26 Ada ms JA, Phillips P, Ahm ad M. The usefulness o f
children: a n epi demiological study. Am J Dis Child 197 5; coJ pos copic pho to graph s in the evaluat io n o f suspected child
129 :689 - 92. sexual abuse. Adolesc Pedia tr Gyneco /1 990; 3:75-8 2.
3 Kemp e CH. Sexu al a buse, a nother hidd en pedi a tri c probl em. 27 Mura m D, J ones CE oTh e use o f vid eo- colposcopy in the
Pedia trics 1978; 62 :382-9 . gynecologic exa minati o n of infa nts, children, and yo ung
4 Finkelhor D. Current in formation on the scope and na tu re of adolescents. Ado lesc Pediatr Gym'col 1993 ; 6: 154-6.
child sexual abuse. The Future ofCIlildren J994; 4:3 1-5 3. 28 McCann Jl. The colposcop ic ge nita l exa min atio n o f the
5 Jones LM, Finkelhor D, Ko pie c K. Why is sexual a bu se sex ua lly abu sed prep ub erta l fe male. Adolesc Pedi atr GYl1 ecol
declin ing? A survey o f state chil d pro tecti on a dministra tors. 1993 ; 6:123 - 8.
Child A buse Et Neg lect 2001 ; 25 : 11 39- 58. 29 Hobb s C, Wy nn e J. Use of the colpo scope in examin ation for
6 Dunne MP, Purdi e DM, Cook MD et al. Is child sex ua l abuse sex ual abuse. Arch iues of Diseas e in Childhood 1996 ;
declining? Evid ence Fro m a popul a tio n-based survey o f men 75 :53 9- 42.
and wo me n in Australi a. Child A buse & Neg lect 2003 ; 30 Heger AH. Twe nty years in the ev aluati on of the sex ually
27:141- 52. a bused child : h as medi cin e helped or hUlt th e child a nd the
7 Hobbs CJ , Wy nne JM. Child sexual abu se: an in creasing ra te fa mily? Child Abuse Et Neglect 1996; 20:893 -7.
of di ag nos is. Lancet 1987; ii: 837- 4 5. 31 Will ott GM, Cross MA. Th e de tec tion o f sperm a tozo a in the
8 Summit RC. Th e child sexual abuse acco mmodati on mo uth. J Fo rel/sic Sci Soc 1986 ; 26 :12 5-8.
sy ndrome. Child A buse & Neglect 1983; 7 : 177-93. 32 Willot t GM, All ard JE. Sp erma tozo a - their persi sten ce a Fter
9 Mo k JYQ, Busuttil A, Hamm ond HF. The j o in t paedia tric sex ual in terco urse. ForeJ1 sic Sci Int 1982; 19:13 3-5 4.
fo rens ic exa min ati on in child a buse. Child A buse ReIJ 1998 ; 33 Davi es A, Wilson EM . Th e persi sten ce of se minal constitu ents
7 :194-203. in the hum a n vagina. Forensic Sc ience 1974; 3 :45-55.
10 Levitt CJ. Sex ua l abu se in children : A compassiona te ye t 34 Christ ian CW, Lavelle JM, De Jong AR et al. Fore nsi c ev id ence
thorough approach to eva lua tion. Postg rad IVIed 198 6 ; fi ndi ngs in prepub erta l v ictims of sex ual assault. Ped iatrics
80 :201-15. 2000 ; 106 : 100-4.
11 De San Lazaro C. Ma king paedi atric assessment in susp ected 35 Fin kel MA , Ricci LR. Docum enta ti on and prese rva ti o n o f
sex u al abus e a therap euti c ex perience. Arch Dis Childh 199 5; visua l evid ence in ch ild a buse. Ch ild Ma ltreatment 1997;
73 : 174- 6. 2 :322-30.
12 Th e Royal CoJl ege o f Pa edia trics and Child Hea lth and Fac ul ty 36 Finkel !VIA. Techni cal condu ct of the CSA medic al
of Foren sic a nd Legal Medi ci ne. Guida nce all pa ediatri c exa min at ion. Child Abuse & Neglect 1998 ; 22 :555-66.
f orensic examina tions in relation to poss ib le child sexu.al 37 Bl ock RW, Palusci V1. Ch ild Abuse Pedia trics: A new pedi atri c
abuse. September 2004. www.rcpch .ac. uk/doc.aspx? id_ subs peci alty. J Pediatr 2006 ; 148:711- 12.
Resource=3 409 38 Brayden RM, Altemeier WA IJ], Yeager T, Muram D.
13 Roya l College of Paediatri cs and Child Hea lth . Child protection Interp reta tio ns o f coJposcopi c photogra ph s: Evidence fo r
companion. Lond on, RCPCH. com peten ce in assess ing sex ua l a buse. Child Abuse Et Neglect
J4 Kendall-Tacke tt KA, Meye r Williams L, Finkelhor D. Impa ct of 1991 ;15:69-7 6.
sexu al abuse on children: A review and synthesis of rec ent 39 Adam s J A, Well s R. Normal versus a bno rma l gen ital findings
empirical st udi es. Psychol Bull 1993; )] 3 : 164 -8 0. in ch ildren: how well do exa min ers agree? Child Abuse &
15 Be rliner L, Elli ott DM. Sexual a bu se of children. In Briere J, Neglect 1993 ; 17:663-7 5.
Be rliner L, Bulkley C et al (eds) The APSAC handbook on child 40 Paradi se l E, Win ter MP, Finkel IVIA et al. Influ ence of hi story
maltreatment. Thousa nd Oaks, CA: Sage, pp. 51-71. on physici ans ' interp ret ation of girls ' genit al finding s.
16 Lev itt CJ. The medica l ex amin a tio n in child sexua l abuse: Pediatrics J999 ; 10 3 :980- 6.
A ba lance betw ee n history a nd exam. J Child Sex Abuse 41 Bu tle r-Sloss E. Report of the inqui ry il1 to child abuse ill
199 2; 1:11 3-21. Clellelan d 19 87. London: HMSO.
17 Myers JEB. The ro le of the physici an in presen tin g verbal 42 Adams JA. Gra nd round s : medica l iss ues and case s tudi es.
evid ence of child abuse. J Pediatr 19 86; 109 :409 -11. Sign ifican ce of medi cal findin gs in s uspec ted sex ual a buse:
18 Ferrell J. Foley cath eter balloon technique fo r visu ali sing th e mo vin g towards con sensus. J Child Sex Abuse 1992 ; 1 :9 1-9.
hym en in fe male ad olescent sexu al a buse vic tims. J f merg 43 Ada ms J A, Ha rper K, Kn ud son S. A proposed syste m for the
Nurs 1995 ; 21 :585-6. cla ss ifi cati on of anogenital findings in children with
19 Starling SP, Jenny C. Fo ren s ic exa mina tion of adolescen t susp ected sex ual abus e. Adolesc Pediatr Gy neco l199 2;
fema le genitalia: th e Foley cathete r technique. A rch Pedia tr 5:73-5.
Adolesc Med 1997 ; 151 : 10 2- 3. 44 Ameri can Aca demy of Pediatrics Committee on Child Abuse
20 Persa ud Dl, Squires JE, Rubin-R emer D. Use of Foley ca theter a nd Neg lect. Gonorrhea in prepu berta l children. Pedia trics
to examin e estrogenised hy mens for evidence of sexual abu se. 1998; 101: 13 4- 5.
J Pediatr Adolesc GYl1ecol 1997; 10: 83-5. 45 Ada ms J A. App ro ach to the in terpreta ti on of medica l and
21 Lauber A, Souma G. Use o f toluidine blue fo r docum enta ti on labora tory findings in sus pected child sex ual abu se: a 2005
of trauma tic intercourse. Obs tet Gyneeol 1982 ; 60:64 4-8. revisio n. Th e APSAC Advisor 200 5; Summer:7-1 3.
22 McC au ley J, Gorma n RL, Guzins ki G. Toluid in e blue in the 46 Royal College of Physi Cia ns of Lond on. Physical Sig ns of
detecti on o f pe rineal la cera ti o ns in pedi a tri c a nd adolescen t Se.rual A bus e in Child re n. Lond on: Lav enham Press , 1997.
sex ual a buse victims. Pediatri cs 1986; 8 6: 1039-43. 47 Ro yal College o f Paediatrics a nd Child He alth. Th e physical
23 Tei xei ra WR . Hy menal col poscopi c ex amina tion in sex ual signs of child sexu al abuse. A n elJ ide nce-based relJ iew and
oFfences. A m J Forensic Mcd 19 81; 2 :209-14. gu idance fo r best pra ctice. Lo nd on, RCPCH.
48 Heger AH, Ticson L, Guerra L et al. Appearance of the 73 Hobbs CJ, Wynne JM. Sexual abuse of English boys and girls:
genitalia in girls selected for non-abuse: review of hymenal the importance of anal examination. Child Abuse Et Neglect
morphology and non-specific findings. ] Pediatr Adolesc 1989; 13:195-210.
Gynecol2002; 15:27-35. 74 Berenson AB, Somma-Garcia, A, Barnett, S. Perianal findings
49 Berenson AB. Appearance of the hymen at birth and one year
in infants 18 months of age and younger. Pediatrics 1993;
of age: A longitudinal study. Pediatrics 1993; 91:820-5.
91 :838-40.
50 Berenson AB. A longitudinal study of hymenal morphology in
75 Finkel MA. Anogenital trauma in sexually abused children.
the first 3 years of life. Pediatrics 1995; 95:490-6. Pediatrics 1989; 84:317-22.
51 Yordan EE, Yordan RA. The hymen and Tanner staging of the 76 McCann J, Voris J, Simon M. Genital injuries resulting from
breast. Adolesc Pediair Gynecol1992; 5:76-9. sexual abuse: a longitudinal study. Pediatrics 1992;
52 McCann J, Wells R, Simon M, Voris J. Genital findings in 89:307-17.
prepubel1al girls selected for nonabuse: a descriptive study. 77 McCann J, Voris J. Perianal injuries resulting from sexual
Pediatrics 1990; 86:428-39. abuse: a longitudinal study. Pediatrics 1993; 91:390-3.
53 Berenson AB, Heger, AH, Hayes, JM et al. Appearance of the 78 Muram D. Child sexual abuse: Relationship between sexual
hymen in prepubertal girls. Pediatrics 1992; 89:387-94. acts and genital findings. Child Abuse Et Neglect 1989;
54 McCann J, Voris J, Simon M, Wells R. Comparison of genital 13:211-16.
examination techniques in prepubertal girls. Pediatrics 1990; 79 Adams J, Harper K, Knudson S, Revilla J. Examination
85:182-7. findings in legally confirmed child sexual abuse: it's normal
55 Gardner JJ. Descriptive study of genital variation in healthy, to be normal. Pediatrics 1994; 94:310-17.
nonabused premenarchal girls.] Pediatr 1992; 12:251-7. 80 Adams JA, Knudson, S. Genital findings in adolescent girls
56 Berenson AB, Chacko MR, Wiemann CM et al. A case-control referred for suspected sexual abuse. Arch Pediatr Adolesc Med
study of anatomic changes resulting from sexual abuse. Am ] 1996; 150:850-7.
Obstet Gynecol 2000; 182:820-34. 81 Heger A, Ticson L, Valesquez 0, Bernier R. Children referred
57 Berenson A, Heger, A, Andrews, S. Appearance of the hymen for possible sexual abuse: medical findings in 2384 children.
in newborns. Pediatrics 1991; 87:458-65. Child Abuse Et Neglect 2002; 26:645-59.
58 Emans SJ, Woods ER, Flagg NT, Freeman A. Genital findings 82 Slaughter L, Brown CRV, Crowley S, Peck R. Patterns of
in sexually abused, symptomatic and asymptomatic girls. genital injuIY in female sexual assault victims. Am ] Obstetr
Pediatrics 1987: 79 :778-85. Gynecol 1997; 176: 609-16.
59 Heppenstall-Heger A, McConnell G, Ticson L et al. Healing 83 Kellogg ND, Menard SW, Santos. A. Genital anatomy in
patterns in anogenital injuries: a longitudinal study of pregnant adolescents: 'Normal' does not mean 'Nothing
injuries associated with sexual abuse, accidental injuries, or happened'. Pediatrics 2004; 113 :e67-9.
genital surgery in the preadolescent child. Pediatrics 2003; 84 Paradise JE, Campos JM, Friedman HM, Frishmuth G.
112:829-37. Vulvovaginitis in premenarchal girls: clinical features and
60 Kerns DL, Ritter ML, Thomas RG. Concave hymenal variations diagnostic evaluation. Pediatrics 1999; 70:193-8.
in suspected child sexual abuse victims. Pediatrics 1992; 85 Straumanis JP, Bocchini JA. Group A beta-hemolytic
90:265-72. streptococcal vulvovaginitis in prepubertal girls: a case report
61 Hobbs CJ, Wynne JM, Thomas AJ. Colposcopic genital and review of the past twenty years. Pediatric Inject Dis]
findings in prepubertal girls assessed for sexual abuse. Arch 1990; 9:845-8.
Dis Child 1995; 73:465-9. 86 Pierce AM, Hart CA. Vulvovaginitis: causes and management.
62 Cantwell H. Vaginal inspection as it relates to child sexual Arch Dis Childh 1992; 67:509-12.
abuse in girls uncler thirteen. Ch ild Abuse Et Neglect 1983; 87 Jenny C, Kirby P, Fuquay D. Genital lichen sclerosus mistaken
7:171-6. for child sexual abuse. Pediatrics 1981; 83:597.
63 White ST, Ingram DL, Lyna PRo Vaginal introital diameter in 88 Handfield-Jones SE, Hinde FR, Kennedy CT. Lichen sclerosus
the evaluation of sexual abuse. Child Abuse Et Neglect 1989; et atrophicus in children misdiagnosed as sexual abuse. EM]
13:217-24. 1987; 294: 1404-5.
64 Kerns DL. Cool science for a hot topic. Child Abuse Et Neglect 89 Muhlendahl KE. Suspected sexual abuse in a 1O-year-old girl.
1989; 13:177-8. Lancet 1996; 348:30.
65 Paradise JE. Predictive accuracy and the diagnosis of sexual 90 Bond GR, Dowd MD, Landsman I, Rimsza M. Unintentional
abuse: a big issue about a little tissue. Child Abuse Et Neglect perineal injury in prepubescent girls: A multicenter,
1989; 13:169-76. prospective report of 56 girls. Pediatrics 1995; 95:628-31.
66 Heger A, Emans SJ. Introital diameter as the criterion for 91 Kellogg ND, Parra JM. Linea vestibularis: a previously
sexual abuse. Pediatrics 1990; 85:222-3. undescribed normal genital structure in female neonates.
67 Hobbs CJ, Wynne JM. Buggery in childhood: a common Pediatrics 1991; 87 :926-9.
syndrome of child abuse. Lancet 1986; ii:792-6. 92 Kellogg ND, Parra JM. Linea vestibularis: follow-up of a
68 McCann J, Voris J, Simon M, Wells R. Perianal findings in normal genital structure. Pediatrics 1993; 92:453-6.
prepubeltal children se.iected for nonabuse: a descriptive 93 Johnson CF. Prolapse of the urethra: Confusion of clinical and
study. Child Abuse Et Neglect 1989; 13:179-93. anatomic characteristics with sexual abuse. Pediatrics 1991;
69 Priestley B. Reflex anal dilatation and abuse. Lancet 1987; 87 :722-5.
ii:1396. 94 Wallis SM, Walker-Smith J. Case report: an unusual case of
70 Stanton A, Sunderland R. Prevalence of reflex anal dilatation Crohn's disease in a West Indian child. Acta Paediatr Scand
in 200 children. EM] 1989; 298:802-3. 1976; 65:749-51.
71 Clayden G. Reflex anal dilatation associated with severe 95 Sellman SPB, Hupertz VF, Reece RM. Crohn's disease
chronic constipation in children. Arch Dis Childh 1988; presenting as suspected abuse. Pediairics 1996; 97:272-4.
63 :832-6. 96 Friedrich WN, Grambsch P, Broughton D et al. Normative
72 Agnarsson U, Warde C, McCarthy G, Evans N. Perianal sexual behavior in children. Pediatrics 1991; 88:456-64.
appearances in childhood constipation. Arch Dis Childh 1990; 97 Friedrich WN, Fisher J, Broughton D et al. Nonnative sexual
65:1231-4. behavior in children: a contemporary sample. Pediatrics 101 :e9.
98 Neinstein LS, Goldenring J, Carpenter S. Nonsexual 108 Embree JE, Lindsay D, William s T et al. M. Acceptability a nd
transmission of sexually transmitted infections: an infrequent use fulness of vaginal washes in premenarch eal girls as a
occurrence. Pediatrics 1984; 74 :2 17-25. diagnostic procedure for sexua lly transmitted infections.
99 American Academy of Pediatrics. Sexually tra nsmitted Pediatr Inject Dis J 1996; 15 :651-66.
diseases. In Pickering LK (ed .) Rep ort oj th e Co mmittee on 109 Mok JYQ. Routine Hrv testing after child sexual abuse? Child
InJectious Diseases, 26th ed n. Elk Grove Village, lL: Ballinger, Abuse ReI! 1998; 7:6 3-9.
2003, pp. 157-67. 110 Muram D, Spe ck PM, Gold SS. Genital abnorma lities in
100 Am erica n Aca demy of Pediatrics Co mmi ttee on Child Abuse female siblings and friends of child victims of sexual abuse.
and Neglect. The eva luation of sexua l abuse of children. Child Abuse ft Neglect 1991 ; 15: 105-10.
Pediatrics 2005; 11 6:506-12. III Bays J, Ch adwic k, D. Med ical diagnosis of the sexually
101 In gra m DL, Everet t D, Lyna PR et al. Epidemi ology of adult ab used child. Ch ild Abuse ft Neglect 199 3; 17:9 1-11 0.
sexually transmitted disease agents in children being 112 Muram D. Child sex ual ab use. Curl' Opin Obstetr GYllecol
evaluated for sexual abuse. Pediatr Inject Dis J 1992 ; 1993 ; 5:784-90.
11 :9 45-50. 11 3 De Jong AR, Rose M. Legal proof of child sex ual abuse in the
102 Robinson AJ , Watkeys JEM, Ridgway GI. Sexually tra nsmitted absence of physical evidence. Pediatrics 199 1; 88:506-11.
organisms in sexually abused ch ildren. Arch Dis Childh 199 8 ; 114 Kerns DL. Triage and referral s for child sexual a buse medical
79: 356-8. examinations: Which children are likely to have positive
103 In gram DM, Miller WC, Schoenbach VJ et al. Risk assessment medical findings? Ch ild Abuse ft Neglect 1998 ; 22:515-18.
for gono co cca l an d ch lamy dial infec tions in young children 115 Chadwick DL, Krous HF. Irresp ons ibl e testimony by medical
underg oing evalua tion for sexua l ab use. Pediatrics 2001: ex perts in cases involv ing th e physica l abuse and neglect of
107 :e73-80. ch ildren. Child Maltreatment 1997; 2: 313 -21.
104 Th o mas A, Forster G, Robinson A, Rogsta d K for the Clinical 116 Adams JA. Th e role of photo do cu mentati on of genital
Effectiveness Group (Association of Genitourinary Medicine findings in medical evaluations of suspected child sexual
and the Medical Society for the Study o f Venereal Diseases). abuse. Child Maltreatment 1997; 2:341-7.
Nation al guideline for th e management of suspected sexu ally 117 Mok JYQ, Busu ttil A. Medical exa minati ons for Child Sex ual
transmitted infections in children a nd yo ung people. Sex Abuse in Scotla nd: good enough practice? Child Abuse
TrailS Inject 2002; 78:324-31 . Revi ew 2004; 2004; 13:324-37.
105 Handl ey J, Dinsmore W, Maw R et aJ. Anogenital warts in 118 Jen ny C. Pediatric fellowships in child abuse and neglect: the
children; sexual abuse or not? Int J STl ft AIDS 1993 ; 4:271-9. development o f a new subspecialty. Child Maltreatment 1997;
106 Si eg fri ed E, Rasnick-Conley J, Co ok S et al. Human 2:356-61.
pap ill o mavirus screening in pediatric victims of sexual abuse. 119 Emans SJ, Woods ER, Allred EN, Grace E. Hymena l findings
Ped iatrics 1998 ; 101 :43- 7. in adolescent wo men: impact of tampon use an d consensual
107 Steele AM , de Sa n Lazaro C. Transhymenal cultures for sexual activity. J Pedlatr 199 4; 125: 153-60.
sexually transmissible organisms. Arch Dis Childh 1994;
71 :423-7.
J. - - - - - - ---- -- -
I CHAPTER 3 I
RADIOLOGY OF CHILD ABUSE
Maeve McPhillips
Child abuse has been a concem of paediatric radiologists since occurred accidentally, or have been identified as incidental
Caffey's description of multiple fractures in children with a findings on a radiograph performed for another purpose.
subdural haematoma in 1946. He describes fractures of the Such cases w ill need to be discussed with the referring
long bones as a 'common complication of infantile subdural clinician .
haemato ma' and, although suggesting that there is a trau When referred a case of possible non-accidental injury,
matic origin, states that 'the causal mechanism remains the radiologist should review the presenting injury, and
obscure'.! In 1971 , Guthkelch, a British neurosurgeon, first discuss the patient with the clinician. A knowledge of nor
proposed shaking as a cause of infantile subdural haematoma, mal skeletal development and variations in ossification can
and the following year Caffey published his paper 'On the prevent unn ecessary investigation. 4 - 7 There shou ld be a
theory and practice of shaking infants'. 2,J Since then, non recognized departme ntal protocol for a skeletal survey for
accidental injury has been a subject of great importance to suspected non - accidental injury and all im ages should be
those who have responsibility for the care, diagnosis and reviewed by the radiologist before the child leaves the
investigation of children who may have been victims of abuse. department. Th is is to ensure high-quality images and to
Teams of clinicians work closely together, aware that only allow for any necessary supplementary views. An assess
when they are in possession of all the information available ment should be made of the age of any injuries. The possi
about the child can the correct diagnosis be reached. bility of an underlying bone disease or other condition,
together w it h the need for appropriate im aging, should be
considered at this stage.
ROLE OF THE RADIOLOGIST The report should be commun icated verb ally to the
responsible clin ician as soon as is practicable and a formal
Paediatric radiologists have expertise in the interpretation written report issued promptly. If there is doubt as to the
of imaging of in fants and children and a familiarity with presence or significance of a lesion, this should be clearly
the appearances of the normal brain and ske leton. The stated together w ith a plan of investigation to clarify the
appeara nces of the skeleton and brain followin g accidental findings. The need for further imaging, in particular neuro
trauma will be well known to them, and they will have an imaging, should be discussed with the clinician at th is
understanding of possible mechanisms of injury, and so be stage. The radiologist should be available for any discus
able to judge whether the given history is appropriate for sions with clinicia ns and members of the chi ld protection
the injury. team, including the police. Because of the possibility of
The first role of the radiolo gist is to be alert for injuries child protection or other legal proceedings, all images and
that, in the light of the history supplied, may not have reports should be retain ed in a secure place.
48 I Radiology of child abuse
RADIOLOGICAL INVESTIGATIONS
A separate exposure should be obtained of each
anatomical area to optimize image quality, reduce geomet
Skeletal Survey ric distortion and allow for the detection of subtle abnor
malities. At least two views of the skull should be obtained.
The radiographic skeletal survey is the mainstay of investi Both oblique views of the ribs should be obtained routinely.
gation of non-acc idental injury in yo ung children and Two perpendicular views should be obtained of any focal
infants. There has been much variability in the quality of injury. Coned views of the metaphyses, in anteroposterior
exa minations 8,9 a nd national stand ards have been intro (AP) and lateral projections, may be helpful to confirm or
duced by the American Academy of Paediatrics, JO the exclude classic metaphyseal lesions.
American College of Radiology,l1 the Royal College of The skeletal survey is not an emergency investigation and
Radiology and the Royal College of Paediatrics and Child should be performed during the standard working day. It
Health. 12 Table 3.1 shows a suggested protocol. requires two radiographers and a lo t of time. It should be
A ' babygram' of the whole body with a si ngle exposure performed in the radiology department unless the child is
or coverage of the whole body using th ree or four films is critically ill , in which case it may be performed using mobile
inadequate. High-quality radiographs must be obtained radiography equipment in the paediatric intensive care unit.
with optimum exposure factors and good coning. High Image quali ty may be compromised in this situation.
resolution cassettes should be used with no grid. The patient Most radiol ogy departments no longer use hard-copy
name, side marker, and date and time of the examina tio n radiographs for reporting. Soft-copy reporting has been
must be clearly visib le on the radiograph . The names of the shown to be superior to hard copy owing to its abil ity to
radiographers must be recorded: they should wo rk in pairs. vary the grey -scale settings and to magnify sections of the
Not only does this allow for increased efficiency in what image selectively. 13
can be a prolonged investigation, but it provides legal safe
guards for both the patient and the ra diographers.
The radiograp hers should obtain positive identifi cation Follow-up Skeletal Survey
of the patient from the accompanying staff or carer and the
identity should be checked on the na me band. It is impor If there is ongoing clinical concern, the skeletal survey should
tant that the responsible clinician has discussed the need be repeated in 10-14 days, apart from the skull radiographs.
for the skeletal survey with the parents or caregivers, as This is particularly relevant if the presentation has been with
they should be allowed to accompany the child during the head injury and the initial skeletal survey has been normal. If
investigation. It may also be appropriate that a member of there are areas of radiological concern, repeat radiographs
the nursing staff is in attendance. should be obtained of these areas, also at lO-14 days (Fig.
3.1). For patients in whom there is no particular diagnostic
unceltainty, it is still important to obtain repeat chest radio
graphs, including obliques, and views of the id entified bony
Table 3.1 Suggested protocol for initial skeletal survey abnormalities. It has been shown that repeat skeletal surveys
can identify a greater number of fractures, particularly of
Skull AP and lateral views ribs, and can confirm suspected fra ctures. 14-16 Lack of change
Towne's view if occipital injury over time can confirm a normal valiant. Radio grap hs taken
suggested clinically on two separate occasions can also help more accurate dating
Spine Lateral views of cervical, thoracic and
of injury and show evidence of different ages of injuries.
lumbosacra l sp ine
Chest AP to include clavicles
Lateral Post-mortem Skeletal Survey
Oblique views of both ribs - each to
include whole chest The same high standards should be applied to the examina
Abdomen AP whole abdomen, to include pelvis tion of the deceased child as to the live child. A 'ba bygram'
and hips should not be performed . Individual exposures of each
Limbs AP both upper arms
anatomical area should be obtained, with particular atten
AP both forearms tion paid to the presence of COITect identifiers on each image.
PA both hands The skeletal survey must be obtained prior to the autopsy
AP both femora (Fig. 3.2). If necessary, the autopsy should be delayed to
AP both lower legs allow this. The reporting radiologist should co mmunicate
AP both feet his or her findings to the pathologist as soon as possible.
Supplementary Additional views of any questionable areas A verbal report will suffice, as long as it is documented in
AP and la tera l coned vie ws of metaphyses the notes and in the final radiological report. Detailed
AP, anteroposterior; PA, posteroanterior. radiography of removed specimens of bone can be very
- - - ~ - - - ---- --- -- - - -
Radiological investigations I 49
Ultrasound
- ~ "'- -
-----~--
50 I Radiology of child abuse
- - -- - - - - --- ---- - -- - - - --
Skeletal injuries I 51
Figure 3.6 Isotope bone scan of the upper body show ing
multiple areas of increased uptake in the posterior ribs,
Figure 3.5 Magnetic resonance scan, Tl-weighted image, representing fractures of the posterior shafts at their necks. A
showing a fresh small linear subdural in the right posterior spi ral fracture of the left humeral shaft is identifiable by the
parietal region and high signal as a result of fresh haemorrhage asymmetrical uptake in the humeri.
(arrows). Overlying this is a scalp haematoma. A skull fracture was
not visible on magnetic resonance imaging. therefore be further assessed by radiography. Bone scintig
raphy is complementary to radiographic skeletal survey,
and may be useful in individual patients.
and venography, and to use transverse Tl-weighted fat
saturated sequences to search specifically for inj ury to the
major vessels in the neck. 28 Spinal cord injury is best assessed SKELETAL INJURIES
with MR, using short tau inversion recovery (STIR) sequences.
Post-mortem MR scanning has been shown to be more Fractures in child abuse are most common in infants and
sensitive than autopsy alone in the detection of a shearing children under 2 years of age. It is convenient to separate
injury, cortical haemorrhage and mastoid fluid, and to be fractures into two groups: (a) those that are seen fre
equal to autopsy in the demonstration of cerebral oedema, quently, but are not very specific for non-accidental injury;
focal contusions and subfalcine herniation. 3D and (b) those that are highly specific for inflicted injury but
are less commonly seen (Tables 3.2 and 3.3).33
Scintigraphy
Periosteal New Bone
Isotope bone scanning, using technetium-99-labelled
methylenediphosphonate (99Tc-MDPl, can show bone Fractures repair by the laying down of subperiosteal new
pathology. In one series,3l 10 per cent of fractures were bone. With no obvious fracture, the presence of sub
seen only on scintigraphy, and Mandelstam has shown that periosteal bone may be due to injury to the periosteum by
50 per cent of rib fractures may be visible only on isotope rough handl ing, or by acceleration-deceleration forces
bone scanning 32 (Fig. 3.6). The yield from radiography may during shaking. Physiological periosteal reaction is seen in
have improved since the introduction of oblique views of normal infants as young as 8 weeks, up to about 8 months
the ribs at presentation. Many fractures also become appar of age. It is smooth, rarely more than 2 mm in thickness
ent on follow-up radiographs. Scintigraphy has a Jow sen and is seen along the diaphyses of the long bones. It is usu
sitivity for skull fractures. Because of the high activity in ally symmetrical, though may be more obvious on one side.
normal growth plates, metaphyseal fractures may not be
apparent on isotope bone scanning, particularly if bilateral Long-bone Fractures
(Fig. 3.7). Fractures may show increased activity on
scintigraphy for up to a year after injury. Any areas of Diaphyseal fractures of long bones are common in non
increased activity detected using scintigraphy should accidental injury; some authors have found them to be four
52 I Radiology of child abuse
(a) Table 3.2 Fractures that are frequent but with low specificity for
PO :;. T non-accidental injury
LT Midclavicular fractures
Simple linear skull fractures
Single diaphyseal fractures
Reproduced with pe rmission from Cart y33
. .- -- - -- -- - - - --
Skeletal injuries I 53
~------
54 I Radiology of child abuse
-- ----- ---- - --
Skeletal injuries I 57
sternal compression on a firm surface and the posterior ribs Figure 3.21 Healing bucket handle metaphyseal fractures of the
showed no change in their relationship to the spine; however, distal tibia bilaterally with disproportionate periosteal new bone
fractures did occur with manual compression. Lateral rib formation. Irregular periosteal new bone on the fibular diaphyses
fractures are due to the compressive force on the ribs and are bilaterally. The appearances probably represent a direct gripping injury.
-... . , - ~
58 I Radiology of child abuse
. -- - --
Skeletal injuries I 59
require CT, but MRI will be necessalY to exclude a ny asso fracture healing. A radiologist whose practice involves paedi
ciated cord injury. atric trauma will have co nsiderable experience in the report
ing of acute and healing fractures, when the timing of the
Digital Injuries injUly is precisely known . This can then be extrapolated to
the appearance of fractures for which there is an inad equate
history. Despite the lack of published data, there is remark
Injuries to the metacarpals, metatarsals and phal a nges are
able agreement between radiologists. Approximate dating of
uncommon, but have a high specificity for abuse. They are
usually torus fractures, and the mechanism is thou ght to be
twisting, bending or hyperextension 58 (Fig. 3.25). Crush
fractures may also be seen.
Dating of Injuries
-
~ -
~-------
-
~
- -- , -
- --~--
60 I Radiology of child abuse
Time Peak
HEAD INJURY
- - - - . - --- -- -- - - -
Head injury I 61
Figure 3.30 Magnetic resonance scans of an infant. (a) Fluid-attenuated inversion recovery [FLAIR) coronal, (b) Tl-weighted sagittal at
presentation and (c) T1 sagittal 10 weeks later. High-intensity interhemispheric subdural haemorrhage is indicated by arrowheads. The
low-intensity posterior fossa subdural collections [stars) are presumed to be older. Secondary herniation of the craniocervical junction
(white arrows) resolved along with the subdural. An ill-defined, subependymal, high-intensity signal suggests a shearing injury [arrow).
There is also a large subgaleal fluid collection [asterisk). Adapted with permission from McPh il lips.75
- 4.: - - -- --- ~ - ~
Head injury I 63
Subarachnoid Haemorrhage
Figure 3.32 (a) Acute computerized tomography scan showing high-density fresh interhemispheric subdural haemorrhage poster iorly.
There is some loss of the normal grey-white matter different iation due to mild cerebral oedema . (b) Follow-up scan 4 weeks later. There is
mild generalized cerebral atrophy.
64 I Radiology of child abuse
Figure 3.33 (a) Fluid-attenuated inve rsion recovery (FLAIR) coronal MR image at presentation shows a thin layer of subdural blood over
the convexity on the right and in the subtemporal region bilaterally (arrows). T2-weighted transverse images. (b) At presentation there is
loss of grey-white matter differentiation on the right. with sl ight swelling and subtle midline shift. (e) Ten days later there is increased
intensity in the right hemisphere and left frontal lobe. with prominent sulci due to early atrophy. (d) Nine months later there is extensive
atrophy on the right. sparing only the basal ganglia. The left hemisphere is normal at this level. Adapted with permission from McPhillips.75
- - --
Head injury I 65
scanning, MR scanning, particularly in the coronal plane, the grey-white matter junction in the frontal or frontopari
demonstrates them well, and gradient echo (GRE) images etal lobes or in the corpus callosum (Figs 3.30 and 8.17,
are very sensitive for blood products, which are seen as low p. 157). At the grey-white matter junction it may actually
intensity. They may be subtle at autopsy, and knowledge of be seen as a tear, which is readily visualized using high
the MR findings will direct attention to the appropriate frequency ultrasound scanning, and is seen as a low
areas 30 echogenicity focus. 22 Tears may be seen on CT, but as they lie
Shearing injury is unusual in head injury but, when seen close to the vertex they are not easily characterized. On MR
in the absence of a history of a high-velocity injury, such as they may be seen as haemorrhagic (Fig. 3.34) tears initially,
a road traffic accident, it is highly suggestive of non-accidental later becoming cystic. Follow-up imaging shows that some
injury. The most common site for focal shearing injury is at persist as cysts 85 (Fig. 3.35), while some show focal gliosis
Figure 3.34 Sagittal Tl-weighted magnetic resonance images. The presentation scan (al shows increased intensity posteriorly, in
keeping with layering in a subdural haemorrhage (arrowheads), and a focal area of high intensity (arrow) over the frontoparietal
convexity, suggestive of focal clot formation. Six days later (bl, an ill-defined, high-intensity signal can be seen in the frontoparietal
subcortical white matter (arrow), representing a tear. Two months later (cl, there is residual focal atrophy (white arrows). Adapted with
permission from McPhilli ps l5
..
-------
66 I Radiology of child abuse
Figure 3.35 Coronal fluid-attenuated inversion recovery (FLAIR) magnetic resonance im ages at (al 3 days, (bl 3 weeks and (cl13
months after presentation. (a) III-defined, high-inten sity signal over the left convexity (arrow) and subtle irregularity inthe underlying
parenchyma. (b) Cortical tear with low-intensity cyst formation (arrow) . (c) The tear now appears as a slit-like lesion (arrowl with
adjacent gliosis (arrowheads) and subtle atrophy. Adapted from with permission from McPhilli ps 85
and others collapse with associated white matter loss 86 Oedema or swelling of the brain may be as a direct result
(Fig. 3.3 4) . Sometimes shearing injury may be seen as tiny of trauma, or may be a result of hypoxic-ischaemic damage
focal haemorrhagic areas on MR scanning, and knowledge of or hyperaemia. Traumatic oedema is usually associated with
these findings may direct the pathologist to a subtle lesion. 3D contusion or parenchymal haemorrh age. Hypoxic-ischaemic
0- _ - - - - -
Head injury I 67
~ ---~-
68 I Radiology of child abuse
VISCERAL INJURIES
-- - - - - - --
Differential diagnosis I 69
SOFT-TISSUE INJURY
Accidential Injury
~-- - --
Differential diagnosis I 71
2 mm, and are not assoc iated with overlying soft-tissue described, and it should be remembered that it can be detected
swelling. for up to 4 weeks following delivery.77,78
The most common fracture sustained during delivery is of the Periosteal reaction and minor metaphysea l or cortical
midshaft of the clavicle. It may not be recogni zed at birth but irregularity may be see n on ultrasound or, sev era l days
it either presents at 2-3 weeks of age, with a palpable lump, later, radiographically. In the absence of a histolY of
or is seen as an incidental finding on a chest radiograph. Most trauma or clinical evidence of infection this may be
h ave healed by 12 weeks and are undetectable. A few fail to ascri bed to non-accidental injury.
unite, leaving a pseudoal1hrosis of the clavicle, which may be
mistaken for a new fracture at a later date. Other diaphyseal
fractures are usually recognized owing to reduced movement Metabolic Bone Disease
of the limb, humeral fractures in shoulder dystocia and
femoral fractures in extended breech delivery. Rib fractures, Several conditions can present with reluctance to use a
resulting from impaction against the maternal symphysis limb, in the absence of trauma. Radiographs may show a
pubis, are usually unilateral. 43 ,44 They may present as a result metaphyseal lucent line. Careful inspection of the radi
of crepitus or respiratOlY distress, or be seen as an incidental ograph will show that this is present at every metaphys is,
finding on a chest radiograph . Metaphyseal fractures may though more obvious at some. There is also no visible frac
rarely be due to birth trauma, when a limb has been twisted or ture. This is a well-recognized presentation of leukae mia
pulled, in the course of Caesarean section,39 an armling deliv and neuroblastoma, and the child is likely to be systemi
ery or a breech delivery. The finding of subdural haematoma cally unwell. Langerhans cell histiocytosis may h ave a sim
in a significant number of asymptomatic neonates has been ilar presentatio n but the bone lesions are unlikely to be
symmetrical and generalized.
Meta physeal fractures are well recognized in scurvy,
Menke syndrome and metabolic disease of the newborn.
Scurvy has typical radiological appeara nces, with osteopenia,
blurred metaphyses and poor trabeculation. Clinical appea r
ances and biochemical investigations will clarify the diagno
sis in most cases.
Copper deficiency is rare, but well described. 108 Prema
turity, mal nutrition, malabsorption and dietary defic iency
are predisposing factors. Copper has been added to modern
milk formulae and parenteral feeds to prevent its defi
ciency in preterm infants. Fractures are seen in copp er defi
ciency only in the presence of widespread symmetrical
skeletal chan ges, showing periosteal reaction, metaphyseal
spurs and cupping and fraying of the metaphyses, visible
especially at the wrists, knees and costochondral junctions.
Biochemical investigations will confirm this disorder.
Rickets is characterized by metaphyseal fraying and cup
ping, with osteopenia and poor trabeculation. Fractures may
occur, but only in the presence of obvious bony abnormal
ity, confirmed by biochemical abnorm alities (Fig. 3.43).
Congenital syp hilis can be confirmed by clinical
features and serology. Metaph yseal in'egularity is seen, but
the appearances are those of undermining of the metaphysis.
Osteogenesis Imperfecta
Figure 3.42 (a) Simple torus fracture of the distal tibia with
slight change in the normally smooth contour of the anterior Osteogenesis imperfecta is a disorder of synthesis of type I
cortex. This fracture is technically metaphyseal, but this is a collagen, with resulting bone and connective tissue abnor
simple accidental injury, not non-accidental injury. (b) Healing malities. It is an inherited condition, with four major types
with sclerosis at the fracture site. described by Sillence. 109
72 I Radiology of child abuse
Figure 3.43 Immigrant child with nutritional rickets and severe osteopenia. (a) Left lower limb shows healing fractures of the tibia and
fibula and a fresh corner fracture of the proximal tibia. (b) Upper limb shows a healing fracture of the radius and abnormal periosteal new
bone extending to the humeral metaphysis. Typical metaphyseal fraying and cupping best seen in the distal ulna. The faint metaphyseal
sclerosis is due to healing, following treatment. A diagnosis of non-accidental injury was not made because of the rickets, although there
were grave concerns about the family.
• Type I is the most prevalent form, with blue sclerae, a distinguish osteogenesis type IV from non-accidental
family histolY of deafness, and sometimes abnormal injury.
dental development. Ten per cent of people with
osteogenesis imperfecta show fractures at birth. This is
an autosomal dominant condition but new mutations It should be remembered that children with osteogenesis
do occur. imperfecta susta in fractures similar to fractures of other
• Type II can be recognized in utero and is usually lethal children from accidental trauma, just with a lesser degree of
in the fetal or perinatal period. It is thought to be force. They w ill therefore have fractures that are similar in
transmitted as an autosomal recessive condition. distribution to those in accidental trauma, and would not be
• Type III is a rare autosomal recessive form of expected to show metaphyseal corner fractures. Simila rly,
osteogenesis imperfecta, resulting in deformity of the rib fractures and skull fractures are not commonly seen in
limbs, spine and skull. Fractures are often present at osteogenesis imperfecta. Because fractures are painful and
birth. cause the child distress, there should not be any significant
• Type IV is regarded as a rare form of osteogenesis delay in presentation to medical attention; therefore, previ
imperfecta. The sclerae are normal in appearance. The ously unidentified fractures in various stages of healing
bone disease is very variable in severity. This group would not be seen. Subdural haemorrhage is a rare feature
may be subdivided, according to the presence of of osteogenesis imperfecta, recognized in severe cases, in
dentinogenesis imperfecta. In a mild case, particularly which the diagnosis is not in doubt. 110 Brain injury is not a
because of the normal sclerae, it may be difficult to feature of osteogenes is imperfecta.
.. - -- -
References I 73
There are rep orts of suspected non-accidental injury 5 Kleinman PK. Schm id-like metaphyseal chondrodysplasia
that have later been proven to be osteogenesis imperfecta . Sim ul a tin g child abuse. A]R )991; 156: 576-8.
that wo uld render a child more susceptible to fractures for non-accidental injury im aging in England, Scotland and
a short period of time, with spontan eo us resolution of the Wales. Ciin Radio/2003; 58:696-701.
28 Barnes PD, Krasnokutsky M. Im aging of the central nervous 51 Kleinman PK, Marks Sc. Vertebral body fra ctures in child
system in suspected or alleged nona ccid ental injury, including abuse: radiologic-histopathologic correlates. In vestigative
the mimics. Topics MR Imaging 2007 ; 18:53-74. Radiology 1992; 27:715-2 2.
29 Suh DY, Dav is PC , Hopkins KL, et a1. Nonaccidental pediatric 52 Kleinman PK, Zito JL. Avulsion of the s pinous processes
head injUly: diffusion-weighted imaging findings. caused by infant ab use. Radiology 1984; 151:389-91.
Neurosurgery 2001; 49 :309-20. 53 Rooks VJ , Sisler C, Burton B. Cervical spine injury in child
30 Hart BL, Dudley MH, Zumwalt RE. Post mortem cra nial MR T abuse. Report of 2 cases. Pediatr Radiol 1998; 28:193-5.
and autopsy correlation in suspected child abuse. Am ) 54 Cullen Jc. Spinal lesions in battered babies. ) Bone Joint Surg
Foren sic NJed Patho/1996; 17: 2 17-24. Br 1975; 57B:364-6.
31 Sty JR, Starsh a k R1. The role of bon e sci ntigraphy in the 55 Lonergan GJ , Baker AM, Morey MK, Boos Sc. Child abuse:
eva luation of the sus pec ted abused child. Radiology 1983; radiologic-pathologic correlation. Radiograph ics 2003;
146:369-7 5. 23:811-45.
32 Ma nd elstam SA, Cook D, Fitzgera ld M, Ditchfielcl MR. 56 Kogut! MS, Swischuk LE, Fagan CJ. Patterns of injury and
Complementary use of radiolog ical skel etal survey and bo ne significance of uncommon fractures in th e battered child
scintigraphy in detection of bony injuri es in su spected child syndrome. Am) Roencgenol Radium Ther Nucl Med 1974;
abuse. Arch Dis Child 2003; 88:387-90. 121:143-149.
33 Carty HML. Fractures ca used by child abuse. ) Bo ne Joint 57 Kleinman PK, Shelton YA. Hangman's fracture in an abused
Surg Br 1993; 75B:849-57. infant: imaging features. Pedian' Radio11997; 27:776 -7.
34 Rao P, Carty H. Non-accidental injury: review of the 58 Nimkin K, Spevak MR, Kleinman PK. Fractures of the hands
radiology. Clin Radiol 1999; 54: 11-24. a nd feet in child abuse: Imaging and pathologic features.
35 Taitz J , Moran K, O'Mea ra M. Long bone fractures in children Radiology 1997; 203 :233- 6.
under 3 years o f age: is abuse being missed in emergency 59 Carty H. Th e non-skeletal injuries of child abuse: part II the
department presentatiolls? ) Paediatr Child Health 2004; body. In The Year Book of Paediatric Radiology, Borsod
40:170-4. County Teach ing Hosp ital, Miskolc, 1991 ; 3:25-34.
36 Worlock P, Sto wer M, Barbor P. Pattern s of fractures in 60 Hechter S, Huyer D, Manson D. Sternal fractures as a
accidental and non-acc id ental injulY ill children: a manifestation of abusive injury in children . Pediatr Radiol
comparative study. Br Med) Ciin Res Ed 1986; 293:100-2. 2002 ; 32 :902-6.
37 Kleinman PK, Marks Sc, Richmond 1M, Blackbo urne BD. 61 Ablin DS, Greenspan A, Reinhart MA. Pelvic injuries in child
lnflicted skeletal injury: a postmortem radiologic abuse. Pediatr Radiol 1922; 22 :4 54-7.
histopathologic study in 31 infants. AJR 1995; 165: 647-650. 62 O'Connor JF, Cohen 1. In Kleinman PK (ed.) Diagnostic Im aging
38 Grayev AM, Boal DKB, Wallach DM, Segal LS. Metaphyseal of Child Abuse. Baltimore: Wi lli ams and Wilkins, 1987, p. 112.
fractures mimicking abuse durin g treatment for clubfoot. 63 Prosser I, Maguire S, Harrison SK, et al. How old is this
Pedin tr Radiol 2001; 3 1 :559-63. fracture? Radiologic dating of fractures in children: a
39 O'Connell AM, Do noghue VB. Can classic metaphysea l lesions systematic review. A JR 200 5; 184: 1282 -6.
fol low un co mpli ca ted caesarean section? Pediarr Radiol 2007; 64 Haviland J, Ross Ru sse ll RI. Outcome after severe non
3 7:488-91. accidelltal head injury. Arch Dis Child 1997; 77:504-7.
40 Kleinman PK, Marks SC, Bl ackbourne B. The metaphyseal 65 Jayawant S, Rawlinson A, Gibbon F, et al. Subdural
lesion in a bu sed infants: a radiologic-histopathologic study. haemorrhages in infants: population ba sed s tudy. BM) 1998 ;
AJR 1986 ; 146:895-905. 317:1 558-61.
41 Bulloch B, Schubert CJ, Brophy PD, et a1. Cause and clinical 66 King SJ, Boothroyd AE. Cranial trauma following bi rth in
character isti cs of rib fr ac tures in infants. Pediatrics 2000; term illfa nts. Br J Radiol 1998 ; 71 :233-8.
)05:E48. 67 Lo TYM, McPhillips M, Minns RA, Gibson RJ. Cerebral atrophy
42 Cadzow SP, Armstrong KL. Rib fr ac tu res in infants: red al ert! following shaken impact syndrome and other non-accidental
The clini cal features, investig ations and child protection head injury. Pediatric Rehabilita tion 2003; 6:47-55.
ou tco mes. ) Paediarr Child HealrlJ 2000; 36 :322-6. 68 Lloyd DA , Carty H, Patterson M, et al. Predictive value of
43 Durani Y, DePiero AD. Images in internal medicine. An n skull radio g raphy for intracranial injury in children with
Em erg Med 2006; 47:210-15. blunt head injUly. Lancet 1997; 349:821-4.
44 Rizzolo PJ, Coleman PR. Neonatal rib fracture: birth trauma 69 Billmire ME, Myers PA. Serious head injury in infants:
or ch ild abu se 7 ) Family Practice 1989; 29:561-3. accident or abuse? Pediatrics 1985; 75:340- 2.
45 Kleinman PK , Marks SC, Adams VI, Blackbourne BD. Factors 70 Warrington SA, Wright CM , ALSPAC Study Team. Accidents
affecting visualization of po sterior rib fractures in abused and reS Ulting injuries in premobile infants: data from tb e
infants. A)R 1988; 150:635-8. AlSPAC stu dy. Arch Dis Child 200 1; 85: 104-7.
46 Kleinman PK, Schlesinger AE. Mechanical fa c tors associated 71 Rob erton DM, Barbor P, Hull D. Unusual injUly? Re cent injury
with posterior rib fractures: laboratolY and case studies. in normal children and children with suspected non
Pediotr Radiol 1997; 27:87-91. accidenta l injury. BMJ 1982; 285:1399-401.
47 Kleinman PK, Marks Sc, Nimkin K, et a1. Rib fra ctu res in 3 1 72 Helfer RE, SJovis TL, Black M. Injuries resu lting when small
abused infants: postmortem radiologic-histopathologic study. children fallout of bed. Ped iatrics 1977;60:53 3-5.
Radiology 1996; 200:807-10. 73 Kravitz H, Driessen G, Gomberg R and Korach A. Accid ental
48 Feldman KW, Brewer DK. Child abuse, cardiopulmonary fa lls from eleva ted surfaces in infants from birth to o ne yea r
resuscitation and rib fra ctures. Pediarrics 1984; 73:339-42. of age. Pediatri cs 1969; 44 [Supp1.):869-76.
49 Sp evak MR , Kl ei nm a n PK, Belanger PL, et a1. 74 Stoodley N. Neuroima ging in non-accidental head injury: if,
Cardiopulmonary resuscitation an d rib fractures in infan ts . A when, why and how 7 Ciin Radial 2005; 60 :22-30.
post- mortem radiologic-pathologic st ud y. )AMA 1994 ; 75 McPhillips M. Non-acc identa l injulY in the yOUllg infant. ln
272:617-18. Rutherford M [ed.) MRJ of the Neonatal Bmin. London: WB
50 Dolinak, D. Rib fractures in infants due to cardiopulmon ary Saunders, 2002 , pp. 261 -71.
resuscitation efforts. Am J Forensic iVied Parlwl 2007; 76 Newton RW. Inrra cranial haemorrhage and non-accidental
28:107-10. injury. Arch Dis Child 19 89; 64:188-190.
... - -- --
References I 75
77 Whitby EH, Griffiths PO, Rutter S, et al. Frequency and 97 Siovis TL, Berdon WE , Haller 10, et at. Pancreatitis and the
natural history of subdural haemorrhage in babies and battered child syndrome: Report of 2 cases with skeleta l
relation to obstetric factors. Lancet 2004; 363:846-851. involvement. Am J Roentgenol Radium Tit er Nucl iVIed 1975;
78 Looney CB, Smith JK, Merck LH, et al. Intracranial 125:456-61.
hemorrhage in asymptomatic neonates: prevalence on MR 98 Sivit Cl, Ingram 10, Taylor GA , et at. Post traumatic adrena l
images and relationship to obstetric and neonatal risk factors. hemorrhage in children. CT findings in 34 patients. AJR
Radiology 2007; 242:535-541. 1992; 158;1299-302.
79 Tung GA, Kumar M, Richardson RC, et al. Comparison of 99 KJeinman PK, Brill PW, Winchester P. Resolving duodenal
accidenta l and nonaccidental traumatic head injury in jejunal hematoma in abused chi ldren. Radiology 1986 ;
children o n noncontrast computed tomography. Pediatrics 160:747-50.
2006; 118:626 - 33. 100 Fulcher AS, Das Narla L, Brewer WH . Gastric hematoma and
80 Duhaime AC, AJario AJ, Lewand er WJ, et al. Head injury in pneumatosis in chi ld abuse. Am J RO(,lltgenol 1990;
very young children: mechanisms, injury types and 155: 1283-4.
ophtha lmologic findings in 100 hospitalized patients younger 101 Ablin OS, Reinhart MA. Esophagea l perforation by a tooth in
than 2 years of age. Pediatrics 199 2; 90: 179-85. ch ild abuse. Pediatr Radial 1992; 22:339-4 1.
81 Ew ing-Cobbs L, Kramer L, Prasad M, et a l. Neuroimaging, 102 McDowell HP, Fielding Ow. Traumatic perforation of the
physica l, and developmenta l findings after inflicted and hypopharynx: an unusual form of abuse. Arch Dis CiJild
Pediatrics 1998; 102 :3 00-7. 103 KJeinman PK, Spevak MR, Hansen M. Mediastinal
82 Hy mel KP, Makoroff Kl, Laskey AL, et al. Mechanisms, pseudocyst ca used by pharyngeal perforation during ch ild
clinica l prese ntations, injuries, and outcomes from infli cted abuse. Am J Roentgenol 1992; 158:1111-1 3.
ve rsus noninflicted head trauma during infancy: resu lts of a 104 Ablin OS, Greenspan A, Reinhart MA. Pelvic injuries in child
prospecti ve, multicente red, co mp arative study. Pedia trics abuse. Pediatr Radial 1922; 22:454-7.
2007; 119:922-9. 105 Carty H. Case report; ch ild abuse - necklace calcification - a
83 Case ME, Graham MA, Handy TC, et aJ. Position paper on sign of strangulation? Br J Radio11993; 66:1186-8.
fata l abusive head injuries in infants and yo ung children. 106 Twomey EL, Naughten ER, Donoghue VB, Ryan S.
Am J Forens Med Path 2001 ; 22: 112 - 22. Neuroimaging findings in glutaric aciduria type I. Pedia tr
84 We lls RG, Vetter C, La ud P. Traumati c low attenuation Radio/2003; 33:823-30.
subdural fluid co llec tions in chi ldren younger than 3 years. 107 Hartley LM, Khwaja OS, Verity CM. Glutaric aciduria type I
Arch Pediatr Adolesc Med 2003; 157: J005 -1 0. and nonaccidental head injury. Pediatrics 2001; 107; 174-5.
85 McPhillip s M. Initial and sequential MRI in non-accidental 108 Shaw JCL. Copper deficiency and non-accidental injury.
hea d injUiy. In Minns RA, Brown JK (eds.) Shaking and other Arch Dis Child 1988; 63: 448-455.
Non-a ccidental Head Injuries ill Children. London: MacKeith 109 Sillence D. Osteogenesis imperfecta. An expanding
Press, 2005, pp. 262-70. panorama of variants. Clin Orthop 1981; 159:11-25.
86 Ordia IJ. Strand R, Gilles F, Welch K. Computed tomography 110 Pozzati E, Poppi M, Gaist G. Acute bilateral extradural
of contusional clefts in the white matter in infants. hematomas in a case of osteogenesis imperfecta congenital.
J Neurosurg 1981; 54:696-8. Neurosurgel)' 1983; 13:66-8.
87 Han BK, Towbin RB, de Courten-Myers G, et al. Rev ersa l sign III Gahagan S, Rimsza ME. Child abuse or osteogenesis
on CT: Effect of anoxic ischae mi c cereb ra l injury in ch ildren. imperfecta: how can we tell? Pediatrics 1991; 88:987-92.
Am J Neuroradiol 1989; 10: 1191-8. 112 Wardinsky TO, Vizcarrondo FE, Cruz BK. The mistaken
88 Kleinman PK, Zito lL, David so n RI, Raptopoulos V. Th e diagnosis of child abuse: a three-year USAF medical center
subarachnoid spaces in ch ildren: normal variations in size. analysis and literature review. Mil Med 1995; 160:15-20.
Radiology 1983; 147:455-7. 113 Paterson CR, Burns J, f\IlcAJlion SJ. Osteogenesis imperfecta:
89 Kapila A, Trice J, Spies WG, et a!. Enl a rged ce reb rosp inal fluid the distinction from child abuse and the recognition of a
spaces in infants with subdural hematomas. Radiology 1982; variant form. Am J Med Gellet 1993; 45:J87-92.
142:669-72. 114 Chapman S, Hall CM. Non-accidental injury or brittle bones.
90 Libicher M, Trager J. US meas urements of the subarac hnoid Pediatr Radiol 1997; 27: 106-10.
space in infants: normal values. Radiology 1992; 184:749-51. 115 Miller MN, Hangartner TN. Temporary brittle bone disease:
91 Wilkinson AG, Cooke R, Tallur KK, et al. Pericereb ra l space association with decreased fetal movement and osteopenia.
measurements in infants: sonographic determination. Paper Calc Tiss Int, 1999; 64:137-43.
presented at Hydrocephalus 2005 Internati ona l Co nsensus 116 Mendels on Kl. Critical review of 'temporary brittle bone
Meeting, Queenstown, New Zealand, August 2005. disease'. Pediatr Radiol 2005; 35:1036-40.
92 Fessell DP, Frankel DA, Wolfson WP. Sonography of 117 Rodriguez Jl, Palacios J, Ruiz A. et al. Morphological
extraaxial fluid in neurolo gica lly normal infants w it h head cha nges in long bone development in fetal akinesia
circumference greater than or equal to the 95th percentile for deformation sequ ence: An experimenta l study in curarized
age. J Ultrasound !vIed 2000; 19:443-7. rat fetu ses. Teratology 1992; 45:213-21.
93 Nimkin K. Teeger S, Wa lla ch MT, et a!. Adrenal hemorrhage in 118 Knight 01 , Bennet Gc. Non-accidental injury in osteogenesis
abused children: imaging and post mortem findings. AJR imperfecta: a case report. J Pediatr Orthop '1990; 10:542-4.
1994; 162: 661-3. II 9 Dunca n AA, Chandy J. Ca se report: multiple neonatal
94 Ng CS , Hall CM, Shaw DG. The ran ge of v isce ral fractures - dietary or deliberate? Ciin Radiol 1993; 48:137-9.
manifestations of non-accidental injury. Arch Dis Child 1997 ;
77:167-74.
95 Barnes PM, Norton CM , Dunstan FD , et al. Abdomina l injury
Further Reading
due to chi ld abuse. Lancet 2005; 366:234-5.
96 Touloukian RJ. Abdomina l visceral injuries in battered Kleinman PK (ed.) Diagnostic Imaging of Child Abuse, 2nd edn.
I CHAPTER 4 I
HAEMATOlOG ICAL ABNORMALITIES
THAT CAN SIMULATE ABUSE
Angela Thomas
INTRODUCTION
abuse does not rule out abuse. 7 When the child has bruising
in a recognizable pattern, such as a belt or hand, then sus
Medical, nursing and other personnel who care fo r children pected abuse must be repo rted regardless of the results of
have a responsibility to be aware of signs and symptoms laboratory tests. B•9 In addition, these children may be at
that are suggestive of child abuse, including non-accidental greater risk of serious injUly secondary to bleeding/,ID as
inju ry (NAI) . .Equally, however, they must also recognize well as greater risk of bleeding secondary to an abusive
that medical and physical conditions may simu late abuse injury.l1 The physi cian who evaluates the bruised child for a
and that appro priate measures must be taken to confirm or disorder of coagulation sh ould always assess the history
rule out these conditions. Cutaneous lesions are by far the and physical examination to determine whether the bruises
most common presenting manifestations of child abuse, I,2 were acqui red spontaneously, accidentally or as the result of
and of the conditions that may simulate ab use haematol abuse. The pattern of bruising or associated findings may
ogical abnormalities are manifest usually as bruising or other yield important information. Knight and Bennett J2 describe
bleeding into the skin or mucosal mem branes. Over-reporting a case of abuse in a child with osteogenesis imperfecta,
of natural disease as NAI will occur, especially if a full which was diagnosed from the pattern of the injuries; in this
assessment of the child is not carried out. An incorrect diag case a periorbital haematoma in association with a spiral
nosis of ch ild abuse, a lthough at times unavoidable, can be fracture of the humerus.
devastating for the family a nd child 3,4 and, exceptionally, Unusual diseases may mimic abuse - misinterpreta tion of
has led to parental suicide .s Another consequence of such ' usual' by the inexperienced, for example a Mongolian blue
misdiagnosis is that a condition such as haemophilia will go spot (see Chapter 8, Fig. 8.7, p. 152), can a lso lead to a mis
unrecognized and therefore untreated. This in itself may diagnosis. 6 Advice from a paediatric haematologist, paediatric
lead to morbidity and even mortality. If the child is revealed dermatologist, paediatrician or haematologist may be crucial
to have a genetic or serious blood diso rd er, regaining the in reaching the right diagnosis. Caution must be exercised
trust of that patient and their family may be impossible. The when fo llowing guidance (frequently given) that abuse
mistaken diagnosis of child abuse has been described as a shou ld be suspected if there is significant bruising or bleed
form of medical abuse. 6 However, it is important to remem ing with no history of trauma or a history inconsistent with
ber that diagnosis of a medical condition that can simulate the severity of the injury.13-16 In a ch ild with a bleeding
- - - -
Introduction I 77
diathesis, the severity of the bruising may be disproportion number as they begin to crawl and walk. How ever, children
ate to the injury sustained or there may be a denial of any with severe bleeding disorders did develop bruising when
significant injury by the parent or caregiver ; a paediatrician non-mobile and had larger bruises than those with mild
who is not familiar with haematological disorders may mis bleeding disorders, who in turn had larger bruises than
takenly interpret this as evidence of abuse. 17 those wit h no bleeding disorder. These observations indi
Clinical identification of the age of a bruise compared cate that children with severe bleeding disorders develop
with the age given from the history has been used to assess bruises before they are mobile and bruise more frequently
the credibility of the history an d to identify whom the child than a control popul ation. 26 Although the finding of
was with at the time the bruise occurred. However, there is bruises of different ages coupled with an inconsistent his
an increasing amount of evidence to suggest that the age of tory has been cited as a hallmark of child abuse,9 not only
a bruise cannot be estimated reliably.18 A bruise is caused is it difficult to age a bruise accurately, but also it is impor
by blood that has escaped from damaged or leaky blood tant to note that a child w ith a bleeding diathesis may pres
vessels, usually capillaries, into the interstitia l tiss ues, but ent, quite normally, with bruising of differen t ages plus an
the appearance of the bruise depends on several factors. inconsistent history.
The amount of blood that escapes depend s upon the Haematological investigation of a brui sed child is
integrity of the coagulation system, the force of the injury mandatory in all cases when the bruising is une xplained or
and the integrity of the vessels. The location of the bruise implausible, and in cases where some explanatio n is given
also determines its characteristics; for instance, periorbital or found but the bleeding that results is disproportionate to
and genital bruises will appear sooner than bruises on the the injUly sustained. In a child who may poss ibl y have
extremities because the tissues are loose and the vessels been ab used, it is essential that investigations are as atrau
poorly supported. 19 The clinical ageing of bruises is most matic as possible and yield the maximum information. In
frequ ently based upon colour of the bruise when compared the case of a bruised or bleeding child under these circum
with an estab lished chart. 2o However, the colour of a bruise stances, whether or not a diagnosis of NAl is enteliained, a
depends not only on age, but also on amount of blood set of screening investigations should be performed on
present, location beneath skin, skin colour and amb ient blood taken from a single ve nepuncture, with the labora
light. 21 Stephenson and Bialas 22 conducted a study on chil tory set up and alerted to analyse the tests with the min
dren with bruises of known ages; photographs of the bruise imum of delay, using sma ll plasma samples. Further
were taken at different time intervals and aged by a 'blind' invasive tests, such as bleeding time, should usually be
observer. They concluded that several different colours avoid ed at this stage. Initial screening and investigation is
could be present at the same time within anyon e bruise aimed at the diagnosis of the commoner causes of bleeding,
and that bruises change colour at very different rates, even and exclusion or confirmation of some of the rarer causes
when sustained at the same time in the same child. Ageing for the safety and management of the child. Further inves
of bruises from photographs is much less precise than tigations might be needed if no explana tion of the bleeding
many textbooks imply.21 ,22 However, assessing the age of a is found or no admission of NAJ is made.
bruise on clinical examination may be easier, as other clues In order to be able to interpret the results of coagul ation
such as tissue swelling or abrasion might be presentY This testing, the fundamentals of primary and secondary
was not the finding in a study by Munang and col haemostasis must be understood. Artefact can significantly
leagues, 2J in which marked variability in colour descript ion distort resul ts and lead to misdiagnosis; utmost care must
between observers of the same bruises, and thus estimation be exercised in the way the specimens are taken and han
of the age of the bruise, was demonstrated; less than one dled prior to processing.
third of descriptions between two observers tallied either in Haemostasis depends upon the integrity of the tissues and
vivo or from clinical photographs. Wide variability in vasculature, the n umber and function of the platelets (pri
bruise development and healing urges caution in ageing mary haemostasisl and the formation of fibrin (secondary
bruises and therefore this should never be used as the sole haemostasis). The first of these is outwith the scope of this
criterion for child abuse;21 the pattern of distribution of the chapter and will not be dealt with in any detail, exceptions
bruising is a key factor that must be linked to the child 's to this being Henoch-Schbnlein purpura, which is not
history and stage of development. Bruises found in atypi uncommon in children and may be mistaken for NAl, and
cal areas in toddlers, such as the trunk, hands or buttocks, vita min C deficiency, which will be discussed briefly. A basic
are of concern, as are bruises in normal infants under 9 scheme of the haemostatic mechanism is shown in Figure 4.1.
months who are not yet mobile. 24 Sibert and colleagues l8 ,25 When a vessel is injured, the subendothelial collagen is
have develop ed a scoring system to evaluate these factors exposed, resulting in platelet adhesion, activation and
and such tools may be useful for discriminating between aggregation forming the primary haemostatic plug. Vascular
abused and non-abused children.18,25 In a prospective injury also leads to vasoconstriction and slowing of blood
analysis of bruising in children with a nd without an inher flow, aiding this process. At the same time, tissue factor is
ited bleeding disorder it has been shown that non-mobile expressed on subendothelial fibroblasts and monocytes ini
babies without a bleeding disorder do not bruise, but once tiating the coagulation process, by which the inactive
they start to roll they develop a few bruises that increase in plasma coagulation factors are converted to their active
78 I Haematological abnormalities that can simulate abuse
Vasoconstriction ....- -
1
Platelet
1
Blood
factor (VWF), a large, adhesive pl asma protein. Conditions
where either of these receptors is absent, for example Glan
acti vat ion coagulation zmann's thrombasthenia (Gpllb-lIIa) or Bernard-Soulier syn
~ Th"~bi"
drome (GpIb-lX), or where there is a lack of VWF, are
1
Reduced
Plait
characterized by abnormal bleeding of variable severity and
will be discussed later in the chapter.
blood flow pug
1
Clot
1
Fibrin
Platelet Activation
Figure 4.1 Haemostasis after injury. Activation of the Platelet activation rap idly foll ows adhesion and results in a
coagulation system and platelets is necessary for formation of a change in shape of the platelets, generation of active medi
stable fibrin plug. ators an d secretion of the g ranul ar conten ts. The activation
process is hi g hly specific owing to the presence of platelet
counterparts by cleavage of peptide bonds. Thrombin is recepto rs that recognize appropriate ago nists, the most
generated, converting fibrino gen to fibrin , which then sta important of these being thrombin and ad enosine diphos
bilizes the pl atelet plug. Activation offactor XIII by thrombin phate (ADP). Thrombin binds to the platelet at several si tes,
allows cross-linking of the fibrin strands and further stabil including the Gplb-IX receptor as well as a sp ecific throm
ization. Thrombin also activates the protein C pathway, bin receptor, which is the most importa nt for activation.
which is inhibitory to coagulation and profibrinolysis, The thrombin recepto r has been identified as a large seven
allowing dissolution of the fibrin clot as healing progresses. domai n transmembrane protein that binds proteolytically
active thrombin, which is essential for the induction of
platelet activation. This results in a change in shape of the
PRIMARY HAEMOSTASIS platelets, platelet aggregation and secretio n of granule
contents. A second agon ist, ADP, is much weaker than
Platelets must be present in ad equate numbers and function thrombin in vitro but is important in vivo as both red blood
normally in order for the primary haemostatic plug to form cells and vascular tissues release ADP in response to
effectively. Platelets are small, disc- like cells that remain in damage and lead to ADP-induced platelet activation and
the circulation for 7-10 days. In their quiescent state, they aggregation.
ex press predominantly neutral phospholipids on their outer
membrane, which are haemostatically inert. However, when
activated, negatively charged phospholipids, predominantly Platelet Aggregation
phosphatidyl serine, are exposed, providing the membrane
surface required for the coagulation reactions. On the cell Once platelets are activated, they aggregate, linking to each
surface there are also specific protein receptors for platelet other to form a haemosta tic plug. The binding of fibrino
agonists and adhesive glycoproteins, which are essential for gen to the GpUb-lIla complex is essential for bridging
effective haemostasis. Storage gran ules are also contained between acUacent platelets and is crucial to platelet plug
within the platelet cytoplasm, ex granules and dense (6) formation. Fibrin is also bound by this receptor, helping to
granules, whose contents are released by fusion of their stabilize the platelet plug.
membrane with that of the platelet.
Platelet Secretion
Platelet Adhesion
When platelets are activated , in addition to aggregation, a
The initial event in haemostasis is the attachment or ad hesion release reaction occurs in which the contents of the ex gran
of platelets to the vascula r subendothelium that is exposed ul es and 6 granules are secreted. The constituents so
after injury to the non-thrombogenic vascular endothelial lin released contribute to the process of primary and second
ing. Collagen receptors on the platelets, thought most likely to ary haem ostasis as well as wound repair and vascular
be the glycoprotein (Gp)la-IJa complex, bind to the exposed remodelling. The a granules release, amongst other things,
subendothelial coliagenY After adhesion, the platelets VWF (synthesi zed by the pl ate let-forming megakaryocytes)
undergo further interactions involving the GpIlli-llia receptor and fib rinogen (synthesized by the liver and taken up into
Secondary haemostasis I 79
the platelet). It is possible that these pro tei ns bind to the receptor that binds fibrino gen but also VWF, facilitating
platelet surface and participate in platelet adhesion and aggregation. VWF is non-covalently bound to factor VIIl
aggregate formation. This may be particularly important in and protects it from protein C proteolysis. Low levels ofVWF
platelet-rich plugs in which th ere is limited access to result in correspondingly low levels of factor VIII, which, if
plasma proteins. Factor V is also contained in the ex gran severe, can result in defects of secondalY haemostasis.
ules and is important for formation, along with factor Xa,
of the prothombinase complex on the ex posed phospho
lipid surface of the activated platel et. SECONDARY HAEMOSTASIS
x
TFPl
.~
.---'----,-_--,-.L---,--_,...~
1 II
_ ___,_vv_~-~a--,-r---. Xa - Va ~ i XI_Xla
Endothelial cell or
monocyte
TF
Vila
l.I Pl ate let
TG~\NF
"\
Va Villa
IX IXa
Activated
platelet
Figure 4.2 Tissue factor (TF) is exposed after injury and binds both factor VII and activated factor VII (Vila), which both activates factor
IX and genera tes a small amount of thrombin from prothrombin (II ) by activating factor X. The thrombin activates factors XI, Vand VIII by
cleavage from von Willebrand factor (VWF). Platelets are also activa ted and surface phosphol ipid (PL) is exposed. Tissue factor pathway
inhibitor (TFPI) binds the tissue factor-Vlla-Xa complex and inactivates it, switching off the ini tia ting pathway.
80 I Haematological abnormalities that can simulate abuse
Endothelial cell or
monocyte
Insoluble
Vila fibrin
~
IX
XIII • Xilia
Figure 4.4 Conversion of fibrinogen to fibrin by thrombin and
thrombin activation of factor XIII forming an insoluble clot by
cross- lin kage.
- . -
Laboratory tests I 81
XII_Xlla
XI_Xla
aPTI
1
IX-IXa VII_Vila
Vllla~~
PT
Com,," p",hw", X }-::
II - Throm bin - - -
Fibrinogen _
1 Fibrin
Figure 4.5 The classical pathway of co agulation: it is proposed that the intrinsic system is activated by cont act with an activating
surface and the extrinsic system by tissue factor released from damaged vessels or tissue. 80th systems activate factor X which via the
final common path way results in the formation of fibrin. The prothrombin time (PT) reflects the acti vity of the extrinsic and common
8ath w ays. The acti vated partia l thromboplastin time (aPTI) is most sensitive to changes in the intrinsic pathway. II, factor II (prothrombin);
Va, activated factor V; Vila, activated factor VII; Villa, activated factor VIII; IXa, activated factor IX; Xa, activated factor X; XI a, activated
factor XI; Xlla, activated factor XII.
LABORATORY TESTS (FOPs) and heparin. The clotting time and the appearance of
the clot are informative. A prolonged IT is seen in hypofibrino
A full blood count will yield important information regard genaemia, both congenital and acquired, as in, for example,
ing platelet number, and some au tomated counters will disseminated intravascular coagulation (DIC), in which there is
give the mean platelet volume. However, there will be no consumption and inhibition by FOPs and liver disease (raised
evaluation of function, an important parameter in primary FOPs) . A prolonged IT is also seen with hypoalbuminaemia.
haemostasis. Haemoglobin and white cell count may also Heparin causes extreme prolongation of the IT, but a 'reptilase'
indicate an underlying haematological disorder such as time, which utilizes a thrombin-like enzyme obtained from a
leukaemia or aplastic anaemia. Examination of the blood snake venom that is unaffected by heparin and relatively
film will reveal morphological abnormalities of platelets insensitive to FOPs, is normal. It is useful therefore in differen
such as size (Wiskott-AJdrich, Bern a rd-Souli er, May tiating contamination by heparin from fiblinogen deficiency
Hegglin, idiopathic thrombocytopenic purpura) or abnormal and also in indicating if a dysfibrinogenaemia is pre-sent when
granulation (Chediak-Higashi and Hermansky-Pudlak it is generally more prolonged than the thrombin time. In the
syndromes). The basic screening tests available for coagu latter case, the nature of the clot is often abnormal , being
lation are the PT, the aPIT (written also as PITK) and transparent and bulky as a result of abnormal fib rin polymer
fibrinogen measu rement and/or thrombin time (IT) . ization. This can be seen in liver disease.
These laboratory investigations are designed to test the Fibrinogen is measured in a variety of ways; some auto
integrity of both the extrinsic and intrinsic pathw ays of the mated machines derive the value from the prothrombin
classical scheme of coagulation (Fig. 4.5) and remain time, which is quick and simple. However, PT-derived
invaluabl e for the understanding of clot formation in vitro. fibrinogen levels appear to show a false elevation in a variety
The PT reflects the integrity of the extrinsic pathw ay (fac of clinical settings. Conversely, if the PT is very prolonged ,
t ors V1I and X), as well as factors II (prothrombin) and V of a fal sely low fibrinogen measurem ent can result. The
the common pathw ay. The aPIT tests the intrinsic pathway Clauss fibrinogen assay is a functional assay, based on the
and is prolonged by deficiencies of factors XII, XI , IX, V11I princip al that the thrombin clotting time is inversely
and X as well as prekaJJikrein and HMWKs. proportional to the fibrinogen concentration )) and is the
Thrombin time is a functional test of fibrinogen, by which most reliable me thod for general use in clinical laborator
mrombin is added to test plasma and the time to clotting is ies.)4 The use of diluted pl as ma and a relatively high con
measured. It is affected by the concentration of fibrinogen and centration of thrombin results in little interference by FOPs
abnormalities of fibrinogen (dysfibrinogenaemia). Inhibitors of or heparin, which are known to influence the technique. If
the reaction include fibrinogen- and fibrin-degradation products a PT-derived method is used, fibrinogen levels of <2.0 gil
82 I Haematological abn ormalitie s that ca n simulate abu se
or >6 .0 gil should be rechecked using the Clauss method. 35 platelets and on platelets after stimulation by agonists in
The go ld standard method is measurement of clot weight controlled conditions.48 Thus platelet activation and reactiv
but this is time-consuming and not practi ca l as a screening ity can be determined. The use of whole blood in the second
test. VelY lo w levels of fibrino gen «0,8 giL) wiiJ begin to method allows for the contribution of red and white blood
prolong both the PT and t he aPTT, especially when auto cells to the haemostatic process . This method measures
mated analysers are used, aggrega tion and the release reaction of pl atelets in whole
blood using the PFA-lOO system 49 ,50 Using this system,
whole, anticoagul ated blood is drawn under a constant vac
uum into a collagen/ADP-coated or collage n/adrena line
MEASUREMENTS OF PRIMARY HAEMOSTASIS coated membrane, which has a small aperture in it. A
platelet plug is formed and obstructs the fl ow through the
In paediatrics, defects of prima ry haemostasis are more com apertu re and both the max imum velocity of flow an d closure
mon than coagu]opathies. 36 Primary haemostasis is depend ent time are recorded. The test can be performed both on resting
upon both number and function of platelets. Other physiolog platelets and after stimul ation by agonists in controlled con
ical variables such as temperature,37 stress,38 anaemia,39 leu ditions. The PFA-lOO is sensitive in detecting classi ca l
copenia 40,41 and the integrity of the connective tissues and defects resulting in maj or platelet dysfunction, such as
vessels also influence primary haemostasis. Many of the va ri Glanzmann's thrombasthenia and Bernard-Soulier syn
ables that might cause abnonnal primary haemostasis such as drome and also von Willebran d's disease (VWD), although a
cardiovascular or renal disease, cardiopulmonary bypass sur full blood count and film will demonstrate a macrothrombo
gery or ingestion of aspirin-like drugs, will be obvious from cytopenia in Bernard-Soulier syndrome and abnormal val
the history and examination of the child, or from the results of ues of VWF antigen and/or activity will be seen in VWD, if
initial blood tests such as full blood count or plas ma urea specifically tested as is recommended in investigation of NAI
level. However, platelet function and platelet-vessel wall (see beloW). False-negative results occur with the PFA-lOO in
interaction are not eluci dated by these methods. milder platelet defects, such as storage pool disorder and
The bleeding time is an in vivo test and dependent upon release defects, which are not detectable by t he routine labor
both haemostatic and other physiological variables outlined ato ry tests 50 and it is not sensitive to vascular-col lagen dis
above. For instance, it is prolonged in patients with the con orders. Although it may be useful as a screening tool for a
nective tissue disease Ehlers-Danlos syndrome, generalized bleeding diathesis in children,51 its use in identifying those
vasculitis and scurvy (vitamin C deficiency), patients who who have a bleeding diathesis in cases of possible NAI has
have a low body temperature (as can occur in prolonged sur not been tested. Clinicians should always perform a full
gical procedures) and in those with leukaemia or uraemia. range of platelet function tests when clinical susp ic ion is
AJthough the bleeding time can be a useful screening test to strong and when exclusion of a platelet defect is essenti al.
di agnose heredita ry bleeding defects that involve platelet If pl atelet dysfunction is suspected from the pattern of
endothelium interaction, the test requires a highly motivated bleeding or needs to be excl uded, formal platelet function
and experienced operator who und erstan ds the many vari testing can be performed. This requires larger volumes of
ables influencing the result and can inform interpretation of blood and arrangements need to be made with the labora
the test. 36 The most widely used method is the Ivy bleeding tory to discuss the extent and range of tests. Screening tests
time 42 using a template. The test itself is invasive; small inci would include aggregation response to the agonists ADP,
sions of a standard width and depth are made in the forean}] adrenaline (epinephrine), ristocetin and collagen and
which can cause scarring and sign ificant distress to the 5-hydroxytryptamine release, indicating the secretory
patient, although there are specially designed templates for resp onse. If abnormal ities are fo und on screeni ng tests, fur
children minimizing th is problem. 43-46 The norma l ranges in ther testing could include t hromboxane genera tion, platelet
children of different ages vary from those in adults and older nucleotides, t10w cytometry or electron microscopy to
ch ildren and between bleeding devi ces. define t he disorder more precisely. There are, however,
Two further methods are available to screen for platelet problems with reproducibility and the overall haemostatic
dysfunction: both use small volumes of whole blood, from condition of the patient cannot be fully assessed by this
2 flL to 1.5 mL, a nd require special instrumentation. The first method as vessel-wall interaction is not taken into account.
method uses flow cyto metlY and utilizes the change in It ca n be difficult to determine the contributio n to bleeding
expression of platelet membrane proteins as an indicator of made by minor abnormalities of platel et fun ction testing.
platelet activation,47 Upon activation, P-selectin is translo
cated to the platelet surface membrane due to degranulation
of the a granules and there is increased expression of the EVALUATION OF A BLEEDING PATIENT
GpJlb-IIIa complex. Conversely, as platelets are activated,
the GpIb-IX complex is internalized and results in decreased When a child presents with bruising or bleeding, the main
expression. These changes in expression ca n be detected differential diagnoses are physiological or accidental
using different monoclonal antibodies both on resting bleeding. NAI or a bleeding diathesis, The sex of the
Evaluation of a bleeding patient I 83
Examination of the Child Figure 4.7 Fingertip bruising in a child with Glanzmann's
thrombasthenia.
The general health and state of the child should be assessed.
In a bruised child, particular points to look for are the distri
bution, size and age of the bruises; whether they are in any
recognizable pattern such as a hand or belt mark; and
whether there is a haematoma indicating the extent of bleed
ing. It should be noted that fingertip bruising is not infre
quently found in children with a bleeding diathesis and is not
pathognomonic ofNAl (Fig. 4.7). Distribution of bruising can
be the key to the diagnosis of Henoch-Schonlein purpura,
which is present on extensor surfaces and is due to a vasculi
tis rather than a coagulopathy. Coagulation screening and a
full blood count will be nonnal and thus confusion with NAl
may arise (Figs 4.8 and 4.9).16.59.60 Presence or absence of
petechial haemorrhages will help to differentiate disorders
associated with thrombocytopenia. Such haemorrhages also
occur in situations when there is raised intrathoracic pressure
in the absence of a bleeding diathesis, such as a severe bout Figure 4.8 Symmetrical bruising on extensor surfaces seen in
of coughing or vomiting or in strangulation. In these situ Henoch-Schbnlein purpura.
ations, the petechiae will be seen in the distribution of the
superior vena cava area or around the eyes or mouth 61 (Fig.
4.10). A swollen, tender joint may indicate bleeding into the enlarged lymph nodes, liver or spleen usually accompany
joint, as is seen in haemophilia or, much more rarely, VWD, bruising and petechial haemorrhages secondary to acute
but tender joints may also be seen in Henoch-Schonlein pur leukaemia . A limp, with or without bruising, can also be seen
pura, acute leukaemia or neuroblastoma. Neuroblastoma may at presentation in this disease. A rare disease, haemophago
present with bilateral black eyes as a result of tumour infiltra cytic Iymphohistiocytosis (HlH) presents with systemic signs
tion; comparatively minor injury can cause the same in and symptoms, such as pancytopenia, coagulopathy and
haemophilia (Figs 4.11 and 4.12). Systemic upset and organomegaly, but may also present with central nervous
Evaluation of a bleeding patient I 85
Fig ure 4.12 Black eye and mild proptosis in a child with
neuroblastoma. (Courtesy of the late Dr J. Pritchard, Edinburgh.)
FIRST-LINE INVESTIGATIONS
system (eNS) manifestations. These can range from irritabil
ity to encephalopathy and coma. Retinal and intracranial Once history-taking and examination have been done, inves
haemorrhages can be found in these circumstances and this tigations can proceed. The way blood is taken and processed is
condition has been mistaken for abuse. 62 Full examination of paramount importance, as activation of the coagulation
and laboratory evaluation, however, will indicate the correct system by a difficult venepuncture and contamination with
diagnosis. Disseminated intravascular coagulation with tissue fluids or by the presence of air bubbles will invalidate
secondary bleeding is generally seen in an ill child with signs the result. Shortening of the PT or aPIT through activation of
and symptoms of infection. a specimen may mask a significant bleeding disorder such as
- -- - - - - -
86 I Haematological abnormalities that can simulate abuse
haemophilia. Conversely, if blood is taken through a cannula above, which includes VWF antigen and activity, and
that has been kept patent with heparin then contamination levels of factors VIII and IX , will add further information
frequently results, giving abnormal values for many of the a nd minimize del ay. The results of the tests will inform the
coagulation tests. Venepuncture of children should be done by choice of fu11her tests to determine the precise diag nosis.
an experienced operator wherever possible, at a time when the
blood can be rapidly dealt with by the laboratOly. Blood
should be taken with the minimum of venous stasis and sub Isolated and Prolonged aPTI (PT Normal,
sequent handling. Spinning, freezing and thawing blood can TI Normal, Fibrinogen Normal, Platelets
cause al1efact; for example, factor XI levels may rise.6) Trans Normal)
port to a laboratory some distance away, even if the specimen
is transported on ice, can also cause artefact, usually a lower Prolongation of the aPIT is probably the commonest abnor
ing of clotting protein values. Some common pitfalls in speci mality found on pelforming the coagulation screen. This may
men collection and processing are shown in Table 4. I. Initial be by only a second or two but can signal a significant bleed
tests include a coagulation screen, PT, aPIT, IT and fibrinogen ing diathesis. Referring to the classical scheme of coagulation
as outlined above and a full blood count and film. A fa ctor (Fig. 4.5, p. 81), which is a good model for in vitro coagul a
VlU, factor IX level and VWF antigen and activity are recom tion, this indicates a defect in the intrinsic pathway.
mended in all cases of suspected NAI, as a normal or margin This pattern is fo und in:
ally prolonged aPIT can be associated with a significant 1. con genital deficiency of factors VIII, IX, XI and XII,
decrease in factor VIII or IX levels.64 Factor XIll may also be as well as prekallikrein and HMWK ;
measured in neonates with intracranial haemonhage. If the 2. VWD, as this may result in low levels of factor VIII;
blood is flowing well , a few millilitres extra can save a second 3. circulating inhibitors, e.g. lupus anticoagulant;
venepuncture if an abnormality is found, for example a pro 4. mild deficiencies of factors II (prothrombin) , V and X
longed aPTT, and further testing is required. Bleeding time is may prolong the aPIT, whereas the PT rema ins norm al
an invasive test and although it will demonstrate the integrity (reagent dep endent);
of the platelet-vessel wall interaction, is usually unnecessary 5. heparin, either as a contaminant or a therapeutic
in the early stages of investigation. It is highly operator agent, is a common cause of a prolonged aPIT; the IT
dependent. One of the newer tests (flow cytometry or the PFA is very sensitive to heparin and will be prolonged; a
100) may be helpful in the future but there is no published reptilase time will be normal.
work in this clinical setting and false-negative results may be
misleading. Thus the prolongation of the aPIT is due either to a
It is sometimes helpful to investigate parents, especially clotting factor deficiency or an inhibitor of coagulation. To
if results are equivocal or subsequent testing for clarifica differentiate between the two, a 50:50 mix of test plasma
tion of a ny abnormality requires large volumes of blood. with normal plasma can be performed . If there is a deficiency
Identification of a child's natural parents is not always cer there should be a correction of the clotting time to within a
tain, or they may not be available or readily located, mak few seconds of normal and at least less tha n 50 per cent of
ing this approach impossible. the difference between the control and test plasmas. If there
is no significant correction then the presence of an inhibitor
is suspected. This might be heparin contamination or therapy,
PATIERNS OF ABNORMAL RESULTS an acquired inhibitor of a clotting factor (velY rare, especially
in children) or the lupus anticoagulant (Fig. 4.13).
The pattern of abnormalities obtained using first-line tests Therapeutic heparin administration or heparin contamin
along with the clinical presentation may indica te an und er ation of the specimen can be ruled out from the history, by
lying defect/disorder. The extended testing discussed checking that tlle specimen was not taken from a heparinized
Patterns of abnormal results I 87
. t . i t i
50:50 plasma mix '+ DIC
Ehler's-Danlos
LA' '+ ? heparin disease
1 cx,antiplasmin
t t t
v\AlD 1 XI ? LA'
1 VIII reagent
(I DRVVT)
llx dependent:
hw 111, V, X
t
HM\AlK' ? factor VIII, IX or
PK' XI inhibitor
(DRVVT N)
Figure 4.13 Algorithm for initial investigation and interpretation of results in a child with a possible bleeding diathesis.
II, factor II; V, factor V; VII, factor VII; VIII, factor VIII; IX, factor IX; XI, factor XI; XII, factor XII; XIII, factor XIII; V+VIII, combined factor Vand VIII deficiency;
HMWK, high molecular weight kinonogen; PK, prekallikreth; cong, congenital; LA, lupus anticoagulant; HSP, Henoch- Schbnlein purpura; SPD, storage pool
disorder; DIC, disseminated intravascular coagu lation; vWD, von Willebrand Disease; PAI-l, plaminogen activator inhibitor-l ; H thrombin time; Rep,
Reptilase; DRWT, dilute Russell viper venom time.
'Not associated with bleeding
Inhibitor
Lupus anticoagulant t t t Correction N N
Heparin t t t No correction N N
Factor VIII and IX antibody t t N n/a N N
Deficiency
Factors VIII, IX, XI and XII t Correction N n/a N N
VWF t or N Correction N n/a N N
Factor II sl t or N Correction t No correction N t·
Factors V, X t Correction t No correction N t <
aPH partial thromboplastin time; DRWT, dilute Russell's viper venom time; n/a, not applicable; N, normal; PT, prothrombin time; 51, slightly; VWF, von
Willebrand Factor.
Note: Specific patterns of bleeding may be seen with specific diagnoses (Table 4.3).
line or put into a bottle containing lithium heparin before as the dilute Russell's viper venom time (DRVVT) can dif
being transferred to a citra ted one. It is also possible that ferentiate this from other factor inhibitors. Russell's viper
heparin may have been given inadvertently or with intent to venom causes clotting by direct activation of factor X. This
cause bleeding.55 If there is uncertainty, a reptilase time can is inhibited by the lupus anticoagulant, but is not affected
be performed, which, by activating fibrinogen directly, is nor by deficiencies or inhibitors of factor VIII or factor IX. A
mal in the presence of heparin. If the pattern of correction is platelet correction test can then be performed, whereby the
one of an inhibitor, and heparin has been excluded, the most DRVVT is repeated but using platelet-associated phospho
likely cause is a lupus anticoagulant (Table 4.2). lipid rather than thromboplastin. Platelet-associated phos
Lupus anticoagulant is a phospholipid antibody that pholipid is protected from the action of the antibody and
interferes with the phospholipid added to the in uitra test the test revelis to normal with a lupus anticoagulant. Defi
ing model. It affects the aPTT rather than the PT, as the ciencies of factors X, V and II will also prolong the DRVVT
phospholipid in the former test is more dilute and usually but these will usually give a prolonged PT and will not cor
more sensitive than that used in the PI. FUliher tests, such rect with platelet-associated phospholipid (Table 4.2).
88 I Haematological abnormalities that can simulate abuse
If the pattern of correction is one of a deficiency FVIII can result in serious bleeding, or the lupus anticoagulant,
and FIX should be assay ed if not already done. If these fac which is likely to be an incidental finding and unconnected
tors are normal then FXI and XII should be assayed. If the with the bleeding and bnlising. The extra blood taken at
only defect is a deficiency of FXII then this is not associ ini tial venepuncture ca n be used to carry out further test
ated with a bleeding diathesis and will not account for any in g such as the DRVVT with platelet correction as described
haemorrhagic symptoms or signs. 66 Prekallikrein a nd above. If the pattern is not that of a lupus anticoagulant, a
HMWK deficiency, which are rare, will also prolong the specific inhibitor such as to factor VIII or very rarely factor
aPTI but, again, are not associated with a bleeding diath XI should be considered. In these circumstances, when the
esis. Von Willebrand factor antigen and activity shou ld also plasma is serially diluted, the clotting times shorten , rather
be checked. Factor VIII is non-covale ntly bound to VWF, than lengthen , as the inhibitor is diluted out. The aeti ology
which protects it from proteolysis in the plasma. Deficien of factor VIII inhibitors in children is uncertain but can be
cies of VWF may therefore be associated with reduced associated with malig nancy or possibly recent use of high
levels of factor VIIl, such that the aPTT is prolonged; however, dose penicilli n 68 .G9 Factor XI inhibitors are very rare and
this is not necessarily seen (Table 4.2). have been described foll owing viral infections, resolving
The pattern of bleeding that is identified on clinical within a few weeks 70 ,71 There are only a handful of
examination may predict the most likely cause of the reported cases of acquired coagulation factor inhibitors in
abnormal test result (Table 4.3). Mucocutaneous bleeding children.
and/or ecchymoses are a common presentation of VWD.
Factor Xl deficiency also presents with this pattern of
bleed ing but is very rare excep t within certain population Isolated and Prolonged PT (aPTT Normal. TT
groups such as Ashkenazi Jews, for whom the gene fre Normal. Fibrinogen Normal. Platelets Normal)
quency is around 8 per cent 67 Factor VIII and factor IX
deficiencies do not necessarily present with a history of Referring again to the classical scheme of coagulation (Fig.
painful or sw oll en joints and, although spontaneous muco 4.5, p. 81), this indicates a defect in the extrinsic pathway,
cutaneous bleeding is rare, bruising is common. As well as that is factor VII.
ecchymoses with or without joint swelling and bleed ing, This pattern is found in:
factor VIII or factor IX deficiencies can present with inter 1. factor VII deficiency, either congen ital or secondary to
mit te nt but persistent bleeding from mucocuta neous injury, liver disease or vitamin K deficiency;
such as a bitten tongue or lip . 2. early warfarin therapy;
Mucocutaneous bleed ing and/or ecchymoses in a child 3. mild deficienci es of factors 11 (prothrombin), V and X
whose aPTI does not correct, may be secondary to an may prolong the PT, wheras the aPTI remains within
acquired inhibitor, such as a factor VIII inhibitor, which the normal range, depending on the reagents used;
Table 4.3 Patterns of coagulation results and patterns of bleeding: possible diagnoses
APTI, partial thromboplastin time; Ole, disseminated intravascular coagulation; PT, prothrombin time; VWD, von Willebrand's disease;
4. lupus anticoagulant (rarely); it can be more sensitive to classical scheme (Fig. 4.5, p. 81). Inhibitors can also pro
reagents used in the PT than those used in the aPTT; a duce this pattern; in paliicular, over-heparinization can
mixing test will not correct. lead to a markedly long aPTT, but with only a minor pro
longation of the PI. Mixing experiments using the PT may
Both warfarin and vitamin K deficiency can cause this be useful if there is no history 0 f anticoagulant use or obvi
picture. Although they affect the carboxylation of all the ous cause of vitamin K deficiency, for example parenteral
vitamin K-dependent coagulation factors (factors ll, VII, IX nutrition or prolonged antibiotic use. Inhibi tors of coagu
and X), factor VII has the shortest half-life and is the most lation factors may be specific, for example to factor V or
sensitive to failure of that step. However, with increasing factor II (prothrombin). or may affect more than one factor.
dosage of warfarin or severity of vitamin K deficiency, the However, they all have the characteristic that as the plasma
aPTT wiH become prolonged, due to the involvement of is serially diluted, the clotting times shorten rather than
other factors (Figure 4.13). lengthen as would be seen in a simple deficiency.
Mucocutaneous bleeding and ecchymoses in association Mucocutaneous bleeding and ecchymoses are seen in
with a prolonged PT suggest warfarin ingestion, early liver severe vitamin K deficiency, over-warfarinization and
dysfunction or early vitamin K defiCiency (Tabl e 4.3)' over-heparinization. Severe liver dysfunction is associated
although none is necessarily associated with bleeding. Vita with a reduced fibrinogen level. Acquired inhibitors of
min K deficiency is a recognized hazard in the neonate, coagulation can cause very severe bruising, muscle bleed
especially in those who are excl usively breast fed, and is ing, and gastrointestinal or urogenital bleeding, but are
corrected by parenteral vitami n K at birth or repeated oral very rare. Factor X, factor V and prothrombin deficiencies,
vitamin K at birth and the early neonatal period. Malab which are also rare, can all present with mucocutaneous
sorption syndromes are also associated with vitamin K bleeding and bruising of varying severity. Factor X defi
deficiency. More severe deficiency ho wever, results in a ciency can also present with haemalihroses, as can factor
prolongation of both the PT and aPTT (see beloW). A trial V and prothrombin deficiency (less commonly). Specific
dose of oral or intravenous vitamin K can be given, which factor assays will need to be performed to differentiate
shou ld begin to correct abnormalities within 4-12 hours if between them.
deficiency is the cause. This pattern of coagul ation tests can be seen in associ
Prothrombin defici ency is rare but usually presents with ation with the lupus anticoagulant and acquired prothrom
mild mucocutaneous bleeding or with post-traumatic bin deficiency. However, sponta neous bleeding is rare and
haemorrhage. The aPTT may be slightly prolonged along mixing tests are indicative of lupus anticoagul ant.
with the prolonged PT, but fibrin ogen is normal. Dyspro
thrombinaemia has also been described with a similar pre
sentation and prolonged PT.72 Prolonged PT and apn, Long n with Low
Ecchymoses and bleeding into joints is seen with con Fibrinogen (Platelets Normal)
genital factor VII deficiency, and thus a factor VII assay
can be performed once warfarin ingestion and vitamin K This pattern is found with:
deficiency have been reasonably excluded and liver func 1. high level s of heparin (with lower levels, TT is long but
tion checked. A trial of vitamin K can be given as above. fibrinogen is normal);
2. hypo- and afibrinogenaemia; dysfibrinogenaemia;
3. systemic hyperfibrinolysis;
Prolonged PT and apn (n Normal, Fibrinogen 4. some cases of liver disease.
Normal, Platelets Normal)
When the fibrinogen level falls below 0.8 giL (measured
This pattern is found in: by a functional assay), as well as a prolonged TT, the PT, and
1. vitamin K deficiency, when the PT is usually more to a lesser extent the aPTT, also become prolonged. This
prolonged tha n the aPTT; occurs with hypo- and afibrinogenaemi a and with the
2. oral anticoagulants, when, again, the PT is usually majority of dysfibrinogenaemias. The reptilase time, which is
more prolonged than the aPTT; based on clotting induced by release of the fibrinopeptide A
3. liver disease, giving rise to multiple factor deficiencies; fragmen t from fibrinogen, is also prolonged, especially with
fibrinogen may be abnormal in severe disease ; the dysfibrinogena emias. The reptilase time is not affected
4. deficiencies of factors II, V, X or combined V and VIII; by heparin and so is useful in determining whether there is
5. lupus anticoagu lant, with acquired prothrombin contamination by this anticoagulant or not. The platelet
deficiency. count is normal. Bleeding in those with afibrinogenaemia
and fibrinogen levels of less than abo ut 0.5 giL is lifelong
For both of these screening tests to be prolonged either and varies in severity between patients. Ecchymoses, muco
there are multiple defects in the coagulation pathway or cutaneous haemorrhage and haemarthroses have all been
there is a deficiency in the final common pathway of the described. Patients with dysfibrinogenaemi a are usu ally
90 I Haematological abnormalities that can simu late abuse
asymptomatic with bleeding (when it do es occur) often PAl-l and Ctrantipl asmin assays or platelet function studies
limited to epistaxis, menorrhagia and mild-to-moderate post may be necessa ry. Abnormalities of collagen vascular dis
traumatic bleeding, including surgely. eases and vascular integrity such as Henoch-Schonl ein pur
pura will also have normal investigations (see Chapter 4,
p.94).
Prolonged PT and aPIT, Long IT with Low
Fibrinogen and Platelets
Von Willebrand Disease
This pattern is found in:
1. disseminated intravascu lar coagulation ; Von Willebrand disease is a bleeding dis order secondary to
2. so me cases of severe liver disease. either a quantitative or qualitative abnormality of VVI/F.
Diagnosis hin ges on demonstrating these defects. Von
Children with DIe secondary for instance to infection , Willebrand factor is a multimeric, high-molecular-weight
such as meningococcal septicaemia, are usually clinically glycoprotein that is synthesized both by endoth elial cells
sick, with evidence of in fection in addition to ecchymoses an d megakaryocytes, and plays an impol1ant paI1 in pri
and mucocutaneous haemorrhage. Th e coagulation scree n mary hae mostasis. It is essential for platelet adhesion to the
will be deranged bu t the platelets will be low. Sequential vessel wall, which, having been secreted from endothel ial
measurements of coagulation and full blood count may cells, it initiates by bindi ng to the GpIb-IX receptor. This
show a continuing fall in both platelet count a nd fibrino ex poses the GpIlb-llla receptor complex to wh ich it binds
gen concentration. Evidence of increased fibrin degrad a lon g with fibrinogen, facilitating platelet aggregation ,
ation will support the diagnosis a nd meas urem ent of the es pecially in high-shear conditions. It acts as a carrier pro
D-dimer level is probably th e most reliable. D-Dimers a re tein for circulating FVI llc to wh ich it is non-covalently
formed followin g the di gestion of cross-linked fibrin by bound, protecting it from proteolysis by protein C and thu s
plasmi n and are raised in DIe. Fibrinogen degradation its rapid clearance from the plasma. Decreased levels of
produ cts (FDPs) can also be measured but are not specific VWF or reduced binding of V\I\fF to factor VIlle are accom
to cross-linked fibrin and can be raised in other conditions. panied by correspondingly low levels of circulating factor
Severe liver dis eas e may also give a pattern of prolonged VIlIc. It can be seen therefore that both qu antitative and
PT, aPTT with a low fibrinogen and lo w pl atelets secondary qualitative defe cts of VWF may affect both primary
to hype rsplenism, a lthough these can be normal. Clinical haemostasis (resulting in mucocutaneous bleeding) and
exa mination will di fferentiate the problem as well as add secondary hae mostasis (resulting in haema to ma formation
itional tests of liver fun ction . a nd , more rarely, hae marthroses). Several types and sub
types of VWD have been described; however, the id entifi
cation of the latter requires specialized testing. Di agnosis
Prolonged PT and aPIT, Platelets Low (IT of the specific subtype, although importan t for correct
I\lormal, Fibrinogen Normal) treatment, is not necessary for the purposes of identifying
if a chUd has a propensity to bl eed. The importan ce of
This pattern is fo und in: screening for VWD lies in the fa ct that, excluding type 3
1. mass ive transfu s ion ; VWD, it is the commonest inherited bleeding disorder, with
2. some cases of chronic liver disease, for exampl e a prevalence of between 0.1 per cent and 2 per cent 73 75
cirrhosis. and that the coagulation screen and full blood count can be
normal. Levels of VWF vary with environmental factors,
The clinical findings and liver function tests will clarify such as stress and exercise. To minimize the risk of misdiag
the cause. nosis, VWF antigen and function must be measured in
samples obtained on at least two occasions with consistent
resu lts. 76.77 However, from a pragmatic point of view, ini
NORMAL COAGULATION SCREEN WITH tial tests will guide the clinician in early management but
A NORMAL PLATELET COUNT should be repeated at a later stage w here possible.
Type 1 V\I\fD is the commonest form of the disease (80
Several conditions need to be considered ifthere is Significant per cent of cases) an d is cha racteri zed by a quantitative
bleeding and screenin g investigations are nom1al. The com defect. It is inherited as an autosomal dominant but with
monest is VvVD but factor XIII deficiency, platelet storage marked variability of both phenotypic penetrance an d
pool disorder and Glanzmann's thrombasthenia can produce exp ressivity. This may be because th ere are VWD modifier
these results, although very rarely. Disorders of fibrinolysis genes unassociated with the VWF locus as well as the VWF
such as cxra ntiplasm in deficiency and plasminogen activator gene S7 .78 A bleeding diathesis secon dary to V\I\fD can usu ally
inhibitor-l (PAl- 1) defiCiency may also predispose to bleed be id entified from evidence of a fa mily history of bleeding,
ing. In these rare instances, clot solubility assays, factor XIII, a clinical histOIY and la boratory demonstratio n of VWF
Normal coagulation screen with a normal platelet count I 91
deficiency. However, without documentation of all three of VWF:Ag is decreased in type 1 VWO and decreased or nor
these features, a diagnosis of type 1 VWD can be more dif mal in type 2. More than 80 per cent of all patients with
ficult. Type 2 VWD is inherited as an autosomal dominant VWD will have an abnormal v'vVF:Ag result BO but, if done
condition when there is a qualitative defect in the VWF. For alone, this will miss a proportion of patients with normal
both type I and type 2 diseases, the clinical picture is one antige n levels but abnorm al function, i.e. type 2 VWD, In
of mucocutan eous bleeding, most commonly epistaxis, addition, in some cases the levels will be borderline and
gum bleeding and bleeding from superficial cuts and may require repeat testing. 77 In those patients for whom
wounds. Petechi ae are rarely a feature, with none reported the diagnosis of VWD is strongly suspected or needs to be
in a study of 1257 patients 79 ,Bo but in 11.5 per cent of excluded, results of several different tests shou ld be
patie nts reported in a study by Silwer. BI The haemorrhagic analysed to y ield the maximum information from which to
tendency is velY vatiab le and depends On the type and draw conclus ions. Further tests include VVllF function,
severity of the disease. In many patients with type 1 or 2 measured using a platelet-based ristocetin cofactor (RiCo F)
disease there may be no history of bleeding and caution and VWF collagen binding assays. These assays improve
should be exercised when attributing bruising in a child the abi lity to detect type 2 variants and more clearly define
with suspected NAI automatically to VWD on the basis of type 1 VWD. Other assays including agarose gel elec
laboratory testing.? VWD may induce increased bleeding trophoresis of the VWF multimer pattern, further platelet
wi th trauma, but not precipitate spontaneous bleeding. agglutination studies and VWF-factor VIllc binding assays
Diagnosis of a bleeding diathesis does not exclude NAl. are required to subclassify VWD but are not always neces
and the histolY and clinical findings remain important salY to establish the basic diagnosis of VW0 77 (Table 4.5).
contributors to determining the cause of bleeding7 An important point to note is that VWF levels vary accord
The results of the screening tests usually reveal a normal ing to blood gro up and also in crease as part of the acute
platelet count, although mild thrombocytopenia may be phas e response. The mean plasma VWF:Ag level is about
found in subtype 2B or platelet-type-pseudo VWD (see 30 per cent lower for those with blood group 0 than those
beloW). The PT is normal, whereas the partial thromboplas with blood group A, and 39 per cent lower for those of
tin time may be mildly prolonged and dependent upon the blood group AB.B2 Whether the ABO group has any effect
plasma level of factor VIllc (Table 4.5). In mild type 1 dis on the specific activity ofVWF is unresolved. Variation due
ease, t he bleeding time may be normal or prolonged and to the acute phase response means that those with VWD
thus may not be helpful in the diagnosis and should be may have intermittently normal levels and , therefore, a
avoided,n However, an in vitro bleeding time using the singl e normal level does not exclude the diagnosis. There is
platelet function analyser PFA-lOO shows good sensitivity no evidence that venepuncture alone results in raised lev
to the defect in primary haemostasis that occurs in VWD els of VWF, but it is very likely that venepuncture per
and may be helpful if the device is available. 50 ,5 1 Tests used formed under stressful circumstances can result in an
for the primary diagnosis of VWD include a factor VIlIc increase in factor VIIIc and VWF. 77 Unfortunately, children
level and VWF antigen (VWF:Ag). Factor VIllc half-life is are often stressed when blood is taken and, if this occurs,
regulated by VWF and is frequently reduced in VWD, results shou ld be interpreted with this in mind and the fact
although factor VJlIc levels are not necessarily reduced. 77 recorded in the notes, As mentioned above, VWF:Ag and
Normal Abnormal
ITP, idiopathic th rombocytopenic purpura: PAI-l, plasminogen activator inhibitor-l ; VWD, von Wi llebrand's disease.
92 I Haematological abnormalities that can simulate abuse
function must be measured in samples obtained on at least secret ion mechanism. They are often associated with rela
two occasions. One subtype, type 2B, is characterized by an tively mild bleeding states and diagnosis may be delayed
increased affinity of the abnormal VWF for the platelet. until later in childhood, or even in adult life. Common pre
This often results in a mildly reduced platelet count with sentations include easy bruising and possibly prolonged
large forms seen on the blood film 83 (Table 4.5). The platelet bleeding from superficial cuts, epistaxis and menorrhagia.
counterpart to this can be seen in platelet-type-pseudo Screening tests are normal, therefore specific platelet func
VWD, in which the platelet Gplb-IX complex has an tion testing, including platelet aggregation and secretion, are
increased affinity for normal VWF and produces the same required to confirm the diagnosis. Again, however, the diag
clinical and laboratory picture B4 Differentiation between nosis of platelet storage pool disorder does not exclude NAI,
the two can be made by mixing experiments. and the history and clinical findings are very important.
A variant of VWD, type 2N, can also present in this way Specific syndromes such as Grey platelet, Hermansky
with a prolonged aPTT on testing. Factor VIII is unable to Pudlak, Chediak-Higashi and Wiskott-AIdrich syndromes,
bind to the abnormal VWF and factor VIII levels are reduced which are associated with other defects, and specific blood
(Table 4.5). The condition can be confused with mild film appearances in addition to the st orage pool disorder are
haemophilia A, particularly as VWF levels and activity can described under separate headings.
be normal. However, the differentiation between mild Deficiencies can occur in either the 0. granules (Grey
haemophilia and VWD type 2N is not necessary for the platelet syndrome) or the 6 granules or both. Abnormalities of
immediate management of the child . It is inherited as an platelet aggregation may be detected but can be normal in
autosomal dominant and can be confirmed at a later stage milder cases. The definitive diagnosis depends on the demon
once family history and investigation has taken place. A fur stration of the deficiency either biochemically (using tests of
ther variant, type 2M, has a type 2 pattern of response, i.e. platelet secretion), by electron microscopy or both.
normal VvVF:Ag and reduced activity but, unlike the other Failure of the secretion mechanism occurs in aspirin
type 2 variants, it has a normal multimer pattern (Table 4.5). ingestion and cyclooxygenase deficiency. Defective aggre
Type 3 VWD has an autosomal recessive inheritance gation and secretion is seen to specific agonists. Most
and is very rare. It results in a severe bleeding disorder with patients with congenital cyclooxygenase deficiency do not
markedly decreased or undetectable VWF/Ag and activity have an accompanying bleeding diathesis, perhaps because
and has factor VIII levels in the region of 0.01-0.05 IU/mL. the defect is balanced by impairment of prostaglandin gen
Acquired von Willebrand syndrome has been described, eration in the vascular endothelial celIs.87
although very rare, and has similar laboratory findings to Defects have been shown in platelet testing t o weak
congenital disease. 8 5 The severity of the bleeding tendency agonists only and described as weak agonist response
varies from mild to severe and has been found in associ defect (WARD) . The clinical significance of this is very
ation with several disease states including systemic lupus uncertain especially since defective responses to weak agon
erythematosus 86 and congenital cardiac defects.8o ists such as epinephrine are found in normal platelets and
results should be interpreted with caution.
Platelet Storage Pool Disorders
Factor XIII Deficiency
These are a collection of disorders that are characterized by
failure of secretion of the contents of the platelet granules on Factor XIII is responsible for the stabilization of the fibrin
stimulation. This may be due to either absence of the gran clot by polymerization 88 and also inhibition of fibrino
ule contents (storage pool deficiency) or to failure of the lysis by the binding of <:X2-plasmin inhibitor to the fibrin. 89
Normal coagulation screen with a normal platel et count I 93
Antiplasmin Deficiency
Glanzmann's Thrombasthenia
Congenital
MAY-HEGGLIN ANOMALY
OTHER MACROCYTOPATHIES
MYELODYSPLASIA
COAGULATION DEFECTS
Haemophilia A and B
THE NEONATE
These include deficiencies of fibrinogen (factor 1), prothrombin Thrombocytopenia in the Neonate
(factor 11), factor V, factors V + vm, factor VIl and factor X.
They occur with a frequency of 1 in 500000 to 1 in This will lead to purpura and, in particular, a petechial rash.
2 million. In congenital fibrinogen deficiency, levels of Differentiation from NAl can be made from examination of
the blood count and film. The causes will include the rare haemorrhage is uncommon. Vitamin K prophylaxis wil l
inherited thrombocytopenias described above, immune not have been given and the infant is often almost exclu
causes such as neonatal alloimmune thrombocytopenia sively breast-fed. Factor VJI has the shortest half-life of the
(NAlT) or maternal ITP and non-immune causes. Non vitamin K-dependent coagulation proteins and therefore a
immune thrombocytopenia is seen in congenital intra coagulation screen will initially show a prolongation of the
uterine infection such as that caused by cytomegalovirus PT. As factors II, IX and X then fall, so the aPTT prolongs.
(CMV) or the human immune deficiency virus (HN), and in A deficient state is demonstrated by correction of the coagula
neonates who are sick from various causes, including sep tion times after a 50:50 mix of test with normal plasma.
sis and neonatal asphyxia, congenital aplastic processes as These screening tests are not specific and further investiga
described above or infiltration of the marrow by leukaemia, tion of individual factors and the carboxylation state of
for example. History and examination are important in prothrombin will confirm the diagnosis. A therapeutic trial
determining the precise cause. of vitamin K will shorten the PT significantly in this condi
tion and may be the most practical approach.
Neonatal Alloimmune Thrombocytopenia Late HDN occurs after the first week of life and is most
common between weeks 2 and 8 after birth; however, there
have been reports of it occurring as late as 15 weeks after
Although NAIT classically presents at or soon after birth it
bilih.1JO A significant proportion of neonates (50 per cent)
should not be mistaken for NAl. The classical presentation is
presents with intracranial haemorrhage and its associated
one of bleeding, which includes intracranial haemorrhage in
morbidity and mortality. A history of exclusive breast
an otherwise well infant. Neonatal alloimmune thrombo
feeding and either lack of vitamin K at birth or a single oral
cytopenia occurs as a result of transplacental passage of
dose, is often elucidated. However, it is also seen with other
maternal IgG antibody directed against the fetal platelets
underlying conditions that res ul t in malabsorption of vita
owing to expression of a paternal antigen that the mother
min K, such as antibiotic therapy. Laboratory investigation
does not possess. The commonest cause of this is antibody
reveals the same results as for classical HDN.
directed against the HPA-Ia antigen expressed on the platelets
of about 98 per cent of the population. This is significant
because the antibody, when bound, blocks the GpIfb-lIIa
DRUGS ASSOCIATED WITH BLEEDING
receptor on the platelet and leads to a platelet function defect
as well as moderate to severe thrombocytopenia. Although
Warfarin
skin and mucocutaneous bleeding are frequent, the most ser
ious complication is that of intracranial haemorrhage, which
Ingestion of warfarin, whether accidentally or for therapeutic
occurs in up to 20 per cent of neonates with this condition;
purposes, will lead to a haemorrhagic di athesis, the extent
around one-half of these bleeds occur in utero. The degree of
depending upon the amount taken. Bruising is the common
thrombocytopenia is most marked in the first day of li fe and
est manifestation but mucocutaneous bleeding, including
platelet numbers gradually rise over the next 2-4 weeks as
haematuria and gastrointestinal bleeding, can also occur. If
the level of the antibody declines in the infant's circulation. It
investigated, a coagulation screen will show a prolonged PT
is essential that the condition is correctly identified so that
with normal or slightly prolonged aPTT and normal platelets.
appropriate treatment may be given to the child and advice
When a large amount of warfarin has been ingested, the aPTT
given to the mother for future pregnancies.
will be prolonged, but not to the same extent as the PT. Such
an overdose may be accidental, but it must be remembered
Vitamin K Deficiency that a child with fabricated or induced illness (FII) may pre
sent with bleeding due to warfarin poisoningYI.132
Vitamin K is essential for the carboxylation of coagulation
factors II, VJI, IX and X. Without this step, these factors are
inert and incapable of activation in the coagulation Heparin
process. Vitamin K deficiency misdiagnosed as child abuse
has been described in several reports, for example Wheeler Heparin must be given parenterally to exert its antithrombotic
and Hobbs 117; Wetzel et al 127 in the neonatal period (haem effect, which is mediated through antithrombin. Owing to
orrhagic disease of the newborn, HDN) and by Carpentieri anti-factor Xa and antithrombin effects, the capacity to gen
et al 128 and Kaplan 129 later in life. Presentation of HDN can erate thrombin is both decreased and delayed. Heparin can
be in the immediate neonatal period and is usually second cause excessive bruiSing but also more serious bleeding,
ary to maternal ingestion of vitamin K antagonists such as including CNS and retroperitoneal bleeding. Heparin can also
warfarin or anticonvulsants. Classical HDN presents at induce thrombocytopenia. If standard heparin has been given,
between 2 and 5 days after birth, with purpura and gas a coagulation screen will show both prolonged aPTI and TI.
trointestinal haemorrhage in an otherwise well infant. The reptilase time will be normal. Low-molecular-weight
Mucocutaneous bleeding can occur, but intracranial heparin is most senstively detected using an anti-Xa assay.
Bone marrow failure syndromes I 99
Many drugs have been associated with platelet dysfunction Inherited Bone Marrow Failure Syndromes
in vitro but either do not have a clinically signifi cant effect
in vivo or are used under strict medical supervision;1 33 such It is outwith the remit of this chapter to discuss bone mar
examples include high-dose penicillin and heparin. row failure syndromes in detail, but Freedman and Doyle 143
have reviewed this topic. When there is significant thro mbo
cytopenia, presentation with bruising and bleeding is pos
Sodium Valproate sible. Many of these conditi ons have accompanying
physical abnorm alities as well as haematological ones. For
The anticonvulsant sodium valproate may also cause example, Fanconi's (aplastic) anaemia is often accompa n
thrombocytopenia and minor coagulation abnormalities ied by short stature, pigmentary skin changes and abnor
due to an effect on platelet function. Richardson 134 malities of the upper limbs. 144 Othe r conditions in this
describes a study in which platelet functio n was shown to category includ e Shwach man-Diamond syndrome, dysker
be abnormal in 6 out of 23 patients on sodium valproate atosis congenita and thrombocytopenia with absent ra dii
but who had normal bleeding times. A previous report had (TAR) . Amegakaryocytic thrombocytopenia presents in
shown a prolonged bleeding time in four out of five infancy with isolated thrombocytopenia due to red uced or
patients, one of whom was thrombocytopenic. A later absent marrow megakaryocytes. An initial diagnosis of ITP
prospective study showed that one-third of children stud may be made (see below)' but with failure of resoluti on of
ied had a fall in platelet count, with the lowest reaching the thrombocytopenia within 3 weeks a bone marrow asp ir
35 X 10 9 /1. 135 Subsequent studies have confirmed that ate is recommended 145 and the diagnosis made. Overall , 45
sod ium valp roate-associated thrombocytopenia occurs in per cent of patients will go on to develop ap lasti c anaemia.
12 per cent to 18 per cent of patients , with th e platelet
counts correlating inversely with age of the patient and
dosage of the dru g. 136 - 138 Whether the mechanism is Aplastic Anaemia
immune or due to impaired platelet production has not
been clearly e]ucidated . 139 ,\40 There are a few other drugs In this condition, a child may present with pallor, overt infec
that have clinically relevant antithrombotic and haemor tion or bleeding problems typical of a primary haemostatic
rhagic effects. defec t. Past history is usually unrem arkab le and physical
examination normal. A fuJ I blood count will show a combina
NON-STEROIDAL ANTI-INFLAMMATORY DRUGS tion of anaemia, leucopenia and thrombocytopen ia. The
Aspirin is rarely given to children because of the risk of diagnosis can be confirmed on bone marrow aspirate and
Reye's syndrome. There are certain specific indications such biopsy.
as Kawasaki 's disease, for whic h there is a risk of thrombosis
and aspirin is given specifically for its ant ipl atelet effects. 141
Leukaemia
Accidental aspirin ingestion or poisoning occurs. Inges tion
of aspirin is the commonest cause of platelet dysfunction in
Leukaemia can present primarily with bleeding problems,
the population as a whole. Aggregation is impaired due to
but the clinician is usually ale11ed to the diagnosis as the
deficien t thromboxa ne A2 formation by aspirin's inhibition
child is often unwell and examination may reveal
of cyclooxygenase. The defect lasts for the lifespan of the
hepatosplenomegaly and lymphadenopathy. A full blood
platelet and causes a mild bleeding tendency. It can con
count may show a pancytopenia or a naemia a nd thrombo
tribute significantly to the bleeding seen in those with coag
cytopenia with a normal or high wh ite cell count. Gener
ulation defects such as haemophilia; maternal asp irin within
ally, blast cells are seen on the blood film and diagnosis
10 days of delivery may contribute to neonatal bleeding.14 2
ca n be confirmed by bone marrow examin ation. This con
O'Hare and Eden 7 describe a case of child abuse in which
dition is rarely misdiagnosed as NAI, although McClain
aspirin resulted in abnormal platelet function tests an d led to
et a!. 146 describe a case of acute lymphoblastic leukaem ia
severe bleeding from a minor injury. Other anti-inflamma
diagnosed at autopsy when abuse was suspected.
tory drugs such as ibuprofen that might be used in children
are rarely associated w ith bleeding problems.
Myelodysplasia
ANTI-PLATELET DRUGS
Some drugs a re g ive n specifically for their antithrombotic Acqu ired storage pool defects can acco mpany the thrombo
effects but a re rarely used in ch ildren. They include prosta cyto penia of myelodysplasia and result in significant
cyclin and prostacyclin analogues, ticlopidine and dipyri skin purpura and mucocutaneous bleeding (Fig 4.22).133
damole. Diagnosis is genera lly straightforward, with a full blood
- - -- - - -- -
- - -- - -- -----
--
100 I Haematological abnormalities that can simulate abuse
(al
count showing pancytopenia commonly with a red cell
macrocytosis and teardrop poikilocytes; neutrophils may
show Pelger-Hue! forms. Bone marrow cellularity is normal
or increased, often with increased reticulin, and one or more
cell lines may be dysmorphic. An iron stain to detect ring
sideroblasts and chromosomal analysis is also useful. Early
in the disease process, however, mild thrombocytopenia
alone may be present, with a disproportionate bleeding ten
dency due to the functional defect. Other investigations such
as fetal haemoglobin (HbF) level and plasma lactate dehy
drogenase measurement may indicate the correct diagnosis.
Vitamin B12 or folate deficiency must be excluded.
Renal Failure
Figure 4.22 Bruising with no history of trauma in a child with Liver Failure
mild thrombocytopenia and platelet dysfunction due to a
congenital myelodysplasia. The liver is the main site of synthesis for most haemostatic
proteins. An exception to this is factor V1II, which has
Conclusion I 101
signi ficant extrahepa tic synthesis . Liver failure results in evidence of microangiopathy on the blood film (red
impaired synthesis of coagulation proteins and reduced cell fragments and reduced platelets) and increased
clearance of activated haemostatic factors. Thrombo levels of FDPs and/or D-dim ers . A chronic localized form of
cytopenia can occur as well as imp aired platelet function. 49 DIC is seen with giant haem angiomas; this is known as the
Owing to the reduction of both procoagulant and anticoag Kasabach-Merritt syndrome.
ulant proteins and activation of both the coagulation
and fibrinolytic systems, thro mbosis, as well as bleeding,
may occur. Factor VII has the shortest half-life of the coag
ulation proteins and therefore disappears from the
Head Injury
circulation first. The result is an initial prolongation of the
PT, foll owed by prolongation of the aPTI, and, lastly, Head injury in children can be associated with coagulatio n
reduced fibrinogen as liver func tion declines. In the abno rmalities secondary to DIc.152.153 In adults, a coagu
absence of DIC, factor VIII is normal owing to its extrahep lopathy has been associated with delayed brain injury sec
atic synthesis. Bleeding manifestations include bruising ondary to bleeding. 154 ,155 The study by Hymel et al 151 of
and petechiae, mucosal haemorrhage and gas trointestinal 265 children with head injury showed that those with
haemorrhage. parenchymal brain damage were more likely to have a
coagulopathy manifested by prolongation of the PT than
those without. Of the children who had died from abusive
Malabsorption Syndromes head injury, which had caused parenchymal brain damage,
94 per cent had prolongations of their PT and 63 per cent
Malabsorption of fat-soluble vitamins, including vitamin K, had evidence of activated coagulation. Hymel and col
can occur in cystic fibrosis ,128 biliary atresia and obstruc leagues concluded that it was highly unlikely that the
tive jaundice. Broad-spectrum antibio tics may alter the coagulation abnormalities were due to an underlying
normal intestinal flora and result in malabsorption of haemorrhagic diathesis. It is important to recognize th is as
vitamin K. Bleeding manifestations are those of vitamin K a secondary phenomenon , not a primary one, so that inves
deficiency. tigation into possible abuse is not terminated on spurious
gro unds.
ACTIVATION OF COAGULATION
Cyanotic Congenital Heart Disease
Disseminated Intravascular Coagulation
Abnormalities of coagulation and platelets can occur in
Disseminated intravascular coagulation is due to patho cyanotic congenital heart disease. 156 ,157 The exact mech
logical activation of the coagulation system and can be anis m producing the coagulopathy is not known. Hypofib
precipitated by a variety of disease processes. Both endothelial rinogenaemia secondary to poor liver function and reduced
injury and release of tissue factor can trigger the activation, synthesis and clearance of clotting factor intermediates can
resulting in both circulating thrombin and plasmin. Throm lead to low-grade disseminated intravascular dissemin
bin generates fibrin fro m fib rinogen, which is deposi ted in atio n. Thrombocytopeni a may result from both DIC and
small vessels and causes microvascular thrombosis. Con shortened platelet survival.
sumption of coagulation factors occurs during this process,
leading to the classical changes in coagulation parameters
of a prolonged PT, prolonged aPTI, decreased fibrin ogen CONCLUSION
and reduced platelets. A bleeding tendency results.
Increased circulating plasmin digests both fibrinogen and In a child presenting with haemorrhagic symptoms, a his
fibrin, resulting in raised FDPs and D-dimers (see p. 91). tory that includes details of ethnic origin, consangu inity
These interfere with both fibrin polymerization and platelet and family history, drug history and clinical examination is
function, thus adding to the bleeding tendency. In children, essential. Initial screening tests should include aPT, aPTI,
DIC is usually acute and occurs in an ill child; it can be sec TI and fibrinogen, a factor VIII and factor IX level and a
ondary to infection, trau ma (such as crush injuries or burns) VWF antigen CVWF:Ag) and activity (RiCoF). If the history,
and malignancy (such as leukaemia and li ver disease). A examination and investigations do not account for the
more chronic, compensated form can occur, with the fol bleeding then further investigation might include other
lowing features : (1) less marked derangement of the coagu specific assays, such as factor XIII or platelet function test
lation parameters when the platelet count may be only ing. Advice on ful1her testing or interp retation of findings
moderately reduced; (2) plasma fibrinogen is often normal should be sought from a haematologist who has experience
or slightly elevated; and (3) the PT and aPTI may be within in this field, especiall y if there is a disparity between the
normal limits. In such patients, the diagnosis is made on clinical and laborato ry findings. If the case is likely to
- - - - -
- ~~
102 I Haematological abnormalities that can simulate abuse
receive a formal legal challenge then even the rarest of 25 Dunstan FD, Guildea ZE, Kontos K et a\. A sco ring system for
causes may require exclusion. Diagnosis of a bleeding bruise patterns: a tool for identi fying abuse. A rch Dis Child
2002; 86:330-333.
diathesis, especially if associated with a mild phenotype,
26 Hamilton M, Jenkin s B, Dunstan F et a\. Prospective analysis
does not exclude NAI and when these are found concur of bruising in children with and without an inherited bleeding
rently the child will be at even greater risk. di so rder. Blood 2006; 108:304a-305 a.
27 Kunicki TJ, Nugent OJ, Staats SJ et a \. The human fibroblast
cl ass \[ extracellul ar ma trix receptor mediates platelet
adhesion to collagen and is identical to the platelet
REFERENCES glycoprotein la-lia compl ex . J BioI Chern 1988;
263 :4516- 4519.
Stewart GM, Rosenberg NM. Co nditions mistaken for child 28 Roberts HR, Monroe OM, Oliver JA et a\. New er concepts of
abuse: part II. Pa ediatri c Emergen cy Care 1996; 12:2 17-221. blood co agulation. Ha ell10ph ilia 1998; 4:331-334.
2 Gallenoh H, Oppenheim W. Th e battered child syndrome 29 Bazjar L, Morser J, Neisheim M. TAFI, or plasma
revisited. Clin OrtllOpaed Related Res 1982; 162:11-19. procarb oxpeptidas e B, couples the coagulation and
3 Wexler R. Wounded il1lIocents: the Real Victims of tile War fibrinolytic casca des through the thrombin-thrombomodulin
Against Child Abuse. Buffalo, NY: Pro metheus Books, 1990. complex. J Bioi Chem 1996 ; 271: 16603-1 6608.
4 Vorenberg E. Di ag nosin g child abuse; th e cost of getting it 30 MacFarlane RG. An enzyme cascade in the bl oo d clotting
wrong. Arch Dermatol 1992 ; 128:844-845. mechanism and its function as a biochemical amplifier.
5 Anh NT. 'Pseudo battered child' syndrome. JAMA 1976; Nature 1964; 202 :498-499.
236:2288. 31 Davie EW, Ratnoff 00. Waterfall sequ en ce for intrinsic blood
6 Kirschner RH, Stein RJ. Th e mistaken diagnosis of child
clotting. Science 1964: 145:1310-2.
abuse: a form of medical ahuse? Am J Dis Child 1985 ;
32 Peny OJ. Factor VlI deficiency. Br J Haematol 2002;
139:873-87 5.
118 :689-700.
7 O'Hare AE, Eden OB. Bleeding disorders and non-accidental 33 Clauss A. Rapid physiological coagulation methods in
injury. Arch Dis Ch ild 1984; 59:860-864. determination of fibrinogen. Acta Haemato/ (Basel) 19 57;
8 Harley JR. Disorders of coagulation misdiagnosed as 17 :237 -246.
nona cc id enta l b!1lising. Pedialr Emag Care 1997; 13 :347 -349. 34 Mackie IJ, Kitchen S, Machin SJ, Low e GDO. Guidelines on
9 Maguire S, Mann MK, Sibert J, Kem p A. Are there patterns of fibrinogen assays. Br J Haemato/2003; 121 :396-404.
b!1lising in childhood which are diagnostic or suggestive of 35 Medical Dev ices Agency. MDA Evaluation Report : fibrinogen
abuse? A systematic review. Archs Dis Child 2005; 90:182 -186. assay reagents and methods. MDA 2000/01 ISBN184J82
10 Johnson CF, Coury DL. Bruising a nd hemophilia: accid ent or
139X. Norwich: Her Majesty's Stationery Office.
child abuse? Ch ild Abuse and Neglect 1988 ; 12:409-415.
36 Sutor A. The bleeding time in pediatrics. Semin Thromb
11 Smith. S.M, Hanson. R. 134 batte red children: a med ical and
Hemostasis 199 8 ; 24:531-543.
psychological study. British IVIedical Journal 1974 ; 3:666-670. 37 Valeri CR, MacGregor H, Cassidy G, Tinney R, Pompei F.
12 Knight DJ, Bennet Gc. Non-accidental injury in osteogenesis Effec ts of temperature on bleeding time and clotting time in
imperfecta: case report. J Pediatr Orthop 1990 ; 10:542-544. normal male and female volunteers. Crit Care Med 1995;
13 Altieri MF. Child abuse. When to be suspicious and what to 23 :698-704.
do the n. Postgrad Med 1990; 87:153-162. 38 van den Burg PJ, Hospers J .E, van Vliet Met a\. Changes in
14 Ladebauche P. Childhood trauma. When to suspect abuse. Reg haemostatic factors and activation produ c ts after exercise in
Nul'S J 1997; 60:38-42. healthy subjects w ith different ages. Thromb Ha emostas is
15 Roberts R. When you suspect abuse: what next? The 1995; 74:1457-1464,
Practitioner 1997; 241:322-326. 39 Lowe GDO. Blood viscosity and cardiovascular disease.
16 Daly KC, Si egel RM. Henoch-Schonlein purpura in a child a t Th romb Haemostasis 1992; 67 :494-498.
risk o f abuse. Arch Ped iatr Ado/esc IVIed 1998; 152:96-98. 40 Jakob E. Sutor A. Interaction of platelet aggregation and
17 Lapo sata ME, Laposata M. Children with signs of abuse: when gra nulocyte function (Abstract). Ann Haemato/ 1991; 62:78.
is it not child abuse? Am J c/in ParI1O/2005; 123(Suppl. I): 41 sohngen D, Hattsteill E. Heye ll A et aI. Hematological
S 119- I 24. parameters in fluencing th e Thrombostat 4000. Semin Thromb
18 Sibert J. Bruising, coagulation disorder and physical child Hemostasis 1995; 2J(Suppl. 2):20-24.
abuse. Blood Coagulation Fibrinolysis 2004; 15(Supp\. I): 42 Ivy AC , Shapiro F, Melnick P. Th e bleeding tendency in
S33-39. jaundice. Surg Gynaeco/ Obstet 193 5; 60:781 - 784.
19 Johnson CF. Inflicted injury versus accidental injury. Pediatr 43 Lethagen 5, Kling S. New bleeding time devices with
c/in North Am 1990; 37:791-814. retractable blades evaluated in children, hea lthy volunteers
20 Wilson EF. Estim a tion of the age of cutaneous contusions in and patients with prolonged bleeding time. Thromb
chi1.d abuse. Pediatrics 1977; 60:750-75 2. Haemostasis 1993; 70 :595-597.
21 Schwartz AJ, Ricci R. How accu rately can bruis es be aged in 44 Fe usner JH. No rmal and a bnormal bleed in g times in neo nates
abused children? Lite rature re v ie w and synthesis. Pediatrics and yo ung children utilizing a full standardized template
1996; 97:254-257. technic. Am J Clin Pa t11011980 ; 74:73-77.
22 Stephenson T, Bialas Y. Estimation of the age of bruising. 45 Andrew M, Castle V, Mitchell L, Paes B. Modified bleeding
Arch Dis Child 1996; 74:53-55. time in the infant. Am J Hemato/1989; 30:190-191.
23 Munan g LA, Leonard PA , Mok JY. Lack of agre e me nt on 46 Andrew M, Paes B, Bowke r J , Vegh P. Evaluation of an
colour description between clinicians examining childhood automated bleeding time device in the newborn.
bruising. J Clin Forensic Med 2002; 9:171-174. Am J Hematol 1990; 35:275-277.
24 Sugar NF, Taylor JA. Feldm a n KW. BIllis es in infants and 47 Michelson AD. Flow cytometlY: a clinical test of platelet
toddlers: those who don 't c ruise ra rel y bruise. Puget Sound function (Rev iew). Blood 1996; 87 :4925-4936.
Pediatric Resea rch Network. Arch Pediatr Adolesc Med 1999; 48 Michelson AD. Platel et function in the newborn. Semin
153 :399-403. Thromb Hemostasis 1998; 24:507-512.
References I 103
49 Kni:il1er R, Weissbach G, Kuhlisch E. Platelet function tests in 72 Shapiro SS, Martin ez J, Holburn RR. Congenital
childhood. Measuring aggregation and release reaction in dysprothrombinaemia: an inherited stru ctural disorder of
whole blood. Semin Thromb Hemostasis 1998; 24:513-521. human prothrombin.] Ciin In vest 1969; 48:2251-2259.
50 Ha rrison P, Robinson M, Leisner R et al. The PFA-IOO: a potentia l 73 Miller CH, Lenzi R, Breen e. Prevalence of von Willcbrand's
rapid screening tool for the assessment of platelet dysfi.lnction. disease among US adults. Blood 1987; 69:454-459.
Clinical Labomtmy Ha ematology 2002: 24:225-232. 74 Werner EJ, Broxson EH, Tucker n, Manaiac LF, Giroux DS,
51 Harriso n P. The role of PFA-IOO® testing in th e in vestigation Abshire Te. Prevalence of vo n Willebrand disease in children:
and management of hae mo stati c defects in children and a multiethnic study. Blood 1991; 78:68a.
adults. Br] Haematol2005; 130:3-10. 75 Castaman G, Fredericic AB, Rodeghiero F, Mannucci PM.
52 Andrew M, Vegh P, J ohnston M et al. Maturation of the Von Willebrand's Disease in tb e year 2003: towards complete
h em atopoie tic system during childho od . Blood 1992; 80: identification of gene defects for th e correct diagnos is and
1998-2005. treatment. Haematologica 2003; 88 :94-108.
53 Lane DA, Scully MF, Thomas DP et al. Acquired 76 Abildgaard CF, Su zuki Z, Harrison J et al. Serial studies in
dysfibrinogenae mia in acute and chronic li ver disease. von Willebrand disease: variability versus 'variants '. Blo od
Br] Haemato/ 1977; 35 :301-308. 1980; 56:712-716.
54 Weigart Ai, Schafer AJ. Uremic bleeding: pathogenesis and 77 Laffan M, Brown SA, Collins PW et al. The dia gnosiS of
th era py. Am] !VIed Sci 1998 ; 316 :94- 104. von Willebrand disease: a guideline from the UK Hae mophilia
55 Mannucci PM, Gi angrande PLF. Acquired disorders of Centre Doctors' Organisation. Haemophilia 2004; 10: 199-217.
coagulation. In Bl oo m AL, Forbes CD, Thomas DP, 78 Castaman G, Eikenboom JCJ, Bertina RM, Rodeghiero F.
Tuddenh am EG D (eds) Haemostasis and thrombosis. London: In co nsistency of associa tion between type I von Willebrand
Churchill Livings ton e, 1994, vol. 2, pp. 249-268. di sease phenotype and genotype in families id ent ified in an
56 Giannelli F, Green PM. The molecular basis o f haemophilia A epidemiological investigation. Tinomb Ha emostas is 1999;
and B. Bail/ieres Clin HaematoI1996; 9:21/-228. 82: 1065-1070.
57 Casaiia P, Martinez F, Haya S, Espinos C, Aznar JA. 79 Frederici AB, Mannucci PM. Actual management of von
Significant linkage and non-linkage of type I von Willebrand Wil!ebrand disease: first report on 880 cases of the Ita lian
disease to th e von Willebrand fac to r gene. Br 1 Haematol regis tlY of vWD. Blood 1997; 90:32a.
2001; 115:692- 700. 80 Frederi ci AB. DiagnosiS of vo n Willebrand disease. Semill
58 Bolton-Maggs PHB, Young Wa- Yin, McC raw AH et al. Thromb Hemostasis 1998; 24:654-660.
Inh e ritan ce and bl eedin g in factor XI deficiency. 81 Silwer J. Von Willebrand 's disease in Sweden. Acta
Br] Ha ematol 1988; 69 :521-528. Pa ediatrica Scandanovia 1973 ; 238:1-159.
59 Brown J, Melinkovich P. Henoch-Schonlein purpura 82 Gill JC, Endres-Broo ks J, Bauer PJet al. The effec ts of ABO
mi sd iagnosed as suspected child abuse: a case report a nd blood groups on the diagnosis of vo n Willebrand disease.
literature review. ]AtVIA 1986; 256:617-618. Blood 1987; 69:1691-1695.
60 Menter A. Symmetrical purpura doesn't mirror ch ild abuse. 83 Ruggeri ZM, Zimmerman TS. Variant von Willebrand 's
lAMA 1987; 257:486. disease. Characteri zation of two subtypes by analysis of
61 Perrot LJ. Masque ecc hymotique. Specific or nonspecific multimeric composi tion of FVllljvWF in pl asma and platelets.
indicator for abuse. Am] Forens ic iVIed 1989; Pathology ] Ciil1 fl1/Iest 1980; 65: 1318-1325.
10:95-7. 84 Mill er JL, Ca stella A. Pl atelet-typ e von Willebrand disea se:
62 Rooms L, Fitzgerald N, McClain KL. Hemophagocytic characterization of a new bleeding disorder. Blood 1982;
lymphohistiocytosis masquerading as child abuse ; 60:790-794.
presentation of three cases and revi ew of cen tral nervous 85 Rinder MR, Richard RE, Rinder HM. Acquired von Willeb rand's
system findin gs in hemophagocytic Iy mphohistiocy tosi s. disease: a concise review. Am] Hematol1997; 54:139-145.
Pediatrics 2003 ; 111:636-640. 86 Sim one N, Co rnet JA. Abildgaard CF. Acquired von
63 Pearson RW, Triplett DA. Factor XI assay results in the Willebrand ' syndrome in systemic lupu s erythe ma tosus. Blood
CAP su rvey (1981). Am] Clil1 Pathol 198 2; 78(Suppl. 1964; 31:806-812.
4) :615-620. 87 Pareti Fl, Mannucci PM, D'Angelo A et aL Congenital
64 UK NEQAS for blood coagulation Survey !l8, September 1999. deficiency of thromboxane and prostacyclin. La ncet 1980;
65 Martin CM, Engstrom PF, Barrett O. Surreptitious self 1:898-901.
administration of heparin. ]AMA 1970; 212:475-476. 88 Duckert F. Jung E, Shmerling DH. A hitherto undescribed
66 Blanchette VS, Dean J, Lillicrap D. Rare co ngen ital co ngen ital hemorrhagic diathesis probably due to fibrin
hemorrbagic disorders. In Lilleyman J, Hann I, Blan chette V stabilizing fact or deficien cy. Throl17b Dia thesis Ha em orrl1agic
(eds) Pediatric Hematology, London: Churchill Livingstone , 1960 ; 5:179-186.
1999,611-628. 89 Sakata Y, Aoki N. Si gnitl ca nc e of cross-linking of a lpha
67 Seligsohn U. Facto r XI deficien cy. Tinomb I-Iaemostasis 1993; 2-pJasmin inhibitor to tlbrin in inhibition of fibrinolysis and
70: 68-71. in hemostasis.] C/in Il1vest 1982; 69:536-542.
68 Green D, Lechn er K. A survey of 215 non-hemophilic pa tients 90 Duckert F. Documentation of the pl as ma fa ctor Xlli
with inhibitors to fa cto r VlII. Thromb Ha el110stasis 1981; deficiency in man. Anl1 New York Acad Sci 1972;
45 :200- 203. 20 2:190-199.
69 Nakashima K, Miyahara T, Fuj ii S et al. Spontaneously 91 Lorand L, Losowsky MS, MiJoszewski KJM. Human factor
acquired fa cto r VllJ inhibitor in a 7-ye ar- old girl. Acta XIII: fibrin s tabilizin g factor. Progress Hemostas is Tl1romb
Haematol 1982; 68:58-62. 1980; 5:245-290.
70 Reece EA, Clyne lP, Romero R, Hobbins J.e. Spontaneous 92 Saito H. Alpha 2-plasm in inhibitor a nd its deficiency states.
factor XI inhibitors: seven additional cases and a review of 1 Lab Ciin Med 1988; 112:671-678.
the lite rature. Arch In t Med 1984; 144 :525-9. 93 Lind B, Thorse n S. A novel missense mutation in the
71 Beck DW, Strauss RG , Kisker CT, Henriksen RA. An intrinsic human plasma inhibitor (alpharantiplasmin) gene
coagulation pathway inhibitor in a three year old child. associated with a bleeding tendency. Br] Haematol 1999;
Am] Ciin Patho11979; 71:470-472. 107:317-322.
104 I Haematological abnormalities that can simulate abuse
94 Schleef RR, Higgins DL, Pillemer E, Levitt LJ. Bleeding diathesis 118 Smith PS. Co ngenital coagul a tion protein deficiencies in the
due to decreased functional activity of type I plasminogen perinatal period. Semin Perinatal 1990 ; 14: 384-392.
activator inhibitor.) Clillinvest 1989; 83:1747-175 2. 119 KJetzel M, Miller CH, Becton DL et 31. Post delivery head
95 Dieval J, Nguyen G, Gross S et al. A lifelong bleeding disorder bleeding in hemophili c neonates. Am ) Dis Clli/d 1989;
associated with a deficiency of plasminogen activator 143: 1107-1110.
inhibitor type J. Blood 1991; 77:5 28- 53 2. 120 Ljung R, Lindgren AC, Petrini P, Tengborn L. Normal vaginal
96 Fay WP, Parker AC, Condrey LR, Shapiro AD. Human delivery is to be recommended for haemophilia ca rrier
pla sminogen activator inhbitor-l (PAl-I) defici e ncy: gravidae. Acta Paediatrica 1994; 83 :609-61 J.
characterization of a large kindred wi th a null mutation in the J21 Kadir RA, Economides DL. Obstetric management of carriers
PAI-l gene. Blood 1997 ; 90:204-208. of haemophilia A. Haemophilia J997; 3:81-86.
97 Nurden AT, Caen JP. An abnormal platelet glycoprotein
12 2 Conway JH, Hilgarter MW. Initial presentations of pediatric
pattern in three cases of Glan z mann's thromba sthenia.
hemophiliacs. Arch Pediatr Adolesc Med 1994; 148:589-594.
Br) Haematol 1974 ; 28:253-260.
123 Schiver W, Brueschke EE, Dent T. Family practice grand rounds:
98 Phillips DR, Agin PP. Platelet membrane defects in hemophilia.) Fam Pmct 1982; 14:661-662,664, passim.
Gl a nzmann 's thrombasthenia. Evidence for decreased amounts 124 Peyvandi F, Mannucci PM. Rare coagulation disorders.
of two maj or glycoproteins.) Gin Invest J977; 60:535-545. Thromb Haemostasis 1999; 82:1207-1214.
99 Gl a n zman n E. Hereditare hamorrhagische Thrombasthenie.
12 5 Bolton-Maggs PH , Peny OJ , Chalmers EA et a!. The rare
Ein Betrag zur Pathologie der Blutplattchen. )alzrbuch
coagulation disorders : review with guidelines from the UK
Kinderheilkunde 1918 ; 18 : 1-42.
haemophilia Doctors Organisation. Hael110philia 2004;
100 Sauslbury FT, Kesler RW. Thrombocytosis in Henoch 10:593-628.
Schonlein purpura. Clinical Pediatr 1983; 22:185-87. 126 Katz JA, Moake JL, McPherson PO et al. Relationship between
101 Weinstein M, Babyn P, Zlotkin S. An orange a day keeps the human devel opment and disappearance of unusually large
doctor away: scurvy in the year 2000. Pediatrics 2001; 108:55. von Willebrand factor multimers from plasma. Blood J989 ;
102 Chatproedprai S, Wananukul S. Scurvy: a case report. ) Med 73: 1851-1858.
Assoc Thailand 2001; U4(Suppl. J):S 106-lJO. 127 Wetzel RC, Slater AJ, Dover GJ. Fatal intramuscular bleeding
103 Stewart GM , Rosenberg NM. Conditions mistaken for child misdiagnosed as suspected non-accidental injury. Pediatrics
abuse: part I. Paediatr Emerg Care 1996; 12:116-12J. 1995; 95:771-773.
104 Aldri ch RA. Steinberg AG, Campbell DC. Pedig ree 128 Carpentieri U, Gustavson LP, Haggard ME. Misdiagnosis of
demonstrating a sex-linked recessive condition characterized neglect in a child with bleeding disorder and cystic fibro sis.
by draining ears, eczematoid dermatitis and bloody diarrhea. South Med) 1978; 71 :854-855.
Pediatrics 1954; 13 :1 33-139. 129 Kaplan SJ. Pelcovitz 0, Salzinger S et al. Adolescent physical
105 Clemetson KH, Clemetson JM. Molecular abnormalities in abuse: risk for adolescent psychiatri c disorders. Am J
Glanzman 's thrombasthenia, Bernard-Soulier syndrome and Psychiatry 1998 ; 155: 954-959.
platelet-type von Willebrand's disease. CUTl' Opill Hematol 130 Sutor AH , Dagres N, Nied erhoff H. Late form of vitamin K
1994; 1 :388-393. bl eeding in Germany. Klinische Padiatrie 1995; 207:89-97.
106 Hermansky F, Pudlak P. Albinism associated with 131 White ST, Voter K, Perry J. Surreptitious warfarin ingestion.
hem orrhagic diathesi s and unusual reticulated pigmented cells Child Abuse Neglect 1985; 9: 3 49-352.
in the bone marrow: a report of tVIO cases with histoch emical 132 Souid AK, Korins K, Keith 0 et al. Unexplained menorrhagia
studies. Blood 1959; 14:162-169. and he m a turia: a case of Munchausen's syndrome by proxy.
107 White JG. Structural defects in inherited and giant platelet Pedin tr Hel17otol Oncol 1993; 10:245-248.
disorders. Adv HlIm Gen et 1990; 19:133-234. 133 Warkentin TE, Kelton JG. Aquired platelet disorders. In Bloom
108 White JG. Inherited abnormalities of the platelet membrane AL, Forbes CD, Thomas DP et al (eds) Haemostasis and
and sec retory granules. Human Pa th ology 1987 ; 1U: 123-139. Thrombosis. London: Churchill Livingstone, 1994, pp. 767-815.
109 Greinacher A. Muelle r-Eckh a rdt C. Hereditary types of 134 Richardson SG, Fletcher OJ, Jeavons PM, Stuart J. Sodium
thrombocytopenia with giant platelets and inclusion bodi es in valproate and platelet function. BM) 1975; 1 :221-2 22 .
the leukocytes. BIt//' 1990; 60:53-60. 135 Barr RD, Copeland SA, Stockwelll\llL et a l. Valproic acid and
110 Djaldetti M, Creter 0, Bujanover Y, Elian E. Ultrastructural immune thrombocytopenia. Arch Dis Childh 1982; 57:681-684.
and functional studies of the platelets in patien ts with 136 Ko CH, Kong CK, Tse Pw. Valproic acid and thrombocytopenia.
May-H egglin anomaly. Haemafologica 1982; 67:530-538. Hong Kong Med ) 2001; 7: J5-21.
III Epstein CJ, Sahud MA, Piel CF et al. Hereditary 137 Conley EL, Coley KC, Pollock BG et al. Prevalence and risk of
macrothrombocytopenia , nephriti s and deafness. Am) Med thrombocytopenia with valproic acid: experience at a
1972; 52:299-310. psychiatric teaching hospital. Pharmacoth erapy 2001;
112 Eckstein JD , Filip OJ , Watts JC. Hereditary thrombocytopenia, 21:1325-1330.
deafness and renal disease. Ann Intern Med 197 5 ; 82:639-645. 138 De Beradis 0, Campanella 0, Matera V et a l.
113 Lilleyman JS. Intracranial haemorrhage in idiopathic Thromb ocytopenia during valproic acid treatment in young
thrombocytopenic purpura. Arch Dis Child 1994; 61:251-253. patients with new-onset bipolar diso rder. ) Ciin
114 Lilleyman JS. Intracranial haemorrhage in chronic childhood Psychopharma col2003; 23:451-458.
ITP. Pedia tric Hematology and On cology 1997; 14:iii-v. 139 Ganick OJ, Sunder T, Finley JL. Severe hematolog ic toxicity
115 Lilleyman JS. Managem e nt of childhood idiopathic of valproic acid: a report of four patients. Am) Pedintr
thrombocytopenic purpura. Br) Haematol 1999; 105: 871-875. Hematol Oncol 1990; 12:80-85.
115 Bolton-Maggs PHB, Moon I. Assessment of UK practi ce for 140 Gesundheit B, Kirby M, Lau W et al. Thrombocytopenia and
management of acute childhood idiopathic thrombocytopenic megakaryocyte dysplasia: an adverse effect of v alproic acid
purpura against published guidelines. Lancet 1997; treatment.) Pediatr Hematol Onco12002; 24:589-590.
350:620-623. 141 Durongpisitkul K, Fururaj VJ, Park JM, Martin CF. The
117 Wheel er OM, Hobbs CJ. Mistakes in diagnosing non prevention of coronary artery anuerysm in Ka wasaki disease:
accidental injury: ]0 years' experience. Medical Practice a meta-analysis on the efficacy of aspirin and
1988; 296:1 23 3-123 6. immun oglobulin treatment. Pediatrics 1995 ; 96: 1057- 106J.
References I 1 05
142 Stuart MJ, Gross SJ, Elrad H, Graeber JE . Effects of 150 Leititis JU, Brandis M, Critical care in uraemic children,
acetosalicylic- acid ingestion on materna l and neon ata l Pediatric Nephrology \ 992; 6:88-95,
haemostasis. N Engl J Med 1982 ; 307:909-912. 151 Kelly D, Summerfi eld J, Haemostasis in liver disea se, Semil1
\43 Freedman MH, Doyle JJ. Inh erited bone marrow failure Liver Dis 1987; 7:182-\91.
syndro mes. In Lilleyman J, Hann I, Blanchette V (eds) Pediatric \ 52 Miner ME, Kaufman HH, Graham SH, Haar FH, Gilden berg PL.
Hematology. London: Churchill Livingstone, J 999, 23- 49. Disseminated intra vascu lar coagul ation fibrin olytic syndrom e
144 Young NS, Alter BP. Clinical features ofFanco ni 's anae mia. In following head injulY in children: frequency and prognostic
Aplastic Anaemia, Acquired 011d fuherited. Phila delphia, PA: implications, J Pedialr 1982; 100:687-691.
WB Saunders, \ 994 , pp: 275-309. 153 Hymel](P, Abshire Te, Luckey DW, Jenny C. Coa gulopathy
\45 Eden DB, Lilleyman JS. Guidelines for the management of in pediatric abusive head trauma. Pe(iialrics 1997;
idiopathic thrombocytopenic purpura. Arch Dis Child 1992 ; 99:371-375.
67: 1056-1058, 154 Stein SC, Young GS, Talucci RC et at. Delayed bra in injury
146 McCla in JL, Clark MM, Sandu sky GE, Undiagnosed, untrea ted after h ea d trauma: sign ificance of coagulopathy. NeurosurgelY
acute lymphobl astic leukaemi a presenting as susp ected child \992 ; 30:160-165.
abuse. J Forensic Sci 1990 ; 35:735-739, 155 Kaufman HH, Moake JL, Olson JD et al. Delayed and recurrent
147 Berrebi A, Ma lnick SDH, Vorst fJ , Stein D, Hi gh incidence of intracran ial hematolnas related to disseminated intravascular
factor Xl deficiency in Gau cher's disease. Am J Hema lol clottin g and fibrinoly sis in head injury, Neurosurgery 1980;
1992; 40:153-161. 7:445- 449 ,
148 Humphries JE, Hess CE, Gaucher's diseas e and acquired 156 Wedemeyer Ai, Ed son JR, Krivit W. Coagulation in
coagu lopathy. Am J Hemato l 1994; 45:347-353, cyanotic congenital heart disease. Am J Dis Child 1972;
149 Kelsey H, Chri stopoulos C, Gray AA. Mach in SI. Acquired 124:656- 660,
pseudo-pseudo Bernard-Soulier syndrome comp licating 157 Waldman JD, Czapek EE , Paul MH et al. Shorten ed platelet
Gaucher'S disease. J c/in Palh ol 1994; 47:162-165, survival in cyanotic heart disease. J Pedialr 1975; 87:77-79.
I CHAPTERS I
BIOCHEMICAL INVESTIGATIONS
ON POST-MORTEM SPECIMENS
Denis R Benjamin
INTRODUCTION
fluids decreases, compounds may dissociate from proteins
and other ligands thus changing their measured concentra
The diagnostic use of biochemical tests on post-mortem tion. This is especially important in the interpretation of drug
specimens dates back to the earliest development of the levels. A recent study, which included a large compendium
technology that enabl ed us to measure the blood concen of drugs detected in post-mortem blood, examin ed the pos
trations of various compounds. I It is safe to assume that sible factors affecting their concentrations. 2 Hormones and
almost evelY analyte measured during life has been other compounds may be similarly affected.
investigated in post-mortem samples. Unfoliunately, the Some compounds are relatively stable for prolonged
changes associated with death can profoundly affect the periods of time. These are of the most value in post-mortem
concentration of many intermediary metabolites. As a ll assessment. Others show relatively predictable changes,
the dynamic energy systems normally maintaining various increasing or decreasing with some definable relationship
concentration gradients begin to fail or run out of sub to the post-moliem in tervaL Numerous efforts have been
strate, compounds rapidly equi librate across the various made to use these for the assessment of the time since death,
body compartments and cell membranes. Such changes or if the time were accurately known, to use some 'correc
occur in the first few minutes to hours after death. Indeed, tion factor ' or equation to calculate what the ante-mortem
many such changes may already be under way before a value would have been. Both uses are fraught with diffi
formal declaration of death by the medical a ttendan ts. culty as the many factors influencing the rate of change are
With time, as cells die, intracellular compounds are usually not known, nor can they be readily controlled.
rel eased into the surrounding tissue and circulation, com Other compounds show erratic and unpredictable behaviour
pounding t he difficulties in the interpretation of results. with increasing time after death, which is of almost no ben
This is well illustrated by the onset of haemolysis in the efit in post-mortem diagnosis. For many analyies, changes
circulation. The rupture of red cells and t he release of in blood concentration are so non-specific that their meas
the haemoglobin and al l the intracellular enzymes in to the urement offers no signifi cant diagnostic information.
bloodstream often signa l the end of the useful period for Despite the obvious drawbacks of the post-mortem
the measurement of many metabolites. Not only does the changes, the investigation of sudden, unexpected death
concentration change in unpredictable ways, but also the in infants a nd children can be substantially aided by the
' matrix ' in which a particular ana lyte is normally measured judicious use of post-mortem biochemical testing on a
may be sign ificantly a ltered and this can have an important variety of body fluids a nd tissues. The recognition, over the
imp act on the analytical method. As the pH of vario us body past decade, of a large number of inherited metabolic
General eva lu ation I 107
defects, which may not be preceded by overt or obvious not militate against the thoughtful use of selected biochemi
clillical disease, has greatly ex pand ed the differential diag cal tests. On the other hand, the use of molecular biologic
nosis of such cases. A high index of suspicio n will ensure tech niques and tandem mass spectrometlY has opened pow
[ha t these are not missed at the time of the autopsy, as erful new avenues for the definitive diagnosis of a host of dis
diagnosis has imp0l1ant implications for the family and orders in infants and children who die unexpectedly.
future pregnancy planning. If ca refully so ught, clues are Failure to appreciate the impact of those factors influ
frequently uncovered in the his tO lY or on the gross and encing results of post-mortem biochemical investigations,
microscopic exam ination alerting the pathologist to a pos or to perform adequate cont ro l studies, has spawned a large
sible underl ying disease. Newer technologies, such as ta n literature in which numerous claims have been made for
dem mass sp ectrometry, now permi t ready scree ning for a the pathogenesis of certain diseases, only to be refuted by
wide variety of metabolic defects, an d molecul ar biologica l subsequent studies. Nowhere is this more eviden t than in
techniques can be used for specific genetic disord ers. These the investigation of sudden infant death sy ndrome (SIDS,
have considerably exp anded our abi lity to establish diag or criblcot death). The 'scientific' literature is filled with
noses from post-mortem material. It is pruden t to coHect all such poorly designed studies . There has been a characteris
of the approp ri ate samples, en sure that they are suitably tic pattern of publication in this field . The first couple of
stored an d then decide on how to proceed, once the various articles usua lly note an interesting chem ical observation in
lines of evidence unfold. a small number of infa nts. This is claimed to be sp ecific
All the caveats pe11aining to post-mortem biochemistry a nd sensitive. A few articles validatin g the observation
in adults apply in children . Most of the studies relating to the may follo w. But soon the naysayers arrive, poin ting out
changes in con centration of various metabolites after death similar findings in a number of other settings. The speci
have been perfo rmed in ad ults, although a number have ficity of the original fin ding co mes under se rious scrutin y
been validated in children. Table 5.1 lists the common fac as more 'controls' are added and the impact of post
tors influencin g the interpretation of bioch emical resul ts. mortem in terval and other factors are assessed. And not
The literature relating to thana tochemi stry is relatively old. infrequently, the test or obselvation that appea red so hope
Many of the studies were carried out in the early yea rs, soo n ful at its inception fades into obscurity.
after clinical patho logy techni ques became available. Many This chapter only covers those analytes or compounds for
have no t been rigo rously repeated with curren t technology which interpretation is reasonably possible and there is some
and this has created potential problems. For example, flam e demonstrated clinical application . Readers are referred to a
photometry, whi ch was the sta ndard for electrolyte measure number of excellen t general reviews of post-mortem bio
ments in the 1950s and 1960s, has been largely repla ced chemistlY if they require information about analytes that
with ion-selective electrodes. Each analytic method is sensi are not dealt with in this discussion. The following are nat
tive to its own particular set of interferences, and one cannot covered in this chapter: lactate, pyruvate, ammonia, serum
necessarily extrap olate results between different methodol o enzymes, lipids, trace metals, immunoglobulins,) methaemo
gies. For this reaso n, reference ranges published in the older globin,4 myoglo bin,5.6 thiamine,? and thyroid and parathy
literature may no longer have currency. roid hormones 8- 14 as well as the other hormones.) It is indeed
For each analyte these variables need to be known and un fortunate that four of the most important indica tors of
appropriately considered. Because of the large number of metabolic diseases in infants and children, namely, elevated
possibilities, great precision is frequently not possible. Inter ammonia, low glucose, abnormal pyruvate an d lactate are so
pretation of the test results often requi res considerable judg unreliable in the post-mortem pelio d. Other fluids, such as
ment and certainty of diagnosis is seldom achieved , pelicardial 15 and synovial16 have had such limited study that
especially with the measurement of traditional intemlediary fe w ge neralizations about their value in the day-to-day prac
metabolites and electrolytes Even so, these difficulties should tice of pediatlic pathology can be made.
In considering the role of biochemical testing in the
post-mortem period in pediatric fore nsic cases, tests can be
Table 5.1 Factors affecting the results of biochemical di vid ed into two broad gro ups. Some tests are us eful for the
investigations assessment of t he general health of the child, whereas oth
ers focus on specific clinical disorders. Table 5.2 lists the
Post-mortem interval
role of these anci lla ry pro cedures in pediatric cases.
Temperature at which the body has been maintained
Sample type (e.g. blood, vitreous humour, urine, cerebrospinal
fluid, etc.) GENERAL EVALUATION
Sample site (e.g . left or right heart, peripheral blood)
Ana Iytica I method Nutritional Status
Interfering substances, e.g. drug s, alcoh ol
Resuscitative attempts
The nutritional statu s of the deceased is an important con
Mode of death (e.g. rapid, prolonged, presence of hypoxia, etc.)
sideratio n in many pa ediatric au topsies. This may have
108 I Biochemical investigations on post-mortem specimens
Table 5.2 Role of biochemical testing in paediatric forensic cases advantages over prealbumin for the assessment of recent
nutritional intake.
General evaluation
Both these proteins are very sensitive to the presence of
Nutritional status
the acute inflammatory response. Hepatic synthesis rapidly
Hypoxia
decreases in the face of any inflammatory stimulus, be it
Inflammation
from infection, trauma, neoplasia or autoimmune disease.
Anaphylaxis
For this reaso n prealbumin should always be combined with
Infection
some measure of the acute phase response. At this time
Dehydration
C-reactive protein is the recommended test (see below).19
Time of death (post-mortem interval)
Specific disarders
Endocrine disorders
LONG-TERM NUTRITIONAL STATUS
Diabetes The long-term nutritional status is much more problematic
Adrenal and/or pituitary insufficiency to assess from a biochemical standpOint and the laboratory
Renal failure probably offers few significant advantages over anthropo
Inborn errors of metabolism metric studies. When the body is adapting to the decrease in
Fatty acid oxidation defects nutrients, including total calories and protein, there is a phase
Other mitochondrial disorders of compensation during which a variety of hormonal adapta
Urea cycle defects tions and changes occur. These are difficult enough to evalu
Aminoacidopathies ate under ideal circumstances during life, and become even
Organic acid urias more so with the post-mortem changes in blood chemistry.
Abnormal carbohydrate metabolism Similarly, many tests of immune function become abnormal
Miscellaneous in the chronically malnourished infant and child but these are
almost impossible to evaluate after death.
Selected serum proteins have long been used as surro
gate measures of nutrition. Albumin is a non-specific and
implications in regard to parental neglect and/or abuse, or
relatively insensitive measure of nutritional status. Its
to a chronic, but clinically unsuspected, underlying dis
serum concentration is dependent on a variety of factors.
ease. Evaluation of the physical characteristics and stan
Post-mortem levels in the serum are close to ante-mortem
dard anthropometric measurements should be carefully
values unless there is significant haemolysis. The problem
performed and compared with the appropriate standards
is not with any post-mortem changes, but with t he ma ny
for age, sex and ethnic group. These will be abnormal in
factors influencing serum albumin concentration during
many chronic diseases, including endocrinolog ica l, meta
life . The serum value is the net result of the protein intake
bolic, immunological and renal disord ers. From the labora
and hepatic synthesis, degradation and loss. Each of these
tory standpoint there are only a small number of tests that
is subject to many influences other than protein intake. It
can be utilized to assess both recent and long-term nutri
should also be remembered that in cases in which caloric
tional status. Obviously, none of the dynamic or functional
insufficiency exceeds the protein deprivation (i.e. maras
tests that is helpful during life can be used.
mus) serum protein levels might be normal.
Because of the long half-life of albumin in the circula
RECE NT NUTRITIONAL INTAKE
tion (18-20 days) and its large extravascular pool, changes
Knowing the recent food intake history of the deceased in serum values are slow to develop. Decrease in serum
sometimes has important diagnostic implications. Perhaps albumin develops late in the course of malnutrition and
the best serum protein for evaluating recent protein and only in the more severe cases. In interpreting the albumin
calorie intake is prealbumin. Prealbumin, also known as level the following factors must be kept in mind:
thyroxin-binding prealbumin or transthyretin, is a transport • Serum albumin is dependent on t he hydrational state
protein with a very short half-life (2 days). It is synthesized of the patient, so that dehydration in a malnourished
by the liver and has a small pool size. Either protein or calo child may result in relatively normal albumin level.
rie deprivation will result in a decrease in the serum level of • Both synthesis and catabolism are dep endent on
prealbumin in 3 to 4 days.17,18 As such it is an indicator of hepatic function and the acute infl a mmatory response.
recent nutritional intake rather than an accurate assessment The presence of abnormalities of either will
of nut ritional status. Any infant or child who has been significantly change albumin levels.
recently sick or who may have had significant anorexia • Renal and gastrointestinal losses of protein can
could have a depressed prealbumin level. There are no good profoundly decrease the albumin levels.
studies of prealbumin in the immediate post-mortem inter
val. Retinal-binding protein, also produced by the liver has If all of these can be excluded then low levels roughly
an even shorter half-life (approximately 12 hours). It has no correlate with the severity of chronic malnutrition.
Anaphylaxis I 1.09
HYPOXIA
appropriate clinica l indications, albeit quite non-spec ific.
Performing a satisfactory ESR test after death is not possi
A laboratory test for the detection of ante-mortem hypoxia ble with all the changes in coagulation and fibrinolysis. In
would be a great boon for the forensic pathologis t. The the last two decades, C-reactive protein has found a very
question arises in a diverse array of circumstances, varyin g useful role as an alternative or a compl ementary indication
from the suspicion of suffoca tion and strangulation to of inflammation; it can be used as a replacement for the
drowning and fire victi ms and to natural causes of death, ESR in the majority of clinical situations or in addition to
such as SIDS. Needless to say, although many tests have it. C-reactive protein has many advantages over the ESR, as
been investigated, none has proven to be reliable or satis it is not influenced by many of the other proteins that
factory. This is hardly a surprise as so many mechanisms of affect the ESR and it is independent of the shape, size and
death involve some element of hypoxia as one of the com number of the red cells. It is produced by the liver as part
mon final pathways a nd the rapidity of the post-mortem of the acute phase response and is a sensitive marker of
change on most of the candidate chemical compounds pre inflammation . It increases dramatically in the serum, rising
cludes their use after more than a few minutes . For exam to over 10 times the baseline value within 12 hours.)2,))
ple, post-mortem oximetry, with blood being samp led from None of the other proteins that also increase in response to
va rious sites in the circulation, was no t fou nd to be partic inflammation, such as a -acid glycop rotein (orosmucoid)
ularly helpful or reliable in defining the cause of death in a nd tra nsferrin , offers any advantages over C-reactive pro
one study of 214 ad ult cases .20 Other compounds have also tein in terms of rapidity of change, ease of technical meas
been investigated from this standpoint,21 such as lactate, urement or stability in the serum afte r death. A C-reactive
which increases in response to anaerobic glyco lysis. How protein level that is greater than 0.8 mg/dL is good indica
ever, there is such wide variation in the res ul ts that inter tion of a pre-existing inflammatory state. The higher the
pretation in th e indivi dual case is all but impossible. valu e, t he greater the chance of a bacterial infection,
Normally, lactate rises quite promptly in the blood - within alth ough elevations in C-reactive protein should always be
a day reaching leve ls of up to 70 times greater than the considered quite non-specific. A number of studies have
ante-mortem va lues .22 The rise of lactate in the vi treous confirmed the relative stability of this protein in the imme
humour is not quite so dramatic, but is sti ll substantial. 2J di ate post-mortem period. )4,)5 A number of recent investi
Hypoxa nt hin e is formed from adenosi ne monophosphate gations have also confirmed its value.)6,)7
in the presence of hypoxia. Its appearance in the vitreous A great deal of recent work, from Europe in general and
humour has been used as indicatio n of ante-mortem France and Germany in particular, has focused on the use
hypoxia. For examp le, some investigators have hypothesized of pro calcitonin as a marker of bacterial infection .)8,)9
that a SIDS death may be preceded by hypoxic episodes of There is as yet no consensus of opinion on whether this is
varying duration. 24 One study investigated the hypoxan more useful t han C-reactive protein. Its behaviour in post
thine levels in four g roups of patients including a group of mortem specimens has recently been studied,4o and some
SlDS victims, ch ildren with congenital heart disease, babies of the changes associated with tempera ture and storage
who succumbed from respiratory distress and a control have also been documented .41
group of accident victims in whom there was no evidence of
hypoxia. A sizeable fraction of the infants and children in
the SIDS, respiratory distress and congenital heart groups ANAPHYLAXIS
had elevated levels of vitreous hypoxanthine.25 This obser
vation has been challenged by other stud ies, which have all An acute anaphylactic reaction is a possible mechanism of
failed to corroborate the results. 26 - 28 At this time there is death in a small minority of paediatric cases. There is usually
conflicting evidence to support a significant difference in the a past history of severe hyp ersensitivity to a particul ar all er
hypoxa nthine concentration between SIDS and other causes gen, such as peanuts or shellfish. However, bee venom and
of death in infants. 29 ,)O Most recently, the meas urement of other allergens may also be responsible. Generally the clini
vascular endothelial growth factor (VEGF) in cerebrospinal cal features of overwhelming anaphylaxis are evident,
fluid (CSF) has been used to assess the incidence of hypoxia although in some cases they can be so acute or confusing as
in SlDS victims.)1 to go unrecognized. A number of tests have been promoted
for the diagnosis of anap hylaxis. Histamine is released from
both tissue mast cells and the circulating basophils. It has a
INFLAMMATION very short half-life and so is of no use in the post-mortem
diagnosis of anaphylaxis. Of the other tests, serum tryptase,
It is freque ntly useful to know if there is a pre-ex isting released from degranul ating mast celis, has attracted the most
infl amm atory state at the time of death. This 'acute phase attention. This enzyme is not released from the circulating
response' commonly accompanies infections, autoimmune basophils. A number of studies in living patients, involvi ng
conditions or maligna ncies. The erythrocyte sedimentation cases of witnessed anaphylaxis, have demonstrated raised
rate (ESR) has been the tradition al screening test with the levels of serum tryptase in association with an increase in
110 I Biochemical investigations on post-mortem specimens
allergen-specific immunoglobulin E (IgE). In the early studies laboratory for assistance in the diagnosis. For example,
of post-mortem specimens the resuHs appeared quite encour toxic shock syndrome has been identified as a cause of
aging; however, these involved only a small number of unexpected death in a child , with the identification of the
cases. 42.43 Subsequent evaluation in a group of 49 patients staphylococcal toxin TTST-1 in brain tissue. 51
known to have died from other causes, with no evidence of The presence of bacterial endotoxin has been sought in
anaphyla xis whatsoever, showed that many samples had a variety of clinical situations. The usual test is based on
levels of tryptase above the usual detectable levels in the the limulus lysate assay, of wh ich there are a number of
serum.44 Of these, five subjects had levels of tryptase that modifications. The level of endotoxin is not significantly
were greater than 10 ng/mL, which was regarded as a positive affected by the post-moJiem interval in the first 48 hours
result, and one had an extremely high value. The mechanisms after death, nor does it correlate with the blood culture
for these increases were quite unclear. There was no correla results. Detectable endotoxin has been noted in a many dif
tion with the post-mortem interval. Further studies have ferent causes of death, including blunt injury as well as
demonstrated raised levels in a variety of circumstances. 45 severe infection. However, it is interesting to note that it
At this time it is advisable to utilize both the senlm has not been found in a study of SIDS victims. 52
tryptase level and an increase in the allergen-specific IgE
for the diagnosis of anaphy laxis. A recent study concluded
that as many of 13 per cent of deaths in adults may be DEHYDRATION AND ELECTROLYTES
accompan ied by the activation of mast cells 46 There has
been one formal investigation into the lise of tryptase in The question of possible dehydration is a frequent consider
sudden unexpected deaths in infants. This study suggested ation , especially in paediatrics when the caregivers may
that mast cell degranulation was more evident in infants in underestimate the extent and severity of vomiting/diarrhoea
t he SIDS groups than in the infants in whom a cause of and there is inadequate fluid replacement. Dehydration
death was found. 47 This was a small study that has not develops more rapidly in infants and young children than it
been corroborated in a systematic survey involving a larger does in adults and is often unrecognized by the parents. The
number of control subjects, with careful attention to the chemical diagnosis of dehydration and the accompanying
post-mortem interval and mechanism of death; this issue electrolyte abnormalities have been well studied. The ability
has yet to be resolved. 48.49 of cells to maintain their normal concentration gradients of
ions is highly energy dependent and this rapid ly decreases
after death. Within minutes, intracellular ions, such as potas
INFECTION sium ions, begin to equilibrate across the cell membrane.
After an hour the serum potassium may well be six times
If there is any suspicion of infection then the appropriate cul the ante-mortem level, even in the absence of haemolysis.
tures should be obtained. The problems of interpreting post Once haemolysis develops, the pot assium concentration
mortem bacterial cultures are well known , eve n if standard soars even higher. Therefore, serum potassium should n ever
techniques are followed. Culturing multiple sites may be use be used to assess the state of dehydra t ion as it increases
ful or separately sampling each ventricle of the heart might geometrically after death. The appearance of haemolysis is
provide additional infomlation that assists in the interpreta quite variable, depending on the temperature and the storage
tion of positive cultures. 5o Viral cultures can be very inform of the body. It can take as long as 48 hours for haemolysis
ative and the use of new polymerase chain reaction (PCR) to be evident. Serum sodium concentrations' tend to
assays for the detection of many human pathogens is becom decrease more slowly after death compared with the rapid
ing more widespread. These should always be considered in and very significant changes with potassium. Sodium
cases of possible myocarditis and meningit is. Over the years decreases at a rate of approximately 0.9 mEq/L per hour
we have been struck by the poor correlation between the after death;5J chloride decreases as well , with an average
gross appearance of the heart and the microscopic presence fall of 0.97 mEq/L per hour. However, there is so much vari
of significant myocarditis. For this reason we tend to err on ation in sodium and chloride ion levels from among
the side of conservatism and obtain viral studies in most subjects that they have not been useful for the assessment
cases of unexpected death. The inherent sensitivity of these of post-mortem interval. In addition, serum is not the ideal
molecular diagnostic techniques raises the difficult issue of body fluid for the assessment of dehydration unless blood
the clinical significance of a positive finding. We are only just can be obtained very soon after death.
beginning to learn about our 'normal' microbial flora as Sodium and chloride, as well as osmolality, are much
defined by PCR methodology. more stable in the vitreous humour. Almost all studies of
Assays for specific bacterial tox ins are described and electrolytes have relied on the measurement of their con
have been quite useful in selected cases. These are not gen centrations in v it reous humour,2J, 54,55 which are similar
erally available in most clinical laboratories. If a particular in both adults and children, 56- 58 The sodium and chloride
infection is suspected from the gross and microscopic find levels in the vitreous humour change in parallel with tbe
ings then the pathologist should seek out the appropriate serum and every study has shown very good correlation
Time of death (post-mortem interval) I 111
between the known concentration of these during the ter 4. The decomposition pattern, in which the sodium and
minal hours of life and the post-mortem values in the chloride are both low but the potassium is markedly
vitreous humour. The levels of these ions are the most elevated (>20 mEq/L).
reliable for the assessment of hypernatraemic dehydration.
Measured with ion-se lective electrodes, concentrations of As can be deduced, the hypotonic and decomposition
sodium of > 165mEq/L and chloride >125mEq/L are patterns are only distinguished by the potassium concen
excellent indi cations of hypernatraemic dehydration. The tration. For this reason it is important to measure the vitre
blood urea nitrogen (BUN) may also be significantly ele ous potassium level as part of any electrolyte panel in order
vated. If renal failure has developed, the creatinine will to separate these t\ovo pathogenetic mechanisms, Vitreous
also be increased. Although severe dehydration is the usual potassium has been well studied, as it shows a linear
cause of hypernatraemia, either as a result of a disease such increase after death and has been utilized for the assess
as gastroenteritis, or neglect by the caregivers, other possi ment of the post-mortem interval 64 (see below).
bilities must be kept in mind. Excess loss of water may Coe and Apple 65 have studied the influence of the
occur through the skin and lungs in infants and children method of measurement on the values of vitreous elec
with fever, or there may be excessive renal excretion, such trolytes and showed very sizeable differences between tra
as occurs in diabetes insipidus (both central and nephro ditional flame photometry and ion selective electrodes. It is
genic), osmotic diuresis (e.g. chronic renal failure, hyper very important therefore for each laboratory performing
glycaemia) or hypercalcaemia. Both accidental salt these assays to establish their own unique reference ranges,
poisoning and inte ntional salt poisoning have also been and fo r the pathologist to be knowledgeable about the
described in chi ldren. 59 Claims that a significant fraction of methodology used for satisfactory interpretation.
infants dying of SIDS have significa nt electrolyte distur Studies on the osmolality of vitreous fluid are quite lim
bance based on vitreous chemistry60 are almost certainly ited; in normal subjects it has been repolted to range from
overestimated. 305 mOsm/kg 66 to 346 mOsm/kg.67 No systematic studies of
In certain cases of dehydration, when fluid witho ut suf v itreous osmolality in the post-mortem period have been
ficient electrolytes is given to the patient to drink, such as performed in children covering a wide enough spectrum of
water or apple juice, hyponatraemic and hypochloraemic pathologies. The anion gap, a useful measurement during
dehydration can occur (e.g. excessive vomiting as might life for the interpretation of electrolyte and acid-base dis
accompany pyloric stenosis) .61 The vitreous sodium is turbances, plays no role in post-mortem chemistry,
generally less than 135 mEq/L and the chloride is less than
95 mEq/L. Hyponatra emic dehydration is much less fre
quent than the hypernatraem ic form. Both adrenal insuffi TIME OF DEATH (POST-MORTEM INTERVAL)
ciency (see below) and salt-losing nephritis are causes of
hyponatraemia in the dehydrated patient. Other causes of Over the years, in addition to all the gross pathological evi
hyponatraemia and hypochloraemia should also be consid dence and the temperature of various body organs, other
ered, including lethal water intoxication due to intentional, methods have been sought to more accurately define the
forced water ingestion. In the oedemato us patient, low time of death. 68 .69 No other question, apart from the cause
sodium frequently accompanies congestive cardiac failure, of death, has provoked so many studies or spilled so much
hepatic cirrhosis and the nephrotic syndrome 6 2 However, it ink, The answer to this forensic question is often crucial to
must be remembered that not all cases of dehydration are the entire case. Tests on blood constituents have largely
accompanied by significant electrolyte disturbances. been abandoned in favour of a few ana lytes in vitreo us
Coe J,6J has described four basic patterns of electrolyte humour. These have included such compo unds as inorganic
abnormalities in the vitreous humour in forensic cases in phosphorus,7o amino nitrogen, non-protein nitrogen,
which they are diagnostically helpful: ammonia and creatinine. 7I Levels of amino acids also rise
1. The hypertonic dehydration pattern, which is the sharply after death, although there is considerable varia
common pattern in most cases of severe dehydration tion between the various amino acids in terms of the rate at
leading to an unexpected death. In this state, both which their levels rise. None of these analytes has been
sodium and chloride are elevated, usu ally to a similar proven to be sufficiently rei iabl e or reproducible for the
degree, and the urea nitrogen may be mildly to estimation of the post-mortem interval on a routine basis.
moderately increased. In the last tvvo decades, potassium and hypo xanthine have
2. The uraemic pattern, in which sodium and chloride are gained the most favour for the assessment of the post
relatively normal, while both the creatinine and the mortem intervaI. 64 ,72,7] In the uncomplicated adult case, both
urea nitrogen are elevated (see below). the potassium and hypoxanthine concentrations in vitreous
3. The hypotonic or low-salt pattern, in which both the humour, show a linear increase in the early post-mortem
sodium and chloride are low, as is the potassium period. 74,75 However, the rate of change after the first 24
concentration; the potassium concentration should hours is different from the initial rate and this is seldom taken
be less than 15 mEq/L. into account. It has been suggested that hypoxa nthine shows
112 I Biochemical investigations on post-mortem specimens
less varia tion in the early post-mortem period, especially in such a sample maybe indicative of hypoglycaemia, depend
the first 12 hours. 76 A variety of studies have investigated the ing on the length of the post-mortem interval. Note, how
relationship of these two analytes to the post -mol1em interval ever, that it has been well shown that high levels of glucose
and the effect of temperature. Also, equations have been con may occur in peripheral blood sampled from a distal
structed to estimate the time of death, ordinarily based on extremity even in patients who are not diabetic.8o .8 ! This has
regress ion analyses from cases in which the time of death was been observed in a variety of deaths due to different etiolo
precisely known 77 The most recent example suggests the use gies and may be the result of a terminal surge of noradren
of both measurements 78 and the formulae constructed from aline (norepinephrine) and adrenaline (epinephrine) and/or
this study are as follows : post-m0l1em interval (hours) = the iatrogenic effects of resuscitation. 82 In interpreting the
4.32 X (potassium [mmol/Lll - 18.35; post-mortem interval glucose level, the terminal therapy received by the patient
(hours) = 0.31 X (hypoxanthine [mmol/Lll + 0.05. The final should be carefully scrutinized, including all the adminis
estimate is based on the mean of these two calculations. tered drugs and the intravenous fiuids. The use of car
Needless to say there are a variety of methodological diopulmonary resuscitation outside the hospital has risen
and conceptual problems with these estimations. Recent dramatically in many Western countries, often being
refinements in the formulae have not been systematically administered by paramedical personnel, and such interven
tested for routine practice in paediatrics. They are far from tion can significantly impact the glucose values. In some
precise, with considerable variation from case to case. They communities it is the rare patient who avoids the chemical
have not replaced body temperature in the early post 'last rites'.
mortem period . Their use in adults shou ld be considered Some investigators, such as Coe,] feel very strongly that
once the body reaches ambient temperature. The rate of a diagnosis of diabetes mellitus should never be based exclu
increase in both ana lytes is not perfectly linear over time sively on a blood glucose measurement. The level of glucose
and this has been ignored in all the regression equations in the CSF (rather than in blood) is more reliable as it is less
published to date. Moreover, the mechanism of death and affected by the terminal conditions that can so dramatically
the length and extent of ante-mol1em hypoxi a have not infiuence the serum. On the other hand, the increasing use of
always been well controlled for in the study popu vitreous fiuid has largely supplanted CSF as the sample of
lations. The environmental temperature at wh ich the body choice for the diagnosis of hyperglycaemia.
remains after death has a major impact on the rate of In the normal situation, the vitreous humour glucose
change of potassium. Finally, the number of studies that level falls after death due to continuing glycolysis. In most
have included paediatric cases is very limited. There is evi uncontrolled diabetics the glucose will remain above
dence that potassium increases much more rapidly in the 200 mg/dL. Ketones invariably accompany the hypergly
vitreous humour of the infant than in that of the caemia and can be detected by the standard qualitative lab
adu]t. 3 ,58.79 oratory procedures in the vitreous fiuid (or peripheral blood).
Some of the problems associated with hypoxanthine are Diabetes has been diagnosed in embalmed cadavers.8] Some
discussed above in the section on hypoxia . have recommended the use of both lactate and glucose
measurements on the vitreous humour, in addition to the
measurement of ketone bodies in the serum, such as acetone,
ENDOCRINE DISORDERS for the diagnosis of diabetes. 82 However, a cl ose examina
tion of the data shows no significant benefit of the addi
Diabetes lVIellitus tional lactate measurement. The circumstances and
environm enta l tempe rature around the time of death may
significantly infiuence the rate of change of the vitreous glu
GLUCOSE
cose. Rapid cooling, as might occur in drowning (cold water
Estimation of the glucose level is important for the detec immersion) reduces the rate of glycolysis, such that the glu
tion of a number of disorders. It is obviously critical for the cose may be higher for a given post-mol1em in terval. This
determination of uncontrolled or untreated di abetes melli may be compounded by the stress response accompanying
tus, during which the severe hyperglycaemia is usually such deaths. 84 ,8 5
accompanied by significant ketoacidosis. When decreased, In recent years, glycosylated haemoglobin (haemoglobin
glucose is one of the key metabolites indicative of a possi A Ic) has been used as a reliable marker of long-stand ing
ble metabolic disease, or may be associated with starvation, hyperglycaemia. It is stable in the post-mol1em blood and is
abuse and neglect. Glucose measurements have been well a refiection of the average blood glucose leve l over the pre
investigated after death. The specimen type and its proper ceding 3 months. It is important to recall that there may be
selection is crucial for evaluation. If blood is taken from the other mechanisms for long-standing hyperglycaemia. For
right side of the heal1 or the inferior vena cava, glucose lev exa mple, one study of 28 cadavers demonstrated raised
els may be very high. This is a result of hepatic glycogenol haemoglobin A Ic in all five patients with known diabetes. 86
ysis. With time, the glucose diffuses into the surrounding In addition, seven subjects with malignancies, who had been
vessels. On the other hand, very low levels of glucose from receiving steroid therapy, also had elevated levels. Other
Endocrine disorders I 113
studies have confirmed the value of measuring glycated which it is associated with hyperglycaemia. On the other
haemoglobin as an assessment of hyperglycemia. B7 - B9 hand, the finding of ketosis with hypoglycaemia is seen
Fructosamine, which is a measure of non-labile glycated in some of the glycogen storage disorders, such as Von
serum proteins, has also been used during life to monitor Gierke's disease. Perhaps even more important than the
the effectiveness of diabetic therapy. It too is an indication finding of ketones in an infant or child with sudden death
of the mean blood sugar level over the preceding weeks. is the absence of ketosis, especially if there is any evidence
One of the advantages of fructosamine is that it is influ of hypoglycaemia. The lack of ketones should trigger the
enced neither by the presence of any abnormal haemoglo investigations for a possible (3-fatty acid oxidation disor
bins, nor the dynamics of red cell turnover in patients who der. Hypoglycemia with negative ketones is also present in
may have haematological problems. However, there are nesidioblastosis and other states of hyperinsulinism.
only a few post-mortem studies on fructosamine. 90. 91
Although sudden unexpected death in childhood due to
undiagnosed diabetes mellitus is very rare, it has been
Addison's Disease: Adrenal Insufficiency
reported. 92 •93 In one particular case an ll-year-old girl died
suddenly after a 2-day history of a minor respiratory tract
Sudden death in infants and children as a result of an
infection. The autopsy was performed 3 hours after death and
Addisonian Gisis is well described. 95 - 99 It may be due to
documented a vitreous humour glucose level of 606 mg/dL,
congenital adrenal hypoplasia or destruction of the adrenal
blood acetone level of 24 mg/dL, marked glucosuria
glands as a result of a number of different pathological
(2073 mg/dL) and positive urinary ketones. Her liver con
processes. These latter cases are more likely to occur in the
tained microvesicular fat and the pancreas showed typical
childhood and teenage years, 100 whereas the congenital forms
changes in the islets. 94 The combination of an elevated glu
generally are present in infancy and may mimic SlDS. JOI
cose in the vitreous humour and acetone in the blood has
The gross and microscopic appearance of the adrenal
proven to be quite reliable for the diagnosis of unexpected
glands normally establishes the diagnosis. The usual meta
diabetes mellitus. Sudden death has also been well described
bolic derangement accompanying an Addisonian crisis is
in children who have been recently diagnosed and are under
usually reflected by low sodium and chloride concentration
going initial therapy.
in the vitreous humour and very elevated potassium, often
The diagnosis of hypoglycaemia in post-mortem sam
in the range of the 'decomposition' pattern described
ples is much more problematic than the detection of hyper
above. As salt wasting and dehydration are frequent fea
glycaemia. This remains a serious problem in paediatric
forensic cases as hypoglycaemia is such an important clue tures, levels of urea nitrogen, and even creatinine, may be
elevated. If urine is available, the elevated sodium level
for a variety of metabolic diseases. Ongoing glycolysis
should be evident. This combination of dehydration ,
generally results in a progressive decrease in glucose level
hyponatraemia and an elevated level of sodium in the urine
in all body fluids, including CSF and vitreous humour.
is generally indicative of adrenal insufficiency or salt
However, the dynamics of this decrease are quite unpre
dictable. Attempts to combine glucose measurements with losing nephritis. If both the adrenal glands and kidneys are
normal then one of the many causes of inappropriate
lactate and other metabolites have not proven to be satis
factory for routine diagnosis. At this time there is no secretion of antidiuretic hormone should be entertained.
acceptable method for the detection of hypoglycaemia. Random cortisol values will be markedly decreased in case
of adrenal insufficiency, regardless of the time of day. Cor
tisol concentration in the blood is relatively stable for up to
KETO NES 18 hours after death, especially if the body is cooled within
4 hours of death.102 Reference values are method, age and
The detection of ketone bodies (acetone, acetoacetic acid
sex dependent, so that interpretation depends on compari
and (3-hydroxybutyric acid) is important for the diagnosis
son with the specific laboratory 's reference values.
of a variety of conditions. Infants and children have a ten
dency to generate ketone bodies more readily than adults,
especially after a relatively short period of fasting. Children
shift their energy production to utilize fat quite promptly, Renal Failure
generating the ketone bodies in the process. The dipstick
test, which is based on the reaction of acetone and ace Renal failure in childhood may be quite occult, with few
toacetic acid with nitroprusside in an alkaline buffer to clinical indications. Sometimes the family or the attending
produce a purple colour, is quite sensitive. detecting as few physicians may not readily detect the growth failure asso
as 5 mg/dL in the urine. Unfortunately, the detection of ciated with declining renal function. Unexpected death in
ketones in body fluids is rather non-specific and should be such situations is quite uncommon but can occur. Gener
interpreted with caution. It accompanies starvation/fasting, ally the gross and/or microscopic appearance of the kid
vomiting and exercise, as well as any impairment of carbo neys and the urinary system will establish the diagnosis.
hydrate metabolism, most notably diabetes mellitus in The severity of the renal impairment can also be gauged by
----=-- --
- .'
114 I Biochemical investigations on post-mortem specimens
measuring the concentration of urea nitrogen and creati an increased anion gap (due the accumulation of unmea
nine in the blood and vitreous humour. Urea nitrogen is sured acids), hyperammonaemia, ketonuria, lactic acidosis
remarkably stable in these body fluids 56 ,IOJ-)05 and the or a respiratory alkalosis. Unfoliunately, neither ammonia
diagnosis of renal failure can readily be established. IOG , IU7 nor lactic acid measurements are reliable after death and
Of all analytes, urea nitrogen is perhaps the most stable, cannot be used. They tend to increase in both the serum
showing no diagnostically significant changes over a and the vitreous humour in a fairly lin ear fashion in the
period of up to 5 days post-mortem. J08 first 24 hours, but there are no systematic studies of post
Creatinine measurements in both blood and vitreous mortem ammonia levels in infants and children evaluating
humour are also quite stable,109 although some investiga their role in diagnosis. Organic acid ana lysis in the urine or
tors have described modest increases in creatinine after v itreous humour is valid after death. Many changes occur
72-96 hours. 106,107 Levels in the vitreous humour are in the plasma amino acids (including an elevation of glut
slightly lower than serum values 54 but parallel the changes amine and alanine) and the neurotransmitters GABA and
that occur in renal fai lure. In well-established renal failure, (1-alanine. This makes post-mortem amino acid measure
both the urea nitrogen and creatinine levels will be ments difficult to interpret.
markedly elevated. In states of acute dehydration and pre Table 5.4 lists a collectio n of metabolic diseases that are
renal failure, the level of creatinine will be normal and associated with 'energy production' and that have reportedly
the renal nitrogen will be elevated. The ratio of urea caused sudden unexpected death. This list will undoubtedly
nitrogen to creatinine is a useful estimate of prerenal ver grow as defects in each step of the complex pathways
sus renal failure. leading to the generation of adenosine triphosphate are
identified.
able 5.3 Genetic metabolic disorders which may present acutely during infancy
Acute presentation Major disorder Abnormal clinical test Disease example Confirming genetic
in infancy test
MSUD Plasma AAs
(l ethargy, coma, acidosis, Amino acid disorder NKH glycinemia Organic acids
hepatomegaly)
Tyrosina em ia-I Tissue enzymes
GTC Plasma/urin e AAs
Rule out: Urea cycle disorder NH 3t Argininosuccinate Orotic acid
• hypog Iycaem ia Citrullinaemia Tissue enzym es
• sepsis acidosis Methylmalonic
• CNS bleed Organic acids
Organic acid disorder NH3t (7) Propio nate/isova leric
• congestive heart Tissue enzym es
urine ketones Glutaric-II
• toxins
Renal loss Plasma/urine
HC0 3 loss 'Iytes
GI loss pH Et HC0 3
CI~S, central nervous system; 'Iytes, electrolytes; MSUD, maple syrup urine disease; NKH, non-ketotic hyperglycinaemia; OTC, ornithine
transcarbamilase; GI, gastrointestinal; Glutaric-II, glutaric aciduria type II; PDH def, pyruvate dehydrogenase deficiency; MERRF, myoclonic epilepsy
with ragged red fibres; MELAS, mitochondrial encephalomyopathy, lactic acidosis, stroke-like episodes; GSD, glycogen storage disease; MCAD, medium
chain acyl-CoA deficiency; LCAD, long-chain acyl-CoA deficiency; LCHAD, long-chain 3-hydroxyacyl-CoA-dehydrogenase; S, syndrome; MPS,
mucopolysaccharide; AAs, amino acids; RBC, red blood cell; bx, biopsy; VLCFA. very long cha in fatty acids ; EM, electron microscopy.
A minor illness, a short peliod of fas ting, an immuniza of the mother, might have triggered the metabolic crisis
~on or other stress might trigger the acute metabolic decom leading to a neonatal death.11 4
pensation. There are a number of rather ordinary situations The biochemistry of this group of disorders is complex
in which physiological fasting might occur and not be obvi and incompletely understood. It involves the transfer of
ous to the parents. The two classic examples occur at the fatty acids into the mitochondria via the carnitine transport
time of weaning and when an infant begins to sleep through mechanism and their oxidation to produ ce adenosine
~he night. It is not unco mmon for in fants to present with triphosphate (ATP). The oxidation of fa tty acids is an impor
sy mptoms at such times and it is useful to specifically tant energy source in many organs, apart from the brain,
~equest this history from the family. In a recent case of car but especially the heart and skeletal muscles. In the normal
:1itine transporter deficiency, it was surmised that the fasting situation, fatty acids enter the cell, are activated to their
stress caused by poor breast-feeding without any supple acyl-eoA esters, linked to carnitine by the action of carni
mentalY formula feedings, and possibly the vegetarian diet tine palmitoyl transferase 1 and II, and then translocated
116 I Biochemical investigations on post-mortem specimens
Disorder Reference(s}
across the inner mitochondrial membrane into the mito is the G958 allele. JJ8, 1J9 Medium chain acyl- CoA dehydro
chondrial matrix. The carnitine is removed and the acyl genase defiCiency accounts for the majority of cases of sud
CoA fatty acid then enters the 0-oxidation cycle. Repetitive den death due to a defect in fatty acid oxidation. It usually
rounds of dehydrogenation by a number of different dehy presents during infancy with lethargy, a Reye-like syn
drogenases (short, medium and long chain) depending on drome, accompanied by hypoketotic hypoglycaemia. As
the length of the fa tty acid, convert them to acetyl-eoA. many as 40 per cent of infants are said to die during their
This then enters the respiratory (electron) chain that gener first episode. Generally, the liver shows considerable fatty
ates the ATP. change, although there is great variation from case to case.
The diagnosis of abnormalities in 0-fatty acid oxidation, Moreover, lipid accumulation in the liver is quite non
which may be due to a defect at any of the many points in specific. Although much less common, the mutation for
the process, usually relies on the measurement of organic long ch ain 3-hydroxacyl-CoA dehydrogenase deficiency is
acids in the urine, carnitine in the serum and the profile of also well describ ed. 140
acyl-camitines. The in terp retation of the organic acid pro Tests for the diagnosis of 0-fatty acid oxidation defects
files on the urine as well as the acyl-carnitine profile requires include:
considerable experience and skill, and this is beyond the J. Urine: organic acids by gas chromatography/mass
scope of this chapter; those interested in pursuing the topic spectroscopy to identify specific glycine conjugates;
further should consult the suggested references. 13J-lJ5 urine acyl-carnitine profile to identify the specific
Depending on the results, more definitive tests involve the acyl-carnitine conjugates. If urine is not avai lable then
measurement of specific enzyme activities or direct detection consider collecting either bile or vitreous humour. Both
of a gene defect. bile l41 and vitreous humorlJl have been used
The use of the serum measurement of dodecanoic acid successfully to establish a diagnosis.
was initially promoted as a sensitive and specific screening 2. Liver: enzyme analysis for both medium- and long
test for the common fatty oxidation defects. JJ6 Soon after chain acyl-eoA dehydrogenase activity. 142 Immediate
the original publication, issues around its specificity sur freezing of tiss ues greatly retards the post-mortem
faced and it was reported to be much less specific than decrease in activity. However, the short-long chain
originally hoped. 137 Until this issue is resolved with addi ratio remains constant so that the difference between a
tional studies and a satisfactory cut-off value estab lished, true enzyme deficiency and post-mortem degradation,
this method of screening cannot be relied upon. However, in which both values will be decreased, can be
it may alert the pathologist to those cases need ing addi distinguished. An alternative approach to enzyme
tional work-up. activity is the analysis of various metabolites obtained
One of the most commonly identified disorders in this from the liver after methanol extraction and
group is medium chain acyl-CoA dehydrogenase deficiency ultracentrifugation. 14J This has been successfully
(MeAD). This is inherited as an au tosomal recessive, and the utilized for the detection of a number of the defects,
most frequent mutation accounting for 90 per cent of cases even in livers that were only obtained 72 hours after
Technical considerations at the time of autopsy I 117
death. This technique assays a large number of the Table 5.5 Other genetic metabolic disorders claimed to be
farry aci d metabolites, together with glucose and associated with sudden death
lactate. The pattern of metabolites present can
Carbohydrate metabolism
sometimes establish a specific diagnosis.
Galacatosaemia
J. D A: medium-chain acyl G985 mutation - a PCR
Hereditary fructose intolerance
assay is available;144 coenzyme A dehydrogenase
Glycogen storage disease, type"
de ficiency (MCAD) - the mutation has been detected in
Fructose-l ,6-diphosphatase deficiency
D A obtained from a variety of different sources,
including frozen blood, llOdried blood spots,145 Urea cycle defects
archived paraffin embedded material, 146 and liver. 147 Carbamylphosphate synthetase
Detection of the mutation associated with long-chain Ornithine carbamyltransferase
J-hydroxyacyl CoA dehydrogenase (G 1528C) is also Argininosuccinate synthetase
possible. 14o Newborn screening cards are an excellent Argininosuccinate lyase
source for this testing. Many DNA tests can be carried
Amino acids
out directly from the newborn screening blood spots,
Tyrosinaemia
as can acyl-carnitine profiling.
Isovaleric acidaemia
~. Cultu red fibroblasts: globa.l [9,1O- 3Hl myristic acid
Non-ketotic hyperglycinaemia
oxidation assay - detects at least nine different fatty
Maple syrup urine disease
acid oxidation defects;117 specific enzyme analysis.
Organic acidaemias
Propionic acidaemia
Other Genetic Metabolic Disorders Methylmalonic acidaemia
3-Hydroxy-2-methylbu tyric aciduria
:'l,number of disorders involving amino acid metabolism Glutaconic acidaemia
may cause unexpected death, especially in the first Isovaleric acidaemia
6 months of life. Some of these are also associated with the Multiple carboxylase deficiency
ab normal excretion of certa in organic acids. Of these, Glutaric acidu ria type II
maple syrup urine disease is an excellent example. 148
Miscellaneaus
Table 5.5 provides a list of potential enzym e deficiencies that
Lysine protein intolerance
remove all the fluid is based on some expelimentaJ data that laboratories prefer plasma for some of the analyses_ It is
demonstrated regional differences in the concentration of important that the laboratory's reference ranges clearly
certain analytes, Because of this, repetitive sampling of the specify the type of specimen , as there may be considerable
vitreous humour is not possible, Replacing the vitreous differences between whole blood, serum and plasma, Blood
humour with a volume of water or saline similar to the collected in EDTA can be used for the preparation of DNA.
amount aspirated can restore the shape of the eyeball by Blood collected in heparin or acid-citrate-dextrose (ACD)
maintaining its turgidity, a courtesy generally appreciated can be utilized for the preparation of white cells for the
by the mortician. analysis of lysosomal enzymes,
Vitreous humour can be used for the following tests; glu
cose, ketones, sodium, urea nitrogen and organic acids. The
differences in levels of organic acids in the blood and vitre Urine Collection
ous humour have been investigated. 111.156 Amino acid analy
sis is also satisfactory if the vitreous humour can be obtained The most convenient way to obtain urine is to perform a
in the first 24 hours after death and is rapidly depro bladder puncture once the abdomen has been opened. This
teinized. 139 With time there is a gradual decrease in the will ensure that the urine is not contaminated. In the event
sodium and glucose concentration, whereas urea (measured that the bladder is empty, attempts can be made to collect
as urea nitrogen) shows a slight rise. The level of magnesium a small quantity of urine by puncturing the renal pelvises.
is quite age dependent and also rises with time. The rate of Other techniques, such as catheterization are more com
decrease of sodium is approximately 0.5-0.65 mEq/l per plex, especially in small infants, as the appropriate size of
hour. Calcium shows no consistent or predictable changes catheter is often not available. Expressing the urine from a
after death. Abnormalities in calcium metabolism have not full bladder is also possible, although it is more difficult to
been successfully diagnosed after death. Glucose declines at collect than a simple bladder tap. Urine should be stored in
a rate of around 0.21-1.6 mg/dl/hour. Urea rises in the vit sterile plastic or clean glass tubes at -70· C, in aliquots of
reous humour at a rate of 0.081-0.62 mg/dL/hour. 56 Bicar 1-5 ml, until analysis.
bonate values are usually low in the vitreous humour, which Urine may not always be available. In one study of
may be an artefact due to exposure of the sample to the air infants dying of SIDS a substantial majority (60 per cent)
with loss of carbon dioxide. Such is apt to occur when a had no urine on opening the bladder. 132 If urine cannot be
small volume of vitreous humour is placed in a large con aspirated from the renal pelvis then Bennett et al 131 recom
tainer and if there is a long delay before the specimen is mend swabbing the surface of the bladder with a cotton
analysed. Other compounds that have been measured in the swab, storing the swab at -20·C and then submitting it as
vitreous humour include amino acids,157 hormones l50 and the specimen for organic acid analysis . However, most
lactate. 23 laboratories will not have the expertise to process such
samples effectively.
In the event that the urine cannot be shipped to the ref
Blood Collection erence laboratory in the frozen state, a few drops of 6N
hydrochloric acid or chloroform can be added as a preser
There is considerable variation in the concentration of cer vative in order to prevent the growth of bacteria. However
tain analytes between the right and left side of the it is preferable to maintain the specimen frozen until the
heart. 159.J 60 Moreover, the influence of post-mortem hepatic time of analysis.
metabolism, such as glycogenesis, can affect a number of Urine can be utilized for the measurement of ketone and
carbohydrates. As the largest body of clinical biochemical organic acids. If urine is not frozen rapidly then the insta
data is based on peripheral blood analysis during life, it bility of some of the urinary organic acids will result in the
makes the most sense to use this source whenever possible. disappearance of some of the key groups, such as the
Femoral or subclavian vein puncture will most accurately oxoacids. There are many potential artefacts and pitfalls in
reflect the ante-mortem values. the post-mortem analysis of urinary organic acids. The
Blood should be centrifuged as soon as possible and the most common and serious is bacterial contamination.
serum separated and stored at - 20· C or -70· C. The longer Freezing or inhibiting the growth with acidic compounds
the serum is in contact with the formed blood elements, the such as hydrochloric acid will avoid this. Bacterial over
more difficult interpretation becomes. 16J Haemolysis will growth can render the entire analysis meaningless. The key
affect the measurement of many analytes and should be for the successful interpretation of urinary organic acids is
avoided. If red cells are going to be analysed (e.g. for the close communication between the pathologists and the lab
presence of a haemoglobinopathy) then some blood should oratory worker who is performing the assay. Knowledge of
be anticoagulated with ethylenediaminetetraacetic acid the infant/child's diet and medication history may be cru
(EDTA) in the standard fashion. Collecting blood in EDTA cial. For example, certain acids such as adipic, furoic and
or heparin is always valuable, as the specimen can be tartaric acid may have a dietary origin. 'Nutramigen', a
immediately centrifuged and the plasma separated. Certain common supplementary formula, may be associated with
Technical considerations at the time of autopsy I 119
an increase in 5-oxoproline. Infa nt feeding formulas that metabolic disease is raised by the clinical and associated
a re supplemented with MCT oil will result in elevations of laboratory findings . In such circumstances, performing the
the dicarboxylic acids: adipic, suberic and sebaric. Many autopsy immedi a tely after death can circumvent the prob
pharmaceuticals profoundly affect the analysis, although lems of post-mortem change in tissue and body fluids.
this can generally be accounted for. Unfortunately, there Autopsy consent should be obtained ante-mortem if at all
a re only a velY few organic acid disorders that can be diag possible, or immediately at the time of death to reduce any
nosed from a single urine analysis, even if one pays very delay, and the body should be expeditiously transported to
close attention to the pattern of additional metabolites. the morgue. Most children 's hospitals offer autopsy serv
ices around the clock to expedite the diagnosis of such
cases. Tissue and body fluids should be collected on open
Fibroblast Culture for Enzyme Analysis ing the chest and abdomen and immediately processed to
limit any of the post-mortem changes. Tissue should not be
allowed to sit at room temperature for any length of time.
If there is any hint of a possible metabolic disease an
Three 5-mm cubes of liver, skeletal muscle, healt, CNS and
attempt should be made to establish culture of fibroblasts.
kidney should be snap-frozen in liquid nitrogen and stored
This is generally successful with the cell culture media
at -70°C. Urine and blood should be collected and stored
available nowad ays for at least 48 hours after death. The
as mentioned above. Very small samples (1 mm maximum)
sk in should be cleaned with an alcohol-based disinfectant,
of liver, skin, CNS, healt and other organ should be fixed
and well dried before excising a couple of2- to 4-mm J por
in glutaraldehyde or other suitable fixative for electron
lion of skin. These can usuaJJy be taken along one of the
microscopy. Thymus/lymph node or spleen can all be used
sta ndard autopsy incisions. The specimen must contain
for DNA studies as indicated above. Once the tissues have
adequate amounts of dermis. Some investigators have used
been colJected and stored the autopsy can then proceed in
:;"chilles tendon biopsies for the source of the fibroblasts;
a more considered and 'leisurely" fashion . ln many institu
this necessitates another incision but may decrease the
tions the formal autopsy may even be delayed to the next
bacterial overgrowth that is common in skin biopsies, even
working day.
with apparently adequate disinfection. The transport media
However, if it is not feasible to perform an autopsy
should contain antibiotics to suppress the growth of any
immediately after death then all is not lost. The largest sac
contaminating bacteria. Generally, the biopsy in the trans
rifices are in enzyme activity and electron microscopy.
port media should be maintained at room temperature prior
Although celtain assays, such as enzyme activity determi
(Q transfer into the growth media. Once fibroblast culture
nations, may become impossible to interpret, considerable
has been established, the cultures can be frozen and stored
information can still be obtained from blood, bile, urine
:n liquid nitrogen. With such appropriate storage, fibro
and vitreous humour.1 6J -1 65 As has been demonstrated for
blast culture can be re-established for many years.
many of the fatty acid oxidation defects, analysis of post
Should culture facilities not be immediately available,
mortem liver, even as long as J days after death, can still
[he skin biopsy itself should be frozen at -70°C and stored
yield helpful and diagnostic information. 14J As DNA tests
in that manner. It is notable that successful culture has
for more and more of these diseases become available, it is
been established from such stored tissue. 162
likely tha t t he need for such expeditious sampling may
diminish . Tandem mass spectroscopy, using blood spots on
newborn screening cards or bile, have also expanded the
Tissue for DNA Analysis potential diagnostic repertoire, as these a re quite stable a nd
can be mailed to a distant laboratory for analysis.
In any case in which there is a suspicion of an inherited dis The following list shows the recommended specimens in
ease, tissue should be obtained for potential DNA analysis. the majority of cases in which death has occurred many
If only limited facilities are available, blocks (approximately hours previously. Not all of these specimens will be needed.
5 g) of tissue (such as thymus, lymph nodes, spleen or liver) The autopsy findings and any clinical history will s hed
should be frozen at -70°C. An alternative to the direct stor light on the possible aetiologies and will guide their subse
age of tissue is to extract the DNA at the time of the autopsy quent use.
and to store the DNA. This probably results in better quality • urine (freeze 1-mL aliquots at -80' C);
DNA for later analysis. There are numerous standard proto - for organic acid analysis by gas
cols for the preparation of DNA from tissue samples. chromatography/mass spectroscopy;
- ketones;
30 Vege A, Chen Y, Opdal SH et al. Vitreous humo" 50 Schonheyder HC, Henriqu es U. Postmortem blood cultures.
hy poxanthine levels in SlDA and infectious death. Acta Eva luation of sepa rate sampling of blood from the right and
Paediatr 1994; 83 :634-9. left ventricle. APMIS 1997; 105:76-8.
31 Jones KL, Krous HF, Nadeau J et al. Vascular endoth elial 51 Bentley AJ. Zorgani AA, Blackwell CC et al. Bacterial toxins
growth factor in the cerebrospinal fluid of infa nts who die of and sudden unexpected death in a young child. Forensic Sci
sudden infant death syndrome: eviden ce for antecedent lilt 1997; 88:141-6.
hypoxia. Pediatr 2003; 111 :358-63. 52 Platt MS. Elin RJ, Hosseini JM, Smialek JE. Endotoxemia in
32 Putto A, Rllllskanen 0, Meurman 0 et al. C-reactive protein sudden infant death sy ndrome. Am J Forensic Med Pathol
in the evaluation of a febrile illness. Arch Dis Child 1986; 1994; 15:261-5.
61 :24-9. 53 Coe JI. Postmortem chemistries on blood: particular reference
33 Peltola H, Homberg D. Rapidity of C-reactive protein in to urea nitrogen. electrolytes and bilirubin. J Forensic Sci
detecting potential septicem ia. Pediatr InJ Dis 1983; 2 :374-6. 1974; 19:33-42.
34 Benjamin DR , Siebert JR. C-reactive protein and prealbumin 54 Naumann HN. Postmortem chemistry of the vitreous body in
in su spected sudden infant death syndrome. Pediatr Parhol man. Arch Ophthalmol 19 59 ; 62: 356-63.
1990; 10:503-7. 55 Coe JI. Postmortem chemistries on human vitreous humor.
35 Uhlin-Hansen L. C-reactive protein (CRP), a comparison of pre Am J C/in Patholl969; 51 :741-50.
a nd post-mortem blood levels. Forensic Sci hrt 2001; 14:32-5. 56 Swift PGF. Worthy E, Em ery JL Biochemical state of the
36 Tsokos M, Reichelt U, Jung R et al. Interleukin-6 and vitreous humour of infants at necropsy. Arch Dis Child 1974;
C-reactive protein serum level s in sepsis-related fatalities 49:680-5.
during the early postmortem period. Forensic Sci Int 200 I; 57 Sturner WQ. Dempsey JL. Sudden infant death: chemical
119:47-56. a nalysis of the vitreo us fluid. J Forensic Sci 1973; 18:12.
37 Fujita MQ, Zhu Bl, Ishid a K et al. Serum C-reactive protein 58 Blumenfeld TA, Blanc WA, Devlin J. Postmortem v itreou s
level s in po stmortem blood-an analysis with special reference humor concentrations of Na, K, CI, Ca and Mg in children.
to the cause of deat h and survival time. Forensic Sci Int 2002; Pediatr Res 1974 ; 8:356.
130:160-6. 59 Zumwalt RE. Hirsch CS. Subtle fatal child abuse. Hum Pathol
38 Chiesa C, Panero A, Rossi N et at. Reliability of procalcitonin 1980 ; 11:167 - 74.
concentrations for the diagnosi s of sepsis in critica lly ill 60 Emery JL, Swift PGF. Worthy E. Hyp ernatremia and uraemia
neonates. Clin I11J Dis 1998; 26:664-72. in unex pected death in infancy. Arch Dis Child 1974; 49:686.
39 Monneret G, Lab a une JM, Issac C et al. Procalcitonin and 61 Coe JI. Use of chemical determinations on vitreous humor in
C-reactive protein leve ls in neo natal infections. Acta Paediatr forensic pathology. J Forensic Sci 1972; 17: 54 1-6.
1997; 86:209-12. 62 Flear CTG, Gill GV Hyp onatremia : mechanisms and
40 Tsokos M, Reichelt U, Nierhaus A, Puschel K. Serum management. Lancet 1981; ii: 26 -31.
pro calc itonin (PCT): a valuable biochemical parameter for the 63 Coe JI. Further thoughts and observations on postmol1em
postmortem diagnosis of sepsis. Jnt J Legal Med 200 1; chemistry. Forensic Sci Gazeffe 1975; 5 :2-6.
114:237-43. 64 Coe JI. Vitreous potassium as a measure of the postmortem
41 Mei sne r M, Tschaikowsky K, Schanbel Set al. Proca lcito nin interval: an historical review and clinical evaluation. Forensic
influence of temperature, storage, anticoagulation and arterial Sci Int 1989 ; 42:201-13.
or venous asservation of blood samples on procalcitonin 65 Coe JI. Apple FS. Variations in the vitreous humor chemical
concentrations. Eur J C/in Chem Clill Biochem 1997 ; values as a result of the instrumentation. J Forensic Sci 1985;
35:597-60J. 30:828-3 5.
42 Yunginger JW, Nelson DR, Squillace Dl e t al. Laboratory 66 Sturner WQ. Dowdey ABC, Putnam RS. Dempsey JL.
investigation of deaths due to anaphylaxis. J Forensic Sci Osmol ality and other chemical determinati on s in postmortem
199 I; 36:857 -60. vitreous humor. J Forensic Sci 1972; 17:387-93.
43 Ani sa ri MQ, Zamora JL, Lipscomb MF. Po st mortem diagnosis 67 Devgun MS, Dunbar JA. Biochemical investigation of
of acute anaphylaxis by serum tryptase levels. Am J c/ill Parh vitreous: application in forensic medic ine esp eci ally in
1993; 99:101-3. relation to alcohol. Forensic Sci Jnt 1986; 31:27-34.
44 Randall B, Butts J. Halsy JF. Elevated pos tmo rtem tryptase 68 Hen ssa ge C. Krompech er T, Madea B et al. The Estimation oj
levels in th e absence of anaphylaxis. J Forensic Sci 199 5; the Time Since Death in the Early Postmortem Period. 2nd
40:208- I I. edn. London: Hodder Arnold, 2002.
45 Edsto n E. van Hage-Ha msten M. Beta -tryptase measurements 69 Schleyer F. Determinations of the time of death in the early
postmortem in anaphylactic deaths and controls. Forens ic Sci postmortem interval. {VIeth ods oj Forensic Science, vol. 2.
lnt 1998; 93:135-42. Chichester: Inte rscience Publ., John Wiley and Sons, 1963.
46 Schwartz HJ, Yunginger JW. Schwartz LB. Is unrecognized pp. 253-93.
anaphylaxis a cause of sudden unexpected death ? Clin Exp 70 Furuno J . Momura S. Inorga nic phosphorus content in human
Allergy 1995; 25:866-70. aqueous humor in rel ation to hours postmortem. Tohoku
47 Holgate ST. Walters C, Walls AF et a l. Th e anaphylaxis J Exp Med 1976; 119:293-5.
hypothesis of sudden infant death syndrome (SIDS): mast cell 71 Henry JB. Smith FA. Estimation of the postmortem interval by
degranulation in cot death revealed by eleva ted concentrations chemical means. Am J Forensic Med Patho11980; 1:341-7.
of tryptase in serum. c/in Exp Allergy 1994; 24: 1I 15-2 2. 72 Stephens RJ, Richards RG. Vitreous humor chemistry: the use
48 Edston E. Gidlund E, Wickman M et a!. Increased mast cell of po tassiu m concentration for the prediction of the
tryptase in sudden infant death: anaphylaxis, hypoxia or postmortem interva l. J Forensic Sci 1987; 32:503-9.
artifact. C/in Exp Allergy 1999; 29: 1648- 54. 73 Adjudanti s G. Co utselinis A. Estimation of the time of death
49 BuckJey MG, Variend S. Walls AF. Elevated serum by potass ium levels in the v itreous humor. Forensic Sci 1972 ;
concentrations of beta- tryptase, but not alpha-tryptase, in 1 :55-60.
Sudden Infant Death Syndrome (SIOS). A n investigation of 74 Madea B. Henssge C. Honig W, Gerbracht A. References for
anaphylactic mechanisms. C/in Exp Allergy 2001; determining the time of death by potassium in vitreous
31: 1696-704. humor. Forensic Sci Jnt 1989; 40:231-43.
122 I Biochemical investigations on post-mortem specimens
75 Saugsatd OD, Oliasen B. Postmortem hypoxanthine leve ls in 98 Sperling !VIA, Wolfson AR, Fisher DA. Congenital adrenal
the vitreous humour. Foren sic Sci Int 1978; 12:33-6. hypoplasia: an isolated defected of organogenesis. J Pedjatr
76 Rognum TO, Hauge S, Oyasaeter S, Saugstad OD. A new 1973; 82:444-9.
hioch emical method for the determina tion of postmortem 99 Batch JA, Montalto J , Yong ABW et a l. Three cases of
time. Forensic Sci Int 1991; 51 :131-46. congenita l adre nal hypopl asia; a cause of salt wasti ng and
77 Madea B, Herm a nn N, Hengbe C. Precision of estimating the mortality in the neonatal period. J Pediatr Child Hea lrh
time since death by vitreous potassium- co mparison of t'.NO 1991; 27:108-12.
different equations. Forensic Sci Int 1990 ; 46:277-84. 100 Molander N. Sudden natu ral death in later childhood and
78 James RA, Hoad ley PA, Sampson BG. Determination of adolescen ce. Arch Dis Child 1982; 57 :572-6.
postmoltem interval by sampling vitreous humour. 101 Favara BE, Franciosi RA, Miles V. Idiopathic adrenal
Am J Foren sic Med & Patl7 ol1 997; 18:158-62. hypoplasi a in children. Am J Clin Pathol 1972; 57:287-96.
79 Mason JK, Harkness RA, Elton RA, Bartholomew S. Cot deaths 102 Finlayson NB. Blood cortisol in infants and adults: A
in Edin burgh: infant feeding and socioeconomic factors. postmortem study. J Pediatr 1965; 67 :248-52.
J Epidemiol Commlln Health 1980; 34:35-41. 103 Sa nders FW. The preservatio n of blood for chemical analysis.
80 Coe JI. Postm ortem periph era l blood glucos e and J Bioi Ch em 1923; 58:1-15.
cardiopulmonary resuscitation. Forensic Sci Gaz 19 75; 6:1-2. 104 Paul JR. Postmortem blood chemical determinations. Bull
81 Gormsen H, Lund A. diagnostic value of postmortem blood Ayer Ciin Lab Pen Hosp 1925; 9:51-62.
glucose determinations in cases of diabetes mellitus . Forensic 105 Leahy MS, Farber ER. Postmortem chemistry of human
Sci Int 1985 ; 28: 103-7. vitreous humor. J Forensic Sci 1967; 12:214-22.
82 Peele t C, Picotte P, Jobin F. The use o f vitreous humor levels 106 Levonen E, Raekallio J, Saikkonen J. Postm orte m
of glucose, lact ic acid and blood levels of acetone to establ ish determination of blood creatinine and urea . J Forensic Sci
ante-mortem hyperglycemia in diab etics. Foren sic Sci Inr 1963; 10:22-9.
1994; 65:1-6. 107 Jensen OM. Diagnosis of uremia postmortem. Dan Med Bull
83 Coe JI. Comparative postmortem chemistries of vitreous humor 1969; 8(Suppl.):1-97.
befo re and after embalming. J Forensic Sci 1976; 21 :583-6. 108 Coe J1. Postmortem chemistries on human vitreous humor.
84 Bray M. The eye as a chemical indic ator of environme ntal Am J Clin Pathol 1969 ; 51 :741 - 50.
temperature at the tim e of death. J Forensic Sci 1984; 109 Pol ayes SH, Hersh ey E, Ledered M. Postmortem blood
29:396-403. chemistry in renal disease. Arch liltem Med 1930;
85 Coe JI. Hypothermia: auto psy findin gs and vitreous gl ucose. 46:283-9.
J Forensic Sci 1984; 29:395-8. 110 Bennett MR, Rinald o P, Millington DS, Tanaka K, Yokota I,
86 Chen C, Glagov S, Mako M et al. Postmortem glycosylated Coales PM. Medium-chain acyl-CoA dehydro ge nase
hemoglobin (HbAlc): evide nce for a history of diabetes deficiency; postmortem diagnosis in a case of sudden infa nt
mellitu s. Alln Ciin Lab Sci 1983; 13 :407-10. death a nd neonata l diagnosis o f an affected sib ling. Pediatr
87 Winecker RE, Hammet-Stabler CA, Chap man J F, Ropero-Miller Patlzol 1991 ; 11 :8139-95.
JD. HbA I c as a postmortem tool to identify glycemic contro l. 111 Bennet l\IlJ, Allison F, Pollitt RJ, Variend S. Fatty acid
J Forensi c Sci 2002; 47: 1373-9. oxidation defects as a cause of unexpected death in infa ncy.
88 Khuu HM , Robinson CA, Brissie RM , Konrad RJ. Postmortem Prog Clin Bioi Res 1990; 321 :349-64.
diagnosis of unsuspected diabetes mellitus established by 112 Duran M, Hofkamp M, Rhead WJ et al. Sudd en child death
determination of dec ede nt's hemo globin Alc leve l. and 'healthy' affected family members with medium- cha in
J Forensi c Sci 1999; 44:643-6. acyl- coe nzyme A dehydrogenase deficiency. Pediatrics 1986;
89 Goulle JP, Lacroix C, Bouige D. Glycated hemog lobi n: a useful 78:105 2-7.
post-mortem referen ce marker in determining diabetes. J 113 Chalmers RA, Stanley CA, English N, Wigglesworth JS.
Forensic Sci 2002; 128 :44-9. Mitochondrial camitine-acylcarnitine transl ocase deficiency
90 Osuna E, Garcia-Villora A, Perez-Carceles!VI et al. Vitreous presenting as sudd en neonatal death. J Pediatr 1997;
humor fructosamine concent rations in the autopsy diagnosis 131:220-5.
of diabetes mellitus. fnt J Legl Med 1999 ; 112:275-9. 114 Rinaldo P, Stanley CA, Hs u BY et al. Sudden neonatal death
91 Osuna E, Garcia-Villora A, Perez-C arcel es M et al. Glucose in carnitine transp orter deficiency. J Pediatl" 1997 ;
and lactate in vitreous humor compared with the 131:181-2.
determination of fructosamine for the post-mortem diagnosis 115 Di onis i- Vici C, Seneca S, Zeviani Met al. Fulminant Leigh
of diabetes mellitus. Am J Forensic Med Pathol 2001; syndrome and sudde n unexpected death in a family with the
22:244-9. T9176C mutation of the mitochondrial ATPase 6 gene.
92 DiMaio V JM, Sturn er WQ, Coe JI. Sudden and unexpected J Inherit Metab Dis 1998; 21:2-8.
deaths after the acute onset of diabetes mellitu s. J Foren sic 116 Ha rpey J-P, Charpentier C, Coude M et al. Sudden infant
Sci 1977 ; 22: 147-51. death sy ndrom e a nd multiple acyl-coenzyme A
93 Sturner WQ, DiMaio VJM. Sudden and unexpected death dehydrogenase deficiency, ethylmaionic-adipic aciduria or
following the acute onset of diabetes mellitus (Abstract). system ic carnitin e deficiency. J Pediatr 1987; 110:881-4.
Foren sic Sci 1975; 5: 168. 117 Lundemose JB, Kolvraa S, Gregersen N et al. Fatty acid
94 Rozin L, Perper JA , J a ffe R, Drash A. sudden unex pected oxidation disorders as primary ca use of sudden and
death in childhood due to unexpected diabetes mellitus. unexpected death in infants and yo un g children: an
Am J Forensic Med Patho11994; 15:251-6. investigation performed on cu ltured fibroblasts from
95 Russel MA, Opitz JM, Visekul C et al. Sudd en infant dea th s 79 children who died aged between 0-4 years. J Ciin
due to conge nital ad renal hypopla sia. Arch Pathol Lab Med Palhol Mol Patho11997; 50:212-17.
1977 ; 101:168-9. 1113 Emery JL, Va riend S, Howat AJ, Vawter GF. In vestigation of
96 O'Don ohoe l\rv, Holland PDJ. Fam i.lial con genita l adrenal inborn errors of metabolism in unexpected infant death.
hypopl asia. Arch Dis Child 1968; 43:717- 23. Lancet 1988; ji:29-31.
97 Jindrich EJ . Adrenal hypofullction and sudden death. 119 Hale DE, Benn ett MJ. Fatty acid oxidation disorders: a new
J Forellsic Sci 19134 ; 29:930-3. class of metabo.lic diseases. J Pediatr 1992; 121: 1-11.
References I 123
120 Ho wa t AJ, Bennet MJ, Variend S, Shaw L. Medium cha in 140 !jIst L, Ruiter JP, Hoovers JM et al. Common missense
fatty acyl -coenzyme A dehydrogenase pres en ting as the mut ation G 1528C in long-chain 3-hydroxyacyl-CoA
sudd en infant death. BM] 1984; 288:976. dehyd roge nas e deficiency. Characterization and expression
121 Howat AJ, Bennett MJ, Variend S et al. Defects in the of the mutant protein, mu tation analysis on geno mic DNA
metabo lism of fatty acids in the su dden infant death and chro mosomal localization of the mitochondrial
syndrome. BlVlJ 1985; 290:1771-3. trifunct ional protein alph a subunit gene.] Clin Invest 1996;
122 HollOn JB , Allen JT, Green CA et a l. Inherited metabolic 98: 1028-33 .
disease in sudden infant death sy ndrome. Arch Dis Child 141 Rashed MS, Ozan d PT, Bennett MJ et al. Inborn errors of
1991; 66:13)5-17. metabolism dia g nosed in sudden infant death case by
123 Miller ME, Brooks JG , Forbes N, Insel R. Frequency of acylcarnt ine analysis of postmortem bile. Clill Chem 1995;
medium-cha in acyl-CoA dehydrogenase G-985 mutation in 41:1J09-14.
sudden infan t death syndrome. Pediatr Res 1992; 31:305-7. 142 Hale DE, Cornell JE, Bennett MJ. Stability of lon g- chain and
124 Chinsky J, Tolsm a T, Cowan T, Blitzer M. MCAD deficiency short-chain 3-hydroxyacyl -CoA dehydroge nase activity in
in SIDS. Am] Hum Gen 1991 ; 49(Suppl.):AI83. the postmortem li ver. Clin Chem 1997 ; 43:273 -8.
125 Arens R, Gozal D, Jain K et al. Prevalence of medium-chain 143 Boles RG, Martin SK , Blitzer MG, Rin a ldo P. Bio chemical
acyl-CoA dehydrogena se deficiency in sudd en infant death diagnos is of fatty acid oxid atio n disord ers by metabolite
syndrome.] Pediatr 1993; 122:71 5-18. analysis of postmortem li ver. Hum Pathol 1994;
126 Lemieux B, Giguerre R, Cyr D e t al. Sc reenin g urine of 3 25:733-4.
week old newb orns: la ck of associ at ion betw ee n sudden 144 Gregersen N, Blakemore AIF, Winter V et al. sp eci fic
infant death synd rome and some metabolic disorders. diagnosis of medium-chain acyl-CoA de hyd rogenase
Pediatrics 1993 ; 85:986-8. (MCAD) deficiency in dried blood spots by a polymerase
) 27 Bonham JR, Dow ning M. Metabolic deficiencies and SIDS. chain reaction (PCR) assay detecting a point mutat ion (G9 85)
] Clin Pathol 1992; 45(Suppl.):33-8. in the MCAD gen e. Clin Chim Acta 1991 ; 203 :23-4.
128 Bennett MJ, Powell S. Metabolic disease and sudden, 145 Matsubara Y, Na risawa K, Tada K. Medium-chain acy l-CoA
unexpec ted death in infancy. Hum Pat/1011994; 25:742-6. dehydrogenas e deficiency. Molecular aspects. Eur] Pedia tr
129 Roe CR, Millin gto n DS, Maltby KF. Recognition of med ium 1992; 151:154-9.
chain acy l CoA dehydrogenase deficiency in asy mptomatic 146 Ding J-H , Roe CR , lafolla AK, Chen Y- T. Medium-chain
siblings of children dy ing of s udden infant death or Reye acy l-CoA dehydrogenase deficiency and sudden infant
like syn drome.] Pediatr 1986; 108:13-18. death (Letter). N Engl] Med 1991; 325: 6 1-2.
130 Rebuffat E, Sottiaux M, Goyens D et al. Sudd en infant death 147 Kelly DP, Hale DE, Rutledge SL et al. Molecular basis of
syndrome, as first expression of a metabolic disorder. In inherited med ium-cha in acyl-CoA dehydrogenase deficiency
Sch aub J, Van Hoof F, Vis HL (eds) Inborn Errors of ca using sudden child death.] Inherit Metob Dis 1992 ;
Metabolism. New York: Vevey/Raven Press, Nestle Nutri tion 15:171-80.
Workshop Series 24, 1991, pp. 71-80. 148 Hallock J, Morrow G, Karp LA, Barness LA . Postmoliem
13 1 Bennett MJ, Hale DE, Coates PM, Stan ley CA. Postmortem diagnosiS of metabolic disorders. The findings of maple
reco gnition of fatty acid ox idation disorders. Ped iatr Pathol syrup urine disease in a cas e of sudden and unexp ected
1991 ; II :365-70. dea th in infancy. Am] Dis Child 1969; 118:649- 5 1.
132 Bennet MJ, Allison F, Pollitt RJ, Variend S. Fatty acid 149 Byard RW, Cohle SD. Sudden Death ill Infancy, Childhoo d
oxid ation defects as causes of unexp ected death in infancy. and Adolescence. Camb ridge: Cambridge University Press,
In Tanaka K, Co ates PM (eds) Fatty Acid Oxidation: Clinical, 1994.
Biochemical and Molecular Aspects. New York: AJan R Li ss, 150 Clayton PT, Hy land K, Brand M, Leonard N. Mitochondrial
1990, pp. 349-64. phosphoenolpyruvate ca rboxykinase deficiency.
J 33 Blau N, Duran M, Blaskovics ME (eds) . Physicia n 's Guide to Eu r] Pedia tr 1986: 145:46- 50.
the Laboratory Diagnosis of Metabolic Diseases. London : 151 Norman MG , Taylor GP, Clark e LA. Sudden, unexpected,
Chapman and Ha ll Medical. 1996. natural death in childhood. Pediatr Patlwl 1990; 10:7 69 -84.
134 Scriver CR, Beaudet AL, Sly WS, Valle D (eds). The Metab olic 152 Burton BK, Roach ES, Wolf B, Weissbecker F. Sudd en dearh
and /VIolecular Bases of Tnh erited Disease. New York: associated w ith biotinid ase deficiency. Pediatrics 1987;
McGraw -Hili, 1995. 79:482-3.
135 Applegarth DA, Dimmick JE, Hall JG (eds). Organelle 153 Danks DM, Campbell PE, Stevens BJ et at. Menke's kinky
Diseases. London: Chapman and Hall Medical, 1997. hair syndrome. An inherited defect in copper metab olism
136 Kemp PM, Little BR , Bes t RO, Dawson DB. Whole blood with widespread effects. Ped iatrics 1972; 50: 188-20 t.
leve ls of dodecanoic acid, a routinely detectable fore nsic 154 Forres t AR (ACP Broadsheet no. 137, April 1993). Obtaining
marker for a gen etic disea se often misd iagnos ed as samples at postmoliem examin a tion for toxicological and
Sudden Infant Death Syndrome (SIDS): MCAD deticiency. biochemical analyses. ] Clin Pathol 1993 ; 46: 292-6.
Am] Forensic NIed Path 1966; 17:79-82. 155 Pounder DJ , Carson DO, Johnston K, Orihara Y. Electro lyte
137 Ross KF, Guileyardo JM, Bennet MJ, Ba rnard JJ. Comment concentration differenc es between the left and right vitreous
on wh ole blood levels of dod eca noic acid, a routin ely humor sampl es. ] Forensic Sci 1998; 43:6 04 -7.
detectabl e foren sic marker fo r a genetic disease ofte n 156 Bennett MJ, Ragni MC, Hood 1, Hale DE. Comparison of
misdiagnosed as Sudden Infant Death Sy ndrom e (SlD S): postmortem vitreous and urinary organic acids. Ann Clill
MCAD deficiency. (Letter). Am ] Forensic Med Path Biochem 1992; 29:541-5.
1966;17:349-50. 157 Pallick W JA, Logan RW. Free amino acid content of the
138 Blakemore AJF, Singleton H, Politt R et at. The frequency of vitreous hum or in cot deaths. Arch Dis Chil d 1988; 63:660-2.
the G985 MCAD mutation in the gen eral populati on. Lancet 158 Chong APY, Aw SE. Postmortem endocrine levels in the
1991; 337:298-9. vitreo us hum our. Ann Acad Med (Singapore) 1986; 15 :606-9.
139 Coates PM. Histori cal perspecti ve of medium chain acyl-CoA 159 Hill E. Significance of dex trose and nondextrose reducing
dehydrogena se deficiency. A decade of discovery. Prog Clin substa nces in postmortem blood. Arch Pat/IOI 1941;
BioI Res 1992; 375:40 9-23. 32: 4 52-73.
124 I Biochemica l investigati ons on post-mortem spec ime ns
160 Lythgoe AS. The act ivity of lactic dehydrogenas e in ca daver Beny J, Alibon e E, McKeever P et al. The contribution of ancillary
sera: a comparison of different sampl ing sites. Med Sci Law pathology tests to the in vestigation of sudden infant death. In
1980; 20:48-53. Fleming P, Blair P, Bacon C et al reds} Sudden Unexpected Dea ths
161 Zhang DJ, Elswic k RK, Mill er WG, Bailey JL. Effect of ill InJancy. The CESDI SUDI Studies. Norwich: The Stationery
serum-clot contact time on clinical chemistry laboratory Office, 2000, pp. 97 -112.
results. Clin Cliem 1998; 44:1 325-33. Bla u N, Duran M, Blaskovics ME (eds). Physician's Guide to the
162 Fowler KJ . Storage of skin biopsies at -70°C for futu re Laboratory Diagnosis oJ M etabo lic Diseases. London: Chapman
fibroblast cu lture. J Ciin Patlioil984; 37:1191-3. and Hall Med ical, 1996.
163 Bennet MJ , Rinaldo P. The metabo lic autopsy comes of age. Brettell TA, Safers tein R. Forensic science. Anal Chem 1995;
Ciin Chem 200 1; 47 :11 45-6. 67:27 3R- 94R.
164 Wilcox RL, Nelso n CC, Stengel P, Stei ner RD. Postmortem Byard RW. Sudden Death in inJancy, Childhood and Adolescence, 2nd
screening for fat ty acid oxid atio n disorders by analysis of edn. Cambridge : Cambridge University Press, 2004.
Guthri e cards with tandem mass spectrometry in sud den Coe J1. Postm ortem chemistries on human vitreous humor.
unexpected death in infancy. J Pediatr 2002; 141 :833 -6. Am J Clin Pathol 1969; 51:741 - 50.
165 Chace DH, DiPerna JC, Mitchell BL et al. El ectrosp ray Coe J1. Postmorte m chem istlY of blood, cerebrospinal fluid and
tandem mass spectrometry for analysis of acylca rnitines in vitreous humor. In Ted esc hi CG, Eckert WG, Tedesc hi LG (eds)
dried postmortem blood spec imens collected at autopsy from Foren sic Medicine, vol. 2. New York: WB Saunde rs, 1977.
infants with unexplained cause of death. Clin (hem 2001; Coe J1. Post mortem biochemistry of blood and vitreous hum our in
47: 1166- 82. paediatric prac tice. In Mason JK [ed.} Paediatric Forensic
166 Karke la JT. Critical evaluation of postmortem changes in 1\1edicine and Pathology. London: Chapman and Hall, 1989, pp.
hum a n autopsy cistern al fluid . Enzymes, elect rolytes, acid 191- 203.
base balance, glucose and glycolysis, fre e amino acids and Coe J1. Post mortem chem istJy upd ate. Emph asis on fore nsic
amm onia. Con-elation to total brain isch emia. J Forensic Sci appli catio n. Am J Forensic Med Pathol 1993; 14:9 1-117.
1993; 38 :603- 16. Scriver CR, Beaudet AL, Sly WS et al (eds). The Metabolic and
Molecular Bases oJ Inh erited Disease. New York: McGraw-Hili,
1995.
Further Reading
NON-ACCIDENTAL INJURY
Harry Willshaw
INTRODUCTION
SCOPE OF OCULAR AND ADNEXAL INJURY
_\11 abused child is most often the victim of damage Fundus changes are often considered the hallmark of ocu
'n flicted by a parent or carer. Although both self-mutila lar involvement in NAl, but in fact any ocular or adnexal
TIo n and Munchausen's syndrome by proxy do occur in structure may be damaged.
children, I they are rarities in the setting of child abuse
globally.
The practice of child abuse was first challenged legally Eyelids
in New York in 1871 2 using legislation written to prevent
cruelty to animals. Since that time the incidence of child Swelling, laceration and burning of the eyelids have all
abuse, or at least its identification, has risen to such an been described. 12 Lid bruising and swelling are relatively
ex te nt that in 1962 it was considered the major cause of common events in a young life and should not be
child death and maiming in the USA. 3 overinterpreted, unless they are seen in combination with
The frequency of ocular involvement in non-accidental other injuries that are suggestive of abuse. In children
injury (NAI) has been variously reported at between 5 and under the age of 12 years, lid injuries are most commonly
6 1 per cent. 4 ,5 Eye damage is usually seen in association caused by implements. The characteristic pattern of belt
':;ith head and facial injuries. 6 In reported series in which buckles etc. may be obvious in the injured site. Typically,
- e injuries resulted in death, the rate of ocular involve after the age of 12, the injuries are inflicted with fists and
ment is high/- 9 suggesting that ocular involvement is a feet. 5
sig nificant marker for severe neurological involvement. Burns to the eyelids are an occasional feature of ab use.
Fatal neurological involvement is particularly common in Unlike their accidental counterpart, non-accidental facial
'ery young victims; all of 13 fatalities attributable to shak burns tend to be deep and multiple. At 60°C, heat must be
in g reported in 1987 were of children aged 24 months or applied to skin for at least 3 seconds to produce even a
less . 10 first-degree burn,13 so that multiple deep cigarette
A proportion of children will suffer life-long visual burns, as seen in Figure 6.1, are not compatible with the
andicap as a consequence of NAl. The likelihood of long child having inadveliently brushed against the hand of a
lerm visual sequelae from a n injury is usually directly smoker.
re lated to the extent of ocular damage; however, it must be Occasionally more exotic agents of abuse are applied to
remembered that some children are blinded by their neuro the lids. Recently, a case has been reported in which super
logical injuries with no evidence of persisting ocular glue was deliberately applied to the eyelids, 14 resulting in a
pathology. II short period of permanent lid closure.
126 l Ocular involvement in non-accidental injury
Figure 6_1 Cigarette burns on the eyelids - inflic ted injury. Figure 6.2 Bilateral subco njunctival haemorrhage and
periocular bruising following assault.
External Eye syste mic investigations are required. This is palii cularly so
for lens subluxation, in which investigations must exclude
Subconjunctival haemorrhage is ano ther rela tively co mmon Marfa n's syndrome, homocystinuria, sulphite oxi dase defi
finding (Fig. 6.2) associated with facial IDJUlY or ciency and Weil-Marchesani syndrome.
Valsalva manoeuvre, and does not in itself indicate When lens damage does occur, it may be possible with
deliberately inflicted damage. On the other hand, persisting the slit lam p to detect small tears in the anterior lens cap
corneal damage is a great rarity in childhood. In otherwise sule. These tears allow t he ingress of aqueous humour into
healthy children with no cause fo r chronic corneal exposure the protein of the lens cortex leading to delayed cataract
(e.g. previous ptosis surgery), corneal epithelial defects usu formation. The resultant cataract may not become clinic
ally settle quickly. Persisting epithelial damage may well be ally app arent for 4 or 5 days after the original injury.
caused by chemical in stillat ion into the eye. This is particu
larly likely to involve the epithelium of the lower hal f to
one-third of the cornea and conjunctival cui de sac. This is Posterior Segment
because an intact Bell 's phenomenon causes t he eyes to roll
upwards as the lids are pulled apart to allow instillation. Features of NAI in the posterior segment are dominated
We have treated a child with unexplained bilateral by the finding of fundus haemorrhages. However, retinal
corneal clouding accompanied by total epithelial loss. The detachments can also occur. These may be caused by giant
corn ea s healed well when the child was isolated from her retinal tears 18 or vitreous base avul sion 19 and may be
parents, and her father subsequently admitted to repeatedly apparent immediately after the injury has occurred.
spitting methadone into the child's eyes. Any type of harmful Because the vitreous gel is rel atively firmly adh erent to the
material may be instilled into the eye - no paliicular sub retina over the peripheral portion (the ora serrata), violent
stance is more common than oth ers. In 1987 the app licatio n movements of the vitreous gel exert strong tractional
of hot peppers (capsicum and capsaicin) to the conjunctival forces on that area of the retina. As a consequence, the
fornices was reported as a type of child abuse. 15 retina may tear over more than 90 0 of its periphery, or the
vitreous base may tear away.
Alternatively, detachments may present later because of
Anterior Segment traction from the condensation of traumatic vitreous haemor
rhage. In one study, post-mortem examination revealed what
The presence of anterior seg ment ocular injUlY, in the form have been called haemorrhagic retinal detachments in up to
of iris or lens damage, tends to suggest severe impact 63 per cent of case. 20 In that particular report, they occurred
trauma and carries a relatively poor visual prognosis (Fig. primarily in the retinal periphery and, though they pathologic
6.3). Pupil sphincter rupture (which manifests as slight ally appeared as retinal detachments, were seen ophthalmo
irregularities in th e pupil and tears at the pupil margin), scopically as accumulations of sub retinal blood. Peri macular
hyphaema , recession of the drainage angle, lens sublux folding of the retina (Fig. 6.4), seen either at the time of firs t
ation and cataract leading to glaucom a have all been assessment or at later fundus ex amination, is extremely sug
described in a single child. 16 These events usu ally follow gestive of NAJ. The folds can result from either direct head
direct injury to the gl ob e,17 but can be seen after the rapid trauma or violent shaking, but are almost unique to the eyes
deceleration caused by an impact injury. of injured children. 21 .22
As with all of the ocular features of child abuse, the Late fundus manifestations of abuse include hole for
changes seen are not pathognomonic of NAJ and appropriate mation in the retina, cysts, gliosis and scarring. 2J Features
Scope of ocular and adnexal injury I 127
Visual Pathways
PR E- RETINA
fin ally, blood may break through the posterior hyaloid face (whiplash shaken baby) are substantial, but they increase
and give rise to a localized or diffuse vitreous haemor greatly with head impact against a solid surface. These forces
:hage. Vitreous haemorrh age is thought to happen 2-3 then give rise to a number of contributolY events:
days after the initial injury,37 a feature that may have con • Blood vessels within the retina may shear because of
siderable forensic significance. The vitreous will usually the relatively firm adherence between the vitreous and
clear spontaneously, but the process can be slow, taking the retina in yo ung children. The violently moving
several months. Since these injuries typi cally affect chil vitreous transmits its mom entum to the layers of
dren in the sensi tive period for visual development,38 the the retina, tending to tear the retina and give rise to
visual outcome following such severe bleeding is often bad splits within it (schisis) and blood-filled cystS. 45
not only because of structural damage to the eye, but also Post-mortem examination of 190 eyes 46 demonstrated
because of stimulus deprivation amblyopia. 39 Studies on that bleeding, even when it extended in to all layers of
children with other causes of stimulus deprivation have the retina, invariably arose from the vessels in the
shown t hat relativel y short periods of asym metric interfer inner layers . This tends to support the view that
ence with visual input, at the right time in life, can lead to tangential forces are responsible for retinal shearing
significant amblyopia. 40 Figure 6.7 illus tra tes the distribu and bleeding.
tion of haemorrhages within different layers of the retina. • Raised intraocular venous pressure results from
No particular pattern of fundus haemorrhage is exclu either raised intrathoracic pressure or raised pressure
sive to NAI, but it is frequentl y observed that the presence within the optic nerve sheath. In either case, venous
of bleedi ng at all depths of the retina, throughout 360 0 of stasis results, and subsequently retinal bleeding. There
the retina and extending from the posterior pole to the ora are several potential causes of raised intraocular
serrata, is strongly suggestive of bleeding caused by the venous pressure in an abusive situation.
severe acceleration and deceleration forces seen in
NAI. 41 - 43 If such extensive haem orrhages are seen in com A Purtscher type of retinopathy secondary to chest com
bination with peri macular folds 44 or haemorrhagic retinal pression has been postulated as the cause of retinal haemor
cysts,45 then anything but the most severe accidental injulY rhage in child sexual ab use 47 or when an excessively firm
can almost certainly be excluded as the cause. grip is taken of the infant chest during shaking. Chest com
pression may well be a contlibutory factor in the re tinal
haemorrhages commonly found in the neonate after normal
Mechanisms vaginal delivery48,49 (see later). Raised intratlloracic pressure
during cardiopulmonary resuscitation (CPR) also has been
The mechanism of blood vessel damage during NAI remains causatively linked with retinal haemorrhage. A number of
debatable, and it may well be that several different mecha individual case reports testify to the possibility of fundus
nis ms all contribute. The acceleration/deceleration forces gen haemorrhage following CPR, but large prospective studies
erated by the vigorous shaking of an unsupported infant head have failed to validate that association (see later) .
130 lOcular involvem ent in non-accidental inj ury
Bleedin g into the optic nerve sheath a nd th e perip apil have detailed bleeding following accidental trauma. In
lary sclera is a commonly reported feature of the more one case, a 13-month-old boy developed subdural haemo r
seve re cases of NAl. Eleve n of s ixtee n children dy ing from rhage and intra- and pre- retin al haemorrhage follow ing a
the central nervous syste m (CNS) complications of NAl fall down 13 concrete steps. The fundus haemorrhages
showed perineural bleeding. The bleeding was into the dis were unilateral and clea red within 3 months. In the second
ta l portion of the optic nerve, indicating that it was not case, a 9-month-old boy fell 1-2 feet, striking his head on
simply forward extension of th e accompanying subdural the floor, and sustained bilate ral reti nal haemorrhage in
haemorrhage. 20 It is entirely possible that perineural bl eed associ at ion with a subdural h aemo rrh age. In the third case,
ing raises the pressure within the optic nelve sheath, thus a 7-month-old gi rl fell through a stair rail onto a concrete
ten ding to impede ven ous return, wi th fundu s bleeding basement flo or and again sustained both subdural and
seco nda ry to the ve nous co ngestion. (unilateral) retinal haemorrhage.
Simila rly, raised intrac ranial pressure with pap illoedema These repo rts serve to emphasize that no ocular feature
is a recog ni zed cause of retinal h ae morrhage with a sim ilar can be considered pathognom onic for NAl. None the less,
mechanism. There is a close correlation between severe in la rge selies, accidental injury has almost a lways seemed
CNS injury and prominent fundus bleeding. 50 insuffic ient to cause retinal bleeding. For example, of 79
A third possible source of fundus h aemorrh age, particu under-3-year-olds admitted to the Sick Children's Hospita l in
larly in the presence of subarachnoid haemorrhage, is Toronto with head inj ury, none of the 75 who had experi
Terson's synd rome, 51 in which subhyaloid blood is seen enced accid ental trauma had fundus haemorrhages S8 Th e
shortly after the development of a subarachnoid haemor three with NAl and the one whose injUlies were of un certain
rhage. Paton 51 postulated that the subarachnoid blood, under cause all had fundus haemorrhages. Similar reports have
pressure, passes forward through the la mina cribrosa to e nter been provided from other instituti ons,59,50 indicating that,
the eye, but a mechanism rel ated to raised intrasheath excluding high-speed road traffi c accidents and falls from
pressure is prob ab ly more likely. Examinati on of children extreme heights, accid ental injUly is most unlikely to provoke
w ith intracrani al haemorrhage from a variety of causes other retinal bleeding. The summary view ex pressed in a recent
than NAl showed that associated retinal h ae morrhage is Roya l College of Ophthalmologists Working Party ana lysis
extremely rare.52 was that 'No absolute values can be given for the a ngula r
acceleration forces required to produce retinal bleeding or
other injUly, but th ere is good evidence that they must be
Forces Required to Cause Bleeding considerabl e'.51
- f of the children with subarachn oid h2e morrhage the severity of retinal bleeding and the associated neuro
wed fundus bleeding as well. This study also showed logical injuries. 50 However, t hough CNS and ocular haem
:::-.a- me babies who had been severely shaken were more orrhage commonly coexist, it is quite possible for either to
d_ - to show multilayered retinal haemorrhages than were occur independently. A 1991 study 63 fou nd that 39 per cent
- _0 e who had suffered blunt trauma. of subdural haemorrhages had no associated ocular haem
One child (subsequently considered to have suffered orrhage, and that seven per cent of children with retin al
- m SIDS) who underwent CPR had a few red cell s in bleeding had no evidence of CNS bleeding.
- e anrerior optic nerve - a findi ng in sharp contrast to
- e extensive bleeding associated with acceleration/
_C{'eleratio n injuries. Whatever the type of injury, cerebral Differential Diagnosis of Fundus Haemorrhages
ema was found in all fatally injured children.
A similar study from Sheffield 20 examined the eyes and Since no ocular pathology is unique to NAl, it is essential that
brai ns of 23 fatally injured children; 16 of these children had the clinician excludes other possible causes of the observed
ead injuries, while the remainder died from suffocation, clinical signs. This needs to be done as quickly as possible so
abd ominal injUlies or other injuries. The children were all that appropriate steps can be taken, whatever the outcome of
Wlder 3 years of age. The authors constructed a 'total eye investigations. A comprehensive list of diagnostic differentials
ore' based on the presence of haemorrhagic retinal detach has been produced by a Working Party ofthe Royal College of
menr. retinal and optic nerve bleeding. They compared this Ophthalmologists G1 ,64 (Table 6.1). Some of these will be obvious
-',-ith an 'intracranial score' based on the extent of the following initial assessment whilst others demand particular
inn-acrani al bleeding, cerebral lacerations and evidence of diligence on the part of the supervising clinici an.
pri mary axonal injury. Though cerebral oedema was com Alternative explana tions that need to be consid ered
mo nly reported from im aging, they found it difficult to iden includ e accidental inju ry, CPR, epileptic seizures, arousal
. pathologically and, therefore, unlike the Philadelphia shak ing, bleeding disorders, bleeding secondalY to infection,
dy, did not include it in the score. genetic metabolic disorders and birth-associated retina l
The authors found a close relationship between the haemorrhage.
-everity of the ocular and t he cerebral injuries, and felt able
to deduce a severity sequence for the damage suffered in
ACCIDENTAL INJURY
_-Al. They suggested that with increasing force of trauma
:0 the head a ch il d is likely to suffer: A hi stO lY of accidenta l injury or injury associated with
• first, subdural haemorrhage, followed by subhyaloid, attempts to aro use or resuscitate is commonly cited to
intraretinal and then perineural optic nerve sheath account for neurological and ocular damage. As discussed
haemo rrh ages ; above, accid ental injury, except the most severe, is almos t
• second, haemorrhagic retinal detachment (more never the cause of significant cerebral or ocular bleeding.
accurately subretin al bleeding);
• third, choro id al and vitreous haemorrhages, which
coincide with subarach noid haemorrhage, intracerebral Table 6.1 Differential diagnosis of retinal haemorrhages in
haemorrhage and cerebral laceration. children 61 •64
As in the case of accidental injury, isolated case reports Arousal shaking is always a difficult area, with many practi
exist to suggest that vigorous CPR, particularly if per tioners considering it inconceivable that the forces required
formed by inexperienced practitioners on very young to cause intracranial and ocular bleeding could be generated
children, may be responsible for retinal haemorrhage. 65 in a reasonable attempt to arouse an unconscious child.
The infant in Kramer's case report was fully documented, Nevertheless, it is a defence used in court with sufficient fre
with fundus examination prior to the resuscitation quency for the British Medical Journal to have published an
attempt showing there had been no bleeding. This, there article warning carers of the possible dangers of baby shak
fore, represents one of the few cases in which the bleeding ing and an admonishment to exercise caution. 73
unequivocally followed the CPR rather than being present
prior to instigating CPR. This particular child received
BLEEDING DISOR DE RS
60 minutes of vigorous CPR, after which she was found to
have intraretinal and pre-retinal haemorrhage out to the Various types of bleeding disorder have been shown to give
mid-periphery of the retina. rise to fundus haemorrhages. Lymphoblastic leukaemia, in
Set against such individual reports, prospective series particular, is recognized to cause widespread retinal bleed
of paediatric patients and experimental studies on piglets ing, involving all retinal layers. This occurs particularly
have shown retinal haemorrhage secondary to CPR to be when the children are thrombocytopenic. 74 Sickle cell dis
extremely unlikely. ease, aplastic anaemias, Henoch-Sch6nlein purpura and
In a large clinical series, Kanter found only a single small vitamin deficiencies are all capable of producing retinal
haemorrhage in 40 children requiring CPR. 66 He concluded haemorrhage and must be excluded before pursuing a
'When retinal haemorrhage is detected in the paediatric diagnosis of NAl.
patient after CPR, prior trauma should be assumed'. More
recently, 169 post-mortem ocular examinations were
performed in children after failed CPR. 67 Sixty-one of BLEEDING SECONDARY TO INFECTION
the children had retinal haemorrhage after CPR, of whom Meningococcal meningitis was reported in 1995 as a cause
56 had suffered head injuries and four had CNS disease or of fundus bleeding?5 In our hospital, we have seen a child
sepsis; in only one case was cause of death undetermined. with massive, unilateral retinal haemorrhage (associated
No case was found with retinal haemorrhage in which CPR with fatal meningococcal disease) in whom the provisional
was the only explanation for the bleeding. In a prospective diagnosis was NAI and the true diagnosis was revealed
study of 43 children admitted to hospital with non only by post-mortem studies.
traumatic problems and requiring resuscitation, small punc
tuate retinal haemorrhages were seen in a single case. 68
Six newborn piglets receiving CPR for up to 50 minutes GENETIC METABOLIC DISOR DERS
with monitored intrathoracic and intracranial venous pres
Although individually rare, a number of genetic metabolic
sure showed no evidence of retinal bleeding.69
disorders (GMDs) have been shown to cause retinal bleed
ing and must be actively excluded in the evaluation of NAl
victims. Galactosaemia may cause widespread bleeding
EPI LE PTIC SEIZURES
into both retina and vitreous humour, particularly when
In our hospital, we examined a series of children admitted associated with significant liver dysfunction?6 Less com
with convulsions. 70 The aetiology of the seizure disorder mon GMDs have attracted attention in recent years, since it
varied but none was traumatic. Even though 40 of the chil became clear that glutaricaciduria could give rise to both
dren were less the 2 years of age and, therefore, in the most retinal and subdural bleeding along with cerebral atrophy.
vulnerable age group, none showed any evidence of fundus Children presenting with both subdural and intraretinal
bleeding. The children examined were all seen within 24 bleeding associated with glutaricaciduria were reported as
hours of admission, and examined in detail with an indir long ago as 1987. 77 These reports further emphasize the
ect ophthalmoscope using pupil dilatation in most cases. need to exclude metabolic causes in all suspicious cases.
In a more recent prospective study of 143 children aged Autosomally recessive protein C deficiency can also be
between 2 months and 2 years who were seen in an acci responsible for intracranial (usually subarachnoid) and
dent and emergency department following seizures and intraocular bleeding. Vitreous bleeding with severely
examined by an ophthalmologist, unilateral retinal haem reduced protein C levels has been reported at birth in a
orrhages was documented in only one child. 71 child of 37 weeks' gestation. 78
As with resuscitation attempts, despite individual case Other entities, such as hypertension, prematurity and
reports72 it should be assumed that, if retinal haemorrhage AlDS-related cytomegalovirus (CMY) infection, should be
is found in a child admitted with fits, the likelihood is that apparent from the history and examination of the child on
injury caused both the fits and the fundus bleeding. admission to hospital.
Fundus haemorrhages I 133
Document all fin dings (including negati ve fin dings) in notes. In clude time
BIRTH -ASSOCIATED RETINAL HAEMORRHAGE
and date of examination. Sign examination record
Retinal haemorrhages are found in the first few days after
delivery in up to 59 per cent of children. 4B The haemorrhages
Do not dilate pupils without discussion with supervising
clear quickly and, of 1238 children reported in 1970,49 18.9 pa edia tri cia n/i nte nsivis t
per cent examined in the first 24 hours showed signs of
having had a retinal haemorrhage. When examination was
delayed until 3-5 days after birth, only 2.6 per cent showed If dilatation not possible, record this limitation in notes
evidence of fundus bleeding, presumably indicating that
most of the haemorrhages clear in the first few days of life.
Notify consultant of findings - immediately if positive
Recent studies found retinal haemorrhages in about one
iliird of babies following spontaneous vaginal delivery and
in 75 per cent or more following vacuum extraction.79.BG Consultant review as soon as possible if positive findings
The incidence of ocular bleeding following Caesarean
de livery is lower, at around 7 per cent/ 9 ,BO suggesting that
the raised intrathoracic and intracrania l pressures associ Photography of fundus pathology if dilatation possible
ated with compression of the baby in the birth canal are
not the only aetiologica l factors . It has been specu lated Bl
chat hypoxia and hypercapnia are also contributory. Print off copies of fundus photographs
Examining the Child With Fundus A second copy - similarly validated - retained in eye department
Haemorrhages
- - - - -. - -
Specify date of review and ensure review undertaken
Though a paediatrician may be the first doctor to suspect
~etina l haemorrhages in a child, it is essential tha t an oph Figure 6.8 A guide to the examination of children with
j]almologist is involved at an early stage (Fig. 6.8). An eye suspected non-accidental injury.
assessment shou ld be requested on any child suspected of
being a victim of NAI, particularly if the head (including
CNS) and neck area are involved. The responsibility of the Once the appearances have been documented in detail,
ophthalmologist is to examine the whole of the retina the child shou ld be reviewed on a regu lar basis, both to
(where possible), and accurately document the distribution observe the changing appearance of the haemorrhages and
and depth of the retinal haemorrhages along with associ to determine whether any therapeutic intervention is likely
a ted optic nerve and retinal features. The tool of greatest to be needed. It is not possible to date an injury on the
':alue is an indirect ophtha lmoscope, w hich allows a view basis of retinal haemorrhages (in the way that skin bruising
of the retina right out to its peripheral extent. Such a view can be used).61 If, however, the appeara nces change with
can be achieved only through dilated pupils. Thus, if neuro time, knowing the ti mescale of the absorption of the
:ogical monitoring precludes pupillary dilatation at first observed haemorrhages may give a guide as to the timing
co ntact, the supervising team must be prepared to notify of their original appearance. Dating the time of haemor
rbe ophthalmologist when pupil dilatation is permissible. rhage with post-mortem specimens is more practical and
In some instances a short-acting mydriatic, such as tropi uses the appearance of haemosiderin. The presence of
(a mide, may be appropriate. However, in other instances, haemosiderin in the eyes of abused children has been con
!Ja n icularly in a child with dark-brown irides, a stronger sidered to indicate that an injury is at least 3 days 01d. 82
and longer-acting agent, such as cyclopentolate or To further aid the documentation of the retinal haemor
~h enylepherine, is likely to be necessary. There is no reason rhages, fundus photography may be useful. Traditional
i...n these circumstances to use agents that cause an effect fundus photography is unlikely to be possible, but photog
~-or several days, such as atropine or homatropine. raphy with a variety of hand-held cameras certainly will
134 lOcular involvement in non-accidental injury
be. The limitation on such documentation is related to the uncommon, except in the child who has also sustained
relatively small angle of view of most cameras and can be serious neurological injury.
overcome using a RetCam device (Massie Research Labora
tories, Inc., Dublin CAl. This recently developed digital
ADDITIONAL OBSERVATIONS
camera allows a uniquely wide, 1200 angle of view and is
combined with a telemedicine facility. 83 It is still being It is a surprising observation that, despite the nature of the
evaluated, but tools of this sort are sure to remove some of injury, strictly unilateral retinal haemorrhages may result
the uncertainty surrounding clinical descriptive recording. from severe shaking. In 1997 we reported three consecutive
When the view of the fundus is compromised by vitreous cases in which the bleeding was unilateral,25 and high
haemorrhage, imaging with ultrasonography can be coupled lighted the fact that such a presentation should not cast
with electroretinography (ERG) to define the location of the doubt on the diagnosis of NAl. Two of those three children
retina and its functional integrity. Fishman and coJJeagues 84 suffered severe long-term visual loss, despite the unilateral
found ERG to be unhelpful in the initial assessment of six nature of their ocular injury.
children with retinal haemorrhages. Their ERG parameters A review of the literature revealed strictly unilateral
were not significantly different from those in six age bleeding referred to in other series, and in those cases, too
matched control subjects. However, with time, one parameter the associated neurological injury was often severe. 52 .G3
improved, suggesting that some subtle neurophysiological The presence of unilateral eye signs does not diminish the
damage was caused by the injury. They also stressed the likelihood that the injuries are a consequence of abuse; nor
importance of ERG in distinguishing between neurological do they predict a more minor level of eye and neurological
and retinal visual loss at an early stage. In the presence of damage.
vitreous haemorrhage, ultrasonographic evidence of retinal Non-accidental injury is a harrowing topic for all clin
detachment and ERG evidence of impaired retinal function icians. We owe it to the children in our care to consider it as
should provoke early referral to a vitreo-retinal surgeon. a diagnosis to avoid the child being exposed to further
Clearing the vitreous blood and surgical relocation of a injury. However, we also owe it to the family to ensure that
detached retina will help long-term visual prognosis. Some the vista of NAI is only raised in appropriate circumstances.
authorities have also succeeded in clearing pre-retinal haem Detailed examination of the eye and modern documen
orrhage using intravitreal injections.85 All such procedures tation wiJJ contribute to the diagnosis, but none of the ocu
need to be considered when the child 's long-term visual lar features can, in isolation, establish a diagnosis of NAI .
development is likely to be compromised . Despite recent reports, domestic accidents rarely, if ever,
cause the level of ocular damage that is commonly associ
ated with severe shaking or impact injuries.
LONG-TERM VISUAL SEQUELAE
monly only one eye was involved . Significant visual victims: autopsy findings. J Forensic Sci 1991 ; 36:741-7.
impairment is difficult to divine from this series, but 8 Rao N, Smith RE , Choi JH et al. Autopsy findings in fourt ee n
may have been limited to one child with cortica l visual fatally abused children. Forensic Sci 1I1t 1988; 39:293 - 9.
impairment. 9 Hadley MN, Sonntag VKH, Rekate HZ, Murphy A. The infant
whiplash-sh ake injury syndrome: a clinical and pathological
It is clear from these studies that, despite the potential
study. NeurosLiTgelY 1989; 24:536-40.
for visual loss secondary to macular folds, retinal detach 10 Duhaime A- C, Gennarelli TA, Thibault LE et al. The sh aken
ment, persistent vitreous haemorrhage and other structura l baby sy ndrome. A clinica l, pathological and biomechanical
injuries to the eye, such visual loss is in fact relatively study. J Neurosurg 1987; 66:409-15.
References I 135
11 Jacobi G. Damage patterns in severe child ab use with and 38 Va ega n, Taylor D. Critical peri od for deprivation amblyopi a in
without fatal seq uelae. Monatssch r Kinderl1eilkd 1986; ch ild ren. Trall s Ophthalmology Soc UK 1979; 99 :432-9.
13 4: 307-15. 39 Mathews GP, Das A. Dense vitreo us hem orrh ages predict poor
12 Harcourt B, Hopkins D. Ophthalmic man ifestations of the visual and neurologica l prognosis in infants with shaken baby
batte red-baby sy ndrome. BM} 1971; 3 :390 . syndrome.} Pedia tr Ophthalmol Strabismus 1996; 33:2 60 - 5.
13 Feldman KW, Schaller RT, Feldma n JA. Tap water scald burns 40 von Noorden GK. New clini ca l aspe cts of stimulus deprivat ion
in children. Pediatrics 1978 ; 62: 1-9. a mblyo pia. Am} Opllthalmol1981; 92:416-21.
14 Landers A, Belfer KF, Jenki ns AD. Supergill e Jids: possibly non 41 Ober RR. Hem orrhagic retin opat hy in infancy: A clinico
acc idental and a medicol ega l problem. Eye 1996; 10:402 -3. patho log ical report.} Pedia tr Ophthalmol Strab ismus 1980;
15 Tom inack RL, Spy ker DA. Capsicum and caps aicin - a review: 17:17-20.
case rep ort of the lise o f hot peppers in child abuse. } Toxicol 42 Lambert SR , Johnson TE, Hoyt CS. Optic nerve sheath
Ciin Toxicol1987; 25:591-60 1. hemorrh ages assoc iated with shaken baby sy ndrome. Arch
16 Tse ng SS, Keys MP. Battered child syndrome simul ating Ophthalmol 1986; 104: 1509.
congenital glaucoma. Arch Ophthalmol 1976; 94:8 39-40. 43 Mun ger CE, Peiffer RL, Bouldin TW et al. Ocular and
17 Fri en dl y DS. Ocu lar manifestations of phys ical child ahuse. associated neuropat hologic observ ations in suspected
Tran s Am Acad Ophthalmol OtolarYllgol 19 71 ; 7 5:318 - 3 1. w hipl as h baby sy ndrome: a retros pective study of 12 cases.
18 Weidenthal DT, Levin DB. Retinal detachment in a battered Am} Forellsic IVIed Pathol 1993; 14: 193- 200.
infant. Am J Ophthalmol 1976; 81:725-7. 44 Mass icote SJ, Folberg R, Torczynski E et al. Vitreoretinal
19 Gonzales CA . Bilateral rhegmatogenous detachments with traction and perimacular retin al folds in the eye of deliberately
unilateral vitreo us base avu lsion as the presenting sig ns of traumatized children. Oph thilimology 1991 ; 98: 1124-7.
child abuse. Am} Op hthalmo/ 1999 ; 127:475-7. 45 Greenwald MJ, Weiss A, Oesterle (S, Friendly DS. Traumatic
20 Green MA, Lieberman G, Milroy CM, Parso ns MA. Ocu lar and retinoschisis in battered babies. Ophthalmology 198 6;
ce rebral trauma in non- accidental injury in infan cy : 93:618-25.
underlying mecha nisms and implications for paediatric 46 Riffe nbllrgh RS, Laks hmanan S. Ocular findings at autopsy of
practice. Br} Oph tl1OImoi 1996 ; 80:282-7. child abuse victi ms. Ophthalmology 199/ ; 98: 1519-24.
21 Massic ote SJ, Folberg R, Torc zynski E et al. Vitreoretinal 47 Tomasi LG , Rosman P. Purtscher retinopat hy in the battered
traction and peri macular retinal fold s in the eyes of child syndrome. A m) Dis Child 1986; 93:1435-9.
deliberately tra umatized children. Ophthalmology 199 1; 48 Ba um JD, Bulpi tt CJ. Retina l and conjun ct ival haem orrhage in
98:1124-7. the newborn. Arch Dis Child 1970; 45:344-9.
22 Munger CE, Peiffer RL, Bouldin TW et al. Ocular and 49 Sezen F. Retinal haemorrhages in newborn infants.
associated neuropathological observa tio ns in suspected Br} Opllthalmol 1970; 55: 248-53 .
whiplash shaken infan t syndrome. A retrospective stu dy of 50 Wilkinson WS, Han DP, Rappley MD, Owings CL. Retin al
12 cases. Am} Forensic Med Pathol 1993; 14:193-200. hemorrhage pred icts neurolog ic injuty in the shaken baby
23 Giangiacomo J, Barkett KJ. Ophth alm oscopic findin gs in sy ndrome. Arch Ophtlwll1lol 1989; 107:147 2-4.
occ ult child ab use.} Pediatr Ophth almol Stra bismus 1985; 51 Paton L. V 11 diseases of the nervo us system: ocular sy mptoms
22:234-7 . in subarachn oid haemorrhage. Trans Ophthalmol Soc UK
24 Levin AV. Oc ular mani festa tions of ch ild abus e. Ophthalmol 192 4; 44: 110-26.
Clin No rth Am 1990 ; 3 :249 -64. 52 Schlo ff S, Mullaney PB, Armstrong DC et al. Retinal findings
25 Tyagi AK, Willsha w HE, Ai nswo rth JR. Unilateral retinal in children with intra crania l haemorrhage. OphtllGlmol 2002;
haemorrh ages in non- acc idental injury. Lancet 1997; 349:1 224. 109: 1472-6.
26 Lin KC, Glasgow BJ. Bilateral peri op tico intrascleral 53 Duhaime A-C, Gennarelli TA, Thibaul t LE et al. Th e shake n
hemorrhages associated with traumatic child abuse. ba by syndrome - a clini cal, pathological a nd biomecha nical
Am} Oph tlw lmol 1999; 12 7 :473- 5. study.} Neurosu rg 1987 ; 66:409-1 5.
27 Ktitzinger EE. Pers ona l com mun ication. 54 Gilliland MG, Fo lberg R. Shaken babies: so me ha ve no impact
28 Kowal L. Ophthalmic manifestati ons of head injury. Aust NZ injuries.} Fore nsic Sci 1996; 41:/14-1 6.
} Oph thalmol 199 2; 20:35-40. 55 David DB, Mears T, Quinlan MP. Ocul ar co mplications of
29 Frank y, Zimmerman R, Leeds NM . Neurologic manifestation bllngeejumping. Br} Op/lthalmol 1994; 70:234-5.
in abused children who have bee n shaken. Dev IVIed Child 56 Lyle DJ, Sta pp JP, Button RR. Ophthalmologic hydrostatic
Neurol 1985 ; 27:31 2-1 6. press ure syndrom e. Am J Opllthalmo/1 95 7; 44: 652 -7.
30 Jensen AD, Smith RE, Olson Ml. Ocular clues to child abuse. 57 Christian CW, Taylor AA, Hertle RW, Duhaime A- C. Retinal
} Pediao' Ophthalmol Strabismus 1971; 8:270- 3. haemorrhages caused by acc idental household traum a.
31 Ball anty ne AJ. Th e nerve fibre pattern of th e huma n retin a. } Pediatrics 1999; 135:1 25 -7.
Trail S Ophtha lm ol Soc UK 1946; 66: 179-91. 58 Buys YM, Levin AV, Enzenauer RW et al. Retinal findin gs
32 Roth M. Uber Netzhau ta ffection en bein Wund feibe rn. Deutsch after head tra uma in in fants and young children.
Z ChiI' 1872 ; 1:471-00. Ophthalmology 1992; 99: 1718-23.
33 Falcone PM, Larrison WI. Roth spots see n on 59 Billmire ME, Myers PA. Serio us head injury in infants:
ophthalmoscopy: diseases with whi ch they may be associated. accident or ab use. Pediatrics 1985; 75 :340- 2.
Conn Med 1995; 59:271-3. 60 Elder JE, Taylor RG, Klug GL. Retina l haemorrhage in
34 Kaur B, Taylor D. Fundus haem orrhages in in fancy. SurIJey accidental hea d traum a in childhood. } Pediatr Child ffealth
Ophthalmo/ 1992 ; 37 :1- 17. 1991 ; 27:286.
35 Fred HL, Chandler FW. Traumatic asphyxia. Am} IVIed 1960; 6 1 The Ophth almology Chi ld Abuse Wo rkin g Patty. Child abuse
29:508-17. and th e eye. Eye 1999; 13 :3-10.
36 Holl en horst RW, Stein HA. Ocular signs and prognosis in 62 Budenz DL, Fa rber MG, Mirchandani HG et al. Ocular and
subd ural and sub arach noid bl eeding in young children. optic nerve hemorrhages in abused in fa nts with intracranial
Arch Ophthalmol1958; 60:187-92. injuries. Ophthalm ology 1994; 101 :559-64.
37 Mushin AS. Ocular damage in the battered baby sy ndro me. 63 Riffenburgh RS, Lakshm anan S. Ocular findin gs at autopsy of
BM} 1971; 3:40 2. ch ild abuse victims. Ophthalmology 1991; 98:15/9-24.
136 l Ocular involvement in non-accidental injury
64 Adams G, Ainsworth J, Butler Let al. Update from the child 77 Hoffman GF, Nau ghten ER. Abuse or metabolic disorder. Arch
abuse working party: Royal Coll ege of Ophthalmologists. Dis Child 1998; 78:399.
Eye 2004; 18:795-8. 78 Pulido JS, Lingua RW, Cristol 5, Byrne SF. Protein C
65 Kramer K. Retinal hemorrhages following cardiopulmonary deficiency associated with vitreous hemorrh age in a neonate.
resuscitation. (lill Perliatr 1993 ; 13 :366-8. Am) Ophthalmol J987; 104:546-7.
66 Kanter RK. Retinal hemorrhages after cardiopulmonary 79 Emerson MY, Pieramici OJ, Stoessi K]\il et al. Incid ence and
resu scitation or child abuse. ) Pediatr 1986 ; 108:430-2. rate of disa ppearance of retinal haemorrhages in newborns.
67 Gilliland MG, Luckenbach MW. Are retinal hemorrhages Ophthalmology 200J; 108:36-9.
found after resusc itation attempts? A study of the eyes of J 69 80 Hughes LA, May K, Talbot JF, Parsons MA. Incidence,
children. Am) Forensic Med Pathol 1993; 14:187-92. distribution and duration of birth-related retinal
68 Odom A, Christ E, Kerr N et al. Prevalence of retinal haemorrh ages : a prospective study. ) AAPOS 2006 ; 10: 102-6.
haemorrhages in pediatric patients after in-hospital 8l Von Barsewisch B. Perilla tal Retinal Haemorrhages:
cardiopulmonary resusc itation: a prospective study. iVlorplw/ogy, Aetiology and Significance. Berlin: Springer
Pediatrics 1997; 99:E3. Verlag, 1979.
69 Fackler JC, Berkowitz !D, Green R. Retinal hemorrhages in 82 Gilliland MG, Luckenbach MW, Massicote SJ, Folberg R. The
newborn piglets foll ow ing cardiopulmonalY resusc itation. medicolegal implications of detecting hemosiderin in the eyes
Am) Dis Child 199 2; 146:1294-6. of children who are suspected of being abused. Arch
70 Tyagi AK, Scotcher S, Kozeis N, Willshaw HE. Can Opi1thaimoI1991; 109:321- 2.
convulsions alone cause retinal haemorrh ages in infants? 83 Schwartz S. Telemedicine and ophthalmology Jules Stei n Eye
Br) Ophtha/mol 1998; 82:659-60. IJlst: Clin Update 1998; 7:3-4.
71 Mei-Zahav M, Uziel y, Ra z J et al. Convulsions and retinal 84 Fishman CD, Dasher WB, La mbert SR. Electroretinographic
haemorrhage: s hould we look fUJther. BM) 2002; 86:334-5. findings in infants with shaken baby syn drome. ) Pediatr
72 Feyi-Waboso AC, Bec k L. Minerva. BM) J997; 314:688. Ophthalmol Strabismus 1998; 35:22-6.
73 Markovitch H. Opening eyes to child abuse. BM) 1999 ; 85 Conway MO , Peyman GA , Recasens M. Intravitrea l tPA and
318:950. SF6 promote clearing of premacular subhyaloid hemorrhages
74 Guyer DR, Sch achat AP, Vitale S et a1. Relationship between in s haken and battered baby syndrome. Ophthalmic Surg
fundus lesion s and haematological parameters at diagnosis. Lasers 1999; 30:435-41.
Ophthalmology 1989; 96:860-4. 86 McCabe CF, Donahue SP. Prognostic indicators for vision and
75 Frase r SG. Retinal haemorrhage in meningitis. Eye 1995; mortality in sh aken baby syndrome. Arch Ophthalmol 2000;
9:6 59 -60. 118 :3 73-7.
76 Levy HL, Brown AE, Williams SE, de Juan E. Vitreous 87 Shokunbi T, Agbeja A. Ocular co mplications of head injuly in
hemorrhage as an ophthalmic complication of galactosemia. children. Childs NeTlJ System 1991; 7:147-9.
) Perliatr 1996 ; 129:922 -5.
I CHAPTER 7 I
THE DEATH SCENE FOLLOWING THE
SUDDEN DEATH OF A CHILD
Anthony Busuttil
INTRODUCTION
they feel have led to the injuries an d the demise of the child.
The police have often found it useful to encourage these wit
The proper, co mprehensive, planned and professional nesses to re-enact quite carefull y and meticulously at the actual
investigation of a suspicious death of a child requires a scene what t hey are suggesting has taken place, using dolls
carefully managed, orde rly and thorough inspection of the as substitutes for the child or infant. These re-enactments,
scene where the death has occurred, wit h its full documen which ha ve been resorted to more frequen tly given the con
ta tion. J This should take place whether the bo dy of the troversies that may arise about the manner of whi ch head
child is still in situ or whether it has already been removed. and neck injuries have been sustained, are videoed by the
Th e scene investigation forms as much an essential part of police. Forensic pathologists should view these and comment
~he death investigation as does the autopsy and an access thoughtfully on the plausibility or otherw ise of what is being
~ o all the medical records of the deceden t. suggested. It has also been known for legal defence personnel
The patho logist investigating the death should always be to do this; if it is known that this tactic will be employed, it
given app ropriate access to the scene, as this will assist fur is essential that the pathologist sees such reconstructions in
iher in the reconstruction of the events that led to the death good time before the case comes to court.
of the ch ild. 2 ,3 Access to the scene may be required again
a fter the au topsy has been completed to attempt to identify
-le vario Lls sites at which blunt trauma may have been SLlS SCENE MANAGEMENT
(ained. If a cou nter- argument of an accidental infliction of
Jl e injuries is being pursued, for example rolling over from It has to be decided at a very early phase, after tile deatll has
a table or couch or a fall down stairs, then it is essential that been brought to the attention of the police, whetller it is
" is hypothesis be tested at the scene in the exact location thought that the death under investigation is suspicious or not.
-,';here it is suggested to have occurred. This reconstruction It is always the counsel of perfection to be safe rather than
should a lso involve the forensic pathologist. sorry, and to upgrade - at least initially - the investigation of
On occasions, carers of a baby or child who has died a scene to suspicious status when one is unsure. This decision
- nexpectedly give a complex narrative of the events that may require an early input from a forenSically trained medical
138 I The death scene following the sudden death of a child
practitioner, though not necessarily a pathologist, and the if looked for and found, thereby proves that such a co ntact
viewing and careful inspection of the body. In situations that had indeed occurred. 4
speak for themselves, in particular when evidence of violence The agenda set before the CSM comprises the following:
is present, this, of course, can be dispensed with. 1. From the time the police first attend the crime scene,
The fact that the body of the child may already have been an accurate and detailed record must be maintained of
taken to hospital for attempted resuscitation does not dispense the comings and goings and of who did what and
with the scene examination, and a close retrospective inspec when. At an early phase, thought should be given to a
tion of the scene where the death has occurred is called for at retrospective timetabled account of events occurring
the earliest possible opportunity. Once the attention of the prior to the arrival of the police.
police has been drawn to the occurrence of a suspicious, vio 2. That the scene itself be documented fully, both
lent or unexpected death, every effoli should be made to return photographically and by video recording, and, if
to the scene where the death took place. In these circumstances, required, by sketches to sca le also. It is an essential good
it is essential that hospital staff and emergency personnel are practice for individuals attending the scene to make their
interviewed soon after the event to obtain from them informa own notes and to make their own rough sketches as
tion regarding: the findings on the child when brought into aide-memoirs. Such documentation will enable 'best
hospital , dead or moribund; the results of any emergency bio evidence' to be produced in front of the courts. All this
chemical , haematological and radiological investigations car documentary material will form part of documentary
ried out; and the various aspects of the resuscitative process disclosure when a criminal procedure is in force. The
that have taken place, including drugs administered to the photographs and videos of the scene will also be
child. It is also useful to obtain access at an early stage to - and available later to refresh the minds of those working on
indeed to secure seizure of, against appropliate receipting the case, which allows better briefings of further
any samples collected from the now-deceased child prior to additional personnel joining the investigative team.
any medication or intravenous infusions having been admin J. All trace evidence has to be collected painstakingly from
istered. These so-called 'pre-transfusion specimens' may prove the scene. A systematic, planned and thorough search
extremely useful for further analysis. has to be made of the entire scene and, initially,
Similarly, if clinical photographs have been taken of the everything that may have some relevance has to be
injuries of the child then police should obtain access to co llected. To be exclusive and selective in evidential
these. Any clothing removed from the now-deceased child, collection at an early phase of an enquiry may be shown
including nappies, should be handed over to the police. later to have been counterproductive and inefficient. The
Slightly more controversial decisions may have to be approach to the collection of evidence should be that
taken, as to when and whether information should go to the each item collected is logged and a 'continuity of the
police, in cases in which a child is admitted with severe chain of evidence' is established for every item. This
injuries and is likely to die as a consequence. If injuries sus refers to the progression of each item from the scene to
tained are serious and, perhaps, 'are likely to prove' fatal, a the forensic laboratory and eventually to couli; each step
decision to inform the police or not has to be taken at the of this procedure has to be accounted for fully and not
earliest possible time by the clinical staff. There should be found to be wanting. Every person handling the item of
no dubiety in the minds of the doctors in accepting that this evidence signs the original label attached to it at the
would not constitute a breach of medical confidentiality scene. Strict attention has to be given to ensure that
due to the patient giving their oveniding obligations as cit appropriate containers, methods of preservation,
izens of a country to ensure that crime is investigated and labelling and packaging are used. The collection,
its perpetrators detected. A senior member of the medical labelling and logging of the items collected is delegated
team should make this disclosure to the police, speaking to to a police officer designated as the exhibits' (or
a senior police officer. Hospitals should produce written productions', in Scotland) officer.
protocols about the procedures that should be followed in 4. All relevant investigative personnel should be given
such circumstances. The child's carers must be told that the access, in turn, to the scene, ensuring that they are all
police have been informed. wearing appropriate protective clothing in the course
of their attendance to decrease scene contaminati on .
They will be briefed to work in unison and in mutual
THE CRIIVIE SCENE MANAGER collaboration; altho ugh, in order to avoid congestion
and overcrowding at the scene, each team may have to
The scene will often be placed under the control of a spe work separately, with the crime scene manager present
cially trained scientist or police officer designated as the at all times. The pathologist's role at the scene should
crime scene manager (CSM). The CSM's g'uiding principle is principally concern the body. However, even if the
that enunciated by Edmond Locard over a hundred years body has left the scene, there is often still scope for the
ago - namely that every contact between two persons or a pathologist to inspect the scene of death, particularly if
person and a scene will leave a trace of this contact, which it is being treated as a suspicious death.
A good look around I 139
5. It is essential that there is adequate lighting at the 9. Specially trained 'scenes of crime' officers inspect the
scene. It is foolhardy to attempt to carry out decent scene and collect trace evidence.
and proper investigations without appropriate lighting. 10. Forensic scientists inspect the scene and collect
The crime scene manager should ensure that portable evidence.
generators or other sources of light are made available. 11. Pathologist is given access to the body.
The commencement of the investigation may have to 12. Trace evidence from the body, in terms of tapings
await the arrival of such equipment and if possible from the exposed surfaces of the body (for fibre
daylight. evidence), is collected, and swabs are taken from
6. Information is sought from the scene that will assist accessible orifices.
with an estimation of the time of death; this requires a 13. The body is transferred on to a clean cotton or
recording of the ambient temperature. If death occurred polythene sheet and tra nsferred to the mortuary for
indoors and sequential changes to room temperature more detailed inspection.
over time are known to occur (e.g. central heating 14. If required, finger print experts cover the scene.
cycle), these should be appropriately logged. If the death 15. The scene is secured.
occurred out-of-doors, meteorological reports should be 16. Further examinations as required - all fully logged.
sought for the area to assist with such estimates. These
will have to include details about humidity, ground
temperature, wind-chill factor and rainfall. UNCLOTHING THE BODY
7. Exposure to carbon monoxide should be considered in
all domestic childhood and infant non-violent deaths, This should be avoided at the scene as it may lead to loss
and any potential sources of this gas, for example a of evidence, to potential contamination and perhaps also
blocked chimney, faulty kerosene heaters, blocked gas cross-transference of evidence to it. It is a better policy to
flue and coke burners, should be carefully looked for postpone the unclothing till later, and for this to take place
and expert advice sought about any possible under controlled conditions, in a good light, in a mortuary.
malfunction. This would enable each item of clothing to be removed
8. If possible, the scene should be secured after all the sequentiaJJy and to be carefully preserved. Contents of
searches have been carried out, as it may become pockets can be removed, identified and considered, and
necessary to return to the scene at a later time to check areas of damage to items of clothing can be identified and
over certain facts or to look for other evidence; new considered, as each layer of clothing is removed in
additional members of an investigative team also may sequence.
wish to see for themselves the scene of the fatal In babies, the clothing is of particular importance in
incident. attempting to assess whether the baby has been subjected
to excessive heat;l headgear, in particular, retains a signif
icant amount of heat. If the clothing is soaked in sweat it is
SEQUENCE OF EVENTS AT THE DEATH SCENE a very good indication of excessive heat exposure (or of a
recent convulsion or hypoglycaemia). Inadequate, flimsy or
Once the crime scene manager has been designated, the wet clothing worn by the child in a cold bedroom may be
sequence of events at the death scene often follows the fol similarly of importance if a death from hypothermia is
lowing pattern of activity:5,6 being considered.
1. If the body is still in situ, establish a path to the body
that will be adopted by all those wanting to inspect
the body, this is referred to as th e Co 111 1110 17 Approach A GOOD LOOK AROUND
Pathway (CAP).
2. The CAP may have to be demarcated and preserved
From the room in which the body is found, the search has
by elevated platforms or stepping plates that have
to proceed to every other room in the house in a systematic
bee n thoroughly cleaned previously.
fashion. In addition to looking for other evidence, a good
3. If in the open, construct protection (e.g. tent) for the look around the house will give a better appreciation of the
body and its immediate surroundings. lifestyle of the parents and the other inhabitants of the
4. Set up a written log of named comings and goings to house, and also the prese nce of pets. The police will be alert
the scene. to signs of chronic controlled substance misuse and alcohol
5. (A medical practitione r or a suitably trained
misuse or overuse. This would also mean some degree of
paramedic formally pronounces life extinct).
attention to sites of waste disposal and garbage containers,
6. Provide adequate lighting or wait till daylight. to gardens and outhouses.
7. Photographs a nd video team document the scene
Special care must be taken in examining any bathrooms
fully.
and kitchens. In the latter, any partially consumed food or
8. Appoint an exh ibits' officer. made-up bottles of formula milk should be retained. These
140 I The death scene following the sudden death of a child
should be refrigerated to prevent fermentation, which leads even more important, if the baby has been co-sleeping, in
to spurious results of ethanol estimations. They can also be order to assess the possibility of overlaying. 2I ,22
tested for medical and illicit drugs as well as alcohol. It is important to retrieve the bedding in order to dis
Any medication found within the household, in particu cover whether this was stained wi th blood or saliva or was
lar antipsychotics, antidepressants, benzodiazepines, anti covered in vomit, and also to assess whether it could have
convulsants and controlled drugs, should all be noted and made the baby's environment too warm or, ind eed, inap
removed. propriately cold.
When various theories related to the possible production
of inespirable and toxic gases from cot mattresses were being
SUDDEN INFANT DEATH SYNDROME OR investigated, it was considered important to retain the mat
NON-SUDDEN INFANT DEATH SYNDROME tress for appropriate testing.2J,24 This is no longer necessary.
Great caution must be exercised, if on investigation it is
The most common cause of post-perinatal motility in the discovered that the 'cot death' under investigation is the
western hemisphere is still sudden infant death syndrome second, or even more concerning, the third sudden infant
(SIDS). However, as the incidence of SIDS has declined death in the same family. Greater suspicion and , therefore,
worldwide,S the number of deaths related to abuse and vigilance should be entertained in these circumstances. 25 ,26
neglect has increased proportionally, and the task of distin
guishing a natural fr om a suspicious dea th is becoming
increasingly frequent. 9,1O,1I In such instances, a thorough EXTERNAL PETECHIAE
scene investigation assists greatly in reconstruction of the
events that preceded the death. 12. IJ ,14 Although there is some dispute as to the percentage inci
The contribution that death-scene examinations make in dence of the presence of petechiae on the serous membranes
relation to the definition of such infant deaths has been a of children dying of SIDS, there is no doubt that th e presence
source of some controversy. In the 1969 defini tion of SIDS of extern al petechiae in the face, behind the ears, in con
produced by Beckwith,15 there is no mention of the scene junct ival membranes and on t he mucous membranes of the
inspection. However, in 1976 there was a recommendation lips and gums of a baby dying of SIDS is exception al. 27 ,2B
by an American Working Party in their Special Report that It is a simple enough test to carry out in a good light
scenes of death should be visited. 16 In his second attempt at with the baby in situ. If petechiae are found localized to
more elaborately defining SIDS, during the second SIDS the upper t runk and face, in the absence of exposure to
International Conference held in Sydney in 1992, Beckwith meningitis or known coagul ation problems, then that par
suggested the introduction of a scene investigation into a ticular baby's death has to be treated as suspicious; it
three-tiered definition; this also failed to meet with general should be considered a result of mechanica l asphyxiation,
approva l. During the second SIDS Glob al Strategy Meeting an accident or a criminal act, until proven otherwise by the
held in Stavanger, a further attempt at making a satisfac investigation.
tory scene investigation an essential criterion for the diag Accidental mechanical asphyxial deaths do occur in
nosis of SIDS also failed.17 However the National Institute babies and young children and only an examination of the
of Child Health and Development's (NICHD's) definition of scene of death would enable the mechanisms of death in
SIDS has, since 1989, required that an examination of the such instances to be elucidated. 29
death scene be performed before a diagnosis of SIDS can be
made. IS
The definition of the scene has been widened 3 to incor BRUISING
porate not only the location where the infant was discov
ered dead or unresponsive, but also the no rmal home The presence of bruises on children, and particularly
environment of the child and any other locations that may on babies, is always a cause for concern and suspicion,
have been visited by the child immediately prior to the particularly in children who have not started to toddle and
death. Protocols have been published from the United if bruises are of different ages . Bruising in unusual sites,
States outlining the scientific and medical input in such such as the back, arms, neck, chest, thighs and genitals, is
scene investigations. 19 also particularly worrying in infants.
If the baby has been found dead in a cot, crib, Moses bas
ket, or something similar, then it is important to discover
the position in which the baby was found prior to the atten ABANDONED NEONATES
da nce of ambu lance crews and paramedics. Life-sized flex
ible dolls have been made use of, in this respect, to enable The abandoned body of a neonate is still a matter that
the carers of the child to recollect the initial discovery of the greatly exercises the sensitivity of the public and the interest
child. Well over 50 per cent of babies dying of SIDS are of the media. On few occasions, it is the consequence of
found in the prone sleeping position. s This reconstruction is homicide. Nowadays, the emphasis is clearly biased towards
Dyadic and multiple deaths I 141
attempting to trace and then assist the mother of the aban decades and which come to light with renovations and dem
doned baby, addressing any medical, social welfare or finan olition of old buildings (see Fig. 10.2, p. 183), a remnant of
cial problems being experienced by the mother, rather than the days where the birth of an illegitimate child was a mat
in seeking punishment. The potential for a criminal tlial, ter of public shame and dishonour and contraception was
however, still exists, particularly if the body shows features primitive. These cases are velY difficult to investigate, not
of injury and the mother is eventually traced. Therefore, the least because with the passage of time the cause of death and
police have to investigate the matter and carefully search the indeed the potential perpetrator cannot be identified. J5 An
scene where the body is discovered. attempt will usually be made to assess the age of the skele
The incidence of neonate abandonment varies between ton to determine whether it is pre- or post-viable.J 6,37
countries but even in Western countries such instances are
not exceptionaI. 2o ,21,JO,JI It is often the case that the baby
has been concealed prior to its disposal. Although putre DEATHS FROM TRAUMA
factive changes are delayed in newborn infants, owing to
the sterile environment from which they have been deliv A careful review of the scene is a sine qua non in traumatic
ered and due to the absence of bacteria in their intestines , deaths, particularly when different scenarios have been
some decomposition, albeit incomplete, may be present. alluded to in statements given by those who were there
Over time, there also may be partial mummification of the when the injUlY took place. The potential veracity of one or
body in dlY and well-ventilated surroundings. Often the other scenario has to be investigated by a careful inspec
body of the child has a lready been attacked by animals and tion of the scene, with measurements being taken as appro
bears the marks of this. These have to be distinguished priate22.J8,39
fioman~-mortemtrauma. A debate still rages over whether a head impact against
In all these instances it is important to ensure that a a solid unyielding surface, causing acute deceleration, is
sea rch of the scene is made for the placenta, and for any essential to produce the pathological changes that a re
other items that may have been discarded together with the ascribed to the 'shaken baby' syndrome.4o ,41 Evidence for
baby, for example distinctive baby clothes, newspapers and such an impact may be scanty, such as small indents in fur
wrappings; this may give a clue to the locus at which the niture or the deposit of a sma ll amount of hair and skin
baby was delivered. Occasionally, notes may have been from the baby, but shou ld be carefully sought.
pinned to the body and sometimes items of jewellelY left
with it; forensic examination of these may ass ist with trac
ing the mother. This may be indispensable in the identifi Deaths in Fires
cation of the abandoned body, which is often a major
problem and often rema ins unsolved.J2.J 3 In Britain, particularly in Scotland,42 there are a number of
Infanticide has been committed in a number of ways, deaths every year in which children are killed by household
such as intentional neglect, exposure, killing with a fires . This incidence has hardly changed over the years
weapon (e.g. stabbing or by the head being struck against despite major educational attempts by fire brigades and cen
a solid surface). suffocation (by the insertion of foreign tral governments to ensure that houses are fitted with work
material into the mouth or with a soft pillow) and by ing smoke alarms, and discouraging the cultures of smoking
drowning (see Chapter 18). In all instances, it may be use in bed or cooking while under the influence of alcoholY,44
ful to attempt to distinguish whether trauma at the time of The deaths may be accompanied by adult deaths,45,46
birth was brought about by inexpeJiise in delivery or by and it is frequently the case that smoke inhalation is the
precipitate labour from intentional injury infliction. 34 cause of the death. The children may be too young to be able
Occasionally, abortuses and products of conception to find their way to safety or may become disorientated
are found and the police may have to investigate. It is and terrified at an early phase and succumb to smoke
essential, in the first instance, to ensure that one is indeed inhalation before reaching safety. The possibility of drug or
dealing with human remains. Products of conception of alcohol admin istration to children in this circumstance
non-human origin may closely reassemble human material, should not be forgotten.
for example rabbits and other small mammals. Some years The dynamics of the commencement and spread of the fire
ago, in many co untries, toy manufactures produced android and the use of accelerants will be reported on by the special
toys, which had the same size and general appearance of a ized fire officers and the forensic team that attends the fire. In
fetus, and many a n investigation had to be abruptly termi all such cases, the possibility of arson should be kept in mind.
nated with a ceJiain amount of 'red face' all around when it
was discovered that it was not a human fetus which had
been discarded but one of these toys, sometimes as a prac DYADIC AND MULTIPLE DEATHS
tical joke, in poor taste, or simply accidentally.
Another common finding is the presence of bones of A dyadic death is one in which the perpetrator of a homi
babies that appear to have been lying undisturbed for cide later commits suicide,47 often at the same site where
142 I The death scene following the sudden death of a child
the homicide occurred, with multiple bodies found in the site where the murder has been perpetrated or whe re the
same scene. Many of these instances of homicid e cum sui body has been stored prior to its disposal. In such
cide are th e consequence of severe, occasionally previously instances, more than one scene will have to be carefully
undiagnosed, mental disorde rs, such as severe dep ression examined once these have been identified, each being
or paranoid schizophrenia. In many instances, there is an organized and supervised by a different CSM.
acute, emotional trigger for such occurrences, for example To enable a ll the various experts to congrega te at the
bankruptcy, prosecution of the fath er for some alleged scene, it may be necess ary to retain the body in situ for a
offence, the discovelY of a n extramarita l affair, an impend lengthy period. occasionally of several hours' duration. The
ing threatened sepa ration or divorce. Men and women can public media may suggest that this is a callous and insen
both be involved in such deaths. Often, an entire family sitive way in which the police de al with such gri m discov
can be eliminated in the same event. Mechanical asphyxial eries ; from personal experience, there is no substitute to
methods, fire arms and hefty sharp weapons, suc h as axes ensuring that such investigations are thorough, compre
or machetes, are made use of. hensive and as all-encompassing as is possible. Delay in
The wea r and tear of modern living appears to have carrying out the autopsy is a sm all penalty to pay for this.
taken its toll, resulting in an increasing incidence of such Id entification of the victi m becomes paramount in such
deaths. These scenes are particularly harrowi ng for the situations, and the assistance of forensic odontologists and
investigators, but in all these instances it is essential to deoxyribonucleic acid (DNA) exp erts are invaluable in suc h
ensure that there need not be a search for another person. instances.
Even more harrowing and soul-destroying are scenes in
which many children have met their death together in a
violent manner. In major incidents of natural or unnatu ra l SENSITIVITY AND STRESS OF
causation, children are often among the fat a lities. It is, THE INVESTIGATION
however, the exception al and unusua l situation in which
all the victims in a majo r incident are children. 48 The noto Deaths of babies and children are becoming increasingly
rious exception in this co unt ry was the incident in Dun uncom mon and thus when such a death occurs, particu
blane primary school; othe r countries, such as the USA larly if it is sudden or violent, it wi ll cause major emotional
have had to deal with several simil ar incid ents . These inci upheaval and grief in the immediate family. If there is an
dents have to be dealt with as if one is dealing wi th multi index of suspicion that has to be investigated and excluded
ple single homicid es, each decedent being treated then great care must be taken to ensure that the fami li es
individually as a vic tim of homicide. The scene in its are treated with resp ect and sensitively at all times, even
entirety, no matter how large, becomes a scene of crime. when there is high index of suspicion re lating to their
direct involvement in the death .
The police have, over recent years, trained officers as
SUDDEN DEATHS OF OLDER CHILDREN 'family li aison officers' who can speak to and communicate
with the families, keeping them informed of the progress of
PsychiatIists are beginning to build up a better profile of the investigation. There wilJ be a certain amo unt of antag
child sex abuse rs and killers. Most child murders are carried onism and animosity shown towards the police and what
out by fami ly members - often mentally ill. In other cases they are perceived to represent. It may be n ecessa ry to
the perpetrators have been abused as a child themselves. liaise with primary medical care rs and the social wel far e
However, there are also a small number of paedophilic crim department to ensure that the family gets best treatment.
inals responsible for child homicides. In such cases, the chil If it is thought that other children in the same family may
dren involved are older children and the motive for their be at risk of physical injury or neglect, the appropriate social
murder is sex ual g rati fication of the perpetrator. These clim welfare legislation must be utilized to safeguard other siblings.
inals will do their best to conceal their crime and it is often The family may wish to view the body of the deceased
the case that the body of the deceased child will not be fo und baby: they have a right to do so, a right enshrined in the
until an exhausti ve search has been carned out; when the European Convention of Hum an Rights. Unless there are
body is eventua lly found, it is often in a state of decomposi very good reasons to the contrary, their wishes should be
tion. The exa mination of such scenes is even more fraught accommodated as soon as possible. The dead baby should be
and requires a meticulous attention to detai l. presented in surroundings that make this harrowing experi
In order to glean as much information as is possible ence less stark. The infant ca n be held and touched and
fro m the sites where such bodies are deposited, in an effort mementos, such as a hair lock or an imprint of the hands or
to solve the crime, it is often the case th at the expeltise of feet in ink, wax or plaster of Paris be provided.
a number of other experts may have to be called into play. Matters relating to post-traumatic stress cannot be
These include, for example, arc haeologists a nd anthropol ignored or side-stepped for any reason , not least the possi
ogists, soil experts, experts in pollen and entomologists. bility of issues of compensation. The pathologist at the scene
The final resting place of the body may well differ from the may feel that the officer in charge of the scene, the senior
Referen ces I 143
investigating officer (SIO), should set in motion appropriate 16 Jones AM, Weston JT. Examination of the sudden infant
provisions, which his police force will p ut in place. In cases death syndrome: investiga tive and autopsy protocols.
J Forensic Sci 1976; 21 :833-41.
involving mutilation of the body, whatever the cause, this
17 Hilton J, Berry JP. Pathol ogy. In Fitzgerald K (ed.) Second
attains an even more acute importance. S/DS Global Strategy Meeting. Norway: Stavenger, August
5-6, 1994, p. 334. In Rognum TO (ed.) Sudden Infant Death
Syndrome. New Trends in the Nineties. Oslo: Scandinavian
University Press, 1995.
INQUESTS AND INQUIRIES 18 Willinger M, James S, Catz C. Definin g the SIDS: deliberat ions
of an expert panel conceived by the National Inst itu te of
All that transpires at the scene will be scrutinized in court. Child Health and Human Development. Pedilltr Pathol J 994 ;
This may be a coroner's court where a n inquisitori al II :677-84.
scheme of thin gs is in operation, or in subseq uen t criminal 19 Center for Disease Control (CDC). Guidelines for dea th scene
in vest igatio n of sudden , unexplained infant dea th s:
cases in which an adversarial and confro ntatio nal system is
reco mmendations of the Inter Age ncy Panel on Sudden Infant
at wo rk, an d all aspects of the investigatio ns may be a nd Death Sy nd ro me. MMWR Rec0111111 Rep 1996; 45:RR-JO.
a re likely to be scrutinized through the fine sieve of cross 20 Iyasu S, Ha nzlick RE, Rowley D, Wi llin ger M. Pro ceedings of
examinatio n. Workshop on Guidelines for Scene Investigation of Sudden
Certa in cases may also come to the attention of Tribunals, Unexp lained Infant Deaths. J Forens ic Sci 1994; 39:
1126- 39.
Public Inquiries or other judicial investigations in which dif
21 Dragon DA, Lan denberg Ai. Infant mechanical suffocation
ferent nuances of procedure may hold, but evidence will stil l deaths in the United States, 1980-1977. Paediatrics. 199 9;
have to be printed under oath or affirmation. 103 :59.
22 Nakamura AS, Wind M, Danello MD. A review of hazards
associated with children placed in adult beds. Arch Pediatr
Adolesc Med 1999;153:1018-23.
REFERENCES 23 Fact Sheet on the Tox ic Gas Hypothesis of Cot Death,
www.sids.org.uk/ed itpics/364-I.pdf
Wagner GN. Crime scene investigation in child abus e cases. 24 Coghlan A. .Mattress bugs may link to cot deaths. New
Am J Forensic Med Pathol 1986; 7:94-9. Scieutist 2002; 10:41.
2 Usher A. Th e role of the pathologist at the scene of crim e. 25 Meadow R. Recurrent cot death and suffocation. Arch Dis
J Fo rensic Sci Soc 1970; 10:213-8. Ch ild 1989; 64(1): 179-80.
3 Bass M, Kravath RE, Class l. Death scene investigation in 26 Emery JL, Gi lbeli EF, Zugu ibe F. Three crib deaths.
sudden infant death. N Engl J Med 1986; 315:100- 5. A babyminder a nd probab le infa nti cide. Med Sci La w
4 Locard E. The analysis of dust traces, Parts I-Ill . Am J Police 1988 ; 28(3) :205 -11.
Sci 1930; 1: 276- 298,405-406, 496-514. 27 Byard R, Petechia l hemorrhages and unexpected infa nt deat h.
5 Vanezis P, Busuttil A. Suspicious Death Scene investigation. Legal Medicine 2003; 1(4): 193-7.
London: Hodder Arnold, 1996. 28 Betz P, Hausmann R, Eisen menger W. A contribution to a
6 Wh ite P. Crime Scene to Court. The Essentia ls oj Forensic possible differe ntiation betwee n SIDS a nd asp hyxiatio n.
Science. Lond on: The Roya l Soc iety of Chemistry, 1998. Forensic Sci Int 1998; 9 1: 147 -52.
7 Stanton AN. Sudden infa nt death. Ove rheat in g and cot death. 29 Shepherd RT. Accidenta l se lf-s trangula tion in a yo un g chi ld.
Lancet. 1984; 2: 11 99-201. Med Sci Law 1990; 30: 11 9-23.
o Byard RW, Krous HF. Sudden Injant Death Syndrome. Lond on : 30 Mend lowicz MV, Jean -Lo ui s G, Gekker M, Rapaport MH.
Arnold, 200 I. Neonaticide in the city of Ri o ele Janeiro: forensic and
9 American Academy of Ped iatrics. Distinguishing sudden psycholegal perspectives. J Forensic Sci 1999; 44:741 - 5.
in fant death syndrome from child abuse fata lities. Pediatrics 31 Cheung PI. Materna l filicide in Hong Kong, 1971-85. jVled Sci
1994; 94( I): I 24 - 6. Law 1986; 26: 185 - 92 .
10 Christoffel KK, Zieserl EJ, Chiaramonte J. Should child abuse 32 Lee ACW, Li CH, Kwong NS, So KI. Neonat icide, newborn
and neglect be considered when a child dies unexpectedly? abandonm ent, and denial of pregnancy - newborn
Am J Dis Child 1985; 139:876-80. victimisation associated with unwanted moth erhood. Hong
II EmelY JL. Chi ld abuse, sudden in fant death syndrome, and
Kong jVled J2006; 12: 6 1-4.
unexp ected infant death. AnI J Dis Child 1993;
33 Herman -Gidd ens ME, Smit h JB , Mittal fvJ et al. New born
1476( 10) : 1097 - 100.
killed or left to die by a pa rent: a population-based study.
12 Bya rd RW, Carm ichae l E, Bea l S. How useful is the post
JAMA 2003; 209:1425-9.
mortem examination in sudden infant death syndrome?
34 Mitchell LB, Dav is JH. Spontaneous births ill to ilets.
Pediatr Patho1 1994; 14:817- 22. J Forensic Sci 1984; 29(2):591-6.
IJ Bya rd RW, Becker LE, Berry 1. Formal retrospective case
35 Kerley ER. Forensic Anth ropology and crime involving
review and sudden infant death. Acta Paediatr 1997;
ch il dren. J Forensic Sci 1976; 21[2):33 3- 9.
86:10 11-1 2.
36 Hargrave DR. A study of chil d hom icide over two decades.
14 Newlands J, Emery JS. Child abuse and cot deaths. Child
Med Sci Law 1992; 32[3): 196- 245.
Abuse Negl 1991; 15(3):275- 8.
37 Kerley ER. The ide ntificatio n of battered- infant skeleto ns.
15 Beckwith JB. Discussion of terminology and defin iti on of th e J Forensic Sci 1978; 23(1): 133- 8.
sudde n in fant death syndrom e. In Bergma n AB, Beck with JB , 38 Kasim MS, Cheah I, Shafie HM. Childhood deaths from
Ray CD (eds) Sudden Infant Death Syndrome. Proceedings oj physical ab use. Child Abuse Negl 1995 ; 19(7):84 7- 54.
the Second international Conjerence on Causes oj Sudden 39 Kellet RJ. Infanticide and ch ild des U'llCtion - the historica l,
Death in Injants. Seattle: Uni versity of Washi ngto n Press, legal and pa thological aspects. Forensic Sci Int 1992:
1970, pp. 14- 22. 53 (10):1-28.
144 I The death scene following the sudden death of a child
40 Alexander R, Sato Y, Smith W, Bennett T. Incidence of 45 Runyan CW, Bangdiwala IS, Lin zer MA et a1. Fatal house
impact trauma with intracranial injuries ascribed to shaking. tires: who dies and who su rvive? ]AMA 1998; 279:
Am] Dis Child J990; 144 :557-63 . 1633- 7.
4J Duhaime AC, Christian CW, Rorke LB, Zimmerman RA. 46 Istre GR, McCoy BS, Barnard JJ, Bolton A. Deaths and injuries
Non-accidental head injury in infants - the 'shaken-baby' from house fires. N Eng] Med 2001; 344(25): 1911-16.
syndrome. N Eng] Med 1998 ; 330:1822-7. 47 Buteau J, Thompson B, Sexton D. Homicide followed by
42 Squires T, Busuttil A. Child fatalities in Scottisb house fires: suicide: a Quebec case-series, 1988 -90. Cal1 ] Psychiatry
a case of child neglect? Child Abuse Neg l J995; 19:6 5-73. 1993; 38(8):552-6.
43 Roberts I. Deaths of children in house fires. EM] 199 5; 48 Busuttil A Lockerb ie Ct Dunblane. Disasters and dilemmas.
311: 1381-2. Med Leg] 1998 ; 66(4):34-42.
44 Squires T, Busuttil A. Alcohol and house fire fat aliti es in
Scotland. Med Sci Law 1997; 37:325-40.
I CHAPTER 8 I
POST-MORTEM EXAMINATION IN
BABIES AND CHILDREN
Jean W Keeling
This chapter considers the investigation of deaths in the Thorough appraisal of the locus is as relevant to deaths
perinatal period, infancy and early childhood. The investi occurring in the home as it is to motor vehicle collisio ns and
gation of deat hs in all th ree groups follows a similar gen other incidents away from the domestic environment. Bass
eral line, but there are some diffe rences in techn iqu e, et al3 and Sturner4 stress the importance of detailed eval ua
emphasis and even in interpretation of the same patho tion of the death scene in unexpected infant deaths. This
logical findings, for example the presence of facial includes the physical state of the infant's place of sleep and
and upper trunk petechiae, among the gro ups. Such differ the precise position of the in fa nt when found. The latter is
ences wi ll be emphasized at appropriate points within the particularly important when positio nal asphyxia is a possi
chapter. bility and when co-sleeping is practised. The use of a life
Sudden death in early life may be the outcome of a var sized doll may assist the recall of witnesses and improve the
iety of natural diseases (see Chapters )0-12) as well as precision of information about position when fou nd. 4
acc idents and homicide. Sometimes, a pre-existing con Evaluation of an inappropriate sleeping environment in
genital or acquired disorder may predispose a child to acci respect of temperat ure, humidity and the presence of
dents and, possibly, to deliberate assault. In an y sudden noxious gases can only be made by prompt evaluation of
death , whether or not a cause of death is precisely the scene. 3 ,5
defined an d particularly when some necropsy findings , Seizure of feeding bottles and domestic items, such as
such as petechial haemorrhages, can be the res ul t of natu jugs and spoons used in preparation of feeds, for toxico
ral disease, it is most important that a full range of inves logical analysis is important. The contribution of drugs and
tigations is carried out and that their results are alcohol to sudden deaths in babies and children is probably
clearly presented. 1,2 Failure to do so weakens any assert ion un de resti mated.
of non-natural death and enhances the scope of cross A visual record of the death scene can aid the interpret
examination. ations of necropsy findings and high-quality photographs
146 I Post-mortem examination in babies and children
or video recording are often extremely useful. Death scene injuries present, particularly in the infant period when
investigation is discussed in detail in Chapter 7. there is much individual variation. Non-mobile infants
should have few, if any, bruises. 7 ,B Awareness of supervi
SOlY arrangements around the time of death is also impor
RECTAL TEMPERATURE tan t. An inexperienced carer may unwittingly permit
hazardous activities or inappropriate freedom and be less
Measurement of rectal temperature at the death scene or on aware than parents of recent recommendations about best
arrival at hospital may be a useful corroborative investiga childcare practice. 9 Carers who are distracted by illness,
tion. If elevated, it raises the possibility of pyrexial illness depression or other adverse events, such as marriage break
or an inappropriately warm sleeping environment. down and bereavement, are likely to be Jess vigilant in
There are no tables or formulae relating rectal temperature respect of both symptoms of illness and environmental
and likely time of death that are appropriate for use follow hazard.
ing the death of babies or children. Surface area in children is
relatively greater and body mass less than that of adults, so
that cooling is likely to be more rapid and probably more RADIOLOGICAL EXAMINATION
influenced by the amount of clothing and other coverings. 6
A fuJJ radiological examination is an essential part of the
MEDICAL AND FAMILY HISTORY evaluation of any sudden death in infancy and childhood,
and this is further discussed in Chapter 3. Necropsy exam
Information about the individual's medical history and ination should not be contemplated in establishments
family background should be available to the pathologist where radiography is not available. It is better to accom
prior to necropsy examination. modate a small delay to necropsy so that radiological evalu
Details of pregnancy, birth and progress in the neonatal ation can be carried out rather than having a major
period are essential in infant deaths, together with a disruption to dissection or an incomplete examination after
detailed account of recent illness and of events in the 48 the post-mortem examination has been completed.
hours prior to death. A single-film 'Babygram' is not sufficient; it is esse11tial
In older children, a brief history of birth and early life to have a complete set of appropriately penetrated regional
may be sufficient; however, information about any ill views. The radiographs should be taken by an appropri
nesses and accidents, particularly those resulting in hos ately qualified and experienced radiographer. This individ
pital admission, are important. ual is important for his/her expertise in producing the
Information about members of the immediate family appropriate views to encompass classical injUly sites and
should include major problems of a social, legal or finan ensuring technical excellence. An experienced radiog
cial nature, recent illnesses and documentation of any rapher may suggest additional films. Take the advice!
familial disorders. Information about sudden death or col Recent rib and skull fractures cause most problems for
lapse in the wider family, particularly in early life, should the radiologist; however, a pathologist is unlikely to miss
be specifically sought. A history of many consultations them as they can be observed directly and should always
with the general practitioner or hospital, either by the be carefully sought. Small fractures in long bones in the
deceased or siblings, is important information for the vicinity of joints (see Chapter 3) are best demonstrated
pathologist. Multiple hospital attendances can be difficult radiographically but require histological confirmation.
to ascertain, as children may have been taken to a variety Histological examination of any area when there is dubiety
of different institutions. about the nature of a radiological abnormality is essential.
The 'At Risk' register, or its equivalent, should be Sections through identified fractures can contribute to
checked carefully, not only for the deceased individual, but assessment of the age of the fracture.
also for siblings and half-siblings, ta king cognisance of the In any suspicious death, it is important that the radio
multiplicity of surnames used in some of today's compli graphs are reviewed prior to necropsy examination by an
cated family units. experienced radiologist, preferably one with paediatric
Information about prescribed medication for the decedent experience. A verbal report will suffice.
and any other family member is required, togetl1er willi It is often useful to X -ray individual bones or regions
information about the availability of recreational drugs and (i.e. rib cage) following removal during necropsy. In partic
alcohol in the decedent's immediate environment. ular, it ensures better views of posterior rib fractures (Fig.
8.1), which, being close to the costovertebral articulations,
are often obscured by thoracic viscera. Localized abnor
OTHER IMPORTANT INFORMATION malities of uncertain aetiology may be positively identified
using this procedure. Re-X-ray of a skull vault with
Information about the deceased's mobility and physical wormian bones ensures an optimal record and avoids later
capabilities are important to the interpretation of any argument.
Photography I 147
PHOTOGRAPHY
(a)
(b)
Table 8.1 Microbiological samples usefully token in sudden Table 8.2 Useful toxicological samples in the investigation of
death investigation in early life sudden death in early life
has extended the period over which samples may be use possibility in mobile infants and small children when pre
ful, 14, 15 but the significance of detection of viral RNA must scribed medication or recreational drugs are present in their
be considered in association with relevant histopathological environment. Alcohol ingestion can produce rapid and
changes in appropriate organs. marked metabolic dis turbance in the young.
Although microbiological investigations are positive in Rapid onset of drowsiness/incoordination and metabolic
only a small proportion of cases, negative results are partic disturbance, such as hypoglycaemia, are indications for full
ularly important in the sudden unexpected deaths in infancy toxicological investigation. Paracetamol (acetaminopohen) is
(SUDl)/sudden infant death syndrome (SlOS) context l6 and it recommended only on presCliption for infants < 3 months of
is important that these investigations are carried out, even age. 19 A single dose may be given following immunization.
when death seems likely to be the result of trauma; failure to Detection of non-prescribed paracetamol in young infants
do so can make for vexatious cross-examination. requires a full explanation.
It is worth remembering that some microbiological causes Suitable samples for toxicological investigation are tabu
of sudden death in early life, such as group B streptococcal lated (Table 8.2).
and some viral infections (such as echovirus), elicit no spe
cific findings at either gross or microscopic level and so that
positive cultures are essential for diagnosis. BIOCHEMICAL AND METABOLIC
INVESTIGATIONS
Prioritization of investigations is usually necessary and outpatient departments. The most widely used charts in the
is based on a combination of history and pathological UK are based on the work of Freeman et al 26 (Appendix B).
findings. Weight and length are important when estimates of
Sodium, urea and osmolality are useful baseline investi blood volume and calculation of administered drug doses
gations JJ and levels are generally stable in vitreous are required.
humour for about 72 hours. Sodium and osmolality pro Caution should be exercised when comparisons are
vide corroborative evidence of hypernatraemia or dehydra made between body weights recorded during life and post
tion. An elevated level of urea is not usually found until mortem weight. Weight in life may include clothing, a
2-3 days into an illness and provides a general indicator of number of different weighing scales may be used and any
subacute illness. may be inaccurate. (When were your mortuary scales last
Glucose levels in vitreous humour are not so stable but inspected and calibrated?)
are useful for up to 12 hours after death,13 which makes Weights may not have been recorded accurately - check
their usefulness at time of necropsy questionable. them yourself if in doubt, especially when using an
The reliability and usefulness of post-mortem investiga unfamiliar mortuary or when examination of babies is
tions is discussed further in Chapter 5. carried out infrequently. Has there been a transcribing
In recent years, molecular methods have replaced error? Go back to the original record and compare with
lengthy biochemical investigations, with somewhat that in the report.
unceliain results, in the identification or exclusion of a Accurate organ weights in babies and children can pro
number of genetic metabolic disorders (GMDs) that might vide pointers towards natural disorders and may be useful
underlie some SUD!. That these investigations can be for corroboration when investigating postoperative deaths.
carried out on frozen tissue or even thick sections from An electric balance calibrated in O.l-g increments is appro
paraffin embedded tissue (especially spleen) is a priate and inexpensive. Gestation-related normal organ
particular advantage, as fluid samples can then be used for weights for comparison from 12 to 42 weeks' gestation are
other tests. those of Singer et al,27 Hansen et al 28 and Maroun and
Molecular methods on tissue samples can also detect Graem 25 (Appendix B) . Age-related organ weights in
the molecular rearrangements of some forms of long QT infants have been collated by Thompson and Cohle 29
syndrome. 21 (Appendix B). Organ weights in older children are more
As the potential for molecular investigation of more and difficult to access. The most recent papers, 30, 31 and Haddad
diverse genetic disorders continues it becomes essential to et ai, are statistical analyses of older data. Age-related
reserve both frozen and formalin-fixed wax-embedded weights of major organs are found in Altman and Dillmer32
material indefinitely when sudden death in infants and (Appendix B). The brain should be weighed before and after
children is initially unexplained or incompletely explained . fixation, being aware of a potential confounding factor
Another sample that has proved useful in the detection when different balances are used.
or exclusion of some GMD is the Guthrie card blood
spot. Wilcox et al 22 undertook tandem mass spectrometry
for acyl-carnitines. They identified three GMDs amongst
247 SUm. EXTERNAL EXAMINATION
Figure 8.5 Body surfaces with infant proportions aid accurate recording of injuries (after ref. 33).
be described carefully and photographed, with the appear Any tears or bruising of the frenula of the tongue and
ance (e.g. fro thy, fluid, dried) noted. When secretions are lips are recorded (Fig. 8.6), as well as scarring suggestive of
bloody, Krous et ae 7 urge careful examination of the mucous old injUly. The impression of teeth (even in an edentulous
membra nes of the nose and mouth with an a uroscope. infant) and patterned bruises of the inner aspects of the lips
152 I Post-mortem examination in babies and children
(b)
Figure 8.7 Mongolian blue spot above the gluteal fold in a baby
of Indian parentage.
'Girls' growth chart (birth- 18 years) 1996/1 Child Growth Founda tion.
Table 8.4 Histological appearance of subdural haematomata of different durations (after ref. 46)
of sum cases are detailed by Valdes Oapena et al40 and (10- to 12-cml, non-toothed forceps to mJlllmlZe tissue
KroUS. 41 Comment will be directed paJiicularly at those damage; a range of tapered, round-ended scissors (Mayos);
organs or regions for which a different approach or empha and a variety of scalpels, rather than an autopsy knife.
sis is appropriate in the older paediatric age group . The task Reference to the need for a digital balance was made in an
is easier when the following equipment is available: shOji earlier section .
154 I Post-mortem examination in babies and children
Forma l neck dissection to detect injUly is carried out in a in hospital with a high concentration of oxygen. Fluid in the
sim il ar fas hion to that in adults and invo lves layer-by-Iayer pleural and pericardial cavities is observed, measured and
examination of the anterior neck muscles. In the infant, it described. A sample of pericardium can be obtained at this
can be achieved following a vertical submental-symphysis stage using sterile inst ruments and placed in tissue culture
pubis incision as tissues stretc h readily but is probably eas Ouid for immed iate culture, if appropriate, or stored at
ier with a shoulder-to-shoulder appro ach. Posterior neck - 280°C for future availability for investigation of putative
muscles are conveniently inspected by ensuring that th e genetic abnormality. This is followed by sampling of heart
posterior scalp refection extends to the cervica l sp ines. blood, lung and myocardium for microbiological purposes.
The thorax is best approached by serial division of costal External examination of the heart is conveniently canied
cartilages, avoiding damage to the osteochondral junctions, out at this point by fully opening the pericardial sac. Cardiac
some of which will be submitted for histological examina abnormalities, both congenital and acqu ired, are common
tion, a nd has the advantage of leaving the ribs intact. Fresh causes of explained SUD! (see Chapter 11, pp. 208-210) and
rib fractures may be apparent following reOection of skin older children (see Chapter 12, pp. 226-235). Cardiac situs is
and soft tissue (Fig. 8.8). Resuscitation Jib fractures are dependent on atrial morphology. The comparative size an d
un com mon, and are usually situated at the anterior ends of relationsh ip of the ascending aorta and pulmonary trunk are
lower ribs; accompanying haemorrhage is minor and there noted - they should be equal in size. The course ofthe an te
is no vital reaction on histological examination. Rib frac rior descending branch of the left coronalY artery should be
tures are discussed further in Chapter 11, pp. 213- 215. observed and the connection of t he vena cavae a nd pul
Relationships of orga ns are observed before disse ction monary veins noted. If all are normal then congenital heart
commences, some abnormal relati onsh ips may suggest a disease is excluded, with the exception of septal defects and
syndrome diagnosis. anomalous pu.lmonary venous return to the coronary sinus.
The thymus is a prominent organ in early life, weigh ing The liver is relatively large in infants and protrudes
10 ::!: 4 g at term. It is of normal size (often described as below the costal margin. Colour and consistency are noted .
large) in sudden death. A reduction in thymic weight below Fatty infiltration and evidence of trauma are sought. The
th e normal ran ge for age is a non-specific indicator of ante former requires frozen section and histochemica l staining
mortem stress, probably in excess of 12 hours' duration 42 for confirmation. Preservation of samples to investigate the
Petechial haemorrhages v isible through the thymic capsule, possibility of GMD and full toxicological exami nation
particularly on its posterior aspect, a re present in about 80 should be considered.
per cent of SIDS 43 (see Chapter 11, p. 212). Larger, blotchy The stomach is often distended, either because of attempted
haemorrhages may be present in asphyxial death. 'bag and mask' ventilation or microbial fermentat ion. The
The appearance of the lungs, noting the degree of expan mesentery is inspected for tears and contusio ns (Fig. 8.9).
sion or collapse, is recorded. The lu ngs usua lly fill the chest in Completeness of intestinal rotat ion is observed, along with
SJDS but are often collapsed following attempted resuscitation the presence of volvulus, intussusception and herniae - all of
wh ich are potential causes of intestinal obstruction, shock
and sudden death - are sought.
Figure 8.8 Fresh rib fractures accompanied by bruising, visible Figure 8.9 Contusion in the mesentery close to the
after reflection of thoracic skin fl aps. duodenojejunal junction.
Dissection (infants and older children) I 155
Organs should be removed in blocks and not piecemeal. the lower ribs (6-10) is common when there are fractures
In particular, the heart should remain attached to the lungs elsewhere and is seen as expansion of the internal aspect of
until completely dissected. Th e Rokitansky technique is the bone adj acent to the cartilage (see Fig. 8.20, p. 160).
suitab le but the following proced ure is easier: removal of Stripping of the pleura to look for rib fractures in
(he intestines from the duodenojejunal fl ex ure to the rec infants is not recommended. The stripping process results
tum (having checked the mesentery for contusions) (Fig. in tissue dam age, including any subperiosteal reaction and
8.9), then rem oval of the thoracic and upper abdominal may render histological evaluation probl ematic.
viscera en bloc thus opening the inferior vena cava and Th e upper airway is examined carefully and may be
abdominal aorta, followed by removal of genitourinary fixed before opening when trauma to the neck has
tract. When sexual assault is suspected, a cuff of perineal occurred. Laryngeal fracture is unlikely but small haemor
skin is removed in continuity with the lower genital tract rhages may be present in the intrinsic lary ngeal muscles.
and anus, usually including the coccyx, the pubic area and FUliher microbiological samples can be taken from the
the med ial aspect of the thighs. Swabs for forensic exam lower trachea/main bronchi.
ination should be taken before dissection is begun . A convenient meth od of opening the heart is shown in
Following removal of the viscera, the thoracic and Figure 8.11. It is opened before separation from the lungs.
abdominal cavities should be cleared of blood and fluid Should a cardiomyo pathy be suspected, a horizontal slice
and their walls examined. The ribs should be carefully through both ventricles, midway between base and apex,
inspected for fractures. Rib fractures are often undisplaced demonstrates it well (Fig. 8.12) without interfering with
in babies and in fants. Attention will be drawn to fresh frac dissection; chamber connections and morphology, appea r
tures by subperiosteal an d intramuscular haemorrhage. ance of valves, septal defects and the origin and course of
Older fracture sites will be supported by callus and are usu coronary arteries are examined. The appearance and thick
ally stable (Fig. 8. 10), although re-fracture can occur. Cal ness of ve ntricular myoca rdium is noted. Any subendocar
lus is particularly marked on the internal aspect of the rib dial fibroelastic thickening or haemorrhage is noted. Heart
and is readily palp able. Fractures at the posterior ends of and lungs are separated and weighed. Pulmonary arteries
ribs are easily missed and should be specifically sought, are examined for evidence of thromboembolus and
with the remova l of pali of the rib cage fo r radiographic hypeliension - both causes of sudden death in early life.
and histological examin ation wh en there is suspicion of The former is most likely to be found in the poorly mobile,
injury. Previous avulsion of the osteochondral junction of postoperatively and follo wing the use of intravenous devices.
... ...
...
\
\
\
\
\
\4
\
\. \
- ...
" ...
\
\
\
... ... \
... ... \
\
\
"
"-,,- 2 ... ... \
" .... \
.... ...
....... ...
......... ...
"
Figure 8.11 Opening the heart, cutting lines are numbered
sequentially (from Fetal and Neonatal Pathology, 4th edn, 2007,
Figure 8.10 Thorax after evisceration; old posterior rib fractures p. 39, figure 2.20, with kind permission of Springer Science and
with ca llus are readily visible. Business Media).
1 56 I Post-mortem examination in babies and children
Table 8.5 External changes of maceration by death to delivery picking apart, and cutting with scissors or bl un t/sharp
interval (after ref. BB) knives - and assessment is aided by dissecting microscopic
examination. It may correspond with the free end of the
Observed feature Death to delivery
cord attached to a placenta discovered elsewhere. The pres
interval
ence or absence of a clip or tie is noted and described.
scrotum.
p. 188) is a useful observation. Some moulding of the cranium
Ch anges of maceration are accelerated by fetal hydrops and retarded in
is nOlmal but excessive moulding of the cranium is an indica
the presence of fetal growth restriction.
tion of long labour. In vertex presentation there is often an
area of localized oedema, usually with marked congestion and
early changes of maceration are ill ustrated in Figs 10.4- 10.6 sometimes with frank haemorrhage, over the presenting part.
p. 184. The presence of excessive meconium staining
indicates hypoxic fetal stress in the mature baby. If there is INTERNAL
much meconium, blood or vernix caseosa on the skin it is
advisable to remove it after initial photography so that the The scalp is incised from behind the ear over the posterior
skin can be examined in more detail. fontanelle to the opposite ear and reflected forwards and
Cutaneous pallor may be evident following significant backwards, sufficiently far to completely expose the bones
fetomaterna l haemorrhage. 53 of the cranial vau lt. The presence of haemorrhage within
The presence and nature of any cutaneous injury is the scalp is noted and a check is made for any corres
recorded and photographed. ponding external injury. Marked congestion of the scalp is
It is not unusual to find cutaneous petechial haemorrhages a very frequent finding in antepartum still births after
over the presenting part in fresh still birth, particularly when vertex presentation. It should not be confused with tra uma.
there is placental abruption. They are commonly found over Histological examination shows no vital reaction. The con
the face, head and neck and, occasionally, on the chest wall tour of the occipital bone is examined. Occipital osteodias
in a mature fetus when vertex presentation is usual. They are tasis , in which the inferior pali of the OCCipital bone is
often particularly prominent following placental abruption displaced inwards resulting in damage to the brain or
(see Fig. 10.10, p. 188) and should not be interpreted as evi venous sinuses, may occur after breech delivery or, occa
dence of stra ngu lation or deliberate airways obstruction siona lly, following forceps delivery;56 it is eas il y missed if
without corroborative evidence. the sca lp is not ful ly reflected posteriorly.
Cutaneous petechiae, and sometimes more extensive The posterior neck muscles are conven iently ex amined
bru ising, are seen on the legs following breech delivery or at this point and any contusion is photographed and sam
the arm if one has prolapsed through a partly dilated celvix. pled for histological examination.
These findings are more usual in the premature fetus, for The atlanto-occipital membrane sho uld then be incised
which breech presentation is more common. 54 At 32 weeks' under direct vision. Gentle pressure on the cranium wiiJ result
gestation approximately 16 per cent of babies present by the in a flow of CSF, which is likely to be heavily blood stained
breech, compared with around 5 per cent at term. when there is selious intracranial injury. The fontanelles and
The presence and nature of any dysmOlvhic features are suture lines are obselved. Congestion or haemon'hage is
noted. These may contlibute to a syndrome diagnosis, directly sometimes seen at the margins of sku ll bones and probably
related to death. The presence oftvvo or more dysmorphic fea indicates the stress of forcible displacement.
tures is an indication for chromosome examination.
The umbilical cord and its insertion are examined and Cranial Cavity
any discolouration or dehiscence is noted. The free end of
the cord may indicate the method of division. Differences The cranial cavity is opened using the modifi ed Beneke
in appearance of the cord ends are described 55 - traction, technique (Fig. 8.19), which allows examination of the
160 I Post-mortem examination in babies and children
Figure 8.20 Birth injury. Tear at the junction of the falx and
tentorium follo wing instrumental delivery, displayed by removal of
the cerebral hemisphere on the affected side.
la rynx
Trachea
Lung - five lobes, PBR reaction on all
Heart - right and left free ventri cu lar wall, interventricular
septum
Kidney - two
Liver - right and left lobes
Small and large intestine
Pituitary
Thyroid
Pan crea s
Adrenal - two
Figure 8.23 Birth injury. Bruising of anterior neck mu sc les. Gonad
Instrumental delivery. Diaphragm
Costochondra l junction
befo re accepting birth injury as a cause for rib fract ures in Any macroscopic lesions
an infant. Rib fractures are discussed further in Chapter 11. Dural fold
The commonest visceral injury is to the liver, when rup Dura - parietal
ture of a subcaps ular haematoma can result in catastrophic
In SUOI (including sudden neonatal death)
haemo rrhage. The baby is usually mac rosomic. A simil ar
Frozen section for oil red 0 staining from:
injulY is seen in preterm infants, particularly following
He art
breech delivelY, as the liver is poorly protected by the rib
Liver
cage at this time. Splenic rupture is a rare birth injury.
Ki dney
Adrenal
HISTOLOGICAL SAMPLES ~L u ng
In perinatal death
Thorough sampling for histological examination is an
Pla centa X 2
essential part of perinatal, infant an d childhood post
Umbilica l cord
mortem examination. In ea rly life, much diagnostic infor
Extra placental membranes
mation is derived from microscopi c examination when
macroscopi c ap pearan ces may be normal, non-specific or
not understood by the inexperienced. Sampling should fol lung maturity. An elastic stain, (Elastic van Gieson is my
Iow a standard protocol (Table 8.6). It is the in vestigation preferred technique) allows assessment of postnatal
most likely to make a positive contribution in SUDl. 2o adaptation of pulmon ary vasculature, ventilation-induced
Norma l his to logical appearances of organs in the infant damage and other residua of neonata l intensive care.
period are illustrated in Valdes-Dapena et al. 40 Exa mination of the ca rdiac conduction system is desir
In addition to convent ional sectio ns of fo rmalin-fixed ab le following instantaneo us death or observed rapid col
para ffin-emb edded samples, it is important to reserve sm all lapse that is unexplained after conventional investigation.
fresh samples of heart, liver, kidney and adrenal gland for It should be considered when sudd en death follows cardiac
frozen sections in order to look for lipid infiltration ; when surgelY and when a second sudden death occurs within the
small in amount, lipid infiltration is probably st ress-related, family. In these circumstances the heart should be retained
but massive lipid infiltration is su ggestive of genetic meta complete with this in mind. Conventional sampling is
bolic disease (see Chapters 11 and 12). These sa mples can be desc ribed and illustrated by Davies et al. BO Michaud et alBI
kept frozen, perhaps wit h a sample of spleen, for molecular and Ashworth B2 describe a simplified method for analysis
studies should the need arise. If inhalation is suspected, a of the conduction system.
froze n sample of lung for lipid staining is also useful. It is imp ortant no t to over interpret minor differences
In SUDI, it is advisable to examine lung sections stained in microanatomy, either in distributio n of conduction tis
fo r iron (PBR) to assess the possibility of previous pulmonalY sue B3 or of the process of so -called 'resorptive degeneration'
haemorrhage. It may be useful in other circum sta nces, Wh en of the atrioventricular nodal tissue and the His bundl e. 83 ,84
death occurs suddenly in the neonata l period, staining for The so-call ed marke rs of SIDS, pulmonary haemosiderin
reticulin is very usefu l in the evalu-ation of microscopic and the significance of minor degrees of infla mmation in
lung anatomy: it permits exclusion of congenital alveolar SUD! are discussed in Chapter 11, pp, 209-212, 215-2 16
dysplasia and better assessment of the appropriateness of an d p. 210, respectively.
Exchange of information and multidisciplinary review I 163
in respect of child protection a nd in our understanding of 21 Ackerman MJ, Tester DJ , DriscoU DJ. Molecular autopsy of
causes and contributing factors in sum, putting in place sudd en unexplained death in the you ng. Am] Forensic Med
Patho1 2001; 22:105-11.
app ropriate support for famili es and in the appropriate plan
22 Wilc ox RL, Nels on CC, Stenzel P, Steiner RD. Postmol1em
ning of selvices for chiJdren.o 1 screening for fatty ac id oxidati on di sord ers by analysis of
Guthrie ca rds with tand em mass spec trom etry in sudden
unexpected death in infancy.] Pediatr 2002; 141 :833-6.
23 Yudkin PL, Aboua lfa M, Ey re JA et al. The influence of
REFERENCES
elective preterm delivery on birth- weight and head
ci rcu mference sta ndard s. A rch Dis Child 198 7; 62: 24-9.
Vujanic GM Cartlidge PHT, Stewart JH. Improving the quality 24 Chambers HM, Knowles S, Staples A et al. Anthropo metric
of pe rinatal and in fant necropsy examinations: a follo w up measurements in the seco nd tri mester fetu s. Early Human Dev
study.] Clin Patho/1998; 51:850-3. 1993; 33:45-59.
2 Sheehan KM, McDonnell M, Doyle EM et a!. The qua li ty and 25 Maroun LL, Graem N. Autopsy standards of body pa rameters
val ue of sud den in fan t death necropsy reporting in Irelan d. and orga n weights in non-macerated an d macerated human
] Ciin Path 01 2003; 56:753 - 7. fetuses. Pedia tr Deuel Patho/2005; 8:204-17.
3 Bass M, Kravath RE, Glass L. Death-scene investiga tion in 26 Freeman N , Co le TJ, Chinn S, Jones PR M, White EM, Preece
sudden infant death . N Engl] Med 198 6; 315:100-5. MA. 1995 Cross-s ecti ona l stature and weight refere nce CUlves
4 St urner WQ. Commo n errors in forensic pediatric pathology. for the UK, 1990. Arch Dis Child 73:17- 24.
Am] Forel1 sic Med Pat110/1998; 19:317 -20. 27 Singer D, Sung CoR, Wiggleswonh JS. Feta l growth and
5 Bass M. The falla cy of the sim ultaneous sudden infant
maturation: with stand ards for body and organ development.
dea th syndrome in rwi ns. Am] Forensic Med Pathol 1989;
In Wigglesworth JS , Singe r D (eds) Textbook of Fetal and
10:200-5.
Perinatal Pathology, 2nd edn. Ma lden, MA: Blackwell Sc ience,
6 Sa ukko P, Knight B. The pathophysiology of death. In 1998, pp. 8-40.
Kn ight's Forel1s ic Pathology, 4th edn. London: Arnold , 2004, 28 Hansen K, Sung CJ , Huang C et al. Referen ce values for
pp. 52-97. second trimes ter fetal and neonata l organ weights a nd
7 Ca rpenter RF. The prevalence and distribution 0 f bruising in meas urements. Pediatr Develop PatI101 2003; 5:160-7.
babies. Arch Dis Cilild 1999 ; 80:363 - 6. 29 Thompson WS, Cohl e SD . Fifteen-year retrospective stud y of
8 Sugar NF, Taylor JA, Feldman KIN et a1. Bruises in infa nts and infant organ weights a nd revisio n of sta ndard weight tabl es.
toddlers. A rch Pediatr Adoles c Med 1999; 153:3 99- 403 . ] Forensic Sci 2004; 49:1-11 .
9 Moon RY, Patel KM, Schaefer SJ. Sudden Infant Death 30 Shan kle WR, Landing BH, Gregg J. Normal organ weights of
Syndrome in child care settings. Pediatrics 2000; 106:295-300. infa nts and children: Graphs of va lue by age, with confide nce
10 Maxeiner H. Demo nstrati on and interpreta tion of bridg ing intervals. Pediatric Patho l 1983; 1:399-408.
vein rup tures in cases of infantile subdura l bleedings. 31 Haddad S, Res tieri C, Krishn an K. Ch aracte rization of age
] Foren sic Sci 2001; 46:85-93. related cha nges in body weight and orga n weights from birth
11 Erlich E, Maxeiner H, Lan ge J. Postm ortem radiological to ado lescence in humans. ] Toxic ol Enviroll Health 2001 ;
in vestigation of torn bridging vein ruptures. Legal Med 54:453-64.
(Tokyo) 2003; 5lSupp1. 1) : 225-7 . 32 Altma n PL, Dittmer DS. Growth: In cluding Reproduction alld
12 Sadler DW. The value of a thorough pro tocol in the in vestigation Morp hological De velop men t. Washington DC: Federation of
of sudden infant dea ths.] Ciin Pathol 1998; 51 :589-94. Ame rican Socie ties for Exp erime ntal Biology, 1961.
13 Wigglesworth JS , Keelin g JW, Rushton D, Berty PJ. Tech nica l 33 Handy TC, Nichol s GR , Buchin o JJ. A pediatric fo rensic
method. Pathological inves tigations in cases of sudd en infan t medicine prog ram. In Dimniick JW, Singer DB (eds) Forens ic
death.] Ciin Pat110/198 7 ; 40:1481-3. A spects ill Pediatric Pathology. Perspectives in Pediatri c
14 Dettm eyer R, Baas ner A, Schlamann M et a!. Coxsa ckie BJ Pathology. Basel: Ka rger, 1995, pp. 87-95.
myocarditis in 4 cases of suspected sudden infant death 34 Kaplan JA, Fossum RM. Patterns of facial resuscitation injury
syndrome: diagnos is by immunohistochemical and molecular in in fan cy. Am ] ForellS ic Med Path o/199 4; 15:187-91.
pa thologic invest igation. Patlwl Res PraU 2002; 198 :689-96. 35 Downes AJ, Crossland DS, Mellon AF. Prevalence and distribution
15 Bajanowski T, Rolf B, Jorch G et al. Detection of RNA viruses in of petechiae in well babies. Arch Dis Child 2002; 86:291-2.
sudden infant death (SID), Int] Legal Med 2003; 117 :237- 40. 35 Byard RW, Krous HF. Petechial hemorrhages and unexpec ted
16 Carmichael EM, Gold wate r PN, Byard RW. Routin e in fant death. Legal Med (Tokyo) 1999; 1:193-7.
microbiological testing in sudden and un expected infant 37 Krous HF, Nad eau JM, Byard RW, Blackbourn e BD . Oronasal
death. ] Pediatr Child Health 1996; 32:412-15. blood in sud den infant death. Am ] Fore nsic Med Path ol
17 Mitchell E, Krous HF, Donald T, Byard RW. An ana lysis of the 200 1; 23 :346-51.
usefuln ess of spec ific stages in the pathol ogic investigation of 38 Mayes C, Macleod C. When 'N AJ' means not act ually injured.
sudd en infant death . Am ] Forensic Med Pat flol 2000; BM] 1998; 318:1127 - 8.
21 :395-400. 39 McCann J, Reay D, Siebert J et al. Postmortem perianal findings
18 Sims MA, Collins KA. Fetal death: a 10-year retrospective in children. Am] Foren sic Nled Patho/2006; 17:289-98.
study. Am] Forensic Med Pathol 2001; 22:251 - 5. 40 Valdes-D ape na M, McFeel ey PA, Damus KH et al.
19 British Med ica l Associatio n. British Natiol1al Formulary, Histopa thology Atlas fo r the Sudden Infant Dea th Syndrome:
Paediatric. Londo n: Roya l Pharm aceutical Society of Great Fi ndings derived from the Natio l1 al/llstitute of Child Health
Britain, 2005, p. 225. and Hum an Develop ment Cooperati1!e Epidemiological Study
20 Berty J, Allibone E, McKeever P et al. The pathol ogy study: of Sudden Infant Death Syndrome (SIDS) Risk Factors.
The contribution of ancillary pathology tests to the Washington : Armed Forces Institute of Pathology, 1993.
investiga tion of unexpected infant death . In Fleming P, Bacon C, 41 Kro us H. The interna tional sta ndardize d autopsy protocol for
Blair P et al (eds) Sudden Unexpected Deaths in Infancy the sudden unexpected infant death. In Rognum TO (ed.) Sudden
CESDI SUDI Stu dies 1993-1996. London: The Statio nery Infant Death Syndrome. New Trends in the Nineties. Os lo:
Office, 2000, pp. 97-11 2. Sca ndin avian Uni ve rsity Press, 1995, pp. 81-95.
References I 165
';1 van Baarlen J, Schuurman H-J, Huber J. Acute thymus 67 Altshuler G. Placenta w ithin the medicolegal imperative. Arch
involution in infancy and childhood: a reliable marker for Pathol Lab Med 199 1; 11 5:688 -95.
duration of acute illn ess. Hum Pathol 1988 ; 19: 1155-60. 68 Macphe rson 1. Fact a nd fancy. What can we really tell from
";3 Beckwith JB. Intrathoracic petechial hemorrhages; a clue to the placenta? Arch Path Lab Med /991; 115: 672 -81.
the mechanism of death in sudd en infa nt death sy ndrome? 69 Wigglesworth JS. Pathology of intrapartum and early
Ann Acad NY Sci \988; 533:37-47. neon atal death in the normally form ed infa nt. In
~4 Muir R, Niven JSF. The local formation of blo od pigments. Wigglesworth JS, Singer DB (eds) Textbook of Fetal and
J Pathol Ba cteriol 1935; 41 :183-97. Perinatal Pathology, 2nd edn . Ma lden , MA: Blackwe ll Science,
-lS Keeling JW. The perinatal n ecropsy. In Keeling JW, Khon g TY pp . 251-68.
(eds) Feta l and Neona ta l Pathology, 4th edn. London: 70 Keeling JW. Intrapartum problems. In Keelin g JW, Khong TY
Sprin ger, 2007, pp. 20-53. (eds) Fetal and Neonatal Pathology, 4th edn. London:
46 Hardman JM. Cerebrospinal trauma . In Davis RL, Ro bertson Springer, 2007, pp. 273-96.
DM (eds) Textbook of Neuropathology, 3rd edn. Baltimore, MD: 71 Teng FY, Sayre JW, Vacuum extraction: does duration predict
Williams Et Wilkins, 1997. scalp if\iury? Obstet Gynecol 1997; 89 :281-5.
47 Parsons MA, Smart RD. Necropsy techniques in ophthalmic 72 Amar AP, Aryan HE, Meltzer HS, Levy ML. Neonatal subgaleal
pa thology. Best Practice 164 J Clin Path 01 2001; 54:417-27. hematoma causing brain compression: results of two cases and
48 Lee WR (ed.). Examination of the globe. Techni cal aspects. [n review of the literature. Neurosurgery 2003; 52: 1470-4.
Oplithalmic Histopathology, 2nd edn. London: Springer 73 Robinson RJ , Rossiter MA. Massive subaponeurotic hae monhage
Verlag, 200 2, pp. 1-33. in babies of African origin. Arch Dis Child 1968 ; 4 3 :684-7.
49 Reichard RR, White CL Ill , Hogan RN et at. Beta-amyloid 74 Doward W, Sgouros S. Acute su bdural haematom as follo wing
precursor protei n immunohistoch emistry in the evaluation of Ventouse-a ssisted delivery. Pediatr Neu rosurg 2001; 35:336,
pediatric traumatic optic nerve inju ry. Ophthalmology 2004; 75 Towne r D, Castro MA, Eby-Wilkens E e t al. Effect of mode of
111 :822-7. delivery in nulliparous women on neonatal intra cranial
50 Wigglesworth JS. Performa nce of the perinatal autopsy. injury. N Engl J Med 1999; 341: 1709-14.
In Wiggleswowrth J S (ed.) Peri/lOtal Pathology, 2nd edn. 76 Vachharajani A, Mathur A. Ultrasound-guid ed needle
Phil adelphia , PA: Sau nders, 19 86, pp. 23-40. aspiration of cranial epidura l hema toma in a neonate:
51 Russell JGB. Radiolog ical assessment of fetal maturi ty. trea ting a ra re complication of vacu um extraction. A m J
J Obstet Gynae Brit Cwlth 1969; 76:208-19. Perinatoi2002; 19:401-4.
52 Ste mpfle N, Huten Y, Fondacci C et al. Fetal bone age 77 Perri n RG, Rutka JT Drake JM et al. Ma nagement and
rev isited: proposal of a new radiographic score. Pediatr Radiol outcomes of posterior fo ssa subdural hematomas in neonates .
1995 ; 25:551-5. Neu rosu rgery 1997; 40:1190-9; discussion 1199- 2000.
53 Biankin SA , Arbu ckle SM, Graf NS. Autopsy findings in a 78 Whitby EH, Griffiths PO, Rutter S et al. Frequency and na tural
series of five cases of feto maternal haemorrh age. Pathology histo ry of subdural haemorrhages in ba bi es a nd relation to
2003; 35:319-24. obstetric factors . Lancet 2004; 363 :846-51.
54 Ha rlow FHD, Spencer JAD. Obstetrics for the neonatologist. In 79 Salman M, Crouch man M. What ca n caus e subdura l
Rennie JM, Roberton NRC (eds) TeXTbook of Neonatology. haemorrhage in a term neo nate? Paediatr Today 1997 ;
London: Churchill Livin gston e, 1999, pp. 157-77. 5:42-5.
55 Morris JF, Hunt AC. Breaking streng th o f the umbilical cord. 80 Davies MJ , Pomerance A, Lamb D. Tec hniques in exa mination
J Forensic Sci 1966; 11:43-9. and a natom y of the heart. In Dav ies MJ , Pomerance A (eds)
56 Wigglesworth JS, Husemeyer RP. Intracra nial birth trauma in The Path ology of the Hea rt. Oxford: Blackwell Sc ientific
vaginal breech delivery: the continued importa nce of injury Pu blica tions, 1975, pp. 26-31.
to the occipital bone. Br J Obstet Gynecol 1977; 84:684-91. 81 Michaud K, Romain N, Taroni F et al. Evaluation ofa
57 Dorovini-Zis K, Dolman CL. Gestational develo pment of brain. simplified method of th e conduction system a nalysis in 100
Arch Pathol Lab Med 1977 ; 101 :192-5. forensi c cases . Forensic Sci Int 2002; 130:13-24.
58 Feess-Higgins A, Lan'oche J-c. De1Jelopmel1 t of the Human 8 2 Ashworth MA. The cardiovascular system. In Keeling JW,
Foetal Brain. An Anatomical Atlas. Pa ris: Inserm CNR S Khong TY (eds) Fetal and Neonatal Path ology, 4th edn.
Masson, 1987. London: Springer, 2007 , p. 613.
59 Ya tes PO. Birth trauma to the vertebral arteries. A rch Dis 83 Ho SY, Anderson RH. Conduction tissue and SIDS. Am NY
Child 1959; 34:436-41. Acad Sci 1988; 533:176-90.
60 Marchet ti D, Belviso M, Marino M, Gaudio R. Evalu ation of 84 Ma tturri L, Otta viani G, Ramos SG, Rossi L. Sudden infant
the pl acen ta in stillborn fetuses to estimate the time of death. death syndrome (SID S): A study of cardiac conduction
A m J Forensic Med Pat/wi 2007; 28:38-43. system . Cardiovasc Pat1lO1 2000; 9:137-45.
61 Pinar H, Sung CJ , Oyer CE, Si nger DB. Reference values for 85 Keeling JW, McKenzie KJ, Ironside J et al. Does detailed
singleton and twin placental weights. PeliiMr Patlwl Lab Med neuropathological ex amina tion o f the fi xed brain in sudden,
1996 ; 16 :901-7. un expected dea th in infa ncy contribute to the evaluation of
62 Naeye RD. Umbilical cord length: clini cal significance. the death? J Pat/wi 2002 ; I 98(Suppl.) :2 4A.
J Pedi atr 1985; 107:278-81. 86 Report of a Working Group . Sudden Unexpected
63 Genest DR. Estimating th e tim e of death in stillborn fetuses: Death in Infancy. London: Royal College of Path ologists
II. Histologi c evaluation of the pla centa; a study of 71 and Royal College of Paedi atrics a nd Child Hea lth,
stillborns. Obstet Gyneco11992; 80:585-92 . 2004.
64 Benirschke K, Kaufmann P, Bae rgen R. Pathology of the 87 Fleming P J, Blair PS, Sid ebotham P, Hay ler T. Investiga ting
Human Place nta , 5th edn. New York: Springer, 2006. sudden unexpected deaths in infancy and childh ood and
65 Fox H, Sebire N. Pathology of the Placen ta, Major Problems in ca ring for bereaved families: an integra ted multiagency
Pat/wlogy, 3rd edn. London: Saunders, 2007. appro ach. BMJ 2004; 328:331-4.
66 Khong TY. The placenta and umbilical co rel. In Keeling JW, 88 Genest DR , Singer DB. Es timatin g the time o f death in
Khong TY (eds) Fetal and Neonatal Pathology, 4th edn. stillborn fetuses: JII. Extern al fetal examination; a study of
London: Springe r, 2007 , pp. 54-89. 86 s tillborns. Obstel Gynecol 1992 ; 80:593-600.
I CHAPTER 9 I
PATHOLOGY OF NEUROLOGICAL
ABNORMALITY IN EARLY LIFE
Waney Squier
INTRODUCTION
may be very strongly held. The findings in the brain are of
paramount importance in such cases but may also be very
Examination of the nervous system forms a critical part of subtle. If trauma is suspected the possible mechanisms
the autopsy exami nation of any infant who dies, whether must be fully considered and due attention paid to exami
suddenly and unexpectedly or following recognized illness. na tion of the stru ctures of the neck and spinal cord as well
The cause of sudden infant death is not always proximate as the brain. Meticulous neuropathological examination
insult or disease; pre-existing diseases of, or damage to, the can identify no t only the cause of neurological disease in
nervous system may cause sudden death. Babies with neuro childhood but, in the case of acquired damage, can also
logical disease, for example, are more pron e to aspiration of assist in evaluating when the damage has occurred. How
feeds and to epilepsy. Even quite extensive congenital brain ever, the effects of the terminal condition of the baby must
damage acquired in utero or at the time of bilih may not be taken into consideration. Hypoxia and resuscitation, as
present clinically until months or years later, while still well as a period of ven tilation , will lead to brain swelling
hav ing the potential to cause sudden death . It may be that a and ge neral organ failure, disturbance of blood clotting,
'secon d insult' is required, for example velY mild brain mal leaking from blood vessels and slowing or cessation of cel
formation (microdysgenesis) is seen in patients with epilepsy lular reactive processes, serious ly hampeling the ab ili ty to
but may not cause seizures until there is another insult such time the injUIY. Careful correlation with brain scans take n
as trauma. Similarly, babies with metabolic or mitochondrial closer to the time of coll apse will assist in distinguishing
disease may become profoun dly unwell a nd die when chal primary injUly from these secondary changes.
lenged by a respiratory infect ion. In these babies careful Neurological abnormalities may resu lt from genetic or
exa mination of skeletal muscle with histochemistlY and meta bolic disease, diseases acquired in utero, during deliv
review of pre-mortem biochemistry is necessary if an accu ery, in the pelinatal period or in early life.
rate diagnosis is to be made. Chronic subdural haemorrhage Intra-uterine damage may not be symptomatic until
may go quite unrecognized for weeks or months until dis weeks or months after birth, for example cerebral palsy is
covered by routine head circumference measurement or not usually definitively diagnosed until 5 years of age. Con
even by parental observation of increasing head size. versely, babi es damaged immediately before or during bilih
The pathologist making an autopsy exa mination in sud tend to exhibit signs or symptoms such as floppiness,
den infant death is faced with an awesome task. Th ere will depression and asphyxia in the first days of life.
often be great pressure from those who have cared for the Once the brain has been damaged there will be tissue
child in life to find a pa rticular cause of death . Opinions loss followed by atrophy. Tissue regeneration and plasticity
concerning the possibility of natura.1 or unnatural causes due to compensatory hyperplasia has been demonstrated in
Timing of injuries by histology I 167
the immature human brain.l Following even static and fragments of cerebellum in the su barachnoid space at mul
non-progressive insult, clinical signs may evolve due to tiple levels of the spinal cord.
atrophy of associated and connected brain areas. Histologically, oedema is better seen in densely packed
white matter tracts than in grey matter.
the brain) are prominent and are recognized by their long, This is stainable in Perls' reaction 48-72 hours after haemor
rectangular nuclei. These are replaced by macrophages in up rhage. In late stages, iron is seen in astrocytes or free within
to 2 days after injury and by 3 days these cells may contain tissues as well as in macro phages. It is common to find
phagocytosed debris or red cells within their cytoplasm. specks of haemosiderin within the dural layers in infants.
Damaged tissue is removed by these phagocytes, which
frequ ently cluster around blood vessels where they may Cyst Formation
remain for months or years . If there has been haemorrhage
then red cells are seen in the cytoplasm of macrophages in Tissue breakdown and removal by macro phages causes cysts
the early stages; they are later converted to haemosiderin, to develop, usually \0 days to several weeks after injury.
which is brown in haematoxylin and eosin (HE) stain but
after about 48 hours may be detected with Peds' stain.
Mineral ization
Endothelial cell swelling is a helpful and sensitive marker Damage to the axonal membrane causes interruption of the
of tissue damage, becoming prominent 1-3 days after intra-axonal transport of proteins, leading to irregular
Acquired intra-uterine damage I 1 69
Early .-.
.
Figure 9.4 Periventricular leucomalacia. Small focus of
Damage in the first 20 weeks of gestation causes malfor periventricular leucomalacia in an infant who died 11 days after
mation. Specific malformations may indicate the timing of severe hypoxic-ischaemic injury. There is a small collection of
the causative insult as structures are malformed only dur glial cells and macrophages, among which are proliferating
ing th eir period of formation . Once formed, structures may capillaries and axonal spheroids. (a) HEtE; (b) C034, showing the
show evidence of destruction. An example is seen in the newly formed capillaries; (c) ~ APP stain, showing the axonal
corpus callosum, which develops from front to back swellings in the infarcted area.
170 I Pathology of neurological abnormality in early life
PORENCEPHALY
by ectopic neurones, and the adjacent cortex may be It has long been accepted that white matter disease results
dysplastic. Schizencephalic clefts are often bilateral, from ischaemia of the developing white matter. However,
and most commonly found in the region of the Sylvian fIs in recent years intra-uterine infection and inflammation
sure. Although originally considered to be a malformation have been implicated in its pathogenesis. Epidemiological
due to a focal defect in the germinal matrix,1O schizen studies have demonstrated an association between placen
cephaly is now generally considered to be a destructive tal and amniotic infection and maternal pyrexia and
lesionY Rare familial cases are described. Mutations in the neonatal neurological abnormality. IS Histological evidence
EMX2 gene were implicated but this association has not of inflammation in the placenta is associated with brain
been confIrmed. 9 lesions and later cerebral palsy. J9
Polymicrogyria is a cOliical malformation resulting Three pattems of white matter damage are described. Com
from damage prior to 28 weeks of gestation. iJ.12 Macro mon to all is gliosis and capillary proliferation. When seen
scopically, gyri are irregular and small , but the diagnosis alone this is described as telencephalic leucoencephalopathy.2o
can be confirmed only by histology. Two histological pat The most common form, pellventllcttiar leucomalacia, is
terns are described: unlayered and four-layered, but their recognized by multiple infarcts in the deep periventricular
distinction has no aetiological significance and they com white matter that can be seen by the naked eye as tiny cysts,
monly coexist. A common and determining feature of often rimmed by a yellow mineralized zone (Fig. 9.6). The
polymicrogyria is abnormal fusion of adjacent gyri, lead infarcts tend to occur in the deep frontal white matter and
ing to continuity of the superficial cortical surface while around the trigone. Histologically, they appear as focal glial
the neuronal laminae form festoons below. Large blood scars - centrally cystic and surrounded by macro phages and
vessels are trapped in the seams of fused gyri. It is not a damaged axons, which are frequently calcifIed (see Fig. 9.3,
uniform entity; several specific patterns of polymicrogyria p. 169). Recent lesions are accompanied by capillary prolif
are described, both by histology and macroscopic distribu eration (see Fig. 9.4).
tion. Acquired polymicrogyria occurs randomly throughout The most dramatic manifestation of white matter dam
the cortex or on the borders of destructive lesions, whereas age is multicystic leucoencephalopathy (Fig. 9.7), in which
symmetrical and bilateral patterns may correlate with spe large areas of damaged white matter are replaced by cysts,
cific genetic disorders. I] Polymicrogyria and schizen
cephaly are both described after ischaemic events and
infections such as CMY.
Classification systems of cerebral malformations based
on MRI studies are not always accurate, owing to the limi
tations of MRI resolution. What may appear to be specific
malformations on imaging may be both structurally and
aetiologically heterogeneous.
Late
- - .- - ~
Birth·related injury I 1 73
BIRTH-RELATED INJURY
Figure 9.9 Acute profound hypoxic-ischaemic injury. Basal
Brain injury acquired at birth is usually due either to trauma
ganglia damage. (a) Fresh brain slice showing dusky
or to asphyxia. Today, birth trauma is relatively uncommon
discolouration of the deep grey nuclei, particularly the lentiform
owing to improved obstetric care and high rates of Cae
nuclei (arrows). The cortex appears as a wh ite ribbon and the
sarean section. Despite a fivefold increase in Caesarean sec
ventricles are narrow, owing to oedema. (b) High-power image of
tions over the last 30 years, the rates of cerebral palsy have
the thalamus, many months after hypoxic-ischaemic injury,
remained constant in industrialized countries? with many
showing neuronal loss and dense gliosis. Many remainin g
of those affected being low-birth-weight or premature
neurones are mineralized (haematoxylin and eosin).
babies who survive with white matter damage.
SUBDURAL HAEMORRHAGE
Method of Delivery
Most subdural haemorrhages follow normal delivery, but
are proportionally more frequent after forceps and ve ntouse
deliveries,37 although, in the above studies, not after Cae
sarean section.
Location
The source of bleeding in bilth-related subdural haemor
rhage remains uncertain. It is striking that the great maj ority
of birth-related subdural haemorrhage is found in the poste
rior fossa or around the occipital poles, close to the tento
rium. 36 .37 The preferential location at the back of the brain is
likely to relate to the origin of bleeding in the dural folds.
The dural sinuses and their tributaries have traditionally
been considered the origin for bilth-related subdural haem
orrhage,J9 even in the absence offrank tentori al tearing. The
Figure 9.13 Chronic subdural haemorrhage. (a) Very vascular
falx has particularly prominent venous sinuses within it 40
membrane (M) adherent to deep surface of dura (D) . Note the
and bleeding here is associated with asphyxia. 41 lntratentor
large numbers of capillaries at the junction wi th the dura and
ial bleeding in asphyxiated infants has been demonstrated
at the free edge (CD34 immunocytochemistry). (b) Subdural
by CT scan and autopsy correlative studies. 42
haemorrhage 4 months after head injury. There are collections of
Another consideration in the location of these bleeds is
fresh red cells and macrophages with pigmented cytoplasm. Peris'
gravitational redistribution when babies are nursed, and
positive material is seen within the macrophage cytoplasm and
indeed scanned, on their backs.43 Blood also tracks into the
free in the tissue, deposited on fibrous strands. These features
subdural space around the spinal cord and is frequently
indicate bleeding of several different ages (Perls' stain).
identified at autopsy as a crescentic collection over the pos
terior cord at lower spinal levels.
Subdural haemorrhage is occasionally found overlying
an area of cerebral infarction in the term infant, usually in
Natural History of Infant Subdural Haemorrhage
the middle cerebral artery territory.44 The pathogenesis is
unknown, but Steinbok et al 44 considered the most likely The natural histOIY of subdural haemorrhage is that it
sequence to be cerebral infarction, with secondary subdural resolves by forming a granulating memb ra ne (see Fig. 9.13).
haemorrhage resulting from rupture outwards of a haemor This membrane contains wide capillaries that may rebleed.
rhagic infarct. Animal studies have suggested an alternative The characteristic histology of a healing subdural membrane
explanation: that subdural blood damages the underlying is a layer of fibroblasts containing fresh and old red cells,
brain. 45 pigmented macrophages and haemosiderin consistent with
Other rare causes of subdural haemorrhage include vas repeated episodes of rebleeding (the chronic active subdural
cular malform ations, meningitis and after neurosurgical haemorrhage) (Fig. 9.13). Birth-related subdural bleeds were
intervention. Very rarely, subdural haemorrhage compli seen to resolve on a single MR scan at 4 weeks in 9 cases. J7
cates metabolic diseases associated with brain atrophy, However the membranes may persist, Rogers et al 47 found
such as Menkes disease and glutaric aciduria in older evidence of chronic subdural membranes in up to
infants. 46 31 per cent of sudden infant death syndrome (SmS) case
176 I Pathology of neurological abnormality In early life
rhage in life.
Some infant subdural haemorrhages develop into
chronic fluid collections th at are visible on brain scans. The
mechanism is unknown but may be related to immaturity
.
of the arachnoid villi or rupture of the arachnoid mem
brane, allowing leakage of CSF into the subdural space,43
osmotic accumulation of fluid or repeated small bleeds. 48
EXTRADURAL HAEMORRHAGE
Early studies have noted spinal cord and nerve root injury Unilateral cerebral artery infarction is common, occurring
after birth.28 ,5 1 Yates 51 described damage to the cervical in 1 in 4000 term infants. The middle cerebral artery is
spine in almost one-half of neonatal autopsies, with nerve most commonly involved. The damage is due to failure of
Infections I 177
blood supply through a m ajo r v essel, which may be due to usually h ave raised lactate in their blood and CSF, which
its blockage by a b loo d clot or tissue fragment, for exam may be exacerbated on exercise or when stressed by inter
ple fronl the placenta, or by externa l compression. Ex tern al curre nt infections. Clin ical presentation includes poor
co mp ression is rare. The cause of vascular obstruction is feeding, vomiting and exercis e-induced lactic acidosis, res
not often demonstrated at autopsy. piratory deficiency and lethal ap noea. 67
Stroke may occur in the an ten atal or perinatal periods Autopsy examination of skeletal muscle may s how typ
period or at the time of birth, and the timing of the damage ical ragged red fibres with Gomori's tr ichrome stain, abno r
may be reflected in the clinical presentation. Ba bies injured mally increased succinate dehydrogenase ac tivity and
in the perin atal perio d tend to present early with neo natal cytochrome oxidase deficiency in many fibres. However,
seizures, whereas infants damaged in utero present later th ese changes a re less frequ ent in infant than in ad ult cases
and develop hemiplegia and intractable epilepsy after sev and diagnosis depends on mitoch ondrial DNA an alys is.
eral months of pos tna tal life. 53 However, many strokes are Brain examination may show a number of abnorma lities.
clinica lly sile nt at the time w hen they occur. In Leigh's syndrome there are typ icall y symmetrical lesions
The aetiology is obscure and prob ably multifactoriaP4.55 of tissue rarefaction, gliosis and capiUary proliferation in
Overall, 30 per cent of neonatal stroke is due to v eno us sinus the basal ganglia and brainstem. In other cases, for example
thrombosis. 56 Factors associated with neonatal stroke pyruvate dehydrogenase deficiency, there may be migration
include: blood clo tting disord ers in fec tions, placental dys disturbances, congen ital cerebral at rophy an d ventricular
fun ction, growth retardation, pre-eclampsia, tw in pregnancy, dilatation. 67
peripartum asphyxia and maternal thrombophilia with
thrombotic lesions in the placen ta. 54.57 ,58 Hypoxic-ischaemic
injlllY is the most common association.59 Lipid Storage Diseases
congenita l clefts in the cerebral mantle. l. Clefts with fused birt h injuri es in term newborn infants. Pediat/' Neurosu rg
11 Barkovich AJ, Kjos BO. Schizencephaly: correlation of clinical 32 Towner D, Castro MA, Eby-Wilkens E, Gilbert VVM. Effect of
findings with MR characteristics. AJNR Am J Neuroradiol mode of delivery in nulliparous women on neonatal
'1992; 13:85-94. intracranial injury. N Engl J Med 1999; 341:1709-14.
12 Bar1h PG. Migrational disord ers of the brain. Curl' Opin 33 Chamnanvanakij S, Rollins N, Perlman JM. Subdural
Neuro! Neurosurg 1992; 5:339-4 3. hematoma in term infants. Pediotr Neural 2002; 26:301-4.
References I 179
34 O'Mahoney F, Settatree R, Platt C, Johnson R. Review of 51 Yates PO. Birth trauma to the vertebral at1eries. Arch Dis
singleton fetal and neonatal deaths associated with cranial Child 1959; 34:436-41.
trauma and cephalic delivery during a national intrapartum 52 Rutty GN, Squier WM, Padfield CJ. Epidural haemorrhage of
related confidential enquiry. BJOG 2005; 112:619-26. the cervical spinal cord: a post-mortem artefact? Neuropathol
35 Geddes JF, Tasker RC, Hackshaw AK et a\. Dural haemorrhage Appl Neurobiol 2005; 31 :247-57.
in non-traumatic infant deaths: does it explain the bleeding 53 Squier W, Salisbury H, Sisodiya S. Stroke in the developing
in 'shaken baby syndrome'? Neuropathol Appl Neurobiol brain and intractable epilepsy: effect of timing on hippocampal
2003; 29:14-22. sclerosis. Dev Med Child Neurol 2003; 45:580-5.
36 Looney CB, Smith JK , Merck LH et a\. Intracranial 54 Arias F, Romero R, Joist H, Kraus FT. Thrombophilia: a
hemorrhage in asymptomatic neonates: prevalence on MR mechanism of disease in women with adverse pregnancy
images and relationship to obstetric and neonatal risk factors. outcome and thrombotic lesions in the placenta. J Maternal
Radiology 2007; 242:535-41. Fetal Med 1998; 7:277-86.
37 Whitby EH, Griffiths PD, Rutter S et a\. Frequency and natural 55 Mercuri E. Timing and aetiology of neonatal cerebral
history of subdural haemorrhages in babies and relation to infarction (Letter). Paediatrics 2000; 106 :615-16.
obstetric factors. Lancet 2004; 363 :846- 51. 56 DeVeber G, Andrew M, Adams C et al. Cerebral sinovenous
38 Tavani F, Zimmerman RA, Clancy RR et al. Incidental thrombosis in children. N Engl J Med 2001; 345:417-23.
intracranial hemorrhage after uncomplicated biI1h: MRI 57 Mercuri E, Cowan F, Gupte G et a\. Prothrombotic disorders
before and after neonatal heart surgety. Neuroradiology 2003; and abnormal neurodevelopmental outcome in infants with
45:253-8. neonatal cerebral infarction. Pediatrics 2001; 107: 1400-4.
39 Volpe JJ. Intracranial haemorrhage. In Neurology of the 58 Marrett S, Lardennois C, Mercier A et a\. Fetal and neonatal
Newborn, 3rd edn. Philadelphia, PA: WE Saunders, 1995, cerebral infarcts. Bioi Neonate 2001; 79:236-40.
pp. 377-8. 59 Evans D, Levene M. MRI Imaging of injury to the immature
40 Tubbs RS, Louis RG , Acakpo-Satchivi L, Salter EG. Anatomy brain. In Squier W (ed.) Acquired Damage to the Developing
of the falcine venous plexus. J Neurosurg 2007; 107:155-7. Brain; Timing and Causation. London: Arnold, 2002.
41 Friede RL. Hemorrhages in asphyxiated premature infants. In 60 Shevell MI, Silver K, O'Gorman AM et a\. Neonatal dural sinus
Friede R (ed.) Developmental Neuropathology. Gottingen: thrombosis. Pediatr Neurol 1989; 5:161-5.
Springer-Verlag, 1989, pp. 44-58. 61 Rivkin MJ, Anderson ML, Kaye EM. Neonatal idiopathic
42 Kibayashi K, Shojo H, Sumida T. Dural hemorrhage of the cerebral venous thrombosis: an unrecognized cause of
tentorium on postmortem cranial computed tomographic transient seizures or lethargy. Ann Neurol 1992; 32:51-6.
scans in children . Forensic Sci lilt 2005; 154:206-9. 62 Fitzgerald KC, Williams LS, Garg BP et a\. Cerebral sinovenous
43 Vinchon M, Noizet 0 , Defoort-Dhellemmes S et al. Infantile thrombosis in the neonate. Arch Neurol 2006; 63 :405-9.
subdural hematomas due to traffic accidents. Pediatr 63 Traill Z, Squier M, Anslow P. Brain imaging in neonatal
Neurosurg 2002; 37:245-53. hypoglycaemia. Arch Dis Child Fetal Neonatal Ed 1998;
44 Steinbok P, Haw CS, Cochrane DD, Kestle JR. Acute subdural 79:F145-7.
hematoma associated with cerebral infarction in the full-term 64 Murakami Y, Yamashita Y, Matsuishi T et at. Cranial MRI of
neonate. Pediatr Neurosurg 1995; 23 :206-15. neurologically impaired children suffering from neonatal
45 Durham SR, Duhaime AC. Basic science; maturation hypoglycaemia. Pediatr Radiol 1999; 29:23-7.
dependent response of the immature brain to experimental 65 Rossiter JP, Anderson LL, Yang F, Cole GM. Caspase-3
subdural hematoma 1. J Neurotrauma 2007; 24:5-14. activation and caspase-like proteolytic activity in human
46 Kemp AM. Investigating subdural haemorrhage in infants. perinatal hypoxic-ischemic brain injury. Acta Neuropathol
Arch Dis Child 2002; 86:98-102. (Berl) 2002; 103:66-73 .
47 Rogers CB, Itabashi HH, Tomiyasu U, Heuser ET. Subdural 66 Lynch N, Roland EH, Poskitt K et a\. Gestational diabetes
neomembranes and sudden infant death syndrome. J Forensic mellitus: risk of newborn brain injury. Dev Med Child Neurol
Sci 1998; 43:375-6. 2007; 49(Supp\. 108):7.
48 Markwalder TM. Chronic subdural hematomas: a review. 67 Brown GK, Squier MV. Neuropathology and pathogenesis
J Neurosurg 1981; 54:637-45. of mitochondrial diseases. J Inherit Metab Dis 1996;
49 Currarino G. Occipital osteodiastasis: presentation of four cases 19:553-72.
and review of the literature. Pediatr Radio12000; 30:823-9 .
50 Volpe JJ. Intracerebellar haemorrhage. In Neurology of the
Newborn, 3rd edn. Philadelphia, PA: WE Saunders, 1995,
pp. 384-9.
I CHAPTER 10 I
FETAL AND PERINATAL DEATH
Jean W Keeling
INTRODUCTION
been translated as 'no risk'. Thus, many consider that there
must be fault attached to any death occurring during or
There are several circumstances in relation to fetal deaths, shortly after labo ur.
still births and deaths in early life when an investigation is
likely to faU within the rem it of a coroner, a procurator fis
calor a medical examiner. These include concealment of DEFINITIONS
pregnancy; unattended delivery, with or without abandon
ment of the newborn infant; and suspected infanticide. The Definitions relating to death in early life differ in different
size of the problem is difficult to assess, although the num countries and in different jurisdictions. This section lists some
ber of infanticides (neonaticides) has fallen in some jurisdic important and relevant definitions and draws attention
tions, as has the number of maternal fatalities attributable to to different usages of the same term. These are related to both
criminal abortion, probably for similar reasons.i Neverthe offences that may have been committed and charges arising
less, abandoned fetuses and infants are still discovered in a therefrom. It is important to be aware of them when
variety of circumstances. Fetal death following maternal comparing statistics or papers from different countries.
injury, sudden fetal death and those fetal deaths in which the • Fetal death is defined as death before 22 completed
attendant midwife or obstetrician is unable to complete a weeks of gestation (World Health Organiza tion [WHO])
still birth certificate also require formal investigation in or before 24 completed weeks of gestation (UK), when
some jurisdictions. 2 the conceptus exhibits no sign of life after complete
A further circumstance in which it has become increas separation from the mother. Fonnal registration of feta l
ingly likely that the pathologist's report will acq uire death is not required, although many maternity units
medicolegal significance is following intrapartum still bilth issue a certificate of fetal death, both as an
or early neonatal death (perinatal deaths) after hospital acknowledgement of existence4 and to facilitate
delivery. The perinatal mortality rate has fallen to lo w funeral arra ngeme nts in accordance with parental
levels in many countries, particularly Western Europe, wishes. Concealment of the death of a fetus is not
North America, and Australia and New Zealand; it was less an offence (in England).
than 8 in 1000 total births in the UK in 2002 with an intra • Still birth is death after 22 completed weeks of
partum death rate of 0.62/1000 total births 3 This low risk of gestation (WHO) or after 24 completed weeks of
labour-related mortality has, in the public's perception, gestation (UK) of a conceptus who exhibits no signs of
---------------~
....
-
The law I 181
life after complete separation from the mother. A Kellett 5 and Mason. 6 They review the effects of sequential
certificate of still birth must be delivered to the changes in legislation in this area in terms of the definition
appropriate registration authority. It can be completed of particular offences and restrictions on medical practice.
by either a midwife or a medical practitioner who was They set out differences in legislation in Scotland com
present at the birth, or who examined the body after pared with England and Wales. The law will be considered
delivery, and is able to certify that the child was not only briefly here.
born alive. If there is any doubt, the procurator fiscal,
coroner or medical examiner should be informed.
• Neonatal death is death following live birth, i.e. The Offence of Procurement of Abortion
exhibition of signs of life after complete separation
from the mother, irrespective of the length of gestation, This is defined in the Offences Against the Person Act 1861,
in the first 28 days (UK) or 30 days (USA) of life. Both Sections 58 and 59. This Act is still extant. The Abortion
birth and death certificates are required. Act 1967 serves mainly to define the term 'unlawfully' in
• Signs of life, from a medical standpoint, are heartbeat, the 1861 Act.
spontaneous respiration and voluntary movement.
Only one of these is necessary to fulfil the definition
of live birth. Child Destruction
• Complete separation from the mother is defined as
complete delivery of all parts of the baby (fetus) from Child destruction is an offence under the Infant Life
the body of the mother. Severance of the umbilical (Preservation) Act J 929, except when performed in pur
cord or delivery of the placenta has no part in suance of 'preserving the mother's life'. It applies to the
definitions relating to completeness of separation from killing of a viable fetus both before and during labour, thus
the mother or to live or still birth. The concept of what offering protection to doctors faced with performing cran
constitutes a separate existence is both important and iotomy during labour for cephalo-pelvic disproportion.
difficult; cardiac function and the ability to breathe This procedure is rarely performed now in developed coun
have been variously considered the gold standard. 5 tries because of the infrequency of contracted pelvises and
However, it is not only the initiation of respiration, but the availability of prenatal diagnosis and Caesarean sec
the ability to sustain respiration, which is essential for tion. The Infant Life (Preservation) Act of ]929 does not
survival,6 that matters. apply in Scotland.
• Abortion is the termination of pregnancy before the
legal limit of viability. Unqualified, the term
encompasses natural pregnancy loss as well as Concealment of Birth
del iberate termination of pregnancy.
• Induced abortion is the deliberate termination of a Concealment of birth, originally a criminal offence, is leg
pregnancy before the legal limit of viability whether by islated in the Offences Against the Person Act 1861, Section
medical or surgical means. 60. It encompasses both live and still births. It is not neces
• Child destruction is the destruction of the life of a child sary to show how death occurred, or even to produce the
who is capable of being born alive, i.e. after the legal infant's body, provided that the mother can be shown to
limit of viability. have delivered a baby in the recent past. It thus dispenses
• Infanticide (England and Wales) occurs when a mother with the thorny problem of demonstration of a separate
(acting alone) causes the of death of a child who is less existence and is a less serious charge than infanticide.
than 12 completed months of age by a wilful act or In Scotland, the Concealment of Birth (Scotland) Act
omission. 1809 additionally requires that the mother conceals both
• Infanticide (USA and elsewhere) is used more generally pregnancy and delivery, and that the child was viable. As
to mean causing the death of a child in the first week in England and Wales, it is not necessary to produce a body
of life. for successful prosecution.
• Neonaticide has no legal definition but is used
variously to mean the unnatural death of a liveborn
child within 24 hours of birth 7 or within the neonatal Infanticide
period.
This offence was introduced in England and Wales by the
Infanticide Act 1922 as an alternative to a charge of mur
THE LAW der or manslaughter. It related solely to death of an infant
caused by the mother acting alone, because of disturbance
The law, in relation to the unnatural death of the fetus and of her mind following birth. It applied originally only to
newly born infant, is considered in detail by Bowen,8 the death of a newly born child. This was amended by the
182 I Fetal and perinatal death
BACKGROUND INFORMATION
CONCEALED PREGNANCY
(a)
WAS THE BABY BORN ALIVE? findings alone, but there are some findings that can answer
this question absolutely. When there is doubt, this must
t may not be possible to distinguish with absolute be clearly expressed and the likelihood of still birth
<:enainty between live birth and still birth from pathological promoted.
184 I Fetal and perinatal death
Figure 10.4 The earliest sign of maceration, skin slippage, over Figure 10.6 Dark-red discolouration of the fetal insertion of the
an extremity. umbilical cord is a sign of maceration which takes about six hours
to develop.
Changes in the skin due to maceration preclude live birth. Deep-red discolouration of the umbilical cord at its fetal
Sequential changes owing to maceration are set out in Table insertion is an indication that fetal death has taken place at
8.5, p. 159. Cutaneous maceration is unlikely to be present least six hours before delivery (Fig. 10.6). Loss of Wharton's
until about six hours after death, when focal loss of superfi jelly from this part of the cord, often accompanied by
cial epidermis is seen over extremities and bony prominences marked twisting of the cord, takes longer to develop and is
(Figs 10.4 and 10.5). This loss of epidermis (skin slippage) also a marker of still birth. Desiccation or separation of
becomes more widespread with the increasing length of time the umbilical cord stump indicates live birth with survival
that the dead fetus spends in utero. Maceration of face, back for a few days. The cord stump progressively desiccates
or trunk develops in about 12 hours. When more than one (Fig. 10.7) over 3 or 4 days and then separates from the
Was the baby born alive? I 185
umbilicus. A vital reaction is apparent w ithin the examp le tissues, cotton w ool or fabric, is occasionally
cord stump w ithin 2- 3 hours of birth on microscopic fo und in the mouth or pha ry nx in deliberate a ttempted a ir
exam ination; this indicates live birth with, at least, brief ways obstruction. Abso rbent material is so metimes found
survival. in or over the mouth, presumably to take up unwanted
fluid secretions, when other findings suggest still birth. The
presence of frothy fluid in the upp er a irways, which has
STOMACH been fo und to withstand freezing of the infant's body, 28 is
a lso an indica tion of li ve birth.
Mil k curd (or other foreign ma terial) in the sto mac h indi
cates live birth with survival. Care must be taken no t to
confuse mucoid secretions, which can be turb id in the LUNG
matu re fetus, particularly in the presence of ascending
Gross Appearance
infection w hen the fet us has swallowed infected liquor
am ni i, w ith milk, which is usually flo ccu lent. Smears of When the lun g has been fully expanded by establishment of
gastri c co ntents stained w ith haematoxylin an d eosin, respiration in a mature baby, it w ill be pink and crepita nt.
Gram 's method for bacteria l organisms and oil red a for fat It is virtually impossible to expel a ll of the air by external
will often clarify any uncertainty. pressure. The lungs are unexpanded in still birth. They are
Rad iographic examination may demonst rate gas in th e smaller, darker in colour and occupy the posterior, paraver
stomach and intestines in liveborn infants. This, however, tebral parts of the thoracic cavity. Knight 29 found the naked
is not abso lute proof of live birth as contamination with eye appearance and feel of lu ngs a better dis criminant
gas- form in g organisms by swallowing infected liquor between live a nd still birth than the lung flotati on test.
am nii ca n result in a gast ric gas bubble. Petech ial hae morrhages are often visible throu gh the
Meconium may be present in the stomach when there pleural surface, but they may be confined to the pleural
has been acute hypo xic stress to the fetus either just befo re fissures. They are considered a sign of acu te hypoxia .
or during labour. Its presence or absence does not help to Petechial h aemo rrhages are p art icularly numerous when
distinguish between live birth an d still birth. death is du e to placental abruption.3o They lose definitio n
and then fade afte r a fe w days, both in the fetus reta ined
in utero and in the delivered baby.
STOOLS
Before birth, the large intestine contai ns meconium, a mix Flotation Test
ture of amniotic squames, intestinal secretions, desqu a
The use of the property of lu ngs to float in water (or
ma ted intestinal epithelium and bile acids. 24 Many babies
buffered formalin) as a determinant of live birth is fra ught
defecate in the first few minutes after birth ,25 although
w ith diffi cul ty. 29 It is un w ise to rely on it as the only deter
many mature ba bies who are subjected to hy pox ic stress
minant of live birth even when so me or any of the pub
before or during labour w ill often defecate before birth.
lished modifications, which all egedly impro ve reliability,
Meconium is dark green and sticky, but sterile. Over the first
are introdu ced. It may be falsely positive because of putre
few days after birth, a mixture of meconium and normal
faction, even to a minor degree. When putrefactio n has
stool is passed. Breast-fed babies have very soft, acidic,
occurred, gas may be in terstitial or both in terst itial and
mustard-yellow stool; bottle fed babies have a firmer, paler
within terminal ai r spaces (see Fig. IO .Sc and d). The value
stool, w hich is less acid ic. 25
of the flotatio n test is negated by mouth-to-mouth or other
Colonization of the intestine wit h bacterial organisms
p ositive-press ure ventilation J
takes place rapid ly. There is geographic variation . More
The flotation test may be falsely nega tive in the pres
than one-hal f of vaginally delivered infants in Pakistan
ence of well-developed hyaline membrane disease, a fre
have positive cu ltures for faec al organisms at 1 day of
quent find in g in the livebo rn preterm baby as well as
age,26 but the proportio n in Swedish infan ts was found to
liveb orn babies born at term who have experienced severe
be much lower. Stool bacteria l flora was well estab lished at
hypoxic stress, or follo w ing ventilation with 100 per cent
4 days of age in babies born in the UKY The stool of milk
oxygen. Despite these dra w backs, Moar 31 eva lu ated lung
fed newborns wi ll cont a in abundant fat globules; bacterial
flotation along with liver flotation, lun g compression, the
organisms will be seen diffusely through the stool sample.
gross appea rance of the lungs and histol ogica l examina
There are di fferences in the bacterial flora of breast-fed,
tion. He fo und it of valu e in carefully controll ed co nditions
compared with form ula-fed, bab ies. 26.27
but stressed the need for corroborative tests.
Plugging of the upper airways by thick meconium pre Both the presence of large amou nts of squamous debris and
cludes estab lishment of respirati on. Foreign material, for mu cus fro m the amnioti c fluid and the presence of meconium
186 I Fetal and perinatal death
.':....
Figure 10.8 Patterns of lung distension. (a) Antepartum still birth at term: there is partial expansion of terminal air spaces up to the
pleural surface as a result of hypoxia-induced inspiratory efforts. There is meconium aspiration. (b) Delivery after motor vehicle collision
(MVC) at 34 weeks' gestation, survival for 13 hours. There is overdistension of term inal bronchioles and alveolar ducts with collapse of
intervening lung parenchyma; these changes are seen in early respiratory distress syndrome. (c) Irregu lar, focally marked, distension of air
spaces and interstit ial tissue as a result of putrefaction, concealed delivery. (d) Putrefaction has resulted in overdi stension of terminal air
spaces and accumulation of gas in pleural connective tissue.
plugging of large intrapulmonalY airways and groups of ter hyaline membranes are classically associated with the
minal air spaces indicate severe hypoxic stress before birth. immature lungs of a preterm infant, hyaline membranes may
The former are readily demonstrated using Alcian be seen in the mature fetus following hypoxic injury to the
blue/phloxine staining, promoted by Attwood 32 to demon type II pneumocytes.
strate amniotic fluid embolism in the maternal pulmonary When respiration has become fully established the lung
vasculature, or an immunohistological marker against high will be uniformly expanded (Fig. 10.S). This can be demon
molecular-weight cytokeratin, such as CK1. 33 strated by histological examination. The terminal air spaces
When the fetus has made strong and repeated inspiratory will not be completely expanded after death, but some
efforts before birth, partial expansion of terminal air spaces expansion should be present right out to the pleura. Pul
is apparent and the alveolar walls create a saw-tooth pattern monary interstitial emphysema has been found to be a use
(Fig. 1O.8a). The presence of hyaline membranes within the ful additional marker of live birth. 34 Septal distension due to
lung on histological exa mination is a marker of live birth . interstitial air trapping and distension of lymphatics can be
They take around 6 hours to develop and up to 12 hours to distinguished using an endothelial marker. It is also impor
become widespread within the lung. Before that time, a char tant to consider that this might be part of the putrefaction
acteristic pattern of overdistension of terminal bronchioles process and indicate the presence of gas-folming organisms
and alveo lar ducts is usually apparent (Fig. 1O.8b). Although (Fig. 1O.8c and d).
Is there evidence of prolonged or difficult labour? I 187
- -- - .a_ ~_~
Are there any significant injuries? I 189
Asphyxia Trauma
Cutaneous Caput
petechiae succedaneum
~--------------------
Haemorrhagic Subaponeurotic
infarction ---"-.A:---'I------------------ (galeal) haemorrhage
Subarachnoid
haemorrhage
""""""""'c------------- Cepha Ihaem atom a
Intraventricu la r
haemorrhage
rf ----+\-\-\--+-\-\-------- Su bd ura I
haemorrhage
Germinal matrix
(subependymal)
haemorrhage
r
Lep<om",;::: J IIII :::0'"'"' (9;''''1
Bone Periosteum
Figure 10.13 Sites of pericranial and intracranial haemorrhage in the newborn by causation. (From Fetal and Neanatal Pathalagy, 4th
edn, 2007, p. 287, Fig. 13.16, with permission, Springer Science and Business Media.)
Subarachnoid haemorrhage is usually focal but can be American Collaborative Study who had suffered non-cata
confluent, usually over the temporal poles (but see Fig. 8.19, strophic trauma during pregnancy and looked at the out
p. 160). It is usually hypoxic in origin. Focal subarachnoid comes. Traumatic episodes comprising blows to the
haemorrhage is often found at the base of a torn bridging abdomen, falls and minor road traffic accidents complicated
vein when the vessel has retracted beneath the membrane. 210 (5.7 per cent) pregnancies, half occurring in the third
Cerebral haemorrhage is rarely traumatic. It is usually trimester. No increase in pregnancy loss was found when
related to bleeding into an area of prior hypoxic injury. compared with non-traumatized control subjects.
The same is true of intraventricular haemorrhage (NH), Fetal loss is more likely following major maternal trauma
which arises from hypoxic injury to small vessels in the (61 per cent) than following minor trauma (27 per cem)54
periventricular germinal matrix or choroid plexus. and when the maternal injury severity score is high 55 57
Although NH most commonly occurs in the preterm baby and the mother is in shock. When trauma is non
postnatally, there are well-documented examples of NH catastrophic, complications are unusual. Fetal loss occurred
occurring in utero. 51 Therefore, its presence in a perinatal in about nine per cent of cases in one study.58 The reason
death should not be interpreted as a marker of live birth. is most commonly premature labour, but placental abrup
tion sometimes occurs. Some advocate careful observation,
even after minor trauma, for this reason. 59 However, a
FETAL DEATH FOLLOWING MATERNAL INJURY
prospective study of nearly 50000 cases found pregnancy
related complications in only 0.001 per cent.50
Fetal deaths following maternal injury are not common,
although trauma complicates 6-7 per cent of pregnancies. 52
The extent of legally directed investigation of the baby in
BLUNT ABDOMINAL TRAUMA
these circumstances depends on a number of factors: the
causation of the injury, whether the baby was liveborn or The most common cause of trauma to the abdomen of a preg
still born and the jurisdiction under which it falls. The procu nant woman is motor vehicle collision (MYC).51 In one study,
rator fiscal (Scotland) is more likely to instruct an investiga maternal injury following MYC accounted for 80 per cent of
tion into a fetal death or still birth than is the coroner cases of abdominal trauma to pregnant women, of whom 10
(England) if the status of the fetus as a person seems some per cent were pedestrians. 55 Other injuries, falls, assaults and
what ambiguous. Fetal maturity is an important factor and is self-harm, are much less common. One institution found that
most impOliant when prosecution of any sort is being con serious assault accounted for 31.5 per cent of trauma to preg
sidered?8 Studies of the outcome of minor trauma in preg nant women over a 9-year period 52 and for 22 per cent of
nancy are few. Fort and Harlin5J identified women from the injury in another study over a similar period. 52
Fetal death following maternal injury I 191
With expanding vehicle ownership, pregnant women are weeks' gestation 70 but is not usually a cause of fetal loss.
increasingly involved in MVCs whether as a driver, passen Uterine rupture is uncommon/! but when it occurs fetal
ger or pedestrian. One study found the highest incidence of mortality is close to 100 per cent. Amniotic fluid embolism
fetal death following MVC in teenage mothers. 63 Occasion is an occasional compJication n
ally pregnant women are involved in accidents as cyclists. In Other causes of blunt abdominal trauma are falls and
my experience, fal ls from bicycles can provoke premature assault, such as ki cking or heavy blows to the maternal
onset of labour and fetal death as a result of prematurity or abdomen. Whilst the velocity component of trauma
placental abruption. Although accidents can occur at any enco untered in MVC is missing, nevertheless these insults
stage of pregnancy, deaths occurring in the last trimester of can result in pregnancy loss and fetal injury even when the
pregnancy (after 28 weeks' gestation) are the most likely to maternal injury severity score is zero. 62 The in cid ence of
be subjected to medicolegal investigation, although the assault during pregnan cy is difficult to assess and likely to
deaths of less mature fetuses are sometimes investigated. be under-estimated. Systematic questioning of mothers
When assessing the likelihood of feta l death being the result attending antenatal clin ic found that an assault had
of any accident, an acquaintance with the incidence of spon occurred at some time during pregnancy in 14.1 per cent of
taneous pregnancy loss at that particular time of gestation is 384 women. 73 In another study, physical violence occurred
useful. Spontaneous fetal loss is less common as pregnancy in 11.1 per cent of 6143 women in the year prior to deliv
advances. In the absence of ma terna l predisposing factors, ery, with an increase in materna l morbidity including
some of which may be revealed on exa mination of the pla preterm delivery, but no other adverse fetal outcome was
centa, co incidental fetal demise or spontaneous onset of noted l4 A study from Sa udi Arabia reported physical vio
labour is unlikely after 32-34 weeks of gestatio n. lence in pregn ancy in 21 per cent of over 7000 records;75
The prime concerns after materna I involvement in an there was an increased rate of placental abruption and pre
MVC are for maternal and fetal well-being, the former bear mature delivery. Confidential enquiries into maternal
ing critically on the latter. 64 Observation of mother and fetal deaths in the UK found that more than one-third of domes
monitoring are necessary for several hours, even when tic abuse cases start during pregnancy and the risk of fetal
maternal injury is slight. 61 Any maternal injuries present loss or still birth is doubled. 76 The subsequen t triennial
should be carefu lly recorded . A careful examination of the report cites continuing vio lence to pregnant women.
abdomina l wa ll for bruising is essential. When the accident Eleven deaths in that triennium were the result of murder
involves the mother as a pedestrian or cyclist, the full cir by the woman's partner, higher than the number of deaths
cumstances and extent of injury to the mother are recorded due to MVC or amniotic fluid embolism and equal to
and are usefully made available to the pathologist before deaths from sepsis. 77 Fatalities are the tip of the iceberg of
fetal examination commen ces. domestic viol ence in pregnancy. Adverse feta l outcome fol
When the mother was the driver or passenger in a vehi lowing domestic violence, including premature delivery
cle, information about the type of restrain ts and whether and a 3.5-fold in crea se in neonatal death, is described in a
they were used is important. Fetal mortality and morbidity North American study.78 A literature review of 30 informa
has been associated with lap strap use. Early reports of uter tive studies of intimate partner v iol ence 9 found an
ine rupture 65 and fetal inju ry66 have deterred some women increase in fetal and maternal mortality and morbidity,
from using restraints. One study found that fetal mortality with a feta l death rate around 16 in 1000 in affected
increased from 14.4 per cent to 16.7 per cent when this type pregnancies.
of restraint was used, a small and comparatively non Chronic subdural haematomas have been demonstrated
significant increase, but maternal mortality was reduced by in the neonatal period in offspring of women subjected to
half67 Despite more recent advice that the risks to the fetus physical abuse, which included kicking and blows to the
are reduced with appropriate use of three point restraints,68 abdomen dUling pregnancy.80,8! In the Stephens et aisl study,
the message is difficult to get across to some women. 55 cerebral atrophy and haematomata of different ages were
Impact severity is the best prediction of adverse fetal out apparent on imag ing shortly after birth. Repeated assault in
come, but appropriate seatbelt use (three-point restraint) the context of domestic violence increases the risk of peri
improved fetal outcome following impacts of mild or mod partl.lm complications. B2
era te severity.69 The position of the restraining straps across Subdural haematoma have been desClibed in 47 offspring
the woman's body is criti ca l to fetal well-being. Defo rma of immigrant Pacific Islanders in New Zealand. s3 Forty-four
tion of the uterus from forcible contact with the restraint is of these babies were still born; some had intraventricular or
the cause of retropla cental haemorrhage producing a shear intracerebral haemorrhage as well as a subdural haematom a.
strain across the uteroplacenta l interface. There may be Some had sca lp or significant subgalea l haematoma. The
additiona l stresses from tensi le failure during rapid deceler haemon'hages were not accompanied by skull fra ctures or
ation of the uterus following impact. 69 Restraints shou ld be tears of the dural folds. Coagulation studies were normal in
positioned above and below the gravid uterus. 6S all mothers tested; a further eight had normal platelet
Premature onset of labour is a common complication counts. The ca use of the intracranial and pericranial haem
(28 per cent) when trauma occurs between 25 and 37 orrhages in these babies was thought to be particularly
192 I Fetal and perinatal death
forceful abdominal massage, resulting in slow head com Maternal death is infrequent. The gravid uterus provides
pression. It is probably significant that more than half of the protection to other maternal organs. As the uterus increases
injured babies were presenting in the breech position. in size, organs, particularly the in testines, are displaced
upwards and may be spared from direct injury. In the case of
gunshot wounds, the uterus, being of dense texture, absorbs
ELECTRICAL INJURY TO THE PREGNANT WOMAN
most of the energy, reducing the risk of damage to other
The literature pertaining to electrical injury, including lig ht maternal abdom in al viscera. loo Intra-uterine fetal death,
ning strikes, is small, with a heavy emphasis on single case premature onset of labour and fetal injury are more likely
reports. The outcome is very variable. Lieberman et al 84 following penetrating abdominal trauma than is the case
report six cases of electrical injury related to domestic acci following blunt trauma. S5 Fetal injury is seen in 60-90 per
dents occurring between 20 weeks' gestation and term. In cent of cases, about half being serious, including penetrating
two cases in which there was an immediate reduction in fetal wounds of thoracic and abdominal viscera. Injury to the
movements, fetal death occurred within one week of the placenta and umbilical cord is also described. IOI Pelinatal
accident. A furth er fetus died 12 weeks after an insult at 2 1 mortality of 47-71 per cent is likely.loo Among Awwad et
weeks' gestation and was growth restlicted at birth. A causal al's96 cases, half of the fetuses died because of maternal
relationship here is tenuous. The other three were born at full hypotension or direct fetal injury. Sel f-directed penetrating
term and survived. The same groups report abortion after trauma has, bizarrely, been used in attempts to terminate
an electrical accident. 8s Jaffe et al 86 report fetal death at pregnancy. Both self-inflicted gunshot wounding 102 and
14 weeks occurring within hours of a domestic mishap. vo luntary repeated stabbi ng l03 are recorded .
Reports of fetal death from the late nineteenth and early
twentieth century (cited by Mehl 87 ) describe immediate
cessation of fetal movements following the incident. Fetal Injuries and Cause of Fetal Death
Interval deaths and chronic problems such as oJigo
hydramnios 84 are more difficult to evalu ate from the point The usual cause of fetal death followin g blunt trauma to the
of view of causation . The frequency of pregnancy loss from maternal abdomen is retroplacental haemorrhage. It usually
other (natural) causes at that particular stage of pregnancy occurs within hours of the injU1Y and can be predicted by
must be taken into account. Yoong88 describes massive increased uteline contractions. 61 Delayed (several days) placen
placental abruption and fetal death 24 hours after electri tal abruption has been recorded. Fetomatemal haemonhage
cal injury. is more common in pregnancies complicated by maternal
Fetal mortality following maternal electrical injury, trauma. 104 It is more likely when the placenta is anterior61 and
when based on accumulated reported cases, is high. can result in fetal dysrhythmias, anaemia and fetal death from
Fatovich 89 found a fetal mortality of 73 per cent among 15 exsanguination. Placental abruption follo wi ng MVC is signifi
reported cases. However, reporting bias is likely w hen the cantly associated with higher maternal injulY severity score
subjects are single case reports. The only prospective study a nd higher vehicle speed. 105 Still birth occurred in more than
of pregnancy outcome following maternal electrical injury 50 per cent of cases in which abruption followed the collision,
followed 31 women and fo und no adverse outcome related with no fetal deaths in the absence of abruption. Rogers et al 106
to the insult. 9o Fetal outcome following matern al lightning also fo und an association between placental abruption and
strike is variable. Of the 12 reported cases 91 - 95 half of the high maternal severity score. They found that fetal cardio
fetuses died. Most mothers sustained burns. Infant survival graphic monitoring could reduce fetal mortality.
bore no relationship to maternal loss of consciousness or Any injulY resulting in maternal hypotension puts the
being thrown to the ground . Loss of fetal movements was fetus at risk. If prolonged or severe, it will result in fetal
observed soon after the incident when the fetus died. There death from hypoxic/hypotensive injury.107 These changes
is a Single case report of taser injury in custody 87 resulting are likely to be seen in the brain and kidney and occasion
in miscarriage 7 days after the incident. There were other ally the heart, liver and intestines. Pathol ogical findings
inauspicious features in that case. are of extensive haemorrhagic infarction. 108 When the
effect on the maternal circulation is less severe, the fetus
may survive, but wit h brain injury. Rapid resuscitation of
the mother gives the fetus the best chance of survival. 59
PENETRATING TRAUMA TO THE MATERNAL ABDOMEN
Blunt injury to the maternal abdomen can result in
The frequency of penetrating trauma to the gravid uterus has direct injury to the fetus, altho ugh it is not common. 58 This
wide geograp hic variation. It is, not surprising ly, enco un is most likely to occur when maternal injury occurs late in
tered more frequently in war zones 96 and those countries pregnancy. The fetal head is afforded protection by the
where gun ownership is common. 97 Fetal well-being is com maternal pelvis but is vulnerable when the pelvis is frac
promised by penetrating injUly to the uterus, but not all tured. Sometimes the fetal skull is fractured when maternal
cases are fatal, even when there is some bleeding into the injury is trivial. Skull bones are the most common sites of
amnio tic cavity or minor fetal injury is sustained. 98 ,99 fractures susta ined by the fetus (Fig. 10.12, p. 189) and may
Can I give a cause of death? I 193
be multiple. 109 Not all are fatal. 110 Skull fractures may be Malformations which should be carefully so ug ht when
complicated by intracranial haemorrhage, perhaps the resuscitation is unsuccessful include laryngeal atresia
more significant injury. 54 Intracranial haemorrhage and (Fig. 10. 14) or severe laryngea l stenosis, laryngeal cleft
tentorial tears are also described. 66 (Fig. 10.15), diaphra gmatic hernia and rena l age nes is.
Subdural and pericranial haemorrhage in t he absence of Cardiac disorders do not usually present as fai lure to
skull fracture and tears of dural folds can cause death within respo nd to resuscitation. Those cardiac anomalies responsi
hours, although some fetuses apparent ly survive for several ble for peJinatal death are usually ductus-d ependent lesions.
days.B3 Not all antepartum subdural haematomas are fatal. Death usually occurs after sudden deterioratio n at 1 or 2
Some babies are liveborn with significant morbidity owing days of age. Both the cardiac anatomical anoma lies and
to cerebral atrophy, hydrocephaly or porencephalyBI .83. 111 some degree of cl osure of the ductus arteriosus w ill be
Whilst the vast majority of in utero subd ural haemorrhage readily apparent on exami natio n of th e thorax.
can be related to tra umatic insult, there are occasional Causes of su dd en de ath in the neon ate are considered
reports of this type of haemorrhage being found in situations furt her in Chapters l1 (p. 209) and 12 (pp. 248-2 50).
of reduced blood coagulability. Bilateral fetal subdural Placental findin gs asso ciated with sudden death are vela
haemorrhage in late pregnancy has been described as a com mentous cord insertion with tearing of major vessels, tight
plication of maternal warfarin therapy, 11 2 although no infor cord knots, meconium staining of the fetal surface
mation about precise timing or repetition of haemonha ge (Fig. 10.16) and retroplacental haemorrhage (Fig. 10.17).
was ava il a ble. It has been described occasionally as a feta l
complication of inherited clotting disord ers, such as factor X CAN I GIVE A CAUSE OF DEATH?
deficiency.113 Usua lly, however, fatal thrombocytopenia and
clotting factor deficiencies give rise to intracranial haemor
Whilst the answer to t his question before post- mortem
rhage in other sites, predominantly intracerebral and in tra
should, in many cases of antepartum still birth, be 'no', few
ventricular. 114 See Chapter 4 (pp. 94- 97).
of these deaths are, in fact, reported to a coroner or procu
Cerv ical injury is rare, bu t lower cervical injury (C5) is
rator fiscal - the ass umption being that examination of the
describ ed follo wi ng ]'vTVC when the mother was using a
placenta, even if fetal exami nation is denied, will provid e
seatbelt. 115 Some authors suggest that protection from
the answer, or at least substantial clues, to t he underl y in g
direct feta l injury in MVC is afforded by an anterior pla
problems.
centa, but the placenta itself is vulnerable, resulting in
retropl acental haematoma. The posterior situation of the
placen ta was thought to be significant in one infant who
sustained lace ration of the liver, ha emoperiton eum and
subdural and subp leural haemo rrhage. 116 Splenic injury
with haemoperitoneum is also described.54.117 Fetal chest
injuries with h ae mothora x and pulmonary contusion can
follo w MCv. 11 8 Scalp bruising was presen t in one infant
foll owing MVC, although the cause of death was rupture of
the placen tao U9
death. They should also selve as a reminder for detailed 16 Funaya ma M, Sagisaka K. Consecutive infanticid es in Japan.
neuropathological examination, which is only possible after Am J Forensic iVIed Pathol 1988 ; 9:9-11.
17 Yamauchi M. Usami S, Ikeda R et a l. Medico-legal stud ies on
fixation of the brain, even in mature babies.
infanticid e: statistics and a case of repeated neon at icide.
It is current practice for obstetricians or neonatologists Forensic Sci Inter 2000; 11 3 :205-8.
to give a copy of the post-mortem report to bereaved parents. 18 Shiono H, Maya A. Tabata N et al. Medicolegal aspects of
For this reason, it is our departmental practice not to issue to infanticid e in Hokkaido District, Japan. Am J Forensic Med
clinicians the clinical summary made by the pathologist for Pathol 1986; 7: 104-6.
Perinatal Patho logy, 2nd edn. Malden Ma: Blackwell Science, 62 Poole GV, Martin IN Jr, Perry KG Jr, Griswold JA, Lambert CJ,
1998, pp. 8-40. Rhodes RS (1996) Trauma in pregnancy: the role of
39 Keeling JW. The perinatal necropsy. In Keeling JW, Khong TY interpersonal violence. Am] Obstet Gynecol 174:1873-7.
(eds) Fetal and Neonaral Pathology, 4th edn. London: 63 Weiss HB. Songer TJ, Fabio A. Fetal deaths related to
Springer, 2007, pp. 20-53. maternal injury. ]ANlA 2001: 286: 1863-8.
40 Huxley AK, Kosa F. Estimation of lunar age from diaphyseal 64 Sugrue M, Kolkman KA. Trauma during pregnancy. Aust ]
length from the fetal materials in the fetal collections at Rural Health 1999; 7:82-4.
Albert Szent-Gyorgyi Medical Uni ve rsity, Hun gary (abst). 65 Rubovits FE. Traumatic rupture of the pregnam merus from
Proc Am Acad Forensic Sci 1999 ; 5:222-3. 'sea t belt' injury. Am ] Obstet Gynecol 1964; 90:828-9.
41 Dorovini-Zis K, Dolman CL. Gestational developm ent of brain. 66 Raney EH. Fetal death secondary to nonpenetrating trauma to
Arch Pathol Lab Med 1977; 101:192-5. the gravid uterus. Am] Obstet GynecolJ970; 106:313-14.
42 Feess-Higgins A, Lan'oche J-C. Del!eiopmel1t oj rhe Humlln 67 Crosby WM, Costiloe JP. Safety of lap-belt restraint for pregnant
Fetal Brain: An Anatomical Atlas. Paris: INSERM CNRS, victims of automobile collisions. N Engl ] Mer! 1971; 284:632-6.
Masson, 198 7. 68 Pearce M. Seatbelts in pregnancy: above and below the bump,
43 Foote GA, Wilson AJ, Stewart JH. Perinatal post-mortem not over it. BM] 1992; 304:586-7.
radiography - experience with 2500 cases. Br] Radiol 69 Pearlman MD, Klinich KD, Schneider LW et al. A
1978; 51 :351-6. comprehensive program to improve safety for pregnant
44 Russ ell JGB. Radiological assessment of age, retardation and women and fetuses in motor vehicle crashes: a preliminary
death. In Barson AJ (ed.) Laboratory In/iestigation oj Feral report. Am] Obstet Gynecol2000; 182:1554-64.
Disease. Bristol: Wright, 1981, pp. 3-16. 70 Williams JK, McClain L, Rosemurgy AS, Co lorado NM.
45 Gould SJ. The respiratory system. In Keeling JW, Khong TY Evaluation of blunt abdominal trauma in th e third trimester
(eds) Feral and Neonatal Pathology, 4th edn. London : of pregnancy: m a ternal and feta l considerat ion s. Obstet
Springer, 2007, pp. 531-70. Gyneco/1990; 75:33-7.
46 Baak JPA, Oort J. A Manual oj Morphometry in Diagnosric 71 Matthews CD. Incorrectly used seatbelt associa ted w ith uterine
Pathology. Berlin: Springer-Verlag; Heidelberg, 1983, rupture follo wi ng ve hicular co lli s ion. Am ] Obstet Gynecol
pp. 131-6. 1975 ; 121:1115-\6.
47 Hilder L, Costeloe K, Thilaganathan B. Prolonged pregnancy: 72 Jud ich A, Kuriansky J, Engel.berg J et aJ. Amniotic fluid
evaluating gestation -speci fi c risks of fetal and infant embolism following blunt abdominal trauma in pregnancy.
mortality. Br] Obstet GYl1aeco/1998; 105:169-73. Injury 199 8; 29:475-7.
48 Keeling JW. Intrapartum problems. In Keeling JW, Khong TY 73 Covington DL, Diehl SJ. Wright BD, Piner M. Assessing for
(eds) Fetal and Neonatal Patho logy, 4th edn. London: violence during pregnancy using a systematic approach.
Springer, 2007, pp. 273-96. Matern Child Health] 1997; 1:129-33.
49 Saunders E. Neo nat icid es following 'secret' pregnancies: 74 Cokkinides VI, Coker AL, Sanderson M et al. Physical
seven case reports. Public Health Rep 1989; 104:368-72. violence during pregnancy: maternal comp li cations and birth
50 Mendlowicz MV, Jean-Louis G, Gekker M, Rapaport MH. outcomes. Obstet GYl1eco/ 1999; 93:661-6.
Neonaticide in the city of Rio de Janeiro: forensic and 75 Rachana C. Suralya K, Hisham AS et al. Prevalence and
psycholegal perspective.] Forensic Sci 1999; 44:741-5. comp li cations of phys ica l vio lence during pregnancy. Eur]
51 de Vries LS, Rennie JM. Preterm brain injury. In Rennie JM, Obstrt Gynecol Reprod Bioi 2002; 103:26- 9.
Roberton NRC (eds) Textbook oj Neonatology, 3 rd edn. 76 Depal1ment of Health. Confidential en quiri es of maternal
London: Churchill Li vingsto ne, 1999, pp. 1252-7 \. deaths in the UK. London: Depal1ment of Hea lth , 2001
52 Connolly AM, Katz VL, Bash KL et al. Trauma and pregnancy. (http//www.doh.gov.uk/cmo/mdeaths.htm).
Am] Perillatol1997; 14: 33 1-6. 77 Drjfe J. Why mothers die. ] Roy Coil Physicians Edin 2005;
53 Fort AT, Harlin RS. Pregnancy o utco me after noncatastrophic 35:332 - 6.
maternal trauma during pregnancy. Obstet GYl1ecol 1970; 78 Lipsky S, Holt VL, Easterling TR, Critchlow Cw. Impact of
35:912-15. police-reported intimate partner violence during pregnancy
54 Rothenberger D, Quattlebaum FW, Perry JF et al. Blunt on birth outcome. Obstet Gynrcol2003; 102:557-64.
maternal trauma: a revi ew of 103 cases.] Trauma 1978; 79 Boy A, Salihu HM. Intimate pal1ner violence a nd birth
18:173-8. outcomes: a systematic review. Int ] Ferti! Wom ens Mer!
55 Shah KH, Simons RK, Holbrook T et al. Trauma in 2004; 49: 159-64.
pregnancy: maternal and fetal outcomes.] Trauma 1998; 80 Soka l MM, Katz M, Lell ME, Fox A. Neonatal sLl lvival after
45:83-6. traumatic fetal subdural hematoma.] Reprod Med 1980;
56 Curet MJ, Schermer CR, Demarest GB et al. Predictors of 24:131-3.
outcome in trauma during pregnancy: identification of 81 Stephens RP, Richardson AC, Lew in JS. Bilatera l subdu ra l
patients who can be monitored for less than 6 hours. hematomas in a newborn infant. Pediatrics 1997; 99:619-21.
] Trauma 2000; 49: 18-24. 82 Pak LL, Reece EA, Chan L. Is ad verse preg nancy outcome
57 Theodorou DA, Velmahos GC, Souter I et al. Fetal death after predictable after blum abdominal trauma 7 Am ] Obstet
trauma in pregnancy. Am Surg 2000; 66:809- 12. Gyl/eco/1998; 179:1 140-4.
58 Good w in TM, Breen MT. Pregnancy outcome a nd 83 Becroft DMO, Gunn TR. Prenatal cranial haemorrh ages in 47
fetomaternal haemorrhage after non catastroph ic trauma . Pacific Islander infants: is traditional massage the cause?
Am ] Obster Gynrcol 1990 ; 162 :665-71. NZ iVIed] 1989; 102:207-10.
59 Siebenga J, van del' Schoot JT, Keeman IN. Trauma during 84 Lieberman JR, Mazor M, Molcho J et al. Electri ca l acc iden ts
pregnancy. Ned Tijdschr Geneesk.d 1999; 143:1144-8 . during pregnancy. Obstet Gynecol 1986; 67:861-3.
60 Runnebaum IB, Holcberg G, Katz M. Pregnancy outcome after 85 Ma zor M, Lieberma n JR. Abortion caused by electric curren t.
repeated blunt abdominal trauma. Eur] Obstet GYllecol Arch GYllecol Obstet 1987; 241 :71-2.
Reprod Bioi 1998; 80:85-6. 86 Jaffe R, Fejgin M, Ben Aderet N. Fetal death in early
61 Pearlman MD, Fintinalli JE, Lorenz RP. Blunt trauma during pregnancy due to electric current. Acta Obstet GYll ecol SClmd
pregnancy. N Eng/ ] Med 1990; 323: 1609 - 13. 1986; 65:283.
References I 197
87 Mehl LE. Electrica l inju ry from tasering and miscarriage. 106 Rogers FB, Rozycki GS, Osier TM et al. A multi-in st itutio nal
Acta Obstet Gynecol Swnd 1992; 71 : 118-23. study of facto rs associated with feta l death in injured
88 Yoo ng AF. Electrical shock susta ined in pregna ncy fo llowed pregnant pati ents. Arch Surg 1999; 134:1274-7.
by placenta abruption. Postgrad Med ) 1990; 66:563 - 4. 107 Drost TF, Ros emurgy AS, Sherman HF et al. Major trauma in
89 Fatovich DM. Electric shock in pregnancy. ) Emerg Med pregnant women: materna l/fetal outcome. J Trauma 1990;
1993; 11:175-7. 30:574-8.
90 Einarson A, Bailey B, ln oce ncion G et al. Accid enta l electric 108 Fries MH, Hankin s GOY. Motor vehicle accident associa ted
shock in pregnancy: a prospective cohan study. Am) Obstet with minimal maternal tra uma but subsequent fetal demise.
Gyn ecoi l997; 176:678-81. AIIII Em Med 1989: 18:30 /- 4.
91 Rees WD. Preg nant woman struck by lightning. BM) 1965; 109 Evrard JR, Stmner WQ , Murray FJ. Fetal skull fracture from
1: 103-4. an automobile accident. Am ) Forensic Med Patho l 1989;
92 Chan Y-F, Sivasamboo R. Lightn ing accidents in pregnancy. 10:232-4.
) Obstet Gynaecol Br Commonw 19 72; 79:761 -2. 110 Hartl R, Ko K. In utero skull fra cture: a case report. ) Trauma
93 Guha-Ray DK. Feta l dea th at term due to li ghtnin g. Am) 1996 ; 41 :549- 52.
Obstet Gynecol 1979; 134: 103-5. III GUIlIl TR , Mo ra JD , Becroft DM. Co nge nital hydrocephalus
94 Fl a nnery DB, Wiles H. Follow- up of a survivo r of intra -uterine seco ndary to pre nata l intracrania l hae morrhage. Aust NZ )
light ning exposure. Am) Obstet GYl1eco11982 ; 142:238 - 9. Obstet Gyl/aecol 1988; 28: 197 - 200.
95 Pierce MR , He nde rson RA , Mitchell JM. Cardiopulmonary 11 2 Robinson MJ, Cameron MD, Smith MF, Ayres AB. Fetal
arrest secondary to lightning inj ury in a pregnant woman. subdural haemorrh ages presenting as hydrocephalus. Br lVIed
Ann Emerg Med 1986; 15: 597-9. J 1980; 281:35.
96 Awwad IT, Azar GB , Seoud MA et a1. High-velocity ]] 3 De Sousa CD, Clark T, Bradshaw A. Ante natally diagnosed
penetrating wounds of the gravid uterus: review of 16 yea rs subdural hemorrhage in congen ita l factor- X deficiency. Arch
of civil war. Obstet GYl1eco l 1994; 83 :259-64. Dis Child 1988; 63:1168 - 70.
97 O'Shaughn essy MJ. Conservat ive obstetric management of a 114 Shere r DM, Anyaegbunam A, Onyeije C. Antepartum
gunshot wound to the seco nd-trimester gravid uterus: a case fetal intracranial hemorrhage, predisposing factors and
repon. J Rep rod Med 1997; 42:606-8. prenatal sonography: a review. Am ) Perinatol 1998;
98 Piers on R, Miha lovits H, Thomas L, Beatty R. Penet rating 15:431-41.
abdominal wou nd s in pregnancy. Ann Ell1erg iVIed 1986; 11 5 Weinberg L, Wyatt P, Busuttil A. Tra umatic intrauterine fetal
15: 1232 - 4. sp in al fracture follo wing seatbelt use: a case repon.
99 Grubb OK. Nonsurgica l management of pe ne trating ute rine J Trauma 200 1; 51 : 1195- 16.
trauma in preg nancy: a case report. Am ) Obstet Gyn ecol 1/ 6 Fakhoury GW, Gibson JRM. Seatbelt ha za rd s in pregnancy: a
199 2; 166 :583-4. case report. Br J Obs tet Gynaeco/ 1986; 93:395-6.
100 Franger AL, Buchsbaum HJ, Peaceman AM. Abdomina l 11 7 Siddall-Allum IN , Hughes JH, Kal er S, Reginald pw. Splenic
gunshot wo und s in pregnancy. Am ) Obstet Gynecoi 1989; rupture ill utero following a ro ad traffic acc ident. A case
160: 1124-8. report. Br ) Obstet Gynaecol 199 1; 98:3 18-1 9.
101 Lavin JP, Polsky SS. Abdom in al tra um a during pregnancy. 11 8 Litmanovitz I, Dolfin T, Arnon S et al. Feta l intrathoracic
Ciin Perinotol1983; 10: 42 3- 38 . injuri es following mild maternal motor vehicle acc id ent.
102 Buchsbaum HJ, Staples PJ. Self-infli cted gunshot wound to J Perill at ivIed 2000; 28: 158 - 60.
the pregnancy uteru s: report of two cases. Obstet Gyneco/ 119 Griffiths M, Hillman G, Usherwood MM. Seat belt injUly
1985; 65:32S. in pregna ncy resulting in fetal death. A need for
103 Sakala EP, Kort DD. Management of stab wounds to the ed ucation? Case reports. Br J Obstet Gynaeco/ 199 1;
preg nant uterus: a case report and a review of the literature. 98:320- 4.
Obstet Gynecol Sum 1988; 43:319-24. 120 Yates PO. Birth tra uma to the verte bra l arte ri es. Arch Dis
104 Rose PG , Strohn PL, Zus pan FP. Feto maternal haemorrhage Child 1959; 34 :43 6-41.
follo wing trauma. Am) Obsrer Gynecol 1985; 153:844- 7. 121 Ito Y, Ts ud a R, Kimura H. Diagnostic va lue of the
105 Reis PM, Sander CM, Pearlman MD. Ab rup tio place ntae after placenta in med ico-legal practice. Forensic Sci lilt 1989;
auto accidents. A case-control stu dy. ) Reprod Med 2000; 40:79 - 84.
45:6-10.
I CHAPTER 11 I
SUDDEN UNEXPECTED DEATH IN
INFANCY: SUDDEN INFANT DEATH
SYNDROME OR SOMETHING ELSE?
Jean W Keeling
INTRODUCTION
In the UK, SIDS became an option as a registrable cause
of death in 1971, and the number of registrations increased
Sudden, unexpected deaths are a large and important over the following decade as the option was taken up by
group of deaths in the post-neonatal infant age group more pathologists. The International Classification of Dis
(1 month to 1 year). These deaths require a consistent high ease separated 'SIDS' (789.0) from 'Sudden Death - cause
standard of investigation, which goes well beyond unknown' (795.0) in 1979; the latter term was included for
anatomic dissection and microscopic examination of tissue the first time in the previous revision in 1968.
samples from a few major organs. Among the deaths pre Whilst the term 'cot death' has been used since the early
senting to the pathologist as sudden and unexpected will 1950s and was refined in 1965 to encompass only unex
be those that are fully explained by a recognized disease plained deaths, I it was not until the mid-1970s that the
process (see Chapter 12), some that are non-natural, general acceptance of Beckwith's definition 'The sudden
some explained by minor or long-standing disease and death of any infant or young child, which is unexpected by
others that occur in the absence of any pathological history and in which a thorough post-mortem examination
abnormality. Cases in the last three groups are often fails to demonstrate an adequate cause of death',2 enabled
problematic and require the fullest, multidisciplinary meaningful comparisons between different studies. Clearly,
investigation. there are ambiguities that beset this definition: How sud
The term 'sudden infant death syndrome' (SIDS) was den? Unexpected by whom? Which historical events
devised to be used in death certificates in cases of (sudden) exclude deaths from the SIDS definition? What constitutes
infant death in which no disease was identified at post a 'thorough' necropsy examination? What type or extent of
mortem examination and death was deemed natural. It was pathological abnormality comprises an adequate cause of
convenient for pathologists - no need to ascribe the death to death?
tenuous 'respiratory tract infection' or 'gastroenteritis' - and Despite these reservations, this definition remains the
it avoided the use of terms such as 'unasceI1ained', which focus of our approach to the problem. However, the
necessitated fulwer inquiries and an inquest or, worse, 'aspir importance of the immediate circumstances of death has
ation of vomitus', which not only provoked an inquest become increasingly appreciated. J Berry et al 4 found a
(non-natural death) but implied poor infant care practices by detailed history, particularly of the precise circumstances
parents or carers. of death, to be a most imp0!1ant pal1 of the evaluation of
The definition of SIDS I 199
sudden unexpected death in infancy (SUD!). The formal to the commencement of post-mortem examination,13
consideration of the circumstances of death as part of the although evaluation of the effect of bed sharing poses prob
definition of SIDS has been proposed,5 but not widely lems for individual pathologists. Fleming et al 14 emphasize
implemented. The greatest benefit of the SIDS designation the role of the multidisciplinary team in the overall evalu
has been that it has permitted much better estimates to be ation of any SUDI, using a non-hierarchical grid.
made of the size of the problem than has hitherto been pos Krous et al 15 have proposed a plan of subdivision of
sible. The number of cases has been very much greater than SUDI!SlDS with precise definitions. They acknowledge the
had been supposed initially and constitutes an important problems of evaluation of cases after prolonged resuscita
public health problem. tion or incompleteness of investigations, both of which
There are, however, disadvantages in the use of the term introduce serious problems for reviewers.
SIDS. One has been the development of an assumption that This chapter will consider those questions arising out of
this 'syndrome' has a single aetiology - 'died of SIDS' is an Beckwith's definition of SIDS. It will explore family back
unfortunate phrase seen in many publications. Another has ground and discuss why prior events and some patho-logical
been a tendency among some in child care to stop thinking abnormalities require a different, sometimes more specific,
about the problem because it is an accepted, registrable designation. It will also describe pathological criteria which
designation. The term has been accepted uncritically and in supp0l1 the SIDS categorization and discuss concerning fea
any particular case, until recently,6 few attempts were tures in presentation or history and asses the appropriateness
made to unravel any predisposing factors. A fUl1her prob and significance of pathological findings.
lem is the unwillingness of some pathologists to investigate
cases fully 'because there is nothing to find'! This is a most
unfortunate view as some of the specific causes of sudden THE DEFINITION OF SIDS
death, which may be missed by slipshod investigation, are
relevant to the medical management of other family mem How Sudden?
bers or of siblings yet unborn. Of equal concern is that
some of these poorly investigated deaths may be the result A very tiny minority of SUDI occur whilst the infant is
of homicide. Bajanowski et al 7 found 5 per cent of non being observed. it is, however, an important group, as a
natural deaths among 339 externally normal victims of high proportion will be explained. This important fact must
sudden infant death. Non-natural deaths accounted for 7.9 be conveyed to the pathologist as additional investigations
per cent of all deaths in the Confidential Enquiry into Still are necessary. The three likely groups of underlying path
births and Deaths in Infancy Sudden Unexpected Deaths in ology which must be carefully considered are trauma, acci
Infancy (CESDI SUDI) studies. 6 dental and non-accidental, cardiac pathology and upper
Gilbert-Barness and Barness 8 suggested that the use of respiratory tract obstruction.
SIDS be repJaced by 'sudden infant death' with case-specific A larger, but still relatively small, group of babies are
qualification. Green 9 has argued for the abandonment of the those who collapse or die suddenly during the course of a
terms 'cot death' and 'SIDS', and for wider use of 'not ascer recognized acute illness, but before a specific diagnosis has
tained' on the death certificate. He points out that with the been reached. In this group, an explanation for death is
considerable reduction in infant deaths following the suc likely to be found if an adequate examination, induding
cessful 'Back to Sleep' campaigns, the proportion of non microbiology and investigation for possible genetic meta
natural deaths is likely to be higher among sudden infant bolic disease, is carried out.
deaths at the end of the 1990s than it was 10 years earlier. The majority of infants dying unexpectedly are found
Meadow lO supports this view and considers that unexpected dead in their usual place of sleep when the rest of the
infant deaths require a multidisciplinary investigation of household wakes. The CESDI SUDI studies 6 found that 64
high standard and that the term SlDS should be more dosely per cent of deaths were discovered betvveen 05.30 and
defined or else abandoned. The opposite view is put by 11.30 hours. A further 17 per cent died during a daytime
Limerick II who argues that the number of non-natural sleep. Few deaths in that study occurred between 18.00 and
deaths is small, and that there is a need for a dear category 24.00 hours. Time of death should not be an excuse for dis
of sudden, unexplained natural deaths. Her statement that a missing the death as natural and for incomplete investiga
case review following post-mortem examination would tion of the death. Important, treatable pathology is found
avoid any need for the pathologist to use the term 'cause in some infants in this group, and it is important that both
unascertained' is, in my view, incorrect. the parents and family practitioners are made aware of
SUDI can fall outside the SlDS definition for reasons such findings.
related to the baby's family history, past medical history, A group of deaths that do not fit easily within the defini
circumstances around the time of death, or because of the tion of sudden are those in which the infants, found col
results of post-mortem investigations carried OUt. 12 Some lapsed, are maintained on life support systems for more than
of the inconsistency of terminology can be related to the 12 hours. These should, in my view, be investigated with the
provision of inadequate information to the pathologist prior same vigour. Clinical management can result in findings at
200 I Sudden unexpected death in infancy
necropsy that are difficult to interpret and a full interven Which Historical Events are Significant?
tional history is essential. Discussion with the medical staff
involved may assist evaluation of findings (see Chapter 19). The most significant points to elucidate from the history
Clearly, there is no necessity to repeat those investigations are previous infant deaths (explained or otherwise) in the
done in hospital, but care should be taken that the full range immediate or extended family, previous admission of the
of investigations is completed. infant to hospital and any documentation of prior injury to
the deceased or to siblings. These and other family events
are considered in more detail later in the chapter.
How Old?
National statistics define the infant period as from birth to What is a Thorough Necropsy Examination?
1 year, but deaths in the first week of life are generally birth
related and not sudden. SUDI studies, anxious not to miss The purpose of necropsy examination in SUDI is to distin
cases, may set age limits widely, 1 week to 2 years in a UK guish between natural and non-natural death, to elucidate a
multicentre study.IG However, most SUD! occur within a cause of death if one is present and to further the under
more restricted age range. Many consider 1 month to 1 year standing of sudden death in the infant period. Published
to be appropriate limits. Most deaths occur prior to the guidelines include those of Wigglesworth et al,J9 the Royal
eighth month. Unexplained sudden deaths before 1 month of College of Pathologists,20 ValMs-Dapena et al,21 Krous 22 and
age are unusual. They are discussed fuliher on p. 209. Sud Howatson.23 It is impOltant that SUDI are investigated to a
den neonatal (0-4 weeks of age) deaths accounted for only consistently high standard. This includes the completion of
three per cent of deaths in a Scottish study. 17 investigations even in those cases where major pathological
Unexpected deaths over I year of age are also unusual abnormality is discovered early in the investigation. How
(three per cent). Most sudden deaths among infants over else is the pathologist to understand the significance of his
6 months of age are likely to be explained. Fleming et al 6 findings in other cases? The need for consistent and thor
found the median age at death of SIDS to be 13 weeks within ough investigation of the family background as well as the
an interquartile range of 7 weeks, 6 days, to 21 weeks, post-mortem examination itself were emphasized by an
3 days. Aggregated data from several large studies derived intercollegiate working group24 and are supported by Landi
the 5th and 95th percentiles as 3 weeks and 9 months, et al, 25 who found that more and appropriate investigations
respectively. IS were done in cases certified as SIDS than those considered
explained. Centres unable to offer full radiological and
microbiological support as well as a high technical standard
Unexpected by Whom? for processing histological samples, including neuropatho
logical samples, should not investigate these deaths. Sadler 26
An important part of the definition of SIDS is the lack of emphasizes the importance of early and thorough microbio
antecedent symptoms. This absence of symptoms requires logical investigations. The relative contributions of different
critical evaluation and is a forceful argument for the components of the investigation of SUDr are evaluated by
involvement of a multidisciplinary team in the invest iga Berry et al 4 and Mitchell et alY
tion of SUD!. An experienced paediatrician or health visitor Toxicological investigations have to be justified in
may unearth symptoms whose significance was not appre many jurisdictions because of a low return of positive find
ciated by carers or police. This might be because of lack of ings. Langlois et al 28 detected drugs in 16 per cent of SUD!
experience of the carer or the non-specific or insidious investigations with drug administration an important, but
nature of signs and symptoms. The apparent lack of symp unsuspected, factor in the death in 2.6 per cent. Arnestad
toms may, in a few cases, be explain ed by the velY rapid et al 29 found that toxicological investigations made a
course of the illness, particularly septicaemia or meningitis. positive contribution to the evaluation of four per cent of
The lack of specific signs of illness in the infant may be SUD!.
another contributory factor.
In some cases, the history is deliberately misleading, to
conceal either real or perceived inappropriate treatment of What is an Adequate Cause of Death?
the infant. Even after thorough, sensitive and timely ques
tioning, it is apparent that some infants exhibit no or only The interpretation of the pathological findings in respect of
trivial symptoms. la A careful and thorough history is adequacy of a cause of death is frequently difficult in SUD!.
particularly important to the pathologist to enable him or The problem does exist in relation to deaths at other ages,
her to include additional investigation pertinent to the but because of the prevalence of degenerative diseases and
particular case 4 and in the overall evaluation of the death. the relative infrequency of tmly unexplained sudden death
It is not infrequent for families to give a history of a recent in older children and aduIts 30 it is less discussed. Over
mild upper respiratory tract infection or of immunization. emphasis of minor pathological findings, particularly in
Epidemiology I 201
poorly investigated cases, is a particular problem in this miscarriages and still bilths 41 and shorter interpregnancy
age group.4,]I,]2 The significance, or otherwise, of partiClI intervals than control subjects. 42 Other pregnancy-related
lar pathological findings is discussed later in the chapter. problems such as urinary tract infection and anaemia are
weaker associ-ations. An elevated maternal serum alphafe
toprotein level in the second trimester of pregnancy is
Multidisciplinary Case Review associated with a higher risk of SIOS .43 Maternal recre
ational drug use is also a risk. 6,17 In Bristol, UK, the propor
A case review following a SUDI is the usual practice in many tion of SIOS victims from deprived socioeconomic
areas]] and is rapidly becoming the norm in many countries. backgrounds rose from 47 per cent to 74 per cent over a
Review groups involved the family doctor and health visitor 20-year period. 37
(with midwifery input, if appropriate), a researcher, when Simultaneous SUDI in twins is most unlikely to be
formal parental interviewing is normal practice, a patholo SlOS. It is an unusual event, but likely to have an environ
gist and a paediatrician. Its purpose is to examine both risk mental cause. A detailed death scene investigation is
factors and the actions of parents or health professionals, mandatory. 44 Sequential SUD! in twins is likely to result
which might usefully be modified either in respect of a par fro m infection.
ticular family or in professional practice. SIOS victims have lower Apgar scores at birth than con
Review of the pathology findings encourages consistent trol infants and a re more likely to be admitted to a special
investigation and goes some way towards introducing uni care baby unit. They are also more likely to have dysmor
formity of interpretation of pathological abnormality. It phic features, malformations and positional deformities.
forms the basis of discussion with each family. It is the Low socioeconomic class, lower parental education level
ideal venue to develop appropriate local strategies to sup and poor housing are consistent findings in the families of
port bereaved families. 24 SIDS infants. The increase in unemployment and of cohab
itation makes it difficult to compare CUlTent and earlier stud
ies, but the CESDI SUDr studies clearly demonstrate that,
irrespective of the parameters used - occupation, family
EPIDEMIOLOGY income, receipt of income support or employment status
an excess of SIOS deaths is seen in families with socio
Epidemiology of SUDI and SIDS economic problems. 17
on their part and, perhaps, have averted death. It is also Table 11.1 Factors in the history causing concern in SUDt
important to remember that young infants often suffer a
Many GP/accident and emergency attendances without clear
succession of minor ailments. This background of frequent,
evidence of illness
minor morbidity must be taken into account when conside
Apparent life-threatening events (ALTEsl. especially if more
ring the relevance of reported symptomatology.
than one and starting at less than 4 weeks of age and if
Stanton et al 49 found that snuffles and colds were
unwitnessed or if the same person finds the baby each time
reported frequently by both SUD! and control groups.
Age > 1 year
However, they found that more serious respiratory symp
Epistaxis
toms, such as cough or wheeze, were more frequent in the
Seizures - especially unexplained
SUD! group. Infants who died were also more likely to have
Death occurs during the day when the baby was apparently
exhibited, in the day or so prior to death, unusual drowsi
well earlier
ness, irritability or excessive crying. They were also more
ALTE in siblings
likely to be off feeds, pyrexial or sweaty. Almost 60 per cent
Previous SUDI in sibling
of SUD! victims had symptoms in the 48 hours before death ,
Death on anniversary of previous SUDI
half of which were considered major. More babies who died
Parental involvement with the media after child's death
than control babies had seen a doctor both within 24 hours
of death and earlier in the course of perceived illness. Half of
the infants with system-related symptoms had evidence
of significant terminal disease at necropsy whilst only one
quarter of those with non-specific symptoms had signifi Adverse Historical Factors (Table 11.1)
cant pathology. 16 In that study, about one-t1Fth of victims of
unexpected deaths (no recognized symptoms) had evidence The occurrence of a previous infant death in the family or
of serious acquired disease at necropsy. in the extended family should alert the pathologist to the
In another study, babies dying unexpectedly in the infant possibility of an inherited disorder. which might not have
period were more likely to have had minor signs of illness in been fully investigated in the earlier death, as well as to the
the previous seven days than control infants and to have possibility of non-natural death 10.53 More than half of
been seen by their general practitioner. 18 This study found Meadow's non-natural deaths were initially certified as
no difference between babies who died and control infants SIDS deaths although the thoroughness of investigation is
with respect to major signs of illness. Stanton and OakJeyS° not set out. A detailed background investigation is required,
found that SUD! victims were more likely to have been and it is helpful to retrieve the necropsy and police reports
admitted to hospital previously and required longer admis on the previous death prior to commencing examination of
sion than control (surviving) babies. Although eight infants the second as well as accessing any clinical notes.
in their study were admitted to hospital because of suspicion A history of unexplained collapse with pallor or cyanosis
of abuse, no babies had unexplained apnoea, but 9 of 71 suf and apnoea is, for me, a cause for concern in SUDI babies.
fered loss of consciousness - repeated in three cases. Concern increases when acute AlTEs have occurred on more
In New Zealand, SIDS babies were found to have more than one occasion 10,54 and when episodes commence after
severe illness resulting in more GP or hospital contact than the neonatal period. 55 A further cause for anxiety is when
control infants.51 The clinical signs that prompted parental investigation of these episodes in a hospital setting fails to
action were not specified. A more recent study47 found that find an underlying problem, particularly if ALTEs cease in
the clinical characteristics of explained SUD! and SIDS vic hospital. 54 .56 It is important that the circumstances under
tims were similar. Apparent life-threatening events which the episodes occurred are fully investigated. Ariagno
(ALTEs), defined as an episode in which the infant became et al 57 found that AlTEs in pre-term infants occurred earlier
apparently lifeless, were reported by parents of both groups and were more likely to be witnessed by professionals (54 per
(9/61 , 14.8 per cent, and 37/317. 11.7 per cent) and were cent) than those in term infants (23 per cent). Similar episodes
more frequent than among control subjects. However. only described in siblings is a further concem. 53- 55 Byard and
two-thirds of these episodes were brought to medical atten Krous 58 list potential natural causes of AlTEs and highlight
tion and less than one-halfwere seen in hospital. Infants in the need to consider imposed suffocation as a basis for such
both groups were more likely to have been unwell than events. In the past, such events have even been called ' near
control infants in the 24 hours before death. Signs seen miss cot deaths', but currently many are thought to be the
more frequently in the SIDS group were drowsiness. result of imposed upper airways obstruction. 10,55.59,60
wheezing and reduced fluid intake in the 24 hours prior to Such a history is most unusual in what, both epidemio
death. Neonatal problems and hospital admissions were logically and pathologically, is a typical SIDS death.
much more common in the SUD! group than among SIDS Having been somewhat sceptical about the frequency of
babies. imposed airways obstruction as the cause of ALTEs, per
In the Netherlands, signs of iJJness were more frequent sonal experience of repeated infant death with or without
in explained SUDI than SIDS or borderline SIDS. 52 AlTEs has changed my view. I wonder whether the history
~ -_.
Sleeping enviro nment I 203
of even a single unexplained ALTE should p reclud e use of Placing the infant in the 'supine' sleep ing position 66 and
the SIDS designation. Unless death was clea rly explained, avoidance of prone and, now, 'side sleeping'65 is advised,
fulfilling a sp ecific diagnosis, then, in these circumsta nces, Avoidance of infa nt restrai nts design ed to maintain the
death sho uld be considered unascertained. initial sleep ing position is current best practice. More
Another unusual symptom in the infant perio d is epi details abo ut safe sleeping and appropriate avoidance of
staxis, defi ned as bl eeding from both nostrils.55 This should co-sleeping have been formulated. 67 Co-sleeping is con
not be confused with blood-streaked mucus from one nos si dered further on p. 204.
tril, usually the resul t of self-inflicted injury. Epistax is is Encouragement of bed sha ring app lies only to non
particularly concerning when accompanied by choking smoking parents who refrain from alcohol and social
and/or tra nsient change in breathing patte rn as it raises the drugs. Excessive tiredness and use of prescribed sedatives
possibili ty of imposed ainvays obstruction. 58 ,61 The pres or tranquillizers a re also contraindications to bed sha ring
ence of blood-tinged mucu s at the nostrils in a su dden as they may result in overlaying of the baby. Sleeping in a
baby death is to be treated suspiciously. cot in the parental bedroom reduces the risk of SIOS. 68
A hi story of seizures is an indication fo r particularly
careful central nervous system (CNS) exa minati on, which
can be done only fo llowing fixation of the brain. Detailed Suboptimal sleeping environments
information about pregnancy and the neonatal peliod is
essentia l when seizures have been reported. Howeve r, the Stanton 69 drew atte ntion to overheating of sleeping infants
possibili ty of a non-natural process should not be forgo tten, as a lisk factor in cot death. He found a number of contribu
Seizures (due to induced cereb ral hypoxemia) are probably tory fact ors likely to promote overheating. These includ ed
the most common presentation of fabri cated illness 62 maternal atti tude, especially duri ng co ld weather, use of
Necropsy examination should include careful h arves ting of daytime, outdoor clothing in bed, swa ddling in blankets,
sampl es for toxi co logical examination ,63 particul a rly urine, use of folded adult bed blankets on cots, use of bonnets or
blood and vitreous fluid (only if the eyes do not need to be hoods and a hot environment, for example central heati ng
examined histo logica lly). on a ll night. In some cases, pyrexial illness in the infant also
Unusual findin gs amo ng parents of babies who had seemed to compound the problem, Excessive bedding was
been smothered were som atizing diso rders, involvement found to be an indep endent risk factor in south-west
with arson or medical litigation and actively seeking media England. 70 Wigfield et al 71 fo und that a low outdoor tempera
publicity in relation to their infant's death.1O ture sti ll prompted mothers to increase infa nts' bedding
when room tempe rature was adequate a decade after Stan
ton's observations were made.
SLEEPING ENVIRONMENT The associatio n of prone sleeping position and cot death
has been reported in the case-control studies of infant deaths
Infan t care practices are culture dependent a nd have since 1970.72 Reports of the association accu mulated with
evolved gra du a ll y over very ma ny years as a result of fam increasing importance of risk from this practice from several
ily experience and loca l contacts. In recent years, opinions centres, as other risk facto rs for SlDS have changed?O,73-76
from many sources have been wid ely dissemin ated; fash Further supp ort for the association has been the marked
ion, fads and consumerism are pressures on you ng, often reduction in SlDS followi ng publicity campaigns to increase
inexpe rienced, mothers who compete with advice from professional and public awa reness. Mitchell et al 77 consider
health professionals. that the practice of 'Back to Sleep' accou nts for about 50 per
cent of the fa ll in SlDS deaths in New Zealand ,
Interactions between sleep ing position and bedding have
Good practice been explored. Pron e sleeping and high tog values of bed
coverings were independent risk facto rs in a UK Study.70 The
The reco mmend ed sleeping environment for young infants is use of quilts was a risk factor for SlDS among supine or side
a cot with firm mattress, with no lise of soft pillows or sleepers in Tasmania. 78 An additional adve rse factor for
bumpers, Whilst a firm, permeable pillow may have had merit prone-sleeping infants may be the adoption of face straight
when used wi th an impermea ble mattress cover when many down or face near-straight down positions. These positions
infants were put down prone,64 it is less useful when most have been shown to introduce an element of airways
babies sleep supine. Preferred cot bedding is a sheet wit h obstruction that makes death more likely in some infants,79
lightweight blan ket(s), which can be tucked in at the sides, Further examinati on of the infants' sleep ing environ
rather than duvets or sleeping pods. It is recommen ded that men t demon stra tes that an increased risk is related to side
the enviro nment be neither too hot or cold (at about 16-18°C) sleepin g pOSition and to head covering as an impediment
and well ven tilated. Pl acing a sm all infant at the end of t he to maintaining thermoneutrality.76,8o Living in a household
cot (foot to foot) rather than in the middle, to avoid burrow where carers smoke is a risk factor for SIOS. This risk is
ing under covers and overheating is current advice.65 related to the number of cigarettes smoked in the local
204 I Sudden unexpected death in infancy
environment. 35 ,75,76 Sleeping in a room separate from par inappropriate for young infants, who were able to push their
ents was a risk factor for SIDS in a New Zealand study. heads between cot slats or rails. Smialek et al 89 and Gilbert
Bamess et aJ83 have described similar cases and a correspond
ing problem when babies sleep on adult beds. Deaths in
Inappropriate sleep surfaces for infants similar circumstances were still occurring 12 years later
because families still used old or broken cotS.93
The use of unstable sleep surfaces is unsuitable for small The use of bedside cots, with bars along only one long
infants. The danger of placing infants prone on surfaces such side of the cot, introduces similar potential for the infant to
as adult-size water beds was flagged up by Ramanathan become wedged between the cot and the parental bed. 94
et al. 81 They reported that 6 out of 34 SIDS cases investigated Bed rails, designed to stop young children from falling out
over a short period had been found face down on water beds. of toddler or adult beds, pose a similar hazard for infants
Bass 82 reported 13 infant deaths on water beds and drew less than 1 year of age. 95 There is a danger of strangulation
attention to the water bed design (free flow, both original and among older infants who can pull themselves up using cot
modified designs). Eleven of these cases had been originaJly sides when loose clothing or dummy cords catch on cot
certified as SIDS. Deep-pile sheepskin rugs may present a p rotru bera nces, 83,88,89,96
similar, unacceptable hazard to young infants.83 In these situ Byard et al 90 draw attention to the risk of asphyxial
ations, the infant is not able to exercise the head movement death when infants are left sleeping in car seats, baby
possible when prone on a firm surface. bouncer or pushchairs (stroller-prams)97 with ill-fitting
The danger to infants of the use of polystyrene-filled restraints, resulting in suspension of the infant. There is a
bean-bags as a resting or sleeping place is clearly danger of asphyxia when impermeable mattress coverings
recorded. 84 ,85 At least 35 deaths have taken place on these in cots are loose rather than being closely fitted. This
cushions, the majority being initially registered as SIDS. applies even when babies are very young and consequently
Babies were predominantly placed face down on the soft, less mobile.
deformable surface and found dead after an interval of Some hazardous sleeping environments are likely to be
0.5-10 hours. Thirty-four of the 35 infants were below the appreciated only when timely death scene investigations
median SIDS age of 13 weeks. Thirteen were only 1 month are conducted. 98 Excessive room heating and probable car
of age. A combination of factors is thought to be respon bon monoxide or carbon dioxide poisoning were problems
sible for these deaths, including rebreathing in a confined discovered by Bass et a[3 when thorough investigations
area, obstruction of nose and mouth, altered permeability were done. Leaving infants sleeping in poorly ventilated
of cushion coverings by wetting with secretions, hyper vehicles on sunny days poses a similar hazard.
thermia and the instability of the bag. The possibility of accidental asphyxia, when mothers,
The use of 'V'-shaped pillows for young infants should who are very tired or sedated, have been breast-feeding in
also be avoided. 86 Movements during sleep can result in bed, has been explored by Bass et al 3 and Byard. 99 The for
wedging in the pillow angle from which the infant cannot mer considered that 3 of 26 SIDS cases resulted from acci
retreat and obstruction of the upper a irway can supervene. dental asphyxia in this situation. The latter found that 3 of
28 infants who died unexpectedly during a I-year period
in south Australia were in bed with their mother, who had
Other unsafe sleeping environments fallen asleep during breast-feeding. Although 'overlying' is
no longer considered a major cause of SUDI, circumsta nces
Other unsafe sleeping environments are some types of may dictate its serious consideration in some cases. 83
swing-cradles87 and cots with ill-fitting mattresses,88-90
which may permit wedging of the baby between mattress
and cot frame with impedance of respiratory movement. The Co-sleeping
continued use of broken cots also constitutes an unaccept
able hazard for infants,91 who can become entrapped Bed sharing with non-smoking parents is enco uraged by
through gaps between broken or loose cot components. many health carers as a means of facilitating breast-feeding,
Parental modification of cots can also be problematic. Byard despite a small increased risk of death among co-sleepers. 100
et a1 92 reported positional asphyxia in two infants whose UNICEF 67 stresses the need for a clean, firm mattress, avoid
parents had introduced an additional mattress into mesh ance of overheating from excessive clothing and hazardo us
sided cots as the original mattresses were thin. This had the situations which might lead to wedging or falls, as well
effect of raising the sleep surface so that it abutted onto the as advising against bed sharing during maternal illness.
mesh section of the side-walls. This area is much more read Concern is expressed by some about the increased propor
ily distensible by the infant and permitted fixation of the tion of SUDI found after a co-sleeping night. Luke 101 drew
head between cot wall and mattress. Inappropriate cot attention to this problem when he found that an increasing
design is an additional hazard. Bass 88 reported 10 asphyxial proportion of SUDI within his jurisdiction occurred following
deaths seen over an I8-month period where cot design was a co-sleeping night. This was four times higher than usual
Pathological findings in SUDI I 205
practice 76 in a Scottish study. The increase in co-sleeping on review of pathological findings by an expert panel,
the night of death might be a marker of minor illness in the although death was considered explained by the examining
baby, well recognized in SIDS victims, but the spectre of pathologist in 30 per cent of cases. 4 Parham et al 109 found
positional asphyxia cannot be dismissed. Unusual bed shar previously unsuspected pathological abnormality in more
ing was also reported in another (small) study52 than half of babies less than 6 months of age in a review of
lt is most important that the sleeping environment is sudden deaths investigated by the Medical Examiner's
examined with great care when an infant is found unexpect Office; death was considered to be explained in 35 per cent
edly dead in this situation. There are two potential problems, of cases. Only 39 per cent of SUDI were considered typical
asphyxiation and overheating. As well as the precise pos SIDS cases in a South Australian study, no although a further
ition of the infant - prone/supine - the level of the infant in 21 per cent were deemed 'undetermined'. Knowledge of the
relation to the parents and covers and the llse of pillows and circumstances of death may alter the pathologist's expect
bed coverings, particularly heavy duvets, should be explored. ation of positive pathological findings, but facilities for full
The state of the mattress and its coverings, particularly the investigation should be available at the outset.
presence of blood-stained secretions and evidence of sweat The findings following the sudden death of older infants
ing, as well as the relative weights of the parents need to be and children are discussed in Chapter 12.
considered. This is in adclition to a clear history about
drug/alcohol ingestion and levels of tiredness. Parental alco
hol consumption and co-sleeping on the night that the baby The Pathology of Explained Natural Death
died were frequent findings in two Scandinavian stud
ies,I02,103 although this was not a factor in one American The range of pathological findings encountered in SUDI are
study.98 I find that lack of detail about many of these factors shown in Table 11.2111 (see also Byard and Krous 58 ). Patho
hampers evaluation of individual cases. Rushton 104 has logical findings in SUDI are discussed in detail by Byard
expressed similar concerns and draws attention to the and Cohle, 112 There are no specific pathological markers for
inevitable lack of important detail related to relative pos SIDS despite the best efforts of many investigators over the
itions of parents and baby during sleeping hours. The loca years.
tion of fixed lividity may be informative in some cases.
Sleeping on a sofa with an adult is a particularly haz INSTANTANEOUS DEATH
ardous situation. l05 In the CESDI SUDI studies, this factor
contributed the greatest risk, with an odds ratio (OR) of When infant death is instantaneous, then the type of patho
31 :25. It was confirmed in a Scottish study (OR 66.9, 95 logical findings that are likely come from a more restricted
per cent confidence interval [CI] 2.8-159.7) 106 and group. Most of the deaths will have a cardiovascular cause.
re-emphasized in a more recent study from BristoL 37 In my Cardiac malformations, particularly ductus-dependent
experience, shared by other pathologists, risk is enhanced abnormalities, are the most common group. One-third of
when alcohol has been consumed by the carer. 83 babies with cardiac malformations were awake at the time of
collapse in one study. 113 These are particularly common in the
first month of life. Nevertheless, infants with some malforma
PATHOLOGICAL FINDINGS IN SUDI tions that are particularly associated with sudden neonatal
death, such as hypoplastic left heart syndrome, do survive
As the number of unexplained, presumed natural deaths in into the post-neonatal period. A surprising number of infants
the infant age group has fallen during the last decade, so the with isolated ventricular septal defect die suddenly.114 Up to
proportion of explained deaths in medicolegal autopsies one-third of affected children have arrhythmias. Detailed
might be expected to rise, Although the increased range and examination of the cardiac conduction system has shown
rapidity and improved accessibility of di,agnostic tests may fibrous interruption of conduction tissueY5 Sudden death in
be expected to enable a pre-mortem diagnosis to be reached infants who have undergone surgical correction of congeni
in more babies. In a UK multicentre study, 1976-79, 18 per tal heart disease (CHD) some months previously demands
cent of victims of unexpected death had evidence of serious detailed examination of the cardiac conduction system. The
acquired disease,16 although this may not have explained distribution of the cardiac conduction tissue is frequently
death in every case. In a retrospective study by Czegledy aberrant in CHD, and inflammation around suture material
Nagy et aI, 107 there was an explanation for SUDI in 20 per can have fatal consequences. A range of other cardiac pathol
cent of cases. Nine per cent were non-natural deaths. Only ogy has been identified in SUDI llG but, in an individual case,
nine per cent of SUDI in an Adelaide study were thought to requires careful consideration of its relevance to causation.
be unexplained. 108 Subjective interpretation of pathological Coronary arteritis, usually a manifestation of Kawasaki's
findings is, to some extent, inevitable. The CESDI SUDI disease, is encountered in the infant age group, more com
investigations during 1993-96 emphasize the effect of sub monly in infants over 6 months of age. ll7 As well as diffuse
jectivity in the interpretation of findings. In that study, only thickening of the coronary arteries (Fig. 11.1), aneurysms
17.8 per cent of the deaths were considered explained after may be found in the proximal parts of the arteries. Other
206 I Sudden unexpected death in infancy
Instantaneaus death
Natural causes
Cardiac malformation
Coronary artery anomalies
Coronary arteritis
Myocarditis
Cardiomyopathy
Dysrhythmia
Acute laryngitis
Laryngeal cysts
Laryngeal papilloma
Laryngeal/tracheal developmental anomalies
Acute intestinal obstruction, e.g. volvulus
Non-natural causes
Accident
Non-accidental injury
Foreign body in airway
Table 11.3 Causes of sudden neonatal death Table 11.4 Typical post-mortem findings in SIDS (after ref. 711)
Congenital heart di sease - particularly ductus-dependent External
lesions Well nourished , normally developed baby
Infection - both bacterial and viral Frothy blood-tinged fluid around the nose (500f0)
Birth-related causes - both hypoxic and traumatic Hypostatic staining often anterior (indicates face-down position)
Microscopic
Thickened laryngeal basement membrane
Pulmonary congestion and oedema
Mild inflammation of the upper respiratory tract
Focal fibrinoid necrosis of the voca l folds
Persistent haemopoiesis in the liver
~
Pathological findings in SUDI I 211
found (60 per cent of cases) in both the upper and lower
respiratory tract in SIDS. The former is not likely to be a
primary cause of death J74 unless it results in major narrow
ing of the airway. The latter provokes the question 'How
much pathology is required to equate with a cause of
death?'. This is a difficult question and one that has been
sidestepped on occasion. 16 Consolidation throughout sec
tions from more than two lung lobes is a yardstick I have
used. Gregersen et al 175 consider the extent of pulmonary
disease a critical factor in deciding whether or not death is
explained. Culture of a single bacterial pathogen from the
organ in question and blood cultures is likely to influence
interpretation of histological findings.
Localized inflammation might acquire significance
because of its localization, for example in the cardiac con
duction system or brainstem. 32 These authors are uncertain of Figure 11.9 SUDI, frozen section of liver. Microvesicular fat is
the significance of focal inflammation in both the meninges present in zone 1; this amount of lipid is a frequent finding in
and myocardium, although an earlier publication appears to unexplained infant deaths. Oil Red O.
accept that significant myocarditis may be focal. 140
Basement membrane thickening in the vocal cord was a
frequent finding in SIDS in one study.176 However, in a
larger study Krous et al J77 found no difference in laryngeal
basement membrane thickening between SIDS cases and a
control group. More serious damage in the form of fibri
noid necrosis of the vocal fold is seen in some SIDS
babies. 17s
In the kidney, glomerulosclerosis may be a striking
finding. It is not significant and is seen frequently in
explained death in this age group. J79
Persistence of haemopoiesis in the liver, which normally
disappears soon after birth, is seen in many SIDS cases and
has been promoted as evidence of repeated or chronic
hypoxia. ISO, lSI
Cytomegalovirus (CIvrv) inclusions have been recorded
in SIDS babies for many years. Their incidence in salivary
Figure 11,10 Reperfusion injury in an infant who was on life
gland or mucous glands in the upper respiratory tract is no
support for hours after collapse at home. There is a small, old
more frequent in SIDS than in explained SUDI. IS2 However,
calcified lesion in the periventricular white matter on the right
their association with microglial nodules in the brain stem
indicative of a pre- or perinatal ischaemic lesion.
has provoked the suggestion that, in this circumstance, the
effect of crvrv infection is likely to be significant. ls3
Evidence of minor old periventricular leucomalacia is
sometimes seen in SIDS babies (Fig. 11.9), present in one
fifth of cases in a Canadian study.184 This is perhaps not A number of changes found at a histological level in SIDS
surprising given the association with SIDS of low birth have come to be regarded as markers of 'stress', i.e. non
weight, pre-term delivery and neonatal morbidity. A more system-specific markers of illness. These changes are seen in
frequent finding is disruption of the ependymal lining of many SIDS deaths, in th e thymus, liver, adrenals and the
the lateral ventricles, which, when recent, results in atten osteochondral interface (most conveniently accessed at the
uation of the columnar epithelial lining and, when longer anterior ends of the ribs).
standing, results in ependymal rosettes and attendant glio A starry-sky appearance in the thymus, perhaps accom
sis. Brainstem gliosis is a relatively common finding. ISS, 186 panied by minor cortical lymphocyte depletion, is compatible
A variety of abnormalities of brainstem serotonergic sys with a SIDS designation, but major cortical lymphocyte
tems have been described in SIDS following detailed mor depletion or reversal of cellularity betvveen cortex and
phometric and immunohistochemical studies, including medulla indicates a longer (and probably more severe)
hypoplasia of the arcuate nucleus and subtle gliosis of sev episode of stress, 166
eral brainstem nuclei. These are reviewed by Kinney and Minor lipid accumulation in the liver (Fig. 11.10) is a
Filiano ls7 and Kinney.18s marker of minor stress. It is microvesicular and usually more
212 I Sudden unexpected death in infancy
Figure 11.12 (a) A rib cage with recent fractures at the anterior ends of three adjacent ribs; there is little associated haemorrhage.
(b) Radiograph of affected ribs. The fractures are accompanied by irregularity of outline at their anterior ends. (c) Internal aspect of the
ribs. Fractures are visible close to the osteochondral junction. (d) Histological examination reveals a fracture with haemorrhage but no
reactive changes. Sudden collapse, attempted resuscitation, anomalous pulmonary venous drainage found at necropsy.
petechiae in the head and neck, and only 2.6 per cent had evidence of contusions, petechiae or intraoral trauma in
three or more petechiae in total. 197 Their presence raises the overlaying and considered that their presence should
possibility of deliberate asphyxia (Fig. 11.14). When pre incline towards inflicted injurY.
sent over the face, neck and upper chest, petechiae are sug Conjunctival petechiae should be very carefully sought.
gestive of chest compression.lg8-200 The infant should be Kleeman et al 168 found small numbers of conjunctival
carefully examined, searching for bruising around the face petechiae in 2.4 per cent of their SIDS cases. This has not
and circumoral pallor, tears at the frenulum (recent or been confirmed by others. Larger numbers of petechiae
healed) and abrasions or bruises on the inside of the mouth were seen in 21.9 per cent of babies dying as a result of
and lips corresponding to teeth. Collins 98 found little trauma, including in five out of six babies dying from
214 I Sudden unexpected death in infancy
Accidental bruises are usually over bony prominences. 206 Pulmonary congestion and oedema, although seen in
Bruises of different ages are most concerning. histological sections, is not a prominent naked-eye finding
in SIDS.
FRACTURES
PULMONARY HAEMORRHAGE/SIDEROPHAGES
Fractures are not an expected finding in SIDS and must be
Histological findings causing concern in SUD! are largely
clearly explained. Fractures with callus always predate ter
confined to the lungs. The abnormalities that have proved
minal events, probably by a minimum of 2 weeks, and are
most contentious are the presence of widespread alveolar
incompatible with the use of SIDS on the death certificate.
haemorrhage and the presence of haemosiderin within the
Rib fractures are sometimes seen in extremely pre-term
lung (Fig. 11.15).214 Minor (focal) pulmonary haemorrhage
babies. They are accompanied by severe growth distur
is common in the lungs of deceased infants and may be
bance at the osteochondral junction (rickets). In most units,
exacerbated by a long post-mortem interval. 215 Coffin et
they are encountered less frequently than 20 years
al 216 found that babies with pulmonary haemorrhage had a
ago. 207,2oa However, rickets was described more recently
high frequency of obstetric and neonatal problems; there
among 39 per cent of very low birth weight babies in one
was a significant association with hyaline membrane dis
unit, with fractures, most commonly of ribs, occurring in
ease and haemorrhage elsewhere.
one-quarter of those affected.209 Fractures are unlikely to
There are numerous causes of alveolar haemorrhage in
occur after the baby is weI] enough to leave hospital.
neonates, including hypoxic stress, sepsis and haemostatic
Should birth injury be offered as an explanation for a frac
problems,216-218 not all of which are easily excluded by
ture, the age of the infant and details of mode of delivery
necropsy. In older infants, sepsis, heart failure (often sec
are important factors. It is worth remembering that the
ondary to CHD) and chronic pulmonary infection, as well as
most common birth-related fracture is clavicular, and a
idiopathic pulmonary haemosiderosis (IPH), can give rise to
history of difficul ty in delivery of the shoulders should be
haemorrhage. It is more prominent in babies who have been
anticipated.
resuscitated. 2i s
Femoral fractures (usually mid-shaft, sometimes bilat
Yukawa et al 2J9 looked at the presence and extent of
eral) can complicate difficult breech delivery but do not
intra-alveolar haemorrhage in a consecutive series of SUD!
occur during vertex presentation. The most common skull
investigated by a forensic pathology department. They
fracture is parietal and linear and runs from the mid-part of
concluded that alveolar haemorrhage involving more than
sagittal margin of the parietal bone down towards the ear
five per cent of alveolar area in histological sections was
and usually follows instrumental delivery.
probably a marker of airways obstruction. It was present in
Rib fractures are exceedingly rare birth injuries and sug
73 per cent of babies thought to have been subjected to
gestion of this mode of causation should be viewed most
involuntary overlaying (accidental asphyxia) and in 45 per
circumspectly. Only nine cases have been reported world
cent of cases in which there were suspicions of deliberate
wide. Taken together, a pattern emerges of high birth weight,
airways obstruction.
shoulder dystocia, delayed second stage of labour, instru
Every consideration must be given to possible natural
mental delivery and symptoms in the first day or two of life.
causes of pulmonary haemorrhage before asclibing the find
Symptoms include crepitus of the chest wall, tachypnoea
ings to imposed airways obstruction. Concerns expressed
and grunting respiration. That the fractures were unilateral is
about the more general application ofYukawa et al's219 find
a useful distinguishing point.210-213
ings are the paucity of typical SIDS cases studied and no sig
nificant difference between number of allegedly asphyxiated
babies and SIDS cases having greater than five per cent
VISCERAL PATHOLOGY
haemorrhage.214 The finding that intra-alveolar haemor
Injuries and foreign bodies in the mouth, pharynx or stom rhage was most common in those babies in whom over-laying
ach, particularly in premobile infants, should arouse suspi was a possible factor in their death invokes the possibility
cion of NAI. 10 Intrathoracic petechiae are seen less frequently that repeated non-fatal hypoxia played an important role in
in traumatic deaths 168 and among bed-shaling infants 10) than the haemorrhage, drawing a parallel with the development
in SIDS cases in all of the three expected sites. In contrast, of petechial haemorrhages. The role of chest compression is
small numbers of larger haemorrhagic subpleural haemor unclear. 21 9 This group of babies was younger than the rest,214
rhages are seen in some asphyxial deaths. When the asphyx a factor associated with more frequent, but scanty, pul
ial insult is very acute, for example a foreign body in the monary siderophages in another study.220 As an isolated
larynx, very few petechiae may be present as death by finding, intra-alveolar haemorrhage is not a clear marker for
parasympathetic stimulation may have preceded the onset of incontrovertible imposed upper airways obstruction but
mechanical asphyxia. In rabbits, pleural petechiae were read should initiate a critical review of the whole case. 214
ily produced by three episodes of sublethal airways occlusion Stewart and Fawcett,221 in an uncontrolled study, found
but not by a single apnoeic episode.2J3 pulmonary interstitial haemosiderin in one-half of 24
216 I Sudden unexpected death in infancy
, ', -I t
be longer than the statutory period for death registration 10 Meadow R. UnnalUral sudden infant death. Arch Dis Child
Barness B in that, when history, circumstances and naked 12 Bacon CJ. Cot death after CESDI. Arch Dis Child 1997;
eye findings are typical , I use the terms SUD! or 'sudden 76:171-3.
unexpected infant death', together with an indication to the 13 Limerick SR, Bacon CJ. Terminology used by pathologists in
registrar that additional information may become available. reporting on sudden infant deaths. ] Clin Pathol 2004;
57:309-11.
In recent years, pathologists have become much more
14 Flemming PJ, Blair PS, Sidebottom P, Hayler T. Investigating
likely to use the terms 'unascertained' or 'undetermined' sudden unexpected deaths in infancy and childhood and
on death certificates, although in the UK practice varies caring for bereaved families: an integrated multiagency
widely.13 This has provoked protest from parents' support approach. Br Med] 2004; 328:331-4.
groups. However, it is the pa thologist, not the counsellor, 15 Krous HF, Beckwith JB, Byard RW et al. Sudden infant death
syndrome and unclassifi ed sudden infant deaths:
whose opinion or competence is called into question when
a definitional an d diagnostic approach. Pediatrics 2004;
an atypical infant death is reinvestigated after the abuse or 114:234-8.
death of another family member. 9 Mitchell et al lio have 16 Knowelden J, Keeling J, Ni choll JP et a1. A Multicentre Study
highlighted the swing away from SIDS to 'undetermined'. It of Post-neona ta l Mortality. London: HMSO, 1984.
is important to consider the potential consequences of the 17 Gibson AAM. Current ep id emiology of SIDS. ] Ciin Pathol
1992; 45(Supp1.):7-10.
use of SIDS, in effect stating that death is due to natural
18 Gilbeli RE, Fleming PJ, Azaz Y, Rudd PI. Signs of illness
causes, on subsequent infants in the family.214 The use of preceding sudden unexpected death in infants. BM] 1990;
'unascertained' in these circumstances ensures that the case 300:1237-9.
is kept open and might alert medical and support services 19 Wigglesworth JS , Keeling JW, Rushton DI, Berry P1.
to the need for vigilance and an open mind when assessing Technical method. Pathological investigations in causes of
sudden infant death.] Clin Patlzol 1987; 40:1481-3.
problems in subsequent infants. The importance of multi
20 Royal College of Pathologists. Guidelines for Post-mortem
disciplinary review of unexpected infant deaths has been Reports. London: Royal Co llege of Pathologists, 1993, pp. 8- 10.
stressed earlier in this chapter. 21 Valdes-Dapena M, McFeeley PA, Hoffman HJ et al.
Histopathology Atlas fo r the Sudden Infant Death Syndrome.
Washington DC: Armed Forces Instit Pathol, 1993.
22 Krous HF. An international standardized autopsy protocol
REFERENCES for sudden unexpected infant death. In Rognum TO (ed.)
Sudden Infant Death Syndrome. New Trends in the Nineties.
Carpenter RG, Shad dick Cw. Role of infection, suffocation Oslo: University of Sca ndinavia Press, 1995, pp. 75-89.
and bottle feeding in cot death. Br ] Pre/} Med 1965; 19: I. 23 Howatson AG. The autopsy for sudden unexpected death in
2 Beckwith JB. Introdu ction: discussion of terminology. In infancy. Current Diagnostic Pathology 2006; 12: 173-83.
Bergman AB, Beckwith JB, Ray CG (eds) Sudden Infant 24 Report of a Working Group: Sudden Unexpected Death in
Death Syndrome. Washington DC: University of Washington Infancy. London: Royal College of Pathologists and Royal
Press, 1970, p. 18. College of Paediatrics and Child Health, 2004.
3 Bass M, Kravath RE. Glass L. Death-scene investiga tion in 25 Landi K, Gutierrez C, Sampson B et al. Investigation of the
sudden infant death. N Engl] Med 1986 ; 315:100-5. sudden death of infants: a multicenter anal ys is. Pediatr Del;
4 Berry J, Allibone E, McKeever P et al. The parhology study: Patl101 2005; 8:630-8.
the contribution of ancillary pathology tests to the 26 Sadler DW. The va lue of a thorough protocol in the
investigation of unexpected infant death. In Fleming P, investigatio n of sudd en infant deaths.] Ciin Pathol 1998 ;
Bacon C, Blair P, Berry PJ (eds) Sudden Une.rpected Deaths 51 :689-94.
in Infancy th e CESDI SUD! Studies 1993-1996. London: 27 Mitchell E, Krous HF, Donald T, Byard RW. An analysis of
The StationeJY Office, 2000, pp. 97-112. the us efuln ess of specific stages in the pathologic
5 Willinger M, James LS, Catz C. Defining the sudden infant investigation of sudden infant death. Am ] Forensic Med
death syndrome (SIDS): deliberations of an expert panel Pathol 2000; 21 :395-400.
convened by the National Institute of Child Health and 28 Langlois NE, Ellis PS, Little D, Hulewicz B. Toxicologic
Human Development. Pediatr Pathol 1991; 11 :677-84. analysis in cases of possible sudden infant death syndrome:
6 Fleming P, Blair P, Ward-Platt M et al. The case-control a worthwhile exerc ise? Am] Forensic Med Pathol 2002;
study: results and discussion. In Fleming P, Bla ir P, 23:162-6.
Bacon C, Berry J (eds) Sudden Une.rpecred Deaths in 29 Arnestad M, Vege A, Rognum TO. Evaluation of diagnostic
lnfa ncy: The CESD! SUD! Studies. London: The Station ery tools applied in the examination of sudd en un expected
Office, 2000, pp. 13-96. deaths in infancy and early childhood. Forensic Sci Internal'
7 Bajanowski T, Vennemann M, Rauch E et al. Unnatural 2002; 125:262-8.
causes of sud den unexpected deaths initially thought to be 30 Keeling JW, Knowles SAS. Sudden death in childhood and
sudden infant death syndrome. Int] Legal Med 2005; adolescence.] Patho11989; 159:221-4.
119:21J-16. 3I Arneil GC, Gibson AA, McIntosh H et a1. National post
8 Gilb ert-B arness EF, Barness LA. Sudden infant death perinatal infant mortality and cot death study, Scotland
syndrome. Is it a cause of death? Arch Pathol Lab Med 1981-82 . Lancet 1985; 1:740-3.
1993; 117: 1246-8. 32 Byard RW, Krous HF. Minor inf1ammatory lesio ns and
9 Green MA. Time to put 'cot dearh' to bed 7 BM] 1999 ; sudden infant death: cause, coincid ence, or ep iphenomena?
319:697-8. Pediatr Pathol Lab Med 1995; 15:649-54.
220 I Sudden unexpected death in infancy
33 Bacon C, Fleming P. Design of the SUDI study. In Fleming P, British Paediatric Association. London: Roy Soc Med Press,
Blair P, Bacon C, Berry J (eds) Sudden Unexpected Death In 1994.
Infancy. Th e CESD! SUD! Studies. London: The Stationery 54 Rosen CL, Frost JD, Glaze DG. Child abuse and recurrent
Office, 2000, pp. 7-12. infant apnea.] Pediatr 1986; 109: 1065-7.
34 Irgens LM. Risk factors for SIDS: Do they exist? In Rognum 55 Southall DP, Plunkett MCB, Banks MW et al. Covert video
TO (ed.) Sudden Infant Death Syndrome. Os lo: Scandinavian recordings of life-threatening child abuse: lessons for child
Uni vers ity Press, 1995, pp. 99-105. protection. Pediatrics 1997 ; 100:735-6.
35 Blair PS, Fleming P J, Bensley D et al. Smoking and the 56 Berger D. Child abuse simulating 'near- miss' sudden infan t
sudden infant death syndrome: results of 1993-5 case death syndrome.] Pediatr 1979; 95:554-6.
control study for confidential inquiry into still bil1hs and 57 Ariagno RL, Guilleminault C, Korobkin R et al. 'Near-miss'
deaths in inFancy. 8M] 1996; 313:195- 8. for sudden infant death syndrome infants: a clinical
36 Aim B, Milerad J, Wennergren Get al. A case-control study probl em. Pediatrics 1983; 71:726-30.
of smoking and sudden inFant death syndrome in the 58 Byard RW, Krous HF. Sudden infant death syndrome:
Scandinavian countries, 1992 to 199 5. Arch Dis Child overview and update. Pediatr Dellelop Patlwl 2003; 6: 1
1998 ; 78 :329- 34. 12-27.
37 Blair PA, Sidebotham P, Berry P J et a l. Major 59 Truman TL. Ayoub CC. Consid ering suffocatory abuse and
epidemiolo gica l changes in sudden infant death syndrome: a Munchausen by proxy in the evaluation of children
20-year population-based study. Lancet 2006; 367: experiencing apparent life-threatening events and sudden
314-19. infant death syndrome. Child Maltreat 2002: 7:138-48.
38 Daltveit AK, 0yen N, Skjaerven R, Irgens LM. The epidemic 60 Galvin HK, Newton AW, Vandevan AM. Update on
of SIDS in Norway 1967-93: chan ging effects of risk factors. Munchausen by proxy. Curr Opill Pediatr 2005; 17:252-7.
Arch Dis Child 1997; 77:23-7. 61 Becroft DMO, Thompson JMD , Mitchell EA. Nasal and
39 Smith Gc, Wood AM, Pell JP, Dobbi e R. Sudden infant death intrapulmonary haemorrhage in sudden infant death
syndrome and complications in other pregnancies. Lancet syndrome. Arch Dis Child 2001; 85:116-20.
2005; 366:2107-11. 62 Barber MA, Davis PM. Fits, faints, or fatal fantasy? Fabricated
40 Malloy MH. Sudden infant death syndrome among seizures and child ab use. Arch Dis Child 2002; 86:230-3.
extremely pre-term infants: United States 1997-1999. 63 Proudfoot A. Is the cause toxicological? Medicine 1999;
] Perinatol2004; 24:101-7. 27(4):4.
41 Leach CE, Blair PS, Flemming P J et al. Epidem iology of SIDS 64 Emery JL, Thornton JA. Effects of obstruction to respiration
and explained sudden infant deaths: CESDI SUDI research in infants, with particular reference to mattresses, pillows,
group. Ped,:atrics 1999; 104:e43. and their coverings. 8M} 1968; 3:209-13.
42 Golding J, Peters TJ. What else do SIDS risk prediction 65 Foundation For the Study of InFant Death s. Reduce the risks
scores predict? Early Hum Develop 1985; 12:247-60. of cot death - an easy guide (Leaflet). FSID, 1996.
43 Smith GC, Wood AM, Pell JP et a!. Second-trimester 66 Foundation For the Study of Infant Deaths. Reduce the ris.~s
maternal serum alphaFeto protein levels and the subsequent of cot death (Leafl et) . FSID, 1991.
risk of sudden inFant death syndrome. N Engl] Med 2004; 67 UNICEF UK. Sharing a bed with your baby. London: UNICEF
351 :978-86. UK,2003.
44 Bass M. The Fallacy of the simultaneous sudden infant death 68 Scragg KR, Mitchell EA, Stewart AW et al. Infant room
syndrome in ('wins. Am] Forensic Med Pat/101 19139 ; sharing and prone sleep position in sudden infant dea th
10:200-5. syndrome. Lancet 1996 ; 347:7-12.
45 Carpenter RG, Gardner A, McWeeny PM , Emery JL. 69 Stanton AN. Sudden inFant death. Overh eating and cot
Multistage scoring system For identifying inFants at risk of death. Lancet 1984; 2:1199-2 01.
unexpected death. ATCh Dis Child 1977; 52:606-12. 70 Fleming P J, Gilbert R, Azaz Y et a!. Interaction between
46 Golding J, Limerick S, Macfarlane JA. Sudden Infan/ Death: bedding and sleeping position in the sudden inFant death
Pattems, Puzzles and Problems. London: Op en Books, 1985, syndrome: a popul ation based case-control study. 8M}
pp. 94-5, 190-5. 1990; 30:85-9.
47 Ward-Platt M, Blair PS, Fleming PJ et al. A clinical 71 Wigfield RE, Fleming PJ, Azaz YEZ et al. How much
comparison of SJDS and explained sudden inFant deaths: wrapping do babies need at night? Arch Dis Child 1993;
ho w healthy and ho w norm a l? Arch Dis Child 2000 ; 69:181-6.
82:98-106. 72 Beal SM, Finch CF. An overview of retrospective case
48 Overpeck MD, Brenner RA, Trumble AC et al. Risk factors for control studies investigating the relationship between prone
inFant homicide in the United States. N Engl] Med 1998; sleep ing position and SIDS: } Paediatr Child Health 1991;
339 :1211-16. 27:334-9.
49 Stanton AN, Downham MAPS, OakJey JR et al. Terminal 73 Beal SM. Sleeping position and sudden inFant death
symptoms in children dying suddenly and unexpectedly at synd rom e. Med] Aust 1988; 149:562.
home. Preliminary report of the DHSS multicentre study of 74 de Jonge GA, Engelberts AC, Koomen-Liefting AJM,
postneonatal mortali ty. 8M) 1978; 2:1249-51. Kostense P J. Cot death and prone sleeping position in The
50 Stanton AN, Oakley JR. Pattern of illnesses beFore cot Netherlands. 8M] 1989; 298:722.
deaths. Arch Dis Child 1983; 58:878-81. 75 Mitchell EA, Scragg R, Stewal1 AW et a!. Results From the
51 Ford RPK, Mitchel! EA, Stewart AW et al. SIDS, illn ess, and first year of the New Zealand cot death study. N Z Med]
acute medical care. Arch Dis Child 1997; 77:54-5. 1991; 104:71-6.
52 L'Hoir MP, EngelbeJ1s AC, van Well GThJ et al. Sudden 76 Brooke H, Gibson A, Tappin D, Brown H. Case-control study
unexpected death in infancy: epidemiologically determined of sudden inFant death syndrome in Scotland, 1992- 5. 8M}
risk factors related to pathological classification. Acta 1997 ; 314:1516-20.
Paediatr 1998; 87:1279-87. 77 Mitchell EA , Hutchison L, Stewart AW. The continuing
53 British Paediatric Association. Report of a Working Party. decline in SIDS mortality. Arch Dis Child 2007; 92:
Evaluation of Suspected Imposed Upper Airway ObSfillction. 625-6.
Referen ces I 221
78 Ponsonby A-L, Dwyer T, Couper 0, Cochrane J. Association 102 Rintahaka PJ, Hirvonen J. The epidem iology of sudden
between use of quilt and sudden infant death syn drome: infant death syndrome in Finl and in 1969-1980.] Forensic
case-control study. BM] 1998; 316: 195-6. Sci Intern 1986; 30:2 19-33.
79 Waters KA, Gonzalez A, Jean C et a!. Face-straig ht-down
103 Norvenius SG. Sudden infant death syndrome in Sweden in
and face-near-straight-down posilions in healthy, prone
1973- 1977 a nd 1979. Acta Paediatr Sco nd Supp/1987;
sleeping infants. ] Pediatr 1996 ; 128:616-25.
333:1-138.
80 Fleming PJ, Blair PS, Bacon C et a!. Environment of infants 104 Rushton Dl. Co-sleeping and sudden infant death syndro me.
during sleep and risk of the sudden infant death syndrome: Lancet 1997; 349:56-7.
results of 1993-5 case-control study for contidential 105 Blair PS, Flemin g PJ, Smith IJ et al. Babies sleeping with
inquiry into still births and deaths in infancy. BM] 1996; parents: case-control study of factors influencing the risk of
3 13:191 -5. the sud den infant death syndrome. BM] 1999; 319:1457-62.
81 Ramanathan R, Chandra 5, Gilbert-Barness E, Fra nciosi R. 106 Tappin 0, Ecob R, Brooke H. Bedsharing, roomsharing and
Sudden infant death syndrome and water beds (Letter to the sudden infant death syn drome in Scotland: a case control
editor). N Eng] Med 1988; 3 18 : 1700. study.] Pediatr 2005; 147:32-7.
82 Bass M. More on sudden infant death syndrom e and water
107 Czegledy-Nagy E, Cutz E, Becker LE. Sudden death in
beds (Letter). N Engl] Med 1988; 3 19:1415.
infants under one year of age. Pedian' Pathol 1993;
83 Gilbert-Barness E, Hegstra nd L, Chandra 5 et a!. Hazards of 13 :671-34.
mattresses, beds and bedding in deaths of infants. Am] lOB Byard RW, Carmichael E, Beal S. How useful is post-mortem
Forensic Med Patl101 1991; 12:27-32. examination in sudd en infant death synd rome? Pediatr
84 Kemp JS, Tha ch BT. Sudden death in infants sleep ing on
Pathol 1994; 14:817-2 2.
polystyrene-tllled cushions. N Engl] Med 1991 ;
109 Parham OM, Savell VH, Kokes CP et a \. Incidence of autopsy
324: 1858-64. findings in unex pected deaths of children and adolescents.
85 Gilbert-Barness E, Emery JL. Deaths of infants on
Pediatr De)} Pathol 2003; 6: 142-55.
polystyrene-filled beanbags. Am ] Forensic IVied Patllol
110 Mitch ell E, Krous HF, Donald T, Byard RW. Changing trends
1996; 17:202-6.
in the diagnosis of sudden infant death. Am] Forensic Med
86 Byard RW, Beal SM. V-shaped pillows and un safe infant
PatllOl 2000; 21 :311-14.
sleep ing.] Paediatr Child Health 1997; 33:171-3.
111 Berry PJ , Keeling J"I/. The in ves tigation of sudden
87 Moo re L, Bourne AJ , BealS e t aJ. Unexpected infant death in unex pected death in infancy. In A nth o ny PP, MacSween
associa tion with suspend ed rocking cradles. Am ] Forel1sic RNM (eds) Recent Advances in Histopath ology 14. London:
Med PatilO11995; 16:177-80. Churchill Livingstone, 1989, pp. 151-279.
88 Bass M. Asphyxial cri b death. N Ellgl] Med 1977; 296:
112 Byard RW, Cohle SD. Sudden Death in Infancy, Childhood
555-6.
and Adolescence. Cambridge: Cambridge Uni versity Press,
89 Smi a lek JE , Smialek PZ, Spitz WU . Accidental bed deaths in 1994.
infants due to un safe sleepin g situatio ns. c/in Pediatr 1977; IJJ Oancea A. Cote A. Roblicek C et al. Cardiac pathology in
11:1031-6. sudd en unexpected infant death.] Pediatr 2002;
90 Byard RW, Beal S, Bourne AJ. Potentially dangerou s s leeping 141 :336-42.
enviro nments and accide ntal asphyxia in infancy and ear ly 114 Cohl e SD, Balraj E, Bell M. Sudden death due to ven tricular
childhood. Arch Dis Child 1994 ; 17:497-500. septal defect. Pediatr Develop Pathol 1999; 2:327-32.
9) Variend 5, Usher A. Broken cots and infant fatali ty. Med Sci 115 Smith NM, Ho SY. Heart block and sudden death associated
Law 1984; 24:111 - 12. with tibrosis of the conduction system at the margin of a
92 Byard RW, Bourne AJ, Beal SM. Mesh -s ided cots: yet ventricul ar septal defect. Pediatr Cardiol 1994; 15:139-42.
anothe r potentially dangerous infant s leeping environme nt. 116 Bajanowski T, Ortmann C, Teige K et a\. Pathological
Forensic Sci Int 1996; 83:105-9. changes of the heart in sudden infant death. In t] Legal Med
93 Cassidy SP. Analysis of crib-related deaths for the years
2003; 11 7: I 93-203.
1989 through 1991 (memorandum). Bethesda, MD:
117 Levi n M, Tizard EJ, Dillon MJ. Kawasaki disease: recent
US Consumer Product Safety Commission, 1994.
advances. Arch Dis Child 1991; 66:1369-74.
94 Foundation for the Study of Infant Deaths. Newsletter FSID 118 Lipsett J, Cohle SD, Berry Pl, Byard RW. Anoma lous
1999; 59:12. cOTOnary arteries: a multicenter pediarric autopsy study.
95 Nakamura 5, Wind M, Danello MA. Review of hazard s
Pediatr Patl101 1994 ; 14: 28 7-300.
associated with children placed in adult beds. Arch Pediatr
119 Berry CL. Myocardial infarcti o n in a neonate. Br Heart ]
AdolescMed 1999; 153:1019-23.
1970; 32:41 2- 15.
96 Moore L, Byard RW. Pathological findings in hanging and
120 Tillett A. Hartl ey B, Simpson l. Parad oxical embQlism
wedging deaths in infants and young chi ldren. Am]
causing fatal myocardial infarcti on in a newborn infant.
Forensic Med Path ol 1993; 14:296- 302.
A rch Dis Child 2001; 35:FI37-B.
97 Bya rd RW, Beal SM, Simpson A et a!. Accidental infant
12 1 Bya rd RW. Vascular conditions. In Byard RW, Cohle SO
death and stroller-prams. Med] Aust 1996; 165:140-1.
(eds) Sudden Death in Infancy, Childhood and Adolescence.
98 Collin s KA. Death by overlaying and wedging. A IS-year
Ca_m bridge: Camb ridge Univers ity Press. 1994,
retrospective study. Am ] Forensic Med Pathol 2001;
pp. 175-261.
22:1 55 - 9.
122 Gilbert -Barn ess E, Barness LA. Nonmalformative
99 Byard RW. Is breast feeding in bed always a sa fe practice ?
cardiovascular pathology in infants and children. Pediatr
] Paedial,. Child Heal th 1998; 34:413-19.
Develop Pathol 1999; 2:499-530.
100 Scragg KR, Mitchell EA. Side sleeping position and bed 123 Fried K. Beer S, Vure E et al. Autosomal recessive sudden
sharing in the sudden infant death syndrom e. Ann Med un expected death in children prob ably caused by a
1998; 30:345-9. cardiomyopathy assoc iated with myopathy. ] Med Genet
JOI Luke lL. Sleeping arrangements of sudden infant death 1979; 16:341-6.
syndrome victims in the District of Columbia: a preliminalY 124 Ed wa rds WO. Cardiomyopathies. HlIm Pathol 1987;
report.] Forensic Sci 1978; 23:379-33. 6:625-35.
222 I Sudden unexpected death in infancy
125 Goodwin Jf. Sudden cardiac death in the young. A family deficiency in su dden infa nt death sy ndrome (SlDS).
history of sudd en death need s investigation. BM] 199 7; ] inherit Metab Dis 1993; 16:99 1-3.
314:843. J 48 Fitzpatrick D. Inborn erro rs of metabolism in the newbo rn:
126 Sua rez V, Fuggle WJ , Ca meron AH et al. Foamy myocardial clinical presentation and investigation. ] R Coli PhYSicians
transformation of infancy: an inherited disease. ] Ciin Patl70l Erlillb 2006; 36: 147-51.
1987; 40:3 29 -34. 149 Emery JL, Variend S, Howat AJ, Vawter GF. Investigation of
127 Shehata BM, Patterson K, Thomas JE et aJ. Histiocytoid inborn errors of metabolism in unexpected infant deat h.
ca rdiomyopathy: three new cases and a rev iew of the Lancet 1988; 2:29-3J.
literature. Pedialr Devel Patl70l 1998; 1 :56-69. 150 Lundemose JB, Kolvraa S, Gregersen N et al. Fatty acid
128 Adgey J, Johnston PW, Math ew TP. Sudden cardi ac death - a oxidation disorders as primary cause of sudden and
predictable, avoidable and treatable event? Proc R Coli unexpected death in infa nts and yo ung children: an
Physicians Edinb 1998; 28 :350-6. investigation performed on cultured fibrobla s ts from 79
129 Towbin JA. Molecular genetic basis of sudden cardiac death. children who died aged between 0- 4 years. ] Clin Patl7ol:
Perliatr Ciin N Am 2004; 51: 1229 -55. Mol Patho11997; 50:212-17.
130 Schwartz PJ, Stramba-Badiale M, Segantini A et al. 151 Boles RG, Buck EA, Blitzer MG et aJ. Retrospective
Prolongation of the QT interval and the sudden infant deat h biochemical sc reening of fatty ac id oxid ation disord ers in
syndrome. New Engl] Med 1998; 338:1709-14. post-mortem livers of 418 cases of sudden death in the first
131 Arnestad M, Crolti L, Rognum TO et al. Prevalence of yea r of life.] Pediatr 1998 ; 132:924-3 3.
long-QT sy ndrome ge ne vari a nts in sudden infant death 152 James CL, Keeling JW, Smith JW, Byard RW Total
syndrome. Circulation 2007; 115:361-7. anomalous pulmonary venous drainage associated with fat a l
132 Ackerman MJ, Tester DJ, Driscoll DJ. Molecular autopsy of outcome in infancy a nd early ch ildhood - an autopsy study
sudden unexplained death in the young. Am ] Forensic Med of 52 cases. Pediatr Parhol 1994; 14:6 65 -78.
Patlwl 200 I ; 22: 105- 11. 153 Keeling JW. The perinatal ne cropsy. In Keeling JW, Khong TY
133 Hanzlick RL. Case report: lingual thyroglossal duct cyst (eds) Fetal and Neonatal Pathology, 4th edn. London:
causin g death in a four-week-old infant. ] Forellsic Sci Springer-Verlag, 2007, pp. 20-5 3.
1984; 29 :3 45-8. 154 Byard RW Idiopathic arterial ca lcitlcation and unexpected
134 Kanawaku Y, Funayama M, Nata M, Kanetake 1. Sudden infant death. Ped Pathol Lab Med 1996 ; 16:985-94.
infant death: thyroglossal duct cyst ve rsus environmental 155 Champ C, Byard RW Pulmonary thromboembolism and
factors. Forensic Sci Int 2006; 156: 158-60. unexpected death in infancy. ] Paedialr Child Health 1994 ;
135 McNamarra VM, Crabbe DC. Tracheomalacia. Paed iatr Resp ir 30:5 50-1.
Rev 2004 ; 5: 147-54. 156 Berry PJ , Nagington J. Fata l infec tion w ith echovirus II.
136 Wong SW, Gardn er V. Sudd en death in children due to Arch Dis Child 1982; 57 :22 - 9.
mesenteric defect an d mesenteric cyst. Am ] Fo ren sic Med 157 Modlin JF. Perinatal echovirus infection: insights from a
Patho11992; 13:2 14-16. litera ture review of 6 1 cases of serious infection and 16
137 Byard RW. Sudden infant death, large intestinal volvu lu s ou tbreaks in nurseries. Re/J Infect Dis 1986; 8:918-26.
and a duplication cyst of the terminal ileum. Am ] Forens ic 158 Wang J, Atchison RW, Walpusk J, Jaffe R. Echovirus
Med Pat/wi 2000; 21:62-4. hepatic failure in infancy: report of four cases with
138 Byard RW, Simpson A. Sudden death a nd intussusception in speculation on the pathogenesis. Pediatr Devel Patl101
infancy and childhood-autopsy cons ideration s. Med Sci La w 2001; 4:4 54 -60.
2001; 41 :41-5. 159 Kaplan MH, Klein SW, McPh ee J, Harp er RG. Group B
139 Ng'wa la li PM , Yonemitsu K, Tsuenari S. Fatal coxsackievirus infections in infants you nger than three
intussusception in infancy: an exp erience in forensic months of age: a serious childhood illness. Rev II/fect Dis
autop sy. Leg Med (Tokyo) 2003; 5:181 - 4. 1983; 5:1019-32.
140 Smith NM, Bourne J , Clapton WK, Byard RW The spectrum 160 Wilcken B, Carpenter KH, Hammond J. Neonatal symptoms
of presentation at autopsy of myocarditis in infancy and in medium chain acyl coenzyme A dehydrogenase
childh oo d. Pathology 1992; 24: 129-3 I. deficiency. Arch Dis Child 1993; 69:292-4.
141 Shatz A, Hiss J , Arensburg B. Myocarditis misd iagnosed as 161 Kirk JM , La ing lA, Smith N, Uttley WS. Neonatal presentation
sudden infant death syndrome (SlDS). Med Sci Law 1997; of med ium-chain acyl-CoA dehydrogenase deficiency in two
37: 16-18. families.] Inheri t N/etab Dis 1996; 19:370-1.
142 de Sa DJ. Isol ated myocarditis in the first year. Arch Dis 162 Vawter GF, Koza kewi ch HPW Aspects of morph o logic
Child 1985; 60:484-5. va riation among SJDS-v ictims. In Tild o n JT, Roeder LM ,
143 Krous HF, Haas E, Chadwi ck AE, Wagner GN. Sudden death Steinschneider A (eds) Sudden Infant Death Syndrome.
in a neonate with idiopa thic eosin ophilic endomyocarditis. New York: Acad em ic Press, J 983, pp . 133-44.
PerliMr DelJel Pathol 2005; 8:587-92. 163 Molz G, Hal1mann H. Dysmorphism, dysplasia, and anomaly
144 Dettm eyer RB, Padosch SA, Madea B. Lethal enterovirus in sudden infant death. N Engl ] /vIed 1984; 311 :259.
indu ced myo card itis and pancreatitis in a 4-month-old boy. 164 Molz G, Brodzinowski A, Bar W, Vonlanthen B.
Forensic Sci i M 2006; 156:51-4. Morphologic variations in 180 cases of sudden infant
145 Losty HC, Lee P, AJfah am M et al. Fatty infiltrat ion in the death a nd 180 controls. Am] Forensic Med Patho11992;
liver of medium chain acyl CoA dehydrogenase deficiency. 13: 186-90.
Arch Dis Child 1991; 66:727-8. 165 Blair PA, Nadin P, Cole TJ et al. Wei ght gain and sudd en
146 Matsubaru Y, Narisawa K, Tada K et al. Preva lence of K329E infant death syndrome: cha nges in we ig ht z sco res may
mutatio n in medium-chain acyl- CoA dehydro genas e gen e identify infants at increased risk. Arch Dis Child 2000;
determ ined from Guthri e ca rds. Lancet 1991; 338: 82:462-9.
552-3. J 66 va n Baarlen J , Schuurman H-J, Huber J. Acute thymus
147 Dundar M, Lanyon WG, Con nor JM. Scottish frequency of involution in infancy and childhood: a reliable marker
the common G985 mutation in the medium-chain Acyl-CoA for duration of illness. Hum Pathol 1988; 19:
dehydroge nase (MCAD) gene and the role of MCAD 1155-60.
References I 223
167 Beckwith JB. Intrathoracic petechial haemorrhages: a clue to 189 Perez-Platz U, Saeger W, Dhom G, Bajanowski T. The
the mechanism of death in sudden infant death syndrome? pathology of the adrena l glands in sudden infant death
Ann Am Acad Sci 1988; 553:37-47. syndrome (SJDS). Inr J Legal Med 1994; 106:244-8.
168 Kleeman W J, Wiechern V, Schuck M, Trager HD. 190 Becker AE, Becker MJ. Fat distribution in the adrenal cortex
Intrathoracic and subconjunctival petechiae in sudden infant as an indication of the mode of intra-uterine death. Human
death syndrome (SIDS). Forensic Sci Int 1995; 72:49-54. Pathol 1976; 7 :495-504.
169 lsakasen CV, Helweg-Larsen K. The impact of attempted 191 Sinclair-Smith C, Dinsdale F, Emery J. Evidence of duration
resuscitation in SIDS: post-mortem findings. In Rognum TO and type of illness in children found unexpectedly dead.
(ed.) Sudden Infant Death Syndrome. Oslo: Scandinavian Arch Dis Child 1976; 51 :424-9.
Unive rsity Press, 1995, pp. 70-80. 192 Kaplan JA, Fossum RM. Patterns of facial resuscitation
170 Werthammer J , Brown ER, Neff RK, Taeusch HW Jr. Sudden injury in infan cy. Am J Foremic Med Pathol 1994;
infant death syndrome in infants with bronchopulmonary 15:187-91.
dys plasia. Pediatrics 1982 ; 69:301-3. 193 Bu s h CM, Jon es JS, Co hle SO, Johnson H. Pediatric injuries
171 Berry PJ. Pathological findin gs in SIDS. J Ciin Path 01 1992 ; from ca rdiopulmonary resuscitation . Anll Emerg Med 1996;
45(Suppl.): 11-16. 28:40-4.
172 Valdes-Dapena M. The pathologist and the s udde n infant 194 Feldman KW, Breweer OK. Child abuse, cardiopulmonary
death syndrome. Am J Pathol 1982; 106: 118-3 1. resuscitation, and rib fractures. Pedia trics 1984 ; 73:339-42.
173 Emery JL. Dinsdale F. Increased incid ence of Jy mphoreticular 195 Spevak MR, Kleinman PK, Belanger PL et al.
aggregates in lungs of children found une xpectedl y dead. Ca rdiopulm o nary resuscitation and rib fractures in infants: a
Arch Dis Child 1974 ; 49:107-11. post-mortem radiologic-pathologic study. JAMA 1994;
174 Krous HF, Nadeau JM, Silva PO, Blackbourne BD. A 272:617-18.
comparison of respiratolY symptoms and inflammation in 196 Price EA , Rush LR, Perper JA, Bell MD. Cardiopulmonary
sudden infant death syndrome and in acc idental or infli cted resuscitat ion- related injuries and homicidal blunt abdominal
infant death. Am J Forensic Med Pathol 2003; 24: 1-8. trauma in children. Am J Forensic Med Patho12000;
175 Gregersen M, Rajs J, Laursen H et al. Pathologic cri teria for 21 :307- 10.
the Nordic study of SIDS. In Rognum TO (ed.) Sudden Infant 197 Dow nes AJ, Crossland OS, Mellon AF. Preva lence and
Death Syndrome. Oslo: Scandinavian University Press, 1995, distribution of petechiae in well babies. Arch Dis Child 2002;
pp. 50-8. 86:291-2.
176 Shatz A, Hiss J, Arensburg B. Basement-membrane 198 Perrot U. Masque ecchymotique. Specific or nonspecific
thickening of the vocal cords in sudden infant death indicator for abuse. Am J Forensic Med Patlrol 1989;
syndrome. Laryngoscope 1991; )01 :484-6. 10:95-7.
177 Krous FH. Hauck FR, Herman SM et al. Laryngeal basement 199 Byard RW, Krous HF. Petechial hemorrhages and unexpected
membrane thickening is not a reliable post-mortem marker infant death. Leg Med (Tokyo) 1999; 1 :193-7.
for SIDS. Results from the Chicago Infant Monality study. 200 Oehmichen M, Gerling I, Meiilner e. Petechiae of the baby's
Am J Forensic Med Patho11999; 20:221-7. skin as differentiation symptom of infanticide versus SIDS.
178 Adelson L, Kinney ER. Sudden and unexpected death in J Forensic Sci 2000; 45:602-7.
infancy and childhood. Pediatrics 1956; 17:663. 201 Betz P, Hausmann R, Eisenmenger W. A contribution to a
179 Valctes-Dapena M, Hoffman HJ , Froelich C, Requeira O. possible differentiation between SIDS and asphyxiation.
Glomerulosclerosis in the sudden infant death syndrom e. Forensic Sci Jnt 1998; 91:147-52.
Pediatr Pathol 1990; 10:273-9. 202 Rao V J, Wetli CY. The forensic significance of conjunctival
180 Naeye RL. Hypoxemia and the sudden infant death petechiae. Am J Forensic Med Pathol 1988; 9:32-4.
syndrome. Science 1974; 186:837-8. 203 Hood I, Ryan D, Spitz WU. Resuscitation and petechiae.
181 Gilbert-Barness EF, Kenison K, Giulian G, Chandra S. Am J Forensic Med Patho11988; 9:35-7.
ExtramedullalY hematopoiesis in the liver in sudden infant 204 McIntosh N, Mok JYQ, Margerison A. Epid emio logy of
death syndrome. Arch Pathol Lab Med 1991; 115:226-9. oronasal hemorrhage in the first two years of life:
182 Smith NM, Telfer SM, Byard Rw. A compa rison of the implications for child protection. Pediatrics 2007;
in cidence of cytomegalovirus inclusion bodies in 120: 1074-8.
submand ibul ar and tracheobronchial g lands in SIDS and 205 Krous HF, Nadeau JM, Byard RW, Blackbolll"ne BD. Oronasal
non-SID S autopsies. Pediatr PnthoI1992; 12:185-90. blood in Sudden Infant Death. Am J Forensic Med Parhol
183 Variend S. Infant morta lity, microglial nodules and parotid 2001 ; 23:346-51.
CMY-type inclu sio ns. Ea,.ly Hum Deuel 1990; 21 :31- 40 . 206 Sugar NF, Taylor JA, Feldman KW, Puget Sound Pediatric
184 Takashima S, Armstrong 0 , Becker LE, Huber J. Cerebral Research Network. Bruises in infants and toddlers. Those
white maner lesions in sudde n infant death syndrome. who don't cruise rarely bruise. Arch Pediatr Adolesc Med
Pediatrics 1978; 62: 155-9. 1999; 153 :399-403.
185 Takas hima S, Armstrong 0 , Becker L, Bryan C. Cerebral 207 Amir J, Katz K, Grunebaum M et al. Fractures in premature
hypo perfusion in the sudde n infa nt death syndrome? infants. J Pediatr Orthop 1998 ; 8:41-4.
Brainstem gliosis and vasc ulature. Ann Neuro l 1978; 208 Koo WW, Sherman R, Succop Pet al. Fracture and rickets in
4: 257-62. very lo w birth weight infants: co nservative management and
186 Kinney H, Burger P, Harrell F, Hudso n RP. 'Reactive gliosis' outcome. J Pediatr Orthop 1989 ; 9:326-30.
in the medulla oblongata of v ictims of the Sudden Infant 209 Dabezies EJ , Warren PD. Fractures in very low birth
Deat h Syndrome. Pediatrics 1983; 72: 181-7. weig ht infants w ith rickets. Ciin Orthop Relat Res 1997;
187 Kinn ey HC, Filiano JJ. Brain research in SlDS. [n Byard RW, 335:233-9.
KrOllS HF (eds) Suddell Infant D~ath Syndrome. Pmblems, 210 Barry PW, Hocking MD. Infant rib fracture - birth trauma or
Progress and Possibilities. London: Arnold, 2001, pp. 118- 37. non-a ccidental injury? Arch Dis Child 1993; 68:250.
188 Kinney He. Abnormalities of the brainstem serotonergic 211 Bullock B, Schu bert CJ, Brophy PO et al. Cause and clinical
system in the sudde n infant death syndrome: a review. characteristics of rib fractures in infants. Pediatrics 2000;
Pediatr Deuel Pathol 2005; 8:507-24. 105:e48.
224 I Sudden unexpected death in infancy
212 Durani Y, Depiero AD, Rib fractu re and birth trau ma, Ann 228 Han zli ck R, Del aney K. Pulm onary haem os iderin in dece ased
Emerg Med 2006; 47 :2 10,2 15, infants: baselin e data for furth e r study of infant mortality.
213 Campbell CJ, Read DJ(, Lung petechiae - their Am) Forensic lVled Pat/wi 2000; 21:319-22,
pathophysiology and diagnostic sig nificance in co t death, 229 Schlu ckebier DA. Coo l CD, Henry TE et al. Pulmon ary
Aust Physiol Pharmacol Soc 197 8; 9 : 143, sid erop hages and unexpected infant death , Am) Forens ic
214 Berry PJ, Intra-al veo lar haemorrh age in sudd e n infant death Med Pat/wi 2002; 23:360-3.
syndrome: a cause for concern? ) Ciin Pa thol 1999; 230 Krous HF, Wixom C, Chadwick AE et aL Pulmona ry intra
52 :553-4, alveo lar siderophages in SIDS and suffocation: a San Diego
215 Hanzlick R, Pu lmon ary hemorrhage in deceased infants: SID S/SU DC project report, Pediatr Deliel Pathol200G;
ba selin e d ata for fu rther study of infant mo rtali ty, 9:103-J4.
Am) Forensic Med Path 01 2001 ; 22: 188-92, 231 Monta na E, Etzel RA, Allan T et aL En vironmenta l risk
216 Coffin CM , Schechtman K, Col e FS, Dehner LP, Neonata l and factors associ ated with pe diatric idiop athic pulmon ary
in fantil e pulmo na ry hemorrhage: an autopsy study with hemorrhage and hemosid eros is in a Cleveland community.
clinica l co rrelation , Pediatr Pathol 1993 ; 13:583-9, Pediatrics 1997 ; 99:3 1-8,
21 7 Golde DW, Drew Wl, KJei n HZ et a l. Occult pu lmona ry 232 Pappas fVlD , Sa rn a ik AP, Meert KL et aL Idiopathic
haemorrh age in leukae mia, BM) 1975; 2: J 66 -8, pulmonary hemorrhage in infa ncy. Clinical features and
218 Keeling JW, Maxwell RS, Busuttil A. Haem osideri n in infant management with high frequency venti lation, Chest 1996;
lun gs, ) Pa thol 1998 ; 184:29A, 110 :553 -5.
219 YlIkawa N, Carter N,RlItty G, Green MA. Intra-alveo lar 233 Cutz E, Perrin DG, Vujani c GM, Ackerly C. Idiopathic
haemorrh age in sudden infant death syndrom e : a cau se fo r pulmonary haemos iderosis (IPH) presenting as sudd en
concern?) Ciin Pathol 1999 ; 52:581-7, unex pected de ath SUD in infancy (Ab stract), Anales
220 Berry PJ, Lee J, Alveolar s id erop hages a nd the sudd en infant Espanoles de Pediatria 1999; 92(Suppl.):52,
death synd rome,) Pathol 1998 ; 184:27A. 234 Mu ir R, Niven J SF. The loca l form ation of bl ood pigments,
22 1 Stewart S, Fawcett J. Interstitia l haemosid erin in the lungs of ) Pa rhol Bacteriol1935; 41:183-97.
sudden infan t death sy ndrome: a histol ogica l hallmark of 235 Sherman JM , Winnie G, Thomassen MJ et al. Time course of
'near-m iss' episodes? ) Patlwl 1985 ; 145 :53 -8, hemos iderin production a nd clearance by huma n pulmonary
222 Byard RW, Stewart WA, Tel fer S, Beal SM, Assessment of macrophages, Chest 1984 ; 86:409- 1I.
pulmon ary and int ra thymic hemosiderin deposition in 236 Este rley JR, Oppe nheimer EH. Massive pulmonary
sudd en infant death sy ndrom e, Ped iatr Pathol Lab Med haemorrhage in the newbo rn, Pathol og ic considerat ions.
1997 ; 17:275- 82. ) Pediatr 1966 ; 69:3-11.
223 Becroft DMO, Thompso n JMD, Mi tchell EA. Pulmonary 237 Rushton DI. Liver and gallbl adder. In Kee lin g.JW (ed.) Fetal
interstiti a l haemosiderin in in fa ncy: a common con seq uence And Neona tal Pathology, 3 rd edn, Lond on: Springer-Verlag,
of normal labour. Pedia tr DelJel PatllOl 2005; 8 :448- 52, 2001, pp, 409-39,
224 Becroft DM, Lockett BK, Intra-a lveo lar pulmona ry 238 McCann J , Reay 0 , Siebert J et al. Postmortem perian al
sid erophages in sudden infant death: a marker for previous findings in children, Am ) Foren sic lVled Path ol 1996;
impose d suffocation, Pathology 1997; 29 :60-3, 17:289-98 ,
225 Milroy CM. Munchausen syndro me by proxy and in tra 239 Case ME, Graham MA, Handy TC et al. The National
al veolar haemosid erin. Int) Legal Med 1999; 1112 : Association of Medical Examiners Ad Hoc Committee on
309-12. Shaken Baby Synd rome, Position paper on fatal abusive
226 Dorandeu A, Peri e G, Jouan H et al. Histolog ical head injuries in infan ts and young children, Am ) Foren sic
demonstration of haemosiderin deposits in lungs and liver Med Path o12001; 22:112-22,
from victims of chronic physica l child ab use. Int) Legal 240 Bodian M, Hes lop B, Sudde n infan t death syndrome , 1n
Med 1999; 11 2:280-6 , Siim J-C (ed,) Proceedings of the Eighth 11Iternational
227 Fagan DG, Ha emosiderin in pulmonary macrophages from Co ngress of Paediatrics, Basel, 1956. Copenhagen:
SUD , SIDS and deaths in an fVTVA.) Patho11997; 182:49A. Williams 8: Wilkins, 1960, p. 91.
I CHAPTER 12 I
SUDDEN NATURAL DEATH IN
INFANTS AND CHILDREN
Dick Variend
INTRODUCTION
heart disease), whereas deaths associated with asthma ,
epilepsy and diabetes mellitus are more likely to affect the
Sudden death refers to the instantaneous death of an appar ado lescent years.
ently healthy person, but in practice the definition is The terminal event may be non-existent, non-specific or
extended to include all deaths within 24 hours from the onset insignificant. Some patients (or their carers) may not fully
of symptoms. The period between the onset of symptoms and perceive the severity of the symptoms that precede death,
the time of death is called the 'terminal event'. The term 'sud and this may delay summoning medical assistance. A lack of
den' generally describes the rapidity of the death and the term self-perception has particular relevance in conditions such
'unexpected' describes the surprising nature of the death . as asthma and hypoglycaemia. Symptoms may be modified
There is a wide range in the degree to which death may by medic atio n. Th e underlying disease mayor may not be
be expected. Some conditions, most notably cardiovascular symp tom atic. vVhen symptoms are present the terminal
in origin, are more prone to cause sudden death, and preced event can be variable in duration, depending on the under
ing knowledge of t heir existence tends to redu ce the level of lying disease. Those dyi ng fro m hypertrophic cardiomyopa
surprise. By their nature, sudden deaths tend to occur at thy or congenital aortic stenosis can be expected to have a
home, in the ambulance or soon after arrival at hospital. As short terminal event. It might be argued that distinction
might be expected, sudden deaths have enormous impact should be made between sudden death in apparently healthy
on the family, the community and medical attendants. They subjects and in those with recognized preceding illness
are relatively uncommon occurrences but, when they occur, (e.g. achondroplasia, cardiovascular disease). It is debatable,
tend to foster suspicion and, in the United Kingdom, are however, whether the latte r group should be included in the
genera lly dealt with by the coroner or procurator fiscal. definition of sudden death.
Many causes of sudden death are age related. Explained The inability of the very young to adequately communi
sudden deaths in the first year of life are more likely to be due cate symptoms may partly account for their increased vul
to an infection or congenital abnormality (e.g. congenital nerability to sudden death. The discrepancy sometimes
226 I Sudden natural death in infants and ch ildren
encountered between the account given and the gravity of may, for instance, vary in their interpretation of the severity
the post-mortem findings may be explained by failure of the and/or exten t of certain diseases (e.g. pneumonia or hydro
patient (or attend ant) to fully appreciate the seriousness of cephalus) and this may have a sig nificant bearing in dete r
symptoms . Sudden death is often accompanied by emotional mining the cause the death in any particular individual.
turmoil in the fam ily, and this may affect fam ily members' Such t1ndings, on the other hand, may be regarded as coin
ability to give a clear account of events leading to death. cidental.
Relevant information may come to light only after the Death ma y also be accelerated when infection affects a
autopsy has been completed. On the other hand, family pa tient with congenital or acqu ired immune deficiency or
members (or med ical attendants) may accurately perceive when the disease is caused by a part icularly vilUlent infec
symptoms but misinterpret their significance. Fo r instance, t ive agent.
the parents of a young child may mistake melaena for diar
rhoea and not feel the need to ask for urgent medical advice.
Severe haemorrhage may thus go unnoticed. CARDIOVASCULAR CAUSES OF SUDDEN DEATH
A history of chronic disease (e.g. diabetes mellitus, aller
gic asthma and epilepsy), w hile clearly very useful, does not Sudden cardiac death in children is far less frequent than in
always clarify the mechanism of sudden death. The cause of the adult population, and condi t ions that predispose to such
death in such a n individual as determined at post-mortem deaths in the young are quite different from those preva lent
may be entirely unrelated to the clinically diag nosed entity. in later years. The preva lence of cardiovascular disease as a
Sudden death is well known to be associated with certain cause of sudden death is difficult to determine because of
syndromic entities (e.g. achondroplasia, La rsen 's syndrome), variation in definition of sudde n death and inconsistency in
and knowledge of the relevant pathop hysiology may point the examination of the conduction system of the heart.
the pathologist to the system affected. Deaths occurring In the series of Lambert et al,] the following heart con
during exercise are more likely to be associated with the ditions accounted for 52 per cent of the cases of sudd en
cardiovascular system and , in such cases, a positive family death: congenital aortic ste nosis, Eisenmenger's syndrome,
history or evidence of preceding cardiac dys rhythmia may cya notic congenital heart disease with pulmonary stenosis
be helpful. and hy pel1rophic obstructive ca rdiomyo pathy. Arrhythmia,
The cause of sudden death at the time of autopsy may be hypo xia and coronary insufficiency are recog nized modes
obvious to the naked eye or on microscopic analysis; those of sudden card iac death but, in a substantial proporti o n of
without discernible findings require further diligent search the cases, the mechanism is ill-defined. I
or special investigations. After available investigation has
been exhausted, those without morphological substrate are
suitably referred to as 'physiolo gical deaths' which theo ret Cardiac Malformation
ically may cover causes such as cardi ac arrhythmia, con
vulsion, laly ngeal spasm or unstable respiratory control. Unrecog nized congenital ca rdiac malformation is an impor
They cover a wide age spa n and may affect the velY yo un g tant cause of morta li ty in the first year of life. 2 Of 185 cases
or o ld er child. of infant death from congenital heart disease investigated,
Medical intervention frequen tly prolongs life following 70 per cent had previously diagnosed heart disease; in the
a sudden loss of consciousness or cardiovascular collapse, remaining 30 per cent, the heart disease was unsuspected or
with death following a period of intensive care. The circu unconfirmed. Of those with unsuspected heart disease about
latory collapse and the medical intervention that foUow one-half had other severe abnormalities, mainly related to
often produce changes at post-mortem that are likely to trisomy 13, 18 and 21. Two babies with severe aortic steno
mask the picture produced by the initial event. Separation sis (both aged 2 months) and interlUption of the aortic arch
of such ischaemic or reperfusion injuries from changes (aged 6 days) died sud den ly at home. The authors speculated
directly caused by the initial event is important in deter that 200 babi es died each yea r in the UK from unsuspected
mining the underly ing cause. The results of biochemica l congenital heart disease. Of these, one-half of the heart
and enzymological studies may be difficult to interpret in defects were unassociated with other malformations.
such 'lingering' deaths. These cases are generally in cl uded After infancy, death from congenital heart disease is more
in the definition of sudden death. likely to be due to abnormalities of the coronary arteries,
The distinction between explained or unexplained sudden post-operative congenital heart disease, aortic valve stenosis,
death often depends on the thoroughness of the autopsy conduction system abno rmalities and dissection of the aorta
investigation, the experience of the pathologist and avail complicating Marfan's syndrome, aortic isthmic coarctation
ab ility of laboratory resources. When facilities are ava il ab le or isolated bicuspid aortic valve. 3 The mechanisms of death in
for t he investigation of genetic metabolic disease, the num this group include co ronary thromboemboli c phenomena,4
ber of explained sudden deaths is more likely to increase. In fatal cardiac alThyth mias and severe pulmonary vascular dis
some cases the relevance and significance of certain lesions ease.] Post-operative congenital heart disease is an imp0l1ant
found at autopsy may be difficult to determine. Pathologists cause of cardiac dysrhythmia leading to sudden death;5 the
Cardiovascular causes of sudden death I 227
defects mostly affected are tetralogy of Fallot, tra nsposition deaths occur in winter months.17 Of 207 cases of sudden
of the great arteries and double-outle t right ventricle. An death among individuals aged between 1 and 21 years,
older age at the time of operation appears to be a significant Neuspiel and Kuller 20 found myocarditiS to be the predomi
risk factor for sudden death. nant cause of cardiac death. Human myocarditis has been
Of 20 patients reported by Steinberger et al 6 who died associated with a number of viruses, most commonly of the
suddenly at the age of less than 12 months, 13 (65 per cent) coxsa ckie group.21 Presenting symptoms are related to the
had a cardiac abnormality, 10 of which were anomalies of age of the patient, and older subjects may complain of chest
the coronary aliery and, of these, 5 were initially consid pain, but symptoms are more often non-specific, especially in
ered to be sudden infant death syndrome. infants.22 Myocarditis may also present with cardiac failure.
Congenital aortic stenosis is a rare but well-recognized Pathological findin gs include cardiac dilatation, mottling
cause of sudden death,l and preceding symptoms such as of the myocard ium and variable opacification of the endo
syncope, fatigue, shortness of breath a nd dyspnoea on cardium. There is usu ally a diffuse interstitial mononuclear
exertion are often present. Dea ths tend to predominate in inflammatory infiltrate with a predominance of lymphocytes
the second decade of life and are often precipitated by on microscopic examination (Fig. 12.1).12,16 Interstitial
exertional exercise. B oedema and scattered foci of necrotic muscle fibres are usu
Intimal tears and aortic dissection leading to intraperi ally present. Both ve ntricles and atria may be involved. Viral
cardi al rupture and cardiac tamponade is a recognized com inclusions are usually absent. The lesion may be foc al or dif
plicat ion of undiagnosed aortic coarctation.] Patients with fuse and may show a predilection for the endocardium or
Eisenmenger's syndrome (pulmonary obstructive vascular subepicardial region. Specialized areas such as the conduc
disease secondary to communication between the systemic tion system may be involved. A subepicardial distribution is
and pulmonary circulations) have a moderately high risk of often ass ociated with pericarditis. The heart weight is often
sudden death, with many of the deaths occurring in the sec mildly to moderately increased, suggesting a latent phase of
ond decade. I myocarditis despite a short clinical history. 12 However, when
A ventricular septal defect is a common cardiac anomaly, involvement is confined to areas such as the conduction sys
and the releva nce of its discovery at autopsy in the sudden tem , the heart weight may be normal. 19
death of an infant or young child is often difficult to
assess. 9.10 Marked cardiac hypertrophy, pleural and intraperi
toneal effusions , pulmonary oedema and 'heart failure' cells
in the lungs are features of decompensation and implicate
the septal defe ct in the cause of death. 9 A multinational
study of 186 cases attributed sudden death to ventricular
septal defect in two cases (one per cent), suggesting its rarity
as a cause. I A ventricular septal defect may also form the
basis of Eisenmenger's syndrome, resulting in sudden death
much later in childhood or in the second decade of life. I
Williams' syndrome is a malformation complex charac
terized by typical facies, mental retardation, mild growth
deficiency and cardiovascular disease. Patients are at
increased risk of sudden death from cardiac deformities
that comprise supravalvar aorti c stenosis, bilateral outflow
tract obstruction and coronary al1elY stenosis . Myocardial
ischaemia, with or without cardiac arrhythmia, from coro
nary aliery stenosis is the most likely mechanism of
death, II but severe bilateral outflow obstruction may also
cause sudden de ath . Abnormal elastin is thought to be
responsible for the cardiovascular disease, and the condi
tion is linked to the elastin gene on chromosome seven. ll
Myocarditis
The heart may be specifically affected ('isolated' Over 50 per cent of cases are inherited as an autosomal
myocarditis) or cardiac involvement may be part of a gen dominant trait (familial hypertrophic cardiomyop athy).33,35
eralized involvement ('incidental' myocarditis). 12, 17 A prob The disease may commence in infancy, and death has been
lem often faced by the pathologist, and for which there is recorded in infants as young as 1 year of age. 3 Studies in
no easy answer, is how many foci of inflammatory cells in affected families, however, do not show Signifi cant num
the myocardium are sufficient to cause death. 16 Noren bers of infant deaths.32 Genes on five loci on separate chro
et al 17 demonstrated the presence of viral myocarditis as a mosomes are now known to be responsible for the familial
coincidental finding in two children who suffered violent disease.32 Of the three genes that have been identified, the
deaths. This calls for caution in attributing death to viral best characterized so far is the one that encodes for ~
myocarditis in children who die unexpectedly. heavy chain myosin, which is found in abou t 50 per cent of
It is important to save samples of heart muscle in aJl cases affected families. At least 30 different point mutations are
of sudden death for virological investigatio n. Diagnostic kno wn for this gene.
methods advocated for clinical investigation are also appli Characteristic of this condition is an increase in ventricu
cable at autopsy. These include serology for specific viral lar muscle mass, with histology marked by myocyte disarray
antibody, viral culture using tissue or fluid and morphologi (disorganization), broad and misshap en individual myocytes
cal determination of the vil1Js in infected cells; the last may and circular alignment of myocytes around central foci of
be au gmented by specific probes for viral antigens or nucleic connective ti ssue 32 (Fig. 12,2). Interstitial and replacement
acid sequences using in situ techniques for probe visualiza myocardial fibrosis and acute or subacute myocardial necro
tion. 22 Molecular techniques such as polymerase chain reac sis may be present. 3 The histological changes are maximal in
tion (PCR) have proved valuable in identifying viral genome areas of macroscopically thickened wall.
in formalin-fixed paraffin-embedded tissue.2J - 25 The exact mechanism of sudden death in this condition is
Contraction band necrosis in 'neuro genic cardiomyopa still unclear, but theories include inappropriate circulatory
thy' and catecholamine-induced cardiomyopathy may cause reflexes leading to haemodynamic collapse,36 arrhythmia an d
minor inflammatory changes that should not be interpreted conduction abnormalities. 37 Cases with the histological char
as myocarditis. 26 ,27 Myocarditis can also occur with any bac acteristics of HCM may occur in the absence of hypertro
terial infection (meningococcus, diphtheria, Staphylococcus phy,16,36 Consequently, extensive histological examination
aureus, pneumococcus, gonococcus and Haemophilus
injlurnzae).28 With the notable exception of diphtheria, car
diac involvement in these cases is likely to be incidental.
In acute rheumatic fever, pancarditis occurs in 50-75 per
cent of children and acute rheumatic carditis may present as
sudden death.29 Sudden death in a 6-year-old girl with acute
rheu matic carditis complicating by thromboembolic occlu
sion of the left anterior descending coronary artery has been
described. 4 Acute myocard itis has also been reported in
patients with dermatomyositis 30 and Kawasaki's disease.31
Cardiomyopathy
may be required in cases of sudden death to confinn or of the endocardium is seen macroscopically. Twelve cases
exclude the di agnosis. (seven per cent) of endocardial fibroelastosi s were included
A spectrum of disorders is responsible for other cases of in the 186 cases of sudden cardiac death in children sur
HCM that are increasingly attributed to metabolic, genetic veyed by Lambert et al. 1 Secondary endocardial fibroelasto
and molecular disease (e.g. fatty acid oxidation defects, sis may be superimposed on cardiac malformation or
mitochondrial disease, glycogen storage disease types 2 and 3, myocardial disease, and sudden death may be related to the
Gaucher's disease, GM 1 and GM2 ga ngliosidosis, sialidosis underlying disease. 4J
and mannosidosis). Cardiomyopathy occurs in about 25 per
cent of patients with Noonan's syndrome J5 and the ARRHYTHMOGENIC RIGHT VENTRICULAR DYSPLASIA
myocardium in affected patients is histologically indistin (RIGHT VENTRICULAR CARD IOMYOPATHY)
guishable from familial HCM.J8 Other associations include
Friedreich's ataxia, Turner's syndrome and some disord ers of Arrthythmogenic right ventricular dysplasia (ARVD) or car
neural crest tissue. 16 Catecholamine-induced cardiomyop a diomyopathy is a diso rd er of heart muscle 44 ,45 of unknown
thy has been reported in patients with phaeochromocytoma. prevalence. 46 ,47 The disease is often familial (about 30 per
Neonatal cardiomyopathy may follow poorly controlled cent of cases) with an autosomal dominant inheritance 46
maternal diabetes. J9 Affected infants are usu ally macrosomic and appears to be related to abnormal developmen t of the
reflecting the growth hormone effect of insulin. An underly myocardium. 47 A gene for ARVD has been mapped to chro
ing anatomica l malformation of the heart must be excluded. mosome 3p23. 48 Sudden death or a rrhythmia of ri ght ven
Ventricular hypertrophy may persist for up to 2 years when tricular origin is the usua l mode of presentation. Symptoms
such patients are followed by echocardiography.J2 Some ra rely appear before 20 years of age and death often occurs
vacuolar and hydropic changes of the myocardium are seen during exertion. 44 ,45 The rate of sudden death is about one
microscopically, but the myocardial fibre disarray and per cent of cases per year. 47
inflammatory infiltration that characterize classic HCM do The usual macroscopic appearance at autopsy is marked
not occur. Cardiac dilatation, which is also sometimes seen, dilatation of the right ventricle with replacement of its free
may be secondary to hypo glycaemia and acidosis?9 waJl by adipose tissue;49 microscopy confinns the presence of
Sudden death can occur from severe left ventricular adipose tissue or fibrofatty tissue (Fig. 12.3). Inflammatory
hypertrop hy from any cause.J2
DI LATED CARDIOMYOPATHY
ENDOCARDIAL FIBROELASTOSIS
cells are seen in about 25 per cent of cases, raising the possi
bility of an infectious or a genetical ly determined immune
process in some of the cases. Heart weight is normal or only
moderately increased.45 Similar histological features are
rarely seen to involve the left ventricle.47 As some degree of
adipose tissue is often seen in the free wall of normal right
ventricle, the amount of adipose tissue required for the diag
nosis of ARVD is sometimes unclear. 49 ,50
ARVD and Uhl 's anoma ly are now considered to be two
distinct morphological entities. 49 Uhl 's anomaly refers to
absence of the right ventricu lar parietal myocardium, the
endocardium and epicardium being directly apposed; the
condition usually presents in neonates and infants with
congestive cardiac failure . (a) (b) (c)
8r
trunk (Bland-White-Garland syndrome) is a rare malforma
tion. The relative low pressure of the pulmonary artery is
LCA
inadequate for perfusing the left ventricular myocardium and
most affected subjects die in infancy from cardiac failure. 29
RCA
the subacute and heali n g stages. Sudden unexpected death may occur in isolation or be associated with other abnor
may occur many years later, following recanalization of malities, for example Epstein 's anomaly, rhabdomyoma of
thrombosed coronary arteries,lo and sudden death has been the heart. These abnormal pathways produce early stimula
reported in a fit athlete during exercise with evidence of pre tion of the ventricle and are associated with a variety of
ceding Kawasaki disease. JI electrocardiographic patterns. Many patients remain asymp
Overlapping features, including the clinical demonstra tomatic however, but sudden death in this group is well
tion of coronary aneurysms, have been reported bet"feen recognized.
group A streptococcal infection and Kawasaki's disease. 66 Heart block may be congenital or result from acquired
causes. Marked degrees of atrioventricular (AV) block may
produce Stokes-Adams attacks and are rarely caused by
Abnormalities of Cardiac Conduction endodermal heterotopia of the AV node (mesothelioma of the
AV node). Some congenital abnormalities of the conduction
While myocardial disease, such as hypertrophic cardio system (e.g. simple absence of the short segment of the right
myopathy, myocarditis, histiocytoid cardiomyopathy and bundle branch or the penetrating portion of the main bundle
tumours, may cause sudden death consequent upon of His) are stable and do not seem to progress. IS
involvement of the conduction pathway, abnormalities of Babies born to mothers with systemic lupus erythemato
the cond uction tract may coexist with certain types of con sus are at considerable risk of congenital AV block but the
genital cardiac malformations or the conduction tissue may prognosis is rem arkab ly good in such cases, and those
be damaged during their operative repair. 29 Fibromuscular affected may not develop syncopal attacks until much later
hyperplasia of the sinoatrial nodal an d atriovent ricul a r in life. Death among infan ts with preceding congenital
nodal arte ri es may cause paediatric sudden death from car heart block or those w h o die in association with maternal
diac arrhythmia. 67 connective tissue disease should be tested for anti-La (SS-B),
On the other hand, fatal arrhythmias may result from anti-Ro (SS-A) and even anti - UIRNP antibodies. 29
intrinsic defects of the conduction tissue that are not obvi Heart block has been described w ith Kearns-Sayre
ous to the naked eye. 30 .6B For precise diagnostic categoriza syndrome, Kartagener's syndrome and certain X-linked
tion, such cases require formal examination involving myopathies. 10
extensive serial sectioning. In practice, this is time-consuming
and associated with such a low positive yield that it does not
gene rally form part of the routine examinatio n of the heart Occlusive Disease of the Coronary Arteries
in cases of sudden death. Conduction defects may thus be
easily overlooked and many such deaths are simply referred Infantile arterial calcification is a rare disease mainly affect
to as showing no structural cardiac disease. 41 ing infants,71 with reports of siblings affected. Low levels of
The long QT syndrome (LQTS) is significantly associated plasma cell membrane glycoprotein-l nucleoside triphos
with ventricular fibrillation and the risk of sudden death. phate pyrophosphohydrolase have been demonstrated in an
The condition can be inherited as an autosomal dominant affected infant. 72 The same group subsequently demon
disorder (Romano-Ward syndrome),69 with ab norm ali ties strated mutations in the EN??] gene 7J Alieries throughout
at six different loci identified.1° A recessively inherited the body are involved, with the exception of the brain and
syndrome w ith deafness is also recognized .7o Vigorous spinal cord. Of the 62 cases reviewed by Moran (1975),71 10
physical exertion often precedes vent ricula r fibrillation. In presented as sudden death, preceded by respiratory distress
these cases there is no discernible abno rmality on histolog in 5. The age of death ranged from 2 days to 28 months, with
ica l examination of the conduction pathway. IS Risk factors 85 per cent of affected infants dying within 6 months.
for sudd en death in such indi vidua ls include a history of Clinical diagnosis is possible with rad iological study. The
syncope, congenital deafness and a fami ly history of sud ECG changes, when available, are mostly compatible with
den cardiac deathS myocardial ischaemia. Cardiac enlargement is common and
Preceding symptoms th at suggest a conduction defect the myocardium is often infarcted. Intra-uterine ultrasono
include syncope and dizziness. A preceding abnormal elec graphic diagnosis has been made in familial cases. 74
trocardiogram (ECG) recordin g or a family history of con Microscopically, there are calc ific deposits with fragmen
duction defect is valuable in these cases, and likely to tation of the internal elastic membrane of arteries along wit h
ex plain an arrhythmic cardiac arrest. In their absence, the variable degrees of fibrointimal proliferation and luminal
significance of any histological cha nges of the conduction narrowing. 71 ·75 An inflammatolY reaction is typically absent.
tract is merely speCUlative. Opportunity for clinical diagno An alieriopathy with features similar to those found in idio
sis in these cases is precluded by the intervention of sudden pathic infantile arterial calcification has been reported in
death of the patient. children with acquired immun e deficiency syndrome
Accessory pathways of th e conduction system may (AlDS);76 one of the cases described by these authors showed
cause re-excitation and re-entry lea ding to fatal arrhyth myocardial infarction associated with thrombus formation
mias. The Wolff-Parkinson-White pre-excitation syndrome in a coronary aneurysm.
Cardiovascular causes of sudden death I 233
FIBROMA
The seco nd most common cardiac tumo ur, fibroma of
the heal1, also ca uses death secondary to outflow tract
obstruction or arrhythmia. A well -circumscribed grey-white
mass betvveen 3 and 10 cm in diameter is_ the usual gross
appearance (Fig. 12.8), cysts may occur. The tumour is most
often located in the interventricular septum. Microscopy
shows fibroblast-like cells in a colla genous matrix, reminis
cent of fibrom atoses of the soft tissues (Fig. 12.9).B6,87 The
tumour may occur as a component of Gorlin's syndrome.
MYXOMA
- - "'. ."
Figure 12.9 Histology of the cardiac fibrom a shows fibrobl ast
like cells in a coll ag enous matrix closely resemb li ng fibrom atosis.
Myxomas are rare in infants and children. s7 Most occur as
sporadic lesions, arise withi n the left atrium and may reach
OTHER CARDIAC TUMOURS
up to 10 cm in diameter. Su dden death is related to dis
turbed cardi ac haemodynamics or systemic emb olization of Cardiac haeman gioma is a rare cause of sudden death in
tumour fragments. Myxomas rarely involve the right side child re n. A previously asympto matic 13 -year-old girl who
of the heart, where they are capable of causing massive was found unresponsive in bed was reported by Krous
pulmonary embo lism. Bs et a\. B9 Autopsy showed haemangiomatous invo lvement of
X-linked hypohidrotic (anhidrotic) ectodermal dysplasia I 235
,,
(histiocytoid cardiomyopathy), 12 cases of cardiac fibroma, #
j
(
I '
•,
!
f J
\.
fI
.."
.1. I • t -
\' ' ,
.!.' .I I -~
Thromboembolism I .
I!
il
~
Pulmonary thromboembolism is an unusual cause of sudden
death in children. Predisposing factors include recent sur
l ,
gery, congenital heart disease, indwelling venous catheters, , I
,.
J
£ ~ , t ..
sepsis, arteriovenous malformation and occult malignancy.91 .
.,
~
Patients with Marfan's syndrome display a characteristic The condition is characterized by an absence or diminution of
habitus: tall stature, arachnodactyly, bilateral ectopia lentis, eccrine sweat glands, eczematoid rash, thin sparse hair,
high arch palate, dolichostenomelia and striae distensae. oligodontia and peg-shaped teeth. The incidence is estimated
Linkage analysis has shown that the fibrillin gene on chro to be in 1 per 100000 live births.94 The condition carries a sub
mosome 15 is associated with Marfan's syndrome. 8 Up to stantial mortality and morbidity, and death occurs in about
236 I Sudden natural death in infants and children
30 per cent of cases. 95 There is an increased risk of chest infec Sudden unexpected death m ay result fro m intracranial
tion and atopic disease. Because of the deficient number of haemorrhage secondary to a generalized bleeding disor
eccrine sweat glands, patients are unable to perspire and con der lOI or the haemorrhage may be related to intracranial
sequently develop fatal heat intolerance 94 Mucus production pathology (Fig. 12. 12). Contrasting with subdural or
is also deficient in the respiratory and gastrointestinal tracts. extradural haemorrhage, sudden unexpected death follow
An inexplicable improvement in the patient's general condi ing parenchymal haemorrhage is more likely to be associ
tion occurs with increasing age. ated with natural disease . Arteriovascular malformations
(Fig. 12.13) and aneurysms a re important sources of cata
strophic bleeding. 2o ,JOI Intra-crani al a neurysms are found
Achondroplasia with uncorrected aortic coarctation, familial multiorgan
cystic disease or autosomal dominant polycystic kidney
Achondropl as ia is the most common type of osteochon disease (ADPKD).
drodysplasia; it is inherited as an autosomal dominant tra it. While fatal subarachnoid haemorrhage in ADPKD usually
Recent studies suggest that sleep apnoea is common among presents in adulthood,I02 infants and children are rarely
affected individuals,96 involving both obstructive and cen affected. A family history is extremely helpful in making a
tral mechanisms. Compressive myelopathy and dysfunction diagnosis. The cerebral anelllysm may be masked or destroyed
at the cervicomedullary junction is caused by a small fora by the haemorrhage and, consequently, may be difficult to
men magnum, and sudden infant and childhood deaths find at post-mortem examination. Intracranial aneurysms in
have been reported. 97 Macroscopic or microsco pic necrosis childhood have been reported to occur as a complication of
of the spinal tissue at the level of the foramen magnum may renal hypertension . 103
be seen (Fi g. 12.11). The intracranial pathology mayo r may not be clinically
apparent prior to sudden death.l04 These authors reported
10 deaths from intracrania l haemorrhage, secondary to
Tuberous Sclerosis Complex tumours (four cases) , berry a neurysm (one case) and vascu
lar malformations (five cases); the presence of four vascu
Tuberous sclerosis is a dominantly inherited condition lar malformations was assumed at autopsy as their positive
characterized by genetic heterogeneity. The birth incidence identification was difficult owing to the marked tissue
is about 1 in 6000. More than 70 per cent of cases will be destruction. Such haemorrhages may be due to micro
new mutations. Early skin manifestations are hypomelanic angiomas. lOl An additional cause of intracranial bleed in
macules and shagreen patch. Subungual fibromas and older female patients is eclampsia.
facial angiofibromas occur later. Cardi ac rhabdomyomas, Four cases of intracerebral haemorrhage reported by Liv
giant-cell astrocytomas and renal disease are other patho ingston and Brown 101 compbcated a pre-existent coagulopa
logical lesions .93 The polycystic renal disease of tuberous thy. In two cases this was thrombocytopenia (associated with
sclerosis usually affects infants and young children, while aplastic anaemia and acute lymphoblastic leukaemia) ; one
the angiomyolipomas tend to occur at or after puberty. At child had haemophilia and one case was thought to be a late
least 80 per cent of children presenting with cardiac rhab manifestation of haemorrhagic disease of the newborn.
domyomas will have tuberous sclerosis. 93 The incidence of Minor trauma was a factor in the case of the haemophilia.
Intracranial haemorrhage, neoplasms and malformations I 2 37
J8,...
Figure 12.12 Cut surface of the cerebellum shows haemorrhage
into a tumour which wa s subsequently demonstrated to be low
grade astrocytoma. The patient was a 4-year-old girl who had
apparently been in good health and had died suddenly. Figure 12.14 The tip of the intraventricular part of a
ventriculoperitoneal shunt that had become ensnared in choroid
plexus. causing obstruction to cerebrospina l fluid egress in a
young child with shunt- treated hydrocephalus.
be enhanced by efforts at resuscitation . 1J7 Other findings Sources of intracranial ha emorrhage have been dis
include increased tracheobronchi al secretions, laryngeal cussed earlier. One study l5 found 13 out of 169 sudden
oedema and pulmonary emphysema. An autopsy diagnosis deaths to be primarily due to haemorrhage; these were
of acute anaphylaxis cannot be made on morphological mainly intra cra nial haemorrhage and usually secondary to
grounds alone and appropriate clinical information is an arteriovenous malformation. One death followed rup
esse ntial. IJ6 In cases of fatal reaction to insect stings, the ture of a tubal pregnancy.
site of the sting may be found. The aetiology of pulmonary haemorrhage has been out
lined by Cutz l47 and, broadly, may be separated into pri
mary idiopathic pulmonary haemosiderosis and second ary
SICKLE CELL DISEASE pulmonary ha emorrhage. The latter includes immunolog i
cal causes and vascular, infectious and bleeding disorders.
Sickle cell disease is more common in the black population Idiopathic pulmon ary haemosiderosis has been recognized
and the clinical course is generally more severe in homozy as a cause of sudden death.lo
gous than in heterozygo us individuals. Most deaths occur Gastrointestinal sources of haemorrhage include oesophageal
in patients with haemoglobin SS, but young subjects with varices, peptic ulceration and vascular malformations.
haemoglobin SC are also at risk.143 Polymerization of
deoxygenated sickle haemoglobin produces sickling of red
blood cells leading to veno-occlusive crises. Predisposing RESPIRATORY CAUSES OF SUDDEN DEATH
factors include dehydratio n, fever, acidosis and hypox
aem ia. Sudden death in infants and older children has been Abnormalities of the Trachea
repolted . 144 Generalized convulsions are also recognized.
Sudden death in infants with a sickle cell trait may be asso Tracheomalacia may be congenital or acquired. The con
ciated with anaesthesia. genital type is exceedingly rare,148 and cases have been
Infarction of the small bones of the hands and feet, reported in association with Larsen's syndrome, pulmonary
sickle cell dactylitis, aplastic and haemolytic clises and vascu lar sling, bronchopulmonary dysplasia and tracheo
splenic sequestration are recognized complications. Splenic oesophageal atresia. Tracheal collapse with life-threatening
sequestratio n crisis, the most severe complication, causes airflow obstruction is a recognized complication following
death from circulatory collapse due to marked pooling of repair of tracheo-oesophageal fistula. 149
blood within the spleen. Th is is particularly common in
infants under 2 years of age. 144 Acute infection is a com
mon precipitating factor. At autopsy the organs are pale
Acute Epiglottitis
except for the spleen, which is enlarged and engorged.
Sickled cells in capillaries may be visualized at microscopy,
but sickling as such does not necessarily imply that it has Acute ep iglottitis, an acute life-threatenin g condition, is
occurred ante-mortem. 144 Purulent meningitis, cerebral one of the most serious manifestations of infection with
infarction, cardiomegaly, splenic infarction and bone mar Haemophilus injluenzae type b. The clinical diagnostic cri
row hyperplasia are other auto psy findings. teria are a red swollen epiglottis (Fig. 12.16), inspiratory
In the USA the peak incidence of death was found to be stridor or difficulties with swallowing and pyrexia. 150 Chil
between one and three years and the major precipitating dren are most commonly affected, but the case fatality is
event was infection. 145 The causative agent most commonly lo w in all age groupS.1 50, 151 Trollfors et ali SO reported six
isolated was Streptococcus pneumoniae. Pneumococcal septi childhood deaths in their series of 485 children. Four died
caemia in childhood is observed less frequently because of at home or were dead on arrival in hospital. The other two
the widespread use of immunization and antibiotics. 143 The arrived in hospital deeply comatosed, With current immu
major cause of death in the second decade is a cerebrovascu ni zation schedules, the incidence of acute epiglottis in the
lar accident. Other causes of death include acute chest syn Western world may be expected to decrease sharp ly.
drome secondary to pUlmonary vasc ular occlusive disease,J46
a condition that tends to affect older subjects and may lead to
chronic lung disease and pulmonary hypertension. Retropharyngeal Abscess
Bacterial infection of the trachea is usually superimposed Bronchiolitis is an infection of the lower respiratory tract
on preceding viral infection. Staphylococcus aureus and caused by one of a number of different viruses. It is prima
Haemophilus influenzae are responsible for the majority of rilya clinical diagnosis 159 and the clinical picture is typified
242 I Sudden natural death in infants and children
deaths is uncertain, but proposed mechanisms involve auto life- threatenin g asthmatic episodes, hospital admissions
nomically related cardiac arrthymias l67 and ictal apnoea .J72 for as thm a in the year preceding death, poor access to
Poor seizure control and poor compliance with medical ca re, inadequate medica l ma nagement a nd psy
an tiepileptic therapy do not ap pear to be significant factors chological and psychosocial problems. IS} Previous stero id
in SUDEP. 173 A toxico logy screen should fonn pa rt of the treatment, by suppressing the ad ren a l axis, m ay lead to
au topsy proto col and the heart should be examined in detai l. adren al insufficiency and a pred isposition to unexp ected
A serum dmg level lower tha n the therapeutic ran ge do es de ath.
not necessari ly imply fa ilure of compliance, and sh ould not Recognized mechanisms of death are severe asphyxia
be in terpre ted by the pathol ogist as a factor in the cause of du e to airflow restriction IS4 or cardiac arrhythmia from
death. 174 Moreover, the relevance of post-mortem blood lev myocardial irri ta bility secondalY to excessive use of ,B2
els of anticonvulsants many hours after death is unclear, as an tag onists. Hypokalaemia, which may be partly expl ained
so me degrade more rapidly than others.1 6S The possib ility of by ~ 2 antagonists, can cause cardia c a rrhythmia or gene r
adverse cardiovascular effects of carbamazipine in SUDEP in ali zed muscle weakness, co ntributing to sudden dea th. 135
children cannot be ignored. 175 Life-threa tenin g eve nts are includ ed among the adverse
When evidence su ggestive of a seizu re is found, for effects of aminophylline. IS)
example a bitten tongue, limb injuries or an empty bladder, The hi gh mortali ty from asthma in t he 1960s was attrib
this sh ould be record ed. However, should this evidence be uted to the excessive use of press urized beta ago nist
absent, a seizure canno t be excluded. It is recommended aerosols, the so-called 'aeroso l hy pothesis'. ls6 Whilst the
that when children are fou n d wi th thei r head immersed in debate concerning the role of th ese aerosols con tinues, it is
water, de ath should be recorded as ep ilepsy- rel ated death, difficult to discount fa ilure to promptly and adequatel y
rather than SUDEP. treat the asthmatic attack in su ch cases. It has been ques
Card iac arrhythmias rarely give ris e to 'epilepsy',1 76. 177 ti oned whether lun g fun ction in such cases is ever normal
and death in such cases may be erron eously attributed . A shortly befor e the fatal attack JB2
prolonged QT interval may be responsible and au topsy Apart from abrupt airway narrowing and medication
examination is unlikely to r eveal any abnorm ali ty. related deaths, an imp ort ant factor may be u nsuspected
pulmonary pathology. lSI Among 13 unexpected asthma
deaths in subjects aged between 9 and 19 years, Kravis and
DEATHS FROM ACUTE ASTHMA Kolski lSI id entified a cause in five cases: pneumothoraces,
histiocytic bronchopneumonia, iso lati on of Kl ebsiella
Asthma is a common disease and, des pite advances in tbe pneumoniae, eosinophilic pn eumonia and foc al bronchio
treatment, there is still a hig b ra te of mortality and morbid litis obliterans.
ity amon g children. Deaths from as th ma are usually as a Macroscopic fi n din gs in acute deaths from asthma are
result of prolonged attacks tha t fail to respond to conve n bulky, hyperinflated lungs, surgical emph ysema and, rarely,
tional measures, status asthmaticus. Sudden unexp ected pneumothorax. Microsco py shows intraluminal bronchial
death is a lso recognized but occurs less freq uen tly.17B and brochiol a r mucinous plu gs, epithelial basement mem
Zach an d Karner l79 reported sudd en deaths in two gi rls, brane th ickening, peribronchial smooth muscle hypertro
one aged 14 years and one aged 9 years, receiving anti phy, mucus gland hyperpl as ia a nd submucosal cellular
asthmatic medication. In both cases the gi rls ' condition infiltration, often with large numbers of eosinophils. 17s
was regarded as moderately severe and well stab ilized. Asthmatics are also at increased risk of developing severe
These deaths might be explained by inappropriate percep or fa tal anap hyl ax is,14 1.142 and this possibility sh ould be
tion of the severity of the attack by the p atie nt. considered in the event of sudden death in a patient who
Of 30 childhood asthma-related deaths reviewed by Car suffers from asthma. Thi s is especially relevan t as acu te ana
swell,l so the mean age at death was 8.5 years; 19 died at phyla xis may present cl inically as respiratory distress. Mea
home or in transit to hospital and 11 died in hospita l. One surement of the semm tryptase level may assist in making
child died within one hour of the sta rt of the attack; three the distinction.
died less than 12 hours after the start and seven died after
more than 12 hours. Deaths occurred mainly at night or in
th e evening. Only 12 patients in this stud y we re known to DIABETES MELLITUS
be receiving prophylaxis.
Unexpected death in asthma patients is especia ll y pro Despi te modern treatment, in sulin-depe ndent diabetes mel
nounced in adolescent an d preadol escent years, and may litus (IDDM) in childhood still carries Significant morta l
be related to treatment non-use/abuse, which is ren owned ity.IB7·IBB Most deaths are attributable to metabolic
to occur at this age. lSI The increased frequ en cy of deaths complications, and those related to diabetic ke toa cidosis
from asthma at night or in the early morning has been a re most common, foll owed by hypog lycaemia. IB9 Sudden
lin ked to diurnal va ria tion in airflow limitation. ls2 Risk unexpected death is reported in cases of di abeti c keto ac i
factors for sudden death include the occurrence of previous dosis,190 and the a dol escent age group is mostly affec ted .
244 I Sudden natural death in infants and children
Cerebral oedema is an important comp li cation in such cases reveal a cause, the diagnosis is usually inferred from t he
and carries a poor prognosis. 191 After death , the blood glu clinical history.
cose is raised and the urine contains ketones, but ketones
may be absent in the aketotic form of diabetic coma.
Post-mortem determination of blood glucose is unreli GENETIC IVIETABOLIC DISORDERS
able. 190 The presence of glucose in the urine may also be
misleading as glucose may have been administered intra Factors that should prompt consideration of genetiC meta
venously. The site of blood sampling at autopsy has an bolic disease in cases of sudden death are parental consan
important influence on the blood glucose level. 192 Thus, guinity, maternal HELLP (haemolysis, elevated liver
samples taken from the right side of the heart often show enzymes, low platelets) syndrome, previo us Reye-like ill
sp uriously high gl ucose levels as a result of glycogenolysis , ness or infant deaths in the family (from known genetic
espec ially if death is accompanied by an 'alarm' reaction or metabolic disease or unex plained), a previous acute life
cardiac massage is performed . Sampling from the periph threatening even t (near-miss cot death), fasting or recur
eral vein, by contrast, often shows a glucose level that is rent hypo g lycae mia, neonatal hypotonia, dysmorphism,
spuriously low due to continuing glycolysis after death. en la rgement of the live r and/or spleen. A pale (fatty) liver
Glucose in the vitreous can be more usefully meas or hepatic fibrosis, cardiomegaly and severe brain oedema
ured, 190.192 but the level decreases rapidly after death. So are important markers of metabolic disease at autopsy. The
lon g as the sample has been taken peri mortem into a fluo possible metabolic causes of sudden death in the paediatric
ride tube, this w ill reflect blood glucose level s at that time group are wide-ranging. 199
and a normal glucose level should exclude hypoglycaemia
providing, of course, that the patient has not received glu
cose infusion prior to admission. 193 Mitochondrial Abnormalities
High levels of vitreous or peripheral venous glucose
probably do indicate uncontrolled diabetes, especially if These comprise defects of fatty acid oxidation (FAO) and
concentrations of ketone bodies in the blood or urine are oxidative phosphorylation (OXPHOS). A Reye-like illness or
also raised. Measurement of glycated haemoglobin has sudden death is exceptionally reported in association with
been suggested as a more reliable indicator. 194 Absence of defects of OXPHOS. In defects of OXPHOS abnormalities are
insulin in ante-mortem blood carries more weight than in likely to involve a single organ or several systems simulta
a post-mortem sample; by contrast, raised insulin values neously. A diagrammatic scheme of intramitrochondrial
in post-mortem blood are useful in excluding diabetic ~ -oxidation of fatty acids an d its interrelationship with the
ketoac idosis. 190 respiratory chain complex is shown in Fig. 12.19. 200
Histology of the pancreas may provide useful informa
tion providing the post-mortem interval is short and
FAm ACID OXIDATI ON DEFECTS
a utolysis is minimal. 195 Changes include acinar atrophy,
reduction in the number and size of Langerhans islets, Enzymes that have been associated with sudden death are
insulitis, distortion of islet architecture and histochemical shown in Table 12.2. All of these inborn errors are inherited
demonstration of a reduced number of insulin-producing B as an autosomal recessive trait. Very long-chain acyl coen
cells. 195 Glycogen may be increased in the renal tubular zyme A dehydroge nase (VLCAD) deficiency usually presents
epithelium (Armani-Epstein nephropathy), but this is not a in early infancy with hypoglycaemia, recurrent vomiting,
constant feature. Othe r histo log ical findings are microvesic liver dysfunction, cardiomegaly and a tendency to cardiac
ular fatty change of the liver and vacuolization of hepatic arrest.200.201 Presentation as sudden infant death is well rec
nuclei; these are, however, non-specific and may be found ognized.202.203 Compared with the other defects, VLCAD is
in other conditions. more likely to be associated with cardiomegaly, although
The 'dead in bed syndrome' constitutes a puzzling group sudden death and severe illness in the neonatal period have
among patients with IDDM. The subjects are without clini been reported. 204 Only a few patients with short-chain acyl
cal evidence of late complications. They are usually on CoA (SCAD) deficiency are recorded, 202 and its association
insulin and there is often a history of one or more noctur with sudden death is unclear. 203
nal hypoglycaemic attacks in the previo us 6 months. 196 Primary carnitine defici ency and carnitine palmitoyl
They are generally observed to be in good health on the transferase type II (CPT II) deficiency often present with
preceding day and are found undisturbed in bed the fol cardiomyopathy, whereas carnitine palmitoyltransferase
lowing morning. 196 ,197 Children and young adults account type I (CPT I) deficiency usually presents with a Reye-like
for some six per cent of all such deaths under the age of 40 illness, but occasionally cardiac arrhythmia occurs in the
years. The most likely cause is thou ght to be hypogly neonatal period. 205 Fat oxidation is impaired as the trans
caemia with associated events such as ca rdi ac a rrhyth fer of long-chain fatty acids across the mitochondrial
mia 198 or respiratory depression. 196 Hypoglycaemia is membrane is dependent on carnitine transporter and carni
difficult to confirm and, as the post-mortem often does not tine palmitoyltransferase. 202
Genetic metabolic disorders I 245
1
acyl-CoA
Ligase
Carnitine palmitavl
transferases I and /I
SCAD, MCAD
vLCAD
Enavl-CoA
hvdratase
-I
:lJ
3-hydroxyacyl-CoA : "
C
Z
NAD+ ~ g
3-Hvdroxvaevl-CoA o
z
dehvdrogenose ,:t>
NADH+H + 1 m
Z
N
-<
3-oxoacyl-CoA ':m
s
Thiolase
acyl(n_2j-CoA
Figure 12.19 Diagram to show the fatty acid 0-oxidation pathway and the link with the respiratory chain complex. (Modified from ref 200.)
Table 12.2 [3-0xidatian of fatty acid defects that have been lead to depleted glycogen reserves, initiating ketone body
associated with sudden death production and muscle fatty acid oxidation.
In the UK, medium-chain acyl CoA dehydrogenase
Very long-chain acyl CoA dehydrogenase deficiency (VLCAD)
(MCAD) deficiency is the most commonly encountered
Long-chain 3-hydroxyacyl CoA dehydrogenase deficiency (LCHAD)
metabolic disorder leading to sudden death in infants and
Mitochondrial trifunctional protein deficiency (TFPD)
children. 206 It is clinically the least severe of the i3-oxidation
Medium chain acyl CoA dehydrogenase deficiency (MCAD)
defects. 20I,202 According to the review by Touma and
Carnitine palmitoyltransferase type 1 deficiency (CPTl D)
Charpentier 207 the average age at presentation is 13 .5
Carnitine palmitoyltransferase type 2 deficiency (CPT2D)
months and the mean age of death is 18.5 months. Patients
Plasma membrane carnitine transporter deficiency (PMCTO)
often present clinically with hypoglycaemia, acute
Multiple acyl CoA dehydrogenase deficiency (MADD)
encephalopathy (mimicking Reye's syndrome), sudden death
Carnitine acylcarnitine translocase deficiency (CACTO)
and an acute life-threatening event. 201.202,208,209 One-quarter
of patients die with a Reye-like illness or experience sudden
infant death. A previous sibling death is often recorded.
A preceding viral prodrome is common.
Long-chain 3-hydroxyacyl CoA dehydrogenase (LCHAD) The gene for MCAD has been identified and sequenced;209
deficiency is associated with a high mortality in infancy.200 up to 85 per cent of subjects with MCAD deficiency are
Cardiomegaly may be severe and hypo ketotic hypogly homozygous for a single gene mutation of an A-to-G
caemia may be a presenting feature. When the fetus is nucleotide at position 985. A recent prospective surveillance of
affected by LCHAD, the later stages of pregnancy may be the prevalence of the condition in the UK210 found 1.3 cases per
complicated by the HELLP syndrome in the mother. 10 000 births. The precise mechanism of death in MCAD defi
Disorders of fatty oxidation tend to be clinically silent ciency is unclear, but hypoglycaernia, cardiac decompensation
in the absence of significant physiological stress. 20 ) Meta from fatty infiltration of the myocardium and arrhythmogerric
bolic events such as prolonged fasting and febrile illness blood levels of acylcamitines have been implicated.
246 I Sudden natural death in infants and children
At autopsy, pronounced fatty infiltration of the liver disease is strongly suspected. The predominant MCAD defi
(Fig. 12.20), kidney and muscle is usually found and should ciency mutation can be detected by DNA analysis of the stored
prompt appropriate investigation. Occasionally, individuals dried blood spot sample taken for mass neonatal screening of
dying with MCAD deficiency do not exhibit significant phenylketonuria,2°9 but DNA may also be obtained from
fatty infiltration of the liver,21I ·212 and its absence should frozen tissue samples and cultured fibroblasts. 193
therefore not preclude investigation along these lines. A
post-mortem diagnosis can be made by demonstrating:
DISORDERS OF OXIDATIVE PHOSPHORYLATION
• a characteristic profile of medium-chain
dicarboxylicaciduria and hexanoylglycine on urinary Compared with ~ -oxidation fatty acid defects, deaths due to
organic acid analysis; defects of OXPHOS are more likely to have demonstrable
• a raised plasma cis4 decenoate; pathology at autopsy (e.g. cardiomyopathy, liver fibrOSis,
• octanoylcarnitine by tandem mass spectrometry cinhosis). These disorders are wide-ranging in their clinical
(MS/MS) on dried blood spots (or using the original manifestation and patients only rarely present as sudden
Guthrie screening card); and death cases. Isolated myopathy, or a multisystem disorder
• ~-oxidation of fatty acids in intact cultured fibroblasts. that includes encephalopathy, liver disease and cardio
myopathy are features of the disorder. 2IJ The cardiomyopa
Urine, blood, vitreous humor, tissues and a skin biopsy thy is usually symmetrical (concentric), without outflow tract
are important for establishing a post-mortem diagnosis 20J obstruction. Death before the age of 1 year from heart failure
(Table 12.3). The minimal requirement of urine is 0.1 mL and, is the usual outcome.2l3 Histology and electron microscopy of
even if the bladder seems empty, swabbing the bladder the myocardium reveals swollen myocytes and an increase in
mucosa with a cotton ball may provide a sufficient volume the number of mitochondria. The mitochondria may also be
for analysis. Urine (or bladder swab) samples should be morphologically abnormal. The most common defects,
stored at -20°e. Organic acid analysis of vitreous humour according to the review by Guenthard et al,213 were deficien
may be useful in the absence of urine. 193 Liver, skeletal and cies of complexes I and lV, in isolation or combination.
cardiac muscle should be collected and stored without fixa An Il-month-old previously healthy boy reported by
tive at - BO°C prior to analysis. Blood should be collected Smeitink et al 214 died following a short Reye-like illness. A
into an anticoagulant and centrifuged; blood cells and defect in the coenzyme Q region of the respiratory chain
plasma should be stored separately at - 20°e. was found. His 6-month-old female sibling had previously
With the advent of gas chromatography-mass spectrome died following a similar illness.
try (GC-MS) technology, bile, blood and cerebrospinal fluid An acute apparent life-threatening event (ALTE) was
can be spotted onto a Guthrie card at post-mortem examina reported as the first sign of respiratory chain complex defi
tion for acylcamitine analysis. 206 Vitreous humour has also ciency in a 4-month-old girl whose parents were consan
been used for the diagnosis ofMCAD deficiency. A small skin guineous. 215 She was found to have markedly reduced
biopsy sample taken under sterile conditions, as soon as pos complex I activity.
sible after death, should be placed in tissue culture medium By the time of the post-mortem examination, it is usu
containing one per cent dimethyl sulfoxide and frozen ally too late to measure the function of the respiratory
at -70°C, or sent directly to the culture laboratory if metabolic chain as enzyme activity deteriorates rapidly.19J However,
Other bacterial infections I 24 7
Recurrent apnoea
Reye-like syndrome
Aminoaciduria
Metabolic acidosis
Hypoglycaemia
Hyperaminonaemia
Dica rboxylicacid uria
A metabolic screen with negative resu lts
Meningococcal infection often has a fulminant onset with a Figure 12.21 A widespread maculopapular rash typical of
high mortality variously attributed to adrenal haemorrhage, meningococcaemia.
endotoxic shock or a generalized Schwalizman reaction. The
disease has two clinical forms: meningitis and meningococ
caemia. 217 The main clinical manifestations are fever, rash
(Fig. 12.21), vomiting, lethargy, meningeal irritation and cir
culatory collapse. The rash may be maculopapular, purpuric
or mixed. 2 18 There was a sustained increase in the incidence
of meningococcal disease in the 1990s.219
An increased incidence of the condition is observed in
late winter and spring. A wide age range is affected, with a
peak incidence in the second year. Infection in the first 2
years of life and meningococcaemia with or without menin
gitis, compared with meningitis alone, are associated with
increased mortality. Other poor prognostic indicators are the
onset of petechiae within 12 hours prior to admission, shock,
normal or low white cell count, and a normal or low erythro
cyte sedimentation rate.217 Autopsy often demonstrates
bilateral adrenal haemorrhage (Fig. J 2.22). This is, however,
not an invariable or specific finding, as adrenal haemor
rhage also may be seen in haemorrhagic shock encephalopa
thy syndrome (see above) and other forms of septicaemia.
Laboratory findings include isolation of Neisseria meningi
tidis in the blood or spinal fluid, or the presence of Gram-neg Figure 12.22 Bilateral adrenal haemorrhage secondary to
ative diplococci in the spinal fluid or a petechial scraping in a men ingococcaemia.
patient with a typical clinical course. Unsuccessful attempts at
isolating the microorganism from the blood and spinal fluid ,
clinically or at autopsy, are often due to preceding antibiotic Sudden unexpected deaths in infants and young children
treatment. Antigen screening ofbJood, cerebrospinal fluid and have been ascribed to overwhelming infection as a result
urine are other diagnostic measures that may be adopted, as is of group A 0-haemolytic streptococcus, or Streptococcus
PCR amplification from blood or cerebrospinal fluid .2J9 pneu111oniae, with the organism being isolated from several
248 I Sudden natural death in infants and children
sites at autopsy.1J,220 However, such reports generally lack bronchopulmonary dysplasia.233 The pathophysiology of the
microscopic evidence of sepsis or associated features and a condition is not well understood but is thought to be
causal association is unclear. acquired rather than congenital. 231 Significant factors in its
Sudden death in infancy due to Haemopi1ilus injluenzae production are low gestational age at birth and a high mean
septicaemia has been reported in infants with asplenia, airways pressure during ventilation in the first week of life.
splenic hypoplasia and polysplenia, occurring in isolation In our own experience of 40 prematurity-related unex
or associated with complex heart disease [with or without pected post-neonatal deaths, two infants were found to
accompanying visceral abnormalities).221 At autopsy, adre have died from conditions unrelated to the prematurity
nal haemorrhage was a common finding in these cases. [RSV-related bronchiolitis and viral myocarditis). Twelve
Colonization or infection with Staphylococcus aureus infants displayed abnormalities directly or indirectly as a
and group A streptococcus is implicated in the toxic shock result of their prematurity; these were post-hypoxic
syndrome. 222 - 224 The condition appears to be toxin medi ischaemic encephalopathy, resolving brochopulmonary
ated; the exact pathophysiology is unknown and the clini dysplasia, tracheal stenosis and stricture of the small
cal course may be rapidly fatal. Identifying the site of intestine, alone or in combination. The cerebral lesions
infection may be difficult. Both adults and children may be included periventricular gliosis, mineralization and
affected. The streptococcal toxic shock syndrome is charac haemosiderin deposits, and periventricular cysts. Infants
terized by the sudden onset of fever, rash, vomiting and with more extreme degrees of prematurity tended to show
diarrhoea, hypotension, conjunctival injection and straw more severe pathology, but even in these instances the role
berry tongue, followed by skin desquamation. of the pathology in the cause of death was often unclear.
Necrotizing fasciitis is a rare, but often fatal, soft-tissue Pathology was not found in the remaining 26 infant
infection with rapidly spreading inflammation and necrosis deaths.
of the muscle fascia and subcutaneous fat, in some instances A limited respiratory reserve in infants with resolving
involving the epidermis. The condition is life-threatening bronchopulmonary dysplasia has been proposed as a
and carries a high mortality rate. Most cases are a result of mechanism of death in such cases, especially when there
polymicrobial or group A streptococcal disease. 225 ,226 Adults was superimposed viral or bacterial infection, producing
and children are affected. Pyrexia, leucocytosis, cellulitis sudden decompensation of cardiac and/or respiratory func
and oedema are the main clinical manifestations. The tion. 23o Injury to the pulmonary vascular bed, a feature of
affected skin becomes progressively warm, erythematous, bronchopulmonary dysplasia, is commonly associated with
swollen and painful. pulmonary arterial hypertension and right ventricular
Invasive group A streptococcal infection [GAS), with or hypertrophy.29 From a clinical perspective, however, the
without 227 necrotizing fasciitis, is the most serious compli picture is quite different, with Tammela and Koivist0 2J4
cation of varicella infection in children. 22B Progression of reporting an absence of fatality in a cohort of 86 low
GAS may be very rapid, with death at home or shortly after birth-weight babies, 23 with bronchopulmonary dysplasia,
arrival in hospitalYs after discharge from hospital and followed over a period of
Some patients with necrotizing fasciitis share many of 12 months, indicating a low mortality among infants with
the features of streptococcal toxic shock syndrome,226 and resolving bronchopulmonary dysplasia.
the onset of streptococcal toxic shock syndrome may be
preceded by necrotizing fasciitis.224 Neonatal omphalitis
can progress to necrotizing fasciitis, followed by rapid dete MISCELLANEOUS CAUSES OF SUDDEN
rioration and death.229 Mortality is increased by delayed NATURAL DEATH
treatment and any underlying debilitating disease.226
Sudden death has been reported in cases of congenital
myotonic dystrophy,235 although the mechanism was
DEATHS RELATED TO OBSTETRIC EVENTS unclear. Cardiac dyslythmia is also known to cause sudden
AND PREMATURE BIRTH death in patients with Emery-Dreifuss muscular dystro
phy.2J6 An occult adrenal phaeochromocytoma as a cause
Babies born pre-term are at risk of sudden unexpected of sudden death in a young boy following a kick to the
deaths at home. The autopsy findings include no anatomi anterior abdomen has been ascribed to a sudden surge of a
cal changes, coincidental disease and residual pathology large amount of noradrenaline into the circulation, produc
from prematurity-related complications. ing violent myocardial contractility.2J7 The sudden death of
Prolonged tracheal intubation and positive-pressure ven an infant who subsequently demonstrated widespread his
tilation in pre-term babies are associated with a range oftra tological features indistinguishable from Kikuchi-Fujimoto
cheal abnormalities including granulations, pseudopolyps, disease has been reported. 238
subglottic stenosis, bronchomalacia and tracheomala Encephalopathy in association with acute pancreatitis is
cia. 230 - 232 Tracheobronchomalacia may cause life-threaten welJ recognized in adults and has been rarely reported in chil
ing episodes from acute airways collapse in infants with dren.239 The condition is thought to be a result of damage to
Sudden death associated with 'inte rm ed iate ' pathology I 249
129:926-3 1.
32 Davies MJ, McKenna WJ. Dil a ted card iomyopathy: an 56 Lipsett J. Byard RW, Carpe n ter BF et at. Anomalous co ro nary
introduct io n to pathology and pathogenes is. Br Ht'Grt J aneries arising from the ao rta assoc iated with sudden death
1994; 72(Supp l.):24. in infancy and ea rly childh ood. Arch Pa thol Lab Med 1991 ;
33 Kelly DP, Stra uss AW. Inherited cardiomyopathies. N Eng l J 115 :770- 3.
Med 1994; 330:913-19. 57 Corrado D. Thiene G. Cocco P, Frescura C. Non
34 Lawson MR. Southwestern Interna l Medicine Conference: atherosclerotic coronary artery disease and sudd en death in
hypertrophic cardiomyopathy: current views on etiology, the young.
pathophys iology and managem ent. Am J Med Sci 1987 ; Br Heart J 199 2 ; 68 :60 1-7.
294:191 - 210. 58 Garfia A, Rodrigu ez M, Chava rria H. Garrid o M. Sudden
35 Burch M. Hypenrophic ca rdiomyopathy. Arch Dis Child ca rdiac death during exe rcise due to an isolated multiple
1994; 71:488-9. anomaly of the left co ronary a rtelY in a 12-year- o ld g irl:
36 McKe nn a WJ, S tewa n JT, Nihoyannopoulos P et al. Clinicopatholog ic fin dings. J Fo rens ic Sci 1997 ; 42:330-4.
Hypertrophic cardiomyopathy without hypertrophy : two 59 ScuJly RE, Mark EJ . McNeely W F. McNeely BU. Case reco rds
families w ith myocardial disarray in the abse nce of of th e Massachusetts Genera l Hospital . Case 22 -1 989. N Engl
increased myocardial mass. Br Hea rt J 1990; 63 :287 -90. J lVled 1989; 320: 1475 - 83 .
37 Gow RM. Sudden cardiac death in the young. Can J Cardiol 60 Thiene G, Pennelli N, Rossi L. Cardiac conduction system
1996; 12: 1157-60. abnormaliti es as a possible cause of sudd en de a th in yo ung
38 Burch M, Mann JM, Sh a rl and M et a l. Myocardial disarray in athletes. Hum Pathol 1983; 14:704-9.
Noonan sy ndrome. Br Heart J 1992; 68:586- 8. 61 Angelini P. Normal and anomalous coronalY arte ries :
39 Ha lliday HL. Hypenrophic ca rdiomyopathy in infa nts of detinitions and classification . A 111 Heart J 1989; 117 :41 8- 34.
poo rl y-co ntroll ed diabetic mothe rs. Arch Dis Child 198 1; 62 Price J . Kawasaki sy ndro me. Br Med J 1984 ; 288:262 - 3.
56 :258-63. 63 Curti s N. Kawasaki disease. Br Med J 1997 ; 315:322-3 .
40 Burch M, Runcima n M. Dilated ca rd io myopathy. Arch Dis 64 Ti za rd EJ, Suzuki A. Levin M. Dill on MJ. Clinical as pects of
Child 1996; 74:479-81. 100 patients wit h Kawasa ki disease . Arch Dis Child 1991 ;
41 Maron BJ, Shirani J, PoJiac LC et a l. Sudden death in young 66:18 5- 8.
competitive athletes: clini cal, demog raphic, and path o logical 65 Nakamura Y. Yanagawa H, Ojima T et al. Cardiac sequelae of
profiles . J Am !VIed Assoc 1996; 276: 199-204. Kawa saki disease among recurrent cases . Arch Dis Child
42 Huber S. Viral myocarditis - a tale of two diseases. Lab 1998; 78:1 63 -5.
hllJesr 1992; 66:1- 3. 66 Hoare S. Abinun M. Cant AJ. Ov erl ap between Kawasaki
43 Be nnett MJ, Hale DE, Pollitt RJ et al. Endocardial disease and Group A streptococcal infection (Letter). Pediatr
fibroelastosis a nd prima lY ca rnitine deficiency due to a /l ljixt Dis J 1997; 16:633 -4.
defect in the plasma membran e ca rnitine tran sponer. Ciil1 67 Bya rd RW. Cardiac co nditions. In Byard RW (ed.) Sudden
Cardiol 1996; 19: 243 - 6. Death in Infancy Childh ood and Adolescellce. 2nd edn.
44 Virman i R, Robin owitz M, Clark MA, McAllister HA. Sudden Cambrid ge: Ca mbrid ge University Press. 2004 , pp. 202 - 48.
death and partial absen ce of the right ventricular 68 Bharati S. Bauernfiend R, Scheinm an M et al. Congenita l
myocardium. Arch Pathol Lab Med 1982; 106:163-7. abnormalities of the conducti on system in two patients with
45 Thi ene G, Nava A, Corrado D e t a!. Right ventricul ar tachyarrhythmias. Circulation 1979; 59:593-606.
ca rdiomyopathy and sudd en deat h in young peopl e. N Engl 69 Splawski I, S he n J, Timothy KW et al. Spectrum of mutations
J Med 1988 ; 318:129-33. in long Q-T syndrome genes KVLQTI. HERG. SCN5A.
46 McKe nna WJ, Thi ene G, Nava A et al. Diagnos is of KCNE I and KNCE2. Circula tiol1 2000 ; 102: 1178-85.
a rrh y thmoge nic ri g ht ventric ul a r dysplasia/ca rdi o myopathy. 70 Towbin JA. Molecular basis of sudden ca rdiac death. Pediatr
Br Heart J 1994; 7 1:2 15-1 8. Ciin N Am 2004; 51: 1229 - 55.
47 Fonta ine G. Arrhythmoge ni c rig ht ven tricula r dyspla sia. Curr 71 Moran JJ. Idiopathic arterial calcificatio n of infancy: a
Opin Cardiol 1995; 10: 16-20. clin icopatho logic study. Pa t/wi Ann 1975; 10:393-417.
48 A hmad F, Li D, Karibe A et al. Localisation of a gene 72 Rutsch F. Vainganka r S. Johns on K et al. PC-I nucleos ide
responsible for a rrh ythmoge ni c right ventricul ar dyspla sia to triphosphate pyrophosphohydrola se defiCiency in idiopathi c
chromosome 3p23. Circu lation 1998; 98 :2791 - 5. infanti le arte rial calcification. Am J Parha l 2001;
49 Gerlis LM, Schmid t-O tt SC, Ho SY, Anderson RH. Dysp lastic 158 :543-54.
cond itions of the right ventricu lar myocardium: Uhl's 73 Rutsch F. Ruf N. Vaingankar S et al. Mutations in ENPP 1 a re
anoma ly versus arrhythmoge nic right vent ricu lar dysplasia . associated with 'idiopathi c' infantile arterial calcification.
Br Heart J 1993; 69: 142 - 50. Nat Geller 2003; 34:379-81.
50 Tansey OK, A ly Z, Sheppa rd MN. Fat in the ri g ht ventricle of 74 Juul S. Ledbetter D. W ight T N, Wood rum D. New in sights
the no rmal heart. Histopath ology 200 5 ; 46 :98- 104. into idiopathic infantile arteria l ca lci nos is. Three patient
51 Krous HF, Ch adw ick AE, Isaacs H. Tum ors associated w ith repons. Am J Dis Child 1990; 144 :229-33 .
sudden infan t and childhood death. Pedia tr Dev Pathal 75 Byard RW. ldiopathic arterial calcification and unexpected
2005; 8:20-5. infant death . Pediatr Pathol Lab Med 1996; 16:985-94.
52 Shehata BM, Patterson K, Thomas JE et al. Histiocytoid 76 Joshi VV. Pawel B. Connor E et al. Arteriopathy in children
cardiomyopathy: Three new cases and a review of the with acquired immune de fi ciency syndrome. Pediatr Pathol
literature. Pediatr Dev Parhol 1998; 1:56-69. 1987; 7:261-75.
53 Suarez V, Fuggle WJ. Cameron AH et al. Foamy myocardia l 77 Siegel RJ. Dunton SF. Systemic occlusive arteriopathy with
transformation of infa ncy: a n inherited disease. J Ciill Pathol su dden death in a 10-year- old boy. Hum Parhol 1991;
1987; 40:3 29 - 34. 22:197- 200.
54 Koponen MA, Siegel RJ. Hi st iocytoid ca rdi omyopathy a nd 78 Im a mura M. Yokoya ma S. Kikuchi K. Coronary
sudden death. Hum Pathol 1996; 27:4 20-3 . fibromuscular dys plasia presen tin g as sudden infant deat h.
55 Ness MJ. McManu s BM. Anomalous right coronary artery Arch Pathol Lab Med 1997; 121 : 159-6 1.
origin in otherw ise unexplained infant death. A rch Pat/101 79 Rosenberg HG. Systemic arterial disease and chronic
Lab Mcd 1988; 112:626-9. arse ni cism in infants. Arch Pathol 1974; 97:360- 5.
252 I Sudden natural death in infants and children
80 Rol fes DB, Towbin R, Bove KE. Vascular dysplasia in a chil d 104 Byard RW, Bo urn e AJ, Hanieh A. Sudden and unexpected
with tube rou s sclerosis. Pediatr Pathol 1985 ; 3 :359-73. death due to hemorrhage from occult central nervo us system
81 Mudd SH, Levy HL, Skovby F. Disord ers of transsulfuration. lesions. Pedia tr Neurosurg 1991 -92; 17:88-94.
In Sc river CR, Beaudet Al, Sly WS, Va lle D (ed s) The 105 Lill ey man JS. Intrac rania l haem orrhage in idiopathic
Meta bolic and Molecular Bases of Il1h erited Disease. New thrombocy topeni c purpura. Arch Dis Child 1994; 71 :251-3.
Yo rk: McGraw-Hili Inc. , 199 5; 1279- 327. 106 Abu AI Ragheb SY, Koussous KJ, Arm SS. Intracra nial
82 Fo rtu in NJ, Morrow AG, Roberts We. Late vascular
neop lasms associa ted with sudd en death : a rep ort of seve n
manifesta tions of the Ru be lla synd rom e. Am J Med 1971;
cases a nd a review of the literature. Med Sci Law 1986 ;
51: 134-40. 26:270-2.
83 Hamilton SJ, Allard MF, Friedm a n JM. Cardiac Findin gs in 107 DiMa io SM, DiMaio VJM, Kirkp atrick J B. Sudden ,
an indiv idual with neurofibromatos is I and sudden dea th . un expec ted death s due to primary intracranial neopl as ms.
Am J Med Genet 2001; 110:95-9. Am J Foren sic Med Pathol 1980 ; 1:29-45.
84 Uno H, Arya S, Laxova R, Gilb ert EF. Menkes' syndro me
108 Bya rd RW, Moo re L. Sudd en and unexpec ted death in
with vascula r and ad ren ergic nerve abn ormalities . Arch
childhood du e to a co lloid cyst of the third ven tricle.
Path ol Lab Med 1983; 107 :286- 9.
J Forens ic Sci 1993; 38 :210 -13.
85 Fleishe r GR , Buck BE, Co rnfeld D. Prim ary intimal 109 Buzzi S, Verdura C, Arlati S, Colecchia M. Sudden deat h in a
fibroplasia in a child with Down's syndrome. Am J Dis Child child due to ra re endo crani al neo fo rm ation . Med Sci Law
1978 ; 132 :700-3 . 1998; 38: 176-8.
86 Cina SJ, Smialek, JE, Burk e AP et al. Primary ca rdiac tumors 110 Leestma JE, Walcz ak T, Hughes JR et al. A prospective study
causing sudden death: a review of the literature. Am J on sudden une xpected death in epilepsy. Ann Neural 1989;
Forensic Med Pa thol 1996 ; 17:271-81. 26:195-203.
87 Siega l GP. Pri mary tumors of mu scl e. In Stocker JT, Askin
II I Byard RW, Keeley FW, Smith CR. Ty pe IV Ehlers-Danl os
FB (eds) Pathology of Solid Tum on il1 Ch ildren. London:
sy nd rome presenting as sudd en infant death. Am J Clin
Chapman and Hall Medical, 1998, pp. 161-81.
Patl101 1990 ; 98:579-82.
88 Parker KM , Embry .IH. Sudden dea th due to tri cuspid va lve 11 2 Elterman RD, Bodensteiner JB, Barnard JJ. Sudden
myx om a with massive pulmonary emb olism in a 15-m onth unexp ec ted death in patien ts with Dandy-Walker
old male. J Forensic Sci 199 7; 42 :524-6. malformation . J Child Neurol 199 5; 10:382-4.
89 Krous HF, Chapman AJ, Alts hul er G. Cardiac hema ngioma: a 113 Vaishnav A, Ma cKinn on AE. Prog ressive hydrocephalus in
rare (or poss ible) cause of su dden death in child ren. J teena ge spina bi fid a patients. Z Kindercilinl rg 1986;
Foren sic Sci 1978 : 23:375-8. 41(Suppl. 1):36-7.
90 Swalwell CI. Be nign intra cardiac teratoma - a case of
114 Stellm a n GR, Banniste r CM, Hillier V. The incidence of
sudd en death. A rch PatilOl Lab Med 1993; 117 :739-42.
seizure disorder in children with acq uired a nd congenital
91 Byard RW, Cutz E. Sudden and un expec ted dea th in in fa ncy hydrocephalus. Z Kinderchirurg 1986 ; 41( Suppl. 1):38-9.
and ch ildhood due to pulm ona lY thro mb oemb olism. Arch 115 Friede RL. Meningoencephalitic pro cesses in th e perinata l
Pathol Lab Med 1990 ; 114:14 2-4. period. Deue!opmental Neuropathology. New York: Sprin ge r
92 Champ C, Byard RW. Pulmonary thromboembolism and
Verlag, 1975 ; 178-87.
unexpected dea th in infancy. J Pediatr Child Health 199 4;
116 Howat AJ, Smith CML, Variend S. Sudden infant death a nd
30:550-1.
encephalitis in two sib lings. Dev Med Child Neu ra l 1987;
93 Pa tton DJ, Ga lliani CA, Johnson WH, Hed lund GL. Sudd en
29:539-40.
death in Marfan syndrome. A m Joumal of Roentgenology
117 Shield s LBE, Handy TC, Parker JC, Burns e. Postmoltem
1995; 165:1 60.
diagnosis of leukodystrophi es. J Forensic Sci 1998 ;
94 Masse J-F, Perusse R. Ectoderm al dysplasia. Arch Dis Child 4 3:1068 -71.
1994; 71: 1- 2. Jl8 Kereny i N. Conge nital ad renal hyp opla sia. A rch Pathol 196 1;
95 Clarke A, Phillips DIM , Brown R, Harpe r PS. Clini cal aspects 71 :336- 3.
of X-linked hypohidrotic ectodermal dyspl as ia. Arch Dis 119 Irvine WJ, Toft AD. Diagnosing adrenocortical insufficiency.
Child 198 7: 62:989-96. Practitioner 1977; 218 :539-45.
96 Waters KA, Evere tt F, Sillen ce D et al. Breathing
120 AI Sabri ANI, Sm ith N, Busuttil A. Su dden death due to a uto
abno rm a lities in sleep in achondroplasia. Arch Dis Child
immun e Addiso n's disease in a 12-yea r-old girl. Int J Legal
1993; 69:191-6.
Med 199 7; 110 :278-8 0.
97 Bla nd JD, Emery JL. Unexp ected death of ch ildren with
121 Fava ra BE, Franc iosi RA, Mil es V. Idiopathic adrena l
achondroplasia a fter th e perin a tal period . Dev Med Child
hypop lasia in children. Am J ((in Pa tl10 I 1972 ; 57 :287-96.
Neurol 1982: 24:489-92.
122 Brown W, Singer DB. Pregnan cy-indu ced hyp ertension and
98 Webb DW, Osbo rne JP. Tuberous scl erosis. A rch Dis Child congenital adrenal hypoplasia. Obstet Gynaecol 1988;
199 5; 72:471-4. 72: 190-4.
99 Nir A, Taj ik J , Freeman WK et a l. Tuberous scl erosis and 123 Bacon CJ , Hall SM. Haemorrh agic s hock encephal opa thy
cardia c rh abdomyosarcoma. Am J ·Cardiol 1995; 76:4 19 - 21. syndrome in the British Isles. A rch Dis Child 199 2;
100 O'CalJaghan FJK, Clarke AC, Joffe H et al. Tuberous scle ros is 67:985-9 3.
complex a nd Wo lff-ParKinson - Wh ite sy ndrome. Arch Dis 124 Li ttle D, Wilkin s B. Hemo rrhagic shock and encephalopa thy
Child 1998 ; 78: 159-62. syndrom e : an unusual cause of sudd en death in children.
101 Livingsto n JH , Brown JK. Intrac erebra l haemorrhage after Am J Forensic Med Pathol 1997; 18 :79-8 3.
the neon ata l period. Arch Dis Child 1986; 61 :538-44. 125 PHLS Communicab le Disease Surveill ance Centre. Joint
102 Gieteling EW, Rinkel GJE. Ch aracteristics of intracranial British Paediatric Association a nd Communicable Disease
a neu rysms and subara ch noid haemorrhage in patients with Surveillan ce Scheme for haemorrhagic sho ck encephalopathy
polycysti c kidn ey disease. J Neural 2003; 250:418- 23. syndrome surveill a nce report for 198 2-4. Br Med J 1985 ;
103 Lilova MI , Petkov DL. Intracranial aneurysms in a child with 290: 1578-9.
autosom a l. recessive poly cysti c kidney disea se. Pediatr 12 6 Glasgow JFT, Mo ore R. Reye's sy ndro me 30 years on. Br
Nephra l 2001 ; 16: 1030- 2. Med J 1993; 307 :9 50-1.
References I 253
127 Anon. Rey e's syndrome and aspirin: epidemiolog ical 152 Coulthard M, Isaacs D. Retroph ary ngeal abscess. Arch Dis
associations and inborn erro rs of metabol ism. Lancet 1987; Child 1991; 66: 1227-30.
ii:429-31. 153 Barrat GE, Koopmann CF Jr, Coulthard Sw. Retropharyngeal
128 Bonnel HJ, Beckwith JB. Fatty liver in sudde n childhood abscess - a ten-year experience. Laryngoscope 1984;
death. Am J Dis Child 1986; 140:30-3. 94:455-63.
129 Forsyth BW, Shapiro ED, Horwitz Rl et al. Misdiagnosis of 154 Brook J. Microbiology of retropharyngeal abscesses in
Reye's-like illn ess (Letter). Am J Dis Child 1991 ; 145:964-6. chi ldren. Am J Dis Child 1987 ; 141:202-4.
130 Green A, Hall SM. Investigation of metabolic disorders 155 Donne lly BW, McMillan JA, Wei ner LB. Bacterial tracheitis:
resemblin g Reye's syndrom e. Arch Dis Child 1992; 67: 13 13-17. Report of eight new cases and review. Rev Infect Dis 1990;
131 Hardie RM, Newton LH, Bruce JC et al.. The chan ging clinical 12:729-35.
pattern of Reye's sy ndrome 1982- 1990. Arch Dis Child 156 Britto J, Habibi P, Walters Set al. Systemic complications
1996; 74:400-5. associated with bacterial tracheitis. Arch Dis Child 1996;
132 Rowe PC, Valle D, Brusilow Sw. Inborn errors of metabolism 74:249-50.
in children referred with Reye's synd rome . JAm Assoc 1988; 157 McKenzie M, Norman MG, Anderson m, Thiessen PN. Upper
260:3167-70. respiratory tract infection in a 3 year old girl. J Ped iatr
133 Byard RW, Bohn DJ, Wilson G et al. Unsuspected 1984; 105:129-33.
diaphragmatic hernia: a potential cause of sudden and 158 Cant AJ, Gibson PJ, West RJ. Bacterial tracheitis in Down's
unexpected death in in fancy and early childhood. J Pediatr synd rome. Arch Dis Child 1987; 62:962- 3.
Surg 1990; 25:1166-8. 159 Pan itch HB, Callahan CW, Schidlow DV. Bro nchiolitis in
134 Pfalzgraf RR, Zumwalt RE, Kenny MR. Mesodivelticular children. Clin Chest Med 1993 14:715-31.
band and sudden death in children. Arch Puthol La.b Med 160 Price JE Acute and long-term effects of vi ra l bronchiolitis in
1988; 112:182-4. infancy; Lung 1990; 168(Suppl.):414-21.
135 Pumphrey RSH, Stanw orth SJ. The clinical spectrum of 161 Meert K, Heidemann S, Abella B, Sarn aik A. Does
anaphylaxis in north-west England. c/in Exp Allergy 1996; prematurity alter the course of respiratory syncytial virus
26: 1364-70. infection? Crit Care Med 1990; 18 : 1357- 9.
136 Pumphrey RS, Robert IS. PostmOltem findings after fatal 162 MacDonald NE, Breese Hall C, Suffin SC et al. Respiratory
anaphyl actic reactions. J c/in Pa thol 2000; 53 :273-6. syncytial viral infection in infants with congenital heart
137 Delage C, lrey NS. Anaphylactic deaths: a clinicopathologic disease. N Eng! J Med 1982; 307:307-400.
study of 43 cases. J Forensic Sci 1972; 17:5 25 - 39. 163 Groothuis JR, Gutierrez KM, Lauer BA. Respiratory syncyt ial
138 Sampson HA, Mendelson MD, Rosen JP. Fatal and near-fatal virus infection in children wit h bronchopulmonary
anaphyl actic reaction s to food in childre n and adolescents. dysplasia. Pediatrics 1988; 82:199 - 203.
N Engl J Med 1992 ; 327:3 80-4. 164 Hall CB, Powell KR, Ma cDonald NE et al. Respiratory
139 Riches KJ, Gillis D, James RA. An autoJlsy approach to bee syncytial viral infection in chi ldren with compromised
sting-related death s. Pathology 2002; 34:257-62. im mun e function. N EI1g1 J Med 1986; 315:77-81.
140 Byard RW. Miscellaneous conditions. In Byard RW (ed.) Sudden 165 Hastleton PA, Ironside JW, Whittaker JS et al. Pulmonary
Death ill Infancy, Childhood and Adolescence, 2nd ed n. veno-occlusive disease. A report of four cases.
Camb ridge: Cambridge University Press, 2004, pp. 469-88. Histopathology 1986; 10:933-44.
141 Assem ESK, Gelder CM, Spiro SG et al. Anaphylaxis induced 166 Cagle P, Langston C. Pulmonary veno -occlus ive disease as a
by peanuts. Br i\lIed J 1990; 300: 1377 -8. cause of sudden infa nt death. Arch Pat/wi Lab Med 1984;
142 Patel L, Radivan FS, David T1. Management of anaphylactic 108:338-40.
reactions to food. Arch Dis Child 1994; 71:370-5. 167 Jay GW, Leestma JE. Sudden death and epilepsy. Acta
143 Powars D, Chan LS, Schroeder WA. The variable expression Neurologica Scandinavica 1981; 63(Suppl. 82): 11 -16.
of s ickle cell di sease is genetica lly determined. Semil1 168 McGugan EA. Sudden unexpected deaths in epileptics
Hematol1990; 27:360-76. a literature review. Scot iVIed J 1999; 44:137-9.
144 Jenkins ME, Scott RB, Bairs RL Stud ies in sickJe cell 169 Shields WD. Status epilepticus. Pediatr Clin North Am 1989;
anaemia. J Pedia tr 1960; 56:30-9. 36:383 - 93.
145 Leikin SL, Gallagher D, Kinney TR et al. Mortality in children 170 Leestma JE. Forensic aspects of complex neural
and adolescen ts with sickJe cell disease. Pediatrics 1989; dysfunctions. In Leestma JE (ed.) Forensic Neuropatho logy.
84:500-8. New York: Raven Press, 1988, pp. 396-428.
146 Clini copathologic Conference. Sudden death in a young 171 Coyle HP, Baker-Brian N, Brown Sw. Coroners' autopsy
woman with sickle cell anaemia. Am J Med 1992; reporting of sudden unexplained death in epil epsy (SUDEP)
92:556-60. in the UK. Seizure 1994; 3:247-54.
147 Cutz E. Idiopathic pu lmonary hemosiderosis and related 172 Nashef L, Walker F, Allen P et al. Apnoea and bradycardia
disord ers in infan cy and childhood. In Rosenb erg HS, during epilept ic seizures: relation to sudd en death
Bernstein J (eds) Respiratory I1l1d Alimentary Tract Diseases. in epil epsy. J Neurol Ne'urosuTg Psychiatry 1996;
Basel: Boehm-Hutter Press, 1987, pp. 47-81. 60:297-300.
148 Stocker JT. The respiratory tracr. In Stocker JT, Dehner LP 173 Lip GYH, Brodie JVIJ. Sudden death in epilepsy: an avoidab le
(eds) Pediatric Pathology Vol. 1. Philadelphia: Lippincott outcome? J Royal Soc Med 1992; 85:609-11.
Company, 1992; 505-73. 174 Brown SW, Mawer GE, Lawler W et al. Sudden death and
149 Schwartz MZ, Fill.er RM. Tracheal compression as a cause of epilepsy (Letter). Lancer 1990; 335:606-7.
apnea following repair of tracheo-oesophageal fistula: 175 Timmings PI.. Sudden unexpected death in epi lepsy: is
treatment by aortopexy. J Pediatr Surg 1980; 15:842-8. carbamazipine implicated? Seizure 1998 ; 7:289-91.
150 Trollfors B, Nylen 0 , Strangert K. Acute epiglottitis in child ren 176 Gars on A Jr. Medicolegal probl ems in the management
and adults in Sweden 1981-3. Arch Dis Child 1990; 65:491 -4. of cardiac arrhythmias in children. Pediatrics J 987 ;
151 Claesson B, Troll fors B, Ekstrom-Jodal B et al. Incidence and 79(1):84-8.
prognosis of ac ute ep iglottitis in children in a Swed ish 177 Brown DC, Godman MJ. Life-threatening 'epi lepsy'. Arch Dis
region. Pediatr Infect Dis 1984; 3:534-7. Child 1991; 66:986-7.
254 I Sudden natural death in infants and children
178 Ch amp C, By ard RW. Sudden death in asthma in chi ldhood . 204 Largilliere C, Vianey-Saban C, Fontaine M et al.
Forens ic Sci Int 1994; 66:117-27. Mitochondrial very long chain acyl-CoA dehydrogenase
179 Zach MS , Karner U. Su dden death in asthma. Arch Dis Child defiCiency - a new disorder of fatty acid oxidation. Arch Dis
1989; 64: 1446-51. Child 1995; 73:FI03-S.
180 Carswell F. Thirty deat hs from asthma. Arch Dis Ch ild 1985; 205 Olpin SE. Allen J. Bonham JR et al. Features of Carnitine
60: 25 -8. Palmitoyltransferase Type I Disease. J Inherit lvIetab Dis
181 Kravis LP, Kolski GB. Unexpected death in childhood asthma. 2001; 24:35-42.
Am J Dis Child 1985; 139:558-63. 206 Chace DH. DiPerna JC, Mitch ell BL et al. Electrospray tandem
182 Benatar SR. Fatal asthma. N Engl J IVIed 1986; 314:423-9. mass spectrom etry for analysis of acylcarnitines in dried
183 Larsen GL. Ast hma in children. N Engl J IVIed 1992; postmortem blood specimens collected at autopsy from infa nts
326:1540- 5. with unexp lained cause of death. Clin Chem 2001 ; 47:JJ66-82.
184 Molfino NA, Nannini LJ. Martelli AN, Sl utsky AS. 207 Touma EH. Charpentier C. Med ium chain acyl-C oA
Respiratory a rrest in near-fatal asthma. N Engl J Med 1991 ; dehydrogenase deficiency. Arch Dis Ch ild 1992 ; 67:142-5.
324:285- 8. 208 Roe CR. Millington OS. Maltby DA. Kinnebrew P. Recognition
185 Haalboom JRE, Deenstra M. Stmyvenberg A. Hypokalaemia of medium-chain acyl-CoA dehydrogenase deficiency in
induced by inhalation of fenot erol. Lallcet 18 May 1985 ; asymptomatic siblings of children dy ing of sudden infant
2 :1125-7. death or Reye-like syndromes. J Pediatr 1986; 100:13-18.
J 86 Pearce N. Crane J. Burgess C et al. Beta ago nists a nd asthma 209 Anon. Med ium chain acyl CoA dehydrogenase deficiency.
mOl1a lity: deja vu. Clin Erp Allergy 1991 ; 21 :401 - 10. Lancet J991; 338:544-5.
187 Warner DP, McKinn ey PA. La w GR, Bod ansky HJ. MOl1ality 210 Shortland G. Besley G, Bonham J et al. Newborn screening for
and diabetes from a population-based register in Yorkshire medium chain acyl CoA dehydrogen ase deficiency (MCADD):
1978-93. Arch Dis Child 1998; 78:435-8. Find ings from a multicentre prospective UK collaborative
188 Dahlquist G, Kallen B. Mortality in childhood-onset type 1 study. J Inherit Metab Dis 2006; 29(Suppl. 1): 19.
diabetes: a population-based study. Diabetes Care 2005; 2JJ Losty He. Lee P, AI Faham M et al. Fatty infiltration in the
28:2381-7. liver in medium chain acyl CoA dehyd rogenase deficiency.
189 Edge JA, Ford-Ad ams ME, Dunger DB. Causes of death in Arch Dis Child 199 1; 66:727-8.
children with insulin- depend ent diab etes 1990-96. Arch Dis 212 Bove KE. Letter to the Editor. Pediatr Patlwl 199 2; 12:621-5.
Child 1999; 81:318-23. 213 Guenthard J. Wyl er F. Fowler B, Baumgartner R.
190 DiMaio VJM. Sturn er WQ, Coe JI. Sudden and unexpected Cardiomyopathy in respiratory chain disorders. Arch Dis
deaths after the acute onset o f diabetes mellitus. J Foren sic Child 1995; 72: 223 -6.
Sci 1977 ; 22: 147-5 1. 214 Smeitink JAM. Fischer JC. Ruitenbeek W et al. Sudden infant
191 Rosenbloom Ai. Kohrman A. Sperling M. Classifica tion a nd death associated with defect ive oxidative ph osp horylation.
diagnosis of diab etes mellitus in children and adolescents. Lancet 199 3; 341: 1601.
Joul11al of Pediatrics 1981; 99:320-3. 215 Konstantopouloll V, Sperl W. Wohl gena nnt J et al. Nea r
192 Coe JI. Post-mortem bioch em istry of bl ood and vitreous missed sudden unexpected infant dea th (SUID) as the firs t
humour in paediatri c practi ce. In Mas on JK (ed.) Paediatric sign of a respiratory chain co mplex I deficiency. J Inh eri t
Forensic Medicine and Pa thology. London: Cambridge lvfetab Dis 2001; 24(SuppL 1):78.
Universi ty Press, 1989, pp. 191-203. 216 Tripp ME. Katcher ML. Peters HA et al. Systemic carnitine
193 Olpin SE. The metabo lic investigation of sudd en infant deficiency presenting as familial endocardial fibroelastosis.
death. A/1I1als of Clinical Biochemistry 2004; 41: N El1gl J Med 1981; 305:385-90.
282 -93. 217 Steihm ER. Damrosch DS. Factors in the prog nosis of
194 John WG . Scott KWM, Hawcroft DM. Glycated haemoglobin meningococcal infection. J Pedialr 1966; 68:457-67.
an d glycated protein and glucose concentrations in necropsy 218 Marzou k 0, Thoms on APJ, Sills JA et al. Features and
blood samples. J Clin Pa thol 1988 ; 41:415-18. outcome in meningococca l disease presenting with
195 Rozin L, Perper JA. Jaffe R, Drash A. Sudden unexp ected maculopapul ar rash. Arch Dis Child 1991; 66:485-7.
death in childhood due to unsuspected diab etes mellitus. 219 Morl ey SL, Lev in M. Bacterial meningitis. Prescr J 1998;
Am J Forensic Med Pa thol 1994; 15:25 1-6. 38:129-4l.
196 Tattersall RB. Gill Gv. Unexplained deaths of type 1 diabetic 220 Sh arief N, Kh an K, Conlan P. Overwhelming sepsis
patients. Diabetic Medicin e 1991; 8:49-58. presenting as sudden unexpected death. Arch Dis Child
197 Thordarso n H, Sovik O. Dea d in bed syndrome in young 199 3; 69:381-3.
diabetic patients in Norway. Diabetic Medicine 1995; 22 1 Dyke MP, Martin RP, Berry PJ. Septicaemia and adrenal
12:782-7. ha emo rrhage in congenital aspl enia. Arch Dis Child 1991;
198 Matyka K, Ford-Adams M. Dunge r DB. Hy poglycaemi a and 66: 636 -7.
co unterregulation during childh ood. Horm Res 2002; 222 Larkin SM, Williams ON, Osterholm MT et al. Tox ic shock
57(Suppl. 1):85-90. syndrome: clinical. laboratory, an d pathologic findings in
199 Emery JL. Howat AJ, Variend S. Vawter GF. Investigat ion of nine fatal cases. Annals of internrillvfedicine 1982 ;
inborn errors of metabolism in un expec ted infant deaths. 96:858-64.
Lancet 1988; ii: 29 -31. 223 Con e LA. Woodard DR. Schlievert PM. Tomory GS. Clinical
200 Pollitt RJ. Disord ers of mitochondrial long -cha in fatty acid and bacteriologic observations of a toxi c shock-like
oxidation. J Inh erited Metob Dis 1995; 18:473-90. syndrome due to Streptococcus pyogenes. N Engl J Mer!
201 Pollitt RJ. Inherited disorders of straight chain fatty ac id 1987; 317:146-9.
oxidation. Arch Dis Child 1987; 62:6-7. 224 Steve ns DL. Tanner MH . Winship J et al. Severe g roup A
202 Anon. Sudden infant death and inherited disorders of fat streptococcal infection s associated with a toxic shock-like
oxidation. Lan cet 8 Nov ember 1986; ii: 1073- 5. syndrome and scarlet fever toxin A. N Eng l J Med 1989;
203 Benn ett MJ, Hale DE, Coates PM , Sta nley CA. Postmortem 321 :1-7.
recognition of fatty acid oxidation disord ers. Pediatr Pathol 225 Stamenkovic I. Lew PD. Earl y recognition of potentially fatal
1991; 11:365-70. necrotizing fasciitis. N Engl J lvled 1984 ; 310:1689-93.
References I 255
226 Chelsom J, Ha lstensen A. Haga T, H0iby EA. Necro tisi ng 235 Reardon W, Newcombe R, Fenton I et aJ. The natural history
fasciitis due to g roup A streptococci in western Norway: of congenital myotoniC dystrophy: mortality and long term
incidence and cI inical features. Lancet 1994: 344: II J 1-15. clinical aspects. Arch Dis Child 1993; 68:177-81.
227 Tyrrell GJ, Lovgren M, Kress B, Grimsrud K. Varicella 236 Thomas PK, Caine DB, Elliott CF. X-linked scapuloperoneal
associated invasive group A streptococcal disease in Alberta, syndrome. J Neurol Neuroslirg Psychiatry 1972; 35:208-1 5.
Canada - 2000-2002. Ciin Infect Dis 2005; 40: 105 5-7. 237 Primhak RA, Spicer RD, Variend S. Sudden death after minor
228 Vugia DJ, Peterson CL, Meyers HB et aJ. Invasive group abdominal trauma: an unusual presentation of
A streptococcal infections in children with varicella phaeochromocytoma. Br Med J 1986; 292:95-6.
in Southern California. Pediatr Infect Dis J 1995; 238 O'Neill D, O'Grady J, Variend S. Child fatality associated
15:146-50. with pathological features of histiocytic necrotizing
229 Sawin RS , Schaller RT, Tapper D et al. Early recognition of lymphadenitis (Kikuchi-Fujimoto Disease). Pediatr Pafhol
neonatal abdomina l wall necrotizing fasciitis. Am J Surg Lab Med 1998; 18 :79-88.
1994; 167:481 -4. 239 Morens OM , Hammar SL, Heicher DA. Idiopathic acute
230 Stocker JT. Pathologic features of long-standing pancreatitis in Children. Am J Dis Child 1974; 128:401-4.
'healed' bron chopulmonaty dysp lasia. Hum Pathol 1986 ; 240 Wilcken B, Hammond J, Silink M. Morbidity and mOltality
17:943-61. in medium chain acyl coenzyme A dehydrogenase
231 Miller RW, Woo P, Kellman RK. Tracheobronchial deficiency. Arch Dis Child 1994; 70:410-12.
abnormalities in infants with bronchopulmonary dysplasia. 241 Traisman ES, Young S, Lifschultz BD et al. Sudden death in a
J Paediatr 1987; 111 :779-82. neonatal as a result of Herpes simplex infection. J Forensic
232 Greenholz SK, Hall RJ, Lilly JR, Shikes RH. Surgical Sci 1988; 33:267-71.
implications of bronchopulmonary dysplasia. J Pedia tr Surg 242 Mercuri E, Cowan F, Rutherford M et aJ. Ischaemic and
1987; 22:1132-6. haemorrhagic brain lesions in newborns with seizures and
233 Doull lJ, Mok Q, Tasker RC. Tracheobronchomalacia in pre normal apgar scores. Arch Dis Child 1995; 73 :F67-74.
term infants with ch ronic lung disease. Arch Dis Child 1997; 243 Norman MG, Taylor GP, Clarke LA. Sudden, unexpected,
76:F2OJ- 5. natural death in childh ood. Pediat,. Pafho11990; 10:769-84.
234 Tammela OKT, Koivisto ME. A I-year foHow-up of lo w birth 244 Variend S. Infant mortality, microglial nodules and parotid
weight infants with and without bronchopulmonalY CMY-type inclusions. Ear/y Hum Del) 1990; 21:31-40.
dysp lasia: hea lth , growth, clinical lung disease, 245 Variend S, O'Neill D, Arnold P. The possible significance of
cardiovascular and neurological sequelae. Early Hum Dev cytomegaloviral parotiti s in infant and early childhood
1992 ; 30: 109-20. deaths. Arch Pat/wi Lab Med 1997; 121 :1272-6.
I CHAPTER 13 I
RECENT ADVANCES IN PAEDIATRIC
TOXICOLOGY
Patrice Mangin and Christian Giraud
During the past decade paediatric toxicology has substan Poisoning agents are usually those found in the home and
tially advanced in line with all other branches of medical may include household chemicals, such as cleansers, hydro
toxicology; such advances should be considered in the carbon fuels, paints and thinners, plants and fungi, and
context of general forensic toxicology, with particular medications that belong to other household members. 3 Poi
attention being directed at the various aspects of the sub soning appears to be frequent in this age group: according
ject that a re so mewhat unique to the paediatric age group. to the Toxic Exposure Surveillance System compiled by the
In this respect, this chapter focuses on several important American Association of Poison Control Centers, 50 per
areas of forensic toxicology that have undergone signifi cent of poisoning cases involved children under 6 years of
cant changes in the past decades, but which also are of par age. In contrast, fatalities in this age group are uncommon,
ticular interest and specific importance to paediatrici ans accounting for only about 4 per cent of all poisoning fatal
and pathologists involved in a medicolegal or forensic ities. 4 As a result, in any given year, the probability of a
practice. forensic pathologist or toxicologist encountering paediatric
After a brief insight into the major issues raised during patients in whom drugs or poisons can be detected is quite
the last few years in this specialty, this chapter will outline lowS Even in these cases, the exact role played by the
what the paediatrician and/or the forensic pathologist can drug(s) or substance(s) in the death of the child is not
expect to learn from toxicological investigations in the always clear; indeed , the presence of the drug or poison
light of the recent advances in analytical toxicology with may be categorized as either causing or contributing to
reference to drug testing, and the circumstances in which it death by direct toxic effect, or by idiosyncratic or hyper
may be app ropriate to initiate laboratory tests looking for sensitivity reaction, or through drug-induced disease.
drug use. The second peak of incidence for poisoning occurs in the
adolescent years, when drug ingestion is more common.
Teenagers are more likely to ingest dangerous substances
SCOPE OF THE PROBLEM intentionally and, in this respect, they generally choose
pharmaceutica l agents. 6 - 10 Within the teenage age group,
The age distribution of children involved in poisonings is girls are more likely than boys to ingest drugs. Agents
bimodaL I Ingestion of, or exposure to, toxic substances by involved in adolescent poisonings are often those found
children before the age of 5 is almost invariably accidental, in the family medicine cabinet, which include aspIrIn,
with the peak of instances occurring in the toddler years.2 paracetamol (acetaminophen), iron supplements (ferrous
Scope of the problem I 257
Table 13.1 Substances that are lethal in young children, even in small doses, as reported by Criddle lO
Table 13.2 Substances most frequently involved in fatal is associated with a high incidence of perinatal complica
exposures in adolescents (7 3- 79 years old) as reported by Criddle lO tions.12,13 Approximately 30 per cent of the substance
for 2004 addicted population in the USA is female, and most are
women of child-bearing age. Among pregnant women
1. Analgesics
aged 15-44 years, 4.0 per cent had used illicit drugs and
2. Sedatives, hypnotics, antipsychotics
1L8 per cent had used alcohol within 1 month of the 2006
3. Antidepressants
National Survey on Drug Use and Health. The complica
4. Stimulants and other street drugs
tions in pregnant women include an increased incidence of
5. Cardiovascular drugs
still birth related to abruptio placentae and placenta prae
6. Alcohol and volatile compounds
via; a reduction in birth weight, birth length and Apgar
7. Gases and fumes
scores; intra-uterine growth retardation; and pre-term
8. Anticonvulsants
birth. 14 - 16 After bilih, withdrawal symptoms including
9. Chemicals
hyperactivity, coarse tremor, poor feeding and poor weight
10. Muscle relaxants
gain are present in almost all infants of heroin-, cocaine
or amphetamine-addicted mothers. 17 Sudden infant death
syndrome (SIDS) is more common among infants of smok
ing mothers. IS, 19 In addition, children born to smoking
sulphate), antidepressant medications, non-steroidal anti mothers develop more slowly, both physically and men
inflammatory drugs (NSAlDs), antipsychotic drugs, drugs tally, through their teen years.20
of abuse and recreational drugs. Inhalants, generally Breast-feeding mothers constitute another group in
volatile hydrocarbons and carbon monoxide, must also be whom there may be a risk of infantile poisoning. Because
considered as well as caustic agents. In this age group, many drugs, especially lipophilic substances, are excreted
fatalities remain rare, with poisoning accounting for less into milk, infants are exposed to adverse effects of these
than 10 per cent of the total fatalities. 4 The 2006 annual drugs if breast fed. During lactation, multiple situations can
report of the Swiss Toxicological Information Centre arise that require maternal pharmacological treatment.
(www.toxi.ch) indicates that the highest number of caJis Generally, paracetamol (acetaminophen), low-dose aspirin
received for human poisoning involved children under (acetylsalicylic acid) (up to 100 mg/day) and short-term
5 years of age (41.4 per cent). However, the proportion of treatment with other NSAlDs, codeine, morphine and
cases with potential health risk was somewhat higher in propoxyphene are considered compatible with breast
adults (35.7 per cent) than in children (21.2 per cent). Boys feeding and lacking significant danger. 21 However, recently,
were more frequently represented among the children, and codeine intake in a breast-feeding mother resulted in a
women were more frequently represented among the ado baby's death. Genotype analysis for cytochrome P450 2D6
lescents and adults. Acute accidental intoxications repre (CYP2D6) indicated that the mother was classified as a
sented the largest group, with children ingesting easily codeine ultrarapid metabolizer. The clinical consequence
accessible household products, pharmaceuticals or plant was a relatively high morphine blood concentration in the
parts. Acute intentional poisoning was mostly due to baby resulting from metabolism of morphine, leading to
attempted suicide and less frequently to drug abuse or central nervous system depressant effect and neonatal
criminal behaviour. death. 22 Diazepam and its metabolite N-desmethyldiazepam
Small amounts of some drugs can be very toxic to enter breast milk. The accumulation of diazepam in breast
young children; one pill or one mouthful can require fed babies whose mothers are taking the drug may result in
aggressive treatment. Table 13.1 lists 12 toxic compounds infantile lethargy and weight 10ss.23 Drugs and substances
or classes of substances which are known as 'one pill can that require a careful assessment of risk before prescription
kill' substances. 10, II Table 13.2 lists the 10 leading sub to breast-feeding women have been published by Ito in
stances responsible for teen fatalities. 2000. The same author listed the main drugs that are pre
There is a growing concern in the medical community ferred for women who are breast-feeding. 24
over the maternal use of illicit drugs during pregnancy, Attention deficit hyperactivity disorder (ADHD) is the
since intra-uterine exposure to drugs of abuse and nicotine most common neurobehavioural disorder of childhood that
258 I Recent advances in paediatric toxicology
affects school-aged children, wi t h a prevalence ranging monoxide, CO) can endanger the life of other people.
between 4 per cent and 12 per cent of the general popula Infants and children are at greater risk of CO poisoning
tion. 25 The traditional therapeutic approach involves the than adults. Fetuses are particula rly vulnerable, because
use of psychostimulants such as methylphenidate. As ther maternal CO crosses the placenta. 37 The rapid identification
apeutic use of methylphenidate has increased, so the risk of of the toxic compound may also help in selecting the
accidental intake by to ddlers and of abuse, misuse or use in appropriate antidote, especially if it is potentially toxic. For
suicide attempts by pre-teenagers and adolescents has example, the standard antidote for acetaminophen (para
become greater. Many cases of poisoning or abuse have cetamol) toxicity is N-acetylcysteine (NAC) . Intravenous
been reported in the scientific 1iterature. 26 ,27 administration of NAC may induce anaphylactoid reac
The use of anabolic steroids among teenagers and tions. J8 In a few rare cases, administration of the false anti
young adults is becoming increasingly popul ar, particu dote may therefore increase the symptoms of toxicity.
larly among those involved in sports requiring great
strength, such as weight-lifting and body-building. 28,29 All
anabolic steroids are derivatives of the male hormone SPECIFICITY OF PAEDIATRIC TOXICOLOGY
testosterone and all are intended for 'tissue building' and to
increase masculini zing effects, Reported psychological Absorption of a drug from the gastrointestinal tract is
side-effects associated with their use include aggression, slower in infants than in older children an d adults.J9 The
psychosis and changes in libido, The real incidence of ana skin of infants is more permeable to drug diffusion than in
bolic steroid consumption is hard to evaluate, but the adults, and the blood-brain barrier of infants is also more
res ults of the National Household Survey on Drug Abuse readily penetrated. 4o Enzyme systems involved in drug
indicate that more than 1 million Americans are current or metabolism are usually functionally immature at birth,
former users.30 In Germany, the estimated number of juve leading to prolonged elimination half-times for drugs. The
nile users is about 100000.3J In the USA, 2 per cent of oxidation P450 and conjugation glucuronidation systems
college-aged men have been estimated to use anabolic require several weeks after birth to reach maturity. In
androgenic steroids. In anonymous US high school sur contrast, the sulphation pathway is more mature and may
veys, as many as 5-6 per cent of adolescent male athletes partially compensate for the deficiencies of the other meta
and 1-2 per cent of adolescent female athletes have bolic pathways41 For multiple enzyme families, a'develop
reported experimenting with anabolic steroids in conjunc mental switch' is observed that characterizes the transition
tion with weight training to improve sports perform between predominant fetal enzyme form to the predomi
ance. 32 ,33 In this context, forensic medicine is involved in nant adult enzyme form. For example, CYF3A7 expression
several different areas. The forensic pathologist examines dominates in the fetus whereas hepatic CYF3A4 expression
doping-associated deaths. The forensic psychiatrist and dominates in the adult. 42 Differences in pharmacokinetics
forensic physician deal with affective and mood disorders between newborn children and adults are presented by
including the results of aggressiveness and Criminality, and Soldin and Steele,43 Soldin and Soldin 44 and Alcorn and
problems of dependence on anabolic steroids. McNamara. 45 There are also important pharmacokinetic
It is perfectly legitimate to raise concerns about the use differences between children and adults that influence drug
fulness of toxicological investigations if poisoning is sus toxicity: (1) the fat compartment of children is generally
pected in a paediatric case. Indeed, recent literature has smaller, providing less storage for lipid-soluble drugs; (2)
confirmed the conclusion that, in most cases, neither com the propoliion of unbound drugs in the blood is higher in
prehensive toxicology nor drugs-of-abuse screening of children, indicating that more drugs are available to pro
poisoned patients significantly influences their clinical duce a pharmacological effect; and (3) the ratio of liver
management. 34 However, these analyses are useful for weight to total body weight in children is 50 per cent
diagnostic confirmation and are a prerequisite in certain higher at 2 years old and 30 per cent higher at 6 years old
forensic investigations, The utility of focused quantitative than for the adult. Drug biotransformation rates are therefore
serum assays to determine serum levels of particular enhanced in children and reduced in neonates compared
poisons is more helpful to paediatric clinicians. 35 In a retro with adults. In addition to pharmacokinetics differences,
spective study of all comprehensive emergency department pharmacodynamics responses also differ. 46
toxicology screenings performed in paediatric patients, 463 A final point concerns the marketing of therapeutic sub
cases were reviewed by Belson et al. 36 In this study, more stances. Only a small minority of approved drugs have
than 550 toxins were screened, Of 234 positive screening received adequate paediatric study in appropriate clinical
tests, only seven were positive without a documented sus trialsY Furthermore, a licensed formulation for the admin
picion of an exposure. In these cases, detection of these istration of certain pharmaceutical products to children is
specific drugs did not result in a change in medical man often also lacking.48 The situation could be even worse
agement a nd their presence did not affect the patient's with off-label/unlicensed drugs. Indeed, many cases of
clinical outcome. However, toxicological analyses are jus adverse events involving off-label/unlicensed prescriptions
tified, particularly if the suspected pOison (e.g. carbon have been reported. 49 To fill these gaps, new regulations
Techniques used in drug testing I 259
and patent protections have been issued for new drugs by IMMUNOASSAY TECHNIQUES
the US Food and Drug Administration, requiring specific
paediatric studies. Immunoassays are very useful to filter out the negative
Another point that deserves consid eration rel ates to samples from the positives, a nd thereby reduce the amount
the clinical research involving children and therapeutic of further analytical work required for toxicological inves
drugs. Several studies of drugs that bring only very minor tigation. 53 Immunoassay techniques are now the most
therapeutic benefits to neonates and children are ethically widely used screening pro ce dures in drug testing. Advan
questionable. 50 ,51 tages include high sensitivity, no extraction steps and
applicability to a large number of samples. Popular ver
sions of these tests are represented by radioimmunoassay
TECHNIQUES USED 11\1 DRUG TESTING (RIA). enzyme-multiplied immunoassay technique (EMIT),
enzyme-linked immunosorbent assay (ELISA) and fluores
Analysis of biological specimens is the necessary step that cent polarization immunoassay (FPIA),s4 These techniques
determines whether or not an individual has been intoxi use antibodies specific to the drug being assayed and a
cated with illegal drugs, with excessive amounts of prescrip labelled form of the same drug, The label itself may be a
tion drugs or with any other toxic substances. It is important radio active isotope (RIA). an active enzyme (EMIT and
to remember that toxicologists, especially in paediatric ELISA) or a fluorescent label (FPIA), which is incorporated
rather than adult cases, often do not know what to look for synthetically. A fixed quantity of antibody and labelled
and therefore have to take into account a vast number of drug are added to the test sample. The binding sites on the
toxicologically relevant substances, which makes the task antibody attract both the labelled drug type and the unla
difficult but very challenging. 52.10 belled drug type in the sample. The amount of labelled drug
bound is inversely proportional to the number of unla
belled drug molecules present. 53,55
Drug Screening In RIA, a rad ioactively labelled drug competes fo r the
same antibody binding site as the unlabelled drug. The ana
The technology of drug screening has greatly improved lytical measurement of radioactivity remaining in the solu
throughout the last two deca des. Idea lly, a screening test tion determines the amount of unlabelled drug in the
for a drug should be rapid (a short turnaround time). highly sample. RIA kits were ava ilable for several classes of drugs
specific (without false-positive results), highly sensitive of abuse but now have been virtually abandoned, being
(allowing the detection of a very low concentration of the replaced by non-radioactive immunoass ays. For instance,
drug in the specimen), reliable (same result from one lab o some larger companies stopped their marketing of most RIA
ratory to another for a given specimen), easily performed kits during the period 2004-2006.
technically and inexpensive. Since none of th e available In EMIT, ELISA and FPIA testing, the analytical measure
analytical technology meets all of these requirements per ment is based on an optically detected change, such as ultra
fectly, drug screening is usually achieved by means of the violet (UV) absorption, fluorescence or luminescence. These
combination of two different techniques; this is done in systems avoid the use of radioactive isotopes but also have
order either to extend the ambit of the screening test to a reduced sensitivity compared with RlAs because the optical
larger group of substances not detectable by the other signal is measured in the presence of the original biological
method or to confirm the result from another ana lysis. In fluid. An EMIT kit is available for opiates, barbiturates,
practice, screening tests are design ed for max imum sensi cocaine, amphetamines, benzodiazepines, methaqualone,
tivity at the expense of selectivity. Confirmatory analyses methadone, phencyclidine, cannabinoids and lysergic acid
are often just as sensitive (if not more so) but additionally diethylamide (LSD) . The advantages of EMIT as a screening
provide the requisit e specificity. It is important, however, to technique include its rap idity, its semiquantitation and its
point out that screening procedures must not give false ability to detect a number of drugs. Disadvantages include
negative results since this inva lidates the testing process, the high cost of the reagents and its susceptibility to both
False-positives at this stage are acceptable because all pre false-positives and false-negatives so that confirmation by
sumptive samples wiU subsequently be reanalysed by a an alternative methodology is recommended for positive and
confirmatory technique. 52 even for negative results. 56 Microplate ELISA tests are also
The methodologies most frequently employed for drug competitive immunoassays. They are speCifically optimized
screening are immuno assays and chromatography, and for use with a number of different biological matrices. Kits
sometimes capillary electrophoresis. Serum (or blood). exist for about 20 pharmaceuticals and 'drugs of abuse'.57 An
urin e and oral fluid (OF) (sa liva) can be analysed by these FPIA is both sensitive and specific; however, the equipment
techniques. The pharmacokinetics of distribution and eJ im is expensive and susceptible to sample alteration, affecting
inati on of dru gs are such that most are detectable in urine its accuracy. The application of the FPIA technique to foren
for a much longer time tha n in blood or OF so that urine is sic blood work and to alternative specimens such as hair or
usually the specimen of choice for screening. bile has been described .58 Additional antibody kits are also
260 I Recent advances in paediatric toxicology
available, expanding the uses of these systems. 55 New rapid clogging of the column, facilitat-ing sample prepara
immunoassays are available to test the OF from drivers at tion and analysis coupling and automation. 51 .52
the roadside and workers at the workplace. For instance, the
Cozaft® drug detection system (DDS) test is used to detect HIGH-PERFORMANCE LI QUID CHROMATO GRAPHY
impaired drivers suspected of driving under the influence of
psychoactive substances. DrugWipe® 5+, developed by The separation of non-volatile substances from each other
Securetec, is another test evaluated for the detection of psy or from other components of an extraction residue can be
choactive drugs in the oral fluid of suspected impaired driv obtained using HPLC. When a mixture of substances is
ers. Improvements are still needed for the similar detection injected onto the column, each component is partitioned
of cannabinoids and benzodiazepines. between the stationary phase (column) and the liquid
(mobile) phase. Molecules with greater affinity for the col
umn spend more time in that phase and, therefore, take
SAMPLE PREPARATION
longer to reach the detector. The time taken from injection
Sample preparation is an important step, which still to the peak maximum is known as the retention time. The
remains the most tedious and time-consuming part of an detector responds in direct proportion to the concentration
analytical method. Optimum sample preparation leads to of material passing through it, hence peak heights and area
enhanced sensitivity and selectivity while reducing the shown on the chromatogram are directly related to the
amounts of interfering matrix components and increasing concentration of each analyte.
analyte concentration. Sample preparation represents a Good sensitivity and high specificity, depending upon
major part of analysis that is capable of taking up to 80 per the detection system used, are shown by HPLC. The most
cent of the total time of a complete separation-based ana common detectors utilize absorption of UV light by the
lytical process, which typically encompasses five steps: drug. A diode array detector (DAD) will allow a full UV
sampling, sample preparation, separation, detection and spectrum of the analyte to be obtained. This can be com
data analysis. Major techniques rely on liquid-liquid, pared to a standard spectrum of the drug, and identifica
solid-phase and headspace extraction. Recent advances tion is then based on both retention time and UV
have been made in miniaturization, high throughput, selec spectrum. The identification of a very broad number of
tivity, integration and hyphenation and automation of unknown substances in blood specimens can be obtained
sample preparation. using HPLC-DAD in combination with a suitable spec
Liquid-liquid extraction (LLE) transfers the target ana trum. 53 .54 High selectivity and sensitivity for molecules
Iytes from a liquid matrix into another immiscible liquid with characteristic excitation and emission fluorescence
phase according to solubility difference. Solid-phase spectra is shown by HPLC coupled with fluorimetric detec
extraction (SPE) is used to extract and concentrate analytes tion. In this way, LSD and its metabolite nor-LSD can be
from a liquid matrix by partitioning the compounds quantified in blood by HPLC fluorescence detection down
between a solid and a liquid phase. An SPE procedure con to 20 pg/ml. 55
sists of four consecutive steps: column conditioning, sam Pre- and post-column reactions for increased sensitivity
ple loading, column washing, and elution of selected have also been described. Owing to its qualities, HPLC can
analytes. The headspace technique is dedicated to the be used to screen a sample for many drugs of abuse. The
extraction of volatile compounds. The analyte is parti drawbacks of HPLC include its expense, the high degree of
tioned between the sample and the gas phase in a closed expertise necessary for operation of the equipment, the
system. Then, the vial is pressurized and the headspace is need for sample extraction and the need for preparation of
sampled and injected into the gas chromatograph. In recent separate specific columns. 55 New filling column materials
years, selective extraction methods with molecularly have been marketed over the last few years. For example,
imprinted polymers (MIPs) and affinity columns with polar metabolites, such as ecgonine or hydroxyl metabo
bound antibodies have also been considered. Template lites of cocaine, can be analysed without prior derivatiza
molecules are used to produce MIPs by creating cavities in tion through hydrophilic interaction chromatography
a polymer that will recognize the target molecule. 59 The (HILlC).57 Similarly, the polar metabolites of morphine
MIPs somehow mimic natural antibodies. Miniaturiza (morphine 3- and 6-glucuronide) can be readily analysed
tion of the SPE methods resulted in the development of on reversed-phase columns, which are compatible with
solid-phase microextraction (SPME). Other materials cur almost 100 per cent water eluent. 58 More recent develop
rently under investigation are restricted-access materials ments in HPLC concern the use of sub-2 p.m particles and
(RAMs).5o The RAMs are used to exclude large molecules, mobile phases delivered at high flow rates. The recent com
such as proteins, and to extract low-molecular-mass ana mercialization of porous hybrid organic-inorganic silicon
Iytes by use of hydrophobic, ionic or affinity interactions, based paliicles with a narrow size distribution in the range
typically from complex blood matrix. They allow the direct of 1.7 pm has enabled a new level of performance, but only
injection of plasma or serum samples on to a high through the use of newly developed pumps that permit
performance liquid chromatography (HPLC) column without pumping and injection of liquids at pressures in excess of
Techniqu es used in drug testing I 261
10 000 pSi. Full implementation of such technology, termed for sepa ration of a large variety of substances, including
ul tra performance liquid chromatography (UPLC), is further those encountered in forensi c toxicology.76 Capillary elec
challenged by the requirement for new detectors fast trophoresis comprises a family of electrokinetic separation
enough to record a minimum number of data-points in a tech niques that separate co mpounds based upon differ
very narrow chromatographic peak (width of a few seconds ences in electrophoretic mobili ty, phase partitioning, iso
only).69.70 The use of UPLC columns speeds up analyti cal electric point, molecular size or a combination of one or
runs to times of as low as a few minutes for complex mix severa l of these propeliies. Electroki netic capillary methods
t ures. Another way to reduce the overall analysis time is to employed for forensi c toxicology include capillary zone
use monolithic supports, which co nsist of a continuous electrophoresis (CZE) and an electrokinetic capillary chro
porous silica rod. Separations with monolithic columns can matography technique called micellar electrokinetic capil
be performed at high flow-rat es, reducing the analysis time lary chromatography (MECC). In CZE, a few nanolitres of a
to less than 10 min.7! sample are applied to the beginning of a fused-silica capil
lary filled with buffer. On app lication of a high-voltage
direct current field, charged solutes begin to separate and
GAS CHROMATOGRAPHY
are swept through the capi llary by the combined action of
Gas chromatography (GC) is another method of separating electrophoresis and electro-os motic bulk flow and are
substances of analytical interest. 66 It is one of the most on-column detected towards the capillary end . In MECC,
efficient techniques available for separating drugs in body the buffer contains charged micelles (dodecy l sulphate
fluids. The separation is carried out on an analytical co l micelles), and uncharged and charged solutes separate on
umn containing a station ary phase (liquid or solid, depend the basis of differential partitioning between the micelles
ing on temperature), which is maintained at a given and the surrounding buffer and, if charged, also by differ
temperature inside an oven. The whole GC system com ential charge effects, including electrophoresis. In both
prises six components: gas supply and flow controllers, techniques, the most commo n detection principles app lied are
injector, oven, column, detector and recording device. In on-column absorbance and fluorescence. Using on-column
drug testing, GC capillary (rather than packed) co lumns are multiwave len gth detection, this technology is well suited
co mmonly used . A com pound is identified by matching its for toxicological drug scree ning. Compared with HPLC and
retention time with that of a drug standard under the same GC, CE has distinct advantages, including automation ,
cond itions. The use of more sens itive and selective detec small sample size, minimal sample preparation , use of very
tors than flame ionizat ion detectors, such as electron cap small amounts of organic solvents and low cost of capillary
ture and nitrogen-phosphorus detectors, has greatly columns. 77 ,78 Recent developments concern the co upling of
improved the reliabili ty of Gc. The method of choice for CE with tandem mass spectrometry and the analysis of
volatile compounds a nalysis is GC, which can be easily biofluids focused on less analysed matri ces, such as amni
coupled with mass spectrometlY detectors. Unfortunately, otic fluid, saliva, cerebrospinal fluid, sweat or airway sur
many drugs and poisons conta in polar functional groups face fluid and sputum. 79
and require chemica l derivatization to improve thermal
stability, volatility and detection. 72 Other dra wbacks include
the time necessary to prepare the specimen (extraction, Drug Confirmation
purification and derivatization), the expense of the equip
ment and the requisite expertise of GC technicians. There DlUg confirmatio n is based upon the use of two different
fore, GC is usu ally reserved for use as a screening and techniques, each confirming the other's results. In this
confirmatory technique in the expert laboratory setting. resp ect, the combination of two of the previously men
Fast GC is a new development in GC technology. Fast GC tioned techniques adds confidence in the accuracy of the
allows rapid analyses and reduction of retention time (up to resul t. 55,80
X 10) while main taini ng acceptable analyte reso lution. This On the other hand, the introduction of the mass spec
is made possible by reduction of column bore size, down to trometry detection method has revolutionized the analyti
100 !Lm, increasing the oven temperature ramp up to ca l tox ico logy in providing the capac ity of a formal
120· C/m in, and by using high-pressure ca rrier gas control. identification of unknown compounds. Gas chromatography
Because peaks a re very narrow, fast detectors and data sam mass spectrometry (GC-MS), combining GC with mass
pling are also required. Applications include analysis of spectrometry, has become the so - ca ll ed 'gold standard ' of
drugs of abuse in urine, 30 different drugs in oral fluid,73 forensic drug testing and doping ana lysis. 8o - 82 In the elec
benzodiazepines in blood 74 and cannabinoids in blood. 75 tron impact ionization mode, the sepa rated compounds are
bombarded with high-energy electrons, causing them to
break apart. The fragments produced are separated on the
CAPILLARY ELECTROPHORESIS
basis of their mass-charge ratio. Under the same condi
During the past decade, capillary electrophoresis (CE) tions, a molecule will fragment in exactly the same way
emerged as a promising, effective and economic approach evelY time, producing the sa me spectrum of fragm ents.
262 I Recent advances in paediatric toxico logy
This spectrum, along with the chromatographic retention determined, namely the calibration model, the accuracy
times, constitutes a virtual 'fingerprint' that is very specific and precision of the method, the low er limit of quantifica
for the particular chemical compound of interest. Detection tion and the limit of detection. 8s .B6 Participation in external
sensitivity may be as low as to measure the volume of the quality control programmes allows laboratOlY performance
drug in nanograms per millilitre. The definitive method for to be evaluatedY
drug identification is considered to be GC-MS.
Chemical ionization (CI), is a softer ionization method
that utilizes a charged reagent gas. Under these conditions SPECIAL TECHNIQUES FOR ANALYSIS OF
of ionization, fewer fragments are formed, increa sing the VOLATILE SUBSTANCES
chance of detecting the pseudomolecular ion. High-tlu'oughput
procedures mean that thousands of relevant toxic sub Epidemiological studies indicate that a substa ntial number
stances can be screened in a single procedure, i.e. the so of children and teenagers worldwide experiment with or
called systematic toxicological analysis (STA). abuse volatile substances with the intention of experienc
Urine still remains the standard specimen for compre ing a euphoric state of consciousness. A large range of
hensive screening, especially when a general screen for vo latile compounds are toxic, especially for young people.
unknown substa nces is required. To detect the largest num They can be classified in three broad classes: the hyd rocar
ber of xenobiotics , conjugated metabolites are first hydrol bons (e.g. butane), oxygenated compounds (e.g. butanol,
ysed and then derivatized prior to GC-MS analysis. amylnitrite) and halogenated compounds (e.g. halothane,
Other confirmatory techniques include HPLC coupled trichloroethylene). In case of suspicion of poisoning with
with mass spectrometry (LC-MS); this constitutes a straight volatile substances, proper sample collection, storage and
forward method for the analyses of polar or thermolabile handling are critical points to guarantee accurate toxico
compounds without a derivatization step as required in logical results. Samples should be taken rapidly and stored
GC-MS. Gas chromatography or HPLC coupled with tan in gas-tight, welI~sea led containers with minimal head
dem mass spectrometers (GC-MS-MS and LC-MS-MS) space. Storage, transport and handling of t he sample
enhance the selectivity and the sensitivity by a specific sec should always occur at low temperature (- 5°C up to 4°C).
ondary fragmentation of selected ions provided by the first The headspace technique is the most appropriate and
fragmentation. All these sophisticated techniques, includ popular method for the extraction of volatile substances in
ing GC-MS, can offer valuable information for the identi body fluids and tissues. Solid-phase microextraction, cryo
fication of unknown compounds. Using these hyphenated genic oven trapping, cryogenic focusing, and purge-and
chromatography and mass spectrometry methods also allows trap extractio n techniques are typical headspace methods.
acute or chronic poisonin g to be definitively excluded, which Usually, electron capture detection (ECD), flame ioniza
might be sometimes as important as the detection of a toxic tion detection (FlO) and mass spectrometry (MS) are used
substance. for detection and quantification purposes, while MS and
Time-of-flight mass spectrometers (qTOF-MS) accu Fourier transform infrared (FTIR) spectroscopy are used for
rately determine the mass-charge ratios of ions by measur identification. 88
ing flight time after acceleration in a vacuum tube by a
high voltage. qTOF-MS highly improves mass resolution
and accuracy allowing exact mass determination. Accurate ALTERNATIVE SPECIMENS FOR DRUG TESTING
mass measurement can be used for qualitative identifica
tion of unknown molecules. Pavlic et a18 ) used ES1-qTOF The rapid growth and development of drug testing technol
MS with mass spectral libraries to identify over 300 ogy has created a number of testing methodologies . that
different drugs. New and very expensive MS tools which can assess a broad range of biological specimens, including
are emerging in toxicology laboratories are Fourier trans besides urine and blood, hair, oral fluid, sweat, meconium
form ion cyclotron resonance mass spectrometers (FTMS) and amniotic fluid. 89
and OrbitrapT". These are now considered as the ultimate
instruments for high mass resolution a nd accuracy84
Urine
Method Validation For many reasons, urine is the most common matrix
analysed for drug testing. The specimen is easy to collect
Correct interpretation of toxicological findings is only pos and is not considered as an invasive sample to request.
sible provided reliable analytical data a re available. There Large volumes can often be collected , allowing extensive
fore, new analytical methods to be used in clinical and screening and storage for any further analyses by other
forensic toxicology require careful method development laboratories if additional expertise is needed. The matrix is
and thorough validation. For method validation, a whole one of the simplest that can be used, and it is the easiest to
bunch of analytical parameters must be considered and analyse compared with other fluids or tissues. Drugs and
Alternative specimens for drug testing I 263
metabolites are usually stabl e in frozen urine, allowing in infants. For instance, if the O-demethylation of dextro
long-term storage of positive samples. methorphan (reflecting CYF2D6 activity) appears to be well
Drugs are usually metabolized by the liver to form polar developed by 2 weeks post-natal age, the N-demethylation
metabolites that are conjugated before their elimination (reflecting CYFJA4 and possibly also CYP2B6 activities)
from the body via the urine. The metabolism of parent increases more slowly over the first year of life. 97
drugs (which are chemically non-polar) is relatively rapid All of these changes have a profound effect on the phar
compared with their urinary excretion. Conversely, the uri macokinetics of drugs across the paedi atric age range.
nary excretion of most metabolites (which usua lly are Because of the non-linear nature of these changes, simple
chemically polar) is more rapid and extensive tha n with the allometric scaling methods based on body weight or body
parent drugs. Drug metabolites are therefore frequently surface area often fail in the prediction of drug dosage,
found in urine in concentrations mu ch greater than the especially in neonates and infants.98 Several sca lin g mod
parent drugs. Therefore, many immuno assay urine screen els in predicting maintenance doses for children from those
ing assays for drugs of abuse utilize antibodies to drug used in adults have been suggested. 99
metabolites. Altered drug metabolism in children may result in adverse
Urine, however, can be easily adulterated since observed drug reactions because of inappropriate high-dose adminis
specim en collection is not a common practice. 90 ,91 Addi tration. Idiosyncratic adverse drug reactions may also occur.
tionally, most drugs are present in urine for only a few For instance, hepatotoxicity after valproate administration is
days after consumption so that only recent drug use will be much more frequent in children. The metabolism of va lproi c
detected. acid in children is increased, resulting in the formation of a
The determination of drug-to-creatinine ratios takes higher amo unt of a toxic metabolite. In some cases, children
account of the potential dilution of urine. Dilution can result are more resistant to drug toxicity than adults. Compared
from deliberate adulteration or through fluctuations in fluid with adults, children appear to be more resistant to the hepa
intake and elimination. Ulinary excretion profiles are nor totoxic effects of paracetamol (acetaminophen) overdose. Sul
malized and smoothed when reported to the creatinine con phation, whic h is more active in children, may diminish the
centration. New intake of drugs can be better predicted by formation of toxic metabolites by P450 enzymes. 2J Hence,
comparing creatine-normalized drug concentrations meas giving a dose extrapolated from adult dosage may result in
ured in two successive Ulinary voids 92 unexpected blood levels in paediabic patients. Therapeutic
ranges for so me drugs are quite different in adults and chil
dren, e.g. therapeutic theophylline concentrations are lower
Blood for neonates (5-15 mg/L) than for adults (10- 20 mg/L). 100
Protein binding can be quite different in neonates, producing
Blood is co nsidered to be an invasi ve sample. It represents vely different therapeutic ranges for highly bound drugs such
a complex matrix particularly if haemolysed. The detec tion as phenytoin.
window of the majority of drugs and metabolites to be The blood (pl asma) drug metabolite concentration (or the
tested is limited from a few minutes to several hours at ratio of parent drug and metabolite) can also be used to dif
most, and much less frequently days. However, the toxic ferentiate between acute and chronic drug administration.
effects or the pharmacological activities of drugs are gen In the case of acute drug administration, the blood (plasma)
erally correlated with their blood concentrations, which metabolite concentration in relation to parent drug concen
allows the effects of these levels on the living person to be tration can be helpful in determining the approximate
predicted, a nd thus the extent of involvement of the drug length of time between drug administration and death, or
in the death to be understood. 9J To assess the significance specimen collection at autopsy or in a living person.
of drug levels measured in blood, the concentrations are Neonates may also show different ratios of the pa rent drug
compared with ranges of therapeutic, toxic and fatal levels and metabolite, the hydroxyl metabolite of phenobarbitone
found in large databases. 94 ,95 These databases compile the and its glucuronide conj ugate being present in much higher
results of toxicological investigations of published case concentrations in neonates than in older subjects. 100
reports. They are mainly based on tox icology data obtained
from adult patients or fatalities. It is assumed that these
compilations can be used to interpret drug levels in blood Hair
specimens taken from young adults and teenagers,
However, extrapolation to the fetus, neonates and infants Detection of dru gs and their metabolites in hair has gained
is probably more hazardous and prone to error. In the much attention over the past decade. Drugs incorporated in
developing chi ld there are rapid changes occurring in terms hair remain in the keratin matrix for a long time, thus open
of organ maturation, changes in body composition and the ing a much wider window of detection than there is for
ontogeny of drug elimination,96 drugs in urine. 101-104 Reported drug recovery from the hair
Genotype and the temporal acquisition of drug bio of ancient Peruvian mummies suggests that hair analysis
transformation are criti cal determinants of a drug response may provide inform ation about drug use or exposure
264 I Recent advances in paediatric toxicology
almost indefinitely under the correct sample storage exceed the concentrations of metabolites even when both
conditions. l05 However, it has also been reported that a the parent drugs and their metabolites are no longer
small amount of the entrapped substances may slowly be detectable in plasma, blood or urine. 114 ,115 However, it is also
hydrolysed spontaneously. lOG possible to detect the metabolites of drugs of abuse in hair,
In addition to the promising diagnostic power of hair which may provide proof of drug ingestion. II G
analysis, hair sample collection is easy to perform without The finding of parent drugs in the hair may prove to be
the embarrassment commonly associated with urine collec an advantage in the use of hair testing by forensic toxicol
tion. Hair samples can be stored or transported without ogists. For example, hair may become the matrix of choice
refrigeration, pH control or the preserving agents that are when differentiation between heroin use and prescription
normally needed for other biological samples, such as use of morphine or codeine is required. IOI , IOJ
blood and urine. The finding of greater concentrations of parent drugs
Hair may also serve as a 'diary' of exposure, producing than metabolites in hair may result from several mecha
a drug history. 107 Scalp hair, in the average person, gener nisms. 1l 7 Parent drugs are generally less chemically polar,
ally grows approximately 1 cm per month ; hair that is 3 cm more lipophilic and better able to cross cell membranes
from the scalp would have been formed by the hair-forming than their metabolites. If substances enter the hair through
cells in the hair follicle approximately 3 months earlier. the hair-forming cells in the hair follicle, then the process
Therefore, it is theoretically probable that a drug incorpo would favour the incorporation of parent drugs. Increased
rated into the hair via the hair-forming cells in the hair lipophilicity may also favour parent drug secretion into
follicle would appear 3 cm from the scalp 3 months after sweat and sebum, where it could come into contact with
consumption. Unfortunately, the interpretation of drug pres the hair shaft distal to the hair-forming cells. In this
ence in hair is not that straightforward. Not all drugs found respect, Henderson et al 118 administered deuterium-labelled
in the hair may enter through the hair follicle. Drugs may cocaine to known cocaine users and found that in some
also enter the hair shaft via blood, sweat or sebum, or by subjects the drug moved in a distinct band while in others
diffusion from the skin surrounding the hair. 108 the labelled cocaine appeared first at a site more distal to
Hair pigmentation 109 and gender differences no also appear the scalp. These data suggest that in some individuals
to influence the concentration of drugs measured in hair. cocaine may be secreted rapidly into the sweat or sebum,
Several investigations suggest that basic drugs bind to and then deposited on the hair. It is also possible that the
melanin, which explains the higher concentrations gener components to which drugs bind in hair favour the binding
ally found in dark hair. On the other hand, levels of acidic of parent drugs over metabolites , as has been shown for
drugs and metabolites, such as ll-nor-9-carboxy-delta codeine, which binds to hair in vitro with a greater affinity
(9)-tetrahydrocannabinol, are unaffected by hair cOIOUr. III ,112 than morphine.
Our understanding about hair as a distribution site for Besides the polarity of drugs, it has been suggested that
drugs is still limited, and much research is necessary before other chemical characteristics should play an important
drug concentrations in hair can be interpreted with accu role for the incorporation of drugs into hair. In this respect,
racy.IJJ Although there is an increasing body of literature Nakahara et al 11 9 and Nakahara and Kikura 120 have shown
being developed about the detection of drugs of abuse in that structural changes that increase the lipophilicity of a
hair, there remain substantial and significant questions compound increase its incorporation into hair (for exam
about the interpretation of the analytical data. ple, an increase in the length of the carbon chain on the
Hair is somewhat unique among tissues in that nitrogen of amphetamine), whereas structural changes that
substances that are incorporated into hair may not have decrease a drug's basicity will reduce its incorporation. On
subsequent access to the blood, and are thus retained long the other hand, subtle structural changes that influence a
after the parent drugs and metabolites have been elimi drug's lipophilicity and ionization to a cation or anion at
nated from the body. Although hair (particularly hair bulb) physiological pH greatly affect the incorporation of the
does contain cytochrome P450 enzymes and has the ability parent drug and its metabolite into hair. At a given plasma
to metabolize drugs, it is considerably less able to do so concentration, drugs that form cations at physiological pH
than most other organs. 107 appear to be incorporated into hair at greater concentra
Substances may become permanently bound in hair in tions than drugs that form anions. The reason for this
the form in which they are distributed into the hair follicle. observation is not known, but evidence suggests that it
Cocaine appears to be an exception in that its hydrolysis to may be a result of the binding with melanin (see below).
benzoylecgonine is thought to occur in the hair shaft. 106 In addition to the impact of changes in the chemical
The blood supply to the hair follicle is extremely rich, and structure of drugs on their incorporation into hair, there is
the hair-forming cells of the hair follicle are among the most also the possibility that the structure of the hair (fine ver
rapidly dividing cells in the body. Drugs in the blood circula sus coarse) or chemical treatments that affect hair structure
tion will be delivered rapidly to the hair follicle and, if they are important for the incorporation of drugs. Blank and
are capable of crossing the cell membrane, will enter the hair Kidwell 12 1 demonstrated that the absorption of cocaine into
forming cells. Parent drug concentrations in hair generally cut hair specimens was greater for thick hair than for fine
Alternative specimens for drug testing I 265
hair. The hair specimens used in these experim ents were extraction medium. However, prior to the extraction step,
black (thick hair) and brown (fine hair), and it is possible removing external contamination is mand atory in order to
that the difference in pigmentation also influenced the avoid misinterpretin g positive results. The most frequent
absorption of cocaine. Cirimele et aj122 measured drug con contaminating factors include passive exposure to dust,
centrations in the hair of a female drug addict who had aerosol, smoke (when other people smoke drugs), drug
brown hair with strands of hair bleached with hydrogen powders (by handlin g drug stuff or any contaminated
peroxide. They found approximately threefold greater con material), and sweat and sebum, which carry drugs actively
centrations of cocaine and codeine in the brown hair than consumed by the subject. Many decontamination proce
in the bleached hair from the same person. dures have been proposed, but all involve washing steps
The role of pigmentation in influencing hair concentra using solvents, detergents and aqueous media, such as
t ions has been clearly demonstrated for several drugs such as buffer solutions, diluted acid solutions or distilled water
cocaine, codeine, methadon e or nicotine. For example, alone. 125
Joseph et al 123 showed that the specific binding of cocaine is The other fundam ental preparation phase is the isolation
157 times greater in female African black hair than in female of the drugs from the keratin matri x. In contrast to body
Caucasian blonde hair. Furthermore, these authors showed fluids, in which drugs and metabolites are dissolved and can
that bleaching black hair decreased the specific binding of be directly extracted and analysed, xenob iotics in hair are,
cocaine J3-fold. Nakahara et al l1 9 fo und a good correlation in fact, entrapped into th e solid keratin, Therefore, keratin
between the in vitro melanin affinity and the incorporation matrix must be digested, extracted, hydrolysed or dissolved
ratio for many drugs of abuse. Hold et al 124 showed that before drug separation and identification, Factors affecting
stanazolol, an anabolic steroid, is incorporated to a greater the choice of the most suitable procedure include the chem
extent in pigmented than in non-pigmented hair. ical structure of the drug, the system in use for particular
If these data clearly show that pigmentation is an dnlg detection, analytical recovery and time of analysis.
important factor in the incorporation of drugs into hair, it
must be pointed out that drugs are also incorporated into
the hair of albino animals who have non-pigmented hair DRUG IDENTIFICATI ON
owing to the absence of ty ros inase. Therefore, one cannot
Methodologies applied to drug detection and quantifica tio n
exclude the existence of a racial bias that could contribute
are simi lar to those usually carried out in laboratories of ana
to the difference in incorporatin g drugs into hair other
lytical toxicology.130 lmmunochemical and chromatographic
than by the degree of pigmen tation of hair. 1l3
methods coupled with mass spectrometry or tandem mass
Proced ures used in hair testing include a preliminary
spectrometry to enhance sensitivity and specificity are com
phase fo r specimen collection and storage, a sample prepa
monly adopted. 126 ,131 To date, more than 60 pharmaceuticals
ration phase for decontamination and drug isolation from
or drugs of abuse have been reported to be detectable in hair,
hair structu re, and a drug identification and quantification
including, besides the classical drugs of abuse (opiates,
phase by instrumental analysis. lOG ,125
cocaine, cannabis, amphetamines, methamphetamine, 3,4
Nails can be also used as an alternative to hair analysis
methylenedioxymethamphetamine [MDMA] 3,4-methylene
for long -term drug detection . It has been established that
dioxyamphetamine [MDA] and other designer drugs), opioids
drugs are incorporated into growing nails at levels similar
(i.e. semisynt hetic or synthetic morphine derivatives), hallu
to those 0 f hair. 126
cinogens, psychostimulants (including nicotine), barbiturates,
benzodiazepines, other sedatives (hypnotics, an tidepressants,
SPECIMEN COLLECTION neurol eptics), cardiovascular drugs, anti-infectious drugs and
other miscellaneo us compounds.67. 131,132
Hair samp les are generally collected by cutting, as near as
possible to the scalp, in the posterior vertex zone. In some
cases, different types of body hair, such as pubic, axillary
CUT- OFF VALUES
and beard hai r, can be used. 127 - 129 The amount of hair
needed for toxicological analysis is about 30-50 mg. If a Two main threshold values should be mentioned. The first
segmental hair analysis is required to evaluate drug use one concerns the analytical thresho ld, which relies on the
history, th e hair sample is generally cut into pieces of limits of detection and quantification of the chromato
}-1.5 cm in length, which represent about 1 month's graphic method used for drug analysis. A second cut-off
growth, Hair samples are to be stored at room t emperature value is used to decide whether an analytical result can be
in paper, plastic envelopes or in plastic or glass tubes. interpret ed with enough reliability with regards to the cir
cumstances of use or exposure. This second cut-off value is
always higher than the first one. A low cut-off value will
SAMPLE PREPARATION
be selected to demonstrate a single dru g expos ure while
Hair specimens are normally cut or pulverized to produce high cut-off values will be chosen in case of repetitive '
small fragments yielding more surface contact with the abuse to exclude external contamination, Several scientific
266 I Recent advances in paediatric toxicology
societies have put forward recommendations for cut-off and physiological effects. It can be anticipated that saliva
values used for interpretation of hair results {e.g. the SoHT testing for drugs will develop further in the near future,
or the Society of Hair Testing) . lOB with new domains of application being found as soon as
the mecbanisms by which drugs enter the saliva have been
c1 ari fied more speci fically. 140
Oral Fluid The marketing of new immuno assays (ElAs) specifica lly
designed for the detection of drugs in OF has enabled the fast
In recent years, OF has attracted much attention as a pos screening and selection of presumably positive samples.
sible alternative to urine for drug testing and, under certain Some of these immunoassay kits are commercially available
conditions, to plasma in the area of clinical investigation, with specific collection devices, e.g. Orasure® microplate,
therapeutic drug monitoring and the assessment of recent Intercept kits and Cozaft® microplate ElA. Confirmation of
drug use. IJJ - 136 Oral fluid is a complex biological matrix presumably positive results is generally performed with
consisting of the secretory products of salivary glands LC-MS methods. Atmospheric pressure chemical ionizat ion
(saliva) mixed with other fluids, substances and cellular (APCl) method is the preferred ionization method since it
debris that are present in the oral cavity. In fact, OF has considerably attenuates ion suppression effects due to OF
specific advantages over both urine and blood in being matrix components. 141
readily accessible for sampling. Furthermore , it constitutes
a non-invasive sample to collect. Drug concentration-time
profiles measured in OF are generally repolied to resemble Sweat
those determ ined in whole blood, although saliva-plasma
ratios may vary considerably depending on the drug prop Since 1911, it has been shown that drugs are excreted by
erties. As such, a positive OF drug test may indicate a the body in sweat, but no one has developed a practical
recent use, probable intoxication and, in some cases, psy solution to the problem of capturing sweat before testing
chobehavioural impairment, whereas urine testing cannot. until recently.142 Occlusive bandages consisting of one to
For these reasons, OF drug testing for forensic purposes has three layers of filter paper or pieces of cotton, gauze or
been shown to be reliable compared with other means of towel were proposed to collect sweat. By using these home
drug testing. 137 In this respect, it is noticeable that the lit made collectors, it was nevertheless possible to identify
erature on OF testing is expanding at a substantial rate, various drugs including quinine, salicylic acid, antipyrine,
which is indicative of the increasing interest in this unique ethanol, methadone, phenobarbitone, morphine, cocaine,
biological fluid. 7J ,'38 cannabinoids, methamphetamine and phencyclidine. 142 ,143
Testing OF for drugs can provide both qualitative and More recently, clothes have been shown to retain opiates
quan titat ive information about the drug status of an indi excreted in sweat, demonstrating that it is possible to
vidual. I39 Generally, drug concentrations in saliva are obtain evidence about the drug use status of the owners of
lower than those found ill urine or blood, and the major pieces of c10tbing (e.g. underwear),144
compound detected is the parent drug, not the metabolites. In practice, systematic collection of sweat specimens is
Initially, oral, intranasal or smoking routes may produce difficult bec ause of unequal distribution of sweat glands on
high concentrations of drugs in OF for several hours owing different pans of the body.'43 Approximately 50 per cent of
to local contamination of the oral cavity. Thereafter, the the total volume of sweat is produced by the trunk, 25 per
concentration of drug in OF is thought to reflect the free cent by the legs and 25 per cent by the head and upper
fraction of drug in blood. Thus, the m ajor disadvantage of extremities. Sweat is approximately 99 per cent water, tbe
saliva is that, after the contamination phase that follows most concent rated solute being sodium chloride. Sweat
local administration, many drugs are retained for a shorter production is irregular and is highly dependent upon an
period of time than they are in urine. Also, many drugs are individual's physical activity and emotional state and the
weak bases and saliva concentrations may be highly ambient temperature and humidity. Furthermore, sweat
dependent on pH conditions, which itself is dependent on collection methods used in most studies actually obtain a
salivary flow. These factors lead to highly variable saliva mixture of sweat and sebum. Drug transport in se bum has
plasma ratios for many drugs. not been examined thoroughly.145
Although highly sensitive methods of detection a re Significant advances have been made in recent years to
required, most drugs can be detected in salivary secretions. develop a sweat patch technology for the routine collection
Therefore, saliva testing is expected to offer many potent ial of sweat samples over an extended period of time. '46 The
applications in the general areas of drug screening and sweat patch collection device PharmChek T.", marketed by
forensic investigations. Saliva drug tests can reveal the Sudormed, Inc., consists of an adhesive layer on a thin
presence of a pharmacologically active drug in an individ transparent film of surgical dressing to which a rectangular
ual at the time of testing. Significant correlations have been absorbent pad is attached. The sweat patch acts as a speci
found between saliva concentrations of drugs, behavioural men container for non-volatile and liquid components of
Pitfalls and limitations of drug screens I 267
sweat, including drugs of abuse. Non-volatile substances Since meconium is often available in only sma ll amounts,
from the environm ent cannot penetrate the transp are nt screening methods require the development of sensitive a na
film, which is a semip ermeable membrane over the pad that lytical procedures and cl ea n efficient extraction tec hniques
allows oxygen, wa ter a nd carbon dio xi de to pass through for the characterization of drugs in this particular matrix. 152
the patch, leaving the skin underneath healthy. Over a In this respect, screening and confirmatory procedures by
period of several days, sweat saturates the pad and drugs immunoassays and GC-MS respectively have been reported
present in sweat are retained. The patch is generally worn for cocaine and its metabolites, 150, 151,158 cannab is,156 opiates,
over a period of 1-14 days, and drugs and metaboli tes including methado ne, 159 phencyciidine,157 cotinine, 160 oxy
accumul a te over this time period. codone l61 a nd benzodiazepines. 162
Patch testing includ es an extraction step followed by a n Generally, the routine anal ysis of meconium fo r drugs of
identification step us ing either immuno assay or GC-MS abuse is recommended for cases in which urin e cannot be
techniques. 146 The predo minant species fou nd in sweat are obtained from the newborn or urinalysis is nega tive for the
parent compounds, not drug metaboli tes. Analyses of substances despite a strong suspicion of matern al use of the
duplicate patches in controlled studies have revealed tha t substances during pregnancy. 149 Liquid chromatography
intra-subject variability is low but inter-subject variability tandem mass spectromehy methods are also ava ilable for
is high. 143 the comprehensive analysis of xenobiotics in meconium. 163
Advantages of this type of testing includ e non-invasive Likewise, a similar method has been presen ted for the
sampl e collection (the patch being worn on the back, biceps analysis of 10 amphetamine- , methamphetamine- and
or chest), ability to carry out normal activities including MDMA (ecstasy)-related an alytes in human meco nium. 164
swimming and showering without removing the patch, and Even at low intake, ethanol can cause adverse effects in
the ability to tell if a patch has been removed and reap newborns and later problems in childhood. Fatty acid ethyl
plied . 146,147 Thus, w hen applied to drug testing, the method esters (FAEEs) are fo rmed in the body by esterificatio n of
increases the window of drug detection to several days and ethanol with free fatty acids and transesterification of
up to several weeks. Furthermore, the test appears to be glycerides. They have been detected in human tissues dam
very sensitive since the administration of low doses of aged by ethanol abuse, and also in blood a nd hair. They
drugs such as cocaine or opiates produces detectable can be used as ma rkers for ethanol use. In this respect,
amounts in sweat after a few days. 147 By offering a cumu FAEEs in meconium have bee n reported to be potential bio
lative estimate of drug exposure over a period of several markers of feta l exposure to alcohol resulting from mater
days, this technology is particularly sui table in t he trea t nal alcohol consumption during pregnancy.16 5 According
ment and monitoring of substance abusers,148 when it is to Moore et al,166 a total FAEE concentration greater than
importan t for medical personnel to obtain information on 10 000 ng/g of meconium may indicate that the newborn
the behav iour and activities of these patients. Patches can has been ex posed to significant a mounts of alcohol during
be worn continuously and constitute a record of drug pregnancy.
intake during that period . Amniotic fl uid, umbilical cord t issue and cord blood
have also been suggested as alte rnative samples. 167-169
Amniotic fluid is present throug hou t gestation and is con
Meconium and Amniotic Fluid stantly diluted owing to fetal urin ation . Fetal swa ll owing
of a mniotic fluid contributes also to the recircul ation of
The determination of drugs of abuse in meconium for the drug metabolites through the fetu s. Many drugs have been
purpose of determining maternal drug use during preg de tected in amniotic fluid. 17o The major disadvan tage of
nancy is becoming increasing ly popular owing to its ability amniotic fluid testing is the difficulty and invasiveness of
to provide a larger historical record of drug exposure.149-154 its collection.
Meconium is a dark green mass of water, cells, mucus, sterol
precursors a nd bile pigments which is formed in the fetal
gut from swallowed amniotic fluid and sloughed gastroin PITFALLS AND LIMITATIONS OF DRUG
testinal epithelial cells from 16 weeks to birth; it is dis SCREENS
charged 1, 2 or J days after birth. 155-157
Drugs and their metabolites a re passed into the amniotic The rapid growth and developmen t of drug testing technol
fluid from fetal urine, and meconium is though t to reflect ogy has created a number of testing methodologies that
the swa llo wed a nd/o r sloughed sta ble metabolites. Analy can assess a range of biological s pecimens, not only to pro
sis of meconium may therefore reflect in utero exposure to vide evidence of recent drug use but also to indicate, as far
drugs. Therefore, the window of drug exposure that can be as it may be possibl e, the pattern of drug use (i.e. rou te, fre
detected by meconium testing is about 20 weeks , compared quency, dose and time of last use), the degree of impa ir
to a 2- to J - day window of de tec tion for a urin e drug abuse ment or the exten t of drug dependence of the individu a l.
screen. 158 Many of these specimens, especially in paediatric forensic
26 8 I Recent advances in paediatric toxicology
medicine a nd pathology, must be used with s pecial atten results for some drugs a nd false-positive results for others
tion bec ause of the potential for false-positive results depending on the type of immunoassay that might be
owing to the ir contamination during passive env ironmen used. 173 Finally, cross-reactive su bstances proved to pro
tal exposure. du ce false-positive drug screen ing include: 66
For the clinician or the pathologist, th e wide variation • Opiates - chlorprom az in e, codeine, dextrom ethorphan,
in the accuracy of drug tests, coupled with t he confusing dihydrocodeine, diph enoxylate, hydromorphone,
rubric 'toxic or drug screen', often leads to considerable levorphanol, meperidi ne, oxycodone, d-prop oxyphene.
confusion abo ut what, in fact, the test used is able to • Ampheta mines - cathinone, diethylpropion, dopamine,
detect. All toxic screen methods have in common, ho wever, ephedrine, p-hydroxya mpheta mine, isoxs upri ne,
a basic desig n th at max imizes sensitiv ity while making labata lol , methy lphenid ate, l-meta mpheta min e,
compromises in specificity. ny lidrin, phenmetraz ine, phentermine, pheny lep hrine,
Immuno assays ge nerally are less se nsitive than GC-MS phenylpropanolamine, propylhex edrine,
or LC-MS, a lthough all are capable of detecting very pseudoephedrine, ranitidine, sergiline,
low quan tities of drugs. The duratio n of a test positivity N-acetylprocain a mide, chloroquine, procainamide.
depends not only, in part, on the sensitivity of the test, • Barbiturates - gJuthetimide, phenytoin.
but also on the circumstances of abuse. Multiple doses • Cocaine (benzoylecgonine) - salicylates.
of a drug taken chronically may redistribute it to deep • Phencyclidine - am itripty line, chlorp romazine,
body compartments w ith slow release back into the blood dextromethorpha n , diphenhydramine, doxylamine,
comp artment, s ig nificantly prolonging the pharmacokinet meperidine, thioridazine.
ics of elimination. This is particularly true for cannabi • Cannabinoids (THC-COOH) - ibuprofen, promethazin ,
noids, methaqualone or phencycl idine. Therefore, issues ribo Ravin.
in dru g pharmacokinetics and test ing methodologies
can easily lea d to both false-positive and false-neg ative Although it may arguably represent genui ne exposure
tests, both with significant consequences. Addition all y, and therefore not a fa lse-positive result, environmental
forensic pathologists must bear in mind that after death exposu re to drugs, such as marijuana smoke, may produce
changes may occur due to post-mortem dru g redistribu tion a positive urin e specimen. Ho wever, it has been genera lly
or degradation. co ncluded that passive inhalation exposure, in the absence
The main pitfalls in forensic toxi co logy and interpreta of extreme conditions of ambient exposure, does not result
tion of ethanol analysis in post-mortem specimens as well in posi tive urine tests at conve n tiona l cut-off condi
as ana lytical pitfalls in hair have been reviewed by tions. 175 The problem of external contamination is more
Rich ardso n, 40 Kugelberg and Jones,I71 and Musshoff and crucial concerni ng hair testing. 172 ,)76 In situ ations in
Madea.172 which dru gs are know n to be present in the environment,
it is well known that passive exposure can produce posi
tive hair analysis resu lts. In this respect, apprecia ble levels
Causes of False-positive Drug Screens of nicotine have been reported in th e unwashed hair of
non-smokers which ca me into contact with the hair of
The two major causes of false - positive drug results concern smokers. 177 In contrast, cotinine (the ni cot in e metabolite)
immunoassay tests and contaminations by envi ronmenta l does appear to be a marker of tobacco use in the smoker
expos ure to drugs. All immuno assay tests are susceptible to population. 17B In a study condu cted by Smith and Kid
fa lse-pos itive results owing to a ntibody cross-reactivity well 179 on children living in a family in which cocaine
wi th substances bearing some structural simil arity to the (crack) was used, and thus present as smoke in the envi
drugs to be detected . 173.174 Included among these are poppy ronment , it has been shown that 85 per cent of the chil
seeds in t he diet (bread), which may conta in opium con dren tested positive for cocaine and benzoyl ecgonine in
ge ners resulting in a drug screen that is pos itive for opiates. their hair. In th is study, it was ass umed that children aged
Also com mon is the abi li ty of nasal decongestants , such as 3-10 were unlike ly to be self-admi nisterin g cocaine, so
ephedrine or phenylprop ano lam ine, to produce a uri ne that any cocaine in their hair must have come from pas
drug screen that is positive for amphetamin es. Newer sive exposure. This was confirmed by the fact that skin
immunoassays for amph etam ines have begun to use mono wipes obtained from the children by wiping their fore
clonal antibody assay that theoretically has a high er speci heads with a cotton swab were also positive for cocaine,
ficity for the amphetamine core. Even with the monoclonal indicating extensive surface contact, whereas saliva test
assay, however, cross-reactivity may still occur; phen ing was negative in most of the ch ild ren, showing th at
metrazine and I-ephedrine may produce a posit ive result. in gestion of coca ine was not likely a so urce of the cocaine
Dextromethorphan and diphenhydramine, commonly used in the hair.
over-the-counter agents, also have the potential to p roduce The problem of hair contam in at io n by environmental
a positive urine test resul t for phencyclidine. Household exp osure to a drug is more complex; experts disagree
and commercial adulterants can produce fa lse- negative about interpretation, with some claiming that external
Pitfalls and limitations of drug screens I 269
contamination can be removed by washing through several the wrong biological specimen is examined or if the wrong
kinds of decontamination procedures, while others contend technology is used. For example, drugs that have a large
that washing removes drugs from inside the hair shaft and does volume of distribution have correspondingly low serum or
not completely remove external contaminants. IBO Caution blood concentrations (cannabinoids or LS~). Immunological
until better decontamination procedures are established drug tests designed for urine specimens are therefore gen
should induce toxicologists to test only short hair lengths erally incapable of detecting these drugs in serum
from close to the scalp to limit environmental exposure. or blood.
Of course, such a limitation would reduce the potential Another pharmacokinetic characteristic, elimination
advantage of hair analysis to provide long-term drug use half-life, also has important implications for drug detection
history for an individual. Notwithstanding this contro capability. With drugs that have a short elimination half
versy, it seems reasonable to point out that not all positive life (e.g. cocaine) the parent compound may be undetectable
hair analysis results must be interpreted as due to passive in blood within 8 hours; however, cocaine metabolites
exposure. Certainly, most positive results are due to the may be detected in urine for several days after a significant
administration of drugs. Nevertheless, the above example exposure because of the drug excretory pattern. Another
illustrates that interpretation of data must take passive factor to be pointed out is that the pharmacokinetics of cer
exposure, especially in young children, into consideration tain drugs may vary according to age. In younger children
to determine the source of drugs in any case of a positive the elimination half-lives are shorter and clearances faster.
result. 176 For example, clearance of morphine reaches adult levels by
6 months of age. IB4 Chlorpromazine is metabolized 2-5
times more rapidly by children under 5 years old than by
Causes of False-negative Drug Screens adults. IBs Subsequently, as a general rule, urine remains the
best biological specimen for drug testing because it takes
Equally problematic in drug testing are situations in which advantage of the fact that kidneys are the primary excretory
a drug test fails to identify the adolescent who has, in fact, organs for most drugs.
been recently abusing psychoactive drugs and in whom a
drug test should be positive. The reasons for a false INTENTIONAL SPECIMEN ALTERATION OR ADULTERATION
negative test can be divided into three general categories:
technological shOitcomings, pharmacokinetic characteris Biological specimens, palticularly urine, can also be inten
tics and intentional specimen alteration or adulteration. tionally altered or adulterated to produce a false-negative
result. IB6 Methods available to accomplish this are numer
ous and varied, ranging from simple dilution of the col
TECHNOLOGICAL SHORTCOMINGS lected specimen to substitution of urine that was produced
One of the most common reasons for a negative toxic screen before any drug use occurred or a fluid that resembles
result, despite obvious drug abuse, is that the clinician or the urine. 90 ,IB7 Another commonly used method consists of
pathologist fails to recognize that the particular screen being drinking a large amount of fluid, or even using a diuretic,
used is incapable of detecting the substance in question. 52 For in order to reduce the concentration of drug in urine so
example, some chemicals abused by adolescents, notably sol that it falls below the detection threshold of the assay and
vents and other inhalants, cannot be detected by routine will produce a negative urine. However, adulteration of the
screening of either blood or urine by commercially available collected specimen with chemical agents is the method
immunoassays. Without prior knowledge of what the test chosen by many users because it requires little sophistica
specifically seeks, the diagnosis may be missed. Similarly, tion and can be easily accomplished in unobserved collec
although drug screens commonly identify all members of a tion conditions . Such substances include vinegar, lemon
drug family, structural differences in specific drugs within the juice, bleach, ammonia-based cleaner, crystalline drain
family may limit their detectability (e.g. fentanyl and its ana cleaner, non-ionic liquid hand soap, methanol, sodium
logues in an opiate screen, or clonazepam in a benzodi chloride, toilet bowl cleaner, ionic detergents and even
azepine screen). Finally, drug tests can be falsely negative whole blood anticoagulated with EOTA IBB Although well
because of poor quality control of the laboratory. Inaccuracy designed collection procedures can minimize the OppOItu
rates (false positivity rates as high as 1.7 per cent; false neg nity for sample adulteration and laboratory tests exist that
ativity rates ranging from 17 per cent to 31 per cent) have may detect certain types of adulterants (temperature, spe
been reported in some selies.IBI-IB3 cific gravity, pH measurement of the sample and measure
ment of urinary creatinine), no system is absolutely
fool-proof. IB9
Commercial adulterants are also available: some con
PHARMACOKINETIC CHARACTERISTICS
tain the fixative glutaraldehyde (UrinAid, ClearChoice). 190
The pharmacokinetic characteristics of drugs have an others contain strong inorganic acid (Amber- 13), such as
important influence on drug detectability, particularly if nitrite and chromate (Urine Luck), or they may contain
270 I Recent advances in paediatric tox 'icoiogy
enzymes, e.g. peroxidase or protease (papain).'91.192 and co-proxamol 217 have been reported. In contrast, other
On-site adulterant detection devices that assess the integlity case reports have determined that there is little evidence that
of urine specimens are also commercially available post-mortem redistribution occurs with tricyclic antidepres
(e.g. Adultacheck 4).193 sants,2lS trazadone 219 and zopiclone. 220
Similarly, drugs can be expected to diffuse from gastric
residue into blood, liver, lung and other nearby organs.
Post-mortem Changes Rats killed and then subjected to gastric instillation of
amitriptyline showed drug diffusion into the liver from
5 hours post administration at room temperature. 221 Over
POST-MORTEM REDISTRIBUTION OF DRUGS
time, highest concentrations were reached in liver lobes adja
The difference in drug concentration in blood samples from cent to the stomach, with significantly lower levels in the
different sites represents a major concern for the forensic right lobe. Similar studies on human cadavers used amitripty
toxicologist and, in general, blood concentrations after line, paracetamol (acetaminophen) , and lithium in quantities
death cannot be interpreted from a knowledge of concen representing 10 tablets of each drug. 222 After 48 hours at
trations in life. This is because, in life, drugs are distributed room temperature, drug diffusion affected the left lobe of the
to different tissues in the body and differences in concen liver and, to a lesser extent, the caudate lobe and variably the
tration between tissues can be built up and maintained. light lobe posteriorly (with the cadaver supine). In the lung,
Active processes, which are responsible for maintaining the left was more affected than the right and the base more
concentration differences, cease after death and concentra than the apex, although this was not always true.
tions can change dramatically. The changes in blood are Post-mortem diffusion of ethanol, facilitated by its high
not necessarily uniform. Concentrations in venous blood hydrosolubility, from the stomach into the blood and the
after death are expected to vary according to the tissues nearby organs has been shown to follow a similar pattern
whose blood they drain. For example, tricyclic antidepres to that observed with drugs 223
sants, narcotic analgesics, local anaesthetics and antihista It is apparent that drug diffusion (including ethanol) in
mines all accumulate in myocardial tissue and may be a cadaver is an important and complex phenomenon that
partially responsible for the observed rise and subsequent can affect a wide range of organs as well as blood . There
high levels found in heart blood post mortem. 194 In this fore, knowledge of sampling site and sampling technique is
respect, blood in the right ventricle drains from the liver a prerequisite for a valuable interpretation of analytical
and other body tissues and is likely to differ substantially results. To facilitate this, pathologists should formalize
from the blood in the left ventricle, which drains the their protocols for toxicological sampling at autopsy. Blood
lungs. The site and time-dependent variability of post samples should be obtained by needle puncture of the
mortem blood and tissues sampling, and the phenomenon external iliac or femoral vein. This may prove difficult in
of post-mortem redistribution have been documented and babies. A liver sample should be obtained from deep within
reviewed in the recent literaturel95-200 after having been the right lobe and any lung sampling should be taken from
described as 'a toxicological nightmare' by Pounder and the apex rather than the basal lobes.
Jones in 1990. 201
Several controlled animal studies as well as case reports
POST-MORTEM DEGRADATION OF DRUGS
have been published that address these issues. Studies with
acetaminophen (paracetamol) in rabbits demonstrated that Post-mortem degradation of drugs and pOisons is a process
post-mortem drug concentrations in blood increased sig that is little understood but may also significantly affect the
nificantly with time for central sampling sites compared interpretation of post-mortem toxicological results. First,
with peripheral blood. 202 The post-mortem redistribution of there is the delay that occurs before the sample has been
amitriptyline was demonstrated in rats, which suggested the taken, and then between sampling and initial presumptive
post-mortem drug release from lungs and other drug-rich drug screening or blood alcohol determination. Second, the
tissues into the blood. 203 Post-mortem redistribution has also subsequent confirmation may not be performed until the
been demonstrated in rats following administration of mor case goes to court for trial or may be done many days or
phine, 204 digoxin 205 and secobarbital. 206 Conversely, rats to weeks after the blood has been taken, especially when toxi
which cimetidine was given did not show statistically signi cological analyses are carried out by separate institutions or
ficant differences in post-mortem changes in cimetidine con at several laboratories each carrying out isolated tests. Usu
centrations in tissues compared with blood. 207 ally additives and preservatives have been added but not
Several human case reports have described similar find always, and samples have been stored at 4°C or at -20°C.
ings. Site-dependent post-mortem changes in blood and tis Degradation before analysis may occur as a result of chem
sues concentrations of digoxin, 20s cocaine,209 methadone 21O ical or physical decomposition owing to the instability of
and methamphetamine 2J' have been reported. Post-mortem the drug, which lacks the protective effect of being bound to
redistribution of chloroquine,212 dothiepin,213 amitripty plasma proteins. 224 For example, diazepam 225 and dilti
line,214 fluoxetine and norfluoxetine,2J5 tranylcypromine 216 azem 226 are susceptible to hydrolysis and flunitrazepam 227
Specific applications I 271
and LSD 228 were reported to significantly degrade within even though the predictive value of vitreous humour in
only 24 hours when exposed to sunlight. Chemical instabil estimating blood alcohol concentration appears somewhat
ity has been also attributed to the time-dependent loss of questiona ble. 24o ,241
drugs, such as chlorpromazine or dothiepin, after death 229
Other drugs, such as paracetamol (acetaminophen), salicy
lates, benzoylecgonine, cannabinoids, morphine, buprenor SPECIFIC APPLICATIONS
phine and amitriptyline, would be more or less stable. 230-232
Degradation may also derive from enzyme activities Drug Screening for Doping Agents
linked to bacterial development that takes place after death
and may continue in an unpreserved sample after collec The misuse of doping agents in sport is concerned with
tion. 233 Enzyme activities in the blood of healthy living substances registered on an annually updated list, 'The pro
persons are thought to be mainly restricted to esterases, hibited list. International sta ndard' (World Anti-doping
which can affect cocaine, procaine and other esters; this is Code, World Anti-Doping Agency (WADA)) . Depending on
the reason why an enzymatic inhibitor must be added as a the survey methods used, it has been estimated that doping
preservative to the blood sample. 224 may involve 3-5 per cent of adolescents. For instance,
After death, intestinal bacteria penetrate the intestinal studies carried out in eastern France with 1501 athletes
wall, enter the blood and lymph vessels and migrate further aged 15-19 years found that 4 per cent had used banned
throughout the body. At 25°C this transmigration can substances at least once in their Iife,242 The same study
occur as soon as 5 hours after death. The possible role of design and setting, i.e. self-questionnaire survey, were used
bacterial degradation has been assumed to explain, at least in a 4-year follow-up of doping prevalence among pre
partially, the continuous decrease in concentration in adolescent athletes,24J This study disclosed that 1.2 per cent
blood after death of many drugs such as cocaine or of the respondents entering the cohort and 3.0 per cent
methamphetamine. Bacteria have been shown also to be 4 years later reported that they had used doping substances
responsible for the post-mortem metabolic changes of at least once during the preceding 6 months. The main sub
many drugs, including the nitrobenzodiazepines (clon stances used as doping agents were salbutamol, cortico
azepam, nitrazepam and flunitrazepam) metabolized to their steroids and cannabis. Cannabis is one of the drugs that is
7 -ami no meta bolites. 224 ,234, 235 most frequently detected in the context of doping.244
With the ever-increasing number of dietary supple
ments used by athletes and the availability of numerous
POST-MORTEM FORMATION OF DRUGS
synthetic steroids and recombinant peptide hormones and
In contrast to the degradation of drugs post mortem, cer modulators, detection of performance-enhancing com
tain compounds may appear to increase in concentration pounds is increasingly difficult,245,246 and needs more and
after death or during storage in unpreserved containers,9] more sophisticated detection methods such as (fast) chro
These include cyanide, which is generated by the degrada matography coupled with tandem mass spectrometry247 or
tion of haemoglobin and ethanol. Ethanol, which tops the carbon isotope ratio determinat ion ,248
list of psychoactive substances encountered in post The paediatrician might suspect such use when the young
mortem toxicology, can be produced by post-mortem fer athlete presents with recent increase in weight or muscle
mentation of glucose, though this rarely ex ceeds 0.5 gOfc. mass, oedema, gynaecomastia in males, deepening of the
Ethanol can, of course, be both formed and destroyed by voice in females, new acneiform rashes, changes in behav
microbial activity236 A recent comprehensive review of iour (initability, labile temperament, depression), or a new
issues relating to the interpretation of ethanol concentra onset of chemical hepatitis. Careful and sensitive question
tions in post-mortem specimens was presented by Kugel ing, in a non-judgemental but knowledgeable manner, can
berg and Jones. llI A chemical preservative, such as sodium sometimes induce the adolescent to confess steroid use so
fluoride (1-2 per cent w/v), should be added to specimens that counselling can take place. 52 According to Dawson,249
intended for determination of ethanol, which should pre the patients who should concern the physician the most are
vent any further production of ethanol. The addition of an not elite athletes but the yo uth who are being increasingly
insufficient amount of fluoride (such as 2-3 mg/mL in drawn to the use of performance-enhancing drugs.
commercially available tubes) may result in very particular Many health problems may arise because of use and mis
cases in the in vitro formation of ethanol, reaching a con use of doping agents, such as intake of counterfeit products,
centration as high as 3.5 giL. 237 Distinction between drug-drug interaction, dermatological disorders, and cardio
ingested ethanol from that formed post mortem can be vascular, hepatic, genitourinary, musculoskeletal, neuropsy
made by the detection of ethylglucuronide, ethylphosphate chiatric, endocrine and haematological adverse side-effects,
and ethylsulphate, three speciflC markers of ethanol not to mention infection resulting from the use of contami
intake,2J8,2]9 nated needl es,249,250
If available, vitreous humour, which is more protected Over 100 an abolic steroids are widely available on
from bacterial contamination, can be used instead of blood , the black market in oral or injectable form. 251 Extensive
272 I Recent advances in paediatric toxicology
metabolic pathways as well as natural occurrence make the Sachs 103 reported the case of a child who had been
detection of these substances difficult to interpret. Positive brought to a clinic with symptoms of intoxication attributed
tests are usually confirmed by GC-MS. Oral steroids can be to clozapine. As clozapine was not freely available, the
detected as long as 4 weeks after last use. Normally, testos mother of the child was suspected of having intoxicated her
terone exists in the body in balance with epitestosterone child with the drug prescribed to her depressive sister. One
with ratios of between 1 and 2. Epitestosterone is a natu year before, a l-year-old boy belonging to the same family
rally occurring epimer of testosterone, thought to be of had been found dead. Owing to the absence of any obvious
extratesticular origin and having no biological activity. cause of death, the case was ruled as sudden infant death
A testosterone-epitestosterone (TIE) ratio greater than 6 (the syndrome. After exhumation, the child's hair, the only tis
threshold ratio used by the International Olympic Commit sue available for toxicological analysis, tested positive for
tee) is considered as evidence of exogenous androgenic clozapine, providing evidence of repeated administration of
steroid use. To further foil ingenious athletes from taking the drug to the child when he was alive. However, only the
exogenous epitestosterone to reduce their TIE ratios below qualitative result was of importance to the judge; as the
6, a maximum allowable urine concentration of 150 ng/mL child could not have been under antidepressive treatment,
epistestosterone has been established. 252 the mother was sentenced to 4 years only because she even
When to use urine testing remains controversial. tually admitted giving the drug to the child.
Although unannounced steroid testing at high-school ath Lewis et al 255 reported the usefulness of hair testing of
letic events would perhaps diminish such substance use, this children living with suspected drug users (crack) and con
testing is prohibitively expensive. Additionally, it would sidered to be at risk. Using the hair of the children to deter
require that a chain of custody be established, and such mine smoke exposure provided 'extra credibility' to the
testing might engender an issue of privacy and consent. child protective services and allowed them to remove chil
Furthermore, besides anabolic agents, peptide hormones, dren from dangerous households. Similar hair testing
including growth hormone, erythropoietin, adrenocorti applications have been reported by Smith and Kidwell. 179
cotropin (ACTH), chorionic gonadotrophin and their releas Strano-Rossi et al 256 described the application of hair
ing factors, are also widely used among athletes. Although analysis in a judicial case to document coercive heroin
detection of these peptides by immunoassay is well estab administration to a 5-year-old child who was admitted with
lished in the clinical laboratory, the acceptability of this overdose symptoms to an intensive care unit. The results of
method of analysis as definitive proof of administration has toxicological hair analysis were accepted by the court with
yet to be established in forensic toxicologY.246 However, the out criticism; only the possibility of the consumption of
unequivocal identification of peptide and protein analytes antitussive drugs was raised by the defendant's lawyers.
used for doping will very likely rely on complex hyphenated However, the presence of monoacetylmorphine (MAMl. the
mass spectrometry methods following very specific extrac specific metabolite of heroin, allowed the toxicologist to
tion and purification methods. In this respect, immuno refute this objection. The relative of the child was found
affinity purification will precede the analysis with tandem guilty and sentenced accordingly.
mass spectrometers, such as quadrupole-linear ion trap or In the same way, Huestis 254 quoted several American
linear ion trap-orbitrap analysers. 253 court decisions ordering that parents' hair be collected and
tested for drugs of abuse (cocaine and marijuana cases) to
determine the fitness of the custodial parent in child cus
Forensic Applications of Hair Analysis In tody disputes or to evaluate their degree of drug addiction
Children and Teenagers in order to determine the best interests of the child in adop
tion cases.
Owing to its ability to provide evidence of drug exposure as Ketamine is a rapid-acting dissociative anaesthetic. It
well as drug use, hair testing for drugs of abuse has been has been reported to be abused by an increasing number of
available for several years in courts and accepted according young people as a 'club drug', and is often distributed at
to the legal conditions of the country or state where trials are 'raves' and parties. Teenagers are the major abusers. Keta
pending. Of course, toxicologists are aware that no sample mine concentrations were found to range between 0.8 and
technique or specimen can provide answers to all toxicolog 92.3 nglmg in hair collected in entertainment places from
ical questions. More information is obtained with a variety 15 ketamine abusers. 257 Ketamine metabolites were also
of analytical approaches and different biological specimens. detected, indicating ingestion of the drug and excluding
However, hair analysis offers 'a unique perspective on the hypothesis of external contamination.
human drug use'254 by providing a wider window of drug Methylphenidate (sold as Ritaline®, Medikinet® or Con
detection and may offer advantages over other drug testing certa®) is used in the treatment of childhood ADHD.
methods in terms of accessibility, resistance to post-mortem Methylphenidate is hydrolysed in the body into ritalinic
degradation and stability of the entrapped substances in it. acid. Both the parent drug and its metabolites can be
The following examples are given to illustrate the potential detected in blood. However, only the parent compound is
value of hair drug test results in forensic practice. present in hair from treated individuals. As hair samples in
Specific applications I 273
children can be easily collected, in contrast to invasive respectively. Many other cases have gone to court having
blood sampling, determination of methylphenidate in hair similar circumstances of a baby dying suddenly and
should be an alternative to check compliance in a wider unexpectedly.
time window than if using blood. Methylphenidate can be Gamma-hydroxybutyric acid is presumed to be one of
measured in hair by LC-MS. 258 the most frequently used substances in drug-facilitated
Hair analysis can be very useful also in the case of sus crimes. Because of its very short half-life, GHB is particu
pected drug-facilitated crime or alleged sexual abuse of larly difficult to detect in blood and urine. Furthermore,
children and teenagers. Drugs generally involved in GHB is also an endogenous substance found in low con
such crimes are sedatives, hypnotics and anaesthetics. Alco centrations in body fluids. Nevertheless, detection of a
hol or drugs of abuse, such as cannabis, LSD, gamma single GHB exposure has been reported to be possible. 267
hydroxybutyrate (GHB) or ecstasy, are also used. lOB Because
the majority of cases are reported to the police after a few
days, the hair is very often the only specimen suitable for Determination of Fetal Exposure to Drugs of
drug detection. In cases of child abuse, the drug can be Abuse
administered once or chronically. A repeated administration
is easier to identify because drug concentrations are gener Maternal self-reported drug history has been shown to be
ally much higher than after single exposure. The analysis is unreliable, as many women who deny use during preg
more challenging with low-dosed drugs, such as buprenor nancy exhibit drug metabolites in their urine. 26B On the
phine. 259 In the case of a single application, very sensitive other hand, systematic urinalysis during pregnancy is
methods are required for drug detection, in concentrations hampered by the sholt elimination half-life of the drugs.
down to the sub-pg/mg hair range. This was made possible A negative result may be a result of deliberate abstinence for
through the introduction of new analytical techniques 2 60 several days before biomedical screening. A positive result
For instance, bromazepam (0.8-28 pg/mg) was measured by reflects only exposure during the preceding 1-3 days.
tandem mass spectrometry methods in three volunteers A common method of estimating the amount of fetal expo
after a single ingestion of 6 mg261.262 Recently, a sex sure to a drug could be provided by measuring the concen
offender has been charged with oral genital contact with an tration in the umbilical blood at birth. 169 The problem is
8-year-old girl. The events were reported by the girl, who tha t the presen ce of drug in umbilical blood reflects only a
experienced insomnia and speech disorders. Hairs were very recent exposure and lack of drug does not rule out
sampled 1 month later and bromazepam was detected in the fetal exposure for most of the pregnancy. Measurement of
hair segment corresponding to the timing of the reported drugs in neonatal urine presents the same disadvantages.
facts. 263 Evaluation of drug concentrations in the amniotic fluid
The usefulness of hair testing was demonstrated in measured during pregnancy or at delivery cannot provide
another drug-facilitated sexual assault case of a 9-year-old information on the duration and degree of fetal expo
girl. In that case, diphenhydramine, an antihistamine agent sure. 167 The same disadvantages are noted with the analy
with sedative effects, was discovered in successive hair sis of meconium, which is only a qualitative test at the
segments, indicating repeated surreptitious administration. moment of delivery.269
The perpetrator admitted charges at court relating to the Neonatal hair begins to form at approximately 6 months
abuse of young girls. 264 The same group reported the gestational age; a positive result indicates use during the
administration of a phenothiazine derivative with sedative last trimester. Hair testing can be collected during the first
properties (trimeprazine or alimemazine) to two children. trimester of life, after which time infant hair replaces
The stepmother, who was the perpetrator, did not challenge neonatal hair. Drugs in neonatal hair could originate from
in court the suggested lise of trimeprazine as a sedative deposition from fetal blood or from contamination of hair
drug. 26 5 by amniotic fluid . 170 Hair from the mother can be collected
Two deaths of babies sedated with methadone by their also for drug analysis and results can be compared with
mothers were reported by Kintz et al 266 Both mothers were those obtained from neonatal hair. Hair analysis may
former heroin addicts under methadone substitution ther therefore remedy the disadvantages previously mentioned
apy. At autopsy, no evidence of violence was noticed. Tox with the other methods, with a wide window of detection
icological investigations by GC-MS in hair and blood ranging from weeks to months, and may provide informa
demonstrated recent and repeated methadone exposure. tion concerning the severity and pattern of an individual's
Hair (6 cm) from the first 14-month-old baby tested posi drug use when a maternal drug history is not available or
(ive at 1.9 and 0.8 ng/mL for methadone and EDDP is in doubt.
(methadone metabolite) respectively. In another case, a Regarding the correlation of drugs in maternal and
comatose 5-month-old girl was taken to hospital, where neonatal hair, concentrations of nicotine and cotinine in
she was declared dead 1I days later. Hair analysis (5 cm) paired maternal and neonatal specimens were found to be
revealed the presence of methadone at a concentration of well correlatedYo Mothers and infants in the smoking
J.O and 21.3 ng/mL in the root and end hair segments groups, both active and passive, had significantly higher
274 I Recent advances in paediatric toxicology
concentrations of nicotine and cotin in e than did those in that several deaths had occurred before that thresh old was
the control, non-smoking group. Th is study documents reached in this particular instance. Each individual unex
the impoltance of hair ana lysis as a tool for measuring plained death occurring in a hospital environment requires
exposure to cigarette smoke. Indeed, passive smoking has careful investigation, and the careful and thorough exclu
been shown to adversely affect the health of infants and sion of poisoning279 (see Chapter 18).
children .271 Fabricated or induced illness by carers (FlI) was first
Prenatal cocaine and methamphetamine exposure is described in 1977 by Meadows, a paediatrici a n in Leeds .
now frequently tested because of their relatively high The condition was, at the time, referred to as Munchausen's
prevalence of use during pregnancy, especially in th e USA. syndrome by proxy280 Among other manifestations, ill
Gestational exposure has been associated with placental nesses are induced by carers of children (mostly females) by
abruption and premature labour, as well as with an ad ministering substances, smothering, withholding nutri
increase in low birth weight, microcephaly, congenital ents and medicines, and by other means, e.g. introducing
anomalies and necrotizing enterocolitis. Neonatal hair was infectious material into the gut or bloodstream, Commonly
found to be a suitable biological marker to demonstrate in used poisons include insulin, salt, antidepressants, anti
utero cocaine and methamphetamine ex posure. 272 ,27J Inter coagulants and bleach. These chi ldren may present as acute
estingly, median cocaine concentratio n was lO-fold higher life-threatening events, as unexplained deaths or as baffling
in the hair of the mothers than in the neonates. 274 clinical problems; the assistance of the tox icologist is of
In utero alcohol exposure also can be detected by hair great value in these instances.281 Occasionally, similar prob
an alysis. The retrospective detection of alcohol consu mp lems arise when others close to the child but not actual car
tion during pregnancy can be performed also by measuring ers, for a variety of potential reasons, induce illness in a
specific markers of alcohol abuse in the mother's hair. Fatty baby by administering therapeutic substances.
acid ethyl esters and ethylglucuronide, which are minor
metabolites of a lcohol , are suitable markers for the detec
tion of heavy alcohol consumption. They may be used in Toxicology and Sudden Infant Death Syndrome
instances of fetal alco hol syndrome or suspicion of regular
alcohol intake during pregnancy to objectiv ate alcohol Among the many hypotheses put forw a rd as explanat ions
abuse,275 for sudden infant death syndrome (SIDS), a number
Recently, analysis of vernix caseosa has been proposed involved poisoning, e.g. antimony leaching from cot mat
as an alternative specimen to hair when hair is not avail tresses. 282 Another hypothesis that the primalY cause of
able or is too difficult to collect. 276 Vernix caseosa is a SIDS is poisoning by toxic gases generated in cot mattresses
thick, white lipid and cell mixture that covers the fetus, was first publicized in the media in 1989 by Richardson, an
starting at about 24 weeks' gestational age. 171 Vernix can independent consultant on biodeterioration of materials. It
eaSily be removed from a newborn's skin with a piece of was first formally published in the Lancet in March 1990. 283
gauze. Cocaine and/o r three of its metabolites were The theory proposed was that a fungus (Scopulariopsis
detected in a 3-cm section of gauze containing verni x brevicaulis) sometimes found in the domestic environment
obtain ed from three out of five neonates. could degrade the chemical compounds of phosp horus,
arsenic and antimony that may be present in fire ret ardants
or plasticizers in PVC cot mattress covers and other cot fur
THE IMPORTANCE OF PAEDIATRIC nishings; this wou ld subsequently release the toxic gases
TOXICOLOGY IN SPECIFIC CASES phosphine, arsine and stibine 284 The hypothesis was inves
tigated by two Independent Expert Groups285,286 that found
Deliberate Poisoning by Health Professionals that Richardson's conclusions could not be substantiated by
independent researchers, and thus concluded in 1991 and
Beverley Gail Allitt, dubbed 'the Angel of Death', was an 1998 that the hypothesis was unfound ed.
English State Enrolled Nurse (SEN) who was convicted of
killing four children and injuring five others, in 1991, on
the children 's ward of Grantham and Kesteven Hospital, CONCLUSIONS AND FUTURE
Lincolnshire, w here she worked. 277 She has since become CONSIDERATIONS IN FORENSIC PAEDIATRIC
one of Britain's most notorious female serial killers. Her TOXICOLOGY
main method of murder was to inject the children under her
care, particularly those in intensive care, with insulin or Use of hyphenated mass spectrometlY methods has permit
potassium to ca use a cardiac arrest. 278 In an environment ted the simultaneous measurement of a large spectru m of
such as a children 's wa rd, where staff are dedicated to pro potentially toxic drugs. Increase in sensiti vity allows a
viding the best possible care, the threshold for recogni zing decrease in the size of the sampled specimens. The use of
covert acts of excess and inappropriate ad ministration of alternative biologica l matrices, such as neonatal hair,
therapeutic substances is high. It is not, therefore, surprising meconium, amniotic fluid or oral fluid, extends the field
References I 275
and time window of investigation. Howeve r, many short 15 Burkett G, Yasin SY, PaIO\'11 D, et al. Patterns of cocaine
co mings ex ist, especially regardi ng the post-mortem redis binging. Effect on pregnancy. Am J Obstet Gynecol 1994;
171 :372-9.
41 Rane A, Tomson G. Prenatal and neonatal drug metabolism a UV spectra library of 2682 co mpo und s. J Anal Toxicol
in man. Eur J Clin Pharmacol1980; 18:9-15. 2003 ; 27 :233 -42.
42 Hines RN. The ontogeny of drug metabolism enzymes a nd 64 Pragst F, Herzler M, Erxleben BT. Systematic toxicological
implications for adverse drug events. Phannacol Ther 2008; ana lysis by high-performance liquid chrom atograp hy with
118:250-67. diode array detectio n (HPLC- DAD). Clin Chem Lab Med
43 So ldin SJ, Steele BW. Mini-rev iew: therapeutic drug 2004; 42:1325-40.
monitoring in pediatrics. Clin Biochem 2000; 33 :333 -5. 65 Grobosch T, Lemm-Ahlers U. Immun oassay screenin g of
44 Sold in OP, Soldin SJ. Review : therapeutic drug monitoring lysergic acid die thylamid e (LSD) and its confirmation by
in pediatrics. Th er Drug Monit 2002 ; 24: 1-8. HPLC a nd fluorescence detection following LSD ImmunElute
45 Alco rn J, McNamara PJ. Pharmacokinetics in the newborn. extraction. J Anal Toxicol 2002; 26: 181-6.
Ad/) Dru g Deliv Rev 2003; 55:667-86. 66 Moore CM. Drug test ing in the 90s. Tn Leikin SB, Paloucek FP
46 Allega ert K, Verbesselt R, Naulaers G, et a J. Developmental (eds) Poisoning alld Toxicology Handbook, 2nd edn.
pharmacology: neonates are not just small ad ults. Acta Clin Chicago, IL:Lexi-Comp inc; Hudson (Cleveland) and
Belg 2008; 63:16-24. American Pharmaceutical Associ atio n, 1996, pp. 11 62 -75.
47 Ward RM, Benitz WE, Benjamin DK, Jr. , et al. Criteria 67 Giro ud C, Michaud K, Spo rkert F, et a l. A fata l overdose of
sup portin g the study of drugs in the newborn . Clin Th e!" cocain e asso ci ated with coin ges tion of marijuana,
2006; 28: 1385-98. bup ren orph ine, and fiu oxeti ne. Body flui d an d tissue
48 Johnson TN, Rostami-Hodjegan A, Goddard JM, et a J. distribution of coca ine and its metaboli tes determined by
Co ntribution of midazolam and its I-hydroxy metabolite to hydrophilic interact ion chro matograp hy- mass sp ectrometry
preoperative sedation in chi ldren : a pharmacokinetic (HILlC-MS). J Anal Toxicol2004; 28:464-74.
pharmacodynamic ana lysis . Br J Anoesth 2002; 89:428-37. 68 Dienes-Nagy A, Rivier L, Giroud C, et al. Method for
49 Cuzzolin L, Atzei A, Fan os V. Off-l abel and unlicensed quantification of morphine and its 3- and 6- glu curonid es,
prescribing for newborns a nd chi ldren in different settings: a codeine, codeine glucuronide and 6-m onoacetyJmorphine in
review of the literature and a consideration about drug human blood by liquid chromatography- electrospray mass
sa fety. Expert Opill Drug Safety 200 6; 5 :703- 18. spectrome try for routin e analysis in forens ic toxico logy.
50 Shamoo AE, Tauer CA. Ethica lly q ues tion able research with J Chromatogr A 1999 ; 854:109-18.
ch ildren: the fenfluramine study. Account Res 2002; 69 Swartz ME . UPLC: An introduction and review. Journal of
9: 143- 66. Liquid Chromatography Et Rela ted Teclmologies 2005;
51 Sharav VH. Children in clillical research: a conflict of moral 28: 1253-63.
values. A m J Bioeth 2003; 3:W 12-W59. 70 Churchwell MI, Twaddle NC, Meeker LR, Doerge DR.
52 Woolf AD, Shannon MW. Clinical toxicology for the Improving LC-MS sensitiv ity thro ugh increases in
pediatrician. Pediatr Clill North Am 1995 ; 42:317-33. chromatographic performance: co mparisons of UPLC
53 George S. Posi tio n of immunological techn iques in screening ES/MS/MS to HPLC-ES/MS/MS. J Chromatogr BAnal
in clinical toxicology. Clin Chem Lab Med 2004; Technol Biomed Life Sci 2005; 825 :JJ4-43.
42: J 288- 309. 71 Bugey A, Staub C. Appli ca tion of mon o lithic SUppOit S to
54 Miyaish i S. lmmunoassays : theolY and in strumenta tion, online extra ctio n and LC-MS analysis of benzodiazepines in
techniques, applicatio ns. In Townshend A (ed.) Encyclopedia whole blood samples. J Sep Sci 2007; 30:2967 -78.
Analytical Sciences. Volume 5. London: Academic Press, 72 Segura J, Ventura R, Jurado C. Derivatization procedures for
1995, pp. 2049-108. gas chromatographic-mass sp ectro met ric determination of
55 Montagne M, Pugh CB, Fink. JL. Testing for drug use, pan 1: xenob iot ics in biological samples, with spec ial attentio n to
analytical methods. Am J Hosp Pharm 1988; 45:1297-305. drugs of abuse and dop ing agents. J Chromatogr B BioMed
56 Baselt RC. Urin e drug screenin g by immunoass ay : Sci Applications 1998; 713:61- 90.
interpretation of results. In Baselt RC (ed.) Adl1allces in 73 Gunnar T, Ariniemi K, Lillsunde P. Validated toxicological
Analytical Toxicology. Foster City, CA: Biomedical determination of 30 drugs of abuse as optimized derivatives
Publications, 1984, pp. 81-123. in ora l fluid by long co lumn fast gas chro matogra phy/
57 Kolbrich EA, Kim I, Barnes AJ, et aJ. Cozart RapiScan Oral electron impact mass spect rometry. J Mass Spectrom 2005;
Fluid Drug Testing System: an evaluation of sensitivity, 40: 739-53.
specificity, and effic iency for cocaine detection com pared 74 Gun nar T, Ariniemi K, Lillsunde P. Fast gas chromatography
with ELISA and GC-MS following controlled cocai ne negative- ion chemica l ionization mass spectrometry with
administrati on. J Anal ToxicoI2003; 27:407-11. microscale volume sample preparation for the determination of
58 Gjerde H, Christophersen AS, Skuterud B, et al. Screening for benzodiazepines and alph a-hydroxy metabolites, zaleplon and
drugs in forensic blood samples using EMJT urine assays. zopiclone in whole blood. J Mass Spectrom 2006; 41:741-54.
Forensic Sci Int 1990; 44: 179-185. 75 Th omas A, Widmer C, Hopfgartner G, Stau b C. Fast gas
59 Pichon V. Selective sample treatment using molecularly chromatography and negative-ion chemical ionization tand em
imprinted polymers. J Chromatogr A 2007; 1152:41-53. mass spectrometry for forensic a nalysis of cannabinoids in
60 Souv era in S, Rudaz S, Veuthey JL. Restric ted access whole blood. J Pharl17 Biomed A nal 2007 ; 45:495- 503.
ma teria ls and large particle suppo rts for o n-line sa mple 76 Thormann W, Thormann W. Progress of capillary
preparation: an attractive approach for biological fluids electrophoresis in therapeu tic drug monitoring a nd clinical
an alys is. J Chro matogr BAnal Technol Biomed Life Sci and forensic toxicology. Th e?' Drug Manit 2002; 24: 222-3 1.
2004; 801:141-56. 77 Thormann W, Zhang C, Schmutz A. Capillary electrophoresis
61 Wille SM, Lamben WE. Recent developments in extraction for drug analysis in body fl uids. Ther Drug Monit 1996;
procedures relevant to analyti cal toxi cology. Ana l Bioana l 18:506-520.
Chern 2007; 388:JJ8J-9 1. 78 Vo n Heeren F, Thormann W. Capi llary electrophoresis in clinica l
62 Chen Y, Gu o Z, Wang X, Qiu C. Sample preparation. and forensic analysis. Electrophoresis J997; 18 :241 5- 26.
J Chromatogr A 2008; 1184: 19 1-21 9. 79 Lloyd DK. Cap illary electroph oresis ana lysis of biofluid s with
63 Herzler M, Herre S, Pragst F. Selecti vity of substa nce a focus on less commonly ana lyzed matrices. J Chromatogr
id entifi cati on by HPLC-DAD in toxicological ana lysis using BAnal Technol Bi0111ed Life Sci 2008; 866:154-66.
References I 277
80 Smith ML, Vorce SP, Holler JM, et al. Modern instrumental 103 Sachs H. Forens ic applications of hair analysis. In Kintz P.
methods in forensic toxicology. J Anal Toxicol 2007; (ed.) Drug Testing in Hair. Boca Raton, FL: CRC Press. 1996;
31 :237-53, 8A-9A. 211 - 222.
81 Maurer HH, Peters F1. Toward high-throughput drug 104 Kintz P, Villain M, Cirimele V. Determination of
screening using mass spectrometry. Ther Drug Monit 2005; trimeprazine-facilitated sedation in children by hair analysis.
27:686- 8. J Anal Toxicol 2006; 30:400-2.
82 Maurer HH. Hyphenated mass spectrometric techniques 105 Springfield AC, Cartmell LW, Aufderheide AC, et a1.
indispen sable tools in clinical and forensic toxicology and
Cocaine and metabo lites in the hair of ancient coca leaf
in doping control. J Mass Spectrom 2006; 41: 1399- 4 13.
chewers. Forensic Sci Int 1993; 63:269-275.
83 Pa vlic M, Libi se ller K, Oberacher H. Combined use of E51 106 Nakahara Y, Kikura R. Hair analysis for drugs of abuse. VII.
QqTOF- MS and E51-QqTOF-MS/MS w ith mass-spectral The incorporation rates of cocaine, benzoylecgonine and
library sea rch for qualitative analysis of drugs. Anal Biaanal ecgonine methyl ester into rat hair and hydrolysis of cocaine
C/wn 2006; 386:69-82. in rat hair. ArchilJes Toxicol1994; 68:54-59.
84 Ojanpera S, Pelander A, Pelzing M, et al. Isotopic pattern and 107 Rollins DE, Wilkins DG, Gygi SP, et al. Testing for drugs of
accurate mass determination in urine drug screening by abuse in hair. Experimental observations and indications for
liquid chromatography/time-of-flight mass spectrometry. future research. Forensic Sci Rev 1997; 9:23-36.
Rapid Commul1 iVIass Spectrom2006; 20:1161-7. 108 Pragst F, Balikova MA. State of the art in hair analysis for
85 Hartmann C, Smeyers - Verbeke J, Massali DL, McDowall RD. detection of drug and alcohol abuse. Clin Chim Acta 2006;
Validation of bioanalytical chromatographic methods. 370:17-49.
J Pharm Bio111ed Anal 1998; 17:193-218. 109 Gygi SP, Joseph RE, Cone EJ, et al. Incorporation of codeine
86 Peters fl, Drummer OH, Musshoff F. Validation of new
and metabolites into hair: role of pigmentation. Drug Metab
methods. Forensic Sci Int 2007; 165:216-24.
Dispos 1996 ; 24:495-501.
87 Pichini S, Ventura M, Pujadas M, et al. HAIRVEQ: an 110 Wilkins DG , Haughey HM , Krueger GG, Rollins DE.
external quality control scheme for drugs of abuse analysis Disposition of codeine in female human hair after multiple
in hair. Forensic Sci Int 2004; 145:109-15. dose administration. J A/wi Taxieol 1995; 19:492-8.
88 Wille SM, Lambeli WE. Volatile substance abuse - post
III Mieczkowski 1. Assessing the potential of a 'color effect' for
mOliem diagnosis. Forensic Sci Int 2004; 142:135-56.
hair analysis of 11-nor-9-carboxy-d el ta(9)
89 Gallardo E, Queiroz JA. The role of alternative specimens
tetrahydrocannabinol: analysis of a large sample of hair
in toxicological analysis. Biomed Chromatogr 2008; 22:
specimens. Life Sci 2003; 74:463-9.
795-821.
112 Mieczkowski T, Kruger M. Interpret ing the color effect of
90 Warner A. Interference of common household chemicals in melanin on cocaine and benzoylecgonine assays for hair
immunoassay methods for drugs of abuse. Ciill Chem 1989; analysis: brown and black samp les compared. J Forensic Leg
35:648-51. Med 2007; 14:7-15.
91 Jaffee WB, Trucco E, Levy S, Weiss RD. Is this urine really 113 Cone EJ , Joseph RE. The potential for bias in hair testing for
negative? A systematic review of tampering methods in drugs of abuse. In Kintz P (eel.) Drug Testing in Hair. Boca
urine drug screening and testing. J Subst Abuse Treat 2007; Raton, FL: CRC Press, 1996, pp. 69- 93.
33:33-42. 114 Goldberger BA, Caplan YH, Maguire T, Cone EJ. Testing
92 Fraser AD, WOlih D. Urinary excretion profiles of 1I-nor-9 human hair for drugs of abuse. N. Ident ificatio n of heroin
carboxy-delta9-tetrahydrocannabinol. Study III. A delta9 and 6-mo noacetylmorphine as indicators of heroine use.
THC-COOH to creatinine ratio study. Forensic Sci Int 2003; J Allal Toxicol 1991; 15:226-31.
137:196-202. 115 Harkey MR, Henderson GL, Zhou C. Simultaneous
93 Gottshalk LA, Cravey RH. Toxicological and Pathological
quantitation of cocaine and its major metabolites in human
Studies on PsychoactilJe Dnlg-inIJolved Deaths. Davis, CA:
hair by gas chromatography/chemical ionizat ion mass
Biomedical Publications, 1980.
spect rometry. J Anal Toxicol 1991; 15:260-65.
94 Druid H, Holmgren P. Compilations of therapeutic, toxic, and lJ6 Kintz P, Cirimele V, Mangin P. Testing human hair for
fatal concentrations of drugs. J Toxicol Clin TO.ricol 1998; cannabis. II. Identification ofTHC-COOH by GC-MS-NCI as a
36:133-4; author reply 5-6. unique proof. J Forensic Sci J995; 40:619-22.
95 Schulz M, Schmoldt A. Therapeutic and toxic blood
117 Henderson GL Mechanisms of drug in corporatio n into hair.
concentrations of more than 800 drugs a nd o ther
Forensic Sci Int 1993; 63:19-29.
xenobiotics. Pharmazie 2003; 58:447-74.
118 Henderson GL, Harkey MR, Zhou C, et a l. Incorporation of
96 Blake MJ, Castro L, Leeder J5, Kearns G L Ontogeny of drug isotopically-labeled cocaine and metabo lites into human
metabolizing enzymes in the neo nate. Semill Fetal Neonatal hair. 1. Dose-response relationships. J Analytical Toxicol
Med 2005; 10:123-38. 199 6; 20:1-12.
97 Blake MJ, Gaedigk A, Pearce RE, et al. Ontogeny of 119 Nakahara Y, K.i.k.ura R, Takahashi F. Hair analysis for drugs of
dextromethorphan 0- and N-demethylation in the first year abuse, X. Effect of physicochemical properties on incorporation
of life. Clin Pharmacol Ther 2007; 81 :510-6. rates into hair. Bioi Pharm Bull 1995; 18:1223-7.
98 Johnson TN. Modelling approach es to dose estimation in 120 Naka hara Y, Kikura R. Hair analysis for drugs of abuse. XIII.
children. Br J C/in Plwrmacol 2005; 59:663 - 9. Effect of struc tural factors on incorporation of drugs into
99 Johnson TN. The problems in scaling adult drug doses to hair: the incorporation rates of amphetamine analogs. Arch
children. Arch Dis Child 2008; 93:207-11. Toxicol 1996; 70:841-9.
100 Walson PD. Therapeutic drug monitoring in special 121 Blank DL, Kidwell DA. External contamination of hair by
populations. Clin Chem 1998; 44:415-9. cocaine: an issue in forensic interpretation. Forensic Sci Int
101 Moeller ME, Fey P, Sachs H. Hair anal.ysis as evidence in 1993 ; 63: 145-56.
forensic cases. Forensic Sci In! 1993 ; 63:43-53. 122 Cirimele V, Kintz P, Mangin. P. Drug concentrations in
102 Kintz P, Tracqui A, Mangin P. Detection of dnlgs in human human hair after bleaching. J Anal To.ricol 1995; 19:331-2.
hair for clinical and forensic applications. hIt J Legal Med 123 Jo seph R, Su TP, Cone EJ. In lIirro binding studies of drugs
1992 ; 105:1-4. to hair: influence of melanin and lipids on cocaine binding
278 I Recent advances in paediatric toxicology
to caucasoid and africoid hair. J Anal Toxicol 1996; 147 Burns M, Basel! RC. Mon itoring drug use with a sweat patch:
20:338-44. an experiment with coca ine. J Anal Toxicol 1995; 19:41-8.
124 Hold KJV!, Wilkins DG, Crouch DJ, et al. Derection of 148 Kintz P. Brenneisen R, Bundeli P, Mangin P. Swea t testing
stanozolol in hair by negative ion chemical ionization mass for heroin and metabolites in a heroin maintenance
spectrometry. J Anal To.ricol 1996; 20:345-49. program. Clin Ch em 1997; 43:736- 9.
125 Chiarotti M, Strano-Rossi S. Preparation of hair samp les for 149 Moriya F, Chan KJV!, Noguchi n, WU PYK. Testing for drugs
drug ana lysis. Forensi c Sci Rev 1996; 8:111-2 8. of abuse in meconium of newborn infants. J AHal Toxicol
126 Irving RC, Dickson SJ. The detection of sedatives in Ilair and 1994; 18:41- 5.
nail sa mpl es using tandem LC-MS-MS. Forensic Sci lilt 150 Lewis DE, Moore CM, Leikin JB, Koller A. M.econ ium
2007; 166:58-67. analysis for cocai ne: a va lidarion slUdy a nd comparison with
127 Mangin P, Kintz P. Variability of opiares concentrations in paired urin e ana lysis. J Anal Toxicol 1995; 19:148-50.
human hair according to their anatomical origin: head, 151 Oyler J, Darwin WD, Preston KL, et al. Coca ine disposition
axillary and pubic reg ions. Forensic Sci Jnt 1993; 63:77-83. in meconium from newborns of cocaine-abusing mothers
128 Offidani C, Strano-Rossi S, Chiarotti M. Dru g distribution in and urine of adult drug users. J Anlll Toxicol 1996;
the head, axillary and pubic hair of chronic addicts. Forensic 20:453-62.
Sci lnt 1993; 63:1105-0. 152 Moore C. Negrusz A, Lewis D. Determination of drugs of
129 Mangin P. Orug analysis in nonhead hair. In Kintz P (ed.) abuse in meco nium. J Chrol11lltogr B Biol11ed Sci Appl 1998;
Drug testing il1 hair. Boca Raton, FL: CRC Press, 1996, 713: 137-46.
pp. 279- 87. 153 Gareri J, Klein J, Koren G. Drugs of abuse testing in
130 Moeller MR, Eser HP. The ana lytical tools for Ilair resting. In meconium. Clill Chim Acta 2006; 366:101 -11.
Kintz P (ed.) Drug testing ill hair. Boca RalOn, FL: CRC Press, 154 Gareri J, Klein J, Koren G, et al. Drugs of abuse testing in
1996, pp. 95-120. meconium. Cli n Chim Acta 2006; 366:101-11.
131 Hegstad S, Khiabani HZ, KJ'istoffersen L, et al. Drug J 55 Browne S, Moore C. Negrusz A, et al. Detection of cocaine,
sc ree ning of hair by liquid chromatography-tandem mass l1orcocaine and cocaethylene in the meconium of premature
spectro merry. J Anal Toxical 2001l; 32:364-72. neonates. J Forensic Sci 1994; 39:1515-19.
132 Tracqui A. Unusu al drugs in hair. In Kintz P (eel.) Drug Testing 156 Moore C, Lewis D, Becker J, Leikin J. The determinarion of
in Hair. Boca Raton, fL: CRC Press, 1996, pp. 191-210. II-nor-69-tetrahydrocannabinol-9-carboxylic acid (THC
13 3 Idowu OR, Caddy B. A review of the use of sal iva in the COOH) in meconium. J Anal Toxicol 1996; 20:50-1.
forensic detection of drugs and other chemicals. J Forensic 157 Moore CM, Lewis DE, Leikin JB. The determination of
Soc 1982; 22:1 23 -235. phen cyclidine in meconium using ion trap mass
134 Caddy B. Saliva as a spec imen for drug analysis. In Baselt RC spectromelly. J Forensic Sci 1996; 41: 1057-9.
(ed.) Ad/J in Analytical To.ricology. Vol 1. fo ster City. CA: 150 Clark GD , Rosenzweig TB, Ra isys VA, et al. The analysis of
Biomedical Publi cations, 1984, pp. 198-254. cocaine and benzoyJecgonine in meconium. J Anal TO.ricol
135 Schramm W, Smith RH, Craig PA, Kidwell DA. Dnlgs of 1992 ; 16:261-3.
ab use in saliva: a review. J Anal Toxicol 1992; 16:1-9. 159 Stolk LM, Coenradie SM, Smit BJ. Analysis of methado ne
136 Dyer KR, Wilkinson C. The detection of illicit drugs in oral and its primary metabolite in meconium. J Anal Toxicol
fluid: another potential strategy to reduce illicit drug-related 1997 ; 21:154- 9.
harm. Drug Alcohol Re/J 2008; 27:99-107. 160 Dempsey D, Moore C, Deitermann D, et al. The detection of
137 Dams R, Choo RE, Lambert WE, et al. Oral fluid as an cotinin e in hydrolyzed meconium samples. Forensic Sci Jnt
alternative matrix to monilO r opiate and cocaine use in 1999 ; 102:167-71.
substance-abu se trea tment patients. Drug Alcohol Depend 161 Le NL, Reiter A, Toml illson K, er al. The detection of oxycoclo ne
2007; 87:258-67. in meconium specimens. J Anlll To.ricoI2005; 29: 54-7.
138 Samyn N, Laloup M, De Boeck G. Bioanalytical procedures 162 Williamson S, Jackson L, Skeoch C. et al. Determination of
for determination of drugs of abuse in oral fluid. Anal the prevalence of drug misuse by meconium ana lysis. Arch
Bioanal Chem 2007; 388: 1437-53. Dis Ch ild Fetal Neonatal Ed 2006; 91 :F291-2.
139 Cone EJ. Saliva test ing for drugs of abuse. Ann NY ACllri Sci 163 Xia Y, Wang P, Banlett MG, er al. An LC-MS - MS method for
1993 ; 694:91-127. the co mprehensive analysis of cocaine and cocaine
140 Kintz P, Cirimele V, Lue\es B. Codeine tesring in sweat and metabolites in meconi um . Anal Chem 2000; 72:764-71.
saliva with the drugwipe. lnt J Legal Med 1998; 111 :82- 4. 164 Kelly T, Gray TR, Hu estis MA. Development and va lidation
141 Dams R, Hu estis MA, Lambert WE, et al. Matrix effect in of a liquid chromatography-atmospheric pressure chemical
bio-analysis of illicit drugs with LC-MSjMS: influence of ionization-tandem mass spectrometry merhod for simultaneous
ionization type, sample preparati on, and biofluid. J Am Soc ana lysis of 10 amphetamine-, methamphetamine- and
Mass Sp ectrom 2003; 14: 1290-4. 3,4-methylenedioxym ethamp hetamine-related (MDMA)
142 Kintz P, Tracqui A, Jamey C, Mangin P. Detection of code ine ana lytes in human meconium. J Chrol11a togr BAnal Technol
and phenobarbital in swea r collected with a sweat patch. Biomed Life Sci 2001l; 867: 194-204.
J Anal Toxicol1996; 20:197-201. 165 Moo re CM, Lewis D. fatty acid ethyl esters in meconium:
143 Cone EJ, Hillsgrove MJ, Jenk.ins AJ, et al. Sweat testing for biomarkers for the detect ion of alcohol exposure in
heroin, cocaine and metabolites. J Alla l Toxicol 1994; neonates. Clin Chim Acta 2001; 312:235-7.
18:298-305. 166 Moore C, Jones J, Lewis D, Buchi K. Prevalence of fatty acid
144 Tracqui A, Kintz P, Ludes B, et al. The detection of opiate ethy l esters in meconium specimens. Clin Che l11 2003;
drugs in non traditional specimens (clothing): a report of ten 49: 133-6.
cases. J Forensic Sci 1995 ; 40:263-5. 167 Ripple M, Goldberger EA, Caplan YH, et al. Detection of
145 Dolan K, Rouen D, Kimber J. An overview of rhe use of cocaine and its metabolites in human amniotic fluid . J A lla l
urine, hair, sweat and saliva to detect drug use. Drug Alcohol To.ricoI1992; 16: 328 -31.
Rev 2004; 23:213-7. 168 Moore C, Browne S, Tebbett I, et al. Determination of
146 Fogerson R, Sc hoendorfer D, Fay J, Spiehler V. Qualitative coca ine and benzoylecgonine in human amn iotic fluid using
detection of opiates in sweat by RIA and GC-MS. high flow solid-phase exrraction columns and HPLC.
J Analytical Toxicol 1997; 21 :451-8. Forensic Sci Jnt 1992; 56:177-81.
References I 279
169 Winecker RE, Goldberger BA, Tebbett 1, et al. Detection of 194 Shepherd MF, Lake KD, Kamps MA. Postmortem changes and
cocaine and its metabolites in amniotic f1uid and umbilical pharmacokinetics: review of the literature and case report.
cord tissue. J Ana! To.rico!1997; 21:97~104. Ann Pharmacother 1992; 26:51O~14.
170 Gray T, Huestis M. Bioanalytical procedures for monitoring in 195 Hilberg T, Ragde S, Morland J. Postmortem drug
litera drug exposure. Ana! Bioanal Chem 2007; 388:1455~65. redistribution ~ human cases related to results in
171 Kugelberg FC, Jones AW. Interpreting results of ethanol experimental animals. J Forensic Sci 1999; 44:3~9.
analysis in postmortem specimens: a review of the literature. 196 Pelissier-Alicot AL, Gaulier JM, Champsaur P, Marquet P.
Forensic Sci Int 2007; 165:1O~29. Mechanisms underlying postmortem redistribution of drugs:
172 Musshoff F, Madea B. Analytical pitfalls in hair testing. Anal a review. J Anal To.ricol 2003; 2 7 :533~44.
Bioana! Chem 2007; 388:1475~94. 198 Drummer OH. Postmortem toxicology of drugs of abuse.
173 Liu RH. Comparison of common immunoassay kits for Forensic Sci Int 2004; 142:!oI~J3.
effective application in workplace drug analysis. Forensic Sci 198 Skopp G. Preanalytic aspects in postmortem toxicology.
Rev 1994; 6:20~57. Forensic Sci Int 2004; 142:75~!OO.
174 Green K, lsenschmid DS. Medical review officer interpretation 199 Yarema MC, Becker CEo Key concepts in postmortem drug
of urine drug test results. Forensic Sci Rev 1995; 7 :42~60. redistribution. C!il1 Toxicol (Phila) 2005; 43:235~41.
175 Cone EJ, Johnson RE, Darwin WD, et al. Passive inhalation 200 Drummer OH. Post-mortem toxicology. Forensic Sci Int
of marijuana smoke: urinalysis and room air levels of delta- 2007; 165:199~203.
9-tetrahydrocannabinol. J Anal To.rico! 1987; 11 :89~96. 201 Pounder DJ, Jones GR. Post-mortem drug redistribution ~ a
176 Kidwell DA, Blank DL. Environmental exposure ~ the toxicological nightmare. Forensic Sci Int 1990; 45:253~63.
stumbling block of hair testing. In Kintz P (ed.) Drug testing 202 Gomez HF, McKinney P, Philipps S, et al. Postmortem
in hair. Boca Raton, FL: CRC Press, 1996, pp. 17~68. acetaminophen pharmacokinetics: an experimental study of
177 Haley NJ, Hoffmann D. Analysis for nicotine and cotinine in site and time-dependent concentration changes. J Forensic
hair to determine cigarette smoker status. Clin Chem 1985; Sci 1995; 40:980~2.
31:1598~1600. 203 Hilberg T, Morland J, Bjorneboe A. Postmortem release of
178 Kintz P, Ludes B, Mangin P. Evaluation of nicotine and amitriptyline from the lungs: a mechanism of postmortem
cotinine in human hair. J Forensic Sci 1992; 37:72~6. redistribution. Forensic Sci Int 1994; 64:47~55.
179 Smith FP, Kidwell DA. Cocaine in hair, saliva, skin swabs 204 Koren G, KJein J. Postmortem redistribution of morphine in
and urine of cocaine users' children. Forensic Sci lilt 1996; rats. Ther Drug Manit 1992; 14:461~3.
83:179~189. 205 Koren G, MacLeod SM. Postmortem redistribution of digoxin
180 Blank DL, Kidwell DA. Decontamination procedures for in rats. J Forensic Sci 1985; 30:92~6.
drugs of abuse in hair: are they sufficient? Forensic Sci Int 206 Quatrehomme G, Bourre F, Liao Z, Ollier A. An experimental
1995; 70:13~38. methodology for the study of postmortem changes in toxic
181 Davis KH, Hawks RL, Blanke RV. Assessment of laboratory concentrations of drugs using secobarbital as an example.
quality in urine drug testing ~ a proficiency testing pilot J Forensic Sci 1994; 39:1300~4.
study. JAIVIA 1988; 260: 1749~54. 207 Imamura T, Nagat T, Kimura K, et al. Pharmacokinetics and
182 Kelly K. The accuracy and reliability of tests for drugs of postmortem changes of cimetidine in body tissues. Nippon
abuse in urine samples. Pharmacotherapy 1988; 8:263~75. Hoigaku Zasski 1994; 48:75~78.
183 Badia R, Segura J, Artola A, de la Torre R. Survey on drugs 208 Aderjan R, Mattern R. Validity of digoxin concentrations in
of-abuse testing in the European Union. J Ana! To.ricol blood determined postmortem. ZeitschriJl Jiir Reci1tsmedizin
1998; 22: 117~26. 1980; 86: !3~20.
184 McRorie TI, Lynn AM, Nespeca MK, et al. The maturation of 209 Hearn WL, Keran EE, Wei HA, Hime G. Site-dependent
morphine clearance and metabolism. Ani J Dis Child 1992; postmortem changes in blood cocaine concentrations.
146:972~6. J Forensic Sci 1991; 36:673~84.
185 Furlanut M, Benetello P, Baraldo M, et al. Chlorpromazine 210 Levine 8, Wu SC, Dixon A, Smialek JE. Site dependence of
disposition in relation to age in children. Clin postmortem blood methadone concentrations. Am J Forensic
Pharmacokinetics 1990; 18:329~3J. Med Patho11995; 16:97~!oO.
186 Mikkelsen SL, Ash KD. Adulterants causing false negatives 211 Miyazaki T, Kojima J, Yashiki M, et al. Site dependence of
in illicit drug testing. Clin Chem 1988; 34:2333~6. methamphetamine concentrations in blood samples collected
187 Cody JI, Schwarzhoff H. Impact of adulterants on RIA from cadavers of people who had been methamphetamine
analysis of urine for drugs of abuse. J Ana! Toxico! 1989; users. Am J Forensic Med Patho11993; 14:121~4.
13:277~284. 212 Kuhlman JJ, Mayers RW, Levine B, et al. Chloroquine
188 Schwarzhoff R, Cody JT. The effects of adulterating agents distribution in postmortem cases. J Forensic Sci 1991;
on FPlA analysis of urine for drugs of abuse. J Anal Toxicol 36: 1572~9.
1993; 17:14~17. 213 Pounder DJ, Hartley AK, Watmough P J. Postmortem
189 Soper JW, Canfield V. Description of a screening assay redistribution and degradation of dothiepin: human case
system to detect adulteration of urine samples. J Anal studies and an animal model. Am J Forensic jl;Ied Pathol
Toxicol 1996; 20:77. 1994; 15:231~5.
190 Goldberger BA, Caplan YH. Effect of glutaraldehyde 214 Pounder DJ, Owen V, Quigley C. Postmortem changes in
(UrinAid) on detection of abused drugs in urine by blood amitriptyline concentration. Am J Patho! 1994;
immunoassay. Clin Chem 1994; 40: I 605. 15:224~30.
191 Cody JT, Valtier S. Effects of Stealth adulterant on 215 Pohland RC, Bernhard NR. Postmoliem serum and tissue
immunoassay testing for drugs of abuse. J Anal To.ricol redistribution of f1uoxetine and norf1uoxetine in dogs
2001; 25:466~70. following oral administration of f1uoxetine hydrochloride
192 Burrows DL, Nicolaides A, Rice P J, et al. Papain: a novel (Prozac"). J Forensic Sci 1997; 42 :812~ 16.
urine adulterant. J Anal To.ricoI2005; 29:275~95. 2J 6 Yonemitsu K, Pounder DJ. Postmortem changes in blood
193 Pearce MR, Tarnai.LD. Performance evaluation of three tranylcypromine concentration: competing redistribution
on-site adulterant detection devices for urine specimens. and degradation effects. Forensic Sci Int 1993; 59:
J Anal Toxico! 2002; 26:464~70. 177~84.
280 I Recent advances in paediatric toxicology
217 Yonemitsu K, Pounder DJ. Postmortem toxicokinetics of chromatography-tandem mass spectrometly. Anal Chem
co-proxamol. Jnt J Legal Med 1992; 104:347-53. 2006; 78:5 884-92.
218 Hanzlick R. Postmortem tricyclic antid epressa nt 240 Cap lan YH, Levine B. Vitreous humor in the eva luation of
concentrations: letha l versus nonlethal levels. Am J Forensic postmoltem blood ethanol concentrations. J Anal To.ricol
Med Pathol 1989; 10:326-9. 1990; 14:305-7.
219 Martin A, Pounder DJ. Postmortem toxicokinetics of 241 Pounder DJ, Kuroda N. Vitreous alcohol is of limi ted value
trazadone. Forensic Sci ll7t 1992; 56:201-7. in predicting blood alcohol. Forensic Sci lnt 1994; 65:73-80.
220 Pounder DJ, Davies JI. Zopiclone poisoning: tissue 242 Laure P, Lecerf T, Friser A, Binsinger C. Drugs, recreational
distribution and potentia l for postmoltem diffusion. Forensic drug use and attitudes towards doping of high school
Sci Jnt 1994; 65:177-83. athl etes. Jnt J Sports lVIed 2004; 25:133 - 8.
22 1 Hilberg T, Bugge A, Beykich KJV!, et at. Diffusion as a 243 Laure P, Binsinger C. Doping prevalence among
mechanism of postmortem drug redistribution: an preadolescent athletes: a 4-year follow-up. Br J Sports Med
experimental study in rats. lut J Leg Med 1992; 105:87-91. 2007 ; 41 :660-3; discussion 3.
222 Pounder DJ, Fuke C, Cox DE, et al. Postmortem diffusion of 244 Sa ugy M, Avois L, Saudan C, et al. Cannabis and sport. Br
drugs from gastric residue. Am J Forensic Med Pat/IOI 1996; J Sports Med 2006; 40 (Suppl 1):i13-5.
17: 1-7. 245 Bowers LD. Analytical advances in detection of
223 Pounder DJ, Smith RW. Postmortem diffusion of alcohol performance-enhancing compounds. Clin Chem 1997;
from the stomach. Am J Forensic Med Pathol 1995; 43: 1299-304.
16:89-96. 246 Cowan DA, Kickman AT. Doping in sport: misuse, analytical
224 Skopp G, Potsch L, Kon ig I, Mattern R. A preliminalY study tests, and legal aspects. Clin Chem 1997; 43:1110-13.
on the stab ili ty of benzod iazepi nes in blood and plasma 247 Thevis M, Schanzer W. Mass spectrometry in sports drug
stored at 4' C.lnr J Leg Med 1998; 111:1-5. testing. Structure characterization and analytica l assays.
225 Mayer W, Erbe S, WolfG, Voigt R. Beitrage zur Analytik und Mass Spectrom Rev 2007; 26:79-107.
Stabilitat einiger pharmazeutisch interessanter 1,4 248 Saudan C, Baume N, Robinson N, et al. Testosterone and
Benzodiazepine. Pharl1lazie 1974; 29: 700-7. doping control. Br J Sports Med 2006; 40 (Suppl l):i21-4.
226 Koves EM, Lawrence K, Mayer JM. Stability of diltiazem in 249 Dawson RT. Drugs in sport - the role of the physician.
w ho le blood: forensic implications. J Forensic Sci 1997; J Endocl'illo/ 2001; 170:5 5-61.
43:587- 97. 250 Sjoqvist F, Garle M, Rane A. Use of dop in g agents ,
227 Benhamou-Batut F, Demotes-Mainard F, Labat L, Vincon G. palticularly anabolic steroids, in sports and society. Lancet
Determination of flunitrazepam in plasma by liquid 2008; 37 1:1872-82.
chromatography. J Pharm Biomed Anal 1994 ; 12:93 1- 6. 251 Musshoff F, Daldrup T, Ritsch M. Black market in a nabolic
228 Augsburger M, Mangin P. LSD, Ie phenix des hallucinogenes. steroids - analysis of illegally distributed products.
Toxicorama 1998; 10:61-6. J Forensic Sci 1997; 42:1119 - 25.
229 Ferner RE. Forensic Pharmacology. Oxford, NF: Oxford 252 Catlin DH, Hatton CK, Starcevic SH. Issues in detecting
University Press, 1996. abuse of xenobiotic anabolic steroids and testosterone by
230 Paterson S. Drugs and decomposition. Med, Sci and Law analysis of athletes' urine. Clin Chem 1997; 43: 1200-8.
1993; 33: 103- 109. 253 Thevis M, Thomas A, Schanzer W. Mass spectrometric
231 Hadidi FA, Ol iver JS. Stability of morphine and buprenorphine determination of insulin s and their degradation products in
in whole blood.lnt J Leg Med 1998; 111:165-7. sports drug testing. Mass Spectrom Rell 2008; 27:35-50.
232 McCurdy HH, Callahan LS, Williams RD. Studies on the 254 Huestis MA. Technical and legal aspects of drugs of abuse
stability and detection of cocaine, benzoylecgonine and testing in hair. In Nntz P (ed.) Drug testing ill hair. Boca
I I -nor-69-tetrahydrocannabinol-9-carboxylic acid in Raton, FL: CRC Press , 1996 , pp. 5-15.
who le blood using Abuscreen radioimmunoassay. J Forensic 255 Lewis D, Moore C, Morrissey p, Leikin J. Determination of
Sci 1989; 34:858-70. drug exposure using hair: application to child protective
233 Robertson MD , Drummer OH. Postmortem drug metabolism cases. Forensic Sci Jnt 1997; 84:123-8.
by bacteria. J Forensic Sci 1995 ; 40:382-6. 256 Strano-Rossi S, Offidani C, Chiaroll i M. Application of hair
234 Robertson MD, Drummer OH. Stability of analysis to document coercive heroin administra tion to a
nitrobenzodiazepines in postmortem blood. J Forensic Sci child. J Ana/ Toxico/1998; 22: 75-7 .
1998; 43 :5- 8. 257 Xiang P, Shen M, ZlllI o X. Hair analysis for ketam ine and its
235 Robertson MD, Drummer OH. Postmortem distribution and metabolites. Forensic Sci lnt 2006; 162: 131-4.
redistribution of nitrobenzodiazepines in man. J Forensic Sci 258 Marchei E, Munoz JA, Garcia-AJgar 0, et al. Development
1990; 43:9-13. and validation of a liquid chromatography-mass
236 Corry JEL. Possible sources of ethanol ante- and spectrometry assay for hair analySis of methylphenidate.
postmortem: its relationship to the biochemi stry and Forensic Sci Tnt 2008; 176:42-6.
microbiology of decomposition. J App! Bacterial 1978; 259 Kintz P, Villain M, Tracqui A, et al. Buprenorphine in drug
44: 1-56. facilitated sexual abuse: a fatal case involving a 14-year-old
237 Hoiseth G, Kristoffersen L, Larssen B, et al. In uitro boy. J Anal Toxicol2003; 27:527-9.
formation of ethanol in autopsy samples containing fluoride 260 Negrusz A, Gaensslen RE. Analytica l developments in
ions. lnt J Legal Med 2008; 122:63-6. toxicological investigation of drug-facilitated sexual assault.
238 Schm itt G, Adeljan R, Keller T, Wu M. Ethy l g lu curonide: an Anal Bioanal Chem 2003; 376:1192-7.
unusual ethan ol metabolite in hum ans. Synthesis, analytical 261 Cheze M, Villain M, Pepin G. Determination of bromazepam,
data, a nd determination in serum and urine. J Anal Toxicol c!onazepam and metabolites after a single intake in urine
1995 ; 19:9 /-4. and ha ir by LC-MS/ MS. Application to forensic cases of drug
239 Bicker W, Lammerhofer M, Keller T, et a l. Validated method facilitated crimes. Forensic Sci Jil t 2004; 145:123-30.
for the determination of the ethanol consumption markers 262 Deveaux M, Cheze M, Pepin G. The role of liquid
ethyl glucuronide, ethyl phosphate, and ethyl sulfate in chromatography-tandem mass spectro metry (LC-MS/MS) to
human urine by reversed-phase/weak anion exchange liquid test blood and urine samp les for the tox icologica l
References I 281
in ves tigation of drug-facilitated crimes. Ther Drug Manit 274 Garcia-Bournissen F, Rokach B, Karaskov T, Koren G.
2008; 30:225-8. Cocaine detection in maternal and neonatal hair:
263 Rey-Sa lmon C, pepin G. (Drug-facilitated crime and sexual implic ations to fetal toxicology. Ther Drug Manit 2007;
abuse: a pediatric observation]. Arch Pediatr 2007; 29:71- 6.
14: 13J 8-20. 27 5 Pragst F, Yegles M. Determination of fatty ac id ethyl esters
264 Kintz P, Evans J, Vill a in M, et a1. Hair analysis for (FAEE) and ethyl glucuronide (EtG) in hair: a promising way
diphenhyd ramine after surreptitious admin istration to a for retrospective detection of alcohol abuse during
child. Forensic Sci 1nt 2007; 173: 171-4. pregnan cy? Ther Drug Manit 2008; 30:2 55-63.
26 5 Kintz P, Villain M, Cirimele V. Hair ana lysis for drug 276 Moore C, Dempsey D, Deitermann D, et aJ. Feta l cocaine
detection. Thn Drug Manit 2006 ; 28:442-6. exposure: analysis of vernix caseosa. ] Anal Toxicol 1996;
266 Kin tz P, Vi llain M, Dumestre-Toulet V, et a J. Methadone as a 20:509-11.
chemical weapo n: two fatal cases involv ing babies. TIler 277 Marks V, Richmond C. Beverly Allitt: the nurse who killed
Drug Manit 2005; 27:741-3. babies.] R Soc Med 20 08; 101 :110-5.
267 Kintz p, Cirimele V, Jamey C, Ludes B. Testing for GHB in 278 Dyer C. Children's nurse convicted of murd er. EM] 1993;
hair by GCjMSjMS after a single ex posure. Application to 306:1 43 1-2.
document sexual assaul t.] Forensic Sci 2003; 48:19 5-200. 279 Appleyard WJ. Murder in the NHS. EM] 1994; 308:287-8.
268 Ostrea EM, Chavez CS. Perinatal problems (excluding 280 Repper J. Munchausen syndrome by proxy in health care
neonatal withdrawal) in materna l drug addiction : a study of workers.] Adll Nlirs 1995; 2 1 :299-304.
830 cases.] Pediatr 1979 ; 94:292 - 5. 28 1 Jones DPH, Boots CN. Factitious lJIness by Proxy. In David TJ
269 Ostrea EM, Brady M, Gause S, et a J. Drug sc reen in g in (ed.) Recel1t AdIJances in Paediatrics J 7. London : Churchi ll
newbo rns by meconium ana lysis: a large-scale prospective, Li vingstone, 1999, pp. 57-71
epid emiologic study. ] Pediatr 1992; 89: 107 - 13. 282 Jenkins RO, Craig P J , Goessler W, Irgolic KJ. Antimony
270 Klein J, Koren G. Hair analysis - a biological marker for leaching from cot mattresses and sudden infant death
passive smok ing in pregnan cy and childhood. Hum Exp syndrome (SJDS). Hum E.rp Toxicol 1998; 17 :138- 9.
Toxicol 1999; 18 :279-8 2. 283 Richardson BA. Cot mattress biodeterioration and SJDS.
271 Kukla L, Hruba D, Tyrlik M. Trends in respiratory morbidi ty Lal1cet 19 90; 335:670.
of children in relation to their passive smoking exposure. 284 Richardson BA. Sudden infant death syndrome: a possible
Cent Eur] Public Health 2006; 14:180- 5. primary cause. ] Forensic Sci Soc 1994; 34: 199- 204.
272 Bar-Oz B, Klein J , Karasko v T, Koren G. Comparison of 285 Department of Health. Expert Group to In ves tigate Cot Delllh
meco nium and neonatal hair ana lysis for detection of Theories: To.ric Gas Hypothesis. Chairman, Lady Limerick.
gestational exposure to drugs of abuse. Arch Dis Ci1ild Fetal Final report. London: Department of Health, 1998.
Neonata l Ed 2003; 88:F98-FIOO. 286 Report of the expert working group enquiring into the
273 Garcia-BoUl'nissen F, Rokach B, Karaskov T, Koren G. hypothesis that toxic gases evolved from matress covers and
Methamphetamine detection in maternal and neonatal hair: cot matresses are a ca use of SIDS. Sudden infallt death
impli cations for fetal safety. Arch Dis Child Fetal Neol1a tal syndrome (STD S) London: Her Majesty's Sta tionery Office,
Ed 2007; 92.F35 1-5. 1991.
I CHAPTER 14 I
HEAD AND NECK INJURIES
DEFINITION
Yager et al 9 assessed the interobserver variability for six
different coma scales used in children an d found tha t
Head injury can be defined as the injury resulting from an the paediatric coma scale had the highest interobserver
external force to the head by penetrating, compression or agreement. In an attempt to standardi ze the way coma is
impact forces, causing damage to the scalp, skull or brain measured in children, the British Paediatric Neurology
(traumatic brain injury). If brain injury results, there is usu Association (BPNA) has recommended the Jam es ad apta
ally an impairment of consciousness and a period of retro tion of the GCS as it takes into account the developmental
grade amnesia. Not all cases of brain damage are associated; immaturity in small chi ldren, uses the sa me number of
however, with a loss of consciousness (e.g. some frontal lobe points irrespective of the child 's age, and is simple to use.
injuries, extradural haematomas and some penetrating an d As the co nsciousn ess level decreases, vital mechanisms
compression injuri es) but, conversely, all cases of loss of such as airway protective reflexes (gag, cough and swa llow
consciousness indi cate brain injury. reflexes) may be compromised, m aki ng the patient vulner
Traumatic brain injury in children may result from able to aspiration 10 and secondary hypoxic ischaem ic
traumatic birth injury, accidental head injury and non injury. The Ad va nced Paediatric Life Suppo rt Guideline
accidental head injury (NAHI). suggests that all head-injured chi ldren with a coma score
The Gl asg ow Co ma Scale (GCS) developed by Teasdale of less than 8 should have endotracheal intubation (to pro
and Jennett i in 1974 has become a widely used clinical tect their airways) and ventilatory suppOl1. Children wi th
tool for assessment of t he level of co nsciousness in adults. mild (G CS 13-15) to moderate (GCS 9-12) head injury may
Furthermore, the GCS enab les a basa l initial sco re to be have impaired gag reflexes and wou ld also be at risk of
estab lished and thus allows for comparison between suc developing aspi ration, and may also require intubation and
cessive examinations of the level of consciousness identi mechanical ventilation.
fying any clinical deterioration. Many of the responses In intubated children, a grimace score has been sug
assessed in the GCS requ ire an adult level of neurodevelop gested to replace the ve rb al component of the GCS assess
mental function and ca nnot be easily graded in children ment. II This modified paediatric coma sca le for intensive
who are under 10 years of age. 2 Modifications of the GCS care practice origin ated from the Sharples' adaptation of
or other coma scales that include age- adjusted verbal and the James ' adaptation of the GCS. Non-verbal communica
motor responses have been described and used to assess the tion and verbal language are not totally independent, and
level of consciousness in .chi ldren with neurodevelopm en an important part of non-verbal commu nication invo lves
tal levels who are und er 10 years of age. 2- 8 facial expression and grim ace. Intero bserver reliability was
Epidemiology I 283
moderate to good, with all components assessed with the ARE presentations due to head injury in children less than
grimace score being better than the verbal score in all of 10 years than in any other age group. I 8
the 73 children studied; this suggested that the grimace Head injury is the co mmon est ca use of death in children
score may be more useful in intubated patients. II aged 1-15 years in England and Wales. 41 The mean mortal
The most common cla ssification for se1!erity of head ity from head injury in children calculated from 18 pub
i njLI1Y in adults is by means of the GCS, I in which a score lished studies was 5.1 ± 2.39 per 100000 children per
of 13-15 is designated mild, a score of 9-12 is designa ted year.9.14-1 6, 18,19,21,24-32,34-39 Although accidental lIlJury
moderate and a sco re of 3-8 is designated severe. Modifica mortality rates have fallen in the UK, groups of children
tions of the GCS for children are similarly used to assign who are at greater relative risk remain within the popula
severity. The duration of unconsciousness or post-traumatic tion, such as children residing in less affluent areas. 31,41 [n
amnesia have also been used to classify severi ty. Post Scotland, accidents to pedestrians were the leadin g cause
traumatic amnesia may be four times as long as the of head injury mortality in children.
duration of coma . Severity based on imaging is useful only Sample statistics for children with head injury attending
for severe injury. Therefore, a GCS, which is reall y a mini ARE at the Royal Hospital for Sick Children in Edinburgh,
neurological exa mination , remains the most useful index over two I-year periods, are seen in Fig. 14.1. 42 The peak age
of severity of brain injury. group was between I and 2 years of age, with a steady decline
in attendance thereafter up to 13 years. Falls accounted for
53 per cent of all attendances, with fails of more than I metre
EPIDEMIOLOGY in height accou nting for 12 per cent. Road traffic accidents
were responsible for 4.6 per cent of attendances; the remain
Accidental Head Injury der (42 per cent) was due to miscellaneous causes.
Cha n et al Sydney, Australia Retrospective January 198 5 0-15 years Admitted to hospital 13 per cent 'serious 2 Not given Pedestrians, MVC
1989 " (W Metropolitan Paediatric to December with trauma. GCS, TS, injuries' with ISS ;;;' 16 occupants, cyclists
Region) Trauma Study 1986 ISS noted 90 per cent
Se rious: TS ". 12 involved cranium
Serious TB I 17
Brookes et al Scotland, UK Retrospective 198411985 Adults, Evidence of brain Admission rate 4011 Children constitu te: 40 MVC 71%
1990 18 Case record s Children damage: altered Admission with brain 50% of ARE attenders,
for 23 Scottish < 5 yea rs, consciousness on injury 290 31 % of admissions,
ARE 5-11 yea rs, arriva l or history Most seve re 5-10 25% of neurosurgical
departments 12-14 years of altered per ce nt unit tra nsfers, 200/0 of
consciousness and 14.5-29 death s on neurosurgical
am nesia on arrival un it, 11 0/0 of operations
for ICH, 750/0 of severe
injuri es in children
are MVC
Yager et al 19909 Virginia, USA 1978 All ages 175 all 5.3 chi ldren
Sha"rpl es et al N Engl and, UK Retrospective 1979-1986. < 16 yea rs All deaths from head 5.3 HI was major sing le
1990 19 review of HI in c. injury in child re n ICD Mortality cause of deaths in
deaths 9 codes N800- NB04, significa ntly related children (1 5% ) 53 % of 25 5 fatal
N850-N854 to depriva tion index: 250/0 deaths 5-1 5 yea rs HI accidents
14.0 with greatest 49% di ed before admission occurred during
deprivation, 0.9 with Most deaths (76% ) due play
least deprivation to MVC: 53% pedestrians,
(P < 0.00001) 140/0 cyclists, 100/0
passengers, 120/135
pedestrian deaths from
unsafe child beh avio ur
lepd> cL al U)A Helrospeclive Not given National Pediatri c 44Ofo ha ve HI Mortality was Fall s
1990 20 Trauma database Trauma Regi ster, lower in children MVC, ped estrian /
of children identified head injury occupant 3.10/0 of 10098
alone, head plu s TBI admissions
extracranial inju ry, only
and extracranial injury
Hansen et al Denmark, 0-14 yea rs Severe TBI 18
1991" Ringkobin g county
Levin et al USA: Retrospective, 14 April 1983 ,,"15 years Severe traumati c head 6 months po st 0-4 years: 31 % fall/ju mp; Not give n
1992 22 San Diego, affluent hospital-based t o 18 April banded : inju ry with in 48 injury mortality 26 0/0 pedestrian; 20%
urban area; 1988 0-4 years hours of approximatel y motor vehicle occupant;
Virginia, inner-city, 5-10years admission, 52 0/0 0-4 years; 17 0/0 assault
suburban , rural ; 11-15 GCS"" 8 180/0 5-10 years; 5-10 years: 49 0/0 motor
Texas, large urban, years Gunshot wounds 22Ofo /1-15 years veh icle occupant; 26%
rural beach resort; excluded pedestrian; 130/0 fall/jump;
Houston, urba n; 9 0/0 cyclist
New York, inner-city 11 15 years: 55 0/0 MVC
occupant; 14% pedestrian ;
12 010 motorcycle; 11%
bi cyc le; 4% fall/jump
Ald rich et al USA Retrospective January 1984 ""1 6 years Severe traumati c 53% died
1992" to September HI with diffu se brain
1987 swell ing
Gabella et al Colorado, USA Surveillance 1991-1992 All ag es TBI defined from ICD Girls Age -adjusted Causes by age not given MVC mortality :
1997 27 study, residents 9 codes: 800 - 80 1, 0-4 years: 70; mortality rate s urban 14.8, rural
only 803-804 (skull 5-14 yea rs: provided 25.5
fracture), 850-854 60 all areas
(intracranial injury) Boys
Data from hosp ital 0-4 years:
discharge records 20 remote ru ra I,
and death ce rtificates 100 all other
5-14 yea rs:
90 metropolitan,
160 rural areas
(Con tinued)
Table 14.1 (Continued)
Reference Country/region Study Study period Age group Inclusion criteria Mean incidence Mortality Main causes Main cause of TBI
for TBI ofTBI rate ofTBI (%J mortality
L ____ . _
--- _._ -- - -
Eman uelson, SW Sweden Retrospective 1987-1991 0-1 7 yea rs ~1 hour of LoC 12 2.6 MVC 60<¥0 MVC 75%
von Wendt clinica l, Falls 22%
. 1997 28 neu rophysiol og ica I, Sport 7%
imaging evidence of Missile 3%
brain contusion Head trauma 20/0
(excludes: con cussion, Unknown 60/0
mild HI)
Suom in en Helsi nki, Finl and Retrospective 1985- 1994 < 16 yea rs Severe multiple trauma, All severe All severe trauma MVC (pedestrians Pedestrian s 21%,
et al 1998 29 inc. marked impaired trauma 14-185.6% 4.8 versus vehicle; MVC 20<¥0
consciousness, head injury 72% bicycle versus Cycl ists 12.7%
admission to ICU, to neurosurgical vehicle)
or auto psy ICU
Durkin et al New York - Retro spective 1983-1995 < 17 years Major MVC 9.3 < 17 years 2.9 < 17 years Falls 24%
1999 30 N Manhattan inc. HI = LoC ~ 1 hour, MVC 16%
with lacerati on :!: (2/3 pedestrians)
haemorrhage
Morrison Scot land, UK Retrospective 1981-1995 0-14 years Morta lity figures Injury mortality MVC mortality 4.8 Injury 140/0 of
et al 1999 31 Regi st rar Genera l, 11.6 Assau lt 0.7 total child
Data on Childhood Falls 0.6 mortality
Injury Mortality
Thurman, USA Annual surveys 1980-1 995 o to > 65 Hospitalization rates 1994-1995 In-hospital In cidence of mild and Not given
Guerrero years stratified : for TBI, from National TBI-related mortality rate, moderate TBI decreased
199932 0-4 years, Hospital Discharge admissions: all ages, remained over study period, but
5-14 years, Survey, CDC, PN CHS. 0-4 yea rs, 105 constant at 5.3 seve re TBI in creased
15-24 years, ICD 9 codes (range 68-142); from 10 to 19
25-44 yea rs, 800.0- 801.9, 5-14 years, 75
45-64 years, 803.0-804.9, or (range 50-100)
;:065 years 850.0-854.1
Benoit R N Virgi nia Retrospec ti ve 1991-1 999 0- 14 years Falls from windows, 3.9<¥0 mortality. 660/0 of 2322 falls resulted
et al 200033 admitted Trauma Centre, All deaths < 4 years in head trauma; children
or died fell onto concrete ";; 4 years more head inju ry
but less extremity injury than
10-14 years
rhurman et al USA, National Retrospecti ve 1997 0-14 years ICD 9 codes: Hospitali zed lBI: Pre-hospital deaths Causes different in Not given
2000 34 Centre Injury 800-801, 803-804, 0-4 ye ars = 2, at all ages different age groups:
Prevention, 850-854, and 873 59 Fatal outcome: 0-4 years: fall s >
Control for fatalities 5-9 years = 7.50/0,0-4 years transport related>
50 and 5-9 years; assault> firearms
10-14 years = 9.20f0, 10-14 years 5-9 and70-years:
51 transport > falls >
assaults/firearm s
Reid et al Minnesota, USA Retrospective 1993 0-19 years Children living in 73.5 ~ 19 years 9.3 ~ 19 years < 7 year: fall s, Non-metropolitan
2001 35 Minnesota with TBI Incidence peaks were Metropolitan 6.5, battering/assau It MVA
re sulting in in groups < 1 year, non-metropolitan 7-4 and 70- 74 years:
hospitalization or 6-7 yea rs and 15.4 fall s; MVC
death 13-19 years 5-9 years: fall s,
ICD 9 codes: Incidence in MVC (pedestrian)
800.00-801.99, Metropolitan 72.4, 75-79 years: MVC, falls
803.00-804.99, non-metropol itan
850.00-854.19 areas 76.1 TBI decrea sed as
and cases identified median hou se hold
from death certificates income and percentage
of high school
graduates in census
block increa sed
Masson et al Aquitaine, France Prospective 1996 All ages All admissions via Overall, 17.3 Overall mortal i ty 5.2 MVC (occupants 53 0/0, Ove rall highest
36 (IS-year emergency service severe TBI Mortality: pedestrians ISO/a, cyclists mortality: firearm s
2001
bands) with a diagnosis of 0-14 years, 8.4 0-9 years, 1.3; 10-19 12%, motorcyclists (but very few
severe head inj ury severe TBI years, 2.1 200/0) most common cases), pedestrians,
Excluded deaths Low-level falls, motorcyclists, falls
at scene high-level fall s on one level,
At 0-14 years, cyclists, high - level
mortality from falls, MVC
MVC-related lBI occupants
3.9; fall s 3.1
Adekoya et al USA Surveillance 1989-1998, o to >75 Mortality from lBI Not given Rates reduced Racial differences: 0-4 years: Falls
2002 37 summary by year years (5-year compiled from death graduall y At ages 0-19 years, 0.3, MVC 3.6,
Mortality only age bands) certificates, all States, 1998 mortality: MVA highest firearms 0.3, other 2.5
using ICD 9 codes 0-4 yea rs, 6.1 ; among American 5-9 years:
5-9 years, 3.3; Indian/Alaska Natives Falls 0.1, MVC 2.8,
10-14years, 5.5; Firearms lBI deaths firearms 0.4, other 0.5
15-19 years, 24.6 highest in black 70-74 years:
Americans Falls 0.1, MVC 3.4,
Fall-related TBI deaths firearms 2.0, other 0.6
highest in black 75- 79 years:
children at 0- 4 years; Falls 0.3, MVC
no difference s at 5-14 15.3, firearms
years; highest in white 12.5, other 1.6
chi ldren at 15-19
years old
(Continued)
Table 14.1 (Continued)
Reference Country/region Study Study period Age group Inclusion criteria Mean incidence Mortal ity Main causes Main cause of TBI
for TBI ofTBI rate of TBI (0/0) mortality
Williamso n et al Scotland. UK HI mortality 1986- 1995 0-14 years National mortality Not give n 290 child deaths MVC (ped estrian) MVC (pedestrian)
2002 38 review inc. data, Registrar from HI over 10 other bo th sexes 1.2
Gene ral for Scotland years M VC (occupants declined over study
ICD 9 codes for Mean annu al HI and cyclists) period
external cause of morta lity 3.0 ; fell Other MVC
inju ry and from 4.1 in 1986 (occupants,
skull fracture, to 1.8 in 1995 cyclists) 0.9
intracrania l inju ry Highest rate 5. 1 Falls 0.3
in most deprived areas; 0.3 Assaults 0.3
lowest 2. 1 in most Other, 0.3
affluen t areas assau lts in infants
Masson et al Aqui taine, Prospective 1996 All ages, Hospita I-admitted, Overa l1 8.5, All ages, 51.6 0/0 died All ages MVC 5.2
2003 39 SW France Traum at ic 0-14 years mas t severe bra i n 0- 14 years 3.2 0-15 yea rs, 42.9<\'0 0- 14 years: MVC
coma inj uries (GCS 8 or died 76 0Al, gunshot wounds
le ss), comatose for 10 0Al, falls 40Al,
> 24 hours, or died other 10%
without waking
Andersson et al W Swede n Prospective 1992- 1993 All ages, All grades or fatal Peak incidence All ages, 4.0 All ages: Morta lity for seve re
2003 40 and lO-year TBI, ide ntified from 0- 9 years Fal l at same level TBI ha s redu ced from
retrospecti ve age band s AEtE discharge register 0-9 years 35% 310/0, fall from height 40% in early 1980s
and death registers 10-19 years 20% 27%, MVC 160/0, hi t to - 10% in 1999
ICD 9 cod es 850-854 Overall, 2.50/0 by object 15<\'0
and 800-804 mode rate or severe
TBI, only 1% admitted
to hospital
CDCP, NCHS, Centers for Di sease Control and Prevention, National Hospital Discharge Survey; GCS, Glasgow Coma Score; HI, head injury; lCD, International Classification of Diseases; ICU, intensive care unit;
in c., inclusive; ISS, injury severity score ; LaC. loss o f con sc iousness; M VC, motor vehicle collision; PTA, post-traum atic amnesia ; TS, trauma score.
Epidemiology I 289
350 r
300 r-
V>
c
.<::>
V>
250
V>
E
"0 200
~
'"
0
150
~
.D
E
::l 100
J J J J I~ I
Z
50
I I I I I I I
0
2 3 4 5 6 7 8 9 10 11 12 13
Age (years)
10 1998-99 • 2002-03 1
Figure 14.1 Admissions to the Accident and Emergency Department, Royal Hosp ita l for Sick Ch ild ren, Edinburgh, with head injury
1998-1999 and 2002-2003, by age. (From ref. 42, wit h permission.)
Incidence of suspected NAHI in Scotland hour. 46 Because birth trauma was often associated with a sec
per 100000 chi ldren, aged 0 -12 months
ondary asphyxia, its precise incidence was difficult to obtain .
35.0
Routine magnetic resonance scanning of newborn
30.0
infants has demonstrated that clinically silent subdural
0
0 25.0 /' ~
haemorrhage does occur in infants in 6 per cent of normal
....-- ~
0
S2 20.0 vaginal deJiveries. However, these haemorrhages resolve
"
C>
"cu 15.0 within 4 weeks and remain clinically silent, and the infants
u" 10.0 develop normally.47 Instrumental delivery was associated
"g
5.0 with a higher incidence of subdural haemorrhage. 47 It
0.0
remains to be established whether these haemorrhages may
1998 1999 2000 2001 2002 explain minor neurological deficits in children. In a fUlther
95% (119.38.41 (17.5,481) (16.9,48.1) (106.37.8) (121.411)
ConL int.
100 100
90
90
80
80
-R 70 ~ 70
.s co
c:
:::>
60 .2:> 60
:~ 50 ~ 50
~
V\
:::> 0
0 40 <U
v
40
·c I::
Q.)
Vl 30
co
.<:; 30
u
20 20
10
10
0
0
3 6 9 20 25 30 35 40 45 50 20 30 40
(a) Speed at impact (mph) (b) Vehicle speed (mph)
Figure 14.3 (a and b) Relationship between veh icle speed and li ke lihood of serious injury or survival in Scotland.
Many earlier studies of severe head injury have concen management of these critically ill patients. HaemolTh age
trated on the importance of raised ICP, current evidence associated with an ICP device is uncommon, with an overall
moves toward emphasising the importance of cerebral perfu reported incidence of 1.4 per cent in the adult population 63 7o
sion pressure (CPP), the principal determinant of cerebral Fewer patients (0.5 per cent) develop significant haematomas
blood flo w (CBF). Thus, maintenance of an adequate CPP is requiring surgical evacuation. 65,67,69 Pople and co-workers 71
essential to maintain cerebral oxygen delivery and prevent examined 303 children who required ICP monitoring and
cerebral ischaemia. The only way to reliably determine CPP in found that only one patient (0.3 per cent) developed intra
the brain-injured patient is to continuously measure alierial cranial haemolThage after insertion of the ICP moni tor; this
blood pressure and ICP (cerebral perfusion pressure = mean patient was known to have a low platelet count prior to the
arterial blood pressure - intracranial pressure). procedure.
Although ICP monitoring is not universally used , virtu Malfunction or obstruction has been repOIied to occur in
ally all major adult head injury centres in the USA use ICP all types of ICP monitors. The mechanical failure rate of the
monitoring in guiding management and it is an integral fibreoptic device used in children with severe brain trauma
pali of intensive care in these centres. The practice varies was reported to be 13 per cent in one series,72 but in another
in Britain, with rou t in e ICP monitoring occurring in 48-57 study t he malfunction rate was much lower at 2,6 per cent. 71
per cent of intensive care units (ICUs) that regularly treat Cerebral pe~fusiol1 pressure is the difference between the
head-injured ad ults. 59 ,GO Similarly, evidence suggests the mean arterial pressure (MAP) and intracra nial pressure
monitoring and managemen t of children with head injury (ICP), i.e. CPP = MAP - ICP. The low systolic blood pres
who are at risk of developing intracranial hype rtension sure in infants means that the perfusion pressure is more
varies considerably in the UK and worldwide. Segal and easily deranged than in adults. The normal cerebral perfu
colleagues G1 conducted a postal survey of the use of ICP sion pressure is 60-70 mmH g in adults and, when reduced
monitoring in children with acute neurological illness. All there is a progressive fall in the CBF to the brain. When the
British !CUs that admit five or more children each year with CBF falls below 40 mmHg there is absolute reduction in
acute neurological illness were contacted. Only one-half cerebral perfusion and ischaemia a nd infarction will result
of the 70 per cent who responded reported the use of ICP at flows below 18- 20 mL!100g of brain per minute.
monitoring. In t his survey, children admitted to paediatric Cerebrovascular autoregulatory mechanisms ensure that
intensive care units (PICUs) were more likely to have ICP there is an adequate CBF to meet metabolic demands by
monitoring performed than those admitted to adult ICUs. 61 changing the cerebrovascular resistance in response to
There are no published guidelines for the use of ICP blood pressure changes over the range of 60-160 mmHg.
monitoring in children, but the major determinan ts in t he The arterioles constrict when the blood pressure rises and
decision to initiate lCP monitoring appear to be the admis- . dilate when it falls. 73
sion GCS score and brain CT scan appearance. GI When ICP Cerebral blood flow also alters in response to the meta
monitoring is clinically indicated in a patient, a decision bolic demands focally or generally within the brain. Both
on the type of monitoring device to be used is required. of these autoregulatory mechanisms are regulated by the
Intracranial pressure monitors that are avaiJable at present small arteries and arterioles. Major blood vessel dimensions
consist of a pressure transduction device that can be placed are reduced by vasospasm, which can be sufficient to cause
in the following intracranial compartments: ep idural, infarction. Small vessels have sympathetic innervation,
subdural, subarachnoid, parenchymal or ventricular space. possibly from the locus coeruleus, and respond to meta
Pressure transduction can be undeliaken t hrough external bolic demands of a paliicular part of the brain (including
strain ga uge, catheter tip strain ga uge or catheter tip fibre during sleep). These local flow changes are all influenced
optic technology. by pH , nitric oxide and lactate production.
The Camino fibreoptic pressure monitoring device is the The Cushing phenomenon is a progressive response of
most fre quently used system in paediatric neurointensive systemic arterial pressure in response to an increasing fCP
care. Initial reporting and validation of this type of trans as it increases to levels approaching the diastolic blood
ducing system have indicated its ab ility to measure the pressure. It is mediated via brainstem reflexes, which
brain tissue pressure directly with a rapid response rate to depend on blood flow to the locus coeruleus. The cardiac
ICP changes. Gambardella's group62 from Italy examined output and peripheral resistance are increased, the latter by
its use in a group of children with severe craniocerebral sympathetically mediated peripheral vasoconstriction and
tra uma, aged 2-16 years, and demonstrated that this an increase in vasomotor tone. Thus there is a sequence of
method of recording ICP was minimally invasive and that comp ensatory events to raised ICP from the initial cerebral
it correlated very closely with the ventriculostomy method, arteriolar dilatation followed by elevation of the MAP, fol
Complications associated with ICP monitOIing include lowing which both MAP and CPP fall in respons e to con
bacterial colonization, infection, haemorrhage, malfunction, tinued ICP elevation. 74
obstruction or malposition of the device. Long-telID morbid Continuous body temperature monitoring in patients with
ity is rarely associated with these complications but they may head injury is a routine practice, as abnormalities in body
affect the accuracy of ICP readings and influence the acute temperature, paliicufarly pyrex ia, occur frequen t ly after
292 I Head and neck injuries
brain injury.75,76 Conventional indicators of core temperature or multifactorial insults. Tasker and co-workers 95 subse
include rectal and bladder temperature, but recent advances quently used the cerebral function analysing monitor
in intracerebral monitoring have enabled brain temperature (CFAM) in conjunction with serial multichannel EEG to
to be measured; this is done by using a thermocouple embed monitor cerebral function in 54 critically ill comatose chil
ded in the tip of an intraventricular catheter used for rcp dren , and demonstrated that CFAM traces could be used to:
monitOling (or a triple bolt inserted into the frontal region (1) identify effects of acute or cumulative cerebral insults;
encompassing an ICP monitor) , a microdialysis catheter and (2) recognize unstable patients exquisitely sensitive to
a Para trend multiparameter (oxygen, carbon dioxide, pH and standard care procedures; and (3) evaluate seizure control.
temperature) sensor. 75 ,77,78 Although maintaining an adequate CPP is important as
The precise influence of pyrexia on the outcome after it provides a pressure gradient governing the CBF, the ulti
brain trauma remains unclear. Pyrexia has been postulated mate determinants of the cerebral metabolism are CBF and
to exacerbate brain injury by increasing cerebral metabolic the oxygen content in the blood. Monitoring of the ICP and
requirements for oxygen and excitatory neurotransmitter CPP alone offers no information on the brain's oxygen
circulation. delivery and usage. At present, methods to measure cere
The clinical use of hypothermia was pioneered in bral blood flow such as PET, Xenon clearance or SPECT
1938 79,80 and since then several laboratory studies 81 - 84 and are too cumbersome for use in an ICU, however, jugular
small clinical trials 85.86 have suggested its association with venous bulb oximetry and transcranial Doppler ultrasound
an improved recovery. In 1994, a multi centre randomized enable a better understanding of the state of the cerebral
controlled study was conducted, in the hope of providing circulation and oxygen consumption .
definitive evidence of the efficacy of hypothermia in adults
with severe brain trauma. The study was halted in 1998 after CEREBRAL BLOOD FLOW
enrolment of 392 of the planned 500 patients when induced
hypothermia, bladder temperature of 33°C, initiated within The CBF depends upon: (I) cerebral perfusion pressure;
6 hours of injury and maintained for 48 hours, failed to (2) cerebral vascular resistance (radius of vessel, length of
improve clinical outcome.87 Furthermore, patients who were vessel) inversely; (3) viscosity ; (4) venous pressure; and
in the hypothermic group and more than 45 years old had a (5) ICP. (CPP = CBF/CVR, CBF = CPP X CVR.)
higher incidence of poor outcome and had more complica The normal CBF is around 50 mL per 100 g of brain tissue
tions than those who were kept nOlIDothermic. 87 Neutropenia per minute. 96 The CBF is a higher percentage of the cardiac
and coagulopathy may complicate induced hypothermia. output in small infants than in adults. It is also highest in
Thus, fuliher studies are required to determine the influence grey matter (100 mL per 100 g) and decreases, sequentially,
of brain temperature on outcome after head injury. It is not through the inferior colliculus, sensory cerebral cortex,
resolved whether hypothermia induced by selective head motor cerebral cortex, geniculate bodies, superior colliculus,
cooling in newborns merely reduces the extracranial blood caudate nucleus, thalamus, cerebellum to the cerebral white
flow without alteration of the intracranial circulation. matter, where the rate of flow is 20 mL per 100 g of brain tis
sue per minute. Impairment of CBF due to raised ICP is clin
ically seen as distension of scalp and retinal veins, pulsation
ELECTROPHYSIOLOGICAL MONITORING
of the fontanelle and a loud systolic cranial bruit, which
Head-injured children requiring intensive care often need appears when the ICP exceeds the diastolic blood pressure
to be pharmacologically sedated and paralysed; this masks and increases in pitch with slowing of the cerebral circula
the motor and other clinical manifestations of seizures and tion, and disappears with a 'carotid stop'.
makes the assessment of cerebral function by conventional Non-invasive methods of assessing the flow velocities of
clinical neurological examination impossible, apart from the intracranial cerebral arteries using transcranial Doppler
pupillaly reactions. Furthermore, some children develop ultrasound (TCD) were first described in the early 1980s. 97
abnormal movements without electrical seizure activity Flow velocities can be measured in the middle, anterior and
after head injUlY and without electrophysiological moni posterior cerebral arteries by placing a probe in the temporal
toring they can receive unnecessary treatment. In addition area, just above the zygomatic arch. As the cross-sectional
to providing valuable information on the degree of cerebral area of the arteries cannot be measured directly, flow can
insult in the critically ill child with an al tered level of not be measured from velocity but the Doppler shift meas
consciousness, electrophysiological monitoring of brain ured is inversely proportional to the diameter of the vessel.
activity has been recognized to have prognostic value in Thus, providing that all of the other factors remain con
children with severe head injury.aa-90 stant, the mean flow volume (MFV) will approximate CBF.
Multichannel electroencephalography (EEG) recording is Vasospasm will result in an increase in flow velocity.
the gold standard in measuring surface brain activity.91-94 The resistive index (RI) or Pourcelot index (S - DIS) is an
It allows distinction to be made between generalized, local alternative ratio for assessing flow velocity and is normally
ized, focal or multifocal seizures and to provide an aetio of the order of 0.7. It increases with increased ICP and is
logical and prognostic guide in relation to single, repeated usually reliable unless there is distortion of the course of the
Epidemiology I 293
MCA. Using transcranial Doppler, MCA spasm was identi an increased production of lactate while the production of
fied in 40 per cent of patients after traumatic brain injury pyruvate decreases; this causes an increase in the lactate
and could begin as early as 48 hours post injury.98 Maximal pyruvate ratio. 102 ,103 Glycerol is an integral component
MCA spasm was found between 5 and 7 days after the of the cell membrane and has been use, therefore, as a
initial trauma. 98 In another study of head-injured adults, marker of cell membrane damage. 104 ,105 Although gluta
severe MCA spasm identified by transcranial Doppler was mate has been proposed as an indirect marker of cell dam
confirmed by angiography.99 age, it is often hard to interpret the dialysate level as the
Daily measurements of the MCA velocity from 121 neuronally released glutamate is mixed with the large
patients with varying severity of head injury showed an metabolic pool of glutamate. A reduction in the dialysate
inverse correlation between the severity of head injury and glucose is often associated with a decrease in brain oxygen
the MCA velocity. 100 In addition, a significantly lower MCA tension suggestive of reduced cerebral perfusion. 106 A
velocity on admission was demonstrated among the head marked increase in the adenosine level of the cerebral
injured patients than among the normal control subjects. 100 interstitial fluid has been shown to occur during jugular
High ICP and low CBF were responsible for the low velocities venous oxygen desaturations suggesting a potential role
in the intracranial circulation after head injury. FUlihermore, for adenosine during the periods of secondary insults after
an admission MCA velocity of less than 28 cm per second brain trauma. 109
correctly predicted 80 per cent of the early deaths. 100
concomitant parenchymal injury. This improves after surgical in the intensive thera py unit (ITU), paediatric neurology
remova l of the clot without opening the dura. La rge volumes department and by the child protection team. The inappro
of blood may be lost, leading to shock and secondary brain priateness and inconsistency of these multiple accounts
injury or death. The risk factors for deterioration include a may be an important indicator of an inflicted injury. A
fracture crossing the middle meningeal artelY, vein or sinus, traumatic explanation is offered in 10-20 per cent of cases
or an increase in the size of the epid ural haemorrhage. to account for the child's condition.
The symptoms at presentation to hospital with thei r rel
SUBDURAL HAEMATOMAS ative frequencies were:
• irritability and crying, 41 per cent;
Thin subdura l haematomas assoc iated wit h diffuse brain
• bruising and su perficia l injury, 41 per cent;
injury or swelling are most common. The indications for
• vomiting and a nore x ia , 38 per cent ;
decompression include midline shift, persistent raised [CP
• apnoea and respiratolY difficulties, 38 per cent ;
and no associa ted severe brain swelling. Subdural haemor
• extensor stiffness, due to fit, anoxic rigidity or raised
rhages in the interhemispheric fissure may be tapped in
intracranial pressure, 34 per cent;
infants via the anterior fontan elle.
• probable fits, 21 per cent;
• pallor or cyanosis, 29 per cent;
CONTUSIONS
• history of a potential traumatic event, 26 per cent;
Focal contusions may not be immediately visible on CT • depressed conscious state, 21 per cent.
scans but are evident in 50 per cent at follow-up.1I2 They
exert a mass effect and contribute to the overa II raised [CP.
ON EXAMINATION
Depressed sku II fractures require repair of the dura to prevent Bruising and oedema of the sca lp , periorbital bruising or
raised pressure, which leads to brain herniation with venous Battle's sig n (bruising over the mastoid) (Fig. 14.4), usually
infarction of the cortex. They also predispose to infection. seen in accidenta l head injury, is rarely present, even in the
Fractures of the skull base, with leakage of CSF throu gh presence of severe intracranial damage.
the nasal cribriform plate or t he ear, result in a compens a
tory mechanism that keeps the ICP low until the leakage Skeletal Injury
ceases. The defi nitive rep a ir is undertaken after the acute
Skull fractures are seen in approximately one-quarter of cases
injUly period.
and evidence of impact is seen in more than one-half of cases.
Penetrating injuries (gun shot, knife, darts, needles) to
Signs of impact, particularly of the head, indicate
the brain in children are rare in the UK. In those cases when
inflicted trauma. As young infants are unable to injure
there is a preservation of consciousness, the object is left in
themselves accidentally, they may be an indicator of intent.
position until after sca nning. Debridement is then under
Clinical evidence of imp act is:
taken with pressure monitoring and antibiotic cover.
• cutaneous or subcutaneous bruising;
• subgaleal haemorrhage;
NON-ACCIDENTAL HEAD INJURY • skull fracture;
• extradural haemorrh age;
A history of preceding events lea ding to hospital admission • focal subdural haemorrhage;
is obtained on several occas ions in the AEtE departm ent, • focal cerebral contusion.
Non-accidental head injury I 295
Overa ll, 53 per cent of cases ha ve no fracture on skeleta l This presentation results from severe shaking forces. Geddes
survey. Rib fractures are seen in one-third of cases and et al" 6 , 11 7 found localized axonal damage at the craniocervi
fractures of the long bones seen in one-qu arter. cal junction, in the corticospinal tracts in the low er brain
stem and the cervica l cord roots; additionally, t here is the
possibility of traum atic thrombosis of the vertebral arteries
BRUISING AND SKIN TRAUMA as they wind through the foram ina of the cervica l vertebrae.
Our experience is that some bruising and superficial abra Damage to these areas is thought to result from hyperflex ion
sions occur in more than 40 per cent of cases, often a and hyperextension movements, a 'whiplash' shaking stem
mixture of new and old bruises. Additionally, abras ions, injury. These cases are infrequently or only briefl y seen by
sc ratches, lacerat ions, hair loss, subconjunctival haemor the clinician (6 per cent) as they a re either dead on a rri val at
rhages and bl eedin g from the ears and mouth are present in hospital or die shortly thereafter, but are more frequently
a small number of cases. seen by pathologists. Th ese young infants, at about 2- 3
A firm adult grasp is necessary in order to shake a child months of age, presen t with apnoea as part of this cervi
and may result in a pattern of bruises. These must be care comedullary syndrome, with only 'trivial' subdural bleeding.
fully documented and photographed. Common methods of The apnoea gives rise to severe seconda ry hypoxic brain
grasping a chiJd (and subsequent injuries) are: injury with cerebral oedema but minimal axonal shearing.
• by the chest - thumb marks at the sides of the nippl es, In this hyperacute presentation, all subdural haemorrhages
fractured ribs and retinopathy; that are seen on imaging and post-mortem exa mination are
• by the arms - spiral fracture of the humerus; sma ll , as a result of ac ute cerebral oedema. Significant sub
• by the legs - bucket handle metaphyseal fracture ; dural haemorrhages do not become evident until 2- 3 days
• by the throat - carotid trauma (ipsilateral infarction later. In summary, the presentation of the hyperacute pattern
and loss of ve nous pulsation in t he retina, contralateral is apnoea, acute respiratOlY fa ilure and death associated wi th
hemiplegia, faci al and retina l petech iae); severe hypoxic ischaemic dam age found post mortem.
• by the abdomen - finger bruising, retroperitoneal
bleeds and bowel bruising;
Acute Encephalopathic Presentation
• by the shoulders - finger and th umb bruises;
• bruising around t he mouth suggests possible This is characterized by seizures, decerebration, homeostatic
suffocation to prevent the infant screami ng; thumb derangements, bilateral large subdural haematomas and
marks should be carefully sought under the mandib le. widesp read haemorrhagic retinopathy. Additional findings,
296 I Head and neck injuries
Subdural Haemorrhage
Bleeding from torn bridging veins in to the subdural space
is the hallmark of non-accidental shaking injUly in the first
year of life. Almost 90 per cent of patients suffer subdural
Figure 14.5 Infant, aged 3 weeks, at necropsy. A thin film
haemorrhage, which is frequently bilateral, and in 20 per
of subdural haemorrhage is present over cortical convexities,
cent of cases it is associated with subarachnoid haemor
following inflicted injury. Bilateral rib fractures and bruises were
rhage. The subdural haemorrhage may be unilateral at
present.
presentation. Subdural haemorrhage may be over the con
vexity, interhemispheric, subtemporal, suboccipital or in
the posterior fossa. The subdural signal intensity on MRI The child may present with a recurrent encephalopathic
scanning may be different in different intracranial com presentation with 'odd turns' or fits, apnoeic attacks, cyan
partments and does not imply repeated bleeding from otic attacks, rigidity or coma.
repeated trauma. By the time of necropsy, only a thin film
of subdural haemorrhage may be present over cortical con
Non-encephalopathic Isolated Chronic Subdural
vexities (Fig. 14.5).
This is a 'late' presentation in a child with an expanding
Retinal Haemorrhages head circumference, subdural haemorrhage and often little
In shaken baby syndrome, retinal haemorrhages occur in else. There may be vomiting and hypotonicity. Other causes
80 per cent of cases and one-third are unilateral. I1 9 Levin's for subdural haemorrhage must be excluded.
groupl20 found them in 84 per cent of cases compared to
16 per cent in accidental head injuries - usually high
DIAGNOSIS OF NON-ACCIDENTAL HEAD INJURY
velocity road traffic accidents with side impact. Retinal
haemorrhages were extensive and involved all of the reti The diagnosis of NARI is made on the history, which is fre
nal layers. They are thought to be due to vitreous traction. quently inconsistent and does not explain the clinical find
The vitreous humour is attached to the retina at its periph ings - ophthalmological and radiological features compatible
ery and with different inertias this is also the site where with, and sometimes characteristic of, non-accidental oligin;
shearing forces tear the vitreous attachments from the and the social pathology in a significant number of cases
retina (or ora serrata) and cause disruption of ocular and in the form of known risk factors (Table 14.3). Evidence that
orbital blood vessel integrity - similar to the shearing shaking is responsible for many NARIs emanates from
forces which cause subdural bleeding and cortical tears. numerous studies and reports of clinical experience, fre
quently acknowledgement or confession with deSCription of
the mechanism by the perpetrator. Biomechanical and animal
Non-encephalopathic Subacute Presentation
studies have confirmed the forces generated by shaking and
With this presentation, there is no acute brain swelling or other mechanisms. Injurious behaviour has been witnessed
diffuse cerebral hypodensities or fits, coma or decerebra by closed circuit TV. Consensus statements by scientific bod
tion. Subdural haemorrhage, retinal haemorrhage, rib and ies such as the American Academy of Pediatrics state the
other fractures and bruising occur in various combinations. extensive combined experience of professionals. 121 As the
Non-accidental head injury I 297
Table 14.3 Risk factors for inflicted head injury with evidence of other malicious injUlY, such as multiple
fractures of different ages, skin incision , cigarette burns or
Single parent or cohabitating partner
repeated beating.
Young parental age
Combinations of any two of the following three factors
Past history of abuse to children
are highly predictive of inflicted head injury (P < 0.001), 122
Drug or alcohol abuse
i.e. head injury plus any two of th e three following :
Domestic abuse
• inconsistent history/physical ex amination;
Parental mental or psychological illness
• retinal haemo rrhages;
Premature birth
• parental risk factors (alcohol or drug abuse, previous
Twin s
soci al service intervention within the family or a past
Re cu rrent medical consultations and hospital admissions
histolY of child abuse or neglect).
Past history of social work enquiry or involvement
Traumatic retinoschisis and retinal folds are thought to
be pathognomonic for 'shaken baby sy ndrome', although
extensive multilayer retinal haemorrhages are most
Table 14.4 Features of the expanded syndrome of
unlikely to be due to any other cause. It is important to dis
non-accidental head injury
tinguish between making a diagnosis of NAHI with raising
Acute encephalopathy concern or suspicion of it. Kivlin 123 quotes 'the mere pres
Subdural hematoma ence of any retinal haemorrhages adequate to raise concern
Age usually less than 1 year of "sh aken baby syndrome", the extent or type of the
Acute cervicomedullary injury haemorrhage is less clinically important.'
Haemorrhagic retinopathy Subdural haemorrhage is more likely to be non-acci
Bruising dental in origin if it is convex and bilateral, interhemi
Sku II fractu re spheric, and is associated with cerebral tears and ruptured
Rib or limb fracture bridging veins that are visible on imaging.
Evidence of malicious injury (bites, cuts, cigarette burns, Rao and co-workers l24 concluded that hypoxic-ischaemic
whip marks) encephalopathy (HIE) plus interhemispheric subdural haemor
A history that is incompatible or inconsistent with rhage was highly specific for the diagnosis of NAI. The inter
clinical findings hemispheric subdural haemolThage and HIE OCCUlTed in 89 per
Acute cerebral oedema or diffuse cerebral hypodensities cent of cases. Zimmennan's group 125 suggested that interhemi
Early cerebral atrophy spheric subdural haemorrhage could be specific for NAI (61 per
Poor long -term prog nosis cent of cases).
There is nothing characteristic about the traumatic
encephalopathy in NAHl, although post-traumatic seizures
occur velY much more frequently in inflicted head injury than
trauma is almost always unwitnessed , it is important that the in accidental head injury and they abate by the fifth day.1 26
clinician is careful about attributing a mechanism of injury in Duhaime et al 127 developed an algo ri thm for determining
a particular case and diagnoses 'inflicted ' or a 'non-accidental' the probability that an injury was inflicted. It results in a
injury (which, however, may be consistent with a shaking diagnosis of 'presumptive inflicted ' injury or 'suspicious
mechanism). inflicted'injury.
Although, theoretically, the features in Table 14.4 repre The presence of anyone of the features in Table 14.3 in
sent the most complete (expanded) syndro me of NAHI, in a child with injuries must raise concerns or suspic ion about
practice common clinical presentations include a combina the possibility of child abuse and should be an indication
tion of several of these components. for investigation.
The component palis do not carry equal weight nor are It is imp0!1ant to consider the questio n of intent, which can
they independent. The more components present, the more sometimes be inferred from the clinical findings , for example
secure the diagnosis. Although a single component, for when the intent was malicious and the perpetrator cannot
example subdural haematoma or retinopathy or isolated have been in any doubt about the injury he was causing to the
haemorrhagic retinopathy in a yo ung infant, is still most child. The clinical findings supporting this intent include:
likely due to NAl, it may be due to other causes. Individual multiple bruises of different ages and patterns in different
signs and symptoms can have different causes and some sites, repeated admissions with physical injuries, multiple
features are more characteristic of NAl and so the diagnosis fractures of different ages, and particular types of injuly pat
only becomes statistically secure when several components tern, such as bite marks, cuts and cigarette bums.
are found in combination. The most secure diagnosis of Disciplinary injuries are not associated with the intent
NAHI is either when there has been an admission from the to injure and there may be no intention of harming the
perpetrator or when the head it\iury is seen in combination child at all. The intention to discipline is not premeditated
298 I Head and neck injuries
or sadistic. AI.though a physician may be sympathetic to severe impact injury. The type of skull fractures reported in
the stressful circumstances surrounding such injury, it a study of 100 consecutive children 127 who were less than
must be reali zed that without intervention , if the stress is 2 years of age revealed that 27 were linear, eight were
repeated then so may the abuse. Other possible explana depressed, three were multiple, one was stellate, three were
tions for the injury include episodic dyscontrol, puerperal bilateral and four were basal.
psychosis, euthanasia, and Munchausen's syndrome by
proxy (fabricated or induced illness).
BIOMECHANICS OF SKULL FRACTURE
ultrasound scanning.142
Foren sic
Figure 14.8 (a) Twenty-five days post admission: fluid-attenuated inversion recovery (FlAIR) coronal sequence showing posterior layering
within the subdural collection and high signal in the left parieto-occipital cortex. There is widespread encephalomalacia (with large subarachnoid
spaces, dilated ventricles and atrophy) and early gliosis. (b) Twenty-five days post admission: T2 transverse magnetic resonance imaging sequence
showing left temporal scalp haematoma, widespread high intensity white matter and encephalomalacia more marked on the right.
SPECTROPHOTOMETRY
Figure 14.10 (a) Seven months post adm ission: fluid-attenuated inve rsion recovery (FLAIR) coronal seq uence showing mixed intensity
loculations bilaterally, suggestive of low-density clot within the collecti on . Overall brain atrophy. (b and c) Eigh teen months post
admission: FLAIR coronal and T2 transverse images demonstrating marked gliosis and residual cystic change particularly in the right
hemisphere. Loculations within the subd ural collections are not now visible.
Pericranial and intracranial birth injuri es are illus trated Cephalohaematomata are haemorrhages beneath the
in Chapter 8. periosteum; they limited by its attachment to cran ial
Caput succedaneum results from scalp oedema secondary sutures and are so metimes associated with skull frac tures
to pressure of the head on cervix. It resolves within a few days. (10-2 5 per cent); 145 they are caused by compression to the
Subgaleal haemorrhage involves bleeding into the sub fetal head and usually resolve without sequ elae.
ga lea l space, often over the whole sca lp . It is an infrequent Skull fractures can result from instrumental delivery.
complication of vacuum extraction or rotational forceps They are usu ally simple linear parietal fractures but,
delivery. A large volume of blo od may be lost, thus requir if depressed, may impinge on the cerebral cortex, cerebel
ing resuscitation. lum or brainstem.
302 I Head and neck Injuries
1.8 - direct;
1.6 - acceleration;
a 1.4 - deceleration;
2- 1.2 indirect;
Q.J
u
c: 1
co - linear acceleration/deceleration;
.D 0.8
a 0.6 - rotational acceleration/deceleration ;
'"
.D
<x: 0.4 • wh ipl ash/s haking injury ;
0.2 • combi natio ns of the above.
0
350 375 405 415 425 460 500 533 575
Wavelength (nm) PENETRATING HEAD INJURY
Figure 14.11 Spectrophotometric graph of a subdural aspirate. Penetrati ng head injuries damage the brain by laceration,
The bilirubin peak indicates that blood has been present for haemorrhage and infection. Children may suffer penetrat
>24 hours. ing head injuries as a result of gunshots, scissors, knives,
darts and pencils (particularly involving the tonsillar fossa
Epidural haemorrhage is a rare bilih injury. It is usually with carotoid artery injury). As they are non-rotational,
associated with a skull fracture and is usually extremely they a re non- concussive.
small because of the tig ht application of the dura to the
skull bones in infancy.
COMPRESSION INJURY
Subdural haemorrhages may follow breech delivery,
cephalope lvic disproportion, vacuum extraction and pre Compression is rarely a primary mechanism of NAH!.
cipitate delivery, which lacerate bridging veins or dural The classical model for compress ion head injury is bilih
venous sinuses. They may require surgical intervention or injury complicating malpresentation or cephalopelvic
develop as a subdural hygroma. disproportion.
Subarachnoid haemorrhage occurs in full term and, The clinical patterns of traumatic birth injuries have
more frequently, in pre-term infants and is usually hypoxic been outlined above . The mechanisms of intracranial dam
in origin. Choroid plexus haemorrhage or subependymal age in compression injury are: (J) compression (impaired
venous haemorrhage extend into the ventricle system and cerebral blood flow, oedema, traumatic asphyxia, coning);
may subsequently extend into the subarach noid space, (2) moulding (l atera l moulding - sagittal sinus entrapment,
around the poste rior fossa structures . There is a risk of sub tearing of bridging veins, anteroposterior moulding
sequent hydrocephalus about 2 weeks later. tearing of the vein of Galen, tentorium and the venous
Brain contusions resulting from forceps or compressive sinuses); and (3) traction (cervical spine, brachial plexus,
damage to the head during bilih are rare but are associated vertebral arteries).
with skull fractures. Compressional injuries are non-rotational and non
Brainstem and spinal cord injuries result from excessive shearing; therefore, they are non-concussive until cerebral
neck movements, particularly during breech delivelY or oedema, raised ICP and surface haemorrhages result (such
other malpresentation. Fractures are not usually evident. as subdural and subarachnoid haemorrhages).
InjUlY results from traction or compression of the brain
stem and spinal cord, or thrombosis of the vertebral or
IMPACT INJURIES
anterior spinal artery. Lower cord injuries are rare but may
result from similar exaggerated spina l movements. Direct acceleration impact injuries are exemplified by a
Peripheral nerve injuries result either from intrauterine straight punch and direct deceleration/impact injuries dur
malposition with oligohydramnios, or traction on the ing a fall. Indirect linear acceleration/deceleration injuries
neck or upper limbs that produces Erb 's (with or without are typically from vehicular whiplash injuries an d indirect
Horner's syndrome), Klumpke's or other peripheral nerve or direct rotational acceleration/deceleration injuries that
palsy. commonly result from motor ve hicle collisions. Shaking is
a specific mechanism of injury that includes rotational
acceleration/deceleration components. In individual cases,
PRIMARY MECHANISMS OF INJURY TO there is often more than one of the above mechanisms
THE BRAIN operating.
Most accidental head injuries in children are impact
There are several mechanisms by which head inju ry may be injuries. In acceleration injuries, the force is dissipated in
sustained: the resultant motion of the head. In deceleration injuries,
• penetrating injury; the force can only be ameliorated by a 'fo rgivin g' surface
• compression injury; or bounce. Deceleration injuri es produce the most severe
• impact injury; clinical features.
Primary mechanisms of injury to the brain I 303
SHAKING AS A MECHANISM OF NON-ACCIDENTAL HEAD importance in shaking injuries. The forces within the skull
INJURY IN INFANTS resulting from shaking against a soft surface, such as a cot
mattress, are signi ficantly less than the forces arising from
There are five major components to the syndrome:
being shaken against a hard surface, such as a wall.
1. shaking causing rotation of the brain inside the sk ull ;
Two biomechanical studies by Duhaime et al 146 and
2. impact against a soft surface, causing acute
Minns et al 147 found the angular accelerations to be of the
deceleration of the brain and vastly increasing the
order of 1139 rads -2 (mean) and 200-892 rads -2 (range)
rotational torque forces inside the skull (no external
respectively.
evidence of impact);
The forces exerted by the shaker are dependent on the age,
3. hyperflexion and hyperextension injury from whiplash,
height, gender and strength of the individual shaking, so that
causing repetitive subluxation of the cervical
young children (3 years) are unable to lift or shake a weight
vertebrae; this results in injury to the cervical cord
equivalent to a 2-month-oJd child. Acceleration patterns
(Fig. 14.12) and/or medulla;
increase gradually to adult values by the age of 10 years.
4. shaking with hard impact, causing skull fractures and
The mean frequency for adult volunteers shaking
contusions in addition to the rotational injuries;
manikins was 3.5- 2.7 shakes per second; for children was
5. primary hyp ox ic-ischaemic brain injury resulting from
2.1-1.8 shakes per seco nd. Males shake significantly faster
apnoea due to brainstem injury.
than females (P < 0.0001) but the frequency is independ
ent of the load.
PREDISPOSING FACTORS In order to produce brain damage, a minimum duration
A number of factors predispose the infant to injury from of shaking is required. It is likely that relatively fewer
shaking: shakes would be necessary to induce a whiplash type of
1. The young infant has a relatively large and heavy head stem injulY than those required to produce damage from
in relationship to the body size. repetitive rotation. There is also a maximum duration of
2. The neck muscles are weak and there is little head shaking that is possible before fatigue prevents further
control in early infancy. As the child gets older the shaking. Experimentally determined maximum possible
neck muscles become stronger so that the head moves durations for shaking may not be exactly similar to dura
with the body; shaking whiplash injuries are rare, tions or rates during situations associated with anger and
therefore, after the second year of life. rage, but for ad ults shaking manikins that a re equivalent
3. The extracerebral space is relatively large in weight to a 2-month-old baby the median duration was
(physiological craniocerebral disproportion). 21.5 seconds; for a manikin equivalent to a 5-month-old
4. Th ere is physiological laxity of the meninges in baby the median duration was 19.75 seconds. Duration
infancy compared with older children, with less sustained was significantly longer in men than in women
anchoring of the brain. and was related to the strength of the weakest arm and to
5. In comp lete myelination in the white matter w ith a full the height of the individual. It is, therefore, not possible
complement of neuronal cell bodies in the grey matter for adults to shake infants for prolonged uninterrupted
means a difference between the specific gravity of the periods.
grey and the white matter and a gradient of tangential The centre of the rotational arc for shaking infants is
accelerations that produ ces shearing. likely to be in the upper part of the cervical spine, between
the craniocervical junction and C2.
The posterior excursion of the head during shaking is
BIOMECHANICAL DETERMINANTS OF SHAKII\IG INJURY
limited by the spinal joint ang les and ligamentous attach
An external force has first of all to overcome the neck mus ments, which decelerate the movement more slowly than
cles and induce anteropos terior movement with rotation. the sudden deceleration from shaking with impact.
This sets up shearing forces within the skull and within the It is possible to shake fast with shOlter angles and slower
brain itsel f (rotational shearing forces). As with impact with greater angular excursions. The resultant brain dam
injuries, the amount of damage will depend upon the rate age may be a product of the speed and degree of rotation.
of acceleration/deceleration before the direction is The pericerebral space in infants can be up to 1 cm in
reversed. Soft impact brings the skull to a sudden halt with depth to allow for subsequent brain growth. This cranio
maximum stretching of the bridging veins, thus the force cerebral disproportion with increased CSF spaces reduces
used is a major determinant of the injury sustained. the inertia for any brain movement consequent on cranial
Sufficient force is necessary to Jift the weight of the infant movement, and more swirling of the brain is possible with
and accelerate it in space and pOSSibly add additional force by stretching of the bridging veins. Measurements of the
active impaction of the child. The force with which the infant sinocortical distance (from the superior sagittal sinus to the
hits the surface will be dependent upon the strength of the cerebral cortex) were made during ultrasound scanning of
individual; the angular forces (i.e. torque), which cause rota the brain, showing a change in this space as a result of a
tion (see above) do not cause fractures but are of paramount change in the child's position, i.e. supine versus sitting.
~===7===A=C=C=ID=EN=T=A=L=H=EA=D=I=NJ=U=R=Y==~~~ll
MALPRESENTATION
or PEN ETR ATl ON
DISPROPORTION
IMPACT
BRAIN COMPRESSION PENETRATING INJURY
(AcceIera ti onl decel era ti on)
Tears of vein of
Galen and venous
Figure 14.12 Mechanisms of brain injury in different situation s. CBF, cerebral blood flow; CPP, cerebral perfusion pressure; EDH, extradural
NON-ACCIDENTAL HEAD INJURY TYPES OF
HEAD INJURY
WHIPLASH
MECHANISMS OF
ROTATION BRAIN INJURY
SHAKIN G IN JURY
ANATOMICAL
CERVICOM EDU LLARY LEVEL
PERCU SS ION AND TRACTI ON
PRIMA RY
INJURY
SECONDARY
CHRONIC SDH ASPHYX IA
INJURY
INTRADURAL
HAEMORRHAGE
SPACE
OCCUPATION
RAISED ICP
(contd) haemorrhag e; ICP, intracrania l pressure; SAH, subarachnoid haemorrhage; SDH, subdural haemorrhag e.
306 I Head and neck injuries
There are many intermedialY pathophysiological mecha A chronic subdural haematoma is often clinically silent
nisms that will determine the final pattern of brain damage with only an inappropriately increas ing head circumfer
(Fig. 14.13). ence. The causes of, and conditions predisposing to, a sub
The esse ntial pathophysiological mechanisms are: dural haem ato ma, are seen in Table 14.6.
• surface shearing - cortical emissary veins - subdural A chronic subdural is not a true haematoma but a subdural
haemato ma; effusion in that it is mainly water with a haematocrit that is
• parenc hymal shearing; usually less than 10 per cent. It is usually due to failure of
- midbrain shearing injury; the acute haematoma to resolve. An acute haematoma may
- shearing of grey-white interface; become cill'onic because of: (1) its large volume; (2) a n
• cervicomedullary whiplash injury ; increased osmolarity of the subdural haematoma, causing
- vascular injury: carotid compression and
ingress of water; (3) continued fresh bleeding; or (4) a true
vertebrals in whiplash;
inflammatory pachymeningitis elicited by red cell break down.
• ischaemia-hypoxia, secondary shock and cerebra l It has two components: a liquid low-haematocrit
oedema with raised rcp; 'haematoma', with evidence of continued fresh bleeding, and
• contusion from impact injury. a membrane. The membrane, which is vascular and easily
bleeds, encapsulates the haematoma and binds it to the dura,
where it undergoes degradation and invasion by fibroblasts.
SUBDURAL HAEMATOMA Calcification may be detected by 3 months. Incorporation of
A haematoma in the subdural space exerts pressure on the the haematoma into the dura as a membrane is the basis of
underlying brain parenchyma with secondary ischaemia the healing process. Brown staining of the dura, resulting
to the immediately underlying white matter and, later, to from deposition of haemosiderin during haemoglobin break
the grey matter. It also exerts pressure on the overlying down, persists for many months (see Fig. 11.1, p. 206).
calvarium and may cause erosion of the inner table of the
skull.
An acute subdural haematoma occurs within 3 days of Table 14.6 Conditions associated with, or predisposing to,
injury a nd is usually associated with severe shock and subdural haemorrhage or hygroma
brain contusion as well as oedema. A subacute haematoma
occurs between 3 days and 3 weeks of injury, and a chronic Birth trauma
subdural haematoma appears more than 3 weeks from the Accidental trauma
time of injury.14B Non-accidental head injury
Aspergillosis
Kawasaki's disease
Osteogenesis imperfecta
Glutaric aciduria type 1
Meningitis (pneumococcal or group B streptococci)
Septicaemia
Haemophilia/factor 5 deficiency/anticoagulant treatment
Idiopathic thrombocytopenic purpura
Malignancy
Atrioventricular malformation or aneurysm
Post-cardiopulmonary bypass
Alagille's syndrome
Disseminated intravascu lar coagulation
Menkes' kinky hair disease
Slit ventricle syndrome
Hyperosmo lar dehydration from mannitol
Wide extra-axial space
Prema tu ri ty
Physiological
Familial 7Autosomal dominant
Renal dialysis
Severe dehydration
Congenital subarachnoid cyst
Figure 14.13 Primary brainstem haemorrhage at pontine level
Chronic subdural haematoma
in a 3-year-old girl, who fell 30 feet.
~ ---=- --
- - --
Secondary mechanisms of brain injury I 307
INTRADURAL HAEMORRHAGE Strich l5J first described primary shearing of whi te mat
Intradural haemorrhages are not specific for NAHI ; they ter. Long axons, such as those found in the commissural
are usually microscopic, venous/capillary in origin and fibres and long association tracts, can be avulsed by trac
equivalent to intradural petechiae, and have been described tion but axonal damage can also be seen in hypoxic
after fetal, perinatal and later neonatal death and have ischaemic damage. Further projection fibres from the cor
been put fonrvard as a source of subdural haemorrhage in tex may be sheared at the grey-white interface and be seen
infants.149 They are likely to be due to agonal anoxia in as petechiae on MR imaging. 14J It is likely that both white
many cases, although other causes of hypoxia may be matter shearing and white matter oedema contribute inde
pendently to the gross cerebral atrophy of the white matter
involved.
and ex vacuo ventricular dilatation.
The role of intradural haemorrhage as a source of sub
dural haemorrhage is an essential component of the 'uni The infant's axons may stretch more than fully myeli
fied theory' of causation of the triad of brain swelling, nated adult fibres l54 but, when sheared by traction forces,
subdural and retinal haemorrhages, which occurs as a they exude axoplasm, which appears as 'retraction balls'
on staining, followed by microglial scars after 2 weeks and
direct result of apnoea in cases when there is no pericranial
Wallerian degeneration after 6-8 weeks.1 55, 156
impact injury or injury elsewhere tha t might be interpreted
as inflicted. 149 A recent study of 82 infants with White matter shearing injuries are seen, in particular, in
hypoxic-ischaemic encephalopathy from a variety of the corpus callosum, the superior cerebellar peduncle and
in the midbrain.
causes found no concomitant subdural haemorrhage in any
case, suggesting that other factors are necessary to produce It has alternatively been suggested that hypoxic
ischaemia, resulting in vascular axonal injury, may be
subdural haemorrhage in infants . 150
the main cause for the axonal damage in shaken baby
syndrome. 116,11 7
SUBARACHNOID HAEMORRHAGE
(b)
Hae matoma
Cerebral oedema
Impairment of circulation
Rai se d ICP
Vasospasm
Infec ti on
Epil epsy
Hydrocephalus
Hypoxia
Hypercarbia
Hypotension
Fever
Ana emi a
Hyponatraemia
Hypoglycaemia
50 per cent during tra nsit and > 65 per cent in the ICU dur
ing th e subsequent 4 hours of continuous monitoring.
From the TCDB da ta , a significant increase in mortality
and morbidity w as associated with hy poxia or hypotension
from injUly throu g h resuscitation in the 717 adults who
were st udied after head injury.1 57 The relationship of out
come and secondary physiological insults occurrin g during
the intensive care management after head injUly was deter
mined in a prospective study involving 124 adults by Jones
and coll eag ues, 53 who used a computerized data collection
system to measure up to 14 clinica lly indicated physiolog
ical v ariables (such as rcp, arterial blood pressure, CPP,
heart rate, oxygen saturation, and core and periph eral tem
perature), minute by minute. 53 Abnormal valu es for each
~- -~~-
--~~~~-~
~
Secondary mechanisms of brain injury I 309
variable falling outside the pre-set threshold limits as of abnormal physiological values, defining raised ICP,
defined by the Edinburgh University Secondary Insult hypotension and hypertension at all ages is too simplistic.
Grading Scheme for ?5 minutes were analysed by insult Jones and co-workers l77 from Edinburgh used age
grade and duration. Ins ults were found in 91 per cent of specific physiologica l norms to quantify secondary phys io
patients, regardless of the severity of head trauma, their logical derangements prospectively in 54 children aged
age or the admission injury severity score (ISS).53 The < 16 years with TBI. They found that secondary physiolog
cumulative durations were much greater than previously ical insults occurred frequently in children after head
recorded and the most significant predictors of m0l1ality in injury, w ith rcp and CPP derangement demonstrated in
this study were durations of hypotensive, pyrexic and 95.45 per cent and 77.27 per cent of patients respectively.
hypoxaem ic insults. 53 Hypotensive insults and pupillary Furthermore, the duration of CPP derangement was found
response on admiss ion were significant predictors of good to predict outcome (dead versus alive, P = 0.003).177 The
versus poor outcome. 53 point at which physiolo gica I derangements become impor
Only a small amount of research has been done with tant, i.e. by inducing ischaemi a in terms of magnitude and
regard to investigating the clinical aspects and patho duration, remains unclear, as does their effect on outcome.
physiology in children with head injury apart from a few A recent two-centre study 178 involving 86 head-injured
specialized studies 167, 168 that primarily examined CBF, children, aged from 2 to 15 years, admitted within 24 hours
cerebral hyperaemia, cerebral metabolic rate of oxygen of injury, had prospective time series physiological data
(CMR0 2) and arteriovenous difference of oxygen (AvDOJ, downloaded from ICU monitors every minute for 6 physio
This study found that cerebral hyperaemia was uncommon logical variables (ICP, MAP, CPP, oxygen saturation, tem
(7 per cent) a nd that the CMR0 2 and cerebral oxygen perature and heart rate) . Critical thresholds of CPP were
extraction was w ithin the normal range in 81 per cent of identified and, ideally, CPP should be kept above 48 mmHg
children with severe head injuries. Cerebral metabolic rate for children aged 2-6 years, above 54mmHg for children
of oxygen and cerebral oxygen extraction fell between the aged 7-10 years, and above 58 mmHg for children aged
first and third day after the injury, and although the CBF 11-16 years. Duration of CPP was found to be the best pre
rose throughout the ac u te monitoring time this was not dictor of survival (P < 0.000l), and a highly significant
significant. These findings led the investigators to conclude discriminator between good (Glasgow Outcome Score
that children were most vulnerable to secondary brain [GOSs] 4 and 5) and poor outcome (GOSs 1, 2 and 3)
damage short.ly after the injury. In addition, they found (P = 0.005). A new cumulative pressure-time index (CPT)
that cerebrovascular autoregulation was preserved (signifi was developed to include severity and duration and the
cant correlation between CPP and central venous pressure CPT was useful in predicting independent/poor outcome. 179
[CVP] , P = 0.0003), the cerebrovascular resistance (CVR)
was normal (in 58 per cent), increased (in 32 per cent) and
CEREBRAL OEDEMA
could be impaired in the most severely injured. The ICP
was inversely propol1ional to the CBF (P = 0.009) and all Cerebral oedema is defined as an increase in the volume of
parameters (CMR0 2, CPP, 0 2/ partial pressure of arterial car the w hole (generalized) or part (focal oedema) of the brain
bon dioxide [Paco2], cerebrovenous pH) were independent due to an increase in the water content. 180 Cerebral oedema
determinants of CVR. 169 is a frequent accompaniment of shaking injuries and has
Another complicating factor for neurotrauma research been estimated in as many as 42 per cent of such infants,
involving children is that the normal ranges for different based on imaging. 181 The frequency is doub led in post
physiological param eters, such as ICp l70 and blood pres mortem studies and Geddes ll6 ,117 confirmed that cerebra l
sure,171-174 are age related and change from birth through oedema was a major cause of death in 82 per cent with evi
childhood and adolescence to finally reach adult levels. Pre dence of hyp ox ic ischaemia in 77 per cent. It takes approx
vious studies by the TCDB 23 grouped all children aged 0-15 imately 6 hours to appear after the injury.!52
years together and used the same end-hour threshold as that Cerebral oedema has several mecha nisms of production,
used in their adult studies. Jackson's groupl75 attempted to as see n in Table 14.7.182 The cause of the oedema in the
quantify the CPP secondary insults in children using only shaking injuries is probably multifactorial, with vascular
three different adult CPP thresholds for their patients who damage (as in the eye) causing vasogenic oedema, white
were aged 16 years or less. Similarly, Sharples and co-workers matter damage. disrupted venous damage from shearing,
used a single threshold value of > 20 mmHg to define raised damage to the blood-brain barrier, high central venous
ICP in their study involving child ren aged 2- 16 years 01d. 166 pressure from chest compression, brain necrosis from
In a more recent study, the cause and incidence of second shock and imp aired perfusion, and secondalY hydro
ary insults in severely head -injured ad ults and children cephalus causing hydrostatic oedema from obstruction of
were examined by Chambers and colleagues. 176 Again, the arachnoid granulations by blood.
unified threshold values for CPP, ICP and arterial blood Cerebral oedema results in a deteriorating level of con
pressure were used on their patients, regardless of their age sciousness. with signs of brainstem dysfunction in pulse,
and developmental maturity.176 Using a standard definition respiration, blood pressure and pupils. which may progress
310 I Head and neck injuries
to signs of 'coning'. Coma itself maybe due to raised ICP or Cerebral Infarction from Impaired Perfusion
to primary midbrain injury. Extensor hy pertonus follo wing Brain infarction occurs as a result of hypotension,
asphyxia may be due to oedema, but is more often a dys raised ICP, brain shifts and vascular occlusions, spasm or
tonia as a result of basal ganglia involvement - i.e. post obstruction.
as phyxial rigidity, which is not influenced by treatment to Glucose is the main energy source for the brain: under
reduce ICP and which is often normal or secondary to aerobic conditions 38 molecules of ATP are produced but
necrotic oedema. I S3 only two are produced under anaerobic conditions. How
Imaging shows a reduction in lateral ventricular size ever, under anaerobic conditions the supply of glucose and
followed by third ventricle size and, eventually, the ambi removal of lactic acid is essential. The ATP produ ced is
ens cistern. The imaging loss of the 'sm iling face' means it required for neurotransmitter formation, transpOli mecha
is imperative that a lumbar puncture is not performed as nisms, maintenance of membrane pumps and polarization
there is some existing tentorial herniation. Scans show of membran es.
hydrocephalic oedema as a hypod ense 'bat wing' around The oxygen supply depends on the POz and the haemo
the lateral ventricles. It is now possible to measure brain globin concentration. Ischaemi a is far more dangerous
water using MRI technology. than hypoxia because there is now no glucose or ketones,
or any way of removing lactic acid, so neuronal necrosis
rapidly occurs.
HYPOXIC/ISCHAEMIC INJURY
Ischaemia initially results in swelling of the mitochondria
Ninety per cent of children dying from accidental head within the neurones and astrocytes as a result of water reten
injuries have evidence of severe hypoxic ischaemic brain tion. The neurones release potassium, which is collected by
damage, 155 whereas 77 per cent of children dying from NAHl the astrocytes; glycogen also accumulates within the astro
are found to have hypoxic ischaemic damage at autopsy. 116, 11 7 cytes, ca using further swelling. ls4 Second, there is leakage of
The hypox ic/ ischaemic damage may be part of the pri protein (and blood - haemorrhagic infarct) through the
mary injury (apnoea or medullary injUly) or result from endothelial cells into the extracellular space. Subsequent
secondary insults such as hypotensive shock, raised ICP, breakdown of the red cells causes iron and bilirubin forma
decreased perfusion pressure and seizures. The 'big black tion in the tissues, which is used to date the infarct. Finally,
brain' indicates particularly severe hypoxic ischaemic dam lysosom al rupture, cell death and biochemical breakdown of
age, which is followed by death or the rapid development tissue proteins results in swelling of the infarct (further com
of cerebral atrophy. Additionally, suffocation may have pressing capillaries), breakdown of the blood-brain barrier,
occurred to quell the child's crying and hypoxic ischaemia loss of autoregulation, and sludging (from thromboplastin
is almost always an agonal event in most infant deaths. release) and liquefaction of the tissues.
Injury to the cervical spinal cord I 311
The necrotic tissue may be absorbed by macrophages vehicular accidents are most common causes in children,
leaving a cyst - as in periventricular leucomalacia - or may followed by sports-related injuries (e.g. rugby and horse
cause an astrocytic reaction, replacing the dead tissue with riding). Although the incidence is lower in children than
the formation of a gJial scar and capillary proliferation. in adults, it carries a worse prognosis and involves a high
Six to twelve hours after acute hypoxic ischaemia there incidence of severe, permanent neurological disability due
may be a therapeutic window, possibly due to a biochemical to spinal cord involvement. 190
cascade triggering apoptosis or a slowing down of metabo The anatomical and biomechanical characteristics of the
lism caused by adenosine and adrenocorticotrophic hormone. child's spine are very different from those of the adult and
the cervical spine is particularly at risk because of the high
weight and volume head-body ratios, ligamentous laxity
Patterns of Infarction and incomplete head control. Younger children have more
cervical lesions - the vertebrae are not yet ossified, the unci
Hypotension does not result in uniform infarction of the
nate processes are flat and facet joints are horizontal, result
whole brain. 18S The four feeding arteries to the circle of
ing in a different pattem of cervical spine injury in children.
Willis means there are areas where the pressures are equal,
Fractures involve the growth plates. 191 Zabramski et al 192
resulting in 'no flow', and ischaemia may occur in any
reported that 71 per cent of children with spinal injuries who
major vessel from Jow perfusion pressure.
were < 10 years old had cervical lesions, and that one-half of
When cerebral blood flow is severely impaired, water
these lesions were between the occiput and C2.
shed zones between adjacent vascular territories are suscep
There are six physical mechanisms that give rise to
tible. In the newborn this is between centrifugal and
spinal and spinal cord injUly:
centripetal arteries, causing periventricular leucomalacia. In
I. flexion dislocation;
the older patient it is between the middle and posterior cere
2. flexion compression;
bral arteries (causing infarcts in the pericentral white mat
3. compression burst fractures in which bone fragments
ter of the optic radiations and posterior temporal lobe) or
are exuded into the spinal canal;
between the anterior and middle cerebral arteries (causing a
4. spinal extension versus penetrating injuries;
wedge shaped infarct in the 'leg area'). This susceptibility is
5. distraction-rotation injury. 193
because cerebral arterioles are 'end arteries' and do not
anastomose.
Upper cervical spine injuries take the form of atlanto
Patterns of infarction also result from brain herniations
occipital dislocation, from birth or other trauma, and odon
pericallosal or subfalcine shifts, posterior cerebral (from
toid fractures that, in small children, are really epiphyseal
tentorial herniation) and posterior inferior cerebellar artery
detachments. Cervical dislocations and compression frac
(PICA) from foramen magnum coning.
tures are most common in older children. Typical Jefferson
There is also selective vulnerability of cell types: neu
fractures, Hangman fractures and cervical strains are
rones are the most sensitive, whereas microglia and blood
exceedingly rare.
vessel cells are the least sensitive. Ischaemia occurs in the
From these mechanisms of injury, primary spinal cord
brain-neocortex, basal ganglia, hippocampus and cerebel
injlllY results from contusion and compression of the spinal
lar Purkinje cells. 186 In head injuries, hypoxic ischaemic
cord, which, within a few hours of injury, results in either
damage occurs to these selective areas as well as watershed
cord oedema with ischaemia or occlusion of the
zone infarction. 12
intramedullary vessels, resulting in haemorrhagic necrosis.
With resolution the necrotic areas leave cavitations or
syrinxes in the cord at the level of the injury. 194
INJURY TO THE CERVICAL SPINAL CORD
Pang and Wilberger llO described spinal cord injury with
out radiographic abnormalities (SCrwORA), which is a
Injuries to the spinal cord in children are rare. They occur
traumatic myelopathy without evident fractures (due to
during birth, and after accidental and non-accidental
hyperflexion or hyperextension, or distraction or repetitive
injuries. In a study by Augutis and Levi 187 the incidence was
flexion-extension injuries), with primary damage to the
found to be 4.6 per million children per year (95 per cent,
cord, or cord injury and ischaemia secondary to hypoten
confidence intervals 3.6-5.5); however, when pre-hospital
sion, or interference with the vertebral artery blood flow
fatalities were excluded the incidence was 2.4 (95 per cent
(bruising of the adventitia with intramural clots or occlu
confidence intervals 1.8-3.1). Children's spinal injuries rep
sion).195 Bollini 196 estimated that almost one-half of such
resent about 5-10 per cent of all spinal injuries.188.189 In
lesions were in the cervical region.
most series, boys predominate.
Spinal shock with a total flaccid paralysis below the
level of the region will last for a variable time after the ini
ACCIDENTAL SPINAL INJURY
tial spinal injury until cord swelling resolves. Painful stim
The aetiology of accidental spinal injury in children is dif uli wiU not elicit flexion of the limbs. The neurological
ferent from that in adults; falls and pedestrian-related findings will point to the classical syndromes of cord
312 I Head and neck injuries
involvement, i.e. complete, posterior cord or anterior spinal column can move in the anteroposterior direction, allow
artery territory, central cord, root involvement or Brown ing a degree of slippage (spondylolisthesis). Whiplash
Sequard's syndrome cord lesion, whether the spine is stable injury can therefore cause cord concussion fro m repe ated
or unstable. 197 partial subluxations, as well as traction with bruising and
Treatment for the above spinal injuries may take the vertebral artery and anterior sp inal artery lesions.
form of halo distraction or other traction, occipitocervical The presentation of infants with cervicomedullary
fixations with bone grafts or progressive traction with injury is with apno ea and seco ndary hypoxic damage.
transcranial tongs. Spine boards are not advisable because Death from hypoventilation and apnoea may therefore be
they tend to force the large head into flexion, which can from a primary injury to the respiratory centre in the
compromise both airway and spine. Spinal cord injuries medulla or from associated vaso pa ra lytic shock w ith loss
in children are prone to result in spinal deformity because of vasomotor tone, also from medullary injury. The result
of imbalance between the paralysed muscles involved in is severe hypoxic ischaemic damage with oedema. These
postural control, or assymmetric lesions of growth plates. cases highlight the need for the MRjCT scans to include the
Syringomyelia is common at the site of the injury but may cervical spine in all suspected cases and for the cord to be
extend over several segments when the classical symptoms examined at necropsy. Cullen 200 and Swischuk 201 described
of head and neck pain, made worse on straining, alert one compression fractures, subluxations, and fracture disloca
to the di agnos is. Survivors have a high frequency of bulbar tions in thoracolumbar levels (and one thoracic level
and associated brainste m deficits. lesion) in 11 in fants suffering NAl.
REFERENCES 25 Arnarson EO, Halld o rsson JG. Head trauma among children in
Reykjavik. Acta Paediatr 1995 ; 84:96-9.
26 Jennett B. Epidemiolo gy of head injury. ] Neurol Neurosurg
Teasdale G, Jennett B. Assessment of coma and impaired
Psychiatry 1996; 60: 362 - 9.
consciousness. A practical scale. Lancet 1974; 2 :81-4.
27 Gabella B, Hoffman RE, Marine WW, Stallones L. Urban and
2 Simpson D, Reilly P. Pediatric coma scal e. Lancet 1982; 2:450.
rural traumatic brain injuries in Co lorado . Ann Epidemiol
3 Raimondi AJ, Hirschauer J. Head injury in the infant and
1997; 7:207-12.
todd ler. Coma scoring and outcome scale. Childs Brain 1984·
28 Emanuelson I, Wendt L. Epidemiology of trauma tic brain
11 :J 2-35. '
injury in children and adolescents in so uth- western Sweden.
4 Jennett B. Teasdale G, Galbraith S et al. Severe h ead injuries
Acta Paediatr 1997; 86:730-5.
in three countri es.] Neu ral Neurasurg Psychiatry 1977 ;
29 Suominen P, Kivioja A, Ohman J e [ al. Severe a nd fatal
40 :29 1-8.
48 Hold en KR, Titus MO, Van Tassel P. Cran ial magneti c 70 Artru F, Terrier A, Gilbert let al. Monitoring of intracranial
resonance imagi ng exam ination of normal term neonates : pressure with intraparenchymal fiberoptic transducer.
a pilot study.) Child Neurol 1999; 14 :708 - 10. Technical aspects and clinical reliability. Anll Fr Anesth
49 Yonas H, Wo lfson SK Jr, Dujovny Met al. Selective Real1im 1992; 11 :424 -9.
lenti culostriate occlusion in the primate. A hig hly focal 7J Pople IK, Mahlbauer MS, Sanford RA, Kirk E. Results and
cerebral ischemia model. Stroke 1981 ; 12 :567 -72. complications of intracranial pressure monitoring in
50 Shaffer L, Rich PM , Pohl KR, Gan esan V. Can mild hea d injUly 303 children. Pediatr Neuro su rg 1995; 23: 64-7.
cause ischaemic stroke? Arch Dis Child 2003; 88: 26 7- 9. 72 Jensen RL, Hahn YS, Ciro E. Risk factors for intrac ranial
51 Ko rs EE, Terwindt GM, Vermeu len FL et al. Delayed cereb ral pressure mon itoring in children with fiberoptic devices:
edema a nd fatal coma after minor head trauma: role of the a critical rev iew. Surg Neuro/ 1997; 47:16- 22.
CACNA 1A calcium channe l subunit gene and relation ship with 73 Meyer JS , Kondo A, Nomura F et al. Cerebral hemodynamics
familial hemiplegic migrai ne. Ann Neurol 2001; 49:7 53 - 60. and metabo lism following experimental head injUlY.
52 Umile EM, Sand el ME, AJvaria A et al. Dynamic imaging in ) Neuro surg 1970; 32:304-1 9.
mild traumatic brain injury: support for the theolY of medial 74 Kaiser AM , VVhitelaw AG. Hypertensive response to raised
tempo ral v ulne rability. Arch Phys Med Rehavil 2002; intracrani al pressure in infancy. Arch Dis Child 1988; 63:1461 -5.
83:1506-13. 75 Rossi S, Zanier ER , Mauri I et a l. Brain temperature, body core
53 Jones PA, Andrews P J , Midgley S et al. Measuring the burden temperature, a nd intracranial pressure in acute cerebral
of seco ndary insults in head-inju red patients during intensive damage. ) Neurol Neuro surg Psychiatry 2001; 71 :448-54.
care.) Neurosurg Anesthesiol 1994; 6:4-14. 76 Albrecht RF II, Wass CT, Lanier WL. Occurrence of potenti ally
54 Jiang JY, Gao GY, Li P, Yu MK, Zhu C. Early indicators of detrim enta l tem perature alterations in hospita li zed patients at
prognosis in 846 cases of severe tra umatic brain injury. risk for brain injury. Mayo Clin Proc 1998; 73:629-35.
) Nellrotrnuma 2002 ; 19:869-74. 77 Hutchinson PJ , al-Rawli PG , O'Connell MT et al. Head
55 Chiaretti A, Pia stra M, Pulitano Set al. Prognosti c factors and inj ury monitoring using cerebral microdialysis and
outco me of children with severe head injury: an 8-year Paratrend multipara meter sensors. leY/trolvl Neurochir 2000;
expe rience. Childs Nerv Syst 2002; 18: 129-36. 61 :88-94.
56 Chiaretti A, De Bened ictis R, Della Corte F et al. Th e impact of 78 Gupta AK, a l-Rawli PG, Hu tchinson P J, Kirkpatrick P J. Effect
initial management on the outcome of children wit h severe of hypothermia on bra in tissue oxygenation in patients with
head inju ry. Childs Nero Syst 2002; 18:54 -60. severe head inju ry. Br) Allaesth 2002; 88: 188-92.
57 Coats TJ, Ki rk CJ, Dawson M. Outco me after severe head 79 Smith LW, Fay T. Obse rva tions on human beings with cancer,
injury treated by an integrated trauma system. ) Acc Emerg maintained at reduced temperature of 75- 90 degree
Med 1999; 16 :1 82-5. Fahrenhe it. Am) Ciin PatllOl 1940; 10:1-11.
58 Zauner A, Bullock R, Di X, Young HF. Brain oxygen, CO 2 , pH, 80 Fay 1. Ea rly experiences with local and generalised
and temperature monitoring: eva lu ation in the fe lin e brain. refrigeration of the human bra in.) Neurosurg 1959;
Neurosurgery 1995; 37:1168-76 ; discussion 1176 -7 . 16:239 - 60.
59 Jeevaratnam DR, Menon OK. Survey of intensive ca re of 81 Clasen RA, Pandolfi S, Russell J et a!. Hy pothermia and
severely head-injured patien ts in the United Kingdom. BM) hypotension in experimental cerebral edema. Arch Neurol
1996; 312:944- 7. 1968; 19:472-86.
60 Matta B, Meno n D. Severe head injury in the United Kingdom 82 Laskows ki EJ, Klatzo I, Baldwin M. Experimental study of the
a nd Ireland: a survey of practice and impl ica tions for effects of hy pothermia o n local bra in injury. Neurology 1960;
management. Crit Care {"fed 1996 ; 24: 1743-8 . 10:499- 505.
61 Segal S, Gall ag her AC, Shefl er AG et a\. Survey of the use of 83 Rosom off HL, Gilbert R. Brain vo lum e and cerebrospinal fluid
intracrani al press ure monitoring in children in the United pressure during hypothermia. Am) Physiol 1955; 18 3:19 - 22 .
Kingdom. intensive Care Med 2001; 27:236 - 9. 84 Rosomoff HL, Shulm an K, Raynor R, Grainger W.
62 Gambardella G, Zaccone C, Cardia E, Tomasello F. Intracranial Experimental brain injury and delayed hypothermia.
pressure monitoring in ch ildren: comparison of ex ternal Surg GYl1ecol Obstet 1960; 110: 27-32.
ventricular device wi th the fiberopt ic system. Childs Nem 85 Clifton GL, AJlen S, Barrowdale P et a!. A phase II study of
Syst 1993; 9:470-3 . mod era te hypothermi a in severe brain injury. ) Neurotrauma
63 Piek J, Bock WJ. Continuous monitoring of cereb ral tissue 1993; 10:263-71; discussion 273.
pressure in neurosurgical practice: experienc es with 100 86 Marion DW, Oozeer R, de Vries L et a l. The use of mod erate
patients. ill tellsil!e Care IVIed 1990; 16: 104-8. therapeutic hypoth ermia for patients with seve re head
64 Yablon JS, Latner HJ, McCormack TM et a l. Clin ica l injuries: a preliminary re port.) Neurosurg 1993; 79:354-6 2.
experience with a fiberopti c intracrania l pres sure monitor. 87 Cli fton GL, Miller ER , Choo i SC et al. Lack of effect of
) Ciill Moni t 1993; 9:171-5. induction of hypothermia after acute brain injUly. N Engl
65 Shapiro S, Bow man R, Callahan J, Wolfla C. The fiberoptic ) Med 2001; 344:556- 63.
intra parenc hymal cerebral pressure monitor in 244 patients. 80 Connell J, Oozeer R, de Vries L et al. Continuous EEG
Surg Neurol 1996; 45:278-82. monitoring of neon ata l seizures: diagnostic and prognostic
66 Friedman WA, Vries JK. Percutaneous runnel ventri culostomy. co nsideratio ns. Arch Dis Cilild 1989; 64:452 - 8.
Summary of 100 procedures.) Nellrosurg 1980; 53:662-5. 89 Dusse r A, Na valet y, Dev ictor 0, Landrieu P. Short- and long
67 Narayan RK, Kishore PR, Becker DP et al. Intracranial term prognostic value of the electroencephalogram in children
pressure: to monitor or not to monitor? A rev iew of our with seve re head injury. Electroellcep iIalogr Clin Neurophysiol
experience with severe head injury. ) Neuro surg 1982; 1989; 73 :85-93.
56:650 - 9. 90 Altafu ll ah I, Asa ik ar S, Torres F. Status epilepticLls: clinical
68 North B, Reilly P. Compariso n amo ng three m ethods of expe ri ence w ith two spec ial devices for continuous cereb ra l
intracranial pressure recording. Neurosurgery 1986; 18:730-2. mon itoring. Acta Neurol Scand 199 1; 84:374 - 8 1.
69 Param ore CG, Turner DA Re lat ive risks of ventriculostomy 91 Aoki Y, Lombroso C. Progno stic valu e of
infecti on and morbidity. Acta Neurochir (Wi en) 1994; electroencephalography in Reye' s syndrome. Neurology 1973;
127 :79-8 4. 23:333-43.
References I 315
136 Helfer RE, Siovis TL, Black M. Injuries resulting wh en sma ll 158 Graham Dl, Ford I, Adams JH. Ischaemic brain damage is
children fall out of bed . Pedia trics 1977 ; 50:533-5. still common in fatal non-missile head injuI)'. J Neural
137 Margulies SS, Thibault KL Infant skull and suture properties: Neurosurg Psyclliatry 1989; 52:346-50.
measurements and implications for mechanisms of pediatric 159 Miller JD , Bec ker DP. SecondaI)' insults to the inj ured brain.
brain injury. J Biol11ech Eng 2000; 122 :364-71. J R Call Surg Edill 1982; 27:292-8.
138 Scarfo GB, Mariottini A, Tomaccini D, Palma L. Growing 160 Gentleman D, Jennett B. Haz ard s of inter-hospital transfer of
sku ll fractures: progressive evolution of brain dam age and comato se head -injured pati ents. Lan cet 1981; 2:853-4.
effecti ve ness of surgica l treatme nt. Ch ilds Nen) Syst 1989; 161 Kohi YM, Mendelo w AD, Teasdale GM, Allard ice GM .
5:163-7. Extra c ra ni al insults and outcome in patients w ith ac ute head
139 Carty H. Training in paedi atri c radiology. Clin Radial 1989; inju I)': relat io nship to th e Gla sgow Coma Sca le. Inju ry 1984 ;
40:227-8. 15 :25-9.
140 Royal College of Radiologists. Guidelinesfor Doctors, 5th 162 Miller JD, Bu tte rworth JF, Gud ema SF et al. Furthe r
ed n. London: Royal College of Rad iologists, 200 3. experien ce in the managem ent of severe head injuly.
141 Zi mmerman RA, Bilianiuk LT, Bruce D et al. Computed J Ne urosurg 1981; 54:289-99.
tomography of craniocerebral injuI)' in the abused chi ld. 163 Miller JD , Sweet RC. Na raya n R, Becke r DP. Early insults to
Radiology 1979; 130:687-9 0. the injured brain. JAMA 1978 ; 240:439-42.
14 2 Johnson H, Fae rber EN (ed.). Trauma. In CNS Magnetic 164 Price DJ, MUlTay A. The influence of hypoxia and hypotension
Resonance Imaging in Infants and Children. Lond on : Mac Oll recov ery fro m head inju I)'. InjUly 1972; 3:218-24.
Keith Press, 1995, pp. 98- I I 5. 16 5 Marmarou A, Anderson RL, Ward JD et al. NIND S Traumatic
143 Barlow KM, Gibson RJ, McPhillips M, Minns RA. Magnetic Coma Data Bank : Intracranial press ure moni toring
reso nance imaging in acute no n-accid e ntal head inju ry. Acta methodo logy. J Neurosurg 199 1; 75 (Suppl.): 21 -7 .
Paediatr 1999; 88:734-40. 166 Andrews PJ, Piper IR , Dea rd en NM, Miller JD . Seco nda I)'
144 Biousse V, Suh DY, Newman NJ et a l. Diffusion-weighted insults during intrahospital transport o f head -injured
magnetic reso nance imagin g in shaken baby syndrome. patients. Lancet 1990; 335:327-30.
Am J Oph tha11110 12002; 13 3:249- 55 . 167 Sha'lJles PM, Matthews DS, Ey re JA. Cerebra l bloo d flow
145 Hovi nd KH. Traumatic brain injuries. In Raimondi AJ, Choux and metabolism in children wit h severe head iIJjuI)'. Part I:
M, DoiRocco C (eds) Head Injuries ill the Newborn and Relation to age, Glasgow Coma Score, outcome, intracranial
Infant. New York: Springer-Verlag, 1986, p. 87. pressure, and time after inju I)'. J Neural Nellrosu rg
146 Duhaime AC, Genarelli TG, Thibau lt LE et aJ. The shaken Psychiatry 1995; 58:145-5 2.
baby synd rome. A clinical, pathological, and biomechanical 168 Muizelaar JP, Ward JD, iVlarmaron A et al. Cerebral blood
study. J Neurosurg 1987; 66 :409- I 5. flow a nd metabolism in severely heael-injured children. Part
147 Morison CN, Minns RA. Biomechanics of shaking. In Minns 1: Relationship with GCS score, outcome, ICP, and PVJ.
RA, Brown JK (eds) Shaking and Other Non-accidental Head J Neurosurg 1989 ; 71:63-71.
Inju ries ill Children. London: Mac Keith Press, 200 5, pp. 169 Sharples PM, Matthews DS, Eyre JA. Cerebral blood flo w
105-46. and metabolism in children w ith severe head injuries. Part 2:
148 Choux M, Lena G, Genitori L. Intracranial hematomas. In Cerebrovascu lar resistance and its determ inants. J Ne urol
Raimondi AJ, Choux M, DoiRocco C (eds) Head Iiljuries Neurosurg Psychiatry 1995; 58:153-9.
in the Newborn and Infan t. New York: Springer-Verlag, 170 Minns RA. Infectious and para infectious encep halopathies.
1985, p. 203. In Minns RA (ed.) Problems of Intracrallial Pressure in
149 Gedd es JF, Tasker RC, Hackshaw AK et aJ. Dural Childhood. London: Mac Keith Press, 1991 , pp. 170- 282 .
haemorrhage in non-traunJatic infant deaths: does it explain 171 Report of th e Second Task Fo rce on Blood Pressure Control
the bl eeding in 'shaken baby syndrome'? Neuropathol Appl in Children: 1987. Task Force on Blood Pressu re Control in
Neurobiol 20 03; 29: 14-2 2. Children . National Heart, Lung, and Blo od Institute,
150 Byard RW, Blumberg P, Rutty G et al. Lack of evidence for a Bethesda, i\IlaI)'land. Pediatrics 1987; 70:1- 25.
causal relationship between hypox ic/ischaemic encep halopathy 172 Donmall Me. Anthropom et ric and blood press ure studies in
and subdural haemorrhage in fetal life, infancy and ea rly children fro m Shetland: reliability of measurement and the
childhood. Pediatr Dev Pathol 2007; 10:348-50. use of heig ht- we ig ht indi ces. Acta Paediatr SC(ll1d 198 5;
151 Ommaya AK, Gennarelli TA. Cerebral concussion and 3 18 (Suppl.) :23-35.
traumatic unconsciousness. Correlation of expe rimental and 173 Ro cce ll a EJ. Update on the 19~7 Task Force Report on Hi gh
clinica l observ ations of blunt heael injuri es. Brain 1974; Blood Pressure in Children and Adolescents: a working
97 :633-54. group report from the National High Blood Pressure
152 Jaspa n T, Narbrou gh G, Punt J A, Lo we J. Cerebral Educatio n Program. National Hig h Blood Pressure Education
contusional tea rs as a marker o f child abuse: detection by Program Wo rkin g Group o n Hypertensio n Control in
crania l sonography. Pediatr Radial 199 2 ; 22:237 -45. Children a nd Adolescents. Pediatrics 1995; 98:649 -58.
153 Strich SJ. Shearing of nerve fibers as a course of brain 174 Rosner B, PrineasRJ , Logg JM, Daniels SR. Blood pressure
damage due to head inju I)'. Lan cet 1961; 2:443-8. nomog rams for children and adolescents, by height, sex, and
154 Crompton R. Closed head injuI)' in children. In Cromp to n R age, in the United States. J Pediatr 1993; 12 3 :871-85.
(ed.) Closed Head InjUly in Children. Lo ndo n : Edward 175 J ackso n S, Piper JR , Wagsta ff A, Souter M. Quantification of
Arno ld , 1986. cerebra l perfu sion pressure (CPP) 'secondaI)' ins ults' in
155 Adams JH, Mitchell DE, Graha m Dl, Doyl e D. Diffuse brain child ren: a study of the effec ts o f using diffe ren t CPP
damage of immediate imp act type. Its relatio nship to 'primaI)' thres holds. British Journal of Neu roSlirgery, 1999; 13: 104.
brain -stem damage' in head injuI)'. Brain 1977 ; 100:489-502. 175 Chambers IR, Treadw ell L, Mendelow AD. The cause and
155 Vowl es H, Scholtz CL, Cameron JM. Diffuse axona l injuI)' in incidence of secon daI)' insults in severely head-injured
ea rly infa ncy. J Clin Pathol 1987 ; 40: 185-9. adu lts and children. Br J Ne urosll rg 2000; 14:424-31.
157 Ches nut RM, Marshall LF, Klauser MR et al. The role of 177 Jo nes PA, A ndrews PJ, Eastman VJ, Minns RA. Traumatic
secondaI)' brain injuI)' in determining outcome from severe brain injuI)' in chi ldhood: Intensi ve care time-series data
head injuI)'. J Trauma 1993; 34:216- 22 . and o utcome. Br J Neurosurg 2003 ; 17: 29 -39.
References I 31 7
178 Jones PA, Easton VJ, Anderson PJD et al. Traum at ic brain 194 Tator CH, Koyanagi T. Vascular mechani sms in the
injury in chi ldren: intensive care time series data and pathophysiology of human spinal cord injury. J Neurosurg
outcome. Br J Neurosurg 1982; 17:29-39. 1997; 86:483 - 92.
179 Ch ambers IR, Jones PA , Lo TYM et al. Critical thresholds of 195 Yates PO. Birth traum a to the vertebral arteries. Arch Dis
ICP and CPP related to age in paediatric head inju ry. J Child 1959; 34:436-41.
Ne urol Nellrosllrg Psych 2007 ; 77:234-40. 196 Bollini G. Thoracic and lumbar spine injuries in children. In
180 Pappius HM, Fe ind el W. Dyna mics of Brai l! Edema. Berlin: Fl oman Y, Fa rcy JPC, Arge nson C (eds) TllOra coiuml1ar Spine
Spri nge r, 1976. Fractures. New York: Rave n Press, 1993, pp. 307 - 25.
181 King WJ, MacKay M, Sirnick A. Shaken baby sy ndrome in 197 Holdsw oli h F. Fractures, dislocations and
Canada: clinical cha racteristics and outcomes of hospital fracture-di slocations of the sp ine. J BOl1e Joint Surg (Am)
cases. Can Med Assoc J 2003 ; 168:155-9. 1971; 52A:1 534-S I.
182 Brown JK, Imam H. Interrelat ions hips of liver and brain with 198 Jones L. Birth trauma and the cervical spine. Arch Dis Child
special reference to Reye syn drome. J Inherit Me tab Dis 1970; 45:1 47 .
1991; 14:436- 58 . 199 Towbin A. Late nt sp in al co rd and brain stem injury in
183 Seshia SS, PN Chow, Sanka ran K. Coma foJlowin g newborn in fants . Del) }\lIed Child Neurol 1969: 11: 54- 68.
cardiorespiratory arrest in chi ldhood. Dev Med Child Neurol 200 Cullen JC. Spina l lesions in bartered babies. J Bone Joint
1979; 21:143-53. SurgelY (Br) 1975; 57B :364-6.
184 Fujimoto T, Walker JTJ , Spatz M, Klatzo I. Pathophys iol ogic 20 1 Sw ischuk LE. Spine and spin al cord trauma in the battered
aspects of ischaemic edema. In Pa ppius HM , Feindel W (eds) child synd ro me. Radiology 1969; 92 :733 - 8.
Dynamics of Brain Edema. Heidelberg: Springer, 1976, 202 Friedman G, Froolll p, Sazbon L et aJ. Apol ipoprotein
pp. 171-80. E-epsilon4 genotype pred icts a poor outcome in surv ivors of
185 Bruce DA, Langfil l m , Mi ll er JD et al. Regional cerebral traumatic brain injury. Neurology 1999; 52:244- 8.
blood flow, intracrani al pressure, and brain metabolism in 203 Lichtman SW, Seliger G, Tycke B, Marder K. Apolip oprotein
comatose patients. J Neurosurg 1973; 38:131-44. E and functi onal recovery from brain injury followin g
186 Dearden NM. ls chaem ic brain. Lancet 1985; 2:255-9. posta cute rehabilitation. Neu rology 2000; 55: 1536-9.
187 Augutis M, Levi R. Pediatric spinal cord injury in Sweden : 204 Utermann G, Langenbeck U, Beisiegel U, Weber W. Genetics
incidence, etiolo gy and ou tcome. Spinal Cord 2003; of the apolipoprotein E system in man. Am J Hum Genet
41 :328-36. 1980; 32:339-47.
188 Hamilton MG, My les ST. Pediatric sp in al injury: review of 61 205 Mah ley RW. ApoJipoprotein E: chol esterol transpOli protein
deaths. J Neurasurg 1992 ; 77 :705-8. with expandin g role in cell biology. Science 19813;
189 Ham ilton MG, Myles ST. Pediatric spinal injury: review of 240:622-30.
174 hosp ital ad miss ions. J Neu rosurg 1992; 77 :700-4. 206 Boyles JK , Pitas RE , Wil so n E et aJ. Apo lip oprotein E
190 Splendiani A, De Amicis R, Pupill o V et a l. Spi nal trauma in associated with astrocy1ic glia of the central nervous system
childho od: a four-y ea r experience at a non-specialis t and with nonmyellnating glia of the peripheral nervous
hospital. RiIJista di Neuroradi%gia 2003; 16:455-7. system. J Clin [n/lest 1985; 76:1501-13.
191 Aufderma ur M. Spinal injuries in juven il es. Necropsy 207 Teasdale GM, Ni cho ll JA , Murray G, Fiddes M. Associati on
find ings in twelve cases. J Bon e Joint Surg Br 1974; of apolipoprotein E polymorphism with outcome after head
56B:513-19. injury. Lancet 1997; 350:1069 -7 1.
192 Zabramski JM, Hadl ey MN, Browner CM. Pediatric spinal 208 Mattson MP. Cellul ar actions of beta-amyloid precursor
co rd and vertebral column injuries. Ban-ow Neural [nst Q protein and its solubJe and fibrillogenic deri va tives. Physiol
1986; 2:11 -17. Rev 1997; 77: 1081-132.
193 Lena G, Bollini G. Spinal inju ries in children. In Chou x M 209 Nicoll JA, Robelis GW, Graham DJ. Apo lipoprotein E epsi lon
et al (eds) Pediatric Ne urosurgery. Lond on: ChurchiJi 4 allele is associated with deposition of amyloid beta-protein
Livings tone, 1999, pp. 313 1-91. fo llowi ng head inj ury. Natu re Med 1995; 1:135-7.
I CHAPTER 15 I
HEAT-INDUCED INJURY OR DEATH
Anthony Busuttil
INTRODUCTION
diminution in the incidence of domestic fire deaths since
1987. Poor housing particularly associated with overcrowd
Deaths in fires are the third leading cause of death in domes ing and poorly built houses present a high lisk of fire deaths.
tic fatalities among children.! In the USA someone dies in a Single-parent families and families from a lower socioeco
fire about every 162 minutes and someone is injured in a fire nomic background, particularly in associatio n with multi
every 32 minutes, and many of these victims are chi ldren ;2 deprivation, show a much higher incidence of fire fatalities.
t he USA ranges only sixth in the incidence of deaths from The misuse of a lcohol, and contro ll ed substances to a
fires among the 25 developed coun tries for which statistics lesser extent, by the adults caring for the child increases
are available.) In retrospect, many of these deaths are found the incidence of deaths in domestic fires, given that the
to have been preventable 4 and this raises a substantial pub adults in such conditions succumb earlier to the fire and,
lic health problem. 5 - 7 Financial and insurance losses are also indeed, being a lready disorientated due to the intoxication
vely substantial. they are much less able to look after t hemselves and after
Thermal injuries are also a common reason for referral the child ren whom they are caring for. This asso ciatio n has
of children to the emergency services, with scalding being been shown to o'ccur both in the UK and in the USA.
the commonest reason for such referrals. Heat-induced Smokers' materials such as cigarettes, matches and
inju ries also form an important part of the gam ut of the lighters are a very frequent cause of house fires. This has
injuries seen in children who have been physically abused led some countries to ensure that cigarettes lighters are
by their carers. manufactured in a child-proof manner.
Deaths in fires usually are the result of smoke inhalation
rather than burns. s Although consumption of ambient
HOUSE FIRE DEATHS oxygen by the fire with environmental deoxygenation is
important, t he inhalation of carbon monoxide (and carbon
In the UK, house fires rank second after road traffic fatalities dioxide) is more frequently what causes death: thus smoke
as a cause of accidental deaths in children below the age of detectors are valued as preventative measures. Smoke
15 years. When a house fire develops, children are more vul detectors are widely used in several countri es and foun d
nerable, partly because they depend on the ad ults around to be effective. Their use is becoming more widespread in
them to assist them out of the fire as they may be too small Europe and in the USA,9,lO and they have been sho wn to
and inexperienced to be capable of helping themselves and have an important role in the prevention of conflagrations
extricating themselves out of the fire. The installation of both in public and private housing. II At present, regul a
smoke alarms in houses has been a major reason for the tions throughout the European Community tend to impose
The pathologist's role I 319
the use of smoke alarms in public buildings rather than 76 fatal fires; carelessly throwing away cigarettes, matches
homes where children live but there may be good reason and lighters caused 39 out of the 76 fatal fires. Smoke
for extending their scope. alarms were fitted in 44 out of the 76 homes where people
The banning of upholstery, certain items of furniture died, but only six alarms were known to have operated
and other fire-retardant items whose combustion produces properly; in the other 38 cases, 15 did not have a battery
cyanide and other toxic compounds has also gone some fitted and three had flat batteries. Fire investigators said
way in reducing deaths from such sources. that 39 out of the 85 house fire deaths were potentially
Fires and flames are a significant cause of death in avoidable if smoke alarms had been fitted and in working
childhood in many countries in Europe. 12 Although there order. Of these fatal fires 34 started in the living-room, 21
has been a striking fall in the number of children who have in the bedroom and 13 in the kitchen. Polyurethane foam
died from fire in recent years, the death rate from fires filled furniture contributed to 14 fatal fires.
remains unacceptably high. A major component of the Over 42000 children under the age of 15 years were
reduction has been the fall in the number of deaths from injured in the UK from burns and scalds in 2002, with
the ignition of clothing following flame-proofing regula 28000 of them being under 5 years of age and 95 per cent
tions and the reduction in use of open fires. There have of the thermal injuries occurring in the home. Over 50 per
been few specific studies of deaths from fire and flames at cent of all severe burns and scalds happened in the kitchen.
the European level. Hot liquids were the cause of 70 per cent of the heat
Each year in the USA, an estimated 700 children aged induced injuries.
5 years and under die in house fires. Representing 20 per cent
of the fire deaths each year, this age group has a fire risk
that is double the national average. Children playing with THE PATHOLOGIST'S ROLE
matches, cigarette I ighters and other fire sources are thought
to have started about 91810 fires per year from 1993 through Key questions to be addressed by the pathologist in fire
1997, which resulted in an estimated 338 deaths and 2624 fatalities are: Can the victim be positively identified? Was
injuries each year. Pre-school children are also the most the victim actually alive or already dead at the time the fire
frequent victims of fires started by children playing with started? Was the death directly related to the fire and
smokers' materials (over 30 per cent of the fires) and they caused by it? What exactly is the cause of death? Why was
constitute 20 per cent of the total number of fire death vic the victim unable to escape the fire?
tims. The United States Fire Administration (USFA) encour To respond to these questions the following information
ages parents to teach children at an early age about the must be correlated: (1) the circumstances preceding the
dangers of playing with fire in an effort to prevent child death; (2) the anatomy of the fire, i.e. evidence of the origin,
injuries, fire deaths and future fire setting behaviour. development and nature of the fire; and (3) the autopsy
Every year, more than 100000 household fires are findings.
reported in the UK and more than 600 people die in these When examining a fire it has to be kept in mind that fire
fires. One in ten of these fire victims are children. The num artefacts occur to varying extents in any body exposed to
ber of house fires in Scotland is 'markedly different' from a fire, irrespective of whether death was a result of the fire
England, Wales and Northern Ireland according to the latest or whether the person was already dead when the fire
government figures. In Scotland 46 per cent of all fires are started. These may be further aggravated at the time of
house fires, compared with 33 per cent in Northern Ireland the rescue and recovery of the body from the site of the fire
and 32 per cent in England and Wales. Broken down by fire occurs. These 'are:
brigade areas, the figures show that all Scottish brigades • The 'pugilistic attitude' of the body. Characteristically,
recorded that at least 38 per cent of all fires were house fires. the flexor muscles in both the upper and lower limbs,
Outside Scotland, the largest percentage of house fires was in which are on the whole stronger than the extensors,
Greater London. The Home Office research, made available will be damaged by the fire and contract, owing to the
to chief fire officers across the UK, shows that around 75 per heat -induced effects on the component proteins of the
cent of all fire casualties are in house fires: 466 people died muscles. This results in the body assuming a pugilistic
in fires in the home and another 14600 people were injured. (boxer) or 'sit up and beg' posture, with the four limbs
The number of firefighters injured - 655 men and women solidly flexed.
was the lowest recorded for over 20 years, with over 50 per • Loss of fluid from the tissues results in their
cent being physical injuries rather than burns. There are desiccation and brittleness, not least the skin and of
more than 100000 fires deliberately started in the UK each any other exposed underlying tissues that thus tend to
year; some of these are targeted at domestic premises for a crumble on pressure and touch.
variety of reasons. • Body weight may decrease with a loss of up to 60 per
In 2003, 111 people died in Scotland because of fires and cent, also mainly due to loss of water.
another 2300 people suffered non-fatal injuries. Misuse of • Bone fractures are due to desiccation and direct heat
alcohol was a major contributory factor in 41 out of the induced effects where they are exposed. The heat of the
320 I Heat-induced injury or death
fire may result in fractures or render them so brittle Blood taken at autopsy is analysed for carbon monoxi de
that they are fractured during recovery of the body_ content and the result expressed as a percentage saturation
Artefact ual fractures are particu lar ly common in the of the haemoglobin. Levels of up to 10 per cent carboxy
sku ll , w here sometimes they can be difficult to haemoglobin saturation may be fo und in the normal popu
disting uish from ante-mortem injury, lation in towns and cities as a res ult of atmospheric
• Body s hortening because the limbs, particularl y the pollut ion. A carboxy haemoglobin saturation level of over
extremities, may be partly burn ed away an d muscles 50 per cent is generally accepted as sufficient in itself to
are fl exed, acco unt for death. Perso ns with pre-existing natural disease
• Distortion of the facial features by bl ackening, skin of the heart or lungs may succumb to levels as lo w as 30 per
co ntraction and tightening, heat-induced damage to cent. In building fires approximately one-half of all victims
the facial features and sin gein g of the hair. have a carboxyhaemoglob in level that is sufficiently high to
• Accumulation of soot and perhaps th e emergence of account for death. Carboxyhaemoglobin levels of 20- 30 per
some blood-tinged fluid in the facia l openings. cent produce dizziness, headache, nausea and fatigu e; levels
• Irregular deep splits involving the sk in of the arms, of 30-40 per cent produce imp aired judgement, confusion
thighs and abdomen du e to ex posure to heat which and later unconsciousness, In a fire situation, levels above
lead s to heat contraction of the skin. (These may be 30 per cent may explain a failure to escape.
misinterpreted as tea rs, lacerat ions and cuts inflicted The presence of a n elevated carboxyhaemoglobin level
during life.) and/or soot in the airways beyo nd the vocal cords provid es
• Accumulation of blood beneath the skull (giving the proof that the victim was a live, but not necessarily conscious,
appearances of an extradural haematoma); this may at the time of the fire, Inh aled carbon monoxide and other
also be mistaken for an an te-mortem injury. noxious gases may have an additive effect and accou nt for
deaths with carboxyhaemoglobin levels below 50 per cent.
Distinguishing between bu rns inflicted during life a nd With the exception of cyanide estimation, laboratory tests
burns inflicted on an already dead body can be difficult, if for the presence of these noxious gases is rarely, if ever, per
not impossible, at autopsy, Furthermore, genuine ante formed on the bodies of fire victims. Cya nide is commonly
morte m nature of burn s may be obscured by the continued produced in fires and results from the burning of materials
ex posure to heat of the body after death. such as wool, silk, horse hair, poly urethane a nd polyaclylo
Fire deaths result from: nitrile. Cya nide is a powerful, rapidl y acti ve and generally
• smoke inhalation; cytotoxic poison, Normal blood cyanide levels are less than
• burns; 8 ~mol/L; non-fa tal tox ic effects begin at about 50 i_,mol/L
• heat shock; and the fatal threshold is above 100 ~mol/L.
• any combination of these, Bodies recovered From fires present severa l problems
of investigation a nd a close integration of info rmation
obtained from the exam inatio n of the scene, the exam ina
Smoke Inhalation tion of the body and the histOiY of the deceased is paliicu
lady important. Several investigators with different areas
Over 50 per cent of a ll fire fatalities occur as the res ult of of ex pertise a re typically involved . In fire-related deaths a
the inhalation of soot and gases generated in the course of fire investigator (usually an offlcer in the loca l fire brigade)
the fire. The hot air a nd steam generated may burn the fa ce is in attendance and can provide valuable information on
and enter into the upper a ir passages where they cause fur the origin, development and nature of t he fire.
ther damage. This heat-induced effect may cause reflex A full investigation of the circumstances preceding the
cardiac death (vagal inhibition). death requ ires the identification of the victim's past med
Fires generate a wide variety of noxious gases that, ical a nd social history.
when inhaled, may ca use confusion and disorientation , If the identity of a charred body is suspect, then identity
incapacitation or dea th , often in combination with other wi ll have to be proved scientifically. Wh en specific identi
factors. With the exception of carbon monox ide, these fication is not possible, circumstantia l evidence of identity
gases rarely reach lethal levels by themse lves. is usua ll y ava il a ble_
Inhalation of soot particles dam ages the airways The majority of fire-rel ated deaths are accidental and
because they are superh ea ted and contain toxic age nts. In there is typically abundant coll ate ra l evidence from police
building fires, evid ence of soot inhalation at autopsy is and fire brigade investigations to exclude suicide or homi
very common (90 per cent or more of cases). Autopsy evi cid e. The yo ung and the elderly are the common victims of
dence of soot in the airw ays below the level of the vocal accid en ta l fires. The deaths are usua lly the result of care
cords is proof that the victim was alive at the time of the lessness: allowing clothing to bru sh aga inst fires; playing
fire, Soot in the oesophagus and admi xed with stomach with matches or other ligh ted obj ects su ch as cigarettes;
contents impli es the swallo wing of soot and has a similar mainta ining faulty electrical and hea tin g appliances, as
significance. well as being unable to effecti vely co mba t or escape a fire.
320 I Heat-Induced injury or death
ftre may result in fractures or render them so brittle Blood taken at autopsy is analysed for carbon monoxide
that they are fractured during recovely of the body. content and the result expressed as a percentage saturation
Artefactual fractures a re palticularly common in the of the haemoglobin. Levels of up to 10 per cent carboxy
skull, where sometimes they can be difficult to haemoglobin saturation may be found in the normal popu
disti nguish from ante-mortem injury. lation in towns and cities as a result of atmospheric
• Body shortening because the limbs, palticularly the pollution. A carboxyhaemoglobin saturation level of over
extremities, may be paltly burned away and muscles 50 per cent is generally accepted as sufficient in itself to
are flexed. account for death. Persons with pre-existing natural disease
• Distortion of the facial features by blackening, skin of the heart or lungs may succumb to levels as low as 30 per
contraction and tightening, heat-induced damage to cent. In building fires approximately one-half of all victims
the facial features and singeing of the hair. have a carboxyhaemoglobin level that is sufficiently high to
• Accumu lation of soot and perhaps the emergence of account for death. Carboxyhaemoglobin levels of 20-30 per
some blood-tinged fluid in the facial openings. cent produce dizziness, headache, nausea and fatigue; levels
• Irregular deep splits involving the skin of the arms, of 30-40 per cent produce impaired judgement, confusion
thighs and abdomen due to exposure to heat which and later unconsciousness. In a fire situation, levels above
leads to heat contraction of the skin. (These may be 30 per cent may exp lain a failure to escape.
misinterpreted as tea rs, lacerations and cuts inflicted The presence of an elevated carboxyhaemoglobin level
during life.) and/or soot in the airways beyond the vocal cords provides
• Accumulation of blood beneath the sku ll (giving the proof that the victim was alive, but not necessalily conscious,
appearances of an extradural ha emato ma) ; this may at the time of the fire. Inhaled ca rbon monoxide and other
also be mistaken for an ante-mortem injury. noxious gases may have an add itive effect and account for
deaths with carboxyhaemoglobin levels below 50 per cent.
Distinguishing between burn s inflicted during life and With the exception of cyanide estimation, laboratory tests
burns inflicted on an already dead body can be difficult, if for the presence of these noxious gases is rarely, if ever, per
not impossible, at autopsy. Furthermore, genuine ante formed on the bodies of fire victims. Cyanide is commonly
mortem nature of burns may be obscured by the continued produced in fires and results from the burning of mate rials
exposure to heat of the body after death . such as wool, silk, horse hair, polyurethane and polyacryIo
Fire deaths result from: nitlile. Cyanide is a powerful, rapidly active and genera lly
• smoke inhalation; cytotoxic poison. Normal blood cyan ide levels are less than
• burns; Bjtmol/L; non-fatal toxic effects begin at about 50f-Lmol/L
• heat shock; and the fatal threshold is above 100 f-Lmol/L.
• any combination of these. Bodies recovered from fires present several problems
of invest igation and a close integration of informatio n
obtained from the examination of the scene, the examina
Smoke Inhalation tion of the body and the history of the deceased is particu
larly important. Several investigators with different areas
Over 50 per cent of all fire fatalities occur as the result of of expertise are typicaJly involved. In fire-related deaths a
the inhal ation of soot and gases gene rated in the course of ftre investigator (usually an officer in the local fire brigade)
the ftre. The hot air and steam generated may burn the face is in attend ance and can provide valuable information on
and enter into the upper air passages where they cause fur the origin, development and nature of the fire.
ther damage. This heat-induced effect may cause reflex A fu ll investigation of the circumstances preceding the
cardiac death (vagal inhibition). death requires the identiftcation of the victim 's past med
Fires generate a wide variety of noxious gases that, ical and social history.
when inhaled, may cause confusion and disorientation, If the identity of a charred body is suspect, then identity
incapaci tation or death, often in combination with other will have to be proved scie ntiftcally. When specific identi
factors. With the exception of carbon monoxide, these fication is not possible, circumstantial evidence of identity
gases rarely reach lethal leve ls by themselves . is usually available.
Inhal ation of soot particles damages the airways The majority of fire-related deaths are acc idental and
because they are superheated and contain toxic agents. In there is typically abundant collateral evidence from police
building fires, evidence of soot inhalation at autopsy is and fire brigade investigations to exclude suicide or homi
very common (90 per cent or more of cases). Autopsy evi cide. The young and the elderly are the commo n victims of
dence of soot in the airways below the leve l of the vocal accid ental fires. The deaths are usually the result of care
cords is proof that the vict im was alive at the time of the lessness: allowing clothing to brush against fires; playing
fire . Soot in the oesophagus an d admixed with stomach with matches or other lighted objects such as cigarettes;
contents implies the swallowing of soot and has a similar maintaining faulty electrical and heating appliances, as
s ig n ifi cance. we ll as being unable to effectively combat or escape a fire.
- --- ~
322 I Heat-Induced injury or death
Evidence of soot inhalation is very common (approxi wood or polyurethane is rapidly fatal at 3000 ppm. Nitro
mately 90 per cent of cases) in building fires but it is gen dioxide from acrylonitrile is rapidly fatal at 2000 ppm.
uncommon to find any soot deposition beyond the primary Hydrogen chloride from PVC is rapidly fatal at 2000 ppm.
bronchioles. Deposition of soot on the tongue, in the nares, Hydrogen sulphide from rubber or wool has toxicity that
the oropharynx or nasopharynx cannot be taken to imply is similar to hydrogen cyanide and is rapidly fatal at
life during the fire. Deposition of soot below the level of the 1000 ppm. Acrolein is an aldehyde product of combustion
larynx indicates that the victim was al ive at the start of the of wood and paper and produces pulmonary oedema after
fire. This soot that coats the mucosa of the tracheo a few seconds exposure at 10 ppm. 14
bronchial tree can be identified through a window cut in Carbon monoxide is a colourless, odourless gas that is
the trachea prior to removal of the neck and thoracic struc formed by the incomplete combustion of carbon com
tures and is most easily seen if the mucus is spread on a pounds and is normally found in the atmosphere in levels
white sheet of paper. The presence of soot in the oesopha well below 1 ppm. Motor vehicle engines are responsible
gus and stomach implies that it has been swallowed and for about 70 per cent of carbon monoxide liberated into the
also indicates life at the start of the fire. On microscopy, the atmosphere.
soot particles within the tracheobronchial tree lie loosely Carbon monoxide poisoning is a distinct cause of inhala
on the mucosal surface or embedded in the mucus. The par tion injury, producing its effects by tissue hypoxia. The
ticles are not incorporated into the tissues and are readily affinity of haemoglobin for carbon monoxide is 200-300
distinguished from anthracitic pigment. times greater than for oxygen so that carboxyhaemoglobin
Smoke poisoning describes the effects of the various concentration is great even when the carbon monoxide con
noxious gases other than carbon monoxide that are pro centration is less than 5 per cent in the inhaled gas. As well
duced by the thermal degradation of both natural and as diminishing the oxygen-carrying capacity of the blood,
man-made materials (Table 15.1). There are hundreds of carboxyhaemoglobin (COHb) also alters the dissociation
such products of combustion. Whereas modern synthetic characteristics of the remaining oxyhaemoglobin, making
materials, for example PVC, may have increased this prob less oxygen available to the tissues. The toxicity of carbon
lem, natural materials, for example wood, wool, and silk, monoxide depends upon: (1) the rate of inhalation of the gas
produce similar gases and the pathological effects of these (i.e. concentration of gas in the inspired air and the duration
noxious gases are difficult to separate from direct particu of exposure); (2) physical activity, which influences oxygen
late injury. With the exception of carbon monoxide and requirements; and (3) individual variations in susceptibility.
hydrogen cyanide, post-mortem analyses for toxic gases Carboxyhaemoglobin is estimated by spectroscopic
are rarely, if ever, performed. Filter masks do not protect methods making use of the fact that oxyhaemoglobin and
against the inhalation of fumes; for this purpose the mask carboxyhaemoglobin have different absorption spectra.
must have a self-contained air supply. The result is expressed as the percentage saturation and
These noxious gases rarely reach lethal levels by them is the ratio of carboxyhaemoglobin to total haemoglobin
selves, but they may cause incapacitation or death in com X 100. The carbon monoxide oximeter is an automated
bination with other factors. Some gases have a synergistic instrument that uses a spectrophotometric method. An
interaction, for exam pie carbon monoxide and hydrogen alternative gas chromatographic method is both specific
cyanide, so that non-fatal levels of each in combination and very sensitive. Carbon monoxide liberated from a
may cause death. The concentration of the noxious gas is known volume of blood is estimated on a gas partitioner
of importance. Hydrogen cyanide produced from wool, and the haemoglobin content estimated spectroscopically.
Gas Source
Table 15.2 The effects on humons of different concentrotions of occur in post-mortem samples of blood and tissue. Artefac
carboxyhoemog/obin saturation tual formation of cyanide in post-mortem blood is more
likely in refrigerated samples than at room temperature. 16
Percentage Effect
will require skin grafting. Heat coagu lation of dermal blood in sta nces co ntact with the hot liquid is therefore momen
vessels leaves the tissue avascular with a characteristic waxy tary and the large surface area of the skin on to which the
white colour. Prolon ged contact of subcutaneous fat with a hot fluid has made contact a llows for rapid cooling, and
flame source produces a leathelY brown, or black, chan-ed thus unless t he temp erature of the water is quite high, there
appearance. There is charac te ristic lack of pain, due to heat would not be a sufficiently lengthy period of exposure for
destmction of all nerve endings. injury to result from such co ntact. These pheno mena are
Burns are nowadays more frequently classified as either quite impoltant in attempting to reconstmct how a child
partial thick ness or full thic kness . Partial-thickness wounds has come by his scalds, palticularly in relation to accepting
contain viable epithelial elements capable of sponta or otherwise the his tory given by the carers when non
neously re-epithelializing the wo und (first-degree, superfi accidenta l injury (NAI) is suspected.
cia l and deep second-degree burns). Full-thickness burns In this respect the temperature of the hot water system
have no viable epithelial elements and always require cuta in households has to be regulated with care to ensure th at it
neous autografting. Partial thickness burns are usually does not exceed celtain levels l9 and legislation and guide
caused by a brief exposure to heat or contac t with hot liq line in terms of plumbing of buildings are in existence.2o
uids and appear pink to mottled red, wet, covered with Other instances of hea t-induced injury involve fire
ves icl es and bullae, and are painful. Full-thickness burns works, palticularly around the time of bonfire night in
result from contact with flame, electricity or chemicals. November in the UK a nd New Year's Day festivi ties in other
They are often dry and charred, may be translucent, a nd countries 2 1,22 Some deaths from overheating have a lso
may have thrombosed superficial veins. They are insen sate. been described in children, particularly infan ts, from the
Burns wi th chemicals can arise both from direct contact use of electric blankets.
and from fumes (e.g. chlorine), either acting directly on the
skin or mucosal surfa ces, or by inhalation. The degree of
tissue damage caused by them is dependent on the type of Pathological Changes
chemical (alkalis are more caustic than acids in general
terms), its concentration, its quantity, the duration of contact Beneath the dead burned tiss ue, there is usually a zon e of
a nd the extent of penetration; if chemicals are lipid soluble ischaemia of marginally viable tissue th at is readily con
or if proteolysis occurs then this results in a greater depth of verted to non-viab le tissue (eschar) by any furth er insult,
penetration and at a quicker rate due to the liquefactive such as hypoxia, decreased blood flow or infection. By this
necrosis that they induce, th us exposing deeper tissues to the process, a deep seco nd-degree burn frequ ently converts
effects of the chemical. Chemicals that induce burns can be over time to a third-degree burn. Prevention of wound
categorized as strong acids, strong alka lis and hydrocarbo ns, conversion of this type is of major importance in the res us
wi th hydrocarbons often having penetrating capacity. c itation period and thus attention to oxygenation and
Heat damage from hot fluids is usually referred to as a decreasing the risk of infection are essential elements of
scald. Water is the compon ent fluid involved, usually in the the early management of burns. 23 ,24
kitchen or bathroom, but other fluid s such as oil, molten A determina tio n of the burn surface area in the living is
metal , molten rub ber, other hot liquids and semisolid or obtained using the 'mle of nines ', each arm constituting
liquid foods (e.g. porridge, molten cheese) may be 9 per cent, each leg 18 per cent, the anterior trunk 18 per
involved; steam as a by-product of the heating process is cent, the posterior trun k 18 per cent and the head 9 per
also very damaging due to its penetrative cap abilities. cent of TBS area. In assessing irregular areas of burn, it is
Scald s do not usually result in charring, carbonization or useful to remember that the surface area of th e person's
singeing of body hairs. Their effects are more like that of palm is roughly equivalent to 1 per cent of the TBS. In chil
first-degree burns from dry heat, with superficial redden dren under 15 years of age the rel ative body surface area of
ing, desquamation and loss of the epidermal layer, swelling the head, upper leg and lower leg differs from an adult; this
and blistering, usually with well-demarcated areas of dam ma y lead to inaccuracy in the estimation of burn injury;
age, with unifo rmi ty all over the site of a rea of contact use of the Lund Et Browder Burn Diag ram provides age
with the fluid. Thus the severity of skin damage is more adjusted data accurate for both adults and children.
uniform in sca lding and from the distribution of the scalds It should be remembered that the extent of burn is often
it may be quite possib le to identify th e position of the over-estimated, and the depth of burn under-estimated
scalded victim in rel ation to the hot fluid; there may be wh en the pati ent is assessed clinically.
trickle or splash scalds, and horizontal fluid levels from In addltion to burn depth and extent of injury, other
immersion in hot water. factors determining morbidity a nd morta lity a re the loca
The scald 's severity depends on the duration of exposure tion of the injury, the age of the victim and the presence of
of the skin to the hot fluid and its tem perature. Clothing injuries or natural disease. Age is a major factor in survival
may prevent direct content of the fluid underlying the skin for children under 2 years and adults over 60 years.
but once it is soaked it may increase the duration of con The dermis acts as a barri er prev,enting loss of body
tact of the hot fluid with the skin's surface. In many fluid by evaporation an d the loss of excess body heat.
The pathologist's role I 325
Consequently, loss of water through burn ed skin and loss viable wound bed is obtained) and excision to fascia . The
of heat play a major role in the pathophysiological disadvantage of tangential excision is primarily that of
changes. The skin is also the primary protective barrier mass ive blood loss.
against invasive infection so that wound infection is a Burns also induce a system ic response to them due to
major cause of mortality and morbidity in the late post the presence of cytokines and other inflammatory media
burn period. Within the superficial dermis are the nerve tors if the area of the burns exceeds the 30 per cent surface
endings that mediate pain. Consequently, partial-thickness area. These substances cause systemic effects:
injuries tha t expose these nerves will be extremely painful, 1. increased capillary perm eability that leads to loss of
whereas full-thickness burns that destroy the nerves are fluid and from proteins within the blood vessels into
usually anaesthetic. the interstitial tissues.
The loca l and system ic responses to burns fo llow a time 2. peripheral and sp lanchni c vasoco nstriction.
course continuum but, for convenience, can be divided in to 3. decreased muscular contractility (probably due to the
three periods: (1) up to 48 hours; (2) 2-6 days; and (3) 7 days effect of tumour necrosis factor alpha).
to wound closure. 4. systemic h ypotension due to primary and seco ndary
In the first 48 hours after severe burns, hypovolaemic fluid losses.
shock an d shock-induced organ failure (primarily renal 5. fall in systemic blood pressure - hypotension.
failur e) are the major threats to life. Hypovolaemia can also 6. hypoperfusion of vital organs.
lead to wound convers ion. Al though the exact pathophys 7. bronchoconstriction.
iolo gy of the post-burn vasc ular changes and volume shifts 8. adult respiratory distress syndrome.
remains to be determined, two processes are involved: an 9. the basal metabolic rate trebl es.
increase in microvascular fluid flux into the interstitium 10. No n -specific downregulation of the immun e system
both local to the burn and genera li zed (remote from the affect ing both antibody- and lymphocyte-mediated
burn wound) - and a generalized impairment in cell mem pathways.
brane function, resu lting in cell swelling.
With modern therapy, adequate initial volume restora
NON-ACCIDENTAL BURNS AND SCALDS
tion is achieved in more than 95 per cent of burns cases.
Correction of hypovolaem ia by the intravenous infusion of Although a significant number of heat-i nduced injuries are
fluid s is comp licated by ge neralized burn oedema forma truly accidental - although often preventable - they result
tion, wh ich may resu lt in (1) further ischaemic insu lt to from temporary lapses in the due care and attention that chil
already damaged cells; (2) chest wall oedema with resultant dren require in the home an d elsewhere. Some, however, are
increase in the work of breathing; and (3) upper airway due to neglect on the part of the carers, who show inadequate
oedema. The last two complications may lead to rapidly and negligent parenting with failure to protect the child.
fatal respiratory failure. Others heat-induced injuries are deliberately inflicted. 2s - 27
Burned skin loses its elasticity. It is less compliant and Sometimes it is eventually suggested by the perpetrato r that
unable to stretch to accommodate an increase in interstitial these are punitive in nature, allegedly punishing the child for
oedema. If burns are circumferen tia l, particularly a round 'bad' behaviour and failure to comply with parental instruc
the distal extremities, a to urniquet effect is produced by tions, for example in relation to 'potty training' or playing
t hem. Initially this impedes venous return , resulti ng in an with matches. They may also be sadistic, although less fre
increase in capillary pressure and further oedema. Impair quently, in their production, with the perpetrator obtaining
ment of arterial blood flow may follow, with the potential pleasure by the infliction of pain and fear on the child. In
for distal tissue ischaemic necrosis. Tissue pressure can be about 10 per cent of physically abused children bums are a
decreased by making an incision through the burned tiss ue, component of the injuries inflicted; about 5 per cent of sexu
i.e. an escharotomy. A chest wall escharotomy may be ally abused chi ldren also show bums. 28 31
required to ease respiratory difficulties; incisions are made Injury induced by liquids is usually referred to as sca lds.
in the bilateral a nterior axillary lin es from the clavicles to These often blister, with peeling of the epidermis when the
the costal margins; a transverse escharotomy at the level of blisters burst. The damaged area often appears soggy and
the costal margins connect ing the two vertical escharo blanched. Their distribution may follow the contour of the
tomies is helpful if the abdomi nal wall is burned to a sig clothes given that the soaking of clothing with hot liquid will
nificant degree. Escharotomy can be a bedside procedure increase the duration of the contact between the liquid and
and does not require anaesthesia because full-thickness the skin and may thus follow the contours of clothing. The
burns are insensa te. If th e deeper viable tissues are not liquid will produce splash patterns - pouring and drip pat
incised then bleeding should be minim al. terns. Domestic hot water, boiling kettles and saucepans, pip
Third-degree or full-thickness burns often require exci in g hot cups of tea and coffee, and baths are frequent causes.
sion prior to cutaneo us autografting. The two commonly Scalds may also be found internally on the lips, gums,
employed techniques of excisi on are tangential excisio n cheeks , tongue an d pharynx as a ctlllsequence of swallow
(sequential shaving of no n-viable tissue until a uniformly ing hot liquids. 32 Large areas of mucosa will peel off, and if
326 I Heat-induced injury or death
oedema supervenes as a reactive change there may be an even involve the loss of limb. There may also be muscle
obstruction to upper air passages and to air exchange. da mage with rhabdomyolysis, which, in turn, may lead to
Contact with metallic, hot, dry surfaces will produce dry acute renal failure. Contact with voltages of over 70000
or contact burns, such as with contact with central heating volts is invariably fatal. In flash electric burns there is
radiators, an oven, electric heate r, clothes' irons, curling exposure to radiant heat of the surface of the body, often
tongs, fire surrounds, light bulbs, heating grills; serious the face or on the arms, but clothing may also be set on fire
burns can result even though the contact with such objects with further burns ensuing.
is usuall y of a very short duration. These burns may The depth of a burn is related to:
demonstrate a branding type of appearance, with the burn • the temperature of the heat source.
appearing dry and well demarcated, reproducing the pattern • the duration of contact between the skin or mucosa
of the object that has caused t he burn. and the warm object.
Lit cigarettes may leave very characteristic patterned
circular or oval marks on the skin and although these may The usual temperature of hot water in the home is about
be accidentally induced , particularly in a crowded house 60°C and contact with water at this temperature results in
hold, they are also often deliberately inflicted on the child. a full-thickness burn in child at about 10 seconds of con
In the latter instances areas of the body that are less easily tact. Higher temperatures will cause burns much quicker
visible to a medical examiner or observer will be used by and at 70°C only 1 second of contact is sufficient for a full
the perpetrator: thus such Cigarette burns may be found on thickness burn.
the lower back, the nape of t he neck, the buttocks and the About 3-10 per cent of burns in British children are due
legs. They are often multiple, and indeed may also be of a to NAI.34 In American studies, 9-25 per cent of children
different duration, and if they present in varying healing who are admitted to burns uni ts have been assaulted.
phases then they are indicative of more than one episode of Younger children in the age range 5 months to 6 years are
such trauma. A pattern that is seen occasionally is that the often affected, with a peak noted in older toddlers around
outline of the main bruise is oval in its configuration with the age of] years; about one-third of these children will be
a roughly triangular ta il of further, usu ally more superfi repeatedly abused. This is in contrast with other forms of
ci al, burning radially confluent with the main burn but abuse when younger children usually feature more promi
located at an angle to it. This is due to move ment by the nently.
cigarette across the skin or by the child attempting to move In assessing whether a heat-induced injury is accidental
away from the point of impact with the lit Cigarette end. or otherwise the presentation of the child and the history
This appearance can thus be found in both accidental and given by the carer at medical presentation is impoJta nt, as
delib erately inflicted cigarette burns. is the interaction of the carers with medical and nursing
Accidental burns are found at sites where the child has staff. If the injury had been inflicted non-accidentally then
handled the lit cigarette or fallen on/brushed against it; the appearance of the injury and the history given by the
thus the back of the hands, the feet and the back are not the carer may not tally: the carer may even suggest that the
usu al sites of an accidental contact. injury was not induced by heat or the explanation given
Radiant burns may occur after prolonged contact with for the contact with the warm solid or liquid is incongru
the sun or a rather hot tIre. These affect a large anatomical ous, for example claiming t hat a two-year-old has opened
area of the body and are often ch aracterized by marked a tight, large, hot water tap in a bath. Siblings may be
erythema. All of the surface that is exposed will show heat blamed and the episode is claimed to be unwitnessed; when
indu ced injury. Clothing may have a protective effect and the child is asked, there is a different version of events
thus patterns may be observed in such burns. given. Associated concomitant injuries and previous heat
Chemical agents, not least acids and alkalis, cleaning induced damage may fe ature in the past medical history in
agents used in the house and also electricity sources will such instances.
also induce burns. The amount of heat generated and thus If t he child has been deliberately immersed into hot
the a mount of tissue da mage is equal to 0.24 x (voltage)2 x water then a clearly delineated tide mark may be seen if the
resistance. 33 Thus voltage is the main determinant of tissue lo we r limbs are involved . Similarly, if the buttocks are
injury, with electric burns thus being divided into low volt immersed then a 'doughnu t' demarcation will be observed
age burns (as with t he domestic current) and high-voltage on the buttocks. The deliberate infliction of scalds on a
burns. Low-voltage burns are often deep-contact burns at child may be the result of chastisement for failure by the
the entry and exit points of the current from the body. child to comply with potty training and in bed wetting. In
When the latter involve voltages of over 1000 volts they many instances, there will also be splash marks elsewhere.
are subdivided further into true 'h igh-tension' injuries Children thus non-accidentally assaulted should be
caused by a high voltage current going through the body exa mined for other signs of abuse and neglect. The differ
and flash injuries caused by tangential exposure to high entiation between accidental and non- acc idental burns is
voltage current with no current actually flowing t hrough not easy and, unless suspected in all >circumstances, non
the body. These may show extensive tissue losses and often accidental injUly will be missed?5
References I 327
BURNS CAUSED BY ELECTRICITY Electricity can cause disruption of the body's normal
electrical activities, with neurological 39 dysfunction being
An electric current is a flow of electrons along the path of
present in some form, even if only temporalY, in virtually
least resistance toward a natural ground with any sub
all patients. 4o Transient nerve injuries resulting in tempo
stance or object along this path decreasing the rate of flow
rary numbness and tingling are most common. Mass depo
(resistors) or increasing the rate of flow (conductors). The
larization of the brain may lead to a loss of consciousness
amnesia and coma. Spinal cord involvement may result i~
skin is a natural resistor to electrical flow: dry skin has a
resistance of 40 000-100 000 ohms, wet skin approximately
transverse myelitis. Transverse myelitis may have delayed
1000 ohms and calloused skin 2000000 ohms. The thin
onset and is associated with poor prognosis for recovery.
skin and high water content of the skin of children has a
Electrical injuries also may affect the heart 41 and about
resistance much lower than that of an adult. J6 The internal
25 per cent of patients with electrical injuries have cardiac
resistance of the body is estimated to be between 500 and
dysrhythmia. Many of these are benign and transient (e.g.
1000 ohms, with bones, tendons, and fat providing most
sinus tachycardia , premature atrial ventricular contractions
resistance. Nerves, blood vessels, mucous membranes and
and conduction disorders). Sudden death from an AC elec
muscle are the best conductors.
trical injury often results in ventricular fibrillation, although
With burns resulting from an electric current, the cross
asystole and other dysrhythmias are common. Ventricular
sectional area is inversely proportional to tissue damage.
fibrillation is three times more likely to occur if the flow of
The pathway of the current plays an important role in
current is arm-to-arm. Tnle myocardial infarction is rare
determining injury, with a vertical pathway being more
unless the patient has pre-existing cardiac disease.
dangerous than a horizontal (hand-to-hand) pathway.J7
Another mechanism of injury is related directly to the
Standard household current in the UK is 230-240 volts
amount of heat generated by the flow of electrical current
(alternating current lAC]) at a frequency of 50 hertz; in the
through body tissue. At higher voltages, higher tempera
USA and Canada it is 110 volts, with a frequency of 60 hertz.
hIres are achieved thus resulting in greater direct thermal
Skeletal muscle is stimulated into spasm and tetany by cur
injury. High-tension voltages cause devastating injuries
rents with frequencies of 40- 110 hertz. Most low- and high
from huge amounts of internal thermal damage.
tension electrical current is AC. Alternating CUITent produces
Vascular injury occurs as a result of vascular spasm, and
tetany and the 'locked-on' phenomenon as the fle xo r muscle
the heat generated can also cause coagulation and vascular
groups are usually stronger and predominate. As a result, an
occlusion. Damage to the vascular wall may produce delayed
individual's grasp is uncontrollably locked on to an object in
thrombosis and bleeding. Compartment syndrome may
which an electric current is passing, which can increase the
develop as a result of acute ischaemic insult to the muscula
length of time the current passes through the body and may
ture. Renal injuries may occur as a result of rhabdomyolysis.
result in greater injury. In contrast, direct current (DC) tends
Rhabdomyolysis causes myoglobinuria from massive release
to produce a single large muscular contraction t hat often
of myoglobin, which, on crystallization in the kidney tubules,
throws the patient away from the source. However, at high
may cause acute renal failure.
voltages, both AC and DC produce similar effects. Contact
with high-voltage CUITents results in injuries that are associ
ated with arc burns and flash burns. Arc temperahlres may
REFERENCES
reach up to 5000· C and are usually responsible for the severe
thermal injuries. Runya n SA, Casteel C (eds) The State of Home Safety in
Lightning involves a single massive current impulse that Americr: Facts about Unintentional Injuries in the Home,
is roughly equivalent to a DC blast of 2000-2 billion volts 2nd edn. Washington DC: Home Safety Council, 2004.
of extremely short duration (0.1-1 ms).J8 The peak tempera 2 Centers for Disease Co ntrol and Prevention. Web-based Injury
Statlstlcs Query an d Reporting System (WISQARS) [online).
ture in the lightning strike channel is 30000·C. However,
2005. National Center for Injury Prevention and Control,
the short duration usually precludes serious direct thermal Centers for Disease Control and Prevention.
injllly. Four modes of lightning injury have been described: www.cdc.gov/n cip c/w isqars
direct strike, side flash (discharge from an object near the 3 Internation al Association for the Study of Insurance
victim)' stride potential (enters one foot and exits via the Economics. World Fire Statistics: Information Bulletin of the
other), and flashover (energy passing outside the body with World Fire Statistics. Geneva: The Geneva Association, 2003.
4 Ma lon ee S, Istee G, Rosenberg M et al. Surveillance and
vaporization of surface water and blast effect to clothing). prevention of reside ntial - fire injuries. N Engl J iVIed 1996;
Lightning may cause full cardiac arrest by inducing either 335:27-32.
asystole or central apnoea. Massive depolarization of the 5 Sorensen B. Prevention of burns a nd scalds in a developed
heart leads to asystole. However, the heart usually restarts counBy. J Trauma 1976; 16:249-58.
6 Linares AZ , Linares HA. Burn prevention programmes for
spontaneously in normal sinus rhythm. Massive depolariza
children: are they effective? Burns 1979; 6:26-9.
tion of the brain is believed to stun the respiratory centre, 7 MacK ay A, Rothman K. The incidence and severity of burn
causing a much longer duration of central apnoea. If artifi injuries following Project Burn Prevention. Am J Publ Health
cial respiration is provided, many patients can survive. 1982; 72:248-5 2. >
328 I Heat-induced injury or death
8 Istre GR, McCoy MA, Osborn L et al. Deaths and injuri es from 24 Wo lf SE, Rose JK, Desai MH et a!. Mortality determinants in
hou se fires. N Engl } Nled 2001; 344: 1911-16. massive pediatric burns. An analysis of 103 chil.dren with
9 McLoughlin E, Marchione M, Han ger L et al. Smoke detecto r ;'80% TBSA burns [;'700/0 full thickness). Ann Surg 1997;
legislation: its effects on owner-occupier homes. Am } Publ 225: 554-65, discussion 565-9.
Health 1985; 75:858-62. 25 Ayoub C, Pfeifer D. Burns as a manifestation of ch ild abuse
to Ahrens M. US Experience with Smoke Alarms and Other Fire and neglec t. Alii J Dis Child 1979; 133:910- 14.
Alanns. Atlanta, GA: Quincy [MA) National fire Protection 26 Sto ne NH , Rinldop L, Humphrey CR et a1. Chi ld abuse by
Associatio n, 2004. burnin g. Surg Clin North Am 1970; 50:141 9- 24.
11 Mill er R, Reisinger K, Blatter M, Wu cher F. Pediatric 27 Showers J, Ga rrison KM. Bu rn abuse; a four -yea r study.
co unselling and subseq uent use of smoke desecraters. J Traumn 1988; 28: 1581 - 3.
Ani J Publ Health 1982; 72 :392-3. 28 Hight DW. Bakalar HR , Lloyd J. Inflicted burns in children:
12 Carlson A, Uden G, Karlsson ED. Burns injuries in sma ll recognition and treatment. JAMA 1979; 242:517 - 20.
children, a population-based study ill Sweden . } Clin Nursing 29 Keen JH, Lend rum J, Wolman B. Inflicted burns and scalds in
2006; 15:129-34. children. BM} 1975; 4:268-9.
13 Anderson RA, Watso n AA, Harland WA. fire deaths in the 30 Hobbs CJ. When are bum s not accid enta l? Arch Dis Child
Glasgow a rea: general cons iderat ions and patho logy. 1986; 61:357-61.
Med Sci Law 1981; 21: 175-83. 31 Lenoski IF, Hunter KA. Speciflc patte rn s of inflicted burns
14 Napier DH. Haza rdou s materials and· the gases they produce. injuri es . J Trauma 1977 ; 17:842.
Med Sci Law 1977; 17 :83-90. 32 Canady JW, Thomps on SA, Bardach J. Oral commissure burns
15 Kojima T, Nishiyama Y, Yashiki M, Une I. Post mortem in chi ldren. Plost Recon struct Surg 1996; 9 7:738-44,
formation of carbon monoxide. Forensic Sci Int 198 2; discu ss ion 745, 74 6-55.
19: 243-4. 33 Luce E. Electrical burns. Clin Plast Surg 2000; 27:133 - 43.
16 Anderson RA, Ha rl a nd WA. Fire deat hs in the Glasgow area: 34 Andronicus M, Oates RK, Peat J et a1. No n-accid enta l burns in
the role of hydrogen cyanide. Med Sci Law 1982 ; 22:35 -40. children. Bums 1998; 24:552 - 8.
17 Hettiaratchy S, Dz iew ulski P. ABC of burns: pathophysiology 35 Benger JR, McCab e SE. Burns and scalds in pre-school
a nd types of burns. BM} 200 4; 828:1427 -9. children atte ndin g accident and emergency: accid ent or
18 Knight B. Burns and sca lds. In Kn ight B [ed.) Forensic abuse? Emerg Med J 200 1; 18:172-4.
Pathology, 2nd ed n. London: Arno ld, 1996, pp. 305- J 7. 36 Ga rcia CT, Smit h GA, Cohen DM, Ferna ndez K. Electrica l
19 Weaver AlVI, Himel MHM, Edli ch RF. Immersion scald burns: injuri es in a pediatric emergency depa ltment. A nn Emug IVIed
strate gies for preventio n. J Emerg lVled 1993; 11 :39 7-402. 1995; 26:604-8.
20 Erdmann T, Felman K, Rivara F et a!. Tap water burn 37 Nguyen BH, Mac Kay M, Bailey B, Klassen TP. Epidem iology
prevention: the effect of legi slati on. Pediatrics 1991: of electrical- and lightning-related deaths and injuri es amo ng
88:572-7 . Canad ian ch ildren and you th. Injurv Pre" 2004; 10:122 - 4.
21 Sm ith GA , Knapp Jf, Barnett TM, Shield s BJ. The rockets· red 38 Jain S, Bandi V. Electrica l a nd li g htning injuri es. Crit Care
glare, the bomb s bursting in the air : fireworks- related injuries Clin 1999; 15:319 - 31.
to children. Pediatrics 1996; 98: 1-9. 39 Lee RC. Injury by electrical forces: pathophysiology,
22 D'Regenio P, Cafaro L, Santon3sras i F, Taggi F et al. manifestations, and thera py. CUlT ?robl Slirg 1997 ;
Capodanno Senza Danno: the e ffects of an intervention 34:677-764.
program o n fireworks injuri es in Nap les. Am J Pub' Healtlr 40 Rai J, Jeschke MG, Barro w RE , Herndon DN. Electrica l
1996; 86:84-6. injuries: a 30-year review. J Trauma 1999 ; 46:933 - 6.
23 Walker AR. Emergency department management of house fire 41 Bailey B, Gaudreau! t p, Thivierge RL, Turgeon JP. Cardiac
burns and carbon monoxid e pOiso ning in chi ldren. Cur Opin mon itoring of children with hou seho ld electrical injuries.
Pediatr 1996; 8:239 -42. Ann Emerg Med 1995; 25:612-17.
I CHAPTER 16 I
ASPHYXIAL DEATHS IN CHILDREN
Anthony Busuttil
The term 'asphyxia' comes from the Greek sphygmos considered to be pathognomonic and an almost sine qua
(meaning absence of a pulse) and is applied to pathological non of asphyxia.
unnatural states in which the body is deprived of oxygen These are haemorrhages arsing from tiny blood vessels,
while there is a concomitant excess of carbon dioxide, i.e. probably mostly venules (capillary haemorrhages may be
a hypoxaemia (hypoxia) and hypercarbia (hypercapnoea) too small to be visible naked eye being 1-2 mm lesions or
together. This may result in loss of consciousness with the smaller), which rupture when the pressure within them is
coma thus produced leading to death. Children die from increased and the endothelial lining is distended beyond its
mechanical asphyxia in many situations and exhibit simi innate elasticity. Contribution by hypoxaemia is likely in
lar pathological changes that one would expect in an adult that the vessel wall is also affected by a decrease in oxygen
dying in a similar situation. tension, becoming leaky and pennitting the extravasation of
In the international classification of diseases, accidents red blood cells. These lesions tend to occur at sites where the
caused by asphyxia are classified under E91O-E913: connective suppot1ing tissue is loose, provided that the pres
• E910 = Accidental drowning and submersion. sure is locally raised in the specific area. 2 Thus an applica
• E911 = Inhalation and ingestion of food causing tion of a sphygmomanometer cuff at a level sufficient to
obstruction of respiratory tract or suffocation. occlude, selectively, the venous supply for a period will
• E912 = Inhalation and ingestion of other object result in petechiae formation distal to the edge of the cuff.
causing obstruction of respiratory tract or suffocation. Another cause of petechial haemorrhages is a blood
• E913 = Accidental mechanical suffocation. clotting problem. A low platelet count from any cause, for
example idiopathic thrombocytopenic purpura, aplastic
anaemia, or leukaemia, may produce petechiae. Excessive
PETECHIAE capillary fragility such as Henoch-Schonlein purpura may
have a similar effect (see Chapter 4).
This pathological feature has long been recognized in the Petechiae may also be found when there is toxic damage
forensic context. As a finding of medicolegal import, to the endothelium, particularly in association with dissemi
petechiae were first described by Ambrose Tardieu in nated intravascular coagulopathy, as in coliform and
1855. 1 It was, at one stage (and perhaps by many still is), meningococcal septicaemia. In septicaemia, petechiae tend to
330 I Asphyxial deaths in children
be widespread. They are also found in situations when environmental catastrophes such as earthquakes and gas
microembolic phenomena occur, such as fat and air explosions, will show features above the level of obstruc
embolism, amniotic fluid embolism and sub-acute bacterial tion associated with t his condition.
endocarditis. In asphyx ia, these haemorrhages tend to be dis Another accidental asphyxial cause of death in yo ung
tributed externally above the level of the obstruction and not children is being run over at low speed by reversing cars,
below the level, and are thus commonly found: on the palpe not infrequently driven by one of the other members of
bral and bulbar conj unctivae ; on the face - mainly around their own family and sometimes in their own driveway. The
the eyes; behind the ears (pinnae); and on the mucosal sur child's short stature precludes their visibility to the driver
faces of the lips and nose. In children it is not uncommon to with a consequent accidental knocking over of the child
find these haemolThages in a single site only and not neces who, having a low centre of gravity, will tend to fall below
sarily the eye. 3,4 They may be present over the face a nd ante the ve hicle rather than being lifted up and over the car.
rior chest wall in still births and in early neonatal deaths of Many children surv ive this type of insult, provided that
babies who die as a result of acute hypoxic/ischaemic insult, the duration of appl ication of the compressive force is not
particularly retroplacental haemorrhage (see Fig. 10.10). prolonged, the velocity of imp act is low and the weight of
They also occur naturally in normal children, albeit in the object trapping them is not considerable. 13 - IS Very
sma ll numbers. In a total of 116 children under the age of often they present with features of cerebral anoxia and
1 yea r, who were fully examined in child surveillance clin convulsions. They frequently have visceral injuries and
ics in Newcastle-upon-Tyne, Downes et al fo und that 27.6 extensive soft tissue crushing injury. Drag marks in the
per cent of children had one or more petechi ae, 8.6 per cent shape of directional scuff abrasions are often found on the
two or more, and 2.6 per cent had more than two. s body, indicating the direction in which the vehicle has
Petechiae can also occur if the child has been experienc moved over the child's body. Many of these children are
ing pronounced and recurrent Valsalva manoeuvres, as yo unger than J years and boys predominate.
with coughing (for example pertussis infection , bronchioli This matter is a serious public health issue. In Aus
tis, vomiting [gastroenteritis], crying, straining, temper tralia, 16 this type of accident acco unts for 8 per ce nt of pae
tantrums). This may also explain why petechiae may be diatric pedestrian fatalities, in New Zealand 10.7 per cent
discovered after active cardiopulmonary resuscitation. 6 and in the USA 20 per cent. 17 ,18 Ultrasonic transceivers
They a lso occur interna ll y with any form of hypoxaemia located on the rear bumpers have been introduced as prox
as the final common pathway for death. They are character imity warning devices.
istic of the sudden infant death syndrome, in which they are
found on the thymus, the epicardium and the pleurae. 7 - 9
ENTRAPMENT ASPHYXIA
object is large enough to occlude the airway completely, it were at play. Knight 27 suggested an overstimulation of the
will lead to immediate asphyxia by preventing any gaseous sympathetic nervous system (resulting in arrhythmias, par
exchange in the lungs and death results within minutes. Pas ticularly ventricular fibrillation) caused death.
sage of the object beyond the calina may still be quite dan In some of these cases in older children, a gas or a
gerous and can cause serious breathing problems; it may vapour, for example solvent vapour, may have been intro
also result in death from bronchospasm in those who are duced into the polythene bag and this may also have influ
susceptible. Common objects include seeds, nuts, bone frag enced directly the mode of death.
ments, nails and screws, small toys and pins. 2J.22
As the angles of bifurcation of the main stem bronchi are
OVERLAYING AND WEDGING
acute, for the first 15 years of life foreign bodies may find
their way into either side of the lower airways.23 Once aspi
Overlaying is the accidental death of a child by smothering
rated, objects may subsequently change position or migrate
resulting from a larger individual sleeping on top of the baby
distally, particularly after spontaneous or external attempts
in the course of deep sleep or sleep induced by extraneous
at removal of the object, after thumping on the back of the
intoxication . This has been a well documented condition
patient or attempting to make him or her cough or retch.
since biblical times 28 and tended to be much commoner in
This may cause delayed obstruction. Inhaled vegetable mate
Victorian times 29 when parents in a state of inebriation may
rial may swell over subsequent hours or days, and cough,
have taken to bed a large number of their children. 3o This
stridor, wheeziness, breathlessness and cyanosis may ensue.
hazard is greatest in infants below the age of 5 months but
Other objects, such as peanuts and other organic foreign
can occur in children of up to 2 years of age. 3J ,32
bodies, may, in addition, excite an acute inflammatory
The rolling over by the adult to co me to lie on top of the
response with tissue swelling, which makes the level of
child exerts pressure on the child 's face against bedding,
asphyxia gradually worse; children aged 1-3 years are more
the mattress or into the body of the sleeping adult or a
at risk and death results at a rate of 0.7 per 100000 per
co-sleeping older child. Owing to the pressure being exerted
annum in the USA. This is due to the tendency of young
on the child's chest, he or she is unable to cry out and
children to put everyth ing into their mouths and the way in
attract attention. Some of these babies show no pertinent
which they chew food. Because their molars are unerupted
clinical signs, often not even petechiae. 33 Some babies
they tend to use their incisors; objects a nd fragments of food
show the presence of contusions and abrasions, as well as
are then propelled posteriorly, thus exciting a reflex reaction
an unusual distribution of the lividity demonstrati ng the
and a tendency to inhalation. 24 ,25
points of compression. Indeed, it is stated that if a child
who appears to have died of overlaying shows other exter
nal blunt force injuries, it is likely that the child has been
PLASTIC BAG ASPHYXIA
the subject of non-accidental injury.34,35
In wedging, the child may be wedged by the bed's
In the later 1950s, polythene bags started to be used for
co-occupants against a wall, between the mattress and the
packaging and as supermarket carrier bags. It was soon
wall, the bed frame or an adjacent piece of bedroom furniture,
reported that children had died directly as a consequence of
the cot sides, railing in the cot sides, the head- or footboard of
these plastic bags being placed over the head, although not
the bed, other co-sleepers including other children etc., with
necessarily secured around their neck. 26 In most countries,
similar consequences. 36,37 Another hazard has been associ
it is now obligatory by law that all polythene bags used for
ated with co-sleeping on waterbeds. Kirchne.-J7 identified 515
wrapping have holes intrinsically cut into them close to the
such deaths in children below the age of 2 years during a 7
base to allow air to enter if they are slipped over the head;
year period. These included 121 deaths due to overlaying by
however, this legislation does not cover such household
a parent, sibling or by other adult - with 77 per cent of the
items such as rubbish bin bags, carrier bags etc., and the
deaths involving children younger than 3 months; 394 deaths
role of child carers cannot be overestimated with respect to
due to wedging in beds - 296 on regular ad ult beds, 79 on
the prevention of these deaths.
adult waterbeds - with onJy two involving previous alcohol
The imm ediate cause of death in these deaths was at one
or drug abuse, 10 on adult day beds, nine in adult-sized beds
time thought to be due to a decreasing oxygen concentra
fitted with bed rails. 38 The larger the number of persons occu
tion within the bag and rebreathing of carbon dioxide, with
pying the beds and the higher the weight of the occupants,
consequent narcosis and loss of consciousness. Physical
the greater is the risk involved. Unsafe sleeping environments
obstruction of the nose and mouth may also have aided
for babies are further discussed in Chapter 11.
and abetted this mode of death. In addition, it was pro
posed that the bag becam e electrostatically charged and
adhered to the face, aided by condensation of water vapour STRANGULATION
from the expired air. However, as there is a dearth of typi
cal asphyxial features and the persons concerned looked Non-intentional, accidental self-sJrangulation of young
pale and placid, it was postulated that other phenomena children with loose wires, cords and other potential ligatures
332 I Asphyxial deaths in children
found commonly around the house and often in close the low oxygen tension causing endothelial cellular dam
proximity to their beds are well documented. 39 - 4 ! Entangle age and hyperpermeabili ty with leakage of body fluids into
ment in such cords was responsible for 14.3 per cent of Amer the extracellular space. 51 Hyponatraemia may also develop
ican childhood deaths. Other causes were plastic bags, bedding if large quantities of fresh water are swallowed or if the
(non-plastic), cords of blinds or curtains, cords by which syndrome of inappropriate antidiuretic hormone secretion
dummy teats (pacifiers) were attached to them and other types occurs.
of cord. 32 An intravenous tubing set was also reported as hav Inhalation of 13 mL/kg of fluid can result in signifi
ing caused such a fatality in a I-month-old boy and a near cantly impaired gas exchange. Fresh water, being hypo
fatality in an 8.S-month-old boy,42 and the wiring of an tonic, damages type 2 pneumocytes and causes a
apnoea monitor, which was non-fatal, in another baby.43 disruption of alveolar surfactant. Sea water draws fluid
from the blood into the alveoli and thus dilutes surfactant,
resulting in lower residual pulmonary functional capacity
HANGING BY A LIGATURE
and pulmonary oedema, with development of an acute res
piratory distress syndrome.
Deaths by hanging in children below the age of 14 years
There may also be plugging of small airways by debris,
are uncommon. Much of the published literature relates to
which increases airway resistance. Release of inflammatory
individual case reports, with no epidemiological studies
chemical mediators from the lungs leads to local vasocon
being undertaken. 44 - 46 Cases usually occur against a back
striction. Biochemical changes may lead to cardiac dys
ground of unhappiness or frank depression. A history of
rhythmia or asystole due to hypoxaemia. There may also be
poor achievement at school, parental and peer bullying and
metabolic acidosis and acute pulmonary coextension.
similar adverse effects come to light after death when a
In the USA, for each death from drowning (about 1500
psychological review is carried out. There may have been
children per annum) an additional four hospital admissions
previous attempts at self-harm.
per drowning death occur after submersion incidents. 52 ,53
Autoerotic unintentional deaths from suspension have
Drowning and near drowning are further discussed in
been reported in children aged 9 years and above. 47 - 49
Chapter 18.
-- -- ----------------------------------
-- --
-- ----
Prevention I 333
It may be extremely difficult to differentiate between a nat • accidents while und er the influence of these
ural phenomenon and abuse phenomenon.63 substa nces, due to loss of judgement and impairment
It can be extremely difficult to find any signs of the of other cognitive functions;
external airways obstruction, although a very careful • burns produced by inadvertent ignition of the
search for bruising on the face, on the inside of the lips and inflammable accelerant used;
cheeks should be made; in the mucosa l sites bruising may • long-term toxic effects of the solvent on the marrow,
be accomp anied by minor abrasions. liver, kidneys and brain;
Internally, the presence of haemosiderin in the lungs, • chronic respiratory problems due to chronic bronchitis
within both alveo lar cells and in the interalveolar septa, is and recurrent inhal ation al or chemical pneumonitis.
of great significance. In the acute phase there will have
been haemorrhage within alveoli and this may give rise to Note that in some older children, these substances may
frothy frank ly blood-stained fluid emerging from the nose be used as sexual aid in the course of autoerotic activities.
and mouth. In cases when such abuse is suspected, there may be lit
tle to find at autopsy. There may be a rash around the
mouth and blisters where the substance had been app lied
ABUSE OF INHALANTS (SOLVENT ABUSE) and produced a solvent degreasing effect on th e skin. The
so lve nt may still be appreciable through its smell at
Inhalants, usually hydrocarbons, are breathable chemical autopsy. One lung should be retai ned in a nylon fibre bag
vapours that produce mind-alteri ng effects similar to the to allow 'a ir' to be sampled from it directly from its airways
effect of alcohol consumption or the use of controlled dnlgs. for gas chromatography assay. The blood , the liver and the
They are an ingredient in many househoJd products, such as kidneys can also be assayed toxicologically. Histology may
agents used in cleaning, decorating, painting (turpentine, confirm chronic changes in the internal organs particularly
white spirit), sta ins, vamishes, g lu es, air fresheners, hair con the lungs. No petechial haemorrhages are found either
ditioners, dry cleaning agents, shoe polish, colour markers, internally or externa lly, except in the few instances of
na il polish removers, spot removers and degreasers and suffocation.
deodorants, as well as in fuels such as petrol (gasoline), butane
and lighter fuel. Anaesthetic agents, such as ether, chloroform,
halothane and nitrous oxide, may be used in a similar manner. REVERSE SUSPENSION
The so lvents are taken in to the body by: squirting
directly into the mouth from cans and canisters, inha ling This is perhaps a rarity, which may occur in children dur
from bags or an aerosol can, sn iffin g directly from the lid ing play activity when death results from exhaustion of
of the container, breathing in directly (huffing) or after breathing du e to splinting of the diaphragm 64 ,65 by the
application of the substance in a rag or handkerchi ef. Plas upwards displacement of abdominal contents. Death occurs
tic bags containing the vapour may be placed directl y into slowly from ex haustion of respiratory effort.
the mouth. There are about 1400 compounds that can act
in this manner, including nitlites, alcohols and halo
genated compounds, in addition to hydrocarbons. CHEMICAL ASPHYXIA
Death may occur in a variety of ways:
• direct acute toxicity of the brain, leading to respiratory This term is applied to situations when irrespirable gases
a nd cardiac arrest; find their way into the child's environment. For example,
• cardiac arrhythmias brought on by the substance carbon monoxide in fires and from exhaust fumes (in
inhal ed, followed by cardiac arrest; this may be closed garages) and from charcoal fires (barbecues) in an
brought about by sudden SPUltS of muscular activity enclosed space, chlorine escaping from swimming pools,
and release of a bolus of catecholamines, as if the hydrogen sulphide and methane from refuse tips, old mines
subject while under the influence is chased and made and slurry pits are among the gases that can cause death
to run away; after inha latio n. These events usu ally involve older, active
• suffocation by the item being used to present the children and, more frequently, boys, in common with many
vapo ur to the mouth, for example plastic bag or environm ental mishaps.
balloon;
• vomiting w hile unconscio us from the effects of the
inhalants; PREVENTION
• suffi xa tion by displaci ng air from the lungs with an
irrespirable vapour; As incidents involving asphyxia are common, public
• acute parasympathetic stimulation with inhibition if health offiCials, community doctors, midwives and health
me substances cause irritation and/or cooling of the visitors have been involved in several programmes aimed
upper gullet or n asophary nx; at alerting parents to the dangers of asp hyxia. 66 In severely
334 I Asphyxial deaths in children
disab.led children accidental asphyxial deaths are more 24 Byard RW. Mec han isms of death in in fa nts a nd young
common. 57 In the neonate and in young children below the children fol.l.o wing fore ign body ingestion. } Forensic Sci
1996; 41 :438-41.
age of 12 months sleeping practices also have to be care
25 Baharloo F, Veyckemans F, Fra nci s C et al. Tracheo-bronchial
fully scrutinized to enquire that they are not potentially fore ign bodies: presentation a nd management in children and
dan gerous to the child. 6B - 72 Co-sleeping in particular adults. Chest 1999; 115 : 1357 - 62.
should be viewed with some discretion, particularly if the 26 Plast ic bags (Ed itoria l). BM} 1959; 1 :1463-4.
parents indulge in smoking and drinking of alcohol. 73 75 27 Knight B, Sa ukko P. Forensic Pathology, 3rd edn. Oxford:
Oxford University Press, 2003 .
28 The Bible, 1 J<jngs Ch . 3 v.19.
29 Templeman C. 258 cases of suffocation of children. Edin Med
REFERENCES } 1891; 38:22- 6.
30 Nakam ura S, Wind M, Dane ll o MA. Review of hazards
Tardieu A. Etude medico -lega le sur les sev ices et ma uva is associated w ith chi ldren placed in adult beds. Arch Pediatr
tra itmen ts exe rces sur des enfants. Ann Hyg Publ tvled Legale Adolesc iVIed 1998; 1 53: 10 19- 23.
1860; 13 :36 1-98. 3 I Drago DA, Dannenberg Ai. Infant mechan ical suffocation
2 Luke JL. Conjunctival petechiae. IV Eng/} iVIed 1971 ; dea th s in the United States 1980-1997. Pediatrics 1999;
28 4:1101. 15 3: 103 -59.
3 Rao VJ , Wetli CW The foren sic sig nifica nce of co njun ctiva l 32 O'Hara M, Harruff R, Smialek J , Fow ler D. Sleep location and
petechiae. Am } Forensic ivfed Pathol 1986; 9:32 4. infa nt s uffocat io n : how good is the evidence? Pediatrics
4 Jaffe FA. Petechial haemorrha ges. Am } Forensic iVIed Patl10l 2000; 10 :9 15-20.
1994 ; 15:203 -7. 33 Moore L, Byard RW Path o log ical fInd ings in hang ing and
5 Dow nes AJ, Crossland OS , Mell o n AF. Preva lence and wedgin g deaths in infants and yo un g children. Am } Foren sic
distribution of petechiae in well babies. Arch Dis Child 2002; ivIed Pa tl101 199 3; 14 :296- 302.
86:291-2. 34 Collins K. Death by overlaying and wedg ing: a 15-yea r
6 Hood I, Ryan 0 , Spitz WU. Resuscitation and petec hiae. Am } retrospective study. Am} Foren sic iVIed Patho1200 1; 15 : 155-9.
Forensic iVIed Pat/101 1988; 9:35-7. 35 Mitchell E, Kraus HF, Byard RW Pathol ogica l findin gs in
7 Beckwith JB. Intrathoracic petechial hemorrhages; a clue to ove rl aying.} Clin Forensic iVIed 2002; 9:133 - 5.
the mechanism of death in SlOS. Ann N Y Acad Sci 1988; 36 Byard RW, Beal S, Bourne AJ. Potentially dangerous sleep ing
533:37-47. environment and accidental asphyxia in infancy and ea rly
8 Krous HF, Jordan J. A necro psy study of distribution of ch ildhood. Arch Dis Child 1994; 71 :497-500.
petechiae in non-SlOS. Arch Pathol Lab Med 1984; 108:75- 6. 37 Kirchner JT. Deaths associated with small children sleeping in
9 Kleeman WJ, Wiechern V, Schuck M, Troger HD . Intrathoracic adult beds. Am Acad Fam Phy 2000; March I ,
and subconjunctival petec hi ae in SlOS . Forel1sic Sci lilt 1995; www.aaf.org/afp.
72:49-52. 38 Nakamura S, Wind M, Danello NA. The pediatric forum:
10 Bass M, Kravath RE, Gla ss L. Death-scene inves ti ga ti on in should infants sleep with their parents 7 Arch Pediatr Adolesc
sudden infant death. N Engl} Nled 1986; 315 : JOO-5. ivIed 2000; 154:1171-3.
II Iyasu I, Hanzlich R, Rowl ey 0, Willin ger M. Proceedings of 39 Nixon JW, Kemp AM, Levene S, Sibert JR. Suffocation,
'Workshop on Guid elin es for Scene Investigat ion of SlO S'. choking and strangulation in children in Eng land and Wales;
} Forensic Sci 1994; 39 :11 26 - 36. epidemiology and prevention. Arch Dis Child 1995; 72:6 - 10.
12 CDC. Guid elines for dea th scene investigation of sudde n, 40 Rauchschwalbe R, Mann NC. Pediatric wind ow cord
unexplained infant deaths: recommenda ti ons of th e strangulations in the United States, 1981-95. }AMA 1997;
interagency pane l on SlOS. MMWR 1996; 45:RR-IO. 72: 1696-8.
13 Haller JA, Donahoo JS. Traumatic asphyxia in ch ildren: 41 Sabo RA, Hanigan We, Flessner K et a!. Strangulation injuri es
pathophysiology and management.} Trauma 1971; 11 :453-7. in children.} Trauma 1996; 40:68-72.
14 Campb ell-Hewson G, Egleston CV, Cope AR. Traumatic 42 Garros D, J<jng WJ, Brady-Flyer Bet a!. Strangulation with
asphyxia in children. Accidel1t El11 erg iVIed 1997; 14(1):47-9. intravenous tubing; a previously und ec id ed adverse even t in
15 Sarihan H, Abes M, Akyazici R et a!. Traumatic asphyx ia in children. Pediatrics 2003; III :732-4.
children. Cardiovasc Surg 199 7; 38:93 -5. 43 Emery JL, Taylor EM, Carp ente r RG, Wa ite AJ. Apno ea
16 Holland AJA, Liang RWY, Sing h SJ et al. Driveway motor monitors and accidental strangulation. BM} 1992; 304 :11 7.
ve hi cle injuri es in children. iVI}A 2000; 173: 192 - 5. 44 Petruk J, Shiels E, Cummings GE, Francescutti LH, Fatal
17 Brison RJ, Wicklund K, Mueller BA. Fata l pedestrian injuries asphyxiations in ch ildren involving drawstrings in cloth in g.
to yo un g children: a differe nt patte rn of injury. Am } Public Can Med Assoc} 1996 ; 155: 1417- 19.
Health 1988; 78 :79 3- 5. 45 Bya rd RW, Ma rcopo ul os D et al. Early adolescent suicide; a
18 Roberts I, Norto n R, J ackson R. Driveway-related c hild co mparati ve study. } Clin Forensic Med 2000; 7:6-9.
pedestrian injuries; a case control study. Pediatrics 1995; 46 Wyatt J, Wyatt PW, Sq uires T, Busuttil A. Hang ing deaths in
95:405-8. children . Am } Foren sic Med Pathol 1998 ; 19 :3 43-6.
19 Sumra ll RE, Petty CS, Holman W. Temperature in closed 47 Feldm a n KW, Simms RJ. Stra ng ulatio n in children:
auto mob iles in hot weather. Forensic Sci Gazette 1976; 7:7-8. epidemiology a nd clinica l co urse. Pediatrics 1980;
20 Surpure J. Heat-rela ted illness a nd the a uto mob il e. Ann 65: 1079-85.
Em erg iVIed 1982; It :263-5. 48 Cooke CT, Cadden GA, Hilton JMN. Hanging dea ths in
21 McGuirt WF, Ho lmes HD, Feehs R et a l. Tracheo-bron c hial children. Am } Forensic Med Pathol 1989; 10:98- 104.
fo reign bodies. Lm)'ngoscope 1988; 98:615- 18. 49 Uva JL. Autoero ti c asphyxiation in the United States.
22 Weiisberg D, Sc hwartz I. Foreign bodies in the tracheo } Foren sic Sci 1995 ; 40:574 - 8 1.
bronchial tree. Chest 190 7 ; 1: 730 - 3. 50 Bya rd RW, Lipset J. Drowni ng deaths in tod d lers and pre
23 Cleveland RH. Symmetry of bronchia l angles in chi ldren. ambulatory children in South Austra lia. Am} Forensic iVIed
Radiology 1979; 11 3:89 - 93. Path ol 1999; 20:328-3 2.
References I 335
51 Nodlkel JH. Drow ning. N Engl ] Med 1993; 328:253-6. 63 Recurrent apparent life- threatening events and intentional
52 American Academy of Pediatrics Committee on Injury and suffocation (Editorial). Arch Dis Child 1999 ; 81:189.
Poison Prevention. Drowning in infants, children and 64 Friberg TR , Weimreb RN. Ocular manifestations of grav ity
Anthony Busuttil
-- - -- ---- ~
Overview of paediatric trauma I 337
Table 17.1 Mortality (deaths per 700000) from injury: ch ildren aged 7- 74 year53
All injuries Transport Fire and flames Falls Poisoning Drowning Firearms Intentional Others
Table 17.2 Mortali ty fram injuries in Organisation for Economic abno lmal an inactive, sedentary young child . This combina
Co-operation and Development (DECO) countries, by age and tion of childhood inquisitiveness, mobility and activity te nds
gender 1991- 7995, per 700000 af the population to be impulsive and experimental. This is superimposed on
a n obvious backdrop of inex perience and a n inherent imma
Age (years)
turi ty of social and intellectual skills to cope wit h adverse
1-4 5-9 10-14 situat ions as and w hen they present th emselves. By their
very nature, children take risks and are sensation seeking.
All of th is compounds their proneness to injury.24
Boys 21 16 16
Falls are therefore inevitable. The graspi ng and pulling
Girls 14 7 7
of objects towards the infant may result in objects being
From the WHO mortality database, with permission. pulled on to the child, and to scalds and burns.25 As the
child's territory expands and manual dexterity and p hysical
The use of the term 'accident' to describ e such traumatic strength improve, play activity becomes more adventuro us
episodes is often a misnomer, as few of th ese injuri es are and the opportunities for injuries increase. As organized
actually due to unpredictable irremedi a bl e ev ents and situ sporting activities are taken up these further raise th e prob
ations; rather they are more often due to interaction with ability of injury, be this in organized and team sporting
the environment in which children are living and playing, activity - participation in team sports such as rugby, soccer
the products that they use and the level of supervisio n t hat and hockey - or, more simply, in such recreational activiti es
they are allocated. Many of these injuries a re preventable as kicking a ball, riding a bicycle or using a sWing. 26 The
and thus the main thrust of the investigation a nd ana lys is more v igoro us the sport (e.g. karate, judo, Tae Kwon Do,
of figures from childhood injuries should be in attem pts at boxing and skateboarding) the more likely is the occurrence
prevention. 19-21 of inju ryY
There is no simple correlation between mOltality staris The role of carers in ens uring that the children do not
tics and non-fatal injury data. 22 FUithermore, the relarive take excessive risks, and indeed to be t here to rescue them
population m ortali ty data are hig her in Eastern Europea n from the hazardous situ at ions in w hi ch they find them
countries. 23 Why this should be so is difficult to determine selves, is essential. When the carers are employed in this
but, at least in palt, it may be re lated to the different capacity by nurseries and schools, there is an ex pectation
perception of the seriousness of injury a nd a populist of a duty of care. Sometimes - indeed , increasingly fre
fatalistic attitude; there may be a tendency to decry the quently - this is called to account in a cou rt of law when
seriousness of injury combined with Jess willingness to this duty is perceived to have been breached. 28 Participa
seek medical help for injury until it is too late. tion in sport also carries with it the possibility and, per
haps, in more viol ent sporting activity, a probability, of
almost inevitable injury. This willing, wholehearted partic
Accident Proneness in Children ipation by the child with the sanction of his pa rents, often
expressed legally in th e max im volenti non fit iniuria ('to one
Intrinsically the behaviour of children is based on curiosity who is willing, no ha rm is done'), m ay decrease the culpa
and constant activity. Many paediatricians would class as bility and answerability of the carer when things go wrong
338 I Accidental injuries in children
but it also bears the expectation and responsibility from cycle tracks in urban areas has also assisted in the decrease
them that all that can be done to supervise children's activ of such incidents. J5 ,36
ities - to decrease the risk, to provide protective clotting Children are most frequently hit while walking along the
and equipment - has been fulfilled in full exercise of their side of the street, during play in the street or while darting
duty of care. 29 into or crossing the street. Children are occasionally hit by
motor vehicles while bicycling, but they may also be injured
when fallin g off their bicycles after hitting objects such as
Causes and Mechanisms of Injuries trees, kerbs or walls. Few of those children admitted to
trauma units after bicycling-related injuries, were wearing
Falls are the leading cause of injuries. Among young chil helmets. 37 Other motor vehicle-related injuries involve
dren these are often from fu rniture, such as cllanging tables, motorcycles, all-terrain vehicles and recreational vehicles.
sofas and couches, cribs, beds or fallin g down stairs, pal1ic Children with hearing impairment from any cause,
ularly in walkers,JO during momentary periods of inattention including chronic ot itis media a nd 'glue ear', are more
by their carers or because of the activity of older siblings. prone to accidents as they may not be aware of vehicles in
As toddlers begin to explore, the oppOl1Unity for falls from their vicinity.
heights increases; this includes recorded falls down stairs.
Less common are falls from balconies, porches or out of
windows;Jl these inciden ts are more frequent in warmer Consequences
climates when children have more open access . Nearly
20 per cent of the children adm itted after falls have fallen Many injuries occur while a child is engaged in some kind of
2.5 metres (8 feet) - the height of one storey or more. activity such as nll1ning, playing, climbing or bicycling. Even
The next major mechanism of injury is motor vehicle when the child is in a more station ary position, such as rid
crashes with children as occupa nts. Most crashes involve ing in a car, the child is part of an activity involving motion
collisi ons with other vehicles. Only one-quarter of those chil and speed. The combined effects of motion, speed and impact
dren injured have been using some kind of protective can injure many regions of a child's body. Two or more body
restraints, such as child safety seats or safety belts. Many regions were affected in about one-half of the children. Chil
unrestrained children were hurled aro und inside the car dur dren typically had multiple injUlies that resulted in several
ing a collision or thrown out of the car through windows or diagnoses: head inj ury is the most frequent diagnosis among
doors. Breast-feeding in the car or holding a child in one's lap children recorded, followe d by fractures to the bones of
is dangerous as the child will be unprotected during a crash. extremities and torso. The third most frequen t diagnosis is
Many other children are injured fatally when they walk, open wounds, mostly lacerations.
cycle or play on the roadway and are struck by motor veh i The severity of injuries is directly linked to how the chi l
cles.J2 Thousands of children are killed annua lly on the dren were injured. Children who are injured in motor vehi
roads but most of the deaths are incurred when the child is cle crashes and as pedestrians are more likely to have four
on foot. There is correlation between the number of child or more functional limitations. The force and speed of col
hood fatalit ies on the roads and the size of the popUlation of lisi ons are factors determining the severity of injuries. Chil
a country : the bigger the child population and the lower the dren injured by falls are less seri ously injured and have
socioecon omic status, the higher is the likelihood that unsu fewer functional limitations.
pelvised children are on the road and thus find their ways
into the path of moving traffic. This may also be the reason
for the disproportionately higher incidence among children BICYCLE HELMETS
of certain ethnic groups.JJ Many of the deaths occur in the
urban location at speeds of less than 30 miles per hour and Experienced, careful bicycle riders crash every 4500 miles
usually within a Sh0l1 distance away from their homes. on the average. Head injury causes 75 per cent of our 800
Most of these deaths occur in the holiday periods and a t plus an nual deaths from bicycle crashes. Medical research
other times during the parts of the day w hen the children shows that a bicycle helmet can prevent 85 per cent of
are not actually at school. Thus more injuries occur in the cyclists' head injuries. 38.39
summer months. The institution of school crossing patrols A helmet reduces the peak energy of a sharp impact. This
manned by retired individual s and the training of children requires a layer of stiff foam to cushion the blow when
in road safety has produced major dividends in terms of crushed. Most bicycle helmets do this with the use of
reducing co llision-related injuries and deaths amo ng chil expanded polystyrene (EPS), the white picnic cooler fo am.
dren; the aims have been to teach children to be vigilant Once crushed, EPS does not recover. Another foam,
and to improve their skills in quickly integrating and expanded polypropylene (EPP), does recover but is much less
assessing movement, relative velocity, distance and depth commonly utilized. A stronger EPS called GECET appeared
in relation to moving motor vehicles, along with the inter in 1992 and is widely used now and another synthetic foam,
pretation of road signs. 34 The setting aside of exclusive expanded polyurethane (EPU), is used in Taiwan as it has a
Falls I 339
unifonn cell structure and crushes without rebound; it is severe force is in volved or in falls from a height, fractured
heavier and requires a manufacturing process that is not long bones a nd closed head injuries may occur 48
environmentally friendly. The spongy foam placed on the The incidence of injury depends to a major extent on the
inside a helmet is for comfort and fit, not for impact. height from which the fall occurs and also on the terrain
The helmet must stay on your head even when the head that the chi ld lands on - the harder and more compact the
is hit or bumps on to a hard unyielding surface, more than latter, the higher is the incidence and severity of injury.49
once in a moving traffic incident: usually there is a bump The height of climbing frames, swings, etc. is another
with a ca r first, and then with the road. It thus needs a factor involved in injury causation and the height of the
strong strap and an equally strong fastener. The helm et fa ll is directly related to the severity of the injuries sus
should sit level on the head and cover as much as possible tained by children in playgrounds. Injuries are more likely
of its surface contour. Above all, with the strap fastened, to occur with equipment that is two 2 metres above the
one should not be able to get the helmet off your head by ground ,5o so that reducing the height of swings, monkey
any combination of pulling or twisting. If it comes off or bars and other equipment improves safety.
slips enough to leave large areas of the head unprotected, Reports of children falling out of windows are all too
the straps must be readjusted otherwise a better-fitting hel frequent in the summer. Height off the ground and landing
met will be requ ired . The strap must feel comfortably snug surface are two critical factors in the injuries to children
when riding the cycle. that result from falls. Height is directly related to the speed
Most helmets made of EPS foam ha ve a thin plastic at wh ich a child falls and the force of impact upon landing.
outer s hell. The shell helps the helmet skid easily on rough Falls are particularly common among children of p re
pavement to avoid jerking of the neck during such an school age who are developing their balance and learning
impact; it also holds the EPS together after the first impact. to walk, climb and run ; however, their physical skill s
Some excellent helmets are made by moulding EPS in the develop sooner than their abilities to anticipate the conse
shell rather than adding the shell later. quences and risks of their actions.
Excessive vents in the helmet mean less actual area con Amongst falls from heights, about one-quarter of chil
tact with the head, which could concentrate force on one dren fell from w indows. Examples of other falls from
point. 'Aero' helmets are not noticeably faster and in a heights include falls from roofs, out of trees, from ski lifts,
crash the 'tail' could snag or knock the helmet aside. off porches or balconies and down flights of stairs. Even
Skinny straps are also less comfortable. Dark helmets are escalator falls have been known to cause serious injuries. 51
hard for motorists to see. Rigid visors can snag or shatter in The incidence of injuries among children h as a distinct
a fall. Helmet standards do not address these problems a nd seasonal pattern, with an increase in the warmer months of
should be reconsidered. the year. This is even more so for falls from windows as
There can be little doubt that bicycle helmets have pre they are more likely to be open in warmer weather. Tod
vent some of the more serious head injuries and undoubtedly dlers and pre-school children are particularly vulnerable as
they have saved the lives of young cyclists. 4o - 42 However, it they lack the judgement needed for risk assessment an d
has been observed that some children, particularly older safety precautions. Falls occurred in th e chi ld's hom e in 96
children, need to be objectively convinced that helmet wear per cent of the cases. Falls from two- to three-storey win
ing is indeed not just useful in preventing injUl)' and that the dows were most common (72 per cent), followed by fo ur
money spent on their purchase rather than other cycl in g to five storeys (18 per cent). Only 8 per cent of children
clothes and accessories is money well spent. Above all cycle admitted to trauma centres fell from firs t-sto rey windows;
helmets had to be shown not to make their wearers less 'cool' however, 3 per cent fell from six storeys or more.
and trendy but vice versa, and also that their use makes the Head injuries, with or without dam age to other body
wearers appear more professional. 43 A variety of strategies regions, occur among 65 per cent of the children . A small
have been devised worldwide to encourage the purchase and child 's head is at special risk for injury from a fall because
wearing of cell helmets with varying success levels. 44- 47 it is disproportionately large and heavy in relation to the
rest of the child 's body. Height, rate of fall and the force of
impact have combined effects that result in injuries to
FALLS multiple body regions in about one-half of injured children.
The greater the height of the fall, the higher the death rate.
All children trip and fall, indeed from the time that initially Among the children who fell from windows, six storeys or
they stand up on their own two feet. Falls a re more com more, 8 per cent died. Among those who fell four to five
mon while the muscle tone is not fully developed but storeys, 5 per cent died; among those who fell two to three
become more frequent once aga in when gait allows chil storeys, 2 per cent died. Non e died who fell from one storey.
dren to explore farther and when their innate inquisitive Upon hospital discharge, 73 per cent of the children who
ness takes them further afield. fell from windows had no identified functional limitations.
Contusions, haematomas, lacerations and abrasions For the 21 per cent who had one to' t hree limitations, these
tend to occur at the site of such fa ll s. OccaSionally, if more were most commonly in bathing, dressing, walking and
340 I Accidental injuries in children
se lF-Feed ing. By contrast, the children with four or more rider, monkey bars, overhead ladder, trapeze/trapeze rings,
functional limitations at discharge had more difficulties in pl ay houses and a maze. Th e scope of many of these items
cogn ition , behaviour speech, vision and hearing, reflecting is to assist children to deve lop physical coordination,
the severity of head injuries in this group. Among t he chil strength, and flexibility, as well as providing recreation
dren who fell from six or more storeys, 42 per cent had one and enjoyment. Common in modern playgrounds are 'play
to three limitatio ns; 8 per cent had four or more limitations. structures' that li nk many different pieces of eq uipment.
More functional limitations were identified as the height of A little more than one-half of the children injured are
t he fall increase d; children who fell from four to five storeys boys, with the largest percentage of injured children in the
had the largest proportion of functional limitations - four or age bracket of between 5 and 9 years 0ld s2
more. Ho wever, because more children died who fell from Protective surfaces (suc h as rubber mats, wood mulch,
fi ve storeys or more, there were fewer surv ivors in that group bark or soft sand) of sufficient depth should be installed
with four or more limitations. und er and aro und playground equipment; 53 there is no
Almost one-half (47 per cent) of the falls occurred on place for concrete, asphalted s u rface or sto ne slabs. 54 .55
playgro unds in recreational areas, one-quarter (25 per cent) This, togeth er with t he removal of monkey bars from parks,
occurred at home, and almost one-fifth (19 per cent) resu lts in a significant reduction of inju ry rate. 56
occurred on sc hool playgro unds. About 90 per cent of these Equipment must be insta lled properly, with sufficient
injuries occurred in falls from playground eq uipment: space a round it. Pl aygro und eq uipment needs to be
sw ings, slides, climbers (monkey bars, jun gle gyms or other inspected periodically for structural defects, peeling paint or
climbing apparatus) and trampolines. Children fell from dif splinterin g wood a nd these should be repa ired. Protective
ferent types of playground equipment depending where the surfaces under playground equipment should also be
equipment was located. At home, children most often fell inspected periodically, as surfaces such as wood chips or
from swings or trampolines, whereas at schoo l and in recre sand lose their protective quality if they get compacted by
ation areas they most often fell from climbing equipment. regul ar usage or are worn awayS7
Children of different ages fell from different types of
playground equipment. The youngest children fell most
often from slides, whereas older children fell most often SPORTS INJURIES ON SNOW AND ICE
from climbing equipment.
Seven of the children died in the hospital: six from head Areas with cold te mpera tures, snow and ice can make ideal
injuries and one as a result of a severed sp in al cord. Three conditions for recreation, but sports on snow and ice create
of these children fell from slides, two fe ll from sw in gs, o ne specia l ha za rds for children. Over one-half of the children
fell from monkey bars and one fell from a trampoli ne. Four with t his type of injury are injured while sledging; the next
of the children who died fell from playground equ ip men t at largest groups have injuries caused by skiing, snowboard
home; the other three fe ll fro m playground equ ip me n t in a ing or playing hockey - fewest children were injured while
recreational area . ice-skating.
Falling or hitting objects such as trees, rocks or fro ze n
ground injured most children. Head injury was the most
PLAYGROUND INJURIES frequent diagnosis. One out of four children had broken or
fract ured bones or had multiple injuries or required sur
Playground s a re considered as oases from road traffic gery. The average length of hospital stay was 6 days. More
places w here children can safely enjoy themselves and than 14 days were spent in the hospital by on ly 10 pe r cent
indeed develop interpersonal, social and team skills. As of the children .
childhood obesity is bandied around as being of epidem ic The combination of speed with a fall or coUision on ice
proportions, t he impo rtance of play activities cannot be and snow increases the risk of serious injuries . Over one
overemphasized. Ho wever, playgrounds - whether at half of the children had head inj uries and few, if any, of
hom e, at schoo l or at rec reational areas - can also be dan these children had been wearing helmets.
gerous. To be safe, all public and private playgrounds must Data from a study of sk i rental fac iliti es in 1999 looking
be well designed , we ll mai nta ined a nd used properly. Chil at 353 rental shops in five European countries (France,
dren need good su pervision , and sho uld be taught how to Austria, Germany, Sw itzerl and and Ita ly) had shown that
play safely. Children enjoy experi men tatio n a nd explo 71 per cent failed to meet the sta ndard and abo ut 13 per
ration and will incur risks, w hether calculated or hapha z cent just about reached an acceptable level. In tota l, 70 per
ard, a nd thus although play areas have to be challenging cent of rental facilities still set the bindings incorrectly
and inn ovative to keep the ch ildren 's interest, there should 66 per cent of bindings in children were found to have been
also be a backdrop of safety built into playgrounds. Mod set dangerously hi gh. 58 - 60 At the biannual [SSS sy mposium
ern playgrounds wi ll often h ave eq uipment such as a held in Japan in 200 5, injury statistics presented from
seesaw, merry- go -round, swings, slide, climbing frame, Scotland, Fra nce and Iran show a rate'expressed as injuries
walking bridge, jungle gym, c hin-up bars, sandbox, spring per 1000 part icip ant days of 1-2 per 1000 fo r alpine skiing
Older children and substance abuse I 341
and 3-4 per 1000 for snowboarding and ski-boarding. forbidding children below the age of 13 years to drive a
More common injuries are knee injuries in alpine skiers, tractor on private ground. Children are also prevented
wrist injuries in snowboarders and lower leg injuries in by law from riding as passengers on farming machinery. This
ski-boarders. Children as a group are twice as likely to be applies not only tractors, but also to harvesters and trailers.
injured as adults, especially if using rental gear and on Quad bikes are quite popular on farms and are often an
their first day of skiing as were those who had received less irresistible attraction to children; crashes with these have
than five formal lessons and those skiing with friends. resu Ited in injury and deaths.
Children were shown to be more frequently injured on lift
systems and as a result of bindings failing to release.
In relation to the use of protective equipment by skiers PREVENTION
the debate about helmet use continues. In Norway, the rates
of use amongst those injured were 31 per cent (alpine ski), A major public health issue 71 is the effective prevention of
26 per cent (snow board) 26 per cent (telemark ski) and 28 per injuries and fatalities from injuries in children and, in most
cent (ski-board). In Sweden, where about 58 per cent of all countries, there is a constant active monitoring of the
people on the slopes now wear a helmet, a 28 per cent rela incidence of serious injury and fatalities resulting from
tive increase in the risk of a head injury was observed in accidents; any patterns that emerge from such studies are
non-helmet wearers. The main protective effect of headgear addressed and this occasionally involves new Jegisla
is against more minor injmies such as abrasions and bruises. tion. 55 ,55.72 Preventative measures that are adopted are reg
The helmets in use at present simply cannot do much in ularly audited in terms of their effectiveness and fine tuned
high-speed impact accidents. Wrist guards were devised to as required.73
reduce the incidence of wrist fractures amongst snowboard The availability of guidelines and legal sanction have
ers but there are currently no international standards for enabled the commencement of, and the frequent successful
them, so it is velY difficult for a boarder to decide which to completion of, litigation procedures in the civil and, more
use and which is the best buy. Release bindings had led to a rarely, the chimerical courts, in addition to public inquests
reduction in the number of lower leg fractures and may and fatal accident inquiries.
reduce the number of knee injuries that were still occurring.
International standards have still to be agreed for them.
OLDER CHILDREN AND SUBSTANCE ABUSE
RIDING INJURIES In the older age groups, alcohol use, solvent misuse 74 and
drugs may creep into the picture; in addition young adoles
Each year children are injured and some die as a result of cents may carry weapons as part of gang culture or for
incidents while riding a horse. An adult horse can weigh alleged personal protection from bullying and other preda
over 500 kg, gallops at speeds up to 65 km per hour and tory behaviour. 75 In other countries firearms may be used
kicks with the force of 1.8 times its weight. About 65-75 in much the same way and because of easy access to them,
per cent 51 - 54 of children admitted after riding accidents injuries and deaths often result from them.
have been riding the horse, with most injuries occurring More than one person per week dies from volatile sol
to the head and in the limbs. Many of the fatalities and vent abuse (VSA). The average home has 30 kitchen and
injuries can be avoided or have a reduction their severity bathroom products that can be abused, including: butane
through a combination of better adult supervision and the gas cigarette lighter refills, liquefied domestic gas, solvent
use of appropriate protective helmets. 55 .55 based adhesives, deodorant aerosols, pain relief sprays,
aerosol air fresheners, hairspray, other aerosols, correction
fl uids, petrol, certain paints, paint thinners and removers,
AGRICULTURAL INJURIES clry cleaning agents, petrol lighter fuel, nail varnish and
varnish remover, some shoe and metal polish, and plaster
Children in farming communities often assist their families remover. In 2001, VSA was responsible for seven times the
during peak harvest times and indeed may seek employ number of fatalities than those related to ecstasy (3,4 meth
ment as fruit and vegetable pickers during vacation peri ylenedioxymethamphetamine [MDMA]). The 2003 annual
ods. In the course of holidays on farms, children may also report from the European Union's drug agency warned that
find themselves in trouble with farming equipment. This VSA is an often overlooked problem, with a big impact on
problem is well recognized in Britain as it is in Australia 57 public health of young people; the use of solvents and
and in the USA;68 between 1979 and 1988, 167 children inhalants is the next common substance misuse after
died in farming accidents in Britain and about 300 children cannabis and alcohol in 15- to 16-year-olds. One in seven
die annually in the USA.69 15- and 16-year-olds in the UK sniffs solvents to get high.
Tractor-related injuries are the most common in More than one person dies every week in the UK from the
Britain,l° even although there is legislation in existence effects of solvent abuse - between 70 and 100 each year.
342 I Accidental injuries in children
The dru gs charity SOLVE IT says so lven t abuse kills 60-75 occur most frequ ently whe n routines are disrup ted, for
youn g people each year. One- thi rd of young people that die exa mpl e moving and vacations. Chi ld safety caps have
from VSA are first-time so lvent abusers; in 1997 in 37 per help ed decrease the number of p oisonings but they are not
cent of deaths there was no prior known history of abuse. 100 per cent effective and should not give a fals e sense of
Most deaths from VSA occurred in the 14-18 age ran ge . security.82,83 It goes without sayin g that a ll potential poi
Solvent abusers can be male or female, al thou gh there are sons should be prop erly lab ell ed, stored out of reach of
higher numbers of solvent-related de aths in boys, probab ly children and locked away.
due to differences in sniffing behaviour. In a household where drugs are misused it is po ss ible for
Sniffing so lvents may cause intox ication similar to the a child , even a tod dler, to gain access to liquids, su ch as
effects of alcoho l,76 so a so lvent sniffer may become methadone, and co ntrolled substances of abuse, such as
drow sy, confu sed, an d aggressive, may take more risks heroin or cocain e powder, which they inges t with the con
than they would when sober, a nd so on. Accidents, there sequence of serious harm as a direct effect of such inges
fore, are quite comm on and so metimes fatal. 77 Over one tion. Passive smoking of cannabis fr om parents smoking
half of the deaths that have been linked to solvent sniffing close to the child in enclosed restricted environm ents has
appear to resu lt from the direct toxic effects of t he chemi been known to result in inhal at ion of enough smoke to
cals that w ere sniffed , but other de aths result from acci result in intoxication of the child .
dents, choking on vomit or suffocati on. Deat hs are often
sudden, and often a mechanism of death invo lv ing cardiac
arrest app ears to be the cause . Gas fuels continue to be HYPERSENSITIVITY
associated with the majori ty of deaths, accountin g for
about two-thirds of VSA deaths. Sniffing the butan e gas Allergies with the potential for a fatal hypersensitivi ty
in li ghters causes a cardiac arrhythmia. 7s .79 Possession or respo nse have become more freq uent over the years S4 - at
abuse of a vo latile substance is not a c riminal act in itself, least in the public perception. According to the Natio nal
although shopkeepers can be fi ned under the Intoxicating Institute of Allergy, food allergy occurs in 6-8 per cent of
Substa nces (Supply) Act 1985 if they knowingly sell to an children und er the age of 4 years, as well as 4 per cent of
abuser who is und er 18 year old. The investigations of such adults. About 30000 episodes of anaphylaxis and 100-200
deaths require carefu l sampling; to enabl e later sampling of deaths per year occur in the USA. At present, the only ways
the air wi thi n the low er airways, it is essential to retain a to manage food allergies are to avoid the food s that cause
lun g, with its main bronchus tied up securel y.8o.sl reactions and to treat the sympto ms caused by allergic reac
tions. About 20 per cent of th e population worldwide can be
considered to have som e form of atopy; this is more common
ACCIDENTAL POISONING in infants and children und er the age of 3 yea rs. B5 ,S6 Nut
allergy is parti cularly important;S7 the inadvertent consump
The innate inquisitiven ess of children and th eir tendency to tio n of nuts by children who happen to be sensitive to them
sample and taste wha t th ey encoun ter in t heir immediate may cause deaths unless immediate action is taken. In the UK,
environm ent may result in accidental poisoning by: con 5 out of the 20 fatal a naphylactic re actions reco rded each
sumption of mushrooms, benies, fruits and vegetation grow year are due to food .s8 In addition to peanuts, the foll owi ng
ing wild, which are intrinsically poiso no us; and the fo ods can result in life-threaten ing reac tions on re-exposure:
sampling of liquids held in bottles, including alcoholic bev tree nuts, seafood, seeds and cows' milk. 89 - 93 A number of
erages, or other containers of liquids found in the house or these children always cany with th em an emergen cy pack
in ga rages and outho uses. They are attrac ted by the colour, that contains t he equipment required for an adrenalin injec
odour and labels of bottles and contain ers. Poisons range tion, along with instructions as to how it shou ld be used if
from med icines (such as a nalgesics, contraceptive pills, iron they were to develop an acute hypersensitivity respo nse;94
tab lets, vitamins an d antidepressants) to com mon fluids or these are carried to school or other extra-domestic sites. 95
items foun d in the ho use (cleaning fluids, a lco hol, cigarettes, Children with bronchial asthma and wh o are kn ow n to
c ray ons and button batteries) to dangerous garden plants respo nd adv ersely to external allergens, su ch as cats, may
(fox glove and lab urnum seeds). Samplin g by children of also develop potentially life-t hreateni ng episodes if exposed
medication not intended fo r th em is a recurring prob lem, acutely to t he antigen .
particularly if the tablets and capsules in questions are
coloured a nd resem ble sweets. Poiso ning is the fourth most
co mmon ca use of accidental deaths in children. REFERENCES
Children under 5 years of age, as well as ado lescents, are
Macke nzie SG, Pless lB. CH1RPP: Canada 's princ ipal injury
pro ne to poisoning but accidental ingestions are most com
surveillance program .il1jul)' Pre/! 1999; 5: 208- 13.
mon in ch ildren under 5 years old . Abo ut 90 per cent of a ll 2 Stone DHG , Morrison A. Smith GS. Emergen cy Depart ment
poisonings occur at ho me, the most common sites being inju ry surveill ance system : the best use of resources? inju1Y
the ki tche n and the bathroom . Unintentional pOisonings Prel) 199 9; 5: J66 -7.
- --
-
-
References I 343
3 ECOHOST. Child Injuries in Industrialised Countries. Report 26 Mott A, Evans R, Rolfe K et al. Patterns of injuries to children
commissioned by UNICEF. London: ECOHOST, 2000. on public playgrounds. Arch Dis Child 1994: 71 :3 28 -30.
4 Malek M, Chang BH, GalJagher SS, Guyer B. The cost of 27 Kujala UM, Taimela S, Antti-Poika I et al. Acute injurie s in
medical care for injuri es to children. Ann Emerg Med 1991; soccer, ice hockey, volleyball, basketball, judo and karate:
20:997-1005. analysis of national registry data. BM) 1995; 311: 1465-8.
5 Barber MA, Sibert JR. Diagnosing physical child abuse: the 28 Swan P. British Schools, Safety Incidents and The Courts.
way forward. Postgrad Med ) 2000; 76:743-9. www.articlecity.com/artic les/lega l/ a rticl e_3 8 5.sh tml.
6 Moyer LA, Boyle CA, PoJJack DA, Validity of death certificates 29 Lane K. General Defences Relevant to the Tort of Negligence.
for injury-related causes of death. Am) Epidel11iol 1989 ; www.city.londonmet.ac.uk/-shlane/Generaldefences.htm.
130: 1024-3 2.
30 Kendrick D, Marsh P. Babywalkers: prevalence of use and
7
Maclntryre CR, Ackland MJ, Chandra raj EJ. Accuracy of relationship with other safety practices. Inju/y Prev 1998;
injury coding in Victorian hospital morbidity data. Ausr NZ ) 4:295-8.
PubI Health 1997; 2:779-83. 31 Spiegel C, Lindaman F. Children can 't tly: a program to
8 Walton W, Jarvi s SN, Lowe P Jet al. Annual incidence of prevent childhood morbidity and mortality from window falls.
unintentional injury among 54,000 children. Injury Prell Am) Publ Hralrh 1977; 67:1143-7.
1996; 2:16- 20. 32 Christie N. Social, Economic and Environmental Factors in
9 Towner EM, Jarvis SN, Walsh SS, Aynsley-Green A.
Child Pedrstrian Accidents: a Research Revirw. Project Report
Measuring exposure to injury risk in school children aged
116 Et 117. Crowthorn e, Berkshire: Transport Research
11-14. Bi\lIJ 1994; 308:449 - 52.
Laboratory, 1995.
10 Levene S. Coron ers' records of accidental deaths. Arch Dis 33 Lawson SD, Edwards PJ. The involvement of ethnic minoriti es
Child 1991; 66:1239-41. in road accidents: data from three studies of young pedestrian
II Murdock R, Eva J. Home accidents to children under 15 years: casualties. Traffic Eng COl1trol 1991 ; 32: 12-19.
a survey of91O cases. BM) 1974; 3:103-6. 34 Boxall JA. School crossing patrols: how effecti ve are they?
12 Lenaway DL, Ambler AG, Beaudoin DE. The epidemiology of Traffic Eng Control J 938; 29:586. Report CR 192. Crowthorne,
school-related injuries: new perspectives. Am) Prep Med Berkshire: Transport and London: Departme nt of TranspOI1.
1992 ; 8: 193-8. 35 Christie N. Social Economi c and Environmen tal Factors in
JJ Latlamme L, Menckel E. School injuries in an occupational Child Pedestrian Accidents: A Research Review. Project Report
health perspective: w hat do we learn from community based no. 116. Craw thorne, Berkshire: Road Research Laboratory,
epidemiological studies? Injul:V Preu 1997; 3:50-6. 1995.
14 Hammarstrom A, Janlert U. Epidemiology of school injuries in 36 Jorgensen E. Bicycle tracks in urban areas in Denmark.
the northern part of Sweden. scal1 ) Soc IvIed 1994; Evaluation of the effect on safety. In Biecheler M, Lacombe C,
22: 120-6. Muhlrad N (eels) Evaluatiol1 85: International Meeting on the
15 Alwash R, McCarthy M. Accidents in the home among Evaluation of Local Traffic Safety Measures, Proceedings of
children under 5: ethnic differences or social disadvantage? the Paris Conference, 20-23 May 198 5, pp. 755-61.
BM) 1988; 296: 1450-3. 37 Bicycle Helmet Safety Institute, 199 5.
16 AvelY lG, Vaudin IN , Fletcher JL, Watson 1M. Geographical http: //www. bhsi.org/webdocs/hend erson. htm.
and social variation s in mortality due to childhood accidents 38 Cameron M, Heiman L, Neiger D. Evaluation of the Bicycle
in England and Wales 1975-84. Publ Health 1990 ; Helmet Wearing Law in Victoria During its First 12 Months.
104: 17I -82. Report No. 32. Victoria, Australia: Accident Research Centre,
17 Campbell H, O'Dri sco ll S. Th e ep id e miology of leisure Monash Unive rsity, 1992.
accidents in Scotland. Health Bulletin 1995 ; 53:280-3. 39 Vulcan A, Cameron M, Watson W. Mandatory bicycle helmet
18 Hapgood R, Kendrick D, Marsh P. How well do use: expe rience in Victoria, Australia. World) surg 1992;
soc iodemogTaphic characteristics explain variation in childhood 16:389-97.
safety practices? ) Publ Health IvIed 2000; 11 3 :285-9. 40 Finch C, Newstead S, Cameron M, Vulcan A. Head Injury
19 David RM, Pless LB. BMJ bans 'accidents '. Accidents are not Reductions in Victoria Two Years After Introduction of
unpredictable. BM)2001; 332:1321-2. MandatolY Bicycle Helmet Use. Report No. 51. Victoria ,
20 Guyer B, Gallagher S, Chang Bet al. Prevention of childhood Australia: Accident Research Centre, Monash University, 1993.
injuries: evaluation of the Statewide Childhood Injury 41 Cameron M, Vulcan A, Finch C, Newstead S. Mandatory
Prevention Program (SCIPP). Am) Publ Health 1989; bicycle helmet use following a decade of helmet promotion in
79: 1521-7. Victoria, Australia - an evaluation. Accident Anal Prey 1994;
21 Gielen A, Wilson M, McDonald E et al. Randornised trial of 26:325- 37.
enhanced anticipatory guidance for injury pTevention. Arch 42 McDermott F. Bicyclist head injulY prevention by hel mets and
Pediarr Adolesc Med 200 1; 155:42-9. man datory wearing legislation in Victoria, Australia. Ann R C
22 Botting B, Crawley R. Trends and patterns in childhood surg Engl 1995; 77:38-44.
mortality and morbidity. In Botting B (ed,) Tile Health oj Our 43 Puczynski M, Marshall D. Helmets! All the pros wear them.
Children: a reuiew in the mid-/990s. Decennial supple ment Am) Dis Child 1992; 146:1465-7.
OPCS DS No. 11,62-81. London: Office of Population 44 Lee A, Mann N, Takliti R. A hospital - led promotion campaign
Censuses and Surveys, 1995. aimed to increase bicycle helmet-wearing among children
23 King A, Wold B, Tudor Smith C, Harel Y. The Health oj Youth; aged 11-15 living in West Berkshire 1992-98. Injury Prev
a Cross-national sumey. WHO regional Publications, 2000; 6:151-3.
European Series No. 69, WHO, 1966. 45 Britt J, Silver I, Rivara F. Bicycle helm et promotion among
24 Bijur P, Golding J, Haslum M, Kurzon M. Behavioural low income preschool children. Injurv Prey 1998 ; 4:230-3.
predictors of injury in school-age children. Am ) Dis Child 46 Abu larrage J, De Luca A. Abularrage C. Effect of educat ion
1988 ; 142: 1307-12. and legislat ion on bicycle helmet use in a multiracial
25 Waller AI, Marshall sw. Childhood thermal injuries in New population. Arch Pediatr Adolesc Med 1997 ; 151 :4 1- 4.
Zealand resulting in death and hospita lization. Burns 1993 ; 47 Dannenberg A, Gielen A, Beilenson Pet al. Bicycle helmet
19:371-6. la ws and educat ional campaigns: an evaluation of strategies
344 I Accidental injuries in children
to increase c hil dren's helmet use. Am} Publ Healt/r 1993; 73 Town er E, Dowswell T, Mackereth C, Jarvis S. What Works in
83:667-74. Preventing Uninte ntional Injuri es in Chi ldren and Young
48 MacKay M. P layground injuries. Injury Prev 2003; 9: 194 -6. Ado lescents 7 An Updated Systemat ic Review. London:
49 Cha lm ers DJ , Ma rshall SW, Lan g ley JD et al. Heig ht a nd Hea lth Develop ment Agency, 2001.
surfacing as risk factors for injury in fa ll s from playground 74 Sva nstro m L, Haglun d BJA. Evidence-based Safety Promotion
equipment: a case co ntrol study. InjU/)' Prev 1996; 2:98 -1 04. and Injury Prevention - a n Introducti o n. Department of
50 Laforest S, Robaitaille, Lesage D et al. Surfa ce charac teristi cs, Publ ic Hea lth Sciences, Di vision of Soc ial Medici ne,
equipment height, and the occurrence and severity of Karolin ska lnsti tutet, Stockho lm , Sweden, 2000.
playground injuries. Injury Preli 2001; 7:35-40. 7 5 Ano nymous. Solvent abuse: Little progress a fter 20 years
51 Platt SL, Fine JS, Foltin GL. Escalator-related injuries in (Editori al). 8M} 1990: i:J00.
children. Pediatrics 19 97: 100 :E 2. 76 McKeganey N, Norrie J. Association between illegal drugs a nd
52 Lilli s KA, Jaffe DM. Playgrou nd injuries in children. Pediatr weapon carry in g in young people in Scotland; schools'
Emerg Care 1997; 13: 149-53. survey. EM} 2000: 320:982-4.
53 Laforest S, Robita ill e Y, Dorval D et al. Severity of fall injuri es 77 Joseph DE, Parke r S. inl7alants. Drugs oj Abuse. New York:
on sand or grass in playg rounds. } Epidemiol Community United States Drug Enforcement Administration , 2005.
Health 2000; 54:475- 7. 78 First Nations and Inuit Health Committee, Canadian Paediatric
54 Ro sevea re CA, Brown JM, Barclay Mcintosh JM, Chalm ers DJ. Soc iety (CPS). Inhalant abuse. Paediatr Child Hea lth 1998;
An intervent ion to red uce playground equipment hazard s. 3:123- 6.
injury Prev 1999; 5: 124-8 . 79 Shep herd RT. Mechanism of sudden death associated with
55 Sosin D, Su rface-specific fall injury rates on Utah school volatile substance abuse. Hum Toxicol 1989; 8:28 7-91.
playgrou nd s. Am} Publ Health 1993; 83:733-5. 80 Anderson HR, MacNair RS , Ramsey JD. Deaths from abuse of
56 Mott A. Rolfe K, Ja mes R et al. Safety of su rfa ces and substances : a national epidemiological study. EM} 19 85 ;
eq uipment for c hildre n in playgro und s. Lancet 199 7: 290:304-7.
349 : 1874- 6. 81 www.tox la b.co. uk/vo la tile.htm and
57 Sub ve rt JR, Mott A, Rolfe K et a l. Preventing inj uries in www.toxlab.co. uk/dasguide.htm.
public playg round s t hro ug h partnership between health 82 S ibe rt JR , Clarke AJ, Mitc hell MP. Improvements in child
services a nd loca l a uth ority: co mmuni ty interwoven study. res ista nt co ntai ne rs. A rch Dis Child 1985; 60:1155 - 7.
BM} 1999; 318:159 5-8. 83 Rodge rs G. The sa fety effects of child-res istant pack aging fo r
58 Mayr J , Russel 0, Sp itzer P et al. Playground accidents. Acta ora l prescripti on drugs : two decades of experie nce. }AMA
Paediatr 1995; 84:573-6. 1986 ; 275: 166 1-5.
59 Macar1hur C, Hu X, Wesson DE, Parkin Pc. Risk factors for 84 Isolauri E, Huurre A, Salminen S, Impivaa ra O. The all ergy
severe injuries associated with falls from playground epide mic ex tends beyond th e past fe w decades . Ciil1 Exp
equipment. Accident Anal Prev 2000; 32:377-8 2. A llagy 2004; 34:1007-10.
60 Langran MA. Summa ry of the research presented at the 16th 85 Host A, Halken S. A prospective study of cow milk a llergy in
In ternationa l Sy mposium of the ISSS held in at Mount Arai, Danish infants during the first 3 years of life. Clinical course
Japan, Ap ril 2005. www.ski-injury.com. in relation to clinical and immun o log ical type of
61 Barone GW, Rodgers BM. Pediatrics equestrian injuri es : hypersensitivity reaction. Allagy 1990; 45 :5 87 - 96.
a 14-yea r rev iew.} Trauma 1989; 29:245-7. 86 Sampson HA, Scanlon SM. Natural hi story of food
62 Bond GR, Christoph RA, Rodgers BM. Pedi atric eq uestrian hypersensitivity in children with atop ic dermatiti s. } Pediarr
injuri es; assessing the impact of helmet use. Pediatrics 1995; 1989; 115:23-7.
95:487-9. 87 Bock SA, Atkins FM. The natural histolY of peanut alJergy.
63 Holland AJA, Roy V, Goh V et al. Horse-related injuries in } Allng.l' Ciill illll11ullol 1989; 83:900-4.
chil dren. Med} Aust 2001; 1756 ;609-11. 88 Pumphrey RS. Lessons for management of anaphy laxis from a
64 Ghosh A, Di Scala C, Drew C et al. Horse-rel ated injuries in study of fata l rea ctions. Clin Exp AI/{:'rgy 2000; 8: 1144-5 0.
pediatric patients.} Pediatr Surg 2000 ; 35: 1766 - 9 . 89 Pa llares D. Allergy to macadamia nut. A 1111 Allergy Asthma
65 Aronso n H, Tough Sc. Horse-rel ated fatalities in the p rovince 1111111ullol 2000; 85:385-6.
of Alberta 1975-1 980. Am} ForC'l1sic fVlt:'d Pat/lOl 1993; 90 YUll gin ge r .liN, Sweeney KG, Sturn er WQ e t a l. Fata l food
14:28-30. induced a naphy laxis. }AMA 1988; 260: 1450-2.
66 Ch itn av is JP, Gibbins CLMH, Hirigoyen M et al. Accidents 9 1 Sa mpso n HA, Mend e lso n L, Rose n JP. Fatal a nd near-fatal
with horses. What ha s cha nged in 20 yea rs? Inj!lI)' 1996; anaphy lac tic reaction s to food in chil dren a nd ado lescents.
27: 103 - 5. N ElIgl} MC'd 1992; 32 7:330- 4.
67 Byard RW, Gilbert J, Lipset J , James R. Farm - and tractor 92 Bock SA, Mun oz - Fu rl ong A, Sampson HA. Fata liti es due to
related fatalities in South Australia. } PC'diarr Child HC'olrh anaphylactic react io ns to foods. J Allergy Clin Immunol 200 1 ;
1998; 32:139-41. 1: 191-3.
68 Cogb ill TH, Busch HM, Stiers GR. Fa rm accidents in ch ildren. 93 Yunginger JW, Nelson DR , Squi llace DL et aJ. Laboratory
Pediatrics 1986; 76:562-6. investigation of deaths due to anaphylaxis. ) Forensic Sci 1991;
69 Rivara FP. Fatal and non-fata l farm injuri es in the United 36:857-65.
States. Pediatrics 1985; 76:5 67 - 73. 94 Simons FE, Chan ES, Gu X, Simons KJ. Epinephrine for the
70 Cameron D, Bishop C, Sibert JR. Farm accidents in children. out-of-ho sp ital (first aid] treatment of anaphylaxis in infants:
EM} 1992; 305:23-5. is the ampOUl e, syringe, need le met hod practical? ) Allergy Clin
71 Millward LM , Morgan A, Kelly MP. NH S Hea lth Development Immunol 200 1 ; 108: 1040-4.
Agency. Health Developm ent Agency. Prevention and 95 Simons FE, Gu X, Silv er NA, Simo ns KJ. EpiPen Jr versus
Reduction of Accidental Injury in Children and Older People. EpiPen in young children weighing 15 to 30 kg at risk for
Evidence briefing, June 2003. anaphylaxis. ) Allergy Clin Immunol 2002; 109: 171-5.
72 Grayling T, Hallam K, Graham D et al. Streets Ahead: SaJe
and Liveab/{:, Streets Jor' Children. London: In stitute for Publi c
Policy Research, 2002.
- --=- -
I CHAPTER 18 I
DROWNING AND NEAR DROWNING
John Pearn
TRODUCTION
child abuse or non- accidenta l injury, homiCide, euthanasia
a nd culpable neglect. The demonstration a nd interpretation
J:o wning remains one of the most common causes of death of secondalY trauma due to attempted cardiopulmonary
:!l early chil dhood . In developed and affluent countries, after resuscitation after immersion also depends upon meticu
?: child has passed th e neonatal period, immersion incidents lous autopsy technique.
~main one of the three principal causes of death.l,2 In many Of all health professionals, pathologists develop the most
:<a es and commun ities within Europe, the USA, Canada and pragmatic perspective of the serio us threats to children's
.-. ~ tra la sia, drowning is the leading cause of unintentiona l health, welfare and safety,7 Both the paediatric pathologist
C:~th in children und er the age of 5 yea rs.) This age group and the forensic pathologist have inescapa ble ethical roles in
:.So ex hibits the highest drowning mortality rate worldwide, advocacy for the reduction of immersion accidents, a signif
=:.xcept in Africa. 4 Immersi on fatalities rank as the seco nd icant proportion of which a re preventable.
~d ing cause of acc idental death for all older children and
-cenage rs and this is the third leading preventable cause of
:'eath.5 [n Australia, a water-orientated society, drowning THE CAUSES OF CHILDHOOD DROWNING:
- . replaced motor vehicle accidents as the leading cause of A PERSPECTIVE
=.:.: childhood deaths from injury.6 Autopsy di agnos is of the
- owned child is challenging for the pathologist and is co n Childhood drowning ra n ks high as a cause of child death,
-:-onti ng for all who work proactively to redu ce the incidence especially in the pre-school years (Table 18.1). The sites of
uC: im mersion accidents. the drowning incid ent include the family bath tub, private
Every case of drowning and near-d rowning has poten and public toilets, buckets and pails, private, family-owned
forensic and legal implications. These app ly not on ly swimming pools, public swi mmin g pools, fish ponds and
- :he wo rk of pathologists, coroners, police officers an d ornamental pools, building site tre nches and drains, agri
G.\. 'ers in the case of immersion fatalities, but also are cultural dams and tanks, resi dential and neighbourhood
apa bl e as part of the clinical skills required by rescuers, canals, lakes, creeks and riv e rs, boating marinas and the
.'-- la nd ers, a mbulance officers, paramed ics a nd the emer open sears) (Table 18.2). Children who drown in each of
=e cy room team. [n addition to the elucidation of the these sites have their own s ite-specific precedents a nd
;mato micopathological features of drowning, and the esta b forensic concomitants B ,9
!i hme nt of precedents and causes of death, the fore nsic The causes, sites, survi val rates and modus operandi of
i sues of immersion incidents may also include those of immersion deaths all differ when child victims are compared
346 I Drowning and near drowning
Table 18.1 Relative rank order (per cent) of traumatic and injury the most difficult in forensic pathology 12 In the case of child
deaths in childhood (1-14 years) victims this difficulty is compounded when there may be no
pathognomonic signs of drowning and where there may be
Cause of death Percentage
no objective post-mOliem signs of intentional injury, even
when such is undoubtedly the case. Alveolar oedema, alveo
Drowning (accidental) 32 lar haemorrhages and emphysema aquosum are often the
Motor vehicle occupant death s 29 only histological featu res of drowning; even these may not
Pedestrian run downs 17 be demonstrable in the case of an infant or child drowning
House/ca rava n fires 8 when the duration of immersion is measured in minutes only.
Homicide, child abuse 6 The second tool, indispensable to the pathologist, is ref
Fa lis, playg round acciden ts, accidenta I 6 erence to the defined syndromic profiles of child drown
strangulati on ings. Much recent research has built up patterns of quite
Other, e.g. poisoning, electrocution 2 distinct child immersion syndromes. lJ Such form indispen
Total 700 sable templates not only for the pathologist, but also for
the investigating police and presiding coroner (Table 18.4).
Data typical of tropica l and temperate developed nations. Comp il ed from
the Austra lian Insti tute of Health and We lfare 6 and other sources. Childhood immersion fatalities do not form a spectrum of
immersion in cidents; rather they constitute subsets of quite
distinct forensic syndromes that are unrelated to each other
Table 18.2 Relative rank order (per cent) ofsites of drowning of
except by the end-point of drowning. Overall, 95 per cent
children aged 0-5 years
of such cases are accidental (unintentional). In homicide
Site Percentage cases, however, drowning is the method of killing in some
10-20 per cent (see Table 18.4). Reference to such immer
sion syndromes in the pre-autopsy analysis of any individ
Private swimm ing pools 64 ual case is of great help to the pathologist in the choice
Family bath tubs 16 of ancillary investigations (Table 18.3) and may be of the
Creeks 11 greatest help to family doctors, coroners and lawyers in
Dams, building trenches, sewers 5 subsequent proceedings.
Waterholes, fish ponds 4 Many childhood immersion fata lities are depressingly
Data typical of tropical and temperate developed nations. Compiled from repetitive as many of these deaths are preventable. Never
the Brisbane Drowning Stud y l 7.43 and other sources. theless, all of those persons who are involved in the pre
mortuary chain - parents, siblings, neighbours, rescuers,
bystanders, paramedics and police officers - are inevitably
with adu lt subjects. The common patterns of adult immersion enmeshed in an intense personal and tragic incident. What
incidents, involving alcohol, suicide lO and boating accidents, happens to such individual s often depends on the determi
are rarely encountered in the case of childhood immersion nation of diagnostic truth, and such in turn depends on the
victims. The common situation of difficulty with cadaver patholo gist's skill in recreating the events surrounding
id entification often encountered in long-immersed adult vic such immersion deaths. Because of the well-recognized, and
tims is very rare indeed; the identity of childhood victims is indeed historic, difficulty in confirming death by drown
usua lly known and not open to question at the time of ing, there exists no greater challenge to the attending
extraction from the water. Immersion times of childhood vic pathologist.
tims are measured in minutes rather than hours in more than
99 per cent of repo rted series. 11 The challenge of child immer
sion deaths to the forens ic pathologist is that every case is THE DROWNING MEDIUM
individual and specific. Yet, however rare might be the cir
cumstances sunounding an individual traumatic child death, In at least 90 per cent of drowning fatalities, death is due to
diagnosis as to the cause of death and elucidation and recon the inhalation of water into the alveoli, with hypoxaemia
struction of its precedents are of the greatest impoliance. and subsequent brain hypoxia. The water in which children
In approaching a case of immersion death or near-death drown contains a case-specific mixture of inorganic sub
in childhood, the pathologist needs to adopt two appro aches. stances, gases, dissolved organic matter and particulate mat
The first is that of the usual detailed history-taking, followed ter. Particulate matter may include bacteria, fung i, diatoms,
by meticulous autopsy examination of the body with, in zooplankton and phytoplankton. Unlike the situation that
some instances, particular attention to post-mOliem radi is more frequently encountered in the forensic analysis of
ographic survey, blood analyses for drugs, chemical analyses adult immersion victims, such water composition does not
of lung tissue, DNA extraction and storage, post-mo rtem usually have clinical, forensic or legal overtones. Very rarely
photography and, rarely, bone marrow extraction for diatom children drown in vessels t1lled with paint or fertilizer, or in
examination (Table 18.3). The diagnosis of drowning is one of agricultural, industrial or domestic chemicals.
The drowning medium I 347
Diatoms Rarely needed; many different diatom genera and species have siliceous skeleton; femoral bone marrow sampled;
tissue acid-digested or macerated, interpretation difficult, diatoms absorbed from the gastrointestinal tract
pre-mortem; water from the putative drown ing site essential for interpretation; concordance of diatom
genera and species is only 65 per cent 12
Diatom analysis is helpful when: cadaver is badly decomposed; body is found on land but suspicion of prior
drowni ng; bod y may have been put into the wa ter after death 12.79
Table 18.4 Relative rank order (per cent) of the modus of child Fresh Water
homicide in the USA in children under the age of 5 years
syndromes in more than 95 per cent of survivors. Experi injUlies (sometimes sustained dUling skylarking in the sur
mental animal studies have not demonstrated significant rounds of swimming pools) can drown in very shallow water.
effects on the surface tension of lung surfactant when In experimental mammals, as little as 1.0 mL/kg of fresh
chlorinated water inhalation is compared with that of pure water instilled into the trachea causes gross pathophysiolog
fresh water. However, any type of fresh water inhalation ical responses in the lungs. Experimental studies, usin g dogs,
elevates the minima l surface tension of tracheal and lung have shown that drownin g mammals may aspirate more than
surfaces very significantly when compared with the affects 22 mi/kg of water. The imp lication is that in the case of a
that follow sa lt water aspiration. typ ical chi ldhood drowning (with a median age of approxi
mately 2 years in most current series) a \ 3-kg child may asp i
rate as much as 300 mL of water. Such ac ute increase in blood
Sea Water volume is not greater than the capacity of the nonnal healthy
child's heart to compensate for this potential flu id overload.
Sea water contains a mixture of inorganic sa lts. Concentra
tions vary widely but the relative prop0l1ions of the differ
ent types of dissolved salts remain surprisingly constant. The 'Dry' Drowning
total concentration of sea-water salts may change over rela
tively small distances, a phenomenon seen particularly in There exists an historic debate abo ut the proportion of cases
estuaries and even during tidal flow. II Sa linity is defined as of immersion accidents that are due to 'dry' drownings, in
the total mass of disso lved solids (in grams) in one kilogram which laryngea l spasm is tho ug ht to be the primary cause
of water. Such is expressed either as parts per thousand (ppt) of fata l hypoxia. Alt hough laryngeal spasm occurs to some
or in grams per kilogram (g/kg). In instances where children degree in every case of fluid inhalati on, water enters the
drown in in land seas, canals, rivers or delta regions (pa11ic peripheral airways in at least 90 per cent of cases. In child
ularJy during floods), the salinity of the inha led water may hood, autopsy findin gs of 'dry drownings' are rare. Such
be very low, approach ing that of fresh water. evidence is supported also by clinical experi ence in the man
In organic sea water, a typical salt ion profile is: agement of near-drowned survivors, where in every case
• sodi um , 10.5 g/ kg; there is evidence of water inhalation. I n adult series, the fre
• magnesium, 1.3 g/kg; quency of 'd ty drowning' has been va ri ously estimated to
• calcium, 0.4 g/kg; occur in 10 per cent to 15 per cent of cases. However, a
• chloride, 18.9 g/kg; recent review of the original studies from which such con
• sulphate, 2.6 g/kg. clusions have been historically drawn has indicated that
such may be without firm foundation 15 and that card iac
Oceanic sea water thus typically contains 34.5 g/kg of arrest, sustain ed coincidentally while the victim is in the
dissolved salts, of wh ich 29.5 g/kg is sodium chloride; this water may be the true cause of death rather than asphyxia.
is equ iva lent to 2.9 per cent sodium ch loride compared Despite this, there is persuasive experimental evidence to
with 0.87 per cent sodium chloride that is the concentra SUpp0l1 the phenomenon of 'dty drowning'. In experimenta l
tion in human plasma. drowning with diatom monitoring, using deliberate immer
Irrespective of the salt content or osmolality of the inhaled sion deaths of aquatic mammals such as mink, muskrat and
water, during the drownin g ep isode there is a one-way, mas beaver, it has been shown that not all fully submerged ani
sive transuda tion of water across the lung membranes into mals inhale significant amounts of water into their lungs. 16
the pu lmonary vasculature. Childhood drowning deaths are When small amou n ts of water enter the larynx or trachea,
exclusively due to cerebral hypoxia fo llowed by acute brain immediate laryngeal spasm results due to an efferent vagal
death. Any acute changes in osmo la lity, consequent upon salt reflex. An immed iate outpouring of thick mucus occurs,
concentration differences of the inhaled water, are not of probably while the drowning victim is still conscious. Froth
forensic relevance. In survivors, because of rapid compensa develops and in some cases a physical mucous plug forms.
tory homeostatic changes - Starling's Law, the Bainbridge When such spasm relaxes preterminally, a significant
Reflex and renal clearance - such diffe rences are not of prac amount of water is thus prevented from enterin g the trachea
tical relevance in the clinical management of near-drowned and lungs by the foam-froth plug, which acts as a physical
victims in the emergency room or intensive care ward. This is barrier. In such cases, loss of consciousness is caused by
not to say that post-mortem electrolytes and osmolality anoxia or carbon dioxide narcosis. Death fo llows as a result
should not be measured, as the issue continues to be a focus of cerebral anoxia.
of legal debate in subsequent cou11 hearings.
How much fluid is needed to drown a child remains
unknown. Depth is no guide, as some infants drown in less Water Temperature
than 20 cm of water in ornamenta l pools, garden fish ponds,
pails or the family bath tub. Drug-intoxicated teenagers, Almost all child ren who drown do so in water in a temper
patients with epilepsy and children with spinal or head ature range of 10-25°(, Water temperatures ab ove 20'C do
--- - ---- ~
Pathophysiology of drowning I 349
not influence the brain-protectin g diving reflex, but low may alter the physiological setting of the breath-h olding
temperatures augment it. The hypothermic, brain-sparing 'break point'. If the hyp ercarbic drive to breathe is lost,
effects of near-drowning in very cold water are mediated unconsciousness from cerebral hypoxia may occur before
through physical chilling of the body core and through an the hypoxia-induced 'break-point' is reached. This is the
augmented diving reflex. In those situ ations when children pathophysiological mechanism leading to drownin g in
drown in cold water, phys ical chilling of the body core is teenagers or adults who, during snorkelling or free diving,
the result of not only the conduction and convection attempt maximum endurance dives. 11
effects of cold water on the child's skin and through the In all cases when the ' break point' is reached and the
involuntary inhalation of cold water, but also from swal first involuntary inspiration is made, arterial hypoxaemia
lowing relatively large amounts of cold water. All children has already developed. Tissue hypoxia and acidosis have
extracted from the water in immersion incidents have a commenced. Involuntary gasp ing then occurs in a sub
reduced body temperature, including those who respond to merged subject, who may stiU be conscious. Glottal spasm
cardiopulmonary resuscitat ion and who may survive the occurs. Even before consciousness is lost, vomiting with
incident. In boating accidents, in accidents involving falls aspiration of gastric contents may occur. Consci ousness is
tho ugh ice and in cold water accidents involving older lost within 3 minutes of invol untary submersion; and the
children who can swim, immersion hypothermia may pre electroencephalogram (EEG) becomes flat within 4.5 min
cede primary drowning asphyxia. Clouding of conscious utes. Breathing movements w ith fluid aspiration , and often
ness occurs in a drowning child w hen the core body the aspiration of sand , mud, g ra vel or vegetable matter,
temperature falls below 36°C; and consciousness is lost continues after the drowning v ictim has lost consciousness.
,-,,,hen core body temperature falls below 34°C. Respiratory arrest follows. Dysrhythmias may follow and
hypoperfusion with hypoxic blood leads to brain death.
During the march of intra-immersion hypoxic events,
Immersion Asphyxia Occurring Other than intense autonomic, catecholamine-meditated blood redis
in Water tribution occurs. The spleens of drowned victims are rela
tively bloodless, as one manifestation of the reflex
Rarely, children drown in non-water media such as paint, liq constriction of sp lanchnic (including splen ic) vessels.
uid fertilizer or industrial chemica ls. Some children die after The duration of brain hypoxia necessary to cause the
falling into wheat or grain silos, or are smothered in falls of death of an otherwise fit and healthy infant or child obvi
earth or sand while playing in 'cubby houses', excavated ously remains unknown. Extensive research from the Bris
·Wendy' houses, or following the collapse of tunnels exca bane Drowning Study,Il.l7 using a 'bracket method' and
vated into the sides of cliffs at the beach or beside rivers or involving an ana lysis of hundreds of immersion fatalities
creeks. In such cases the asphyxiating medium behaves as if and survivors, has indicated that (I) children who are
it were a fluid. Inhaled paint or sand within the airways, for immersed for 3 minutes or less are likely to survive an
example, may be a dramatic macroscopic feature at autopsy; immersion incid ents and (2) children are unlikely to
the foreign material may be demonstrated histologically respond to card iopulmonary resuscitation if the immersion
(Table 18.3). However, in a ll such cases the mechanisms of time is lon ger than 10 minutes in the most common water
anoxia are secondary to foreign substance inhal ation. The temperatures in which they drown (10-20°C).
mechanisms of oxygen exclusion (first with brain death and
then with rapid somatic death) are identical to those encoun
tered in cases of fatal water immersions. The Diving Reflex
'window of salvage' of perhaps 30 seconds of immersion that are approximately iso-osmolar (e.g. 0.87 per cent
hypoxia. sodium chloride) with plasma, do not denature pulmonary
surfactant but may dilute it or wash it out of the alveolar
sacs. Fresh water, or inhaled fluid that is significantly
Lung Pathology hypo-osmolar, causes acute degradation of surfactant
activity. Whether the lining surfactant is lost or denatured,
As soon as water enters the lungs, a chain of pathophysio the end result is the same and alveoli collapse.
logical events occurs, as follows.
FROTH FORMATION
PERIPHERAL AIRWAY RESISTANCE INCREASES
Exudate outpouring from the laryngeal and tracheal mem
Aspiration of even small amounts of water (1.0 mL/kg) is branes, surfactant washout from the alveolar sacs and dis
followed by pulmonary vasoconstriction, with immediate rupted alveolar membranes result in the formation of froth.
development of pulmonary hypertension due to parasympa The production of froth is characteristic, in variable degree, of
thetic reflexes. The composition of the inhaled water is a high proportion of the lungs and upper airways of drowned
important in this context. Vasoconstriction occurs, particu individuals. In some cases, it may be the only forensic mani
larly following aspiration of fresh water into the mammalian festation that drowning has occurred. In cases of neonaticide,
lung; it occurs to a lesser extent in the case of sea water aspi froth in the air passages is a valuable sign; when interpreted
ration, but does not occur with aspiration of amniotic fluid. together with changes in the lungs it is indicative that a
newly delivered fehls was born alive. The forensic evidence
of post-immersion froth in the upper air passages is preserved
LARYNGEAL SPASM
if freezing has occulTed, and can be demonstrated after
Laryngeal spasm follows, with an immediate outpouring warming even following freezing temperatures as low as
of thick mucus, followed by froth development. The conse -18°c. 21 It is important to appreciate that any prolonged
quent degree of airway obstruction probably varies from freezing of the body of a drowned subject will cause tissue
subject to subject; the larynx relaxes preterminally. distortion owing to ice crystal formation. This compromises
the interpretation of subsequent histological examination.
- - - - - - - - - -
---~----
Forensic immersion syndromes I 351
• cardiac arrest; will fOlm the substance of the final medicolegal report.
• brain death; Whereas the clinician is concerned reactively in combating
• somatic death. the clinical consequences of hypoxia, the pathologist is con
cerned with the recreation of the events that led to the pri
Brain death after immersion accidents follows a primary mary immersion anoxic episode. In more than 95 per cent of
state of altered neuronal metabolism. Research studi es of childhood immersion incidents, unlike other forms of child
changes in the hypoglossal nucleus (in the brainstem) indi hood trauma, the immersion incident is not witnessed. This
cate that compared with other quicker forms of asphyxia fact, combined with historic difficulties of confirming death
(e.g. hanging, strangulation and choking) pre-mortem by drowning and the relative absence of specific forens ic
damage to proteins such as the c-fos gene protein and the signs, places the pathologist in a position of great responsi
72-kDa heatshock protein occur in this phase. 22 bility. For these reasons, there is an imperative to approach
With increased percentages of the population trained in the pathological investigation against the background of all
cardiopu lmonalY resuscitation (CPR) skills and the evolu possible syndromic templates. There are at least 16 such
tion of better ambu lance services and more sophisticated age-site paediatric drowning syndromes, each with specific
intensive care units, recent years have seen a number of types of antecedent, different risks for the potential of
childhood immersion victims maintained on life support unlawful child killing, different approaches to investigation
prior to their delayed death due to drowning. The forensic and, ultimately, different stratagems for prevention. The
pathologist is involved in such cases and coronial autop importance of such syndromic definition is cognate to the
sies are required in almost all jurisdictions. Exa mination generation of a clinical and forensic differential diagnosis
of the brains of these children may demonstrate any or all when each new case is being assessed. Review of the child's
of the signs of asphyxial brain death. Although cerebral entire past medical history is essential as a prelude to inter
oedema may develop in survivors, it is not a feature of pretation of the forensic autopsy results and indeed may
autopsy studies in the drowned victim who dies within modify the extent of technical investigation (Table 18.3)
minutes of immersion. Besides petechial haemorrhages used in port-mortem analysis.
on the brain surface, there may be features of cerebral In many nations the crimes of neonaticide and infanti
oedema, sometimes to the point of hemispheric swelling cide are distinct from those of filicide, child homicide or
with resultant mass effect. Such massive oedema has not unlawful child killing. Such jurisdictions recognize that the
been reported in 'CPR survivors' who died within 24 hours syndromes of neonaticide and infanticide have specific
after rescue. Delayed-death subjects, in whom there exists features not only in terms of aetiology, modus operandi
the legal imperative of coronial autopsy, may also show a and sociocul tural implications, but also in prevention.
wide constellation of lung changes. Such autopsy features There exists also the rare problem of false confessions to
range from relatively normal lungs, perihilar pulmonary the drowning of children .24 In these circumstances the
oedema, generalized pulmonalY oedema, pneumonia, col recreating of events leading to the child's death is crucial to
lapse and consolidation, disrup tion of alveoli, alveolar the legal protection of a parent or other party who may be
haemorrhage, foreign inhaled particles in the airways, and psychiatrically disturbed.
signs of unilateral and bilateral pneumothorax. At post In practice, the forensic examination of childhood
mortem , such 'delayed drowning' lun gs may manifest both immersion victims is straightforward. However, in evelY
macroscopic and microscopic features of the adult respira case judgement is required with respect to the choice of
tory distress syndrome (ARDS). 23 In such cases there are post-mortem ancillary tests and investigations. Most juris
often other features of pneumonia, due either to nosoco dictions operate under conditions of scarce, and sometimes
mial infection or infection with waterborne bacteria or inadequate, resources. This applies particularly to patholo
colonization by single-celled organisms. In the case of gists or other medical examiners who are called upon to
children who have survived with intensive care or ve ntila investigate childhood immersion deaths away from sophis
tor support for hours or days following extraction from the ticated fo rensic centres. 25 A small minority, less than 5 per
water, there may be uncommon (indeed, exotic) micro cent of all childhood immersion accidents, have the poten
organisms found in pneumonia consolidates or lung tial for criminal overtones. However, with the increasing
abscesses. In the case of salt-water near-drownings, when tendency for civi l litigation, the pathologist may become
death has occurred later in hospital, marine Vibrio bacteria enmeshed in highly complex and prolonged civil law suits,
or algae may be demonstrable. in which grieving and aggrieved parents seek compensa
tion or damages from parties who are the site owners of
water hazards in which children have drowned . For this
FORENSIC IMMERSION SYNDROMES reason, there is an increasing tendency for forensic pathol
ogists to undertake tests a nd investigations that will leave
From the perspective of the fore nsic pathologist, the site and no doubt as to the recreation of factual events that led to
circumstances of the immersion incident are of crucial death in the water, or after failed attempts at post-rescue
importance in the reconstruction of events as these details resuscitation.
352 I Drowning and near drowning
Classification of Drowning in Children young single teenager, ignorant of physiology and without
antenata l care, de livers herself of her infant. In the context
There are two paradigms by which child drowning may be of fear, pain and ignorance, a proportion of such mothers
classified. The first of these uses the primary discriminator of may not perceive the risk of immersion death. Having said
intentionali ty, thus separating accidental drownings from the t his, neonaticide is the deliberative taking of the newborn's
syndromes of unlawful child killing. The second approach is life and t he related crimes of the deliberate exposure or
to classify drownings by site, each with its site-specific syn abandonment of t he newborn have been classic and not
dromic boundaries. Most medicolegal approaches to child uncommon crimes known since antiquity. In this context,
hood immersion incidents use both approaches. drowning is simply the modus of such unlawful child
killing, the incidence of which varies, as it is always done
with t he sociocul tural mores and the economic circum
Child-killing Immersion Syndromes stances of the parents of the neonate concerned.
Forensic autopsy of such infants needs to establish
whether spo ntaneous breathing has occurred and whether
NEONATICIDE
the airways are patent, and whether asphyxia by water or
Neonaticide is defined va riously, in different jurisd ictions, as other agents was the cause of death. As a proportion of such
the killing of a n infa nt in the first 24 hours of life or the new ly born infants a re found dead in places such as public
killing of an infant in the first 4 weeks of life. In the forensic to ilets, DNA extraction and storage is essential. The tracing,
literature it is usually taken to mea n the unlawful child finding and identification of mothers is important in these
killing of a subject delivered naturally and of sufficient distressing circumstances, as post-neonaticide identification
developmental maturity to have been capable of independent of the biological father may ha ve important forensic or legal
survival, whose killing was perpetrated within the first 24 implications. Such may occur if there is assa ult on putative
hours of life. Neonaticide by dro wni ng is not rare, but it is, fathers by the girl 's family (particularly by her father or
fOltu nately, decreasing in frequency in many Western soci brothers) and in the rare cases when putative or biological
eties. The syndrome is very specific and has been long recog fathers may be enjoined in the act of neonaticide itself.
nized because of its sad sociofamilial overtones. 26 In general,
neonaticide is in almost always perpetrated by the mother.
INFANTICIDE
In some legal jurisdictions, the crime of neonaticide, with its
connotations of diminished responsibili ty, is only recognized The crime of infanticide is the unlawful killing of a child
as a specific crime if perpetrated by the mother. under 1 year of age by its mother. The designated crime of
The mothers are almost always young (95 per cent of infanticide has evolved as the judicial recognitio n that
cases), often teenagers. They are almost alw ays single. there is a subset of unlawful killings that are the result of
Neonaticide by drowning usually occurs in the context of a diminished responsibility.
concealed pregnancy and in the context of a concealed, The deliberate killing of an infant, by a mother often dis
solitary labour and delivery. The mothers are often mem abled by psychosis, occurs not at birth but in the weeks or
bers of ethnic minority groups, often in religious or lan months following birth. Under these circumstances, drowning
guage isolates living in Western society. Such families is, in one sense, a non-specific modus, as the means of ending
typ ically are t hose with religious or tra ditional cultural the child's life. 28 Most such perpetrators are suffering from
condemnation of premarital sexua l re lations . These tragic post-natal depression, with a smaller proportion afflicted with
incidents are also sometimes encountered in white or black schizophrenia. 29 The intra family dynamics in cases of infanti
families, particularly in those of lower socio econom ic sta cide by immersion 3o differ from those encountered in cases of
tus, in which there is a very strict, male-dominated ethos in the deliberative, repeated , sub -fa tal U'auma, which is a feature
the microsociety in w hich the parturient mot her is trapped. of the crescendo child ab use syndrome that, of course, may
In these circumstances it is not rare for a Single teenage ultimately lead to the death of the child concerned.
girl or yo ung woma n to conceal the pregnancy. At the time Post- nata l depression is a dangerous syndrome for both
of the ensuing delivery a nd during labour, the victim will infant and mo ther. When depressed mothers kill their
seek a priva te bathroom or public toil et in which to deliver infants, the prox imi ty of the family bath tub or of was hing
her child secretly.26 The mother adopts the least uncomfort machines, buckets or pails makes this cause of dea th well
able position for such so li tary delivery. Und er these circum recognized. The syndrome of infanticide-suicide is well rec
stances, the neonate is often born into a toilet bowl or, less ognized by all forensic pat hologists. Sometimes, a mother
commonly, in to a bath. In Western and oriental societies this will kill one, severa l or a ll of her children before taking her
usually occurs in a private or public toilet cubicleP own life. 3 1.32 Infanticide by drow ning may reveal, at forensic
There is often doubt about the degree of intent of the autopsy, skin features or lesions co nsequent upo n the force
deliberate taking of the baby's life, allow ing for the dimin exerted by the perpetrator to drown the struggling child.
ished responsibility of a dist ressed, usually lonely and Some parents have attempted to drug their children before
always ignorant primigravida. In many such instances, the immersing them as the fi nal act of killing.
Forensic immersion syndromes I 353
The forensic pathologist has an important preventative component of the chemical and toxicological approach
role in infanticide, not of course in realized cases but as that is so important in these accidents.
an advocate for increased surveillance and help to those
mothers afflicted by psychosis (either by endogenous bipo
BATH TUB DROWNINGS
lar disorder or by schizophrenia) when the risk of infanti
cide may be hi gh. Such risk remains high in subsequent Immersion incidents in the family bath tub and in domes
pregnancies, and the syndrome of sequential infanticide, tic spa pools have special implications for the forensic
sometimes by drowning, is well recognized. pathologist. From both the clinical and the forensic point
of view there are eight defined syndromes of bath tub
drowning and near-drowning (Table 18.5). Of these specific
FILICIDE
and definable bath tub or bath spa syndromes, accidental
Filicide is that crime in which the offender is a biological, immersion is the most common in childhood. In adult
adoptive or de facto parent. The method of killing is culture series, suicides figure promin ently.39
specific. In European, Asian, Canadian and Australasian A special challenge to both general and forensic patholo
societies the cause of such deaths are head injury, drown gists, and also to clinicians, is the generation of a differen
ing or suffocation. 33 In the United States, homicidal tial diagnosis in respect of causation to encompass the key
asphyxia is less common in some regions, where gunshot prese nting feature - the fact that the bath tub is the site of
murder is more frequently employed. 34 the immersi on incident. If the clinician or path ologist does
Mothers (60 per cent) kill their children more often than not consider the various syndromes that constitute such a
do fathers. Eighty per cent of such victims are between 1 and forensic differential diagnosis (Table 18.5) then it is certain
5 years of age, with a median age of between 2 and 3 years. 35 that lawyers in subsequent coronia I, civil action or criminal
Particular 'at risk' times for such immersion killings are in the courts will do so. The fact that perhaps 10 per cent of child
early hours of the evening, particularly during weekends. 36 hood bath tub drownings mask the tragic fin al result of
crescen do child abuse necessitates full post-moltem X-ray
Site-specific Immersion Syndromes skeletal survey in all such cases.
Al most all bath tub immersions involve fresh (tap) water,
chlorinated to the standard 1 palt per million (ppm). Diatom
BUCKET OR PAll
analysis plays no part in the interpretation of forensic bath
A particularly difficult forensic immersion scenario is that tub analyses. By contrast, soap products are inhaled with
which involves a child who has drowned in a bucket or water in many cases of accidental bath tub immersions
pai1. 37 ,38 The majority of such incidents are undoubtedly involving infants and toddlers, unlike the situation in adult
accidental, but there remains a subset, of unknown propor suicides 39 and some homicides 4o for which the bath tub is
tion, in which attempted or realized infanticide or child the site of death.
homicide has occurred. Bath salts are used in many children's baths, especially
The age range of such victims is 9-20 months. Toddlers in the USA and Japan 14 and in many affluent families of all
are at risk, but infants who are able to pull to stand , but nations. Bath salts contain fluorescein, which is highly sensi
cannot yet wa lk , are also at risk. The infant or toddler may tive to detection by high-performance liquid chromatography.
become wedged in the bucket. No complete unselected Confirmation of fluorescein in lung tissue (and other tissues),
(fatalities and non-fatalities) series for this type of accident in parallel with its demonstration in any residual bath tub
has yet been published. A review of all published papers water samples, may be helpful and specific. Its demonstration
indicates that the mortality rate approaches 60 per cent and parallels the selected and rare use of diatom marrow demon
that the risk of post-accident neurological sequelae is high stration following occasional fatal immersions in open water.
amongst survivors. A related and rare type of immersion
involves those toddl ers who climb up beside and fall into
Accidental Child Drowning in Bath Tubs
washing machines or washing tubs.
Buckets and pails are usually ofJ-S gallons (40-70 L) in Overall, 80-90 per cent of childhood bath tub drownings
capacity. The contents of the buckets are almost always are accidental. The usual scenario of such accidents is the
water, soiled nappies, dirty mop water, bleach, detergent, vely human and universal situation in which a mother, tired
soap or antiseptic. 38 These fluids have the potential to at the end of the day, is attempting to coordinate the feed
cause intense bronchospasm and life-threatening laryngeal ing and bathing of a large and vigorous family. As part of
spasm , irrespective of the immersion time. In survivors of this, an infant who is in the bath tub becomes overlooked
such immersion incidents, there is always severe pneumo for several minutes before his or her well-being is checked.
nia and often systemic complications that require the most The syndrome of accidental infant bath tub immersion
sophisticated management in the inten sive care ward over is quite specific. Such fatalities and near-fatalities occur
subsequent days. In the case of fatalities, analysis of bucket only, or virtually only, in working class, labouring and
or tub contents and of lung tissue forms an important poorer families. 41
354 I Drowning and near drowning
Table 18.5 Site-syndromes of bath tub and spa pool drownings: forensic and clinical implications
Accident 20 per 100000 infants aged 0-" Poorer and work ing class families
months, annually Large family size
8 per 100000 infants aged 12-23 One-th ird are single-paren t families 50
months, annua lly High successful resuscitation rates (60 per cent)
Child unsupervised due to sudden unexpected break in routine
Parenta l dichotomy of care, each believing the other is caring
for the infant
'Vu lnerable periods' such as acute ill ness or marital discord
Child abuse 10-15 per cent of bath tub Families often known to ch ild protection agencies
immersions estimated to be Step-parents or de facto relationships inc rease risk
attempted or realized homicide Often other types abuse uncovered
Some perpetrators 'draw back' after the child loses consciousness
Sometimes precipitated by acute stress in a step-parent or de facto
partner who is left alone with a toddler at a time of acute
interpe rsonal discord
Child homicide Rare - 12 per cent of all child Pa rent psychotically dep ressed or with low 10
homicides 10 Child homicide-su icide syndromes fall in this group
High rate of realized death
Only a small percen tage survive
Bath tub epilepsy Rare Older children and youn g teenagers
Victims may have uncontrolled epilepsy
Children with a history of epilepsy who lock themselves in the
bath room for ablutions are pa rticula rly at ri sk
Euthanasia Rare A proportion of physically disabled toddle rs and yo ung child ren
Often not concealed from police, altruistic intent
Bath tub drowning with Very rare; sudden cardiac death May be a history of cardiac abnorma lities; especially those of bundle
cardiac pathology (all causes) in apparently normal of His and right bundle branch block or long 0-Tsyndrome 56
children has a frequency of Although very rare, commonly raised as a defence against a charge of
1.2 per 100000 patient-years homicidal drowning or in civil claims for compensation after drowning
These accidents affl ict infants and todd lers in a very someone calls unexpectedly at the door. The mother lea ves
defined, age-specific window of 8-18 months; the modal two or three children, including an infant, in the family bath
age of such accidents is 9-11 months.42 The mean number tu b,44,45 leaving the youngest in the care of an older, but st ill
of siblings of victims of this type of immersion is greater pre-school, child - the older children may hop out of the
tha n that of the population-matched average. The victims bath when the mother is gone, leavin g the infant alone.
are usually the you ngest or second-youngest child in the Another factor in some infant bath tub drownin gs is the
family - often the second-youngest, when there is a new use of bath seats, wh ich confer inappropriate assurance of
baby claiming a significant part of the mother's attention. safety,45.46 despite clear instruction from the manufacturers
This peculiar vulnerability to accidents, to which children that the baby should not be left unsupervised.
in higher birth ranks within the family are especially prone, Unlike other childhood drowning accidents that all
is known to app ly to other childhood accidents as well, show a preponderance of males, accidental bath tub
especially accidental poisoning.43 immersio ns show an equal sex ratio. Of those in fan ts and
More than one-half of bath tub immersion in cidents toddlers found unconscious in the family bath tub, some
occur during a specific 'vulnerable period' when the family 60 per cent respond to CPR, with residual intact intellect
routine is suddenl y or unexpectedly broken, such as that and without neurological sequelae. This reflects the rela
which occurs during acute sickness afflicting either the tively short immersion times invol ved in such incid ents.
parents or children or in the context of martial strife.41 A
typical scenario is that which involves a stressed mother,
Bath Tub Drownings and Child Abuse
who is tired at the end of the day, attempting to cope,
unaided, with the control, bathing and feeding of several All reviews of consecutive, unselected series of ch ildhood
high-spirited or fractious, but always vigorous, yo ung bath tub immersions with detail ed follow-up have revealed
children. The telephone rings, or an appliance breaks, or th at some 10-25 per cent of such cases are in fact the result
- -. -
Forensic immersion syndromes I 355
of non-accidental injury35,42,47 Forensic studies in Ger violence or strangulation, but the body is sometimes dis
many have shown that in 5-10 per cent of all child homi posed of in a dam, creek or watercourse.
cides in the under-5 age group, the modus operandi is Infant homicide by bath tub drowning approaches the 'per
drowning, usually in the bath tub. 48 fect crime' in that forensic skills usually cannot distinguish
The immersion assault is usually one such in a non-accidental submersion from accident. Two series of bath
crescendo series of acts of child abuse perpetrated on the tub immersions have revealed that child homicide is a subset
infant or toddler. Step-parents and de facto partners are of the bath tub drowning syndrome - known only because of
almost exclusively the perpetrators. Some 'lose their nerve' later, unsolicited confessions by the perpetrators. 24 ,50
during the assault and draw back and may summon help
for resuscitation. A clue to the specific syndrome of child Childhood Bath Tub Immersion in Epilepsy
abuse by bath tub immersion may be that the child is older
than the modal age for true accidental drowning in this A small proportion of children and teenagers drown, or
site (9-11 months); or ou tside the typical 8-18 months age almost drown, in the bath tub as a result of an epileptic
specific window of vulnerability to bath tub immersion seizure sl Enquiry about a past his tolY of epilepsy is impor
homicide. The median age of normal infants to be able to tant in both the clinical and forensic workups of all immer
pull to stand is 9 months and the media n age for walking sion accidents. There exist several specific issues relating to
in normal full-term infants is 13 months. Usually, children this syndrome.
older than these ages ca n support their heads out of water It is very rare for children to drown in the sea or swim
in a bath tub, even if abandoned or left unattended for ming pools as the result of in-water seizures. 52 This applies
short periods of time. There should be a high index of even to children with uncontrolled epilepsy. The risks are
forensic or coronial suspicion in all bath tub immersions significantly greater if there is a sub-therapeutic serum
involving toddlers older than 11 months, particularly if concentration of anticonvulsant dnlg present. The inci
there is no history of developmental delay or epilepsy. The dence of post-seizure bath tub fatality is less than 0.1 per
median depth of water in bath tub immersion incidents is cent of all childhood drownings.
approximately 20 cm. There is a higher specific risk to epileptic children, espe
Although deliberate hot water scalding by immersion in cially teenagers with epilepsy, to fatal and near-fatal drown
the bath tub is not uncommon in reported series of child ings in the family bath tub. All parents (and children) in
abuse cases, the pathology of bath tub immersion incidents families in whom uncontrolled epilepsy is present are
is primarily that of hypoxia and the two injuries are not warned not to take private plunge baths in the family bath
reported coincidentally. It may be that the perpetrators do tub, but rather to shower standing up. Several cases of flan
not wish to scald their own hands. nels or 'washers' occluding the drainage plug, even during
such 'stand-up' showering ablutions, have been repOlied in
forensic sel1es of fatal bathroom immersion acciden ts related
Bath Tub Drowning and Child Homicide
to epilepsy.
There exists the separate and distinct syndrome of child Teenagers, being what they are, are often insistent on
homicide, as a 'one-off' event, by bath tub drowning. As a absolute privacy in the bathroom and insist on locking the
single act, not in the context of ongoing child abuse, this bathroom door before taking a plunge bath or shower. If a
assault is relatively uncommon. Syndromically, it almost seizure occurs, and the unconscious victim slides beneath the
always involves a parent afflicted by psychiatric illness or water, the chance of a successful resuscitation is small. Such
by low intellectual ability. Mothers are usually the perpetra cases are represented in all forensic series of childhood
tors, have a psychiatric history and are living in disadvan immersions. It is believed that the (misguided) practice of
taged socioeconomic environments. Mothers who kill their placing feverish infants into a hot bath - raising a high risk
children prefer to strangle or drown their victims. Psychol of febrile convulsion - may be the cause of immersion in a
ogists believe that in a large proportion of such cases the small proportion of cases of childhood bath tub fatalities.
psychodynamics are those of surrogate suicide. 49 It is some Water impinging on the face, involving the sensory dis
times the modus operandi of the familiar and tragic 'child tribution of the trigeminal nerve, is the afferent trigger
homicide-parent suicide' doublet. In this latter incident, for the diving reflex. 18 This leads to a complex series of
tragically familiar to all forensic pathologists, one parent is dynamic cardiovascular changes, including' increased
almost always psychotically depressed, or in unremitting vagal tone. There is some evidence that this may be one
despair, and kills his or her child or several children and cause of 'bath epilepsy', initially described in the Indian
then immediately takes their own life. Survival rates for medical literature but known to occur in all races.
children involved in such incidents are low; less than 10 per
cent sUlvive this tragic scenario. Child homicide by drown
Bath Tub Drowning and Euthanasia
ing, in the context of sexual assault followed by homicide,
always perpetrated by a stranger or non-biological relative, All series of unlawful child killings include cases of
is rare. The mode of killing is almost always by physical euthanasia. 53 They are usually perpetrated by a parent but
356 I Drowning and near drowning
rarely by a grandparent or other close family member. One individual temporarily col lapses. However, if the individual
method of euthanas ia is bath tub immersion. In the major is under water, secondary vagal stimulation from laryngeal
ity of such cases the chi ld has a congenital disability, such irritation may compound the hypoxia that is already present.
as a neural tube defect, mental retardation , cerebral palsy In this co ntext, it is important to note that any drowning
or a chromosome abnorma li ty. In occasional cases, the par episode itself may provide an outpouring of catecholam ines
ent is psychotic, wrong ly believing that a normal child is (the 'sympathetic storm'), whi ch, w ith hypox ia , may result in
disabled or suffering when such in reality is not the case. 49 microscopic changes to the myocardium , including myocyte
In instances of euthanasia by drowning, the motive is contraction banding and foc al myocyte hypereosinophilia s 8
always (albeit tragically) an altruistic one. The parent believes
that the child is suffering and t hat it would be kinder if the
SWIMMING POOL DROWNINGS
child was put out of his or her pain or misery, or that they
could be helped from a t ragic black or wicked world. There In most temperate and tropical countries of the developed
may be no attempt to conceal the crime. Some perpetrators world, the principal site of drowning in the childh ood years
proactively contact the police to report the incident. is the backyard swimming pool. Salt-water imm ersions
involving children are now less common than fresh-water
immersions, even in littoral regions, for this reason. In the
Bath Tub Drowning and Organic Pathology
USA alone there are now over 12 million pla st ic wading
A final bath tub drowning syndrome is one in which pre pools and over 5 million surface swimming pools, of which
mortem pathology of the heart, recognized or unrecognized, an estimated 2 million are of the more dangerous, in-ground
is the cause of loss of consciousness during bathing. This variety. Proportionate rates may be even higher in other
syndrome is rare, but is included both for completeness and countries, such as Australia and New Zea la nd. In many
because the question commonly arises in the legal defence countri es, pool drownings constitute 70 per cent or more of
mounted by those charged w ith homicide by drowning. Sud all consecuti ve unselected series of immersion deaths in
den cardiac deaths in childhood and during adolescence ch ildhood. In-ground pools cause 80 per cent of swimmi ng
are we ll known to forensic pathologists. 54 Myocarditis was pool fatalities. Motel , hotel , caravan and trailer park pools
demonstrated in 5 out of 22 cases of accidental drowning are a particular hazard in all reported series. 59 Even in
and underlines the need for full investigation of drowning countries such as Finland, as many near-drownings take
deaths wherever they OCCUr. 55 Importantly, the demonstra place in swimming pools as occur in lakes, and twice as
tion of pre-existing cardiac abnormalities in a drowned child many in swimming pools as in the sea.
may protect any innocent person on whom suspicion of cu l The current status of swimming pool drownings has been
pable neglect has faUen. This applies not only to situations a feature of forensic series only since the early 1970s. 47 At
where ch ild homicide is being considered in the differential t hat time, engineering technology and social affluence made
diagnosis, but also to situations such as deaths in public private in-ground pools widely accessible. In many commu
swimming pools where lifeguards may be subject to cliticism nities in many nations, one in five homes now has an in- or
of neglect, sometimes of culpable neglect, in subsequent above-gro und swimming pool filled with water at least
legal proceedings. throughou t the warmer months of t he year. Even in temper
Congenita l or acquired cardiac abnormalities are also ate climates, where winters may be free zing, pool-house
the antecedent to drowning in some cases of older victims ratios may be as high as 1: 10. In the State of California, some
who drown in bathtubs,56 swimming pools or open water. 100 toddler dro wnin g deaths continue to occur annually.
Anomalies of the bundle of His, ab normalities of the right The age spectrum of such victims is between 12 and 40
cardiac bundle branch and congenital abnormalities of the months, with a modal peak between 18 and 24 months. The
coronary arteries have all been reported in cases of teenage social class 'risk curve' is U-shaped, with deaths occurring
drowning, including bath tub drowning, in which toxicolog disproportionately w ithin richer families who, most com
ical tests for alcohol and illegal drugs have proved negative. monly, have in-ground pools, and in poorer families where
In rare cases, genetic cardiomyopathies may be present. 57 above-ground pools are often bought as an impulse purchase,
The mechanism of such sudden and un expected death dur installed and then poorly maintained. Above-ground pools
ing swimming or bath tub bathing is conjectural. One pos are not as dangerous as in-ground pools. However, some
sibility is that the diving reflex (which follows facial children in the 18- to 36-m onth age range will climb a pool
imm ersi on) induces cardiac dysrhythmia with heightened ladder, or will place a box or other object beside an elevated
sens itivity, i.e. a priori abnormal hearts. It h as also been pool to gain access. Overall, 70 per cent of toddlers who
suggested that cardiovascular damage may follow reduced drown do so in their own pools. Other 'at lisk' pools are those
venous return due to venous dilatation in hot water immer of neighbours, motel, caravan or t rail er park pools and the
sions; if this is superimposed on an 'at risk' heart then car pools at homes of relatives whom children visit.
diac output may be insufficient to sust ain consciousness. Toddlers and yo un g children in this age-vulnerable
Under normal circumstances, the cardiac output wi ll be window do not fear water. Many simply crawl into it, or
sufficient to return brain function to co nsciousness if an step into such pools if no barrier is present. The Brisbane
Forensic immersion syndromes I 357
Drowning Study 47 showed that often such vIctIms were parents. They are no substitutes for parental vigilance and
attracted to the water by a floating toy, mbber ring or ball , close supervision when children are in the water. 61 •62 A small
or by an object lying on the bottom of the pool. The follow proportion of such cases involve children who have been
ing factors lead to higher survival rates following pool-side skylarking in the pool surrounds, fall and hit their heads or
CPR: clothing (with its buoyancy), the diving reflex (pre become winded and cannot extricate themselves when they
served in infants and toddlers) and children found floating become submerged. Another subgroup is probably second
(with residual aeration) as opposed to those discovered on alY to cardiac abnormalities, often congenital in nature but
the bottom. hitherto unrecognized; this group includes those with con
Toddlers virtually never drown in fenced pools with genital coronalY artery or valvular abnormalities or those
safety standard-approved self-latching gates with high, with hitherto unrecognized dysrhythmias, with mechanisms
hidden locks. Intense advocacy, often supported by pathol similar to those described in the case of bathroom immer
ogists, to introduce policed, safety legislation to protect sions. Some rare cases are consequent upon cervical spine
toddlers from such water hazards has been unsuccessful injuries following diving accidents. A very small sub-group
in many communities. Most toddlers who drown in such are consequent upon epileptic seizures in the water. In this
pools do so within 30 m of their own home. Immersion latter context, however, it must be said that swimming pool
times are almost always under 20 minutes. Successful drownings due to proven epileptic seizures are extremely
resuscitation rates for consecutive, unselected series of all rare. This is not to say that children do not have epileptic
such swimming pool immersion accidents approach 60 per seizures in the water but, in practice, most are recognized
cent and rise to 70 per cent if a trained first aider happens, and the victim is extracted without significant hypoxia and
coinCidentally, to be involved in the attempted resuscita without any subsequent post-hypoxic neurological defects.53
tion. Survivors do well, and some 97 per cent of children Another subset comprises those young teenagers who have
who survive this near-drowning syndrome function nor taken alcohol or drugs, and who lose coordination in the
mally. Some 30 per cent of such survivors have wide sub water, or sometimes who are illegally or clandestinely swim
scale disparities on formal psychometric testing but do not ming in pools, often as members of group 'larks', sometimes
manifest clinical neurological signs. involving 'skinny dipping' (nude swimming) at night.
Autopsy findings of such children are typically those of a Another subset consists of children performing the danger
completely fit and well toddler who is alive at one moment ously underrated practice of hyperventilating before diving
and dead 10 minutes later. In such cases there may be mini and underwater swimming in such public pools.
mal anatomical changes in the lungs. The body often shows Because of the extended differential diagnosis that is
post-mortem signs of CPR trauma, including fractured ribs, inescapable in such cases, often dissected in meticulous
contused pericardium or injection needle marks from the detail in subsequent court proceedings, it is absolutely
frantic use of cardiostimulant drugs by ambulance officers, essential that the widest array of appropriate post-mortem
paramedics, firemen (in the USA) or physicians. investigations be undertaken. This includes radiographic
studies (especially of the skull, chest and spine), alcohol and
drug assays and meticulous forensic autopsy techniques in
PUBLIC SWIMMING POOL DROWNINGS
the examination of heart, lungs and brain (Table IB.3). In
A small proportion of children, almost always in the age practice, it is uncommon for such cases to result in positive
range of 3-1B years, drown in public access swimming findings in any of these investigations, but the careers, rep
pools or aquatic parks. One US study draws attention to an utation and professional security of many individuals who
excess of deaths in this situation to black, adolescent are involved in the rescue, the attempted resuscitation and
males 60 Such individuals are typically found on the bot the management chain may well be dependent on the
tom of the pool by another casual swimmer. Very com pathologist's findings and defended opinion. Perhaps most
monly, such drownings are followed by major medicol.egal, importantly of all, 'closure' of the grieving process by par
civil, insurance and regulatory consequences. A high pro ents and loved ones is facilitated by the forensic patholo
portion of such drownings becomes the focus of prolonged gist's confident recreation of the events that have led to
court cases, often involving claims for large compensatory post-immersion somatic death.
sums of money; commercial pool owners, lifeguards, para
medics, emergency room physicians and forensic patholo
RIVERS, CREEKS AND LAKES
gists may all become enmeshed in complex discussions
about pathophysiology and the differential diagnosis of the Children occasionally drown in rivers and creeks and in
causes of drowning in general. lakes. 63 Those who do are almost always (90 per cent) boys,
The majority of such cases are simply due to the fact that and the modal age falls in the age window of between B
individuals, usually poor swimmers, cannot stay afloat and and 12 years. There is often an element of disobedience
so they inhale and the 'drowning chain' ensues. Swimming involved in such cases, when children are swimming far
lessons and, perhaps, optimistic evaluation of swimming from supervision. The typical scenario is when a boy or
abjJj!y, may engender a faJse sense of security amongst maJe teenage! experiences diff]cuJties in the water and his
358 I _ Drowning and near drowning
friends find that they are unable to rescue him and run secondary drowning consequent upon pain, panic and the
for help. Because of the distances involved, resuscitation ensuing incoordination even amongst proficient swimmers.
attempts are unsuccessful in more than 90 per cent of The victims of stonefish envenomation, which are com
cases. Estim ated immersion times are relatively lon g and monly encountered throughout the Indo-Pacific littoral in
often exceed an hour or more before the body is extracted both tropical and temperate waters, often become mani aca l
from the water. The forensic autopsy often reveals gravel, with pain . If this occurs during snorkelling, scuba-diving
sand or aquatic flora in the air passages. or reef walking, drowning is an ever-present threat. A sea rch
for, and identification of, barbs or other foreign material
from the venom apparatus is important in such cases. Severe
DRAINS, TRENCHES AND SEWERS
envenomation by cubomedusans can cause death by direct
Children who drown in drains, trenches and sewers are envenomation while the victim is still in the water, or from
almost always boys, and sometimes more than one victi m is a combination of envenomation and drowning. Box jelly
drowned in the same episode. The victims are often playing fishes (sometimes called 'sea wasps') are amongst the world's
or swimming in forbidd en areas. Often, extraction from the most venomous creatures and many of the recorded victims
water is difficult and bodies may be wedged in outlet pipes of such deaths, a disproportionate number of whom are chil
or municipa l watercourses. A series of such drownings dren, occur before the victim can be extricated from the
always include that subset of children who are playing in water. In the forensic examination of such victims, a strip of
fast-movi ng and dangerous waterways during floods. Under adhesive tape applied to areas of the skin thought to be
such circumstances the novelty of the situation may mask affec ted and then examined under the microscope may
the threat of immersion and the bodies of the victims may reveal not only the characteristic nematocysts, but also will
never be recovered. Other are found wedged in or entrapped enable both genus and species identification to be made.
in pipes or outlet drains or under submerged trees. The foren sic autopsy of salt-water drowned victims usu
ally shows no specific distinguishing features if the body is
retrieved w ithin several hours after the drowning episode.
SEA DROWNING
As in adults, the bodies of long-immersed child victims
Many nations have hi gh shoreline-area ratios, with a pop may show secondary changes due to crustacean or piscine
ulation who are sea-oriented from infancy. In such regions, post-mortem damage. In such cases forensic identification
children grow up in an environment in which respect for may be difficult and dental X -rays, post-mortem finger
the ethos of both water safety and water danger is a natu printing or DNA extraction and analysis are essential.
ral part of culture and folklore. Under these circumstances,
sea drownings involving toddlers are relatively uncom
AFTERMATH
mon. 64 Older children are particularly at risk.
Child victims of sea drownings are found disproportion Only a minority of immersion incidents result in fatali
ately amo ng immigrants, tourists or other ethnic sub ties. 67 ,58 An understanding of the pathophysiology of the
groupS.17,65 The children and their parents may not be aware immersion sequence leading to somatic death, taken in con
of the threat of the sea or surf. Occasionally, young teenagers junction with the documentation of post-mortem changes,
drow n during sailing excursions but, in general , boating and will mean that both more efficient prevention and better car
the use of surfboards, in current practice, are lo w drowning diopulmonary resuscitation will be possible in the future.
threats to children. Childhood drownings in the sea occur The documentation and subsequent compilation of child
almost always duri ng daylight hours and, as a manifestation hood immersion statistics is very important from the point
of the recreational use of the sea, most often occur durin g of view of future prevention. In this context, case finding of
weekends or holidays. Childhood salt-water immersion rates immersion fatalities undertaken for epidemiological research
are unaffected by tidal state. Most sea and surfing beaches is compromised because of the many different causes and
are separated by road, dune strip and a beach from residen syndromes of drowning in childhood. As many as 1 in 10
tial houses. The result of this is that it is very unusual indeed cases of childhood immersions may not be retrieved for
for toddlers or preschool children to drown in the open sea; analysis. For example, some are coded as motor vehicle
the modal age for this type of sea drownings is 8 years. The injuries in those cases when vehicles have crashed into the
phenomenon of 'secondary drowning' is most often identi water.59 Multiple-cause coding is essential if a community's
fied after salt-w ater drownings. In this latter syndrome an or nation's statistics are to be fully exploited in the context
indiv idual may be extracted, pulseless, from the water and of injury prevention. 70
may respond well to cardiopulmonary resuscitation, only to The pathologist's professional and ethical responsibility
manifest life-threatening deterioration hours later due to is to the dead child and to the determination of the circum
washout or denaturation changes in pulmon ary surfactant. stances of that death. The evid ential value of such autopsy
Children are sometimes envenomed by jellyfish or poi findings, however difficult may be their interpretation in the
sonous fishes while swimming.66 Under such circumstances, current stage of knowledge, is of the greatest impOltance.
as in the case of adults, there is a very serious risk of Ultimately, the integrity of not only the professions of
-- ~
-
References I 359
medicine and of law depends upon such a fo rensic service, 10 Byard RW, Houldsworth G, J a mes RA, Gilbert JD .
but also that of society. The deliberative killing or preve nta Cha racte ri stic feat ures of sui ci dal drownings : a 20-y ear st udy.
A m J Forensic Med Pathol 2001; 22: 134-8.
ble death of a child defines, in part, the society in which such
11 Pearn JH. Drowning. [n Pearn J, Leditschke JF, Marsha ll V
tragedies occur. From the professional point of view, the et al. Th e Science of Firs t Aid. Theore tical and Experimental
pathologist also has both a professional and an ethical Basis of Modem First Aid Practice. Canberra: St Joh n
responsibility to any living siblin gs of children who have Amb ulan ce Australia , 1996, pp. 138-47.
died from immersion, whether accid ental or no t. 12 Ludes B. Coste M, NOlih N et al. Diatom analysis in victim's
tissues as an ind ication of the site of drowning. Int J Legal
Much remains to be ach ieved in fos terin g closer coll ab
Med 1999 ; 11 2 :163-6.
oration between clinici a ns ca rin g for the survivors of near 13 Pearn JH. The manageme nt of near drowning. In Aochi 0,
drowning in cidents and the path ologists who investigate Ama ha A, Takesh ita H red s) Intensive Critica l Care Medicine.
fatal imm ersions. 7l •72 Paediatric a nd forensic pathologists Amsterdam: Excerpta Med ica, 1990, pp. 139 -46.
have an important role to play in preventative medicine 14 Muka ida M, Kimura H, Takada Y. Detect ion of bathsa lts in the
lungs o f a baby drowned in a bathtub: a case report. Forensic
a nd in promoting advocacy for public health stratagems
Sci Int 1998; 93:5 -11.
that relate, in particular, to home and water safety in com 15 Modell JH, Bellefleur M, Dav is JH. Drownin g withou t
munities in which drownings are common. asp iration: is this an appropriate diagn osis? J Forens ic Sci
A child's death from drowning is, in one sense, from the 1999; 44:1119- 23 .
family's po int of v iew, the beginning of a new era. The role 16 Gi lbert FF, Gofton N. Te rm ina l di ves in min k, muskrat and
beaver. Physiol Beh av 1982; 20:8 35-40.
of the pathologist is a crucial one in helpin g parents in
17 Pea rn JH. Dro wning. In Di cke rmann JD, Lucey JF red s) The
t he immediate aftermath of death. To be able to recreate Critica lly III Child, 3rd edn. Philad el phia, PA: WB Saunders,
exactly what has happened often brings long-term resolu 198 5, pp. 129 -5 6.
tion to the aftermath ofa child's drow ning, which is always 18 Gooden BA. Mechanism of the human diving response.
a t ragedy. Integ rative Physio l Beha l! Sci 1994; 29: 6-1 6.
19 Torre C, Varetto L, Tappi E. Scann ing el ectron microscopic
ultrastructural a ltera tions of the pulmonary alveolus in
ex perimental drowning. J Forensic Sci 1983; 28: 1008-1 2.
20 Gu MN, Xia o JF, Hua ng YR et al. Study of di rect injury by
ACKNOWLEDGEMENT sea water in canine models. Di Yi JUI1 Yi Do X ue Xue Bao
2003; 23:201-5.
I thank Dr Terry Sinton, a senior for ensic pathologist of the 21 Tabata N, Morita M, Azu mi J. A frozen newbo m infant: froth in
John Tonge Cent re, Queensla nd Health Pathology Services , the ai r-passage afte r thawing. Forensic Sci Int 2000; 24:67-74.
22 Gotoh da T, Kubo S, Kitamura 0 et aL HSP70 an d c-Fo s
Brisbane, Australia, for mu ch encouragement.
expression of brain stem hyp oglossal nucl eus in drowning.
J Med Inl!est 2000; 47:76-9.
23 Sac hd eva RC. Near drowning. CritCare Gin 1999; 15: 28 1-96.
24 Heinemann A, Pus chel K. Discrepancies in homicide statistics
REFERENCES by suffo cation. Arch KrimilJo/1 99 6; 197:129-41.
25 Norman MG, Smialek JE, Newman DE, Horemba la EJ. The
Centers for Disease Contro l and Prevention (C DC). Nonfa tal post- mo rtem exam in ation o n the a buse chil d. Pathological,
a nd fata l d row nings in rec reatio nal wa ter settings - United radiograp hic, a nd legal aspects. Perspect Pediatr Pathol 1984;
States. 200 1-2002. MMWR 2004; 5 3:447- 52 . 8: 313 - 43.
2 Meyer RJ, Theodorou AA, Be rg RA. Chil dhoo d drownin g. 26 Saund ers E. Neonati cides followin g 'secret' pregnanc ies: sev en
Ped iatr Rev 2006; 27:16 3- 9. case reports. Public Hea lth Rep 1989; 104:368-72 .
3 Fin gerhut LA, Cox CS, Warner M. Internat iona l co mparative 27 Shiono H, Maya A, Tabata Net aL Med icolega l aspec ts of
ana lyses of injUlY mortality. Findings from the ICE on inj ury infant ici de in Hokkaid o District, Japan. Am J Forensic !VIed
sta tistics. International Collaborative Effort on Injury Path ol 1986; 7: 104- 6.
Statistics. Adv Da ta 1998; 303: 1-20. 28 Scott PD. Fatal ba ttered baby cases. IvIed Sci La w 1973;
4 Peden MM , McGee K. Th e epidemiology of drowning 13:1 97 - 206.
wo rld wide. Injury Cont Safety Promotion 2003; 10:1 95- 9. 29 Kondo T. Ohshim a T. Re trospective in vestiga tio n of medico
5 Laraque D, Barlow B, Durkin IVI. P reve ntion of yo uth injuries. legal autopsy cases involv ing mentally handicapped
J Natl Med Assoc USA 19 99; 9 1 :557-]. individu als. Nippon Hoiga ku Zasshi 199 5; 49:478-0 3.
6 Aus tralian Institute of He alth an d Welfare. Australia's Hea lth 30 Geertin ger P, Voigt J. Death in the bath. A survey of bath tub
2000. The Seven th Biennial Hea lth Report of the A ustra lian deaths in Copen ha gen, Denmark a nd Go th en burg, Sweden.
In stitute o f Health a nd Welfare. 4.2 Childre n a nd Young J Forel1Sic Med 1970; 17:136-47.
People. Canb erra: Aust lnst Health Welfare (AIHW Ca t. No. 31 Bourget D, Gagne P. Matern al fili cide in Quebec~ ) Am
19), 2000, pp. 184-92. Psychia try Law 2002 ; 30 :345-51.
7 Byard RW. Accidental childh ood death and the rol e of the 32 Rouge-MaiHart C, Jou sset N, Gaudin A et a l. Women who kill
pa th o log is t. Pediatr Dev Pat/101 2000; 3 :405-1 O. their chil d ren. Am J Forensic Med Parhol 2005 ; 26:320-6.
8 Pearn JH, Nixon J. A n analys is of the causes of freshwater 33 Vanamo T, Kauppi A, Karkola K et aL Intra -familial child
immersion accid ents involving chiJdren. Accident Anal Prell homicid e in Finland 1970-1994: incid ence , caus es of death
1979; 11:173-8. and demographic chara cteristics. Forel1sic Sci Int 200 I;
9 CoUins KA, Nic hols CA. A decade of pediatri c ho mi cid e: a ll7: 199-204.
retrospecti ve study a t the Medical Uni versity of So uth 34 DiMaio VJ. Hom icid a l asphyx ia. Am J Forel1Sic Med Patilo l
Carolina . Am J F01'ensic Med Pathol 19 99; 20:169 -72. 2000; 21: 1-4.
360 I Drowning and near drowning
35 Gillenwater JM, Quan L, Feldman KW Inflicted submersion in 63 Editor. Drowning in a private lake, NOlth Carolina, 1981 -9 0.
childhood. Arch Pediatr Adolesc !\lIed 1996; 150:298-303. MMWR 1992; 41:329-31.
36 Schmidt P, Grass H, Madea B. Child homicide in Cologne 64 Patrick M, Bint M, Pearn J. Saltwate r drowning and
(1985-94). Forensic Sci Int 1996; 31: 131-44. near-drowni ng acc idents involving ch ildren. Med J Aust
water in the home. J Pediatr 1980; 96:282 -4. 65 Schul pen TW. Migration and child hea lth: the Dutch
38 Wa lker S, Midde lkamp IN. Pail immersion accidents. Clill experience. Eur J Pediatr 1996; 155:35 1-6.
Pediatr 1981; 20:341-3. 66 Pearn JH. Th e sea, stingers a nd surgeons: the s urgeon's role in
39 Copeland AR. Suicide by drowning. Am J Forensic Med Path prevention, first aid and management of marine
1987; 8: 18- 22. envenomations. J Paediatr Surg 1995; 30: 105-10.
40 Schmidt P, Madea B. Homicid e in the bathtub. Forensic Sci IIlI 67 Pearn JH. Surv ival rates after se rious immersion acc id ents in
1995; 72:135 -4 6. childhood. Resuscitation 19 78; 6:271-8.
41 Pearn JH , Brow n H, Wo ng R, Bart R. Bathtub drownings: 68 Henderson H, Wilso n RC. Water incident related hospiral activity
report of seve n cases. Pediatrics 1979 ; 64:68-70. across England between 1997/8 and 2003/4: a retrospecti ve
42 Kemp AM, Mott AM, Sibert JR. Accidents and child abuse in descripti ve study. Bi\;lC Public Health 2006; 16:2 10.
bathtub submers ions. Arch Dis Child 1994; 70:435-8. 69 Wintemute GJ, Kraus JF, Teret SP et al. Deaths resu lting from
43 Pearn J. Predisposing fact ors leading to child trauma. motor vehicle immersions: the nature of the injuries, personal
J Epidemiol COI11I11 Hlth 1978; 32:190-3. and environmental contribu ting facto rs and potentia l
44 Byard RW, deKoning C. Blackbourne B et al. Shared bathing interventi o ns. Am J Publ Healrh 1990; 80:1068-70.
and drowning in infants and you ng children. J Pediatr Child 70 Smith GS , Langley JD. Drowning surveilla nce: ho w we ll do E
Health 2001; 37:542- 4. codes identify sub mersion fa ta lities. Injury Prev 1998;
45 Somers GR, Chiasson DA, Smith CR. Pediatric drowning: a 20 4: 135-9.
year review of autopsied cases: III Bathtub drownings. Am J 71 Start RD , Delargy-Aziz Y, Dorries CP et al. Clinicians and the
Foren sic Med Patho12006; 27:113-16. coroni al system: ability of clinicians to reco gnise reportable
46 Sibert J , John N, Jenkins D et al. Drowning of babies in bath deaths. BMJ 1993; 306:1038-41.
seJts: do they pro vid e fal se reassurance? Child Care Health 72 James DS , Leadbeatter S. Detectin g homici de in hospital. J R
Dev 2005; 3 1 :255-9. Coli PhYSicians Lond 1997; 3 1:296-8.
47 Nixon J, Pearn J. Non-accidental immersion in the bath: 73 Fornes p, Pepin G, Heud es D, Leco mte D. Diag nosi s of
anoth er extens ion to the syndrome of child abuse and drowning by combined computer-assisted histo morphometlY
neglect. Child Abuse Neglect 1977; 1 :455- 8. of lun gs with blood strontium determination. J Fore nsic Sci
48 Griest KJ, Zumwalt RE. Child abuse by drowning. Pediatrics 1998; 43 :772-6.
1989; 03:41-6. 74 Puschel K, Schulz F, Darrmann I, Tsokos M. Macromorphology
49 Ma rl eau JD , Roy R, Laporte L et al. Infanticide committed by and histology o f intramuscular haemorrhages in cases of
the mother. Can J Psychiatry 1995; 40: 142 - 9. drowning. Int J Legal ili/ed 1999; 112:101-6.
50 Pearn JH, Nixon J. Swimming pool immersi on accidents. An 75 Hooper WD, Johnson LP. Antiepi leptic drugs: pharmacology
analysis fro m the Brisb ane Drowning Study. Med J Aust and therapeuti cs. In Eadi e MJ, Vajda FJ (cds) Handbook of
19 77; 1:432- 7. Experimental Pharma cology. Berlin: Sprin ger- Verlag, 1999.
51 Brenin gsta ll GN. Mortality in pediatric epilepsy. Pediatr pp. 173-83.
Neuro12001; 25:9- 16. 76 Azparren JE , Vallejo G, Reyes E et al. Study of the diagnosti c,
52 Pearn J , Bart R, Yamaoka R. Drow nin g ri sks to epilep tic chloride, haemoglobi n and dia toms in imm ers io n cases.
children. A study from Hawaii. BMJ 1978; 4:1284-5. Forensic Sci Int 1998; 9 1 :123-32.
53 Wilkey I, Pearn J, Petrie G, Nixon J. Neonaticide, infanticide 77 Boles JM, Mabil le S, Scheydecker JL et at. Hyopglycaemia in
and chi ld homicide. Med Sci Law 1982; 22:31-5. salt water near-drowni ng v ictims. Intens ilJe Care jvIed 1988;
54 Wilkinson JL. Sudden cardiac death in childhood and 14:80.
adolesce nce. J Paediatr Child Health 1994; 30:384-5. 78 Zhu BL, Ishida K, Quan L et al. Post-m orte m urinary
55 Somers GR, Smith CR, Wi lso n GJ et al. Association of myoglob in levels with refe rence to the causes of death.
drown in g and myocarditis in a pediatric popU lation: an Forensic Sci Int 2001; 115 : 183-8.
autopsy-bas ed study. Arch Pathol Lab J\!led 2005: 79 Pollanen MS. Diato ms and homicide. Forensic Sci Int 1998;
129:205-9. 9:29-3 4.
56 Stumpp JW, Schn eider J, Bar W. Drownin g of a girl with
anomaly of the bundl e of His and the right bundle branch.
A 111 J Forensic !\lIed Pat/wi 1997 ; 18 :208 -10.
57 Schmidt P, Madea B. Death in the bathtub involving chi ldren. Further Reading
Forensic Sci Int 1995; 72:147-55.
58 Lunt DWR, Rose AG. Pathology of the human heart in
DROWNING: PATHOLOGY AND PATHOPHYSIOLOGY
drowning. Arch Parhol Lab iVIed 1987; 111:939-42.
59 Pearn JH , Brown J, Hsia EY. Swimmi ng pool drownings and Colebatch HJH, Halm agyi DFJ. Reflex pulmonary hypoten s ion of
near-drownings involving chi ldren. A total population study fresh-water aspirati on. J Appl P!JysiolI963; 18:179-8"5.
from Hawaii. Milit Med J980; 145 :15- 18. [A classic paper]
60 Brenner RA, Trumble AC, Smith GS et a l. Where children Good en B. Why some people do not drown. Med J Aust 1992;
drown, United States, 1995. Pedia trics 200 1; 108:85-9. 157:629-32.
61 Brenner RA, Saluj a G, Smith GS. Swimming lessons, Modell JH. The Path ophysiology and Treatment of Dro wning and
swimmin g ability and the risk of drowning. Injury Cont Safety Near-drowning. Springfie ld, IL: Charl es C. Thomas, 19 71.
Promotion 2003; 10:211-16. [A classic book]
62 Moran K, Stanley T. Pa renta l perce ption s of toddler water Pearn JH. Drowning and nea r-drowning. In Black JA (ed.). Paediatric
sa fety, swimm in g ability and sw immi ng lessons. lil t J Cont Emergencies, 2nd edn. London: Butterworths, 1987, pp. 40- 9.
Safety Promotion 2006; 13:139-43. Pearn JH. Pathophysi ology of drowning. iVIed J Aust 1985; 142 :586-8.
References I 361
INTRODUCTION
strictly hospital death s, but because they share many fea
tures with other anaesthetic de aths they are incl uded in this
Investigation of the sudden an d unex pected death of a chapter. The causes of sudden death of newborn babies in
baby or child in hos pital is a challenge that requires the hospital are different from those of older babi es a nd chil
combined skills of paediatric and forensic pathology. The dren and so will be discussed separately.
death may be due either to the condition for which the Sudden hospital deaths pose interesting lega l questions.
patient was bein g treated or to so me other unreco g nized [s death due to a recognized complication of treatmen t for
condition. Sometimes death is due to a complication of a lethal disease, natural or unnatural? When is an error
treatment or failure of a medical device. Occasionally, sud negligent? When is an error so serious that the increasingly
den death is due to an error by professiona l staff. Rarely, us ed criminal charge of manslaughter is appropriate?
death is due to the deliberate act of a carer. Whatever the answers to these questions, the post-mo rtem
Sta ff are often upset and understandably defensive, but it report is likel y to be closely scrutinized , and so the investi
is the pathologi st's role to tease out the facts of what hap gation and post-mortem examination must be canied out
pened in a non-judgemental manner so that parents receive a to the highest stan dards. Studies have shown the value of
full account of their child's death, and if errors have occurred autopsy examina tion in revealing major unexpected find
then the hospital is able to lea rn from them. A pathologist is in gs t hat might have affected outcome in 9 per cent of chi l
an advocate for the child and his or her parents, not the insti dren from paediatric inten sive care units (P[CUs), 7 per cent
tution. A post-mortem report indicating medical error seldom from emergency departments, 2 per cent from gene ral
results in a coronial findin g of negligence, but the penalties wards,I,2 25 per cent of pa ediatric oncology patients,J and
for failing to highlight poor practice may be heavy. 8.5 per cent of paediatric cardiac patients.4 ,5
The possible causes of sudden unexpected death in hos
pital are even more numerous than the many diseases,
drugs and procedures th at children en counter th ere. No DEFINITION AND FREQUENCY
attempt will be made to describe every eventuality that can
lead to sudden death in hospital. The ca uses are broadly The definition of sudden and unexpected death of children
grouped in Table 19.1. Deaths in ' the dental chair are not in a hospital context is necessarily subjective because these
Deaths: natural diesase I 363
Table 19.1 Causes ofsudden death ofchildren in hospital septic shock, embolism, cerebra l oedem a and cardiac dys
rhythmia. For exa mple, Manto n et al8 described two child ren
Due to natural disease
who died suddenly and unexpectedly in hospital with
Un recogn ized natural disease
haemolytic uraemic syndrome complicated respectively by
Complications of known natural disease
septicaemi a and haemorrhag ic cerebral infarction. Perhaps
Haemorrhage
surprisingly, unexpected hospital deaths from conditions
Septic shock
associated with su dden death in the community such as
Embolism
asthm a 9 and epilepsyl° appear to be rare. This is probab ly
Cerebral oedema
because sudden death in these two conditions is associated
Ca rdi ac dysrhythmia
with poor disease control and inadequate management of
Other
the acute attack, both of which are more likely to occur out
Failure to monitor
side hospital. Similarly, the 'dead in bed ' syndrome in young
Th erapeutic misadventure persons with type 1 diabetes ge nerally occurs at home and
Adverse drug events may be due to untreated hypoglycaemia or ketoacidosis. I I - I)
Errors of drug, dose or rou te of administration
Anaphylaxis
Other adverse reactions Fatal Haemorrhage
Complications of medical devices and procedures
Dea ths associated with surgery Fatal haemorrhage as a complication of an underlying con
Surgical dea ths dition may be overt or if it is internal it may be completely
Anaesthetic deaths unrecognized prior to autopsy. Haemorrhage can occur in
Accidents unrelated to medical care the course of haematological disorders, such as acute
Suicide leukaemia, or can fo llow erosion of a major vessel by an
abs cess, tumour, fore ign bodyl4 or vasculitis. IS Spontaneo us
Murder
rupture of the spleen with ensuing haemorrhage typically
Filicide
occu rs in infectious mononucleosis, but ca n occur in acute
Homicide by carers and others
leuk aem ia,16 malaria 17 and other conditions where the
spleen is enlarged . Rarely death is due to concealed bleeding
into a large tumour. 18 Haemorrhage in hospital is usually
children are usually already suffering from sign ificant ill
treatable but may be ovelwhelming, occasionally, resulting
n ess. Buchino et al 6 considered 'sudden' to be death within
in sudden death.
1 hour of clinically apparen t distress excluding time spent in
resuscitation, and 'u nexpected' to mean that no matter how
ill the child, death was no t considered imminent. They esti Septic Shock
mated that a children's hospital with 200 beds wo uld have
two to four such sudden and unexpected deaths per year.
Bacteraemia and fungaemia may complicate such recognized
Li fe -threatening collapse is surprisingly common in
infections as those of the urinary tract, meningitis, endo
specialist children's hospitals. Two per cent of more than
card itis, gastroenteritis, pneumonia, osteomyelitis, infective
6000 children admitted to a I22-bed university children's
arthritis or they may arise from a n unknown cause. There
hospita l during a I2-month period received cardiopul
may be predisposing factors such as chemotherapy, steroid
monary resuscitation; a susta ined circulation was restored
therapy, cyclic neutropen ia, HN infection, extensive burns,
in two-thirds of these children, 33 per cent were still alive
eczema or aspleni a . Indwelling vascular catheters and pro
at 24 hours and 15 per cent at 1 year. The most common
longed infusions are other strong associations. The overall
causes of cardiopulmonary arrest, which was not always
mortality in one study of bacteraemia and fungaemia in
unexpected, were respiratory fa ilure and shock. 7
childhood was 19 per cent, a nd risk factors for death included
neutropenia, hospital-acquired infection, polymicrobial
DEATHS DUE TO NATURAL DISEASE infection and ineffective empirical antibiotic treatment. The
Single strongest association was septic shock, WHich had a
Sudden unexpected death in h osp ital may be entirely nat mortality of 60 per cent. 19
ural. either as a result of a complication of an already diag Post-mortem diagnosis rests on suspecting the diagnosis,
nosed disease or due to a separate unrecognized disease. recognizing the source a nd co llecting fresh bacterial cultures
In either case, post-mortem exami nation plays a major role from several sites, especia lly a blood culture and a swab of
in determining the cause of death. sp leen. Some treatments such as steroids can completely
There are ma ny possible complicatio ns of natural disease suppress the symptoms of disseminated infection in children
that can cause sudden unexpected death, but a limited and so cultures must be a routine part of the examination of
number of final common pathways including haemorrhage, any child who dies suddenly and unexpectedly. 20
364 I Sudden death in hospital
Toxic shock syndrome occurs in children with burns 21 Fat embolism occurs in sickle cell disease, and may be a
and after even minor surgical procedures. It has recently cause of the acute chest syndrome in that condition. 53-55
been recognized as a complication of varicella infection Gray et al 56 cited a 9-year-old boy with sickle cell disease
and is sometimes fatal. 22 25 and acute chest syndrome who was found dead in his
hospital bed.
A related disorder occurs from lipid overload during
Fatal Embolism in Childhood intravenous feeding, causing respiratory compromise,
hepatosplenomegaly, abnormal liver function and coagu
Massive pulmonary thromboembolism in childhood is rare, Jopathy.57.58 The condition is generally reversible, but may
but lesser emboli are more common than is generally recog be fatal. Several studies have demonstrated intravascular
nized. 2G ,27 A study at the Hospital for Sick children Toronto lipid in the Iungs post mortem in infants who had been
found only eight cases of massive pulmonary thrombo receiving parenteral lipid emulsion, but its significance in
embolism causing sudden death in 17500 autopsies (0.05 per the absence of clinical signs is not known. 59- 62 Fatal lipid
cent),27 but Buck et al 26 found pulmonary emboli in approx accumulation in the brain has also been reported. 63 Salmon
imately 1 per cent of paediatric autopsies and considered pink discolouration of the spleen at autopsy is a clue to
that it contributed to death in 31 per cent of these cases. lipid overload - the fat can be stained using Sudan black in
Pulmonary thromboembolism may complicate prolonged routinely processed paraffin sections. However, this syn
immobilization, sepsis, dehydration,28 cardiac surgery, 29 drome should be rare with modern preparations of lipid
cardiomyopathy, nephrotic syndrome,30-33 haematological and clinical awareness of the hazard.
disorders with enhanced coagulation, occult malignancy Gas embolism is mentioned here for completeness, but is
and large vascular malformations. 34 -36 Vascular malforma usually a therapeutic misadventure resulting from catheter
tions may also be the explanation for reports of pulmonary accidents,64 open heart surgery, mechanical ventilation 55 or
embolism in conditions such as Proteus syndrome. 37 In chil neurosurgely, especially in the head-up position. Small
dren, pulmonary emboli may arise from cardiac, mesenteric amounts of venous air embolism are seldom fatal, but much
or cerebral veins and from the superior and inferior vena smaller volumes of air are necessary to cause death in
cava, as welJ as from pelvic and leg veins, which are the infants than in adults and older children. Venous air
usual sources in adults. Vascular catheters are a potent embolism has been demonstrated in more than 80 per cent
source of thromboemboli, which may be infected. of children undergoing craniosynostosis repair, although it
Inherited thrombophilias such as deficiency of protein C, resulted in hypotension in only one-third of cases and there
protein S or the presence of a Iupus anticoagulant are very were no deaths in this study66 Topical application, acciden
common in children with thrombotic events, whether or not tal ingestion or infusion of hydrogen peroxide can result in
there is another condition predisposing to the thrombosis. 3B fatal embolism of oxygen produced by the action of blood
Paradoxical emboli or quite small emboli arising in the and tissue catalase.57 ,58 Buchino et al 69 describe how a nurse
left side of the heart, for example in bacterial endocarditis mistook a central arterial catheter for a feeding tube and
and rheumatic fever,39 can cause sudden death by coronary injected a bolus of air with fatal consequences. Cardiovascu
artery occlusion and will be missed unless the coronary lar gas may be seen by computerized tomography (CT) after
arteries are carefully examined. Fatal coronary artery failed cardiopulmonary resuscitation, and is thought to be
embolism is also described in newborns. 40 .41 introduced during venous catheterization. 7o
Tumour embolism typically occurs in nephroblastoma, Diagnosis of fatal gas embolism depends on prompt
either spontaneously42 or as a complication of surgery,43.44 autopsy before oxygen can be absorbed,?1 radiology, and
but may also complicate other childhood renal tumours correct technique including inspection of the large veins
and tumours with large vein involvement. 45 .4G for bubbles, aspiration of the heart in situ under water, and
Fat embolism occurs after trauma , including non examination of the cerebral vessels before removal of the
accidental injury47 and orthopaedic procedures such as brain.
scoliosis surgery.48 Circulating fat is frequently found in
venous blood after fractures and bone operations without
apparent ill effects, but once cerebral and cardiovascular Cerebral Oedema
symptoms have appeared the mortality approaches 90 per
cent. 49 Reports describe sudden death due to fat embolism Cerebral oedema may complicate diffuse hypoxic-':ischaemic
after closed limb lengthening and also after accidental brain damage due to any cause, and so may be erroneously
fractures during manipulation of the hips in cerebral palsy, assumed to be the consequence of cardiorespiratory arrest
muscular dystrophy and severe dermatomyositis. 49 - 52 Fatal rather than its cause. However, it may be a very significant
fat embolism appears to be a particular risk for such chil finding in children who have collapsed suddenly and without
dren who have severely porotic bones and expanded obvious explanation.
marrow cavities as a result of immobilization or steroid Cerebral oedema is a common concomitant of severe
therapy. head injury, but it also occurs occasionally in young
Deaths: failure to monitor I 365
children following relatively minor head injuIY and can with other illnesses87 and during anaesthesia,88 Myocardial
cause unexpected death after a symptom-free interval. 72 infarction is more common in childhood than is often appre
The mechanism of delayed cerebral oedema in children is ciated,89.9o occurring in congenital heart disease, anomalous
unknown, but in rare older patients it has been associated origin of the coronary arteries, cardiomyopathy, as a result
with familial hemiplegic migraine. 7J Delayed deterioration of cardiac surgery, as a complication of myocarditis and vas
after head injury may also occur as a result of the develop culitis such as Kawasaki's disease and in cystic t1brosis. 91
ment of subdural or epidural haematoma .74 The effect of Sudden death may follow unrecognized cardiac contusion
space-occupying lesions such as tumours and abscesses due to trauma. Histiocytoid cardiomyopathy affects infants
may be exacerbated by local cerebral oedema, and they may and young children, especially girls, and, characteristically,
cause sudden death in hospital from acute hydrocephalus, causes recurrent arrhythmias or sudden death, sometimes in
brain swelling and coning 75,76 hospitaJ. 92 .93 Dysrhythmias and cardiac conduction defects
Cerebral oedema is a well-recognized but poorly under are common features of some of the disorders of fatty acid
stood complication of the treatment of ketoacidosis in oxidation, but not typically medium-chain acyl Co-A dehy
young children with diabetes in whom it is the most com drogenase deficiency.94 Recurrent dysrhythmia as in the long
mon cause of death. 12 ,77-79 Cerebral oedema develops in QT syndrome may be mistaken for breath-holding or
about 1 per cent of paediatric cases of diabetic ketoacido epilepsy95 and the author has encountered an example of
sis despite what is considered 'best practice' management, this latter situation causing sudden unexpected death in hos
and in one study had a mortality of 21 per cent. 79 About pital. Sudden death in anorexia nervosa may be due, inter alia,
one-half of these deaths follow sudden respiratory arrest in to sudden dysrhythmia, and has been associated with
hospital. Traditionally, the development of cerebral oedema acquired prolongation and dispersal of the QT interval. 96 - 98
has been attributed to rapid changes in the blood glucose The cause of cardiac arrest may be immediately apparent
level and administration of intravenous fluid, causing at the time of post-mortem examination. If it is not and
osmotic swelling of the brain. A case-control study did not death appears to have been due to cardiac dysrhythmia then
confirm these associations and showed hypocarbia, raised the heart should be retained with the necessary authoriza
serum urea and administration of bicarbonate to be risk tion for further detailed examination of the myocardium,
factors. 78 Evidence for and against is discussed by Glaser 80 conducting system and coronary arteries. Appropriate sam
Post-mortem examination occasionally shows that acute ples should be sent for virology, vitreous electrolyte values,
deterioration of consciousness and death in childhood dia and retained for possible metabolic and toxicological stud
betic ketoacidosis is due to other cerebral catastrophes, such ies. The genetic basis of many of the inherited defects of
as subarachnoid haemorrhage, basilar aliery thrombosis or cardiac conduction is now becoming understood, and
multiple thrombi. 81 retained genetic samples may sometimes prove as useful as
Cerebral oedema also occurs in other metabolic disorders examination of the heart itself. 99
such as acute liver failure in which it is a major cause of
death,82-84 and in some inherited metabolic diseases such as
maple syrup urine disease when it can cause sudden death DEATHS DUE TO FAILURE TO MONITOR
in hospital. 85
Cerebral oedema should be sought in any hospitalized It may not always be easy for a pathologist to recognize
child who dies after an unexpected deterioration in con that there has been a failure to anticipate complications
scious level. Signs of herniation and coning are a better and adequately monitor a sick child, because this some
guide than brain weight, and in babies these signs may dis times requires intimate knowledge of standards of clinical
appear if the brain is put to one side before examination. care. Failure to monitor is not always due to negligence
rather it may be due to lack of facilities or equipment fail
ure. As a rule of thumb, when intensive care facilities are
Cardiac Dysrhythmia adequate an acutely ill child should not die on a general
paediatric ward.
Cardiac arrest is the dramatic common endpoint of numerous Failure may involve simple observations such as pulse,
processes leading to hypoxia or electrolyte imbalance. Some temperature, blood pressure, level of consciousness and
times it is due to primary disease of the heali. The manage fluid balance so that a patient's gradual deterioration is not
ment of cardiac arrhythmias in children is not always appreciated. Failure may also involve more complex mon
straightfolVVard and adverse events resulting in death or itoring, for example of blood gases, electrocardiogram or
brain damage may become the subject of litigation. 86 intracranial pressure. Many routine treatments require
Sudden death from cardiac dysrhythmia can occur in monitoring using laboratory tests such as electrolyte values
myocarditis, cardiomyopathies, myocardial infarction, con during intravenous fluid therapy.
genital heart disease, and in abnormalities of the conducting In a retrospective review of case notes, Arieff et al 100
system and electrolyte imbalance. Unrecognized myocarditis described hyponatraemia due to hypotonic infusions after
can cause sudden death in patients who are hospitalized surgery in 0.34 per cent of children, with a consequent
mortality of 8.4 per cent. In a 6-year prospective study DEATHS DUE TO DRUG TREATMENT
- ---
- -
Deaths: drug treatment I 367
Table 19.2 Doctors occust:d of monsloughtt:r involving childrt:n following drug or onot:stht:tic t:rrors: UK 1970-99
in years (sex)
death or life-threatening toxic effects in 15 children. Koren chaJlenged, 1)7, 138 Dysrhythmia and even death has occurred in
et al 121 describe the sudden death of a newborn baby due to children und ergoing chemical skin peeling using phenol. 139
a tenfold error in a digoxin injection, and cite serious Respiratory depression caused by sedation used for
effects from tenfold overdoses of phenobarbitone, pal1 minor procedures also causes sudden death in children,
curonium and salb utamol. Buchino et al 6 quote the case of Adverse outcomes from sedation were more likely in non
a 4-month-old infant who received about 10 times the hospital settings, using multiple agents and when there was
intended dose of ketamine because the verbal order was no monitoring by pulse oximetry, There was no relationship
given in cubic centimetres rather than milligrams. Medica to the type of sedative or the route of administration, 140 A
tion errors in paediatrics have been reviewed by Sullivan syndrome of metabolic acidosis, lipaemia, rhabdomyolysis
and Buchino. 126 and fatal myocardial failure has been linked to long-term
infusion of propofol for sedation of children in intensive
care units,141 - 143 It has recentl y been suggested that this
Adverse Reactions to Drugs syndro me is due to impaired fatty acid oxidation caused by
the drug,144,145
Sudden deaths due to correctly administered drugs are
uncommon in infancy and childhood. However, infants may
react in ways not seen in adults because of differences in Anaphylactic Reactions to Drugs
drug metabolism. Immature drug metabolism explains the ----------------
grey baby syndrome due to chloramphenicol in neonates, Fatal anap hylactic reactions are very uncommon in children.
and probably why children who are under 3 years old Pumphreyl46 estima ted that there are about 20 anaphylactic
are particularly susceptible to sodium valproate-induced deaths of all ages per year in the UK, of which one-half are
hepatotoxicity.) 27 iatrogenic. A study of anaphylactic reactions in children
Dysrhythmias can occur during treatment with many gave a fatality rate of 2 per cent. 147 Most anaphylactic reac
drugs including tricyclic antidepressants and as a complica tions in children occur out of hospital and are due to food,148
tion of cardiotoxic drugS. 128 ,129 For exam ple, dysrhythmia Anaphylactic reactions in children may be biphasic in 6 per
may follow the intravenous administration of cardiotoxic cent of cases, leading to relapse after apparently sllccessful
chemotherapy such as doxorubicin. Although the great treatment. 147,1 49
majority of dysrhythmias in this situation are benign, sudden Almost any drug given by any route can cause a poten
deaths have been reported both acutely and as a late effect of tially fatal anaphylactic or anaphylactoid reaction, even
such therapy,130-1 33 Tumour lysis syndrome caused by rapid something as apparently innocuous as a bovine gelatine
destruction of tumour by chemotherapy may also cause containing chloral hydra te suppository, 150 Other causes include
sudden death by hyperkalaemia-induced dysrhythmia,1 34,1J5 vaccines, 151, 152 anti-venom,1 53 leukocyte infusions,154 skin
Cisapride, widely used for the treatment of gastroesophageal prick tests, 155-157 cryoprecipitate, 158, 159 gammaglobulin,160
reflux, has recently been implicated in sudden deaths due anaesthetic agents, radiological contrast medium,161 ,162 and
to dysrhythmias,136 although the relationship has been parenteral alimentation,1 63 Children with chronic illnesses
368 I Sudden death in hospital
and those requiring multiple procedures including bladder unless the device is very close.186- '90 Two-way radios used
catheterization may develop hypersensitivities. For example, by emergency crews are more likely to interfere with med
anaphylax is on exposure to latex and ethylene oxide occurs ical equipment than cell phones. '91 Addi tio nal hazards of
in spina bifida 164,165 and hypersensitivity to ethylene oxide infusion include administration of the w rong fluid 105 or
used for sterilization has been reported in renal dialysis fluid contaminated by bacteria. 192 Culture negative intra
patients. 166,167 Children with tumours may become sensitized venous fluids contaminated with endotoxin caused the
to chemotherapeutic agents such as asparaginase.'68.169 death of 36 neonates in an incident in Brazil. I93 Blood
Patients with cystic fibrosis may become sensitized to anti transfusion can be complicated by sudden dea t h due to
biotics, pancreatic extract and vascular catheters. 170, 171 transfusion reactio ns, either as a result of faulty grou ping
Post-mortem findings in anap hylaxis are often non and cross-matching or administration of blood intended
specific, but in children may include oedema of the respi for another patient. Patients who have received multiple
ratory tract and asthma-like changes in t he lungs. 146 .172 , I73 transfusions of blood products are particularly at risk of
Deaths due to an ap hylaxis are very rare, but failure to adverse reactions to further tra nsfusions. Frontela et al 194
diagnose anaphylaxis post mortem is often due to failure to describe an immunofluorescent technique developed to
think of the diagnosis and retain the appropriate specimens establish the diagnosis of mismatched blood transfusion
for measure ment of specific immunoglobulin E (IgEl and using paraffin-embedded tissue from a child at more than
mast cell tryptase, altho ugh the latter may give both false 2 years after death . Sudd en death has followed exchange
positive and false-negative results. 173 - 17 6 transfusion usin g blood damaged by overheating.
---
Deaths: medical devices and procedue s I 369
Conin g in meningitis may sti ll occur despite a noml al CT responses allow a smaller margin for error. Problems posed
scan 21J or without lumbar puncture. 214 Herniation may be by their vulnerability to heat loss, narrow a irway, small
reversed by vigorous treatment with hyperventilation and extracellular vo lume and age-depe nd ent responses to anaes
m annitol, a nd so it may be hard to confirm post mortem thetic agents make paediatric a naesthesia a field for special
unless there a re irreversible signs, such as necrosis of the ists. Paediatric surgery often involves children with multiple
cerebella r tonsi ls. 2l1 Coning may also cause sudden death in complex problems, thus addin g to the operative risk.
hospi tal when there is an undiagn osed space occupying Un expected deaths associated with surgery may be due
intracrani al les ion, such as a tumour or abscess,l5,76 and has to the patient's original condition. In that case questions
been describ ed in childhood acute leukaemia with central arise about the appro priateness of the decision to operate,
nervous system involvement. 21 5 and the adequacy of attempts to prepare a nd stabilize the
child for surgery. The 1999 report of the UK Nationa l
Confidential Enquiry into Perioperative Deaths found that
Ventriculoperitoneal and Ventriculoatrial most dea ths within 30 days of sur gery in children were in
Shunts those with congenital anomalies, necrotising enterocolitis,
tumours or trauma. These children gene rally had respira
Shunt failure in treated hydrocephalus has been a significa nt tory or cardiovascu lar problems in addition to their under
problem and continued to be so in the 1990s.216 Children lying diagnosis prior to surgery. In total, 16 per cent of
with shunted hydrocephalus may die sudden ly, sometimes deaths took place on the day of surgery. 226 Lack of experi
during in vest igation in hospital for a pparently minor symp ence and inad equate superv ision of junior staff are recu r
toms.217 Acute deterioration due to obstruction of the sh u nt rent themes in studies of anaesthetic deaths.
is often preceded by a history of headaches, and the child When unexpected death is due to the surgery itself, the
may be found de ad in bed.216 However, symptoms may be cause is often obvious at post-mortem examination. In con
non-specific, and shunt failure is not invariably accompa trast, anaesthetic deaths a re usually du e to respiratory failure
nied by v entricu lar enlargement. 218 Death can also occur or cardiac arrest, whic h may leave no trace, and so the ca use
due to recurrent pulmonary embolism from vent riculo atrial often ca nnot be determined by autopsy examination alone.
shunts a nd is usually heralded by respiratory sym ptoms or Pathologists cannot have the expertise to assess all aspects of
pulmonary hypertension .217 •219 A ventriculo atria l shunt has an anaesthetic and should seek independent expert advice.
caused coronary sinus thrombosis, myocardial infarction
and fatal coll apse in hospital. 22o Byard 221 describes sudd en
death in a child with a ventriculoperitoneal shunt that had Deaths Due to Surgery
perforated the transverse colon, leading to meningitis.
Post-mortem examinatio n should include careful ex am Deaths due to the surgery itself are either due to general haz
inatio n of the s ite of th e shunt for swellin g - both ends of ards of surgery, such as acute haemorrhage, or specific com
the sh unt in situ and the va lve for signs of obstruction, plications of pal1icular operations beyond tile scope of this
examin ation of any intravascular component for t hrombus chapter. Out of 289 paediatlic pelioperative cardiac arrests,
and careful cultures to rule out infection. The help of a 48 per cent were considered to be unrelated to tile anaes
neurosurgeon may be necessary to fully eval uate the func tIletic; of these, 22 per cent were cardiac surgery patients who
tion of the shunt and valve. could not be weaned off by-pass, anotller 17 per cent suffered
ca rdiac arrest as a result of u ncontroll able surgica l haemor
rhage, 22 per cent had miscellaneous ca rdiovascular eve nts, 4
Other Devices per cent had air embolism and 4 per cent had complicatio ns
of central venous li nes.227 Complications due to vascular
Pacemaker failure is a rare cause of sudden death in hospi cath eters inserted prior to the commencement of surgery are
tal but in all cases when a pacemaker is present it should be similar to those already described (see 'Vascula r catheters and
saved, with its lead intact for expert ex am in ation, and the infusions', above), but misplaced catheters may have disas
site of the tip should be examin ed histologically.222-225 It is trous effects duling surgery if tIley cause intrave nous anaes
important not to let the bare tip of the wire touch the bare thetic agents or resuscitative fluids to be delive red into the
body of the dev ice because electrical settings may be lost. wrong compartment.
Cardiac arrest may follow su rgical traction o~ the viscera
(the viscerocardiac reflex) or inte rference with the eye (the
Deaths Associated with Surgery occulocardiac reflex). 228 Bietti 229 reported such a case of a
and Anaesthesia child operated on for recess io n of the medial rectus w ho s uf
fered a fatal cardiac arrest. Pathologists investigating indi
Children are affected by the same surgical and anaesthetic vidual cases should not accept such possible explanations
accidents as adults but have a higher operative mortality uncri tica lly, and w itllout excluding every other possi ble
because their sma ll size and immature physiological ca use o f death.
~
-
Examination of children who die after cardiac surgery were the sickest group.237 The remarkable safety of paedi
requires specialist knowledge of cardiac morphology and the atric anaesthesia in a specialist setting is shown by a report
surgical techniques used to treat individual malformations. of 24165 paediatric anaesthetics in which there was a rela
Cardiac catheterization in infants and children carries a small tively high incidence of adverse events in neonates, but no
(about 0.14 per cent overall) risk of death, both from diagnos anaesthetic-related deaths in any age group.2JB
tic and interventional procedures, as a result of complications Cardiac arrest during anaesthesia or surgery is most often
such as perforation of cardiac chambers and bradycar due to hypoxia, which has many causes. Induction of anaes
dia. 2JO-232 As a general rule, surgical mortality increases with thesia may be complicated by broncho- and laryngospasm,
the length of time on cardiopulmonary by-pass, and a long failed intubation, or leakage around the endotracheal tube,
by-pass time is often an indication that problems were causing inadequate ventilation. The tube may be kinked, or
encountered during the surgery. Difficulty weaning the become blocked by a defective inflatable cuff or mucus.
patient off the pump usually reflects myocardial damage due Accidental oesophageal intubation is not necessarily fatal
to prolonged cardioplegia,233 although if the patient dies unless spontaneous respiration is prevented by muscle relax
intraoperatively or soon postoperatively there may be no ants, and may sometimes be suspected from erosions in the
morphological evidence of myocardial necrosis. Myocardial oesophageal mucosa even when the tube has been removed
necrosis may also be present preoperatively in congenital post mortem. Delayed detection of accidental oesophageal
heart disease. The risks of sutures obstructing branches of the intubation is rare, but in one study was more common in
coronary al1eries or impinging on the conducting system are infants and in association with emergency surgery.239 Many
common to several procedures. Allwork 234 gives helpful anaesthetic agents cause respiratory depression, either in
advice about likely surgical difficulties encountered in spe therapeutic doses or in over-dosage.
cific ca rdiac operations and their recognition post mortem. Pulmonary aspiration of gastric contents is a much
feared complication of general anaesthesia. However, it
occurred in only 0.04 per cent of 56] 38 consecutive
Complications of Anaesthesia patients who were less than 18 years old and who under
went 63 180 general anaesthetics for procedures performed
In one US teaching hospital 40 per cent of all anaesthesia in all surgical specialities at the Mayo Clinic over a period
related perioperative cardiac arrests were medication of 12 years. The risk of aspiration was highest in emer
related events, 20 per cent due to complications associated gency procedures, and those involving children who were
with central venous access, 20 per cent were attributed to younger than 3 years of age with bowel obstruction or
problems with airway management, and 13 per cent were ileus. Of the 24 patients who aspirated, only three required
due to unknown or possibly vagal reaction. The risk of mechanical ventilation and none died .240
death related to anaesthesia-attributable perioperative car Faulty anaesthetic equipment can cause fatal hypoxia.
diac arrest was 0.55 per ]0 000 anaesthetics. 235 A study of Incorrect plumbing of piped gases, failure of the oxygen
infants and children found an anaesthetic-related cardiac supply, incorrect f10w rates or leaks in the anaesthetic cir
arrest rate of 1.4 per 10 000 anaesthetics, with 55 per cent of cuit, faulty connections and accidental disconnections have
all arrests occurring in babies who were under] year of age all caused unexpected anaesthetic deaths.
and medication and cardiac causes accounting for 69 per Cardiac arrest may also result from dysrhythmias due to
cent of all cases. Cardiovascular depression from halothane, anaesthetic agents such as halothane, electrolyte abnor
alone or in combination with other drugs, was responsible malities or vagal inhibition caused by stimulation of the
for two-thirds of all medication-related cardiac arrests 22 7 In respiratory tract by irritant gases or passage of an endo
a study of over 700 000 anaesthetics, Morita et al 236 found tracheal tube. Ventricular fibrillation results from release of
the incidence of critical peri operative events such as cardiac adrenaline in patients who are inadequately sedated.
arrest and severe hypoxaemia to be greatest in babies who Collapse or death during anaesthesia and surgery in chil
were less than 1 month old, and that these babies also had dren may be the first manifestation of an unsuspected inher
the highest incidence of critical events related to anaesthetic ited or developmental abnormality. Death has been caused
management. However, although their mortality during and by perioperative fasting in a child with unrecognized very
within 7 days of anaesthesia was also the highest of any age long-chain acyl Co-A dehydrogenase deficiency2 4J Anaes
group, this was predominantly due to coexisting disease, thesia may precipitate cardiac arrest in children with
and no death was attributable to anaesthesia in children unrecognized congenital myopathy242-244 or cardiomyo
under 5 years of age. Causes of anaesthetic deaths in older pathy.227.245 Cucchiaro and Rhodes 246 reported a 9-year-old
age groups included anaesthetic overdose, toxic effects of boy who was found to have long QT syndrome after he
local anaesthetics, improper airway management, incom suffered ventricular fibrillation and cardiac arrest following
patible blood transfusion and errors in spinal anaesthesia. accidental intravascular injection of bupivocaine. Surgery is
Another study found 'major' perioperative events, which a potential hazard in children with sickle cell disease.247
included cardiac arrests and death, to be most frequent in McGarry and Duncan 248 described four sudden deaths asso
babies who were less than 1 month of age, although these ciated with surgery due to red blood cell sickling in children
Sudden death in the newborn I 371
with undiagnosed sickle ceLl trait. Acute intermittent por dental procedures resulted in 100 deaths in the UK from
phyria can present for the first time in childhood as a result 1970 to 1979 264 and 42 from 1980 to 1989,265 one-half of which
of an anaesthetic and cause death.249 Cardiac abnormalities occurred in children. 266 The number of dental anaesthetics
have recently been highlighted as a cause of otherwise unex administered in surgeries and those given by operator
plained sudden deaths during anaesthesia and surgery.250 anaesthetists has declined as a result of a flurry of reports
Malignant hyperthermia is a condition characterized by and stricter guidelines. 267 ,268 In the late 1990s approxi
sudden onset of tachycardia, acidosis, muscle stiffness, and a mately 300000 patients underwent general anaesthesia for
rapid rise in temperature, which, in susceptible individuals, minor dental procedures, with around 2- J deaths per year,
is precipitated by suxamethonium and certain inhalation usually in fit young adults or children who presented with
anaesthetics, including halothane. It is inherited as an auto no particular anaesthetic risk,269
somal dominant gene with variable penetrance, and recent Historically, about one-half of deaths in the dental chair
work has shown mutations of the RYRl gene in 30-50 per were due to respiratory causes and one-half were due to
cent of susceptible subjects. 251 The incidence of malignant sudden cardiac arrest. Causes included unobserved respira
hyperthermia in children is greater than that in adults, per tory depression due to premedication and multiple anaes
haps because of the anaesthetic agents used. Dantrolene is a thetic agents, sometimes in excessive doses, Cardiac arrest,
specific treatment. 252 At post-mortem examination, the mus due to either hypoxia or anaesthetics, such as halothane, is
cles may be paJe. Microscopy shows focal muscle necrosis hard to manage in a dentist's surgery.
and myoglobin in renal tubules, which may be necrotic. 253 ,254 The use of local anaesthetics in dental surgery, as in
The fulminant form still occurs, although it represents a fail general surgery, can cause collapse and even death due to
ure of diagnosis and treatment. 255 Anaphylactoid reactions to rapid absorption, overdose, intravascular injection or ana
neuromuscular blocking agents are quite distinct from malig phylaxis. The frequent simultaneous injection of a vaso
nant hyperthermia, but may also be fata1. 256 constrictor such as adrenaline can also cause adverse
Deaths in the recovery room are uncommon and are effects. Accidental over-dosage with local anaesthetic is
likely to be due to the delayed effects of surgery such as particularly easy in children, especially in those who are
haemorrhage or failure to reverse the anaesthetic, resulting undergoing multiple procedures on the same occasion. This
in respiratory depression. The airway may become possibility is exacerbated by the use of multiple agents
obstructed by mucus, blood clot or a forgotten throat pack. with names ending in '-caine' requiring different doses.
About 1 in 1500 children have an inherited deficiency of Overdose of local anaesthetic can be fatal in both children
cholinesterase, which greatly prolongs the effect of the and adults and causes nausea, anxiety, excitement, vomit
muscle relaxants succinyl choline and mivacurium, thus ing, convulsions and cardiorespiratory arrest. 270 Most cases
delaying the onset of spontaneous respiration after anaes of 'anaphylaxis' that are attributed to local anaesthetic
thesia. 257 ,258 Postopera tive cardiac arrests in recovery or agents are in fact probably due to inadvertent intravascu
the intensive care unit are rarely fata1. 259 lar injection.
Excessive volumes of intravenous fluid cause postoper Fortunately, death in the dental chair has become rare
ative pulmonary oedema, which most commonly presents since stringent standards have been set for the administra
without warning as cardiorespiratory arrest with normal tion of general anaesthetics in dentists' surgeries, but each is
electrolyte values. Post-mortem examination shows pul particularly tragic because it affects a fit child undergoing
monary oedema only, with no other cause of death.260 relatively trivial procedures. Post-mortem examination
Postoperative hyponatraemia and cardiorespiratory arrest should include not only toxicological studies, but also the
due to brain swelling is often caused by administration of possibility of unrecognized cardiac, skeletal muscle or meta
intravenous fluids that are low in sodium, possibly exacer bolic disorders that were exacerbated by the anaesthetic.
bated by inappropriate secretion of ADH, and autopsy Very rarely, oral surgery itself may cause death. Fatal
shows cerebral oedema. 100,261 A meta-analysis confirmed intracranial haemorrhage has followed accidental perfora
the association with hypotonic solutions,262 and a survey tion of the base of the skull during surgery on the temporo
of anaesthetists has shown that children in the UK are still mandibular joint in a child.271
at risk of this complication, especially in non-specialist
hospitals. 263 Irrigation of the operative site can lead to sub
stantial unrecognized absorption of fluid. SUDDEN DEATH IN NEWBORN BABIES
There have been few systematic studies of the incidence death later in infancy, may also cause collapse and death in
and causes of sudden death in the first week of life. Early the early newborn period.
neonatal sudden hospital death of newborns considered pre Sudden unexpec ted death in hospitalized infan ts with
viously healthy acco unted for 11 per cent of total neonatal bronchopulmonary dysplasia after prolonged mechanical
mortality in a Swedish study;274 a more recent study gave ven tilation has been reported as a significant cause of late
the rate of 'early neonatal sudden death syndrome' as 0.14 mortality, despite appropriate monitoring and prompt car
in 1000 live births, with approximately one-third of infants diopulmonary resuscitation. 281
dying within the first hour after birth.275 Quite similar fig When a newborn infant is found dead in hospital in bed
ures were found in a French study that noted that many with its mother then the possibility of accidental asphyxia
deaths occurred at night, and that some babies had signs of arises. A familiar scenario is tha t of a mother, perhaps
illness that were not recognized or acted upon. 276 exhausted by childbirth, who falls asleep while breast
Sudden death of newborn babies, either in special care feeding her new baby and wakes up to find it lifeless. 282
baby units or maternity wards, is usually due to age group Byard and Burne1l 283 have demonstrated the ease with
specific causes, or those related to labour, delivery and the which the huma n breast can cause accidental asphyxial
transition from intrauterine to extrauterine life. In term episodes, but such a possibility should be approached with
babies such likely causes include delayed effects of birth extreme sensitivity in the context of a neona tal death.
asphyxia or trauma, infection, genetic metabolic disease Similarly, the possibility of neonaticide has far-reaching
and malformations, especially ductus-dependent cardiac consequences, both for the mother and hospital staff, and
malformations such as hypoplastic left heart syndrome. 277 should be approached with care. Neonaticide is more com
In pre-term babies who are receiving intensive care, addi mon outside hospital but remains a consideration when
tional consideratio ns are complications of prematurity and there is no medical explan atio n for a newborn's death in
adverse effects of procedures and treatment.64.278 hospital , and there are other social and medical pointers
Early neonatal death due to frank birth trauma is now such as concealment or denial of pregnancy.
rare, thankfully, but minor trauma such as a fractured clavi Although no explanation may be found in a small pro
cle, excess moulding of the skull or minor tearing of the falx portion of babies who die suddenly and unexpectedly in
may indicate that the delive ry was not quite as simple as the the early neonatal period, such unexplained cases are
clinical records suggest. Unrecognized birth asphyxia may exceptional in the practices of specialist perinatal patholo
result in sudden unex pected death in the 'lying-in ' ward. gists. Careful review of the pregnancy and labour, com
Most fatal malformations are easily recognized at post bined with a detailed post-mortem examination, including
mortem examination but careful technique is necessary if appropriate laboratory investigation and examination of
subtler anom alies, such as those of the upper airway 279 and the placenta, will provide an explanation in most cases.
larynx, coronary arteries or pulmonary venous drainage,
are not to be overlooked. Minor atrial septal defects a nd
probe patency of the ductus arteriosus are not causes of ACCIDENTS
death in the neonatal period.
Infection may be acquired before, during or after birth . Fatal, non-medical accidents are seldom reported in hospi
The effects of infection are exacerbated by lack of mater talized children and are probably rare, although patients
nally acquired specific antibodies and immaturity of the and visitors are exposed to many of the same hazards as
immune system in newborns. Bacterial infection is tradi exist at home. For example, a 21-month old child was elec
tionally divided into early and late sepsis, the former prin trocuted in hospital by an uncovered lamp switch which he
cipally caused by enterovaginal organisms and the latter took into his mouth,284 and an infant aspirated the nipple
by nosocomially acquired bacteria. Group B streptococcal of a makeshift pacifier in a hospital nursery.28S
infection in newborns is usually an example of the former, A study of eight hospitals, includin g specialist children's
and can cause sudden collapse and death with only mini hospitals, over an 18-month period reported 781 non
mal signs at post-mortem examination. Viruses such as iatrogenic accidents involving patients and visitors under
herpes simplex and the enteroviruses cause devastating 16 years of age. Overall, 41 per cent of accidents to inpa
disease and sudden death in newborn babies, especially tients happened when their parents were present. Most
when they have no passively acquired maternal antibodies. involved falls, slips, striking, and scalds; three were seri
Severe infection in small babies is not always accompanied ous, resulting in fractures or bruising in children with
by symptoms obvious to the new mother. bleeding diatheses 286 There were no fatalities, although
Numerous inborn errors of metabolism can cause death one entrapment was potentially lethal. Falls from beds
in the neonatal period and are listed in appropriate text and cots in hospital usually result in only minor injuries
books. z8o Some, such as galactosaemia, are precipitated by even though beds are often higher than those at home,
the intro duction of particular comp ounds into the diet. It is although there are occasional skull fractures.287.288 Toys are
worth noting that medium-chain acyl Co-A dehydrogenase second only to beds and cots as a cause of equipment
deficiency (MCAD), which is familiar as a cause of sudden related hospital accidents, and attention has been drawn to
Filicide/homicid e in hospital I 373
th e dan ger to other children of toys brought into the hos In recent years it has been recogni zed that rare devian t
pital by visitors. 284 medical professional s may target vulnerab le pati ents caus
Finberg et af89 described the death of 6 out of 14 ing serial collapses and deaths. Frank discussion of this
infants in a hospital nursery, who were accidentally poi topic is hampered by the historical difficulty of obtaining
soned when their formula feed was made up using salt sustainable conviction s. How ever, children have fea tured
instead of sugar. 289 ,290 The most severely affected survivo r prominently amon g these so-called carer- assoc iated serial
had a serum sodium of 274 mmol/L and made a good killings (CASKs) because, like the elderly and tho se in
recovery. A similar episode occurred in Austria in which intensive care units, they are vulnerable and un able to
five newborn babies died. 291 speak for themselves. Some cases invo lv in g children are
de ta iled in Table 19.3.
In a comprehensive review on which much of this
SUICIDE account is based, Forrest 302 esti mated the incidence of
CASKs to be approximately 1 per million carers per year.
Of the 30 000 suicides per year of all ages in th e USA, 5-6 Nurses are the professional group most often involved, per
per cent occur in hospitals. 292 Suicidal ideation and action s haps because they are the most numerous; however, there
are well recog nized in children and yo ung people in are many instances of murderous doctors. 303 Park and
psychiatric wards, a lthough actual suicid es app ear to be Kh an 304 list 13 cases involving nurses from the UK and USA
uncommon. 293 ,294 Noren et al 295 describe a 15-year-old between 1975 and 1998. Perpetrators may be particularly
mental health patient who jumped from the third floor of a attracted to medical work because of the opportunities it
hospital and was found to have incidental lymphocytic affords, and may be motivated by a desire to control events
myocarditis at post-mo rtem exa mina tion. Children with or to show themselves in a good light in emergency si tua
chronic or terminal diseases such as cystic fibrosis may tions. In one adult series the motive appears to have been as
occasionally take their own lives. Repo rts of such deaths in trivial as wishing to win an unofficial sweepstake predicting
hospital are rare, but the possibili ty should be kep t in mind. the time when patients would die.
Bya rd reported 296 the case of a 14-year-old girl with cystic These deaths are extremely difficult to recognize because
fibrosis who injected hair conditioner via an intravenous medical professionals often have knowledge of, and access
infusion pump intended for systemic antibiotics, although to, methods that are almost undetectable. The victims are
suicidal intent was not proven in this case. pati ents a nd so post-mortem examination shows natural
disease that might have caused death and does not ra ise
concern. The alarm is usually raised by cli nical staff, but
FILICIDE AND HOMICIDE IN HOSPITAL typically not until the number of patients affec ted by col
lapses or death has reached double fig ures. The trigger may
Babi es and children in hospital may be harmed by their be the rec ognition of a train of unusual events or increased
paren ts, hospi tal staff or visitors. The motive may be mortality, or a member of staff may become suspicious after
uncl ear, but examples include mercy killing, abuse due to witn essing some strange behaviour by the perpetrator.
Mun chausen 's syndrom e by proxy, and frank psychosis . Yorker3 05 suggested a number of clues to early recognition:
Children may be admitted to hospital as a result of • a significant increase in cardiopulmonary arrests, or
deliberate harm that mayor may not have been recognized deaths or both , particularly in patients not tho ught to
for what it is, such as physical injury by a parent or simu be in immedi ate danger;
lated illness. Physical abuse may then continue in hospital, • an unusually high rate of successfu l resuscita tion;
and has been clearly documented by covert video su rveil • multiple events in the same patient;
lance.297 Rarely, a paren t will go to extraordinary lengths • events occurring more often in a particular shift.
to cause real or simulated illness in their child in hospital,
using methods such as suffocation , administration of drugs It is notable that in so me cases, the staff themselves
and poisons, tampering with samples, interference with have id entified a 'death shift' or a particular member of
equipm ent and introducing infected material into infusion sta ff involved in a large number of events, but their con
fluid s.298-301 It has been said that the techniques used are cerns have not been taken serio usl y by senior personnel.
limited only by the intell igence or imagin ation of the per As soon as con cerns are raised they must be shared with
petrator. Some of these methods are capable of resulting in senior medical and mana geria l staff, w ho will carry out
sudden unexpected death, but it is believed th at this is not further investigations and take the decision to inform the
usually the parent's objective and that such deaths in hos authorities. It would be wise to draw on the experience of
pital are rare. Very rarely a parent is implicated in the independent experts and those who have investigated sim
'mercy ki lling' of a severely handicapped child . The scope il a r cases before at an early stage.
for filicid e in hosp ital has been increased by the greater Possi ble action s that may be taken when such a death
availability of rooms in which parents can stay overnight occurs are shown in Table 19.4. The over-riding immediate
with their child. priority after protecting other children is to secure the
374 I Sudden death in hospital
Table 19.3 Some well-documen ted corer-associated serial killings involving children
Prelimin ary Toro nto Paediatric cardiac Dig ox in Uncertain; increased Case against nurse Susan
hearing 313 ward of children's mortality (x 4) on unit Nelles dismissed
hos pital over 9 months; nurse No proceedings taken
charged with four murders aga inst another known
only as nurse 'A'
Roya l Commission found
that out of 33 deaths 8
were certainly due to
digoxin and 15 others were
suspicious
Texas v. San Antonio Paediatric inten sive Hep arin, Su bsta nti a I exc ess Convicted and sentenced to
Genene ca re unit and suxa methonium mortality in PICU; 99 yea rs and 60 years,
Jones3 14 paedi at ric pri vate charged with one murde r respect ive ly, on each cou nt
practice an d one assault
R v. AllitP15 Grantham, UK Paed iat ri c wa rd s of Lign oca ine, in sulin , Charged with four murders Convicted
general hospital potassium ch loride and nine assau lts Clothier Inquiry316
Table 19.4 Some immediate actions that may be taken when a Vital information may be obtained from used syringes,
child dies suddenly and unexpectedly in hospital in suspicious 'empty' ampoules, and discarded samples for bl ood gas
circumstances to preserve evidence prior to the post -mortem analysis, all of wh ich may have to be recovered from the
examination sharps bin, taki ng appropriate precautions against injury.
Old samples already in the laborato ry and those from other
Control and limit access to the room
patients who may have been affected must also be secured
Institute log of perso ns entering the room
against being routinely disca rd ed.
Secure all samples and equipment (includes sharps containers
Additional samples must be taken at post-mo rtem
and waste)
examination including tissue from any possible injection
Ensure that no items on body are removed (vascular lines,
sites. Toxicological analysis will be informed by circum
end otrach eal tubes, pacemakers, etc.) and postpone
stances, also remembering that some of the agents used in
'la st offices'
prev ious cases, such as insulin and heparin, are no t
To preserve settings and memory, do not switch off
included in a routine toxicological screen . Some cases have
electronic equ ipment
involved inj ection of air.
Secure clinical records, fluid charts, drug records and records
Confinnation that serial killin gs have occurred and iden
of resuscitation
tification of the perpetrator often depend heavily on ep idemi
Ask all staff present at the event to make an immediate
ologica l evidence of increased death rates in a particular part
detailed written record of the event
of a unit at particular times when an individual was known
Notify laboratories to save all earlier samp les (haematology,
to be on du ty (so me drugs with a delayed action, such as
blood bank, chem istry, microb iology, etc.) until
digoxin, may be used to produce clinical deterioration and
notified otherwise
death after the perpetrator has left the wa rd). When finally
Keep any videotape from security cameras and any
identified, the perpetrato r is often found to have history of
op erative procedures
medical and employmellt problems. Such is the difficulty of
Inform coron er, medical director, management or police
identifying these individuals that they might never be discov
as approp riate
ered if they move around. In retrospect, some are found to
Consider taking immedi ate sa mpl es, subject to
have offended before in another centre.
coroner's permission
Prevention of such rare events as CASKs is 'extremely
difficult. The Clothier report into the case of Beverley AJlitt
noted that 'a determi ned and secret criminal may defeat the
scene, sampl es and equipm ent, to prev ent anyo ne remov best regulated organization in the pursuit of his or her pur
ing any tubes or other equipment from the body, and to pose '. Nevertheless, sou nd practices such as daily ch ecking
record who was present and invi te them to write down of ward drug stocks, ce ntral preparation of drugs adminis
their recollection of events as so on as possible. The imp or tered by intravenous infusi on, exclusion of ampoules
tance of preserving sa mples can not be overemphasized. of potassium chloride from general wards, the use of two
----~ -~ .
- -
Investigation of sudden unexpected death in hospital I 375
witnesses for all injections and the use of swipe cards judgement requires a very detailed clinical history and it is
to limit access to critical areas, not only reduces the important to speak to several members of staff before
omnipresent risk of accidents, but also makes the very rare accepting the history at face value. The parents' account of
occurrence of malfiscence more difficu lt. James and events may con tain important details unknown, overlooked
Leadbeatter 306 suggest that hospital policies could include: or even suppressed by staff. Start and Cross 307 provided
• Detailed recording of incidents of unexpected collapse general guidance for pathological investigation of deaths
to include those present at or around the time of following surgery, anaesthesia and medical procedures.
collapse. Pathologists seldom have the opportunity to carry out a
• Taking and storing of a blood sample following such a scene investigation with the body still in situ and so the
collapse for possible future analysis. best chance of establishing what went wrong may be lost
• Regular mortality review including analysis of place of long before he or she is involved if the hospital does not
death and persons in atten dance. have procedures in place for securing specimens and
• Detailed monitoring of potentially harmful drugs on equipment after any sudden and unexpected death. The
wards (drugs such as insulin, potassium chloride and natural urge to clear up the scene and complete 'last
other arrhythmogenic drugs should be monitored as offices' should be resisted in favour of a careful appraisal
closely as controlled drugs). of the scene by a senior member of staff. Where indica ted,
used syringes, ampoules, infusion solutions and blood bags
Pathologists may contribute to early detection by being should be preserved for possible labora tory analysis. Waste
scrupulous in assigning causes of death. It was notable in bins and sharps containers may contain useful samples, but
the Allitt case that 'status asthmaticus' and 'status epilepti should be searched with great care to avo id injury. Infusion
cus' were used in circumstances in which they did not pumps, temporary pacemakers, monitors, ventilators,
really fit the facts. In another case sudden infant death anaest hetic machines and other equipment should be set
syndrome (SmS) was used inappropriately. as ide for testing, but e lectroni c equipment should not be
switched off or unplugged until any settings or memory
have been checked by an expert . All tubes and catheters
INVESTIGATION OF SUDDEN UNEXPECTED should be left in the body, the reasons for so doing care
DEATH OF CHILDREN IN HOSPITAL fully explained to the parents. Staff should be asked to
write down their recollection of events, who was present,
When an infant or child dies suddenly and unexpectedly in t he position of the body as it was found and other details.
hospital, parents will be understandably distressed and may Th e post-mortem examination itself should only take
be angry. Hospital staff will also be upset and sometimes place when there is a full history and, if the cause of death
defensive. In the UK all such deaths should be reported to the is truly unknown must be a hybrid of the best of paediatric
coroner or procurator fiscal. Although it is often easier for and forensic practice (Table 19.5). It is essential that the
pathologists to suggest that the post-mortem examination pat holog ist is fully inform ed about the child's illness and
should be carried out by someone from outside the institu any procedures undergone. In the past, coro ner's autopsies
tion, an in-house pathologist is more likely to obtain full have been carried out by pathologists who seemed unaware
clinical cooperation and wi ll be better placed to provide of the complexities surrounding sudden death in the par
feedback to the parents and the hospital. When there is real ticular disease concerned. 56 A clinician who looked after
concern about serious error or negligence then it is wise to the child in life should be present to explain treatment,
seek an independent pathologist. procedures and any special con ce rns. In surgical deaths the
The investigation follows the same pattern as in adults, anaesthetist should be consulted and the surgeon should
but with additional considerations due to young age. In par attend in order to explain the procedure and ass ist (but not
ticular, small children are more likely to suffer accidental carry out) the examination of the surgica l site.
over-dosage of drugs, and unexpected deaths may be due to Radiographs obtained soon afte r death are necessary to
unrecognized congenital or inherited conditions, such as document a ny gas in vessels, and are useful as an objective
inborn errors of metabolism, myopathy and cardiac conduc re cord of t ubes and catheters. The value of routine photo
tion defects exacerbated by therapy for another condition. gra phs to supplement a full externa l examination and
The vulnerability of small children occasionally makes it nec description may only become apparent in retrospect. How
essary to think the unthinkable and consider deliberate harm. ever, photographs are not a substitute for a cohtempo rane
The scope of the examination varies depending on ous written and drawn record on a pre-printed body chart
whether a natural cause for death is very likely, an accident (those used in clinical departments for recording injuries to
is suspected or there is suspicion of foul play. Deaths clearly children are suitable). All punctures and injection sites
due to natural causes can be deal t with in the usual way, but should be recorded with an estimate of their age. Standard
suspicion of accident or other unnatural death should trigger measurements ane! body weight should be measured
a graded response adopting some or all of the measures accurately. When compared with the chi ld 's we ight on
in Tab le 19.4, depending on the degree of suspicion. This admission to hospital, the post-mortem body weight is
376 I Sudden death in hospital
Table 19.5 A checklist for post-mortem examination of a child who has died suddenly and unexpectedly in hospital
Gas embolism
Thromboembolism
Culture blood, spleen, lung, cerebrospinal fluid, any infective lesion, thrombus on catheter tips, etc. for bacteria
Frozen tissue samples, e.g. skeletal muscle, kidney, myocardium, liver [possible enzyme, muscle or toxicological studies)
Serum, urine, stomach contents, bile, injection sites, etc., for possible toxicology as advised by local toxicolog ist
sometimes a useful clue to excess or inadequate adminis examin ation will include all body cavities (any excess fluid
tration of intravenous fluids. should be measured an d saved for analysis and comparison
The body should be opened with care, noting the posi with infusion fluids) an d all major organs. Special atten
tion and condition of the tips of all tubes and catheters. The tion should be taken during dissect ion to exclude gas
• -
References I 377
embolism, pneumothorax and cerebral oedema. Trauma, In England an d Wales a co roner's pathologist is often
such as chest wall contusion, retroperitoneal haematoma, asked to determine w hether the death was natural or
stomach perforation and even rib fractures may sometimes unnatural. There is unfort una tely no definition of 'na tural
be attributable to vigorous resuscitation. 308 ,309 Subdural causes'. Deaths after surgery or medical treatment often fall
haemorrhage may be enco untered at post-mortem exami into the g rey area between natural and unnatural causes of
nation if there has been unsu ccessful open heart massage death. 311 Coroners themselves do not agree on how to use
during card iac surgery. these terms, and so it is as well to discuss borderline cases
In addition to standard histological sampling, it is nec individually. A majority of coro ners will accept natural
essalY to sample any lesion and any operative site. Snap causes for deaths following appropriate med ica l interve n
frozen samples of urine, serum, li ver, myocardium, skeletal tion for natural disease from which the patient was likely
muscle an d kidney will prove valu ab le if a n inherited to die, provided tha t there was no element of negligence
metabolic disorder or myopathy becomes a consid eration and the relatives do not object.312 Great care should be
late r, bu t when metabolic disease is suspec ted it is adv is taken in assigning the cause of death and in any accompa
ab le to con sul t a clinical biochemist over sa mplin g in nying comment not to go beyond the pathologist's expert
advance and to save a source of DNA. ise. A si gnificant propOJ1ion of medical negligence claims
Bacteriological samples should include blood and cere are attributable to inaccurate post-mortem diagnoses and
brospinal fluid as a minimum. Vitreous humour biochem speculative comments in post-mortem reports.
istry may point to an unexpected electrolyte disturbance.
Samples must be saved for toxicology in case a drug error is
subsequently suspected. Sub-optimal specimens hampered
ACKNOWLEDGEMENT
one study of anaesthetic-related deaths an d so a standard set
of samples that will include blood, urine and gastric contents
[ would like to acknowledge the assistance of Professor
should be agreed in advance with the lo cal toxicologist.
ARW Forrest with the section on carer-associated serial
Samples taken in life an d already sto red in the labo ra
killings.
tory may prove invaluabl e for metabolic, toxicological and
serological examination . In problematic cases the patholo
gist must take steps to secure these from routine disp osal.
REFERENCES
If the death remains unexplained after the gross post
mortem examination and was consistent with sudden car Go ld stein B, Metlay L, Cox C, Rubenstein JS. Association of
diac arrest then the heart should be saved with authorization pre morte m di agno sis and autopsy findin gs in pediatric
or, at the very least, extensively sampled, with the conduct intensive care unit versus emergency department versus ward
ing system saved for possible serial sectioning and samples patients. Crit Care iVIed 1996 : 24:683-6.
saved for 'molecular autopsy'.99 Any implanted pacemaker 2 Stambouly JJ, Kahn E, Boxer RA. Correlation be tween clinical
diagnoses and autopsy fin dings in critica ll y iLl children.
should be submitted for testing with its leads, although in a Pediatrics 199 3; 92 :248-51.
series of 38 surgically induced cases of complete heart block 3 Koszyca B, Moore L, Toogoo d I, Byard RW Is postmortem
in children treated by permanent pacemaker implantation examinat ion usefu l in pediatric oncology? Pediatr Pat/wi
den death is possible from the history then the fresh brain
necropsy in paediatric card iology. Heart 1996; 75:626-31.
should be examined closely for fla tten ing of gyri, herniation 5 Russell GA, Berry P J. Postmo rtem audit in a paediatric
and coning and authorization sought for fixation and formal card iol ogy unit. J Clin Path 01 1989; 42:9 J 2-18.
neuropathological examination. 6 Buchino JJ, Corey TS, Montgo mery V. Sudden unexpected
Ifthe cause of death is still undetermin ed when all inves death in hospitalized children. J Pediatr 2002; 140:46 1-5.
13 Thordarso n H, Sovik O. Dead in bed sy ndrome in 35 Bohm N, Bre ndlein F, Niederhoff H, Sutterle H. Diffuse
youn g diabeti c patients in Norw ay. Diab et Med 1995; hemangiomatosis of the sm all pelvis. Death from lung
J 2 :782-7. embo lism in an ll-yea r-old girl. Med Welt 1978; 29:839-43.
14 Grey TC, Mittleman RE, Wetli CV, Horowitz S. 36 Machin GA, Kent S. PulmonalY thromboembolism from a
Ao rtoes ophageal fistul a and sudden dea th . A rep olt of two large hemangioma in a 4-week-old infant. Pedjatr Pathol
cases and review of th e literature. Am J Fore nsic Med Fath ol 198 9; 9:7 3-8.
1988 ; 9:19-22. 37 Slavo tinek MI, Vacha SJ , Peters KF, Biesecker LG. Sudden
15 Cohle SO, Colby T. Fatal hemoptysis from Behcet's disease in death caused by pulmonary thromboembolism in Proteus
a child . Cardi ollClSC Pathol 2002 ; 11 :29 6-9. syn drome. Clin Genet 2000; 58:386-9.
16 Lash ley PM , Ramesh J , Ferdinand S. Spontaneous rupture of 38 Nuss R, Hays T, Manco-Johnson M. Childhood thrombosis.
th e spleen. West Indian Med J 1998; 47:172-3. Pediatrics 1995; 96:291-4.
17 Vidyashankar C, Basu A. Kulk arn i AR, Chou dhury RK. 39 Sta hl J, Santos lD, Byard RW. Coron ary artery
Spontaneous rupture of sp leen in falc iparum malaria. India'i J thromboembo lism a nd un expec ted death in childhood and
Gastroentero/2003; 22:10 1- 2. adolescence. J Forensic Sci 1995; 40:599-601.
18 Koszyca B, Moore l, Byard RW. lethal manifestations of 40 Muraskas J, Besinge r R, Bell T et al. Perinatal myocardial
neurofibrom atos is type I in ch ildhood . Pediatr PatllOl 199 3; infarction in a newborn with a structurally normal heart.
13:573-81. Am J Perinatal 1997; 14:93- 7.
19 Ashkenazi S, leibovici l, Samra Z et al. Risk factors for 41 Tillett A, Hartley B, Simpson J. Paradoxical embolism causing
morta lity due to bac te remia and fungemia in child hood . Clin fatal myocardial infarction in a newborn in fant. Arch Dis
I'~fect Dis 1992; 14:949 -51. Child Fetal Neol1atal Ed 2001; 85:F 137 -8.
20 lascari AD , Pearce JM, Swanso n H. Sudden death due to 42 Zakowski MF, Edwards RH, McDonough ET. Wilms' tumor
disseminated cly ptococcos is in a child w ith leukem ia in presenting as sudden death due to tumor embolism. Arch
remission. South Med J 1997; 90: 1253-4. Pathol Lab Med 1990; 114:605-8.
21 Brown AP, Khan K, Sinclair S. Bacterial toxico sis/toxic sho ck 43 Moore L. Byard RW. Fatal paradoxical emb ol ism to the left
synd rome as a contributor to morbidity in children with burn carotid artery during partial resection of Wilms' tumor.
injuries. Bums 2003; 29:733-8. Pediatr Parhol 1992; 12:551 -6.
22 Doctor A, Harper MB, Fleisher GR. Group A beta- hemolytic 44 Shurin SB, Gaud erer MW, Dahms BB, Conrad WG. Fatal
st reptococcal bacteremia: hi sto rical overview, changing intraoperative pulmonary embolization of Wilms tumor.
incidence, and recent association with varicell a. Pediatrics J Pediatr 198 2; 101 :559-62.
1995; 96:4 28 -33. 45 Arthur OS, Stephe ns CA, Brummitt WM et al. Fatal tumor
23 Hidalgo-Carballal A, Suarez - Mier MP. Sudden unexpected embolism during examination under anesthesia. Surgery
death in a chi ld with varicell a caused by necrotizing fasc iitis 1973 ; 74:466-8.
and strepto co ccal toxic sho ck syndrome. Am J Forensic Med 46 Chan KW, Fry er CJ, Fra ser GC, Dimmick JE. Sudden cerebral
PatllOl 2006; 27 :93-6. death in malignant presacral teratom a. Med Pediatr Oncol
24 Tseng HW, Liu Cc, Wang SM et al. Complications of 1985; 13:395-7.
varicella in children: emphasis on skin and cent ral nervous 47 Nich ols GR, Corey TS, Davis GJ. Nonfracture-associated fatal
system disorders. J Microbiol Immunol Infect 2000; fat embol ism in a case of child abuse. J Forel1sic Sci 1990;
33:248-52. 35:493-9.
25 Yang YJ, Liu CC, Wang SM et al. Streptococcal toxic shoc k 48 Shapiro G, Green OW, Fatica NS, Boachie-Adjei O. Medical
synd rom e complicating varicella in children. J Formos Med compli ca tions in scoliosis surgery. Curl' Opin Pedi atr 2001;
A ssoc 1997; 96:749-53. 13 :36-41.
26 Buc k JR, Co nnors RH, Coon WW et al. Pulmonary embolism 49 Shulm an ST, Grossman BJ. Fat embolism in childhood. Review
in children. J Pediotr Surg 1981; 16:385 -91. with report of a fatal case related to physica l therapy in a ch ild
27 Byard RW, Cutz E. Sudden a nd unexpected death in infancy with dermatomyositis. Am J Dis Child 1970 ; 120:480-4.
and childhood due to pulm ona lY thromboembolism. An 50 Edw ards KJ, Cummings RJ. Fat embolism as a comp li cation of
autopsy study. Arch Pathol Lab Med 1990; 114:142-4. clo sed femoral shortenin g. J Pediatr Orthop 1993; 12:542-3.
28 de la Grandmaison Gl, Durigon M. Pulmonary embolism: a 51 Ganel A, Israeli A, Horoszowski H. Fatal compli cation of
rare cause of sudden infant death. Am J Forensic ivIed Pathol femoral elongation in an achondroplastic dwarf. A case
2002; 23 :257 -9. repo rt. Clin Orthop 1984 ; 185:69-71.
29 Baudet EM, Puel V, McBride JT et al. long- term resu lts of 52 Pender ES, Pollack CV Jr, Evans OB. Fat embolism syndrome in
valve replacement with the St. Jude Medical pro sth esis. a chi ld with muscular dystrophy. J Emerg JvIed 1992; 10:705-11.
J Thorac Cardiovasc Surg 1995; 10 9:8 58-70. 53 Horton DP, Ferriero DM, Mentze r We. Nontraumatic fat
30 Citak A, Emre S, Sairin A et al. Hemostatic problems a nd embolism sy ndrom e in s ickl e cell anem ia. Pediatr Neurol
thromboembolic com pli cat ions in nephrotic ch ildren. Pediatl' 1995; 12:77-80.
Nephrol2000; 14:138-42. 54 Vi chinsky E, William s R, Das M et al. Pulmonary fat
31 Jones CL, Hebert D. Pulmo nalY thrombo-embolism in the embolism: a distinct cause of severe acute ches t syndrome in
nephrotic syndrome. Pediatr Nephro! 199); 5:56-8. sick le cell anemia. Blood 1994; 83:3107-12.
32 Mehls 0, Andrassy K, Koderisch J et al. Hemostasis an d 55 Vichinsky EP, Neum ay r lD, Ea rles AN et al. Causes and
thromboembolism in children with nephrotic syndrome: outcomes of the acute chest syndrome in sickle cell disease.
differences from adults. J Pediatr 1987; 110:862-7. National Acute Chest Syndrome Study Group. N Eng! J Med
33 Zimmerman Rl, Novek S, Chen JT, Roggli Y. Pulmona ry 2000; 342:1855-6 5.
thrombosis in a lO-yea r-old child w ith minimal change 56 Gray A, An ionw u EN, Davies S(, Brozovic M. Patterns of
disease and nephrotic syndrome. A clinical, radiologic, and mortality in sickle cell disease in the United Kingdom. J Clin
pathologic correlation with literature review. Am J Ciin Patho! Pathol 19 91; 44 :459-63.
1994; 101:230- 6. 57 Haber lM, Hawkins EP, Seilheimer OK, Saleem A. Fat
34 Baskerville PA. Thromboembolic disease and congenital overload syndrome. An autopsy study with eva luation of the
venous abnormalities. Phlebologie 1987; 40:531 - 6. coagulop athy. Am J Clin Patho! 1988; 90:22 3-7.
References I 379
58 Landry BA, Melhem RE. Pulmonary nodules seco ndary to 80 Glase r N. New perspec tives on the pathogenes is of cerebra l
total parenteral alimentation. Pediatr Radial 1989 ; 19:456-7. edema co mplicati ng diabetic ketoacidosis in children. Pediatr
59 Leve ne MI, Batisti 0, Wigglesworth JS et al. A prospective E17do crinal Re/J 2006 ; 3:3 79-86.
study of intrapul mon ary fat accumulation in the newborn 81 Rose nbloom Al. Intracerebral crises during treatment of
lun g following intra lipid infusion. Acta Paediatr Scand 1984; di abetic ketoacid osis. Diabetes Care 1990; 13:22-33.
63 Schulz PE, Weiner SP, Haber LIvI et al. Neurological edema causing death in children with map le syrup urine
complicatio ns from fat emulsion therapy. Ann Neural 1994; di sease. ) Pediatr 199 1; 119:42-5.
64 Vald es -D ape na M. Iatrogeni c disease in th e perinatal period as of cardiac arrhythm ias in children. Pediatrics 1987;
65 Lau KY, Lam PK. Systemic air embolism: a comp lication of 87 Sm ith NM, Bourne AJ, Clapton WK, Byard RW. The spectrum
ventilator therapy in hyalin e membrane disease. Ciin Radial of prese ntation at autopsy of myocarditis in infancy and
1991; 43:16-18. childh ood. Pathology 1992; 24: 129- 31.
66 Fab erowski LW, Black S, Mi ckl e JP. Inciden ce of venous air 88 Critchley LA. Yet ano ther report of anesthetic death due to
embolism during craniectomy for craniosynostos is repair. unsus pe cted myo carditis.) Ciin Anesth 1997; 9:676-7.
Anesthes iology 2000; 92:20-3. 89 Celermajer DS, Sholler GF, Howman-Giles R, Celermajer JM.
67 Cina SJ , Do wns JC, Conradi SE. Hydrogen peroxide: a so urce Myoca rdi al infarction in childhood: clinical ana lysis of 17
of lethal oxygen embolism . Case report and review of the cases and medium term follow up of survivors. Br Heart)
literature. Am) Forensic Med Pathol 1994; 15:44-50. 1991; 65:332- 6.
68 Lubec B, Hayn M, Denk W, Bauer G. Brain lipid peroxidation 90 John srude CL, Towbin JA, Cecc hin F, Perry Jc. Postinfarction
and hydroxy rad ical attack fo ll owi ng the intravenous infusion ventricular arrhythmias in children. Am Heart) 1995;
of hydrogen peroxide in an in fa nt. Free Radical Bioi Med 129: 1171 -7 .
1996 ; 21 :2 19- 23 . 91 Zebrak J, Skuza B, PogorzelskJ A et al. Partial CFIR
69 Buchino JJ, Kee nan WJ, Pi etsch JB et at. Malpos itioning of genotypin g an d characterisation of cystic fibro sis patients
the endotrac heal tube in infants with tracheo esophageal with myo ca rdial fibrosis and necrosis. C/in Genet 2000;
70 Shiotani S, Kohno M, Ohashi N et al. Cardiovasc ular gas on 92 Sheh ata 8M, Patterson K, Thomas JE et al. Histiocytoid
non-traumatic postmortem co mputed tomography (PMCT): ca rdi omyopathy: three new cases and a review of the
the influence of card iopulm onary resuscitation. Radiat Med literature. Pediatr De v Pat/JOI 1998; 1:56-69.
2005; 23 :22 5-9. 93 Stahl J, Couper RT, Byard RW. Oncocytic cardiomyopathy: a
71 Rudd PT, Wiggl esw orth JS. Oxygen embo lus during rare cause of unexpected early childhood death associated
mechanical vent ilation with di sappearance of signs afte r with fittin g. Med Sci Law 199 7; 37:84-7.
death. Arch Dis Child 1982; 57:237-9. 94 Bonnet D, Martin D, Pascale DL et a1. Arrhythmi as and
72 Denton S, Mileusnic D. Delayed sudden death in an infant conduction defects as present in g sy mptoms of fatty acid
followin g an accide ntal fall: a case report with rev iew of the oxidation disord ers in children. Circulation 1999 ;
73 Cox C. When ca n nurses be charged with manslaughter? Nurs 95 Allan WC, Gosp e SM Jr. Seizures, sy nco pe, or breath-h olding
Times 2001; 97 :21. presenting to th e pediatric neurol og ist - when is the etio logy
74 Hahn YS, McLone DG. Risk fact ors in the outcome of children a life-threatening arrhythmia? Semil1 Pediatr Neural 2005;
with minor head injury. Pedia tr Neurasurg 1993 ; 19: I 35-42. 12:2-9.
75 Ch un CH, Johnso n JD , Hofste tter M, Ra ff MJ. Brain abscess. 96 Cooke RA. Chambers JB. Anorexia nervosa and the hea rt. Br
A study of 45 co nsecuti ve cases. Medici ll e (Baltimore) 1986; ) Hasp Med 1995; 54:313 -17.
65:4 15-31. 97 Swen ne I, Larsson PI. Heart risk associated with weight loss
76 Shemie S, Jay V, Rutka J, Armstrong D. Acute obstru ctiv e in anorexia nervosa and eating diso rders: risk factors for
hydrocephalus and sudden death in children. Ann Em erg Med QTc interval prolongation and di spersion. Acta Paedia tr
1997; 29:524- 8. 1999; 88:304-9.
77 Edge JA. Cereb ral oedema durin g treatment of diab eti c 98 Va nn acc i A, Baronti R, Masini E et al. Anorex ia nervosa and
ketoacidosis: a re we any nearer finding a ca use? Diabetes the ri sk of sudde n death. Am ) Med 2002; 112:3 27-8.
Metab Res Re/J 2000 ; 16:3 16- 24. 99 Teste r DJ, Ackerm an MJ. The role of molecular a~topsy in
78 Glase r N, Barnett P, McCaslin I et al. Risk factors for cerebra l unexplained sudden ca rdiac death. Curr Opin Cardiol 2006;
edema in children with diabetic ketoacidos is. The Pedi atric 2 J: 166-72.
Emergency Medicine Co llaborative Researc h Committee of th e 100 Arieff AI, Ayus JC, Fraser CL. Hyponatraemia and death or
American Academy of Pediatrics. N El1gl ) Med 2001; permanent brain dama ge in healthy children. BM) 1992;
344:264-9. 304: 1218- 22.
79 Marcin JP, Gl ase r N, Barnett P et al. Facto rs associated with 101 al Sab ri AM, Smith N, Busu ttil A. Sudd en death due to auto
adv erse outcomes in children with dia betic ketoacidosis immune Addison's disease in a 12-yea r-old girl. Int) Legal
related cerebral edema. ) Pediatr 2002; 141 :793-7. Med 1997; I JO :278- 80.
380 I Sudden death in hospital
102 Berlatzky Y, Shiloni E, AnneI' H, Weiss Y. Delayed rupture of 126 Sullivan J E, Buchino JJ. Med ic ation errors in pediatrics: the
the spleen or de layed diagnosis of the splenic injury? lsI') octopus evading defeat.) Surg 011co12004; 88:182 - 8.
lVled Sci 1980; 16:659-64. 127 Johnson TN. The development of drug metabolising enzymes
103 Olsen WR, Polley TZ Jr. A second look at delayed splenic and their influence on the susceptib ility to adverse drug
rupture. Arch Surg 1977 ; 112:42 2- 5. reactions in children. Toxicology 2003; 192:37- 48.
104 Sta mbouly JJ, McLaughlin LL, Mandel FS, Boxer RA. 128 Swanson JR, Jones GR, Krasselt \IV et al. Death of two
Co mplications of ca re in a ped iatric intensive ca re unit: a subjects due to imipramine and deSipramine metabolite
prospective stud y. Intensive Ca re Nled 1996; 22: 1098- 104. accumulation during chronic therapy: a review of the
10 5 Yo ung RS , Tru ax BT. Hypernatremic hemo rrh agic li terature a nd possib le mechanisms.) Forellsic Sci 1997;
enceph alopathy. Ann Neurol 1979; 5:588-91. 42 :335 - 9.
106 Rouse DA, Hargrove R. An unusual case of gas embo lism. 129 Varley CK. Sudden death of a child treated with imipramine.
Am) Forensic lVled Parhol 1992; 13 :268-70. Case study. ) Child Adolesc PsychopiJarmacol2000; 10 :32 1-5.
107 Bagwell CE, Salzberg AM, So nnin o RE, Hay nes JH. 130 Gup ta M, Thaler HT, Steinherz L. Presence of prolonged
Potentially letha l comp lications of central ve nous catheter dispers ion of QT in terva ls in late surv ivo rs of childhood
placement.) Pediar Surg 2000; 35:709- 13. anthracycl ine the rapy. Pediatr Hematol Ollcol 2002;
108 Na rsingha ni U, Chadha M, Farrar HC, Ana nd KS. Life 19 :533 -42.
threatening respiratory fa ilure followin g accidental infusion 13 1 Krischer JP, Epste in S, Cuthbertso n DD et al. Clinica l
of polyethylene glycol elec tro lyte so luti on into the lu ng. card iotoxicity fo ll ow in g anthracycline treatment for
) Toxicol Ciin Toxicol 2001; 39 : 105-7. chil dhood cancer: the Pediatric Oncology Group experience.
109 We therton AR, Corey TS, Buchino JJ , Burrows AM. Fata l ) Clill Ol1col 1997 ; 15: 1544-52.
intravenous injection of potassiu m in hosp italized patients. 132 Ma ssin MM, Dresse MF, Sc hmi tz V et al. Ac ute
Am) Forensic lVled Patho12003; 24: 120 - 31. arrhyth mogen icity of fir st-dose chemo therap euti c age nts in
110 Cohen MR. Potassium chloride injection mix- up. Am ) Hasp chil dren. Med Pediatr 011co/2002; 39:93-8.
Pharm 1990; 47:2457-8. 133 Wortman JE, Lucas VS Jr, Schuster E et al. Sudden death
III AJ Ferayan A, Russell NA, AJ Wohaibi ]\II et al. Cerebrospinal dUJi ng doxorub icin administration. Cal1cer 1979; 44:1588-9 1.
t1uid lavage in the treatment of in advertent intrathecal 134 Arseneau ]C , Bagley CM, Anderson T, Canellos GP.
vincristine injection. Childs Nerv Sysr 1999; 15 :87 - 9. Hyperkalaemia, a sequel to chemothera py of Burkitt 's
112 Alcaraz A, Rey C, Concha A, ]\Iledina A. Intrathecal lymphoma . Lancet 1973; 1: 10-14.
vincristine: fatal myeloencephalopathy despite cerebrospi nal 13 5 Chasty RC. Liu-Yin JA. Acute tumour lysis syndrome. Br)
t1uid perfusion.) Toxicol Ciin Toxicol 2002; 40:557- 61. Hasp Med 199 3; 49 :488-92.
113 Dettmeyer R, Dri eve r F, Becker A et al. Fatal 136 Anon. Severe cardiac arrhythmia on cisapride. Prescrire JJ1f
myeloencephalop athy due to acc id ental intratheca l 2000; 9: 144-5.
vincristine administration: a report of two cases. Forensic 137 Levy J, Hayes C, Kern J et al. Does cisapride influence
Sci Int 2001; 122 :60-4. card ia c rhythm? Results of a United States multicenter,
114 Fernandez CV, Esau R, Ham il ton D et al. Intrathecal double-blind, placebo-controlled pediatric study. ) Pediatr
vincristin e: an analysis of reaso ns for recurrent fata l Gastroellterol NutI' 2001; 32:458-63.
chemotherapeutic error with reco mm end ations for 138 Ramirez-Mayans J, Garrido-Garcia LM, Huerta-Tecanhuey A
prevention. ) Pediatr Hemarol Oncol 1998; 20 :587 - 90. et al. Cisapride and QTc interval in children. Pedia trics 2000;
115 Kwack EK, Kim DJ, Park TI et al. Neural toxicity induced by 106: 1028-30.
accidental intrathecal vincristine administration . ) Korea n 139 Unlu RE, Alagoz IVIS, Uysal AC et al. Phenol intoxication in
Med Sci 1999; 14 :688-9 2. a child.) CmJ1ioJac Surg 2004; 15:1010-13.
116 Seale JR. Not again! Erroneous intratheca l. inj ect ion res ul ts 140 Cote CJ, Karl HW, Notterman DA et al. Adverse sedation
from a problem with protocols. BM) 200 1; 322:548. events in pediatrics: analysis of medications used for
117 Walter AW. Errors involving pediatric patients rece iving sedation. Pediatrics 2000; 106 :633-44.
chemotherapy: a literature review.1Vfed Pediatr Oncol 1997; 141 Cannon ML, Glazier SS, Bauman LA. Metabo li c ac idosis,
29: 155-6. rhabdomyolysis, and cardiovascular collapse after pro longed
118 Dyer C. Doctors face trial for manslaughter as criminal propofol infusion. ) NeuroSL/rg 2001; 95: 1053 - 6.
charges against doctors continue to rise. BM) 2002; 325:63. 142 Parke TJ, Stevens JE, Rice AS et al. Metabolic acidosis and
119 Ferner RE. Med ication errors that have led to manslaughter fatal myocardial failure after propofol infusion in children:
charges. BlVl)2000; 32 1:1212 - 16. five case reports. BM) 199 2; 305:613 - 16.
120 Kausha l R, Bates DW, Landrigan C et al. Medication errors 143 Vasile B, Rasulo F, Candiani A, Latronico N. The
and adverse drug events in pediatric in patients. )AMA 2001; pathophysiology of propofol infusion syndrome : a simple
205:2 11 4- 20 name for a complex syndrome. Intensive Care Med 2003;
121 Ko ren G, Barzil ay Z, Gree nwald M. Tenfo ld errors in 29: 1417-25.
adm inistratio n of drug doses : a neg lec ted iatroge nic disease 144 Wo lf A, We ir P, Segar P et al. Impaired fatty acid oxidation
in pediatri cs. Pediatrics 1986; 77:848- 9. in propofo l infu sion sy ndrome. Lal1cet 200 1; 357:606-7.
122 Koren G, Haslam RH . Ped iatri c med ication errors: predic tin g 145 Wo lf AR, Potter F. Propofo l infusio n in ch il dren: when do es
and preventing tenfo ld disas ters. ) Ciin Pharmacal 1994; an anesthetic too l become an intensive care liability?
34: 1043-5. Paediatr Al1aestil 2004; 14 :435- 0.
123 Conroy S. Unlicensed and off- label dllJg use: issues and 146 Pumphrey RS. Lesso ns for manageme nt of a nap hylax is from
recommendations. Paediatl' Drugs 2002; 4:353-9. a study of fata l reac tions. Clin Erp Allergy 2000;
124 Kozel' E, Scolni k D, Macpherson A et al. Variables assoc iated 30: 11 44-50.
with medication errors in pediatric emergency medicine. 14 7 Lee JM. Greenes DS. Bip hasic anaphylactiC reactions in
Pediatrics 2002; 11 0:737 - 42. pediatrics. Pediarrics 2000; 106:762- 6.
125 Kozer E, Scolnik D, Keays T et al. Large errors in the 148 Novembre E, Cianferoni A, Bernard ini R et al. Anap hy laxis
dosing of medications for chi ldren. N Engl ) Med 2002; in ch il dren: clinical and all ergologic features. Pediatrics
346:1175-6. 1998 ; 101 :EO.
-------- - - - -
- --- . """'""
References I 381
149 Novembre E, Calogero C, Mori F et al. Biphasic anaphylactic 172 Pumphrey RS. Fatal anaphylaxis in the UK, 1992-2001.
reaction to ketorolac tromethamine. Int J Immul1opathol Nouartis Found Symp 2004; 257:116-28.
Pharmacal 2006; 19:449-50. 173 Pumphrey RS, Roberts IS. Postmortem findings after fatal
150 Sakaguchi M, Inouye S. Anaphylaxis to gelatin-containing anaphylactic reactions. J Clin Patho/ 2000; 53:273-6.
rectal suppositories. J Allergy Ciin Immunol 2001; 174 Caughey GH. Il)'ptase genetics and anaphylaxis. J Allergy
108: 1033-4. Clin Immunol2006; 117:1411-14.
151 Bohlke K, Davis RL, Marcy SM et al. Risk of anaphylaxis 175 Horn KD, Halsey JF, Zumwalt RE. Utilization of serum
after vaccination of children and adolescents. Pediatrics tryptase and immunoglobulin E assay in the postmoltem
2003; 11 2:815-20. diagnosis of anaphylaxis. Am J Forensic Med Pathol 2004;
152 Ponvert C, Ardelean-Jaby D, Colin-Gorski AM et al. 25:37-43.
Anaphylaxis to the 23-valent pneumococcal vaccine in child: a 176 Yunginger JW, Nelson DR, Squillace DL et al. Laboratol)'
case-control study based on immediate responses in skin tests investigation of deaths due to anaphylaxis. J Forensic Sci
and specific IgE determination. Vaccine 2001; 19:4588-91. 1991; 36:857-65.
153 Singh A, Biswal N, Nalini Pet al. Acute pulmonary edema as 177 Co.llier PE, Blocker SH, Graff DM, Doyle P. Cardiac
a complication of anti-snake venom therapy. Il1dian J tamponade from central venous catheters. Am J Surg 1998;
Pediatr 2001; 68:81-2. 176:212-14.
154 Zylberberg R, Schott RJ, Fort J et al. Sudden death following 178 Collier PE, Goodman GB. Cardiac tamponade caused by
white cell transfusion in a premature infant. J Perinatal central venous catheter perforation of the heart: a
1987; 7:90-2. preventable complication. J Am Call Surg 1995; 181 :459-63.
155 Devenney I, Faith-Magnusson K. Skin prick tests may give 179 Suarez-Penaranda JM, Rico-Boquete R, Munoz JI et al.
generalized allergic reactions in infants. Ann Allergy Asthma Unexpected sudden death from coronal)' sinus thrombosis.
Immunol 2000; 85:457-60. An unusual complication of central venous catheterization.
156 Lockey RF, Benedict LM, Iurkeltaub PC, Bukantz Sc. J Forensic Sci 2000; 45:920-2.
Fatalities from immunotherapy (IT) and skin testing (SI). J 180 Pezzati M, Filippi L, Chiti G et a!. Central venous catheters
Allergy Ciin Immunol 1987; 79:660-77. and cardiac tamponade in preterm infants. Intensive Care
157 Reid MJ, Lockey RF, Iurkeltaub PC, Platts-Mills IA. Survey Med 2004; 30:2253-6.
of fatalities from skin testing and immunotherapy 181 Beardsall K, White DK, Pinto EM, Kelsall AW. Pericardial
1985-1989. J Allergy Ciin Immullol1993; 92:6-15. effusion and cardiac tamponade as complications of
158 Burman D, Hodson AK, Wood CB, Bmeton NF. Acute neonatal long lines: are they really a problem? Arch Dis
anaphylaxis, pulmonal)' oedema, and intravascular haemolysis Child Fetal Neonatal Ed 2003; 88:F292-5.
due to cl)'oprecipitate. Arch Dis Child 1973; 48:483-5. 182 Byard RW, Bourne AJ, Moore L, Little KE. Sudden death in
159 Harper JL, Gill ]C, Hopp RJ et al. Induction of immune early infancy due to delayed cardiac tamponade
tolerance in a 7-year-old hemophiliac with an anaphylactoid complicating central venous line insertion and cardiac
inhibitor. Thromb Haemost 1995; 74: 1039-41. catheterization. Arch Patitol Lab Med 1992; 116:654-6.
160 Shemin ER. Anaphylactic reaction to gamma-globulin. 183 Leibovitz E, Ashkenazi A, Levin S, Nissim F. Fatal pericardial
JAMA 1968; 203 :59. tamponade complicating total parenteral nutrition via a
161 Lasser EC, Lyon SG, Berl)' Cc. Reports on contrast media silastic central vein catheter. J Pediatr Gastroel1terol NutI'
reactions: analysis of data from repolts to the U.S. Food and 1988; 7 :306-7.
Drug Administration. Radiology 1997; 203 :605-10. 184 Rogers BB, Berns SD, Maynard EC, Hansen IW. Pericarclial
162 Shehadi WH. Death following intravascular administration tamponade secondal)' to central venous catheterization and
of contrast media. Acta Radio/ Diagn (Stockh) 1985; hyperalimentation in a vel)' low birthweight infant. Pediatr
26:457-61. Pathol 1990; 10:819-23.
163 Pomeranz S, Gimmon Z, Ben Zvi A, Katz S. Parenteral 185 Hermansen MC, Hermansen MG. Intravascular catheter
nutrition-induced anaphylaxis. J Parenteral Enteral NutI' complications in the neonatal intensive care unit. Ciin
1987; 11:314-15. Perinatoi2005; 32:141-56.
164 Bimlingham PK, Suresh S. Latex allergy in children: diagnosis 186 Anon. Radiofrequency interference with medical devices. A
and management. Il1dial1 J Pediatr 1999; 66:717-24. technical information statement. IEEE En9 iVIed Biol Mag
165 Moneret-Vautrin DA, Laxenaire MC, Bavoux F. AJlergic 1998; 17:111-14.
shock to latex and ethylene oxide during surgel)' for spinal 187 Betts IR, Simpson lAo Inhibition of temporary pacing by a
bifida. Anesthesiology 1990; 7J :556-8. mobile phone. Heart 2002; 87:130.
166 Bommer J, Ritz E. Ethylene oxide (EIO) as a major cause of 188 Robinson MP, Flintoft lD, Marvin AC. Interference to medical
anaphylactoid reactions in dialysis (a review). Artif Organs equipment from mobile phones. J Med Eng Technol 1997;
1987; 11:111-17. 21 :141-6.
167 Hurley RM. Anaphylaxis during hemodialysis. Int J Pediatr 189 Iat FH, Wah KC, Hung YH. A follow-up study of
Nephro11984; 5:53-4. electromagnetic interference of cellular phones on electronic
168 Dellinger CI, Miale ID. Comparison of anaphylactic medical equipment in the emergency department. Emerg
reactions to asparaginase derived from Escherichia coli and Med (Fremantle) 2002; 14:315-19.
from Erwil1ia cultures. Cancer 1976; 38:1843-6. 190 Iri JL, Severson RP, Firl AR et al. Cellular telephone
169 Zanotti KM, Markman M. Prevention and management of interference with medical equipment. Mayo Ciin Proc 2005;
antineoplastic-induced hypersensitivity reactions. Drug 80: 1286-90.
Safety 2001; 24:767 -79. 191 Boyle J. Wireless technologies and pati,ent safety in
170 Ganier M, Lieberman P. IgE mediated hypersensitivity to hospitals. Telemed J E Health 2006; 12:373-82.
pancreatic extract (PE) in parents of cystic fibrosis (CF) 192 Ihong ML, Iay LK. Septicaemia from prolonged intravenous
children. Clil1 Allergy 1979; 9: 125-32. infusions. Arch Dis Child 1975; 50:886-9.
171 Haworth CS, Niven RM, Moorcroft AJ et a!. Acute 193 Garrett DO, McDonald LC, Wanderley A et aJ. An outbreak of
anaphylaxis following midline catheterisation in a patient neonatal deaths in Brazil associated with contaminated
with cystic fibrosis. Thorax 1999; 54:747. intravenous fluids. J Infect Dis 2002; 186:81-6.
References I 383
239 Chinachoti T, Surase ranivongse S, Pengpol W, Valairucha S. 262 Choong K, Kho ME, Menon K, Bohn D. Hypo to nic v ersus
Delayed detecti on of esophageal intubation: Thai Anest hesia isoto nic saline in hospitalised children: a systematic revi ew.
Incidents Study (THAI Study) d atabase of 163,403 cas es. Arch Dis Chi ld 2006; 91 :828-35.
J Merl Assoc Thai 2005; 88(Suppl. 7):S69-75. 263 Way C, Dhamrait R, Wade A, Walker 1. Perioperat ive fluid
240 Warner MA, Warner ME , Warner DO et a1. Periopera tive the rapy in c hildren: a s urvey of cu rrent presc ribing pract ice.
pulmon ary aspiration in infants and chi ldren. Anesthesiology Br J Anaesti1es ioi2006; 97:37 1-9.
1999; 90:66-71. 264 Copl ans MP, Curson I. Dental anaesthesia a nd mOliality.
241 Roe CR, Wiltse HE, Sweetman L, Alvarado LL. De a th caused Br Med J (Ciin ResEd) 1982; 284:661 -2.
by perioperative fasting and sedat ion in a child with 265 Copl a ns MP, Curs on I. Dea ths associate d w ith dentistry an d
unrecognized very lo ng chain acy l- coenzyme A d en ta l di sease 1980-19 89. Anaesthesia 1993; 48:43 5- 8.
dehydrogen ase deficiency. J Perlia tr 2000; 136 :397-9. 266 Worthin g ton LM, Flynn PJ, Struni n L. Dea th in the de nta l
242 Breucking E, Rei mnitz P, Schara U, Mortier W. Anesthetic chair: a n avo idab le ca tastrophe? Br J Anaesth 1993;
co mplication s. The incidence of severe anesth etic 80: 13 1-2.
com plication s in patients and families with prog ressive 267 De nta l Anaesthesia Comm ittee of the Royal College of
muscular dystrophy of the Duchenne and Becker types. Anaesthetists. Standards and Guideli nesjor General
Anaesthesist 2000 ; 49 : 18 7-95. Anaesthesia jar Del/tistry. London: The Roya l College of
243 Larach MG, Rose nberg H, Gron e rt GA, Allen Gc. Did A na esthe tis ts, 1999.
anesthetics tr igge r cardiac arrests in patients with occult 268 Poswillo D. General Anaes thesia Sedation a nd Resusc itat io n
myopathies? Anesthesiology 2001; 94:933- 5. in dentist ry: Repon of a n Expert Working Party for the
244 Oudesluys-Murphy AM, van Berke l JH, van den Bran d HM. Standing Dental Advisory Commi ttee. London: Dep artment
Unexplain ed sudden deatb und e r genera l anaesthesi a. Lancet of Health, 1990.
1985 ; 1 :696-7. 269 Cartvl'right DP. Dea th in the dental ch a ir. A naesthesia 1999;
245 Schmidt GN , Burmeister MA, Lilje C et a l. Acute b ea rt failure 54: 105-7.
during spin a l s urgery in a boy with Duchenn e muscul ar 270 Malamed SF. Morb idity, mortality and local anaesthesia.
dystrophy. Br J Anaesth 2003 ; 90:800-4. Primary Dent Care 1999; 6:11-1 5.
246 Cuccbiaro G, Rbodes LA. Unu s ual presenta tion of long QT 271 Lifsc hultz BD, Kenn ey JP, Sturgi s CD , Donoghue ER. Fatal
sy ndrome . Br J A naesth 2003; 90:804 -7. intrac ranial hem orrhage follo win g ped iat ric oral s urgica l
247 Lagard e MC, Tun e ll WP. Surgery in patients w ith proced ure. J Fo rensic Sci 199 5; 40: 13 1-3.
hemoglobin-S disease. J Pediatr Surg 1978; 13 :605-7. 272 Keelin g JW. Fetal and Neonatal Pa thology, 4th edn. Lo ndon:
248 McGarry P, Dunca n C. Anesthetic risks in sickle ce ll trait. Springer, 2007.
Pediatrics 1973; 51:507 - 12 . 273 WiggleswOlih JS. Perina tal Path ology, 2nd edn. Lond o n:
249 Dover SB, P le nderleith L. Moore MR, McC oll KE. Safety of WB Saund ers, 1996.
ge ne ra l anaesthes ia a nd s urgery in acute he pat ic porphy ria. 274 Po lberger S, Svenningsen NW. Ea rly neonatal sudden infant
Gut 1994; 35: 111 2-15. dea th a nd near death of full-term infa nts in mate rni ty
250 Tabib A, Lo ire R, i\Iliras A et a l. Unsuspected card iac lesion s wards. Acta Pa ediatr Scand 198 5; 74:861-6.
associated with sudden un ex pec ted periop e rative death. Eur 275 Rodriguez-Alarcon J, Melchor Jc. Linares A et a1. Early neo natal
J Anaesthes iol 2000; 17:230- 5. sudden death or near death syndrome. An epidemiolo g~cal study
25 t Galli L, Orrico A, Loren z ini Set al. Frequen cy a nd of 29 cases. Acta Paediatr 1994; 83:704-8.
localization o f mutation s in the 106 exo ns o f the RYR J gene 2 76 Deh a n M, Imbert MC, Benisvy C et a l. [Sudden newbo rn
in 50 indiv idu als w ith malignant hyperthermia. Hum Mulat infa nt deat h in mate rnity. Anatomo-c1inica l study of 3 1
2006; 27:830. casesJ. Arch Fr Ped iatr 1992; 49: 609- 16.
252 Hallid ay NJ. Malignant hyperthermia. J Cran iojac Surg 277 Pina r H. Postm onem findings in term neonates. Sel11in
200 3; 14 :800-2. Neonato l 2004 ; 9 :289-302.
253 Karger B, Te ige K. Fatal m a lig nant hyperthe rmia: delayed 278 Keeling JW. Iatrogenic disease in the newborn. Vi rchoUis
onset and aty pical course. Forensic Sci lilt 2002 ; Arch A Pathol A nat Histo1 198 1; 394: 1-29.
129:1 87 - 90. 279 Byard RW, Kennedy JD. Dia g nost ic difficulties in cases of
254 Prescott RJ, Roberts SP, Wi.lIiams G. Malig nant s udden death in infants with mandibular hyp op la sia.
hyp erpyrexia: a rare cause of postoperative death. J Clin Am J Forensic Med Pathol 1996; 17 :255-9.
Path o11 992; 45:361-3. 280 Wigglesworth JS. Performan ce o f the perin atal a utopsy. In
25 5 Ellis FR. Malignant hype rpyrex ia . Arch Dis Child 1984 ; Perinatal Pathology, 2nd edn . Lo nd on: Saund ers, J 996, p. 36.
59: 1013-1 5. 28 1 A bman SH, Burc he ll MF, Schaffer MS, Rosenberg AA. Late
256 Watkin s 1. Adverse rea ct ion to neuromusc ular blockers : s udden unexpected deaths in hospitalized infants with
frequ e n cy, investiga tion, a nd epidemi ology. Acta bronchopulmonary dyspl asia. A m J Dis Child 1989;
Anaesthes iol Scand 1994; 102(Suppl.):6-10. 143:815- 19.
25 7 Fox MH, Hunt Pc. Prolonged neuromusc ular block 282 Espagne S, Hamo n I. Thiebau geo rges 0, Hascoet JM. [Sudd e n
associated with mi vacurium. Br J Anaesth 1995; death of n eo n ates in the delive ry roo mJ . Arch Pediatr 2004 ;
74: 237-8. 11 :436-9.
258 Pasqu ariello CA, Schwartz RE. Plasm a ch olinesterase 283 Byard RW, Burn ell RH. Appare nt life-threatening events and
defici ency in a neon ate. Ca n J Ana esth 1993; 40:529- 3 1. infant h olding practices. A rch Dis Child 199 5; 73 :502-4.
259 Hoffman TM, Wern ovsky G, Wieand TS et al. The in cid ence 284 Frank DJ, Drobish NL. Toy safety in hospi tals: or beware of
of a rrhythmias in a pedia tric ca rdiac inte nsive care unit. parents bea rin g gifts. Ciin Pediatr (Phila) 1975; 14:400-2.
Pedia tr Cordial 2002; 2 3:598-604. 28 5 M illun ch ic k EW, MeAno r RD. Fatal a sp ira tion o f a makesh ift
260 Arieff AJ. Fatal pos toperative pulmonary edema: pathogenesis pacifier. Pediatrics 1936; 77 :369-70.
and litera ture revi ew. Chest 1999; 115:1 371-7. 286 Levene S, Bo nfield G. Accidents on hospital wards. Arch Dis
261 Brouh Y, Paut 0, Tsim a ratos M, Camb o ulives J. Postoperative Child 199 1; 66: 1047-9.
hyponatremia in chi ld re n: pathophysiology, diagnosis an d 287 Lyons TJ, Oates RK. Falling out of bed: a relatively benig n
treatme nt. Ann Fr Anesth Reanim 2004 ; 23 :39-49. occurrence. Pediatrics 1993; 92:125 -7 .
384 I Sudden death in hospital
288 Nimityongskul P, Anderson LD. The likelihood of injuries 302 Forrest ARW The Investigation and Prosecution of Health
when children fall out of bed. ) Pediatr Orthop 198 7 ; Care Workers Who Systematically Harm Their Patients. LL.M
7: 184-6. thesis, Uni versity of Wales, 1992.
289 Finberg L, Kiley J, Luttrell CN. Mass accidenta l sa lt 303 Kinnell HG. Serial homicide by doc to rs: Shipman in
poisoning in infancy. )AlI!LA J963; 184:187-90. perspective. BM) 2000; 321: 1594-7.
290 Finberg L. Unfo rgettable patients.) Pediatr 1992; 304 Park GR, Khan SN. Murder and the ICU. Eur) Anaesthesiol
121 :323-4. 2002; 19:621-3.
291 Walter GF, Maresch W [Acc id ental sa lin e poisoning in 305 Yorker Be. Nurses accused of murder. Am ) Nurs 1988 ;
newborn infants. Morphologic findings and pathogenetic 88: 13 27-8, 1332.
discussio n]. Klin Padiatr 1987; 199 :269-73. 306 James DS, Leadbeatter S. Detecting homicide in hospital.
292 Busch KA, Fawcett J, Jacobs DG. Clinical correlates of ) R Call Physicia11S LOlldol1 1997 ; 31:296-8.
inpatient suicide. ) Ciin Psychiatry 2003; 64: 14-19. 307 Start RD , Cross SS. AC P. Best practice no 155. Pathological
293 Pfeffer CR, Plutchik R, Mizruchi MS, Lipkins R. Suicidal investigatio n of deaths following surgery, anaesthesia, and
behavior in child psychiatric inpati ents and outpatients and medical procedures.) C/il1 Patlwl 1999; 52:640-52.
in nonpatients. Am) Psychiatl)' 1986; 143:733-8 . 308 Bush CM, Jones JS, Cohle SD, Johnson H. Pedia tric injuries
294 Wo lfersdorf M. [Patient suicide in the psychiatric hospita l: from cardiopulmonary resuscitation. Anl1 Emerg Med 1996;
selected results of the Clinic Suicide Wo rking Group Study 28 :40-4.
1/11 1970-1992 of the 'Su icidal Behavior and the Psychiatric 309 Maguire S, Mann M, John Net al. Does cardiopulmonary
Hospi ta l' Study Group]. Psychiatr Pra_r 1996; 23 :84-9 . resuscitation cause rib fractures in children? A systematic
295 Noren GR, Staley NA, Bandt CM, Kaplan EL. Occurrence of review. Child Abuse Negl 2006; 30:739-5 1.
myocarditis in sudden death in children. ) Forensic Sci 310 Peterson DB, Beckwith JB, Benson EA. The sudden infant
1977; 22:188-96. death syndrome in hospitalized babies. Pediatrics 1974;
296 Knight DM, James RA, Sims DN et at. Sudden death due to 54: 644 - 6.
intravenous infusion of hair conditioner. Am ) Forensic Merl 3 II Gill JR, Goldfeder LB, Hirsch CS. Use of 'therapeutic
Pathol 1998; 19:252-4. complication' as a manner of death. ) Forensic Sci 2006;
297 Southall DP, Plunkett MC, Banks M et at. Covert video 51 :1127-33.
recordings of life-threatening child abuse: lessons for child 3 12 Roberts lSD, Gorodkin LM, Benbow EW What is a natural
protection. Pediatrics 1997; 100:735-60. cause of death? A survey of how coroners in England and
298 Feldman KW, Hickman RD. The central venous catheter as a Wales approach borderline cases.) Clin Pathol 2000;
source of medical chaos in Munchausen syndrome by proxy. 53 :367 - 73 .
) Perliatr Surg 1998; 33:623-7. 313 Buehler JW, Smith LF, Wallace EM et al. Unexplained deaths
299 Goldfarb J, Lawry KW, Steffen R, Sabella e. Infectious in a children'S hospital. An epidemiologic assessment. N
diseases presentations of Munchausen syndrome by proxy: El1gl) Med 1985; 313:211 -1 6.
case report and review of the literature. Clin Pediatr (Philo) 314 Istre GR, Gustafson TL, Baron RC et al. A mysterious cluster
1998; 37:179-85. of deaths and cardiopulmonary arrests in a pediatric
300 Kamerling LB, Black XA, Fiser RT. Munchausen syndrome by intensive care unit. N El1gl) Med 1985; 313:205-11.
proxy in the pediatric intensive care unit: an unusual 315 Hunt J, Goldring J. The case of Beverley AJlitt. IVIed Sci Law
mechanism. Perliarr Crit Care Med 2002; 3:305-7. 1997; 37:189-97 .
301 Meadow R. Munchausen syndrome by proxy. Arch Dis Child 316 Clothier e. The AliitT InquilY. London: HMSO, 1994.
1982; 57:92-8.
~~ --------_. - ------ ~
- -
I CHAPTER 20 I
ROAD TRAFFIC ACCIDENTS
IN CHILDREN
Anthony Busuttil
ROAD TRAFFIC FATALITIES Table 20.1 International comparisons: road accident fatality
rates per 70 6 papUlation year - 2000 (aged 0-14 yearsj4s
Road traffic deaths are still common, although rates among
countries vary considerably (Table 20.1). They deserve the Iceland 0 Ireland 27
same level of investigation as any other unexpected and Sweden 12 Australia 29
violent death; however, this does not always occur because Japan 13 Belgium 29
of prioritization of other fatal ities. In the thorough investi United Kingdom 15 Canada 30
gation of any road traffic death , particularly one that Italy 17 Spain 31
involves a child, a number of sources of inform ation must Germany 19 Czech Republic 32
be considered and integrated to enable a useful and faith Netherlands 19 France 33
ful reconstruction of the individual incident and, perhaps Austria 20 Greece 33
more importantly, the development of appropriate preven Finland 21 Luxembourg 37
ta tive measures. The detailed information obtained about Scotland 21 Poland 37
the collision from each potential source must be carefully Norway 22 USA 40
collated to ensure that the incident can be reconstructed in Switzerl and 23 New Zealand 47
detail - not only to assist the bereaved in coming to terms Denmark 26 Portuga l 52
with the death, but also for court inquiry (criminal and Hungary 26 Republic of Korea 58
civil) and insurance purposes. 1, 2 If the death can be shown
to have been preventable then steps should be taken to
ensure that similar deaths in t hat particular location or
manner could be avoided in the future. from a high-income family to be killed ,on the road . Chil
In the UK nearly two out of three road accidents happen dren from ethnic minorities are involved in up to twice as
when children are walking or playing and almost two many accidents as the national average while walking or
thirds of child accident victims are boys. As a child gets playing. Also, the risk of being involved in a road accident
old er, the risk of a road accident increases. A child from a when walking or pl ay ing is more than 10 times greater for
low-income family is five times more likely than a child a child with hearing difficulties. Child pedestrian deaths
Fatal road traffic co llis ions I 387
others are in use and no restraint is available. Children docum ent the scene fully by sti ll photographs and on video
must not sit in the front seat withou t the right child seat. to ensure that the findin gs at the site of the incident are
• In France, children under 10 years of age are not availabl e for futu re reference, briefing and reconstmction.
allowed to travel in the front seat of a car unless the It is often the case that the deceased has not been pho
child concerned is an infant of 9 months old or tographed in situ , as the attendance of ambulance person
younger and weighs less than 9 kg; however, the infant nel and paramedi cs usually results in the child being taken
must be placed in an appropriate baby seat (rear away to hospital for at least atte mpted resuscitation. Pub
facing) unless there is an air bag fitted, in which case lic access to the locus and ready visibility of the scene in
the infant is not allowed on the front seat at all. In the the interest of decency may also require the early removal
rear seats, a ll children under 10 years must be in the of the body from the scene of the collision .
right restraint; if the child's weight is between 9 and Specialists in road traffic collision investigations are
15 kg then he or she must use a child seat and if over nowadays available7 in most police forces. The scope of their
15 kg then a booster type seat should be used. activity is to document as much as possible at the scene, tak
• In Spain, children aged 3 or under must use a suitable ing measurements of such items as skid and tyre marks,8 oil
child seat, and child ren up to 12 should use one if and other fluid leaks and any other data that can be g leaned.
ava il able. If not they can use ad ult seats. They will log the location of di scarded items of clothing and
• In Italy, it is illegal for children under 3 years to sit in the shoes on the roadway, bicycles, broken portions and debris
front or rear seats without the proper child seat. Children fro m a vehicl e (for example paint fra gments,9 side mirrors,
aged betvveen 4 and 12 years cannot be a passenger lamps,1O glass fragme nts and pools of blood) in an attem pt
unl ess they are using a su itable safety restraint seat or an at a full reconstmction of the incident. They wi ll eventua lly
adaptor for a seat belt. If no seat is available for children produce a sketch of the incident and, if the incident warrants
aged between 4 and 12 years then they can use the ad ult it, a computerized reconstmction may be attempted. Valu
seat belts in the rear if accompanied, in the rear, by able information about such ma tters as the condition of the
another passenger aged over 16 years. road, street lighting and prevailin g weather conditions will
be collected and later included in a detailed report.
The design of cars and the restrain ts within them has An important part of the eventua l investigation, is the
exercised engineers and car designers for severa l decades, determination of the speed at which the vehicle{s) inlJollJed
with the emp hasis on design ing safer vehicles. Data col in the incident was trauel/ing. lI It has to be kept in mind
lected from actual incidents have assisted in ensuring that that the injuries on injured or deceased individuals invo lved
proposed designs are effective 4 6 in t he collision are of little or no help in determining this .
One of the main pointers to impact speed are the markings
left on the roadway by a vehicle attem pting emergency
INVESTIGATION OF A FATAL ROAD TRAFFIC
braking and the movement and transport of objects a long
COLLISION
the road. Any contact with walls, street furnitu re (such as
lamp- posts or signposts) and the defo rmation caused to
The inves tigation of a fatal road traffic collision has a vari
such items is more useful in the estimation of imp act speed.
ety of face ts, described as foll ows.
Similarly, any deformation in the actua l framework of vehi
cles involved can , after an assessment by a mechanical
The Scene of the Collision engineer and appropriate calculations, assist in answering
t his importa nt question. The pathologist or clinician should
Data collection should always start at the actual scene of be wary about being led into giving definitive answers to
the fate" road in cident. The deceased child may have been the question of speed estimates, i.e. responses that cannot
a passenger in a veh icle, a pedestrian or a pedal cyclist; be later substa ntiated scientifica lly on cross-examination.
more rarely, the deceased child may have been a pillion These investigators would also collect any trace evidence
rider on a motorcycl e, a passenger on a pedal bicycle or the that many be useful to identify the vehicle involved, particu
occupant of a sidecar of a motorcycle. Establishing this larly in 'hit and mn' accidents. Even a small fragment of pain t
information should be the point of departure of the inves from the offending vehicle can be analysed to enable such
tigation of the particular death. detailed information as to the make and year of manufacture
AJthough it may not be fe as ible for a forensic pathologist of the vehicle, any recent res prays of the bodywork, etc. 12
to attend at evelY such scene of a fata l road traffic collision ,
if the accident appears to be likely to resu lt in an eventual
criminal prosecution then it is imp ortant in such cases for Eyewitness Accounts
the pathologist to actually attend the scene. When the acci
dent is a 'hit and run inc ident' it is a lmost imperative that The second source of data are the accounts given by eye
the pathologist attends, together wit h t he rest of th e inves witnesses to the elJent. Th ey will be interviewed by police
tigative team. In each case, evelY effort sho uld be made to officers as early as possible after the incident and while
386 I Road accidents and ch il dren
constitute the most common accidental cause of child A chi ld aged from 3 to 12 years or under 135 cm (about
fata li ties in the UK. 4 feet 5 inches tall) must use the appropriate child restraint
The speed at which the vehicle is travelling has an essen in the front seat and, if there is one availab le, in the rear.
tial bearing on the number of pedestrian fatal ities and speed Children over 3 years old may wear an ordinary ca r seat belt
reduction measures that have been used on roads (e.g. speed in a taxi . If there are no seat belts in the rear of the car, by
bumps across the road and limiting the speed limit) have all law a child over 3 years old can travel unrestrained. Chil
been shown to benefit from a diminution of child pedestrian dren who are aged 12-13 years old, or younger children
fatalities. An impact speed below 20 mph results in signifi who are over 135 cm tall, must use the ordinary car seat
cantly fewer fatalities. The faster the traffic, the greater the belt, if available. Children aged 14 and older a re classed as
11sk of death and serious injury. When children are hit by cars adult passengers. They must use the ordinary car seat belt
travelling at 20mph, 5 out of 100 are killed; most suffer only and it is their own responsibility to do so.
minor injuries and about 30 out of 100 suffer no injury at all. Car seats a re so ld in stages appropriate to the chi ld 's
At 30 mph nearly half of all children are kjJled and many are age, and they range from rear-facing carders for bab ies up
seriously injured. At 40 mph 85 out of 100 children are killed. to the weight of 13 kg to forward-facing seats that are suit
Children make unpredictable pedestrians. Traffic-coping able from 9 months to 4 years. After this age, booster seats
skill s are complex and children do not have the ability to can be used until the child reaches the height of 135 cm.
judge speed, distance and danger accurately until around The new law states that it is illegal to install a rear-facing
the age of I I years. Even children above this age are easily child seat in the front of any car when the air bag feature
distracted and may not always behave as drivers expect. has not been disabled. Instead, it is recommended that
Young children are also much smaller than adults and so instead any rear-facing infant seat should be insta lled
are difficult for drivers to see. behind the driver on the rea r seat.
Children within vehicles who are sitting in the correct Frontal ai r bags and side air bags are fitted as sta ndard
child seat for their size and one that had been fitted properly practice in new cars within the UK and are designed to dou
usually suffer only minor injuries in a car crash, but arou nd ble up on the effectiveness of the traditiona l three-point seat
one-half of all child seats are still not properly fitted. Rear belts, whic h have been comp ulsory in new cars since 1967.
facing infa nt seats reduce the risk of fatal injury in a crash Containing 35 L of propellant, air bags inflate to full capac
by more than 70 per cent, fon-vard-facing toddler seats by ity within 25 milJiseconds, at a speed of 160 mph, which
more than 50 per cent and safety belts by 45 per cent. At one goes some way to explain why it is imperative to deactivate
time in the USA, only 10 per cent of children under the age any air bag if the child is to be carried in the front. In a crash,
of five travelled unrestrained, but they accounted fo r more the air bag infl ates very quickly. It could hit anything close
than one-half of child deaths in cars. It is important that to the dashboard with enough force to cause severe injuries
child restraints are used properly if they are to save young or even death. Because the back of a rear-facing child seat
lives. A study by the Automobile Association (AA) and sits very close to the dashboard , the seat could be struck with
Devon County Council found that 24 per cent of child car enough force to cause serious, or even fatal , injuries to a
seats were too loose, 21 per cent of the harn esses were too baby. Even older children (who have outgrown child seats)
loose and 21 per cent of the buckles were not fitted properly. are at risk from a deploying air bag if they are not properly
The law app lying to child car seats changed in September restrained with a lap/shou ld er belt. Reports of eight deaths of
2006 in the UK, meaning that all children under 14 years child passengers in crashes involving air bag deployment are
should have some form of extra restraint as appropriate to of special concern because they involved low-speed crashes
their weight and age while travelling in a car. Every year that the children otl1ervvise might have survived. 3
around 30 children aged up to I I years are killed during car Up to the age of 3 years it should always be observed
journeys, 400 are seriously injured and over 7000 more are that no child should travel unrestrained unless in the
injured; many of these deaths and iqjuries could have been exception of traveLling in a taxi or licensed car hire vehicle
prevented if the children had been strapped in correctly. in which no restraint is availab le.
Children who are not secured with the appropriate child The use of the co rrect car seat for the child's weight is
restraint, such as a chi ld seat, may be seriously injured in a essential, although it is a lso velY important to make sure
car accident. A ch ild should never be carried on an adult's that yo ur child 's seat is in sta ll ed properly and is the right
lap. Also, an adult sho uld never try to put an ad ult seat belt design for the particular car. It is important to note, how
around both him- or herself and a child; in the event of an ever, that the law varies in other co untries in rela t ion to
accident the child would be crushed by the adulfs weight. child restraints:
As from 18 September 2006 it became law that a child • In Genl1any, for example, children under 12 years of age
under the age of 3 years must have an appropri ate child and under 1.5 m in height must use an appropriate child
restraint when travelling in the front or rear seat. The only restraint (e.g. a child seat or booster seat while in the
exception to this is travel in a taxi - by law children aged front or rear of a car). Rear-facing seats cannot be used
under 3 years may travel unrestrained in a taxi because the in front seats that are equipped with air bags. Children
ordinary ('adult') seat belt in the car is not suitable for them. may travel in the rear seats without restraints if all the
- .....
388 I Road accide nts and children
events are still fresh in their minds; they may be inter problems with its components, seat belts, tyres, steering
viewed again at a later stage to enable the police to go over mechanism and other features relevant to safe vehicle opera
their initial statement with them once again, in more detail tion. A history of the vehicle may be usefully be sought
and in the light of results of other investigations. through maintenance garages.
It has to be remembered than these incidents are always The deformation and denting on the metallic body sur
the cause of major emotional trauma to the onlookers, and faces of a vehicle that have occurred in the course of the
thus it may often be useful to wait until they have had a time fatal incident can be extremely usefu l in determining, with
to compose themselves before such statements are taken; at some degree of accuracy, the speed at impact. Mechanical
the scene all that needs to be done is to obtain names and engineers are able to examin e the vehicle and determine
addresses. Great sensitivity has to be exercised by the inves by mathematical calculation, often assisted by computer
tigating police officers in the timing of taking statements, aided programs , the speed at impact.
particularly from next-of-kin eyewitnesses. On the other The vehicle will also bear in it dents and breakages at
hand, delayed attempted rationalization of the perhaps jum sites of direct impact with the human body. If a child pedes
bled recollections of the rapid incident may result in prob trian has been knocked down, an inspection of the vehicle
lems in distinguishing fact from rationalization. may well indicate where the initial contact with the child's
In this aspect of the data collection , it is always useful body was; this co uld be the front bumper or bonnet of the
to interview the paramedics , police officers and sometim es car. This inspection may be able to determine whether the
also fire officers who are called to the scene. The informa child has been elevated after the initial impact and pro
tion that they may have picked up in the course of their pelled upwards to strike the car still moving in his direction
scene atte ndance, i.e. their involvemen t afte r the collision, or wheth er the child had fallen beneath the car.
the appearances of the locus of the collision and the posi Impacts with pedal cyclists often result in the child being
tion of the body when found, is particul arly useful. They thrown up in the air if the combined speed of impact is over
may have had to move vehicles or the body of the child or 16-20 mph (25-32 km per hour). The body will often land
to extricate the child 's body from within or under the vehi on the vehicle with second imp act: for example a primalY
cle. It is then most important to know exa ctly the position impact may be with the windscree n and with the secondary
and appearances of the body prior to its movement. impact with the A-pillars or windscreen of the car. At such
In this respect, if any resuscitation has been carried out sites of impact, there may have been transfer of hair, blood
then it is also essential to interview the ambulance crew, and tissu e from the cyclist.
paramedics and the members of the nursing and medical staff When the deceased child has been an occupant of a
who have been directly involved. This will enable them to vehicle, it is important to determine where he or she was
give information as to the methods of resuscitation used, any placed inside the vehicle, whether the child h ad been prop
drugs administered, and any other changes effected by them. erly restrained and wheth er the restraints available in the
vehicle were appropriate, properly fitted and functional,
and whether there were any other occupants or other items
Clothing of the Decedent inside the vehicle that might have resulted in fuliher
impact and injury.
The clothing of the injured parties may also have useful trace
evidence 13 on it, as well as features that indicate that certain
matters had taken place - for example, crushing under a Information from Others Involved
wheel, contact with the greasy undercarriage of the vehicle,
etc. If the child was taken to hospital, it is often the case that FUliher useful data sources are the narratives from other
clothing is cut off with scissors and sometimes has been persons invoJved in the co llision, not least the driver of any
completely disposed of before the police have had a chance vehicle involved. In such instances, in addition to docu
to claim it. It is essential that such items of clothing be men ting any injuries which they may have sustained, it is
retrieved and that those responsible for clothing removal are also impoliant to determine the driver's state of health and
instructed to be aware of its potential value as evidence. whether or not they were incapacitated in any way, or
intoxicated, at the time th at the collision took place.
In all road traffic collisions the drivers of any vehicle
Inspection of the Vehicles Involved will invariably be breatha lysed for alcohol at the roadside
and a general observation made of them by police officers
Another accessible source of data in fatal road incidents is the present as to their state of orientation\ coordination, pupil
vehicle (or veh icles) involved in the incident. If a mecha nically Jary size and reaction. If the situation warrants it, statute
propelled vehicl e is involved it is essential that exp erts care provides for their medical examination a nd the collection
fully check its roadworthiness, including the brakes. The of further samples for the estimation of alcohol and drugs
vehicle is usually towed or taken away to a police-manned in body fluids. More detailed medical examinations and the
garage where it is fully inspected and examined in terms of collections of specimens for drug testing wi II be carried out
_. - - - - - - - -
------- -
Fatal road traffic collisions I 389
at the police station rather than at the scene, or in hospital A full autopsy must be carried out and all internal
if the driver has been injured. injuries documented fully being careful to examine closely
the vetiebral column, the rib cage in its entirety and the
scapulae - fractures at these sites are often missed.
The Decedent If physical abnormalities are present then these should
be carefully recorded. It is also important to check that the
The main source of evidence accessible to the pathologist is child that has not suffered from any previous significant
the body of the deceased. If the child has not been unclothed, medical histolY by look.ing through the child's general
it would be essential that the clothing is removed carefully practitioner records; if the family doctor is not known then
to avoid any further damage to it and to avoid any loss of this should be checked through the police. If such a history
evidence from it as well as any cross-contamination. exists, it is important to give due weight to any relevant
The clothing, including the shoes, should be separately information. If the child suffered from learning difficulties,
bagged into brown paper bags and given to the police for epilepsy, visual impairment or hearing problems then these
further examination as required. may all be relevant; with regard to a history of hearing
The child's external injuries, no matter how trivial, problems the middle ears have to be opened and conditions
should be carefully measured and documented. It is useful such as otitis media and 'glue ear' have to be excluded.
to photograph all aspects of the body. This may not always The autopsy should enable a full documentation of
be feasible and, in such instances, the importance of a care all injuries, externally and internally. If any injuries show
ful description of these injuries is crucial. Injuries with pat any patterns in them (e.g. tyre mark imprints) then these
terns to them must be documented fully photographically. have to be documented fully by photographs and accurate
This may have to be followed, depending on local legal measurements.
custom, by the formal identification of the decedent by Any paint or glass fragments or any other trace evi
members of his family and by police officers who were actu dence should be collected, particularly if there is some
ally present at the scene. Even if there is no legal require dubiety as to the vehicle involved .
ment, it is essential than the parents are aJlowed early access Even in autopsies on children, particularly in the late
to their child after they have been fully briefed about the teens, it is important to collect the following samples for
appearances of their child by those who have been trained to further analysis:
so do, for example police family liaison officers. 1. Blood for DNA profiling, particularly in 'hit and run'
It is essential that the body of the child is presented to accidents.
them in an aesthetic way, perhaps with partial covering or 2. In appropriate cases, it may be useful to exclude acute
bandaging to obscure injuries or even following some clean inhalation of solvents prior to the collision. In addition
ing and reconstruction of the body. The identification to the collection of blood, this may require the
process is fraught with emotional overlay for those who are retention of a lung, with its main bronchus tied off
doing this in an official capacity, and indeed much more securely and placed inside a nylon fibre bag to avoid
poignantly so to the members of the family of the deceased. loss of intrapulmonary air content. This enables assay
The latter should be given enough time with the body of the solvent content within the bronchial air.
and not officiously rushed. They should not be discouraged 3. In some cases, it may be necessary to test for alcohol
from touching the body and if, indeed, the family asks for and other drugs. For these assays, appropriate
mementoes, such a locks of hair or handprints, then these specimens of blood, urine, vitreous fluid and liver will
requests should be met, with the consent of the appropriate have to be retained on the off chance that these
legal personnel. Some families may wish to obtain a Polaroid become necessaty. In the older teens, it is always wise
photograph of the face of the deceased child. Religious ritual to consider intoxication as a potential complicating
may also be requested as part of the acute bereavement pro factor in road traffic collisions.
cedure. All reasonable requests made by the family should be 4. (Control) samples of plucked scalp hair may also come
heard and, if at all possible, agreed. in useful for matching with any hairs found at the
It is extremely useful to document bony injuries using scene and to exclude any previous exposure to
radiology before the autopsy commences. This ensures a controlled drugs.
much fuller documentation of such injuries, particularly of
the ribs and vertebrae and in younger children in whom Recent problems with organ retention from necropsies
ossification may be incomplete. may make it difficult for the brain to be retained for neuro
Another useful investigative adjunct, which is now avail pathological examination. In cases whe,n the death is very
able to most police forces, is the availability of a portable laser acute, retention of the brain has only limited value in terms
source of light that enables the child's unclothed body to be of documenting injury. In delayed deaths, and in cases where
scanned in a dark room; this shows any marks on the body that there is known to be a pre-existing cerebral problem (e.g.
have been caused by tyre imprints or patterned injuries that epilepsy, hyperactivity syndromes, autism, etc.) it may be
have resulted from contact with specific parts of the vehicle. necessary to suspend and fix the brain for an appropriate
--- -. - ~ -
~ ~
390 I Road accidents and children
period in form alin and to examine it as soon as possible girJ. 1s However, such incidents still occur and they should
afterwards. Diffuse axonal injury does occur in babies and warrant the same amount of investigation that any other
children. If the child has surv ived long enough after the acci sudden suspicious or accidental death is given in the medico
dent, this pathological damage is discernible with appropri legal context. Unfortunately, this is not always the case and
ate special staining techniques an d microscopy. important information of potential preventative and safety
In deaths in the intens ive therapy unit, bacteriological improvement importance may remain uncollected.
samples may be required to exclude nosocomial infections Car occupants are the largest category of road users
that may have contributed to the death. This may be of who sustain injulies if they becom e involved in collisions,
importance if civil litigation is in progress in terms of a although fatalities are more common amongst pedestrians
novus actus interveniens and the apportioning of blame for (Table 20.3). This has been the reason why large amounts of
the death. research, time, money and effort have been spent in ensur
In some deaths, with the consent of the family and of the ing that these injuries are prevented by improvement in car
legal authorities, organs have been harvested for tra nsplanta design, stre ngthening the fra mework of vehicles, the intro
tion purposes. The current dearth of organs compared with duction of internal safety devices - seat belts, air bags - and
the demands of patients, combined with the current major through legislation improv ing road worth iness of cars and
successes of transplanta tion, has made this phenomenon general safety of cars.
more common than it once was. The contlibution of the Car collisions cannot be rationalized simplistically from
pathologist in such cases is more limited given that it is an isolated assessment of the forces invol ved; in what
assumed that all the organs removed had been intact, healthy appear to be entirely simil ar incidents, one car occupant may
and functional. In such instances to complete the autopsy survive with little injury, whereas the other may experience
examination, it is essential to speak with and to obtain a fatality. The physical size, injury tolerance and the age of the
report from tbe transplant coordinator as to by whom, wben occupant are all important in the outcome of the collision.
and wh ich organs were removed and whether these have Thus the different physical build of children is an importan t
functioned adequately in their seco nd hosts. If it is considered consideration in predicting the injuries that they will sustain
that some organs or tissues may be required for other pur in road traffic collisions (Table 20.4). The bead of a youn g
poses it is essential that fully informed consent is obtained child forms a higher proportion of the overall body mass
from the family and that tbis matter is approached in as sen than that of an older child or an adu lt. The neck of a child
sitive and measured a manner as is possible. cannot support the weigh t of the head as adequately as in an
adult. The lib cage is smaller and more flexible a nd pliable,
and leaves a larger unprotected area of the abdomen; the
VEHICULAR COLLISIONS child's pelvis is not as resilient as that of an adult and its cur
vature is different. All of these factors will have a bearing on
The data in Table 20.2 show that, by and large there has been the kinematics in the collision situation and the possible
a major welcome decrease in fatal and serious road incidents types of injury that are sustained and their outcome.
involving children. Unfortunately, in relation to less selious Most collisions are frontal, with a tendency for the offside
non-fatal transportation injuries there is serious underreport fron t of the vehicle to become involved. In side impacts, t he
ing of such collision-induced injuJies, particularly if the door of the vehicle on the side of the collision remains sup
injuJies are minor. One estimate placed this at 25 per cent of ported by the striking object dUling the impact phase and
all child pedestrian casualties l4 and another at 20 per cent acts as an unyi elding barrier that is pushed inwards into the
with injuJies in boys being less frequently reported than in vehicl e's compartment and against which the occupant may
Table 20.2 Child fatalities and casualties in Scotland, England and Wales J997-2005
Child casualties Fatal Fatal and serious All severities Fatal Fatal and serious All severities
-- -
Supervening problems in collisions I 391
Table 20.3 Average number of road traffic fatalities per year for Obviously, the mass ratio between the two vehicles
1997-2001 in Scotland involved and the velocity change that occurs on impact are
further critical factors, and these are crucial elements that
Deaths Age (years) Total
have to be considered in assessing the force of the impact
0-4 5-11 12-15 and the injuries caused therein. If the struck object is station
ary and unyielding then the velocity change in terms of
acute deceleration will take place in the impact vehicle only.
Pedestrians 2 (0.01) 8 (0.02) 6 (0.02) 16
In terms of injury, it is self-evident that the reduction and
Driver or rider 1 (0.00) 1 (0.00) 1 (0.00) 3
restriction of movement of the occupants of a vehicle would
Passenger vehicle! 1 (0.00) 3 (0.01) 3 (0.01) 7
tend to decrease or remove propulsion and thus secondary
pillion
impact against the interior of the vehicle. The rational e of a
Figures in parentheses indicate rates per 1000 of the population. seat belt, which is fitted properly and functioning well when
required , is to enable the wearer to 'ride down' the impact
when the rapid deceleration occurs; thereby the energy of
the impa ct is extended over a slightly longer timescale;
Table 20.4 Site affatal injuries in child car occupants by age the longer this time frame, the greater the dissipation of the
Site of injury Head Neck Chest Abdomen Other forces at the time of the impact. The webbing stretch of the
(no.) belt enables this deceleration rate and interval to be length
ened over a somewhat longer period , thus allowing the
energy of the impact on transportation casualties to dissipate
0-9 months (4) 4 0 0 0 0 plior to contact of the body with an unyielding surface. An
10 months to 19 2 4 3 evaluation of the impact of front seat belt legislation in
6 years (29) Britain showed a marked reduction in injuries to children
7-14 years (33) 20 4 4 4 aged II to J 4 years who were restrained in the front seats of
Total (66) 43 6 8 2 7 vehicles when collisions took place. 20
'Most of the children were unrestrained and about one-third of them If the child has been restrained inside a vehicle, there may
were ejected from the ve hicle. be injuries on the body that indicate that this restraint has
Modified from Andersson and Jonansson 4 6 been in position and functi oned: friction-induced abrasions,
bruising and even fractures - of the clavicle, sternum and
ribs 21 - may follow the line along which a seat belt would
sustain injuries. Collisions with trees, lamp-posts and other have been applied with force at the time of impact. Internal
street furniture, and walls produce a more concentrated load, damage to the body such as intra-abdominal injuries 12,22,23
with the possibility of an increased intrusion into the (including ruptures of the liver and mesentery and major
passenger compartment and thus a higher risk of serious blood vessels 24 ) and intrathoracic injuri es (such as pul
and even fatal injuries being sustained, particularly trunk and monary and cardiac injuries) may indeed be associated with
pelvic injuries in front-seat passengers as well as head and such external injuries; fatalities due to seat belt use have
neck injuries. In addition to the initial direction of the been recorded in children involved in transpoltation injuries.
principal force at the time of the collision, there will also be In telms of these considerations, given the relative mass of
post-impact vehicle rebound, depending of the type and pedestrian to a vehicle, particularly a high-sided vehicle, the
magnitude of the force of the collision and the part of the car relatively greater speed of the vehicle than of the pedestrian,
that has been struck. This also depends on the size, resilience the decreased resilience of the human body as contrasted
and safety features of the vehicle involved. These phenom with metal, and the Jack of protection in the pedestrian are all
ena will introduce a sideways and lateral component to the factors that render such collisions very traum atic, even at low
impact in the immediate post-impact phase, which, in tum, speed. 2s
will have a further direct bearing on both the configuration
of the impact and the contact between the body and other
parts of the inside of the vehicle as a seco ndary impact phe OTHER SUPERVENING PROBLEMS
nomenon. Such an understanding of the kinetics of a colli IN COLLISIONS
sion has led to improvement in the inner padding in cars and
the removal from the inside of the vehicle of any pointed Occasionally, the occupant of the car who had been poorly
surfaces and edges. Hyperextension of the neck, secondary restrained inside it is thrown out of the ,car and injuries are
to forehead and facial impacts, may result in serious neck sustained on external impact. If the car catches fire with the
and brainstem injuries that could be instantaneously occupants still inside then secondary fire-induced injuries
fatal; 16, 17 in severe intrathoracic injuries in particular there and inhalation of smoke will occur. These changes will result
may be traumatic rupture of the aorta 18 and compressive in a variety of other injuries to the child involved in such an
injuries to the heart. 19 incident.
392 I Road accidents and children
PEDESTRIAN INJURIES
child to fall into the path of the oncoming veh icle, and an
impact below the centre of gravity would cause the child to
If the child is a pedestrian then the primary impact is usually be lifted up into the air, then, as the vehicle proceeds, on to
by the front of the car with one or other side of the body that the vehicle, and after impact with the vehicle to fall behind
happens to be closer to the vehicle, particul arly if the child is the vehicle. If the child is on a bicycle then the centre of
in the process of walking or mnning in front of the vehicle. gravity at the time of the collision has been so displaced
The distribution of the casualties and fatalities in these inci further up (and this is so the larger the bicycle) that on
dents are examined further in Tables 20.5 and 20.6. impact the child will be thrown over the vehicle. From an
Depending on the child's height above the ground, dif examin ation of that body and the veh icle, the points of pri
ferent varieties of impact are present. This will determine mary and secondary impact may be worked OUt. 26
whether the prim ary impact with the vehicle is above or The size of the vehicle involved , the age and build of the
below the centre of gravity of the child's body and in this child, projecting parts of the front part of the veh icle and
the kinetics of the child's body after the collision takes the speed at impact are all factors that have a bearing on
place, vary. By and large, the umbilicus is at about the level the severity of the injuries sustained by the child. Even
for the centre of gravity. A collision producing a primary minor impacts at low speed may produce severe or fatal
point of con ta ct above the centre of gravity will cause the head and neck injuries in a young childY
Most pedestrian childhood fa tali ties (about 80 per cent)
occur in built-up areas with in a I-km radius of the child's
Table 20.5 Casualties by speed limit at site of incident, mode of home ; a number of such fatalities occur on the way to or
transport and severity, 7997-2007 (Scotland) from school so that most deaths occur at the times of
school opening and closure during weekdays. Injuries are
Speed limit (mph) more prevalent during the summer months when children
are more likely to be outside their homes.
30 40 50 60 70 Other and
The psychological capabilities and developmental level of
unspecified
the child have a major bearing on their abili ty to cope with
road conditions; they have generally less well-developed
Fatal ski lls relating the perception of movement and velocity, the
Pedestrian 53 5 2 15 9 0 ability to judge distance and depth, the abi li ty to OIientate
Pedal cycle 4 3 2 0 themselves in relation to the source of sound and the ability
Motorcycle 6 0 0 30 2 0 to concentrate for a sufficient length of time, putting aside
Car 22 7 146 21 0 play activities and concentrating while croSSing or using
Bus/coach 0 0 0 0 0 1 roads. They are also not equipped with the ability to inte
Other 2 0 12 3 0 grate all of this information swiftly and efficiently, and come
to the appropriate evasive conclusions.
Serious
Pedestrian 893 37 8 49 12 2
Peda I cycle 141 6 29 2
Motorcycle
CHILD CYCLISTS
114 10 7 190 2
Car 482 73 23 1187 175 1
These accoun t for about 17 per cent of deaths on the roads.
Bus/coach 54 2 0 1 0
Arguments have raged for the last decade or so as to the
Other 52 9 5 102 33 0
usefulness of cycle helmets for pedal cyclists and the design
Table 20.6 Child pedestrian (0-15 years) casualties in single-vehicle accident: mean per annum, 1997- 2007 (Scotland)
On pedestrian Within zigzag area Within SOm Crossi ng road Other Total
crossing of a pedestrian of a pedestrian in other areas
crossing crossing
of such protective head gear. 28 There can be little doubt that extended head and neck are thrown backwards until they
most pedal cyclists die because of head injuries 29 a nd that hit the headrest with the consequence that the soft tissues
helmets will afford some degree of protection, particularly of the neck are stretched and torn. This is also known as
in collisions where little force has been expended. hyperextension followed by hyperflexion, and it occurs in
In this vein, a number of individual states and countries a frontal collision when the neck and head continue to
have suggested that legislation may be required on this move forward while the body remains relatively still. 35
matter in concert with legislation concerning the use of seat Rear-end collisions can result from any number of sce
belts. In Victoria, Australia, following on from a massive narios. For example, the driver in front may suddenly brake
10-year-long educational campaign, legislation was intro to avoid a cyclist that pulls out in front of him or the car
duced ensuring that all cyclists wore helmets. 30 ,31 The behind may accelerate more than the leading car at a round
helmet-wearing rate tripled and a 70 per cent reduction in about. The impact of hitting another car is the equivalent of
head injury admissions occurred during the periods hitting a firm surface, such as a wall, at half the speed. Th at
1989-90 and 1991-92. There was also a corresponding is, rear-ending a stationary car while travelling at 30 mph
decrease in non-head injuries of 23 per cent in the first year has the equivalent force of hitting a wall at 15 mph. Overall,
and 28 per cent over the first 2 years. 43 per cent of people who sustain whiplash injuries from car
From a comparison of statistics between countries where accidents are involved in rear-end collisions. Side collisions
cycling is very popular (The Netherlands, Belgium) and those involve the side of one or more cars being impacted. These
where it is a less frequent activity and more of a leisure are common in car parks and at T-junctions.
pursuit, the incidence of injuries is lower in the former. This Personal injuries suffered as a result of side collisions
has been extrapolated to the fact that in the Low Countlies are determined by the part of the car that is struck, the
cyclists and motorists are segregated and cyclists have their speed of the vehicles involved, the presence of safety fea
own cycle paths and do not mix with other vehicles on busy tures such as side-impact air bags, and the weight and con
roads, thereby decreasing the likelihood of collisions. struction of the object that strikes. In total, 35 per cent of
people who have been involved in car accidents and sus
tained a whiplash injury were involved in side collisions.
DIFFUSE AXONAL INJURY A head-on collision involves the front ends of two cars
crashing into one another. This type of collision may result
Brain trauma produces a diverse spectrum of injuries in
from a dIiver coming round a bend on a narrow country lane
which diffuse axonal injury (OAl) is a well-recognized
too quickly or a motOlist swelving into oncoming traffic on a
major contributing factor to long-term disability. Magnetic
B-road to avoid an animal. Head-on collisions are often fatal
resona nce imaging and computerized tomography have
but only 32 per cent of people who receive whiplash injuries
long been thought to cause underestimation of the extent from car accidents sustain them from head-on collisions.
of OAl because of their lack of sensitivity and therefore
newer and more sensitive imaging techniques have been
developed. Using these techniques with magnetic reso INJURIES TO CHILDREN IN UTERO
nance spectroscopic imaging, an increased rate of detec
tion was shown in children in whom at one stage it was Road incidents may involve pregnant mothers who are
thought that OAI did not OCCUr. 32 ,33 passengers, drivers and, less frequently, pedestrians. 36 ,3 7
Injuries to the pelvis and abdomen may result in severe
damage to the uterus, rupture of the uterus and abruption
WHIPLASH INJURIES 34
of the placental site with dire, often fatal, consequences to
the unborn child. Litigation may focus on whether or not
Generally, very young people and elderly people are those
the road incident has directly caused the death of the fetus
less at risk of sustaining whiplash injury from a car acci
or unborn infant. 38 ,39 The potential contribution of the seat
dent. Research suggests that 17 per cent of under-21-year
belt to these injuries may have to be considered 40 - 42 and
olds and just 2 per cent of over-60-year-olds are less likely
indeed pregnant mothers are legally exempt from wearing
to be involved in an accident resulting in whiplash than in
a seat belt for this reason. This topic is considered in more
any other type of motoring accident. These results have
detail in Chapter 10.
come to light despite the fact that younger people, espe
cially small children, have far smaller and less well
developed neck muscles. The action of a car coming to a OTHER VEHICULAR ACCIDENTS
sudden halt usually causes the whiplash motion to occur.
The impact of a rear-end collision (when one vehicle hits Childhood accidents on the farm,4J,44 often involving vehicles
the rear of the vehicle in front) shunts the car forwards, or machinery, are not uncommon, particularly in rural envir
forcing the body forward with it while the h ead and neck onments; they may require intervention from governmental
are thrown back. The head then tilts downwards towards agencies involved in sentimental and work-reared health and
the steering wheel and the neck extends forwards. The safety issues, as well as requiring public inquest.
394 I Road accidents and children
REFERENCES 24 Riches KR. James RA, Gilbert JD , Byard RW. Fatal childhood
vascular injuries associated with seat belt usage. Am)
Forensic Med Pat/lOi200]; 22: 193 - 5.
Basham DJ. TraJfic Account i\lIanagement. Springfield, IL:
INTRODUCTION
numbered 1-22 in decreasing order of size, and one sex
chromosome, in this case an X chromosome. The father's
For centuries philosophers have deliberated which came spenn contributes another 23 autosomes, again numbered
first, the chicken or the egg. In the context of forensic 1-22, plus either an X or a Y chro mosome, which determines
paediatrics, the egg - or rather t he fertilized egg - bein g the the sex of the offspring.] When an ovum fuses with a spenn
precursor and genesis of the livin g child, is the appropriate car!)'ing an X chromosom e, resulting in two X chromosomes
place to start off the discussion on the role of genetics. It is, in the fertilized egg, the conceived child is female, denoted
after all, the fertili zed egg that develops into all 100 trillion XX. When an egg ce ll fuses with a sperm carrying a Y chro
cells that make up the human being, and it is the genetic mosome the offspring is male, denoted XY.4 The chromoso
information that comes together from mother and father at mal complement of the fertilized egg and of the offspring is,
the point of fertili za tion that provides the instructions for therefore, 46 (Fig. 21. 1), which makes up the individual's
the creation and development of the child.] While the result genome. Figure 21. 1 shows these chromosomes as a human
is obviously of biological relevance, it is the analysis of the male ka!)'otyp e in which the 22 pairs of autosomes are
individuality of the inh erited complement of genetic mate neatly aligned in descending order of size, followed by the X
rial from person to person that has proved of significan t and smaller Y sex chromosomes. In a femal e ka!)'otype, the
value in a forensic context. In a forensic investigati on, Y chromosome is replaced by a second X chromosome.
genetic evidence may be used to confirm or refute a known Because it is the male that carries the Y chromosome in
individual as being the likely so urce of trace material or to humans, the presence of Y chromosome DNA in a genetic
identify a body or body remains. This chapter will focus on sample indicates that a male has contributed to the sample,
the above use of DNA in paternity testing. 2 and thus, this is a useful way of determining the sex of ori
gin of a sample.
At the point of fertilization it is not possible to predict
INHERITANCE OF GENETIC MATERIAL whether the sperm cell that fuses with an egg cell carries an
X or a Y chromosome and, hence, which sex the offspring
Humans inherit half of their genetic material from each bio will be; likewise, one cannot predict which one of the par
logical parent in the fo nn of deoxyribonu cleic acid (DNA) ents' two copies of each of the other 22 chromosome pairs
molecules neatly packaged up into chromosomes. The will end up in the egg and sperm cells (also known as
mother's egg contrib utes 23 chromosomes, 22 autosomes gametes or germ cells) that combine. The actual combination
396 I Forensic DNA profiling
2 3 4 5
6 8 9 10 11 12
13 14 15 16 17 18
19 20 21 22 x y
Figure 21.1 Karyotype of a normal human male showing chromosomes in a condensed state. Chromosomes were stained, photographed,
then cut out, paired with their partner and the pairs aligned in descending order of size. The sex chromosomes, the Xand smaller Y
chromosome, are positioned at the end of the pairs of autosomes. (Courtesy of RMurray, South East Scotland Cytogenetics Service.)
is the result of the parental chromosome pairs randomly and us in terms of the diversity of features and characteristics
independently segregating during gamete cell production by that we all possess. It gives us an understanding of why we
the process of meiosis.4 It is this randomness, in part, that feel justified in saying that, with the exception of identical
gives rise to the diversity of genetic material from individual (monozygotic) twins, it is highly probable, thoug h not real
to individual. Because which of the two possibl e copies of istically feasible, to prove definitively that every individual
each of the 23 chromosomes of the parent that enters the egg has a unique combination of chromosomes and a unique
or sperm cell is random, there are 223 possible combinations genetic sequence.
of the mother's chromosomes in a human egg cell and, like Monozygotic twins are the exception, because they are
wise, 223 possible combinations of the father's chromosom es derived from a single fertilization event when the embryo
in a hum an sperm cell. Therefore, when a human egg cell splits to form two individuals at an early stage of develop
and sperm cell randomly fuse, the result is one of 223 x 223 ment. Thus, monozygotic twins have the same chromoso
possible different combinations of the parents' original chro mal and DNA complement. This precludes existing forensic
mosomes, which equates to one of over 713 possible combi DNA analysis techniques from discriminating between their
nations of the parental chromosomes in the resultant DNA, and is a factor that should not be overlooked when
offspring. This helps to explain why siblings, despite being claims of the power of DNA evidence are made. Conven
deJived from the same parents and inheJiting half of each of tional forensic fingerprint evidence is increasingly coming
their parents' genetic mateJial, have a different combi nation under attack for its alleged lack of scientific basis and cred
of chromosomes and a different genetic make-up. ibility an d question ab le reliability, as exemplified by recent
In actual fact there is a further level of genetic variation. high-profile cases of misidentification, such as the Madrid
This arises because a process aki n to chromosomal shuf bombing and the Shirley McKie case. 5 - 11 One feature of
fling takes place between pairs of chromosomes, known as conventional fingerprinting that is often cited in its favour
homologous chromosomes, such that the members of a pair is its abi lity to differentiate between identical twins. How
swap segments of DNA to give rise to novel sequences ever, such cases are equally, if not more, susceptible to the
along the length of the chromosome compared with the assumptions and un certainties of the technique. 12
original chromosomes. In this way, one can appreciate the Nearly all body cells of a human possess the full comple
biological mechanisms that contribute to the interindividual ment of 46 chromosomes and, as such, are said to be in the
genetic variation, the result of which we can see all around diploid state. The chromosomes are comp artmentalized into
Inheritance of genetic material I 397
an orga nelle of the cell known as the nucleus. Any cells with
a nucleus are, therefore, a potential source of this nuclear
DNA, which can be used in an investigation. Notabl e excep Vt=:::ILC:;;:;~_ Sugar
tions include the egg and sperm germ cells previously men phosphate
tioned, which only have half the number of chromosomes backbone
and are said to exist in the haploid state, and mature red
blood cells (erythrocytes), which are anucleate and lack
~~~~~~'- Adenine
Deoxyribonucleic Acid
~=TIt:::~~i\- Nitrogenous
base
While morphologically the members of a homologous
chromosome pair in any individu al look the same (see Fig.
L....l_ _ _ _ Base pair
21.1) and possess a high degree of similarity across their
length, they are not identical. This is due to the specific Figure 21.2 Structure of deoxyribonucleic acid (DNA). This
make-up of the DNA that is transmitted by way of the figure represents the double-stranded he lica l structure of DNA.
chromosomes. DNA lies at the heart of genetic analysis, The molecule comprises two strands of linked sugar and
and in order to appreciate the techniques employed and the phosphate molecules. The sugar-phosp hate backbone is held
evidence derived from forensic DNA analysis a basic level together by nitrogenous bases forming pairs between adenine
of understanding of the molecule is required. and thymine or cytosine and thymine. These linked strands
DNA is made up of three components: a sugar part then twist to form a helical spiral as shown. (Courtesy of
(deoxyrib ose), a phosphate part and a nitrogenous base Talking Glossary of Genetic Terms, National Human Genome
(Fig. 21.2). The sugar and phosphate components are joined Research Institute, National Institute of Health, USA:
by phosphodiester bonds to form the sugar-p hosphate www.genome.gov/10002096.)
backbone of the molecule, but it is the bases that link two
of these strands together, in what has been likened to rungs
of a lad der bridging two strands of linked sugar-phosphate
units, that are of interest in terms of conferring attributes the genome called genes, which provide the instructions to
and individuality on living beings. make specific proteins that build and run our body, for
There are four bases, adenine, cytosin e, guanine and example the pigments that give hair its natural colour, the
thymine, commo nly referred to as A, C, G and T. These enzymes that digest our food and metabolize ingested
bases hold the two strands of DNA together in a very spe drugs, the ion channels found in cardiac cells, or compo
cific way: A can pair only with T (and vice versa) and C can nents of the immune system. The last two examples may be
pair only with G (and vice versa). When correctly paired , of paliicular interest in the investigation of sudden infant
the two complementary strands can spi ral into the recog death syndrome (SmS) as particular variants have been
nizable double-helix form that is so familiar (Fig. 21.2). implicated as putative contributors to a subset of SIDS
It is the specific sequence of the bases along the length deaths.13
of the chromosomes that may vary between members of a Di fferent variations of a paliicul ar gene's sequence,
homologous chromosome pair in an individual and between known as alleles, may typically take the form of a change in
chromosomes from different individuals. It is this sequence one or more base to an alternative base(s), the insertion of
variation that differentiates us at the genetic level and which extra bases or deletion of sections of DNA. These differences
may be used as a means of tracing a DNA sample to its source. mayor may not alter the fun ction or production of the pro
tein product. For example, the different colours of eyes of
different people result from different alleles of the genes that
Genes. Alleles and Genetic Variation determine eye colo ur. 14 Whether a child suffers from the
genetically inherited condition cystic fibrosis depends on
Some of these vari ations may confer phenotypic effects, i.e. which combination of allelic variants of the gene encoding
characteristics that can be observed. These lie in regions of the cystic fibrosis transmembrane conductance regulator
398 I Forensic DNA profiling
Restriction Fragment Length Polymorphisms (Fig. 21.3). Because of the analogy of comparing bands on
an electrophoresis DNA gel with points of comparison of
From a historical perspective, the advent of forensic DNA conventional fingerprints, Jeffreys named the technique
analysis as a means to identify, visualize and trace the 'DNA fingerprinting'. He claims it was coining this term that
interindividuality of genetic material, enabling a sample attracted attention to the potential forensic application of
to be matched to its source, was somewhat serendipitous. the technique. Using RFLP analysis, it is possible to compare
Alec Jeffreys, dubbed the 'father of DNA analysis', was not DNA extracted from biological samples from a crime scene
aiming to develop a forensic investigative tool when he with those of a suspect; if the DNA patterns match, the sam
discovered a potentially groundbreaking way of visualiz ples most probably came from the same individual. If the
ing individual genetic specificity in the 1980s; in fact, he DNA patterns do not show the same pattern of bands, the
was looking for a human version of the myoglobin gene in suspect can be eliminated as a likely source of the crime
seal meat. 21 scene sample. This was, in fact, the outcome of the first
Jeffreys used a technique called restriction fragment application of the technique in a forensic case.
length polymorphism (RFLP) analysis in which genomic Two 15-year-old girls in Narborough, Leicestershire,
DNA is cut or digested in a sequence-specific way by UK, were sexually assaulted and murdered. RFLP analysis
restriction enzymes; this results in a mass of fragments of showed that DNA extracted from semen from both crime
DNA of differing lengths. These fragments can then be sep scenes came from a common source, and thus a single man
arated on the basis of their size and electric charge by a was deemed to have been responsible for both crimes. How
process called gel electrophoresis; specific fragments that ever, the crime scene DNA did not match DNA from a blood
share similarity in sequence are visualized by binding a sample taken from the prime suspect. Subsequently, the first
labelled DNA probe of complementary sequence to the mass DNA intelligence screen (DNA man hunt) was
sought-after sequence, which is shared by a subset of the launched. However, in an interesting twist in this investiga
fragments. 22 ,23 The result is somewhat akin to a 'barcode', tion, the DNA screen failed to identify a match to the crime
with DNA from different individuals forming different-sized scene profiles. The offender, Colin Pitchfork, was eventually
fragments that are represented by bands positioned at a dif identified and convicted after it was overhead in a pub that
ferent location along the length of the developed 'barcode' a colleague had been persuaded to give a DNA sample on
Forensic DNA ana lysis I 399
his behalf. Subsequently, Pitchfork's DNA was found to individual's DNA. An individual who has an SIR, for exam
share the same RFLP pattern as the crime scene samples. ple THOI, which resides on chromosome II and has the core
Pitchfork was convicted of the murders and sentenced to sequence AATG, repeated six times on one copy of chromo
life imprisonment in 1988. 24 ,25 This case provides the first some 11 and nine times on the other chromosome 11 would
example of a suspect being exonerated on the basis of DNA be denoted as having a genotype 6, 9 for that SIR. Using esti
evidence, an outcome of many subsequent analyses that mates of the frequency at which each known allele ex ists
should not be overlooked. Although DNA evidence has in the population, one can determine the probability that the
often been hailed as being responsible for Pitchfork's con particular allelic combination in question would arise in a
viction, some argue that it was, in fact, 'an old-fashioned person chosen at random from that population. For example,
tip-off from an informanf 26 that resulted in the case being using data from the US Caucasian population, one can esti
solved. However, there is little doubt that the probability mate that this allelic combination at THOI would be found at
that two non-related individuals will show the same RFLP a frequency of 0.053 (or 5.3 per cent) in the population. 2s
pattern by chance is extremely low; this has led to many If a crime scene sample sho ws the genotype 6, 9 at
convictions when samples from a crime scene and suspect THOI, it wiJl be appreciated that looking at this single locus
matched. or region alone will not be particularly helpful in identify
ing the individual who left the DNA because many people
Short Tandem Repeat Profiling in the population are likely to share this genotype. This is
similar to the situation that may arise using conventional
blood group typing as an investigative tool: many people
PRINCIPLES
have the same blood group and thus the discriminatory
Although RFLP analysis is highly effective and shows a power of blood group typing is limited. 29 However, if one
very high level of discrimination, practical issues, namely also looks at other STR loci on different chromosomes, then
the requirement for large amounts of high-quality DNA the more loci that are examined, the less chance there is
(which is often a luxury in forensic investigations), the need that any two unrelated individuals wi ll have exactly the
for radioactive materials in RFLP analysis initially, the com same combination of a ll eles at every STR examined.
plexity of analysing and interpreting the results (Fig. 21.3)
and the development of new techniques to analyse DNA
TECHNIQUE
variants, mean that the majority of forensic DNA analysis
today focuses on a specific type of variation called micro The result, in terms of STR repeat numbers, of analysing a
satellite repeat polymorph isms, rather than the longer mini selection of SIR loci is known as an SIR profile. This can be
satellite repetitive DNA on which RFLP was based. illustrated graphically, with the different alleles being repre
A polymorphism simply refers to the occurrence of more sented as peaks at different positions along a horizontal axis,
than one allele or variant at a particular region or locus in an example of which can be seen in the electrophoretogram
the genome within a population. Strictly speaking, to count shown in Fig. 21.4. These profiles are obtained by extracting
as a polymorphism the allele must be present in the popu DNA from biological samples from a wide range of possible
lation at a frequency greater than 1 per cent; rarer variants sources, the proviso being that the sample contains cells that
or de novo events giving rise to new sequences are termed contain a nucleus and that these have DNA of sufficient
mutationsY The important point to be aware of is that dif quantity and quality. Although the erythrocytes in blood
ferences exist and, in terms of differentiating between indi lack the requisite nuclear DNA, the white blood cell compo
viduals, any difference is of relevance. nent makes blood a particularly good source of DNA for
Interspersed througho ut the genome are stretches of forensic analysis. Blood is frequently analysed in a range
microsatellite DNA in which short core sequences (4-6 of investigations, for example, child murder or assault, or for
bases) of DNA are found to be repeated in tandem, multiple diagnostic purposes in relation to genetically inherited con
times, the exact number of repeats varying from chromo ditions. Semen, saliva or fingernail scrapings may be exam
some to chromosome and from individual to individual. A ined in the case of alleged child sexual abuse; buccal swabs
number of these so-called short tandem repeats (SIRs) have may be taken from subjects in a paternity dispute or from
been found to be highly polymorphic with lots of different suspects apprehended by legal autho rities. Bones, skeletal
alleles, corresponding to the number of times the core remains or teeth are particularly useful in cases in which
sequence is repeated. STR polymorphisms have proven to be body remains are badly decomposed, have suffered fire dam
an ideal tool to use to compare DNA samples and look for age or have been exhumed, because of the increased resist
similarities and differences between individuals. ance of DNA from these sources to degradation. Muscle
The greater the level of variation in repeat number at an tissue, skin, fetal material, placental tissue, urine or faecal
SIR locus in t he population, the greater the power of dis material and inanimate objects QJ1 which DNA traces may
crimination. By analysing the number of times a particular have been deposited (such as cigarette butts, postage stamps,
selection of STRs is repeated on each homologous chromo chewing gum and toothbrushes) are also suitable for DNA
some in an individual, it is possible to build up a profile of the ana lysis (see Butleilo for further references). It is of note that
Forensic DNA analysis I 399
his behalf. Subsequently, Pitchfork's DNA was found to individual's DNA. An individual who has an STR, for exam
share the same RFLP pattern as the crime scene samples. ple TH01, which resides on chromosome 11 and has the core
Pitchfork was convicted of the murders and sentenced to sequence AATG, repeated six times on one copy of chromo
life imprisonment in 1988. 24 ,2 5 This case provides the first some 11 and nine times on the other chromosome 11 would
example of a suspect being exonerated on the basis of DNA be denoted as having a genotype 6, 9 for that STR. Using esti
evidence, an outcome of many subsequent analyses that mates of the frequency at which each known allele exists
should not be overlooked. Although DNA evidence has in the population, one can determine the probability that the
often been hailed as being responsible for Pitchfork's con particular allelic combination in question would arise in a
viction, some argue that it was, in fact, 'an old-fashioned person chosen at random from that population. For example,
tip-off from an informanf 26 that resulted in the case being using data from the US Caucasian population, one can esti
solved. However, there is little doubt that the probability mate that this allelic combination at TH01 would be found at
that two non-related individuals will show the same RFLP a frequency of 0.053 (or 5.3 per cent) in the popuJation. 2o
pattern by chance is extremely low ; this has led to many If a crime scene sample shows the genotype 6, 9 at
convictions when samples from a crime scene and suspect THO]' it will be appreciated that looking at this single locus
matched. or region alone will not be particularly helpful in identify
ing the individual who left the DNA because many people
Short Tandem Repeat Profiling in the population are likely to share this genotype. This is
similar to the situation that may arise using conventional
blood group typing as an investigative tool: many people
PRINCIPLES
have the same blood group and thus the discriminatory
Although RFLP analysis is highly effective and shows a power of blood group typ ing is limited. 29 However, if one
very high level of discrimination, practical issues, namely also looks at other STR loci on different chromosomes, then
the requirement for large amounts of high-quality DNA the more loci that are examined, the less chance there is
(which is often a lux UlY in forensic investigations), the need that any two unrelated individuals will have exactly the
for radioactive materials in RFLP analysis initially, the com same combination of all eles at every STR examined.
plexity of analysing and interpreting the results (Fig. 21.3)
and the development of new techniques to analyse DNA
TECHNIQUE
variants, mean that the majority of forensic DNA ana lysis
today focuses on a specific type of variation called micro The result, in terms of STR repeat numbers, of analysing a
satellite repeat polymorphisms, rather than the longer mini selection of STR loci is known as an STR profile. This can be
satellite repetitive DNA on which RFLP was based. illustrated graphically, with the different alleles being repre
A polymorphism simply refers to the occurrence of more sented as peaks at different positions along a horizontal axis,
than one allele or variant at a particular region or locu s in an example of which can be seen in the electrophoretogram
the genome within a population. Strictly speaking, to count shown in Fig. 21.4. These profiles are obtained by extracting
as a polymorphism the allele must be present in the popu DNA from biological samples from a wide range of possible
lation at a frequency greater than 1 per cent; rarer variants sources, the proviso being that the sample contains cells that
or de novo events giving rise to new sequences are termed contain a nucl eus and that these have DNA of sufficient
mutations. 27 The impoliant point to be aware of is that dif quantity and quality. Although the erythrocytes in blood
ferences exist and, in terms of differentiating between indi lack the requisite nuclear DNA, the white blood cell compo
viduals, any difference is of relevance. nent makes blood a paliicularly good source of DNA for
Interspersed throughout the genome are stretches of forensic analysis. Blood is freque ntly analysed in a range
microsatellite DNA in which short core sequences (4-6 of investigations, for example, child murder or assault, or for
bases) of DNA are found to be repeated in tandem, multiple diagnostic purposes in relation to genetically inherited con
times, the exact number of repeats varying from chromo ditions. Semen, saliva or fingernail scrapings may be exam
some to chromosome and from individual to individual. A ined in the case of alleged child sexual abuse; buccal swabs
number of these so-called short tandem repeats (STRs) have may be taken from subjects in a paternity dispute or from
been found to be highly polymorphic with lots of different suspects apprehended by legal authorities. Bones, skeletal
alleles, corresponding to the number of times the core remains or teeth are particularly useful in cases in which
sequence is repeated. STR polymorphisms have proven to be body remains are badly decomposed, have suffered fire dam
an ideal tool to use to compare DNA samples and look for age or have been exhumed, because of the increased resist
similarities and differences between individuals. ance of DNA from these sources to degradation. Muscle
The greater the level of variation in repeat number at an tissue, skin, fetal material, placental tissue, urine or faecal
STR locus in the population, the greater the power of dis material and inanimate objects 0)1 which DNA traces may
crimination. By analysing the number of times a particular have been deposited (such as cigarette butts, postage stamps,
selection of SIRs is repeated on each homologous chromo chewing gum and toothbrushes) are also suitable for DNA
some in an individual, it is possible to build up a profile of the analysis (see Butler.30 for fUliher references). It is of note tha t
400 I Forensic DNA profiling
03S1358 _ vWA _
• A t 1500
--~~ImUJ;---------------/}fE~----------~~----.L[ ~~~O
j -
,.------~
m
08S1179
j >----_
III
021S11
jm;---_~
018S51
~
m Ilil
2000
1500
1000
500
----~
J m----------------~~m~------------'m m
A ~H~
Figure 21.4 A short tandem repeat (STR) profi le. An electrophoretogram show ing the deoxyribonucleic acid (DNA) profile of a sample
analysed at nine STR loci (three on each horizontal row) and a sex-determining region, ame logenin (labelled AMEL on the second row),
using the comm ercial Profiler Plus® kit (Applied Biosystems, Foster City, CAl. Each peak represents a different allele and the number in
green below the peaks denotes the number of times the particular STR sequence is repeated. Where two peaks are seen at a particular
STR locus, for example, 14 and 15 for D3S 1358 on chromosome 3, the individual has a diffe rent number of repeats on their homologous
chromosome 3s; where a singl e peak is seen at a particular STR locus, for examp le, 13 for D8S1179 on chromosome 8, the individual has
the same number of repeats on both copies of their homologous chromosome 8s. In this example a single peak representing the X
chromosome is seen at the amelogenin locus, indicating that the sample was from a female. (Courtesy of Dr JR Gilder, Forensic
Biomatics, Inc., Fai rborn, OH, USA.)
with the advent of lo w-copy-number (LeN) DNA analysis, differentiated from each other. The kits have been designed
it is possible to obtain a DNA profile from as little starting and validated for specific conditions and quantities of tem
material as a single cell or the few epidermal cells left behind plate DNA, conditions not met by the aforementioned LeN
in a fingerprint, and trace quantities of DNA that may sub analysis. In order to compensate for the trace levels of DNA
sequently be analysed may be deposited simply by touching that would not otherwise be detected, LeN analysis increases
articles or surfaces. 31-35 However, the evidential value should the number of rounds of peR amplification, but the cost is
be treated cautiously and the serious technical and interpre an increase in potential artefacts. Spurious alleles can be
tative issues relating to such evidence should be taken into detected and genuine alleles may fail to show Up,34 casting
account. severe doubts on the acceptability of the results as being of
Once a DNA sample has been extracted, it is amplified a sufficient evidential weight for a court of law.
by a process called the polymerase chain reaction (peR) in Amplification of selected STRs of the sample DNA results
order to generate a sufficient quantity of DNA for detec in a mixture of amplified STR fragments of different length
tion. 36)7 peR is a technique that revolutionized molecular that need to be separated and visualized. This is done by a
biology w hen it was developed in the 1980s, and its intro process of gel electrophoresis or, more recently, capillary
duction enabled the shift from RFLP to STR profiling. It electrophoresis, which separates the labelled fragments on
provides a simple and readily automated way to replicate the basis of the electric charge and size of the fragments. 38
specific preselected regions of the genome so that, from a The fluorescent dye-l abelJed fragmen ts are passed through a
small number of initial copies of the specific DNA sequence reten tive material (either a polyacryla mide gel or viscous
of interest, millions of copies of that sequence can be gen polymer), wh ich retards their migration through the material
erated in a cyclical process. During each cycle of the reac under the action of a potential difference. Shorter fragments
tion, the number of copies of the targeted templ ate pass through the material more quickly and are detected by
sequence of interest is doubled, resulting in an exponential a laser detector and registered as a peak on the electrophore
in crease in the tota l amount obtained after multiple cycles. togram before the longer fragments. In this way profiles
In the case of STR profiling, the target sequences within from different samples can be generated and compared.
the genome are a selection of STRs distributed on different In addition to the SIR loci that are analysed, a gene on the
chromosomes of the genome. With the use of commercial sex chromosomes called amelogenin is also typed in order to
kits, it is possible to amplify as many as 10 SIRs in a single identify the sex of the source of the sample. Although much
reaction tube . These kits app ly a fluorescent label or tag of the X an d Y chromosomes differ from each other and are
to the different STRs to enable them to be detected and non-homologous, there are some stretches of the chromosomes
that show homology, and the amelogenin gene is one of M,Xa.,.,:_ •••••••••••••••
these. There is a difference in size between the X amelogenin
-:1=-_____-::::u'--____.J. .P:~
gene and the Y amelogenin gene, which can be differentiated
on the basis of the size of the PCR products when the amelo
genin gene is amplified in the same way as the SIR loci
BI:J\....___
IIil IIil m
described above. When only a single peak corresponding to
the X version of the gene is detected, the sample is taken to m m
be from a female; when two peaks corresponding to the X
and the Y copies of the gene are detected the sample can be
considered to be from a male.
Considering any locus that is typed, we cannot be
::JJ m
u u
~--~IIil------~mm~-------L
n~~
absolutely confident in our interpretation in every case. m
Genetics is no exception to the rule that biological systems
have their anomalies, and there are times when a sample
shows only an X amelogenin signal when the true source is
in fact male. As for all the STR loci, it is possible for DNA
to be present at too Iowa concentration or to be too badly
degraded to be detected, or for the reaction to fail; thus, a
~____~J~------~u~------~ti~~
particular the Hardy-Weinberg equil ibrium. 43 ,44 Analysis RMPs of the order of one in billions or less is debata ble, but
has shown that in a stable, randomly mating popul ation, for a calculation based on an incomplete profile in which
devoid of selective pressures that would alter the frequency not all the STR loci tested have generated a result, thus
at which existing alleles are found in the population , it is resulting in hig her probabilities fo r the RMPs being calcu
possible to predict the frequency of the possible genotypic lated, discrepancies may be more significant. The issue is
combinations of these alleles in the population. 45 ,4G Using ge nerally addressed by calcul ating an RMP based on the
statistics, one can get an idea of the likelihood th at the population da tabase that generates the most co nservative
given profile would arise by chance in the population from results, i.e. those most favourable to the defence in cases in
which the frequency estimates of the different a lle les have which a suspect's DNA is found to match a DNA sample
been derived and, hence, the weight of the evidence. One fro m a crime scene.
should not lose sight of the fact that the data derived from Further issues regarding population structure may be
the population databases are based on predictions take n in to account when deriving the predicted genotype
established from relatively small samples of individuals fr equencies for population databases when one is dealing
in the popUlation; the results genera ted are based on with populations with a hig h level of inbreed ing. It will be
probabilities. apparent that the closer the relationship of two indiv idu als,
The RMP is calculated by mult iplying together each of the more DNA they will have in commo n and, thus, the
the predicted fr equen cies of occurrence of the profiled more likely it is for individuals from inb red populations to
genotype for the STRs that have been typed in the sample share a particular STR allele than would be the case for two
using t he product rule to determine the probability that the non-related individuals in a randomly ma tin g population ,
given genotype would occur by chanceY The product rule i.e. the conditions of random mating fo r Hardy-Weinb erg
can be used because each STR typed is situated on a differ equilibrium are not satisfied. If a susp ect a nd the true per
ent chromoso me and, as describ ed at the start of this chap petrator of a crime are blood relatives then their genotype
ter, c hromosomes segregate to the ga metes randomly and frequencies will not be independent and a correction fa ctor
are inherited independently of each other. or para meter measuring the popul ation differentia tion/
Th e number of STR loci analysed depends on whichjuris e
substructure, denoted as or FsT , is required. 47 ,48.50,51
diction one is in and, thus, the pow er of discrimin ation ofthe An alterna tive to the RMP for the presentation of DNA
resultant RMP. In the UK, 10 STR loci are routinely exam ined evi dence is likelihood ratios. A likelihood ratio expresses the
at present, whereas in the USA 13 loci are looked at, the ratio of two alternative probabilities of the same evidence
latter number providing a greater power of discrimination. under different hypotheses. In the context of DNA evidence,
Despite the fact that such analyses are typically prese nted in when a suspect's DNA profile matches a crime scen e sample
terms of extremely sma ll RMP chance matches, in the order profile, there are two possible hypotheses : either th e sus
of one in a billion in the UK COUlts 48 and smaller in the USA, pect's profile matches because his biological sample was
it is of note that Jeffreys has argued that we should be found a t the Clime scene or the sample came from some
analysing as many as 16 STR loci to increase the power of unrelated individual who has not been arrested and who
discrimination in light of the increasing number of individ happens to match the suspect by cha nce. The like lihood
uals entered on the national DNA databases aga inst which ratio is usually calculated by dividing the likelihood of the
the crime scene profiles are compared. 49 This is because the prosecu to r's hypoth esis (the first above hypothesis) by the
greater the number of profil es on the DNA databases with likelihood of the defender'S hypothesis (the second above
which unknown samples are compared, in order to identify hypothesis).47, 5! In its Simplest form, the likelihood ratio is
a match, and thus a source of the sample, the greater t he the inverse of the RMP. However, things become more com
probabi lity of a chance match. plicated when alternative scenarios for the evid ence that is
One only needs to look aro und at differe nt ethnic and seen are presented, or if the DNA sample is a mix ture from
racial populati ons to see that interpopulat ion genetic dif more th a n one contributor. To give a feel for the numbers, a
fere nces exist. As such, it is important to generate popul a likel ihood ratio of 1000 or greater provides velY strong sup
tion databases for different race groups because certain port for the prosecution's hypothesis, a likelihood ratio of
alleles will be found to ex ist at a higher frequency in some 100-1000 provides strong supp ort and a likelihood ratio of
populations than in others, for exa mple, the frequ ency 10-100 provides moderate supportY
of the nine-repea t allele at the D1 3S317 STR lo cus is
4.5 times higher in the US Hispanic popul ation than in the
US African-American popul ation (Hisp anic frequen cy = SAMPLE COLLECTION AND PROCESSING
0.15357; African-American = 0.0 32955) (data taken fro m
Butler2B ) . If one were to calculate the RMP for a sample The ge nera l principles underly ing DNA evidence as applied
based on data from a popul ation other th an the true source to a crim inal case involving a child, e.g. the identification
of that sampl e, one may end up with an RMP that misrep of a n unknown child's body or remains or other forensi c
rese nts the true likelihood of a match. Whether the differ investigations, are no different from those of an adult con
e nces would be statistically significant w he n dealing w ith text. Typically, a DNA sample believed to be central to the
---
--
.
---
-~~~~~~~~ - ---~~~~~~~~~~~~--
--- -
DNA evidence/child sexual offence I 403
crime or investigation - be it a blood sample, a sample of depending on the nature of the sample, a perpetrator may be
saliva swabbed from a bite mark or, in the case of alleged able to suggest a plausible reason for the presence or his or
child sexual abuse, semen extracted from underwear, bed her DNA at a Clime scene, leading to termination of an inves
clothing or an intimate sample - is analysed to produce an tigation. 57 It is crucial, therefore, that there is evidence to link
STR profile, which is used to link a suspect to a crime scene. samples directly to the event; that all persons entering the
The profile is compared with a reference profile taken from crime scene and crime scene officers, scientists or other indi
a suspect, identified on the basis of additional evidence in viduals collecting samples take precautionary steps to avoid
the case or, if no leads are originally present to identify a contaminating evidential samples with their own DNA, be
suspect, with profiles lodged on a national DNA database. A this from sneezing, coughing, shedding skin or the like on to
profile on its own is of no investigative value unless there is the sample; that clean equipment and gloves are used when
a profile with which to compare it. Samples on such DNA handling each sample; that protective clothing is worn to
databases may be associated with a known and named indi limit contamination from the sampler; and that the samples
vidual or they may be from anonymous samples collected are securely sealed and appropriately transported and stored
from previous unsolved crime scenes, which enables sus to prevent degradation of the sample. 53 During processing of
pects to be matched with crime scene profiles or crime the samples, it is imperative that case samples and reference
scenes to be associated via matches. samples are handled separately and that samples that have
The legislation governing the power to collect DNA sam not been amplified by PCR are kept away from those that
ples and retain profiles on the databases varies from nation have been amplified. It takes only a minute amount of ampli
to nation. In the UK such powers are wide-ranging, and fied contaminant DNA to swamp any DNA present in a case
DNA can be collected from anyone suspected of a record sample, resulting in only the contaminant amplified DNA
able offence and profiles retained indefinitely, regardless of being detected, at the expense of the genuine sample DNA,
conviction or acquittal of the suspect (Police and Criminal when the two are simultaneously subjected to PCR. Plas
Evidence Act 1984 [amended]), but in other jurisdictions the ticware and consumables used in the analytical steps are
power of the criminal investigative authorities may be another potential source of contamination as they may be
much less. contaminated when they enter the laboratory,55,68 and the
From the point of identification of a crime scene or sub problems of contamination are particularly apparent when
ject from which a sample for DNA analysis is collected to LCN analysis is conducted. 34 ,54,59 Forensic laboratories have
the point at which a profile is generated and interpreted, sbict standard operating procedures to which they adhere,
strict adherence to protocols, an unbroken chain of custody such as the DAB Quality Assurance Standards of recommen
of the evidence and careful sample handling must be dations for forensic laboratories, in order to standardize and
adhered to. 53-55 validate procedures. 70
The risk of contamination of a sample and the need for Despite all these efforts, errors will arise, be they the
correct handling to avoid it cannot be overemphasized; pro result of human factors, the nature or quality of the sample
fessional bodies have described procedures and guidelines to or technical or biochemical artefacts, and this should not be
help ensure quality control and quality assurance in these overlooked. 71 - 73 There is debate as to whether the exceed
processes, for example, the Federal Bureau of Investigation ingly small RMP figures that are presented in court are
(FBIl-appointed Technical Working Group on DNA Analysis unfairly prejudicial when they are not accompanied by esti
Methods (TWGDAM), now known as the Scientific Working mates of these errors, particularly when the RMPs presented
Group on DNA Analysis Methods (SWGDAM), and the DNA are smaller than the probability of laboratory error. How
Advisory Board (DAB).53,55,57 A case may be lost if doubt can ever, some argue that a meaningful statistic for an error in
be cast on whether samples have been correctly handled a particular case cannot be provided. 74
risk of contamination, compromise of a sample or poor
handling - the most noteworthy of which is the trial of
O.J. Simpson, accused of the murder of Nicole Brown Simpson DNA EVIDENCE AND CHILD SEXUAL OFFENCE
in 1994;30,58 false convictions may be secured as a result of
bad practice. 59 The major problem facing the forensic investigation in cases
DNA is ubiquitous: we are all continuously shedding DNA of alleged child sexual abuse, and which is not unique to
and leaving a trace of our movements in this way. In fact, we DNA evidence per se, is that the majority of cases arise ret
need not even have been at a location ourselves in order for rospectively, which presents several difficulties in terms of
our DNA to be found there. It may be deposited by secondary gathering forensic evidence (see Chapter 1). A number of
or tertiary transfer, being carried and subsequently deposited other issues that relate to DNA analysis in such cases are
elsewhere by an animate or inanimate object. 5O - 55 Individuals also of relevance to DNA evidence in a more general sense,
who pass through a crime scene after the event could unwit for example mixed samples, DNA quantity considerations
tingly leave their DNA, implicating themselves in the process. and Y chromosome STR typing; discussion and introduction
Likewise, it is possible for DNA to be present at a scene prior of these in the context of sexual abuse investigation seems
to the event and collected when samples were taken,55 or, appropriate.
404 I Forensic DNA profiling
In the typical case of sexual abuse that comes to attention the abuse, was azoosperrnic, oligos permic or used barrier
only months or years after the event, the opportunity to protection; and whether the victim subsequently washed,
obtain DNA samples has passed by the time investigation changed his or her clothes, ate, rinsed his or her mouth, uri
begins. However, other indications, such as evidence of nated or defecated, all of which can diminish the quality or
pregnancy or sexually transmitted infections, wh ich may be quanti ty of detectable DNA, or even remove all traces of
investigated using DNA techniques, can be highly probative detectable DNA. In the event that a child does present in an
should the opportunity for collection of immediate post acute case, conflicting opinions have been expressed about
coital samples not present itself. The occurrence of sexually the value of ta king intimate fore nsic samples up to 72 hours
transmitted infection in a child is unlikely in the event that after the event in prepuberta l victims 85 as recommended by
sexual contact had not occurred, though vertical transmis the American Academy of Pediatrics. 86 It is important to
sion is a possibility, and one also needs to be aware that con weigh the distress and discomfort that may be inflicted on
sensual sex with an infected partner may have taken place, the victim against the likelihood of evidence being found.
rather than abuse. In one study, genital human papillo Christian et al 85 argue that, if one is seeking evidence of
mavinls (HPV) infection, detected by the presence of HPV semen or sperm in a vaginal swab sample to provide objec
DNA from perineum and vaginal samples, was found in tive evidence of sexual contact with a child who does not
some of the sexually abused preadolescent girls examined have a previous sexual histoty, data detived from adult cases
but not in children in whom sexual abuse was excluded, the is inappropriate to prepuberta l situations. While the litera
majority of the HPV DNA-positive cases being subclinical ture reveals marked differences in survival time of sperm
infections.75 It is noteworthy, however, that the incidence of cells in different body orifices, the generally accepted maxi
sexually transmitted infections in victims of child abuse is mum times for observin g sperm heads range from 24 hours
low, a reflection of the generally low incidence of infection in the mouth to 7 days in the vaginal cavity, and in the
in perpetrators. In a recent study in Edinburgh, Scotland, region of 2-3 days in the anus and rectum .87,88 This argu
sexually transmitted infection was seen in less than 10 per ment 55 may be va lid in the case of cytological observation of
cent of referrals. 76 sperm cells or the detection of semen by, for example, acid
The appropJiateness of DNA testing for sexually trans phosphatase activity or the presence of phosphate-specific
mitted infectious agents using nucleic acid amplification test antigen (PSA), which have a shorter period of detection BG ,89
(NAATs)77 in suspected cases of child abuse has been ques However, g iven the sensitiv ity of DNA analysis and its abil
tioned because of t he potential for false-positive results, with ity to detect trace material refractoty to other detection
a possible consequence of false convictions. 78 .79 However, it methods, particularly using LCN techniques,31-34 one should
is paramount that the well-being of the child is prioritized in not discount the possibility of obtaining DNA evidence
any investigation and that the child's health is not compro beyond the time periods suggested by Christian et al. 85
mised at the expense of gathering forensic evidence. While In one study, evidence of male DNA, detected by Y-STR
NAATs may not be as specific as traditional culture tests, ana lysis, was found in over 90 per cent of vaginal or ana l
they show increased sensitivity. From a medical perspective, swabs taken from child sexual assault victims up to 72 hours
the downside of a false- positive result is limited to the cost after the incident,9o supporti ng the fll1dings of others that
and side-effects of superfluous antibio tic treatment com Y-STR DNA amplification can produce positive resu lts in
pared with the long-term potential serious health conse swabs cytologically negative for spelm cells. Such negative
quences for the individual of untreated infection. 8o From a results wo uld usu ally have led to tennination of biological
legal perspective, given that culture tests are sti ll the only investigation,91 illustrating the va lue of conducting the DNA
test currently accepted by the UK courts 81 and some labora analysis. The presence of Y-chromosome DNA in a high
tories no longer conduct culture tests, having implemented vaginal swab from a fe male would be interpreted as tanta
NAATs alone,82 there are issues relating to obtaining suitable mount to evidence of intercourse, and in the case of a minor,
forensic evidence. The recommended approach is to take tvvo where consensual sex is not a defence aga inst statutoty rape,
swabs so that a n initial NAAT-positive result can be con is highly probative evidence of abuse.
firmed by culture,81 though this involves additional distress There are practical considerations, in the case of a sma ll
and discomfort to the child. If pregnancy follows sexual child, in ensuring that internal samples genuinely represent
abuse, it may be possible to gather evidence as to the likely material exclusively of an internal nature. The anatomy of
father by way of DNA paternity testing of offspring or a young girl does not lend itself easi ly to collection of such
aborted fetal material ,83,84 a technique that is discussed later samples, and care has to be taken to avoid contact with
in this chapter. external regions of the child 's body. There may be perfectly
Although child sexual abuse includes a spectrum of acceptab le re asons for male DNA to be present on a child's
activities from intercourse to physically less intrusive sexual body, particul arly if male relatives have frequent contact
ab use of a minor, whether samples taken for DNA analysis with a child, and sampling of t his material could lead to
are informative depends on many factors. These include the inappropriate conclusions being drawn, incrimin ating an
nature of samples in relation to the offence; the time since innocent person. In this respect the use of the colposcope
t he offence; whether a male perpetrator ejaculated during with the magnification and illumination that it affo rds has
- .
Y chromosome STR typing I 405
assisted in the collection of non-contaminated intimate son, essentially unchanged in sequence, so all male offspring
samples. through the generations descended from one particular male
will share the same Y chromosome and, hence, the same
Y-chromosome haplotype. Thus, given a Y-chromosome
Y CHROMOSOME SHORT TANDEM haplotype alone, it would not be possible to conclude that
REPEAT TYPING the man who was the origin of a DNA sample was its source,
or whether it came from a blood relative such as his brother,
As mentioned above, if one can detect the presence of son, father, grandfather, grandson, or uncle on his father's side
Y-chromosome DNA in a sample then, barring syndromes of the family.
with abnormal chromosome complement or structure, one A number of internet-accessible Y-SIR databases have
can conclude that male DNA is present. Such evidence is been established,92 which vary in the number of SIR loci for
particularly useful in the case of all eged sexual abuse by a which informatio n is avail ab le. For example, the frequ ently
male against a female, although typing the Y chromosome of used and largest Y-SIR haplotype reference database (YHRD)
a sample has add itional valuable applications, including was established in 2000 with the objectives of generating
paternity testing and the identification of missing persons. ' reliable Y-SIR haplotype frequen cy estimates for minimal
If an internal vaginal sample is taken from a girl under the and extended Y-STR haplotypes to be used in the quantitative
legal age for sexual intercourse and Y-chromosome DNA is assess ment of matches in forensi c and genealogical case
detected, this is convincing evidence of abuse. work' and assessing the 'ma le population stratification
In theory, one could test for male DNA using the amelo among world-wide populations as far as reflected by Y-SIR
genin locus by conventional SIR typing, the X and Y amelo haplotype frequency distributions'.95 The database can be
genin alleles being of different size. The problem with this searched for frequency data and geographical distribution
approach is the frequently encountered imbalance in the rel data of a particular haplotype, both of which may aid in a
ative quantity of any male DNA that would be present in sam forensic investigation ; however, the haplotypes entered cur
ples routinely taken from the female victim in an alleged case rently vary in the number of SIR loci for which data are
of sexual abuse, i.e. vaginal, anal or oral swabs. Such samples available. In 1997, a 'minimal haplotype' was recommended
typically comprise a vast excess of female DNA, usually from that typed seven 10ci 96 for which data are ava ilable for all
host epithelial or inflammatory cells, with only traces of male entries on the YHRD SIR Database. Since then, European
DNA, typically in the form of sperm cells, although epithelial labo rato ries have moved towards typing nine loci , and
and inflammatory cells may also be present, the last two SWGDAM recommends typing 11 loci. 97 Kits for these and
being morphologicaJly indistinguishabl e from the female additional loci are commercially ava ilable, and current prac
equivalents. Given the imbalance in the amount of male and tice is moving towards 17 loci being analysed routinely,
female template DNA added to the PCR, there would be pref further increasing the discriminatory power of the analysis.
erential amplification of the female DNA to the extent that It is likely that the foc us in the future will move towards
male DNA may go undetected. This is analogous to the situa supplementing the information derived from Y-SIRs with
tion described earlier of a DNA sample to be typed being con single-base sequence variations, called single nucleotide
taminated with PCR-amplified DNA, due to poor laboratory polymorphisms (SNPs).
practice, and swa mping the sma ll amount of DNA present in Various ways are used to try to enri ch a fema le/male
the sample, leading to preferential re-amplification and detec mixed sampl e for the male cell component in order to assist
tion of the contaminant DNA. As ever, absence of detectable in the analysis and maximize the chance of getting an inter
DNA is not evidence of absence. pretable Y-SIR haplotype data. For example, differential
One way to get round this problem is to type a series of DNA extractio n techniques may be employed, which take
SIR loci which are specific to the Y chromosome (Y-SIR advantage of the increased resistance of sperm cells to par
typing),92,93 so that female DNA is not amplified in parallel ticular reagents routinely used to digest cells during the DNA
and a male-specific profile or Y-SIR haplotype results. There extraction step of ana lysis. Epithelial or other non-sperm
are hundreds of STRs distributed on the Y chromosome,94 cells are initially lysed and separated from residual sperm
and by typing a chosen set of these one can identify the par ceJls, which can then be processed separately?O,98 Antibodies
ticular Y haplotype of an individual male, i.e. the closely specific to sperm cell antigens coupled to magnetic beads 30
linked alleles of that individual at the chosen loci. One gets or laser microdissection (LM) techniques 30 ,99.lOo may be used
more information about the source of the samp le using to directly capture sperm cells and, more recently, fluores
Y-SIR profiling than if one was just to type the amelogenin cence in situ hybridization in conjunction with LM has been
locus to determine maleness because of the variation that developed as an effective way to isolate not only sperm cells,
exists in the Y chromosome between non-related males and but also male epithelial and inflammatory cells. 101 , I02 The
between different ethnic, racial and geographic groups. latter technique presents a solution to the problem of
However, the power of discrimination of Y-STR typing in no deposited semen samples that are devoid of sperm cells as a
way approaches that of conventional autosomal STR typing. result of a medical conditions or vasectomy, and offers a
This is because the Y chromosome is passed from father to way to include not only sperm cells, the subject of previous
406 I Forensic DNA profiling
analysis, but also additional male cells, which will boost the Sharing a surname in such societies significantly elevates
quantity of extractable male DNA in small samples. the probability of sharing a Y-chromosome haplotype,
As with conventional STR typing, the result of Y-STR leading to the proposal that the establishment of a large
typing can lead to an individual being excluded as the surname-Y-chromosome haplotype database might be of
source of a sample, if the two Y haplotypes do not match. intelligence value in a case devoid of other leads and when
The results may be inconclusive, for the same reasons as dis male DNA central to the case has been identified, for exam
cussed previously, or they may lead to an individual being ple, in a rape case. By comparing the Y-chromosome haplo
included as a possible source of the sample, when the sam type of such a sample against the database, a potential
ples show sufficient similarity that they could have origi surname of the source might be identified and lead to iden
nated from the same individual. In this case, in a court tification of a pool of potential suspects. Because many indi
setting, the findings are typically presented as being unable viduals share a Y-chromosome haplotype, existing databases
to exclude the suspect (the individual found to match the do not attempt to attach identity to an entry. There are unde
haplotype) as the donor of the crime sample, but also unable niably many barriers to the success of this proposed
to exclude all patrilineally related male relatives, as well as approach, including non-paternity events, surname chang
an unknown number of unrelated males, as potential ing, adoption, the existence of more than one founder of a
donors.92 Courts often seek some kind of statistical meaning particular surname, whether the theory holds true for com
to this conservative statement. The RMP calculation mon surnames and mutations. Civil liberty and privacy
described above is not appropriate because, unlike the auto issues in establishing and using such a database need to be
somal STR loci that are typed for conventional STR profiling, addressed. The method would not be able to conclusively
which are located on separate chromosomes and, hence, exclude or include any individual as a perpetrator, and more
inherited independently of each other, the Y-STR loci are research is required if the proposed approach is to become a
inherited together as a group of linked loci on all but rare realistic option, but it is interesting to be aware of new
occasions. Instead, an indication of the probability of a coin opportunities to exploit genetic information that may come
cidental match may be evaluated using the counting method to fruition.
or a Bayesian-approach haplotype surveying method. 92 • lo]
The counting method is the simpler and expresses the num
ber of times the given haplotype is found in the database MIXED SAMPLES
that has been searched and upper-bound confidence limits
that can be placed on the haplotype's frequency. The signif Y chromosome-specific short tandem repeat typing is of
icance of inferences made from these calculations is highly value when more than one male perpetrator may have con
dependent on the size of the database used. The 20th release tributed to a sample because it reduces the number of possi
of the aforementioned YHRD STR Database included 46831 ble allelic combinations that one is trying to analyse and is
seven-loci 'minimal haplotypes'. Of these, 44974 were typed useful in the event of a mixed male and female sample.
for the nine European recommended loci and 17935 for 11 Where more than one individual contributes to a sample.
SWGDAM recommended loci. 95 The US National Center for then multiple peaks representing multiple alleles may be
Forensic Science is currently in the process of creating a seen on the electrophoretogram of a conventional STR
National US Y-STR Haplotype Reference Database, which analysis. If two individuals contribute to a sample then up to
will contain data on ll-loci haplotypes. The database is to be four different alleles could be detected at anyone locus if
created by compiling and consolidating existing govern both individuals were heterozygous at the locus; in the case
ment, commercial and academic Y-STR databases and will of three contributors there could be up to six different alle
allow continuous updating of haplotypes entered on the les. Such scenarios introduce additional practical and inter
database by retyping existing samples as new STR loci are pretative problems over conventional sample analysis. 107 - 11l
introduced to create extended haplotypes.lOJ.l04 Making How does one know which of the possible allelic combina
allowances for advancing technology and new typing tech tions to attribute to the different contJibutors? One approach
niques in this way shows foresight, but raises some of the is to attempt to allocate alleles on the basis of the strength
highly contentious issues that exist in relation to whether of the signal (peak height or peak area on the electrophore
only profiles of samples or actual DNA samples of DNA togram), working on the assumption that if one of the con
database entries should be retained once an entry is lodged tributors' DNA was in excess of the other at the start of the
on a DNA database. PCR, then the alleles initially present in greater quantity
In the context ofY-STR typing and forensic applications, would produce larger peaks after amplification. This may not
interesting data are coming forth suggesting a way in which hold true in every instance, however, and in cases where
the common biological attribute of the Y chromosome near-equivalent quantities of DNA from more than one indi
shared by male relatives might be exploited in conjunction vidual are present, all signals detected being of similar inten
with another commonly shared feature of many male rela sity, the approach will not be possible. If one considers the
tives, namely their surname, in societies in which the custom scenario of five or more peaks, things get even more compli
is for the family name to pass through the male lineage. 105.106 cated in trying to detem1ine how many different individuals
406 I Forensic DNA profiling
analysis, but also additional male cells, which will boost the Sharing a surname in such societies significantly elevates
quantity of extractable male DNA in small samples. the probability of sharing a Y-chromosome haplotype,
As with conventional STR typing, the result of Y-STR leading to the proposal that the establishment of a large
typing can lead to an individual being excluded as the surname-Y-chromosome haplotype database might be of
source of a sample, if the two Y haplotypes do not match. intelligence value in a case devoid of other leads and when
The results may be inconclusive, for the same reasons as dis male DNA central to the case has been identified, for exam
cussed previously, or they may lead to an individual being ple, in a rape case. By comparing the Y-chromosome haplo
included as a possible source of the sample, when the sam type of such a sample against the database, a potential
ples show sufficient similarity that they could have origi surname of the source might be identified and lead to iden
nated from the same individual. In this case, in a court tification of a pool of potential suspects. Because many indi
setting, the findings are typically presented as being unable viduals share a Y-chromosome haplotype, existing databases
to exclude the suspect (the individual found to match the do not attempt to attach identity to an entry. There are unde
haplotype) as the donor of the crime sample, but also unable niably many barriers to the success of this proposed
to exclude all patrilineally related male relatives, as well as approach, including non-paternity events, surname chang
an unknown number of unrelated males, as potential ing, adoption, the existence of more than one founder of a
donors. 92 Courts often seek some kind of statistical meaning particular surname, whether the theory holds true for com
to this conservative statement. The RlVIP calculation mon surnames and mutations. Civil liberty and privacy
described above is not appropriate because, unlike the auto issues in establishing and using such a database need to be
somal STR loci that are typed for conventional STR profiling, addressed. The method would not be able to conclusively
which are located on separate chromosomes and, hence, exclude or include any individual as a perpetrator, and more
inherited independently of each other, the Y-STR loci are research is required if the proposed approach is to become a
inherited together as a group of linked loci on all but rare realistic option, but it is interesting to be aware of new
occasions. Instead, an indication of the probability of a coin opportunities to exploit genetic information that may come
cidental match may be evaluated using the counting method to fruition.
or a Bayesian-approach haplotype surveying method. 92 , IOJ
The counting method is the simpler and expresses the num
ber of times the given haplotype is found in the database MIXED SAMPLES
that has been searched and upper-bound confidence limits
that can be placed on the haplotype's frequency. TI1e signif Y chromosome-specific short tandem repeat typing is of
icance of inferences made from these calculations is highly value when more than one male perpetrator may have con
dependent on the size of the database used. The 20th release tributed to a sample because it reduces the number of possi
of the aforementioned YHRD STR Database included 46831 ble allelic combinations that one is trying to analyse and is
seven-loci 'minimal haplotypes'. Of these, 44974 were typed useful in the event of a mixed male and female sample.
for the nine European recommended loci and 17935 for 11 Where more than one individual contributes to a sample,
SWGDAM recommended loci. 9s The US National Center for then multiple peaks representing multiple alleles may be
Forensic Science is currently in the process of creating a seen on the electrophoretogram of a conventional STR
National US Y-STR Haplotype Reference Database, which analysis. If two individuals contribute to a sample then up to
will contain data on ll-Ioci haplotypes. The database is to be four different alleles could be detected at anyone locus if
created by compiling and consolidating existing govern both individuals were heterozygous at the locus; in the case
ment, commercial and academic Y-STR databases and will of three contributors there could be up to six different alle
allow continuous updating of haplotypes entered on the les. Such scenarios introduce additional practical and inter
database by retyping existing samples as new STR loci are pretative problems over conventional sample analysis.107-lll
introduced to create extended haplotypes. 103 ,104 Making How does one know which of the possible allelic combina
allowances for advancing technology and new typing tech tions to attribute to the different contlibutors? One approach
niques in this way shows foresight, but raises some of the is to attempt to allocate alleles on the basis of the strength
highly contentious issues that exist in relation to whether of the signal (peak height or peak area on the electrophore
only profiles of samples or actual DNA samples of DNA togram), working on the assumption that if one of the con
database entries should be retained once an entry is lodged tributors' DNA was in excess of the other at the start of the
on a DNA database. PCR, then the alleles initially present in greater quantity
In the context of Y-SIR typing and forensic applications, would produce larger peaks after amplification. This may not
interesting data are coming forth suggesting a way in which hold true in evelY instance, however, and in cases where
the common biological attribute of the Y chromosome near-equivalent quantities of DNA from more than one indi
shared by male relatives might be exploited in conjunction vidual are present, all signals detected being of similar inten
with another commonly shared feature of many male rela sity, the approach will not be possible. If one considers the
tives, namely their surname, in societies in which the custom scenario of five or more peaks, things get even more compli
is for the family name to pass through the male lineage. lOS, 106 cated in tlying to determine how many different individuals
--
Mitochondrial DNA I 407
might have contributed to the sample. There are numerous MITOCHONDRIAL DNA
accepted that, with the exception of mutation events, all As for Y haplotypes, a statistical estimate of the signifi
siblings and maternal relatives have the same mtDNA, so it cance of a match is needed when mtDNA sequencing fails to
is not possible to identify the source of a particular mtDNA exclude an individual as being the source of an unknown
sample to the degree of specificity with which nuclear DNA sample. The process is similar to the Y haplotype evidence,
may be attributed. being a reflection of the frequency at which the given
For forensic purposes mtDNA is analysed by sequencing mtDNA sequence occurs in established mtDNA databases. 121
two specific sections of the mitochondrial genome that show
the most interindividual variation, namely hypervariable
regions I and II. The sequence results of an analysis are PATERNITY TESTING
reported in terms of differences relative to a reference
sequence l27 of the hypervariable regions. 121 When an Inheritance of mtDNA and the Y chromosome provides a
unknown sample and a known sample are compared and means of tracking blood relationship through the genera
they share the same sequence, it is not possible to exclude tions, but often one wishes to have a more definitive means
the known sample as being the source of the unknown sam of detelmining the biological parentage of a child. Proof
ple, but, analogous to the case ofY-STR typing, nor can one of paternity or maternity may be needed to settle inheritance
exclude all individuals originating from the same maternal disputes, child maintenance responsibilities or immigration
lineage. If more than two differences in nucleotide sequence applications, not to mention the psychological and relation
are seen, the known sample is excluded as the source of the ship issues and problems that go along with doubts in trust
unknown sample. If there is one difference in the sequences, and fidelity of relationships. The principles behind paternity
the result is said to be inconclusive to account for mutations testing may also be employed to assist in the identification
that have been seen to occur from mother to child. 128 Mito of the remains of an unknown when no reference sample of
chondrial DNA has a higher mutation rate than nuclear the individual is available, if there is evidence to support
DNA, and tracking the conservation of a particular sequence parentage by individuals who can provide samples.
through generations and the emergence of mutations has DNA paternity testing uses STR typing and the princi
been a useful tool in genealogy studies, acting like a molec ples behind the manner by which DNA is inherited from
ular clock, and has provided support for the 'Out of Africa' our parents. It will be appreciated that, because any indi
hypothesis of human genesis from a woman who lived vidual inherits half of his or her genetic material from each
200000 years ago. 129. 130 biological parent (one copy of each of the chromosome
A consequence of mtDNA's susceptibility to mutation can pairs), the presence of genetic material that is not consis
give rise to a situation of heteroplasmy whereby an individ tent with this mode of transmission, i.e. of SIR alleles not
ual can possess mtDNA molecules with different sequences. present in the alleged parents, is strong evidence that the
As there are multiple mitochondlia and multiple mtDNA alleged parents are not in fact the biological parents.
molecules, if a viable mutation alises in a subset of these In most cases, the biological mother is known and one is
molecules they may be passed on as the cells divide, giving seeking to determine whether an alleged father could be the
rise to mixed populations of mtDNA molecules in an indi biological father, though it may be the father who is known
vidua1.121.131 Whether more than one mtDNA sequence is and it is maternity that is being investigated. Alternatively,
detected on sequencing a sample from an individual exhibit one may be faced with the more difficult situation that only
ing heteroplasmy will depend on the relative proportions of a parent and alleged child are available for analysis 135 or it
the different populations of mtDNA molecules in the sample is necessary to determine, in the case that the mother is
analysed, which may vary from cell to cell and from tissue known, whether two children have the same or a different
to tissue. Heteroplasmy appears to be a more common phe father. 136 In the first of these scenarios in which the mother
nomenon than previously thought 13l and the consequences is known, if one SIR types the mother, the alleged father and
for forensic analysis should be taken in to account when the child, and compares the mother's alleles with those of the
interpreting results. This appears to be particularly relevant child at each SIR locus, it may be possible to determine the
in relation to hair samples, which are often found at crime allele that the father must have passed to the child, the 'obli
scenes and used as evidence in a case. Because the root of gate allele'. For example, if the mother has alleles 15, 16 at a
the hair is needed for nuclear DNA analysis and is lacking in particular SIR locus and the child has alleles 13, 15 then the
many hairs from crime scenes, mtDNA is more frequently child must have inherited allele 15 from the mother and,
performed and heteroplasmy has been found to be more hence, must have inherited allele 13 from the father. If the
common in hairs than in other tissues. 132 Probably the most father has alleles 13, 14 then he could not be ruled out as the
famous example in which DNA evidence was found to father on the basis of DNA evidence, but nor could any other
exhibit heteroplasmy was in skeletal remains identified as man carrying allele 13 at this locus. If, on the other hand, the
Tsar Nicholas II from exhumed remains believed to belong to mother was 13 , 15 and the child was 13, 15, then it would
members of the Russian Royal family murdered at Ekaterin not be possible to nalTOW the father'S contribution to a sin
burg in 1918. The investigation concluded that both Tsar gle allele because the mother could have passed on either
Nicholas II and his brother exhibited heteroplasmy.1 33,134 allele 13 or allele 15. A man with either a 13 or a 15 allele
Paternity testing I 409
could have been the father, i.e. there would be more possible identified had he been the only putative father tested. 142
men who cou ld have fathered the child. Obviously, one looks Other studies also highlight the problem of relatives being
to more than a singl e STR locus when conducting this analy mistaken as a biological father; 14J a way to determine more
sis to determine if the alleged father has alleles consistent definitively if a given allele is inherited from a given indi
with fatherhood at each locus typed; the greater the number vidual would be of great value in preventing potentially mis
of loci examined, the greater the evidential weight. taken conclusions.
Tf a n alleged father cannot be excluded as the father, Determining the parental origin of alleles would also be
on the bas is of his genotype being inconsistent with those of help in cases in which the mother is not available for test
required to fi t the genealogy, then one needs to have a sta ing and DNA is availabl e only from the child and alleged
tistical measure of the weight of this evidence as indicative father, in the case that t he mot her and child share the same
of paternity. This is most commonly presented in terms of heterozygous genotype and when multiple male relatives
the combined paternity index (CPl), though the probability are suspected of incest. A novel test has been proposed that
of paternity, which is based on a Bayesian approach, is also migh t be app licable to these cases. It would look at genomic
w idely used. 121, IJ7 - 139 imprinting that resul ts fro m epigenetic modifications, which
The paterni ty index is a likelihood ratio w here the proba change the chemical structure of the DNA without altering
bility of obtaining the observed genotype in the child at a the underlying DNA sequence.144-146 Certain genes have
particular locus, given that the alleged fa ther is t he biologi been shown to exhibit this phenomenon, commonly by way
cal father, is di vided by the probability of obtaining the of DNA methy lation, with the changes being parent specific.
observed genotype, g iven that a random man is the father. Tn It is possible to detect these changes; thus, there may be a
order to determine the probability of obtaining the observed way of determining the parental origin of a particular allele.
genotype, given that a random man is the father, one uses The approach has been proposed for forensic investigations
the allele frequencies of the obligate allele from population and paternity testing, 147, 148 and it will be interesting to see if
databases of the race of the alleged father. Tn order to obtain it becomes a routine part of forensic DNA analysis, add ing
the probability of obtaining the observed genotype, given strength to support the RMP, where not only the same alle
that the alleged father is the biological father, one looks at les are shown to match a suspect but a common geno mic
his genotype. If he is homozygous for the obligate allele the imprinti ng pa ttern is observed. On a related vein, analys is of
probability would be 1 because there is no alternative allele epigenetic differences has also been proposed as a potential
that he could have passed on to a child; ifhe is heterozygous way to differentiate between monozygotic tw ins. As dis
for the allele the probability would be 0.5, because there is cussed above, monozygotic twins inherit the same ge ne tic
an equal chance that he passed on either of his altern ative sequence, and it has been shown that they are epigenetically
alleles. 121, IJ8 Having calculated the paternity index for each indistinguishable at birth and in the early years of life ; how
locus, a CPI is calculated by multiplying the paternity ever, older twins exhibit detectable differences in epigenetic
indexes for each locus analysed. A cpr of 100 is accepted to modifications of particular sequences t hat were acqu ired
establish 99 per cent probability of paternity and a CPI of during life. This may be one way to differentiate between
greater than 1000 indicates that the probability of paternity identical twins.149
is greater than 99.9 per cent. A CPI of 100 is the genera lly While the increase in shared alleles between relatives
accepted minimum standard for inclusion,140 although some may lead to problems in identifying the real father in a
laboratories use a hi ghe r cpr as their minimum.141 paternity case, it is worth mentioning that, because relat ives
In the context of paediatric cases, paternity testing is have more alleles in common, the effect of th is in incestu
not just applied to the situation where one wishes to iden ous relationships gives rise to an increased level of homo
tify the father of the subj ect of the case; t he subject may be zygosity in the offsprin g, which may be used as an ind icator
the mother, for instance in the case of statutory rape. By of incest. 150
DNA profiling feta l material from a borted pregnancies and CPT or alternative statistics may be reported when an
from the child (mot her) one can determine the likelihood alleged man is not excl uded as the biological father. But on
that a given suspect was t he father an d, hence, guil ty of the basis of ge nea logy, how different do the alleles need to
statutory rape. 8J be from the alleles we would expect the father to have in
It is important to appreciate that no CUlTent DNA test is order to lead to an exclusion? DNA paternity testing works
definitive and there a re a numbe r of issues relating to pater on the same basis as blood group testing, proposed in the
nity testing that need to be considered. As stated earlier, 19205, 151 and late r refined, as a means of investigating
relatives share a highe r proportion of their genome than paterni ty disputes, but may add a greater degree of discrim
random men, so problems may arise if a relative of the real ination. Rather than looking at the phenotypic effect, i.e. the
father, for example, father, brother or son, is being consid observed effect of our genetic material, namely what blood
ered as the biological father. One study found that, in over group an individual is, 29 DNA typ ing looks at the genetic
31 per cent of cases, an un cle could not be excluded as the material itself, which provides more information. Despite
father, with five cases showing no mismatches. These five frequently being overl ooked today, blood group testing was,
cases would have resul ted in paternity by the uncle being and still is, an effective way of, relatively quickly and
410 I Forensic DNA profiling
cheaply, excluding an individual as being the source of a might lead to a conclusion of non-paternity being
samp le or the biological father of a child. If the blood groups dra wn. 166- 169 Mutation rates vary across the SIR loci and
did not match a suspect when trying to identify the source of some may be as high as 0.3 per cent of meiotic events. 165
a sample, then a suspect could be excluded or, in the case of Evidence of two mutations occuning between generations in
a paternity dispute, if the offspring's blood group was not several cases has led to a minimum of three independent
consistent with the possible combinations that could have allel ic inconsistencies being required for a putative father
arisen from the mother and alleged father, then the individ to be excluded as the biological father. 167 ,168 When there is
ual in question is excluded . doubt whether inconsistent alleles are due to mutatio n or a
Despite this appa rentl y clear-cut log ic, things may not genuine excl usion of the alleged father, typing the Y chromo
be quite as simple as far as DNA profiling is co ncerned, for some SIRs of the child, if male, may provide further evidence
reasons to be explained below, and, from histo rical cases, to show a genuine exclusion or not, as the case may beY6,1 70
court decisions have not always followed the scientific The mutations discussed above relate to ge rmline events,
theory and evidence for blood group testing. For instance, which are passed from parent to child and affect all the cells
in the famou s paternity dispute in 1946 involving the actor of the offspring, though somatic mutations may also arise
and director Charlie Chaplin, the apparent clear impossibil in cells other than the parental gametes. If a mutation arises
ity of fatherhood, on the basis of blood group testing, was at an STR immediat ely post fertilization and that STR is
not viewed as bei ng conclusive, and Chaplain was ordered subject to forensic ana lys is, then the resulting cells of the
to pay child support to the child of Ms Joan Berry.152.153 offspling will have a profile that differs from that expected
In 1983, Shaw l52 stated 'Motherhood has always been a on the basis of genetic inheritance. All the child 's cells will
biological certainty; now fatherhood will be as well. We will carry the mutated allele and not the parental one. This has
have come one step closer to equality of sexes.' Little did consequences for pat ernity investigations.
he anticipate the consequences of errors that have been, and It is not just in paternity investigations that mutations
no doubt will continue to be, made in conducting in vitro can affect the outcome of t he investigation. Somatic cell
fertili zation (NFl. A growing number of cases have come to mutations in SIRs can lead to anomalous profiles with more
light in which embryos have been transferred to the wrong than two alleles being seen per locus 171, 172 and discrepan
. woman, so that she has been impl anted with another cies in results of comparison of profiles, even if the profiled
woma n's eggs; the wrong spelID has been used to inseminate sa mples originated fro m t he same individual. If a mutation
eggs; or both errors have occurred. 154 ,155 It has been admit a rises in a subset of cells that are profiled , and the mutated
ted that these mix-ups are a regular occurrence in fertility cells are at a level above the threshold of detection, a com
clinics in the UK;156 similar cases have arisen in other coun parison between pro files of cells of a different tiss ue, or a
tries. 157 ,150 Th e majority of cases t hat have come to light sample of cells lacking a sufficient proporti on of mutated
involve white parents having a black child, or vice versa. cells for detection, can lead to the possibility of a common
This is the most readily apparent way of detecting any mis source of the samples, and thus a match, being missed. This
takes. How many other cases go undetected is open to ques may be of relevance if, for example, blood cells sampled at
tion. Paternity testing is one way that ca n bring the errors to a crime scene have a different profile from those of buccal
the fore . The ramifications of such events for the parties cells taken as a reference sample from a suspect.
involved cannot be underestim ated, encompassing psycho Individuals who possess cells with different genotypes
logical trauma, legal parenthood and custody issues as well that have arisen from a single zygote by way of mutation
as compensation considerations. 155 ,158-1 60 are known as mosaics. 173, 174 An individual may also possess
cells of differen t genotypes because of ch imerism resulting
from natural anomalous events in which more t han one
Mutation and Anomalous Genotypes zygote fuses to create a single individual; 175-1 78 when cells
derived from a different zygote are present in a single indi
While the decision in the Charlie Chaplin paternity case has vidual (e.g. if dizygotic twi ns exchange blood across a pla
been cliticized, given what appeared to be an impossibility of centa l79 ,lool or as a result of medical intervention, be it bone
blood group inheritance, using DNA analysis, failure to marrow or organ donation.181-184 In the case of bone mar
exclude an individual as a parent, even when alleles of the row transplantation, for example, it is possible for false
alleged father do not appear t o fit those expected, do es have identification to ar ise when the donor may, in fact, be the
scientific reasoning. The fact is that SIR loci, like any region guilty party, but because the recipient harbours cells shown
of the genome, are susceptible to mutation. 161 In fact, these to match the donor's cells left at a crime scene the recipient
regions of the genome are particularly susceptible to muta may be falsely accused or convicted. 185
tion and show a high rate of mutation during gamete forma It will be apparent that chimerism and mosaicism can lead
tion, when the cells undergo meiosis, which can lead to allel ic to inappropliate exclusion of the true source of a sample,
change from one generation to the next. 162-165 This can result when looking for a match, and to the exclusion of a biologi
in an allele in the offspring being different from the parental cal parent as being so if the possibility of the phenomenon
allele from which it was derived, and without further testing is not taken into account during an investigation. 175 ,176, 186
Identification: body remains/missing persons I 411
The problem of the perception of the infallibility of DNA evi parent. The best it can do is provide evidence in terms of sta
dence was highlighted in the media-grabbing case where a tistical probability and it is up to a jury or those concerned
mother was accused of committing benefit fraud because it to determine if it fulfils their criteria of proof.
was alleged that the she could not have mothered the children
in question and risked them being taken in to care. It was only
after her third child's birth was witnessed, and DNA evidence IDENTIFICATION OF BODY REMAINS
also showed that baby appeared not have been her child, that AND MISSING PERSONS
it was discovered that she was a chimera. The cells giving rise
to her offspring were of different genotype to those typed The loss of a child is devastating for any parent, and one can
during the maternity investigation. 177 not overemphasize the importance of closure for families in
the event of such tragedies by way of positive identification
that their child has actually died. If the remains of a deceased
The Use of Mitochondrial DNA and the person can be identified, this can put an end to uncertainties
Y Chromosome in Paternity Cases or hopes that a child may have sUlvived and can enable
return of the body or remains to the next of kin for appropri
As mentioned above, the Y chromosome may be used to pro ate burial, events that may be of deep psychological and reli
vide additional information in cases where there may be gious significance and comfort at a time of distress.
uncertainty over whether a mutation has arisen from father DNA analysis is one way that can assist in the identifi
to child. There may be occasions when paternity is in dispute cation of human remains, but the approach is not without
and samples of the putative parent are not available, which problems. Not only does one have to address the state in
causes practical problems. Sometimes, material may have which the remains are found , which is often far from ideal
been taken for medical investigations and, if it has been because of degradation or decomposition, but one needs to
stored, it may be accessed and used as a source of DNA.166 acquire a reference profile with which to compare that of
Obviously, this is not always possible, and one approach that the remains. There is also the issue that following some
might be adopted in the case of a male offspring is analysing incidents, for example, explosions, body parts may be frag
the Y chromosome haplotype of male relatives of the puta mented and widely dispersed, adding to the difficulty of
tive father, given that the Y chromosome is passed from reunify in g an identified whole body.
father to so n and, barring mutation, aU these family relatives If the remains show few signs of decomposition, it will
would share a common Y chromosome. This was put in to generally be possible to obtain DNA from blood samples or
practice in the attempt to address the long-standing histori internal soft tissue which has been largely unaffected. As the
cal controversy over allegations that the US President degree of decomposition increases, blood becomes less of an
Thomas Jefferson fathered the sons of one of his slaves, option and superficial soft tissue may exhibit signs of putre
Sally Hemmings. The first source that on e would look to in faction, in which case tissue or muscle deep er within the
order to type a deceased's Y chromosome would be known body and/or bone marrow may yield DNA suitable for STR
male offspring of the alleged father, but Jefferson had no profiling. When remains are at an advanced state of decom
surviving sons. Instead data were gathered from male-line position, bone marrow may still be able to provide material
descendants of two sons of the president's paternal uncle, for a DNA profile but, as decomposition progresses, the
male-line descendants of the two men Jefferson was alleged extraction process increasingly faces problems of cell debris,
to have fathered (Thomas Woodson and Eston Hemmings) decomposition products and contaminants that can inhibit
and male line descendants of the sons of Jefferson's sister, the PCR reaction, and thus need to be removed. If bone mar
also suspected of fathering one of Hemming's children. row does not yield a profile then hair may be STR typed, if the
Analysis of the Y chromosome haplotypes excluded Jeffer root rema ins, or mtDNA from the shaft may be sequenced.
son of fathering Thomas Woodson, but the younger son, Alternatively, DNA could be extracted from skeletal struc
Eston Hemmings, sha red the same Y chromosome haplotype tures, which is also the option when remains are fully skele
as Jefferson 's relatives, providing support that he may have tonized. The resistance of structures such as bone and teeth
been Jefferson's illegitimate son. 187 However, the very fact offers protection to DNA contai.ned in the bone matrix or
that Jefferso n's male relatives had a common Y haplotype tooth pulp; however, this, in turn, requires special proce
also means that any of Jefferson's male relatives alive and dures to extract and purify it. The level of nuclear DNA that
capable of fatherhood at the time co uld not be ruled out as - can yield an STR profile from these sources is low; sequenc
the father and the president was only one of several candi ing the more resistant mtDNA may be the only option avail
dates, despite the incriminating publicity the results of th e able. On rare occasions one may be faced with naturally
study attracted. 188 The lack of conclusive evidence has mummified or desiccated remains, which present their own
meant the descendants of Jefferson's daughters have not challenges, particularly if there is a degree of decomposition.
accepted the relationsh ip,1 89 highlighting the fact that DNA Extracting DNA from soft tissue of such remains has given
evidence cannot conclusively identify an individual as being little, or marginal, and inconsistent success in STR profiling
the source of a sample or an individual as being a biological (Jason Eshl eman, personal communication; Carlos Morales,
412 I Forensic DNA profiling
personal communication), though extracting and typi ng alleles cannot be identified at evelY 10CllS, th e evidential
DNA from bones and teeth of such remains does not appear weight is lower but va luable information can still be
to offer significant problems over other sources of bone and obtained, as illustrated by this case.
teeth (Tim Clayton, personal communication; Jason Eshle Having an ante-mortem reference sample of a mlss1l1g
man, personal commu nication) . DNA has been extracted and individual with which to compare DNA profiles of remains
sequenced from dried superfici al parts of Egyptian mummi goes a long way to aiding the process of identification, and
fied remains, suggesti ng that it is possible to get genetic over the years such samp les have been taken from tooth
information from such material, but the DNA in this case brushes, hairbrushes, clothes, towels and leftover food with
was in a fra gmented state, making its suitability for STR bite marks.30.193 It is important to realize, however, that
analysis questionable. It was only one of many samples sllch items are mobile and may have been used by someone
studied that indicated that recoverable DNA might be other than the owner and, hence, might cany DNA of
obtained. 190 It is possible tha t natural and artificial mummi someone other than the individual in ques tion. Archived
fication may affect the success of extraction. medica l samples are a very useful source to turn to for ref
Despite these issues and potential problems, DNA analy erence material if they are available.
sis has been of Sign ificant value to investigating teams in The paternity-type approach, using pedigree analys is to
major incidents or mass disasters over the years. Although identify remains when reference material is ava ilable from
techniques such as visual identifi cation, odontology, anthro the pa rents of a ch ild suspected to have di ed, is commonly
pology, fingerprint analysis, radiology and facial reconstluc used. Knowing each parent's alleles, one can predict the
tio n can go a long way in helping to identify remains, they range of possible genotypes that their child could have and
may not always be appropriate or possible. DNA analysis has eliminate remai ns that do not cOITespond to these. If a match
succeeded in many investigatio ns, in conjunction with these is identified, the li kel ihood that the remains are the biologi
tech niques, or in cases in which other techniques have been cal child of the couple rather than the child of an unrelated
unsuccessful. co uple can be determined statistically. Admittedly there can
For example, in 1998 a tragic fire in Manila resulted in be a velY large number of possi ble combinations (for a 10
the death of 23 children between the age of six months and loclls profile this cou ld be as high as 1048576), but the
eight years. The bodies were initially buried unidentified, approach has been used successfully in many mass disasters.
but three mon ths later the burial site was exhumed in an For example, after the Waco, Texas, incident of 1993 - when
attempt to identify the remains. 191 Exhumation recovered the remains of 61 individua ls were recovered in varying
22 bodies, and a combination of autosomal and Y chromo states of preservation, some in a n advanced sta te of putre
so me typin g was carried out on 21 of the bodies that had faction and others badly charred - 26 positive identifications
not been otherwise identified. The identification process were made in this way.194 It is also possible to gather suf
was assisted because th e identity of the children present at ficient inform ation when a single paren t is available as a
the tragedy was known, though this may not always be the reference sample, 195 though the statistical weight of the evi
case in such investigations. It was also possible to deter dence will be less, and comparison with a sibling's genotype
mine the approx imate age range of the re mains, which was may also be helpful. In aeroplane traged ies, families are
of further benefit. Ten of th e exhumed bodies were believed often present together on the same flight and the common
to be sibling pairs and 11 were unrel ated to each oth er. 191 ality of alleles across the generation can be used in identify
Despite the bodies hav ing been burnt, buried an d ing family members.
exhumed, DNA analysis succeeded in matching 18 of the The 9/11 disaster of 2001, when 2792 people were killed
remains with a known ch ild. In two of the cases mothers in a terrorist attack on the World Trade Center, must surely
were able to provide reference mate ria l in the form of stored have been one of the most challenging investigations, and
umbilical tissue from their child, which had been kept in those w ith the task of recoveling and identifying the remains
accordance with Philippine custom an d which provided a faced lo gistical, technical, practical, scien tific and medical
positive identification . 192 difficulties. Although direct matches with referen ce samples
Iden tificatio ns of the other children, for whom ante were used where possible to identify the remains, other DNA
mortem reference sa mples were not ava ilable, was achieved techniques, including the paternity approach, were also
by patern ity-type ana lysis using autosomal STR profiling, employed. 196
as well as by comparison ofY haplotypes, by analysing DNA It will be appreciated that in paternity analysis there
reference samples from 10 mothers, three fathers and a is always the chance of a coincidental match. A particular
paternal grandfather of two children for whom a paternal all ele at a locus is shared not only by a parent and child,
sample was unava ilab le. Y haplotypes were valuable in iden but also by approximately 1 in 1000 random pairs of indi
tifying not only father/grandfather to son/grandson rela viduals, so an attempt was made to minimize the number
tionships but also those of sibling brothers. 191 of false-positive matches obtained during the identification
In the Manila fire tragedy, the remains had been subject process. 197 The number of false-positive results is related to
to extreme conditions so it was no t su rprising that a full the size of the reference list and the number of victims, and
profile was not obtained for evelY sample typed. Where therefore is a major consideration in investigations such as
Identification of an unknown profile I 413
the 9111 tragedy, but it should also not be overl ooked in nationa l DNA database can be searc hed agai nst the profile
investigations invol ving smaller numbers of victims. in question or from a woman suspected of being the
As men tioned previ ous ly, the state of the remains can mo ther during the investigation.
affect the amount of information that can be obta ined from Placental material is an ideal source of maternal DNA in
STR analysis. Poor-qu ality DNA may not provide results for such circumstances, but it may not always be available. In
the alleles present at all loci typed, or it may resu lt in iden another repolied case, a live newborn was fou nd abandoned
tifica tion of only a sin gle allele in a hete rozygous individual in a box but no pl ace ntal materia l was found with the baby.
when there really should be two alleles; one always needs to It was, however, possible to obtain a profile of the mother
be aware of the possibility of this so- ca lled allelic dropout. In by extracting maternal DNA from vernix caseosa on the ha ir
many cases, the reason for poor SIR resu lts is that the STR and body of the newborn, which conta ined maternal
fragment in the sa mple is no t intact along its whole length, blood. 83 One should be aware of the potential for such mate
as it should be. As DNA degrades it fragments into smaller rial being relevant and useful in similar investigations.
pieces a nd breaks may occur within the SIR; thus, the true A more problematic scenario is the case of the abandoned
alleles are not detected. It may be possible to design PCRs for baby fou nd long after deli vely. The rem ains may have
STR profiling that look at smaller sections of DNA, 198 and decomposed, with the concomitant difficulties of obtaining
commercial kits adopting this approac h are no w available ; DNA profiles discussed previ ously, or there may be circum
the sma ller t he targeted DNA being identified, the greater the stances in which the deceased has been concealed or the
chance it remains intact. Where there are deficiencies in SIR environment has been conduci ve to mummification. As men
typing, mtDNA may be sequenced, as discussed ab ove, or the tioned earlier, DNA profiling of desicca ted material is prob
in format ion gathered from partial STR profiles may be sup lematic. [n the case of a mummified baby, there is the
plemented by, or rep laced, with SNP data. 199 SNPs are single addition al issue that bone and teeth, a commonly used source
base sequence variations between indiv iduals at a particular of DNA in the case of remains, will be unavailable if the
positio n in the gen ome. These sites are ab undant (estimated neonate's bone is cartilagi nous and teeth have not erup ted.
10 million) throughout the genome, but because they are Despite the difficulty in obtaining suitable genetic informa
mostly biallelic they lack the polymorphic variation across tion from desiccated tissue, profiles have been obtain ed from
the population that SIRs exhibit. 2oo However, if one geno mummified tissue from babies, though in these cases there
types a DNA sample at many SNPs, one can approach the was no accompa nying decomposition of the bodies, which
discriminatory power of SIR profiling. Estimates vary as to may have contri buted to the successful results.
how many are needed, but it is likely that between 50 and [f no maternal DNA is found with the fet al remains but a
100 may be required,199 consistent with the 70 used in rela DNA profile of the baby can be obtained, genetic analysis
tion to the 9/11 investigation. 196 resorts to the pedigree type of analysis: comparing the baby's
profile with those of suspected parents in a paternity testing
approach to look for consistent transmission of alleles. As
IDENTIFICATION OF THE 'ABANDONED BABY' stated above, the statistical weight of such an analysis is less
OR FETAL MATERIAL AND AVENUES than if information is available from at least one parent. 195
FOR IDENTIFYING THE SOURCE The problem is obtaining DNA from a suspected mother
OF AN UNKNOWN PROFILE when there may be no lead to the parents. However, this is
true of any investigation centring on DNA. An SIR profile
The approac hes described above, which are used to identify alone provides no meaningful information other tha n the
the remains of a ch ild , ha ve been geared towards identify sex of the individual. What fo llows desc ribes possible means
ing a previously know n livi ng individual, either by directly of identifying a candid ate so urce when there is no match
matc hing a profil e with a reference sam ple from the indi with existing profiles on national DNA datab ases or with
v idual or through pedigree analys is. However, one may be suspects in the case, these being the first two options to
faced with the scenario of an aba ndoned newborn or fetus consider.
requirin g investigation. In such a scenario t here will be an Typ ing the less specifi c mtDNA or Y chromosome might
endeavour to iden tify the mother. Whethe r maternal DNA be of assistance in spread ing the net wider: identifying a
can be recovered from the scene to ass ist will depend on match might go so me way in identifying a possible pedigree
the post-partum circum stan ces and the conditi on in which that may lead to a parent, but this type of investigation may
the newborn is found. also lead to nothing. A similar technique, so-called 'familial
Ma ternal DNA has successfully been recovered and pro searching', is also increasin gly being used. In this case, rather
filed from placental material found abandoned with a than searching existing databases for a profile that matches
wrapped dead newborn, as well as from blood found on the all alleles at all loci, a less stringen t search is employed, allow
wrappin g material. Using the profile, it was possible to in g profiles with some allelic mismatches to come to light.
determine the mothe r of the baby in this case. 83 Obviously, The logic beh ind this approach is that such profiles might
this outcome requires that a reference sample is available belong to relatives of the true source, who would be expected
fro m the mother fo r comparison. For exam ple, entries on a to share more all eles in common than a random individual,
--. ~;
414 I Foren sic DNA profiling
and it may be possible to home in on the true source of the DNA DATABASES
50 Lu cy D. DNA I. In Introduction to Sra listics for Forensic 71 Thom so n WC, Ford S, Doom T et aJ. Evaluating foren sic
Scientists. Chichester: John Wiley and Sons, 2005. DNA evidence, Part 2: Essential elements of a co mpe tent
51 Lucy D. Eva luation of evidence in practice. In Introduction fO defense review. Champ ion Magazi ne, 24-28 May 2003,
Statistics fo r Forens ic Sci entists. Chichester: John Wiley and www.nacdl.org/_852 570 IC00653 9F8.nsf/0/9 C5F 7C6D724A4
So ns, 2005. 7518 52 56E540074C156?Open (accessed April 2007).
52 Evert IW. Towards a uniform fram ework fo r reporting 72 Teichroeb R. .Rare look inside state crime labs reveals
opini ons in forensic science casework. Sci Justice 1998; recurring DNA test problems. Seattle PosHntelligencer,
38: 198-202. Thursday 22 July 2004, http://seattlepi.n wsource.com/
53 Forensic Science Service, Association of Police Surgeo ns and local/183007 _crimel ab22 .html (accessed April 2007).
Association of Chief Police. Officers' Guidelines for the 73 Pompa non F, Bonin A, Bellemain E, Taberlet P. Gen otyp ing
Collection of Specimens. London: Forensic Science Service, errors: causes, consequences and so lution s. Na t ReIJ Genet
2000. 200 5; 6:847- 59.
54 Inman K, Rudin N. Good field practice - processing a crime 74 Faigman DL, Kaye DH, Saks MJ, Sand ers J. Modem
scene. In Principles and Practice of Criminalistics: the Scientific E,Jidence: The Law and Science of Expert
Profession of Forensic Science. Boca Ra ton, FL: CRC Press Testimony. 25-1.2 DNA Typing; Th e legal rele/Jance af DNA
LLC, 2001. tests: Objections ro DNA eIJidence. St Paul , MN: West
55 National Criminal Justice .Reference Servi ce. DNA evidence: Pu blishi ng Co., 2002 , pp. 210-36.
what law en forcement officers should know. NCJ 200908. 75 Stevens -Simon C, Nelligan D, Breese P et al. The prevalence
Nat In st Justice J 2003; 249: 10- 15. also Available online: of geni tal human papillomavims infections in abused and
www.ncjrs.gov/pdffilesl/jr000249c.pdf (accessed April 2007). non ab used preadolescent girls. Pediatrics 2000 ; 106:645-9.
56 Kaye DH, Sensabaugh Jr GF. Reference guide on DNA 76 Thompson C The GUiltl/Paediarric il/ le /joc e. Presentation at
evidence. In Reference Manual on Scienfific Evidence, 2nd edn. 22nd lusn Europe Meeting. Versaill es, Abstract 0.048, 2006.
Washington, DC: Federal Judicial Center, 2000, pp. 485-57 6. 77 Kl aus ner JD. The NAAT is out of th e bag. Clin Infect Dis
57 Fa igman DL, Kaye DH , Saks MJ , Sanders J. Modem 2004 ; 38:820-1.
Sc ientific Evidence: Th e Law and Science of Expert 78 Hammerschlag MR. Use of nucleic acid amplification test in
Testimony. 25 -2.5 DNA Typing; The scientific statu s af DNA inves tigating child sexual ab use. Se.r Tran sll1 Infect 2001;
typing: Laboratory performance. St Paul , MN: West 77: 153-7.
Publishing Co., 2002, pp. 257-63. 79 Hammerschlag MR. Appropriate use of nonculture tests for
58 Lynch M. The discursiv e production of uncertainty: the the detection of sexually transmitted disea ses in children
OJ Simpso n 'Drea m Team ' and the sociology of knowledge and adolescents. Semin Pediatr Infect Dis 2003; 14:54-9.
machin e, Soc Stud Sci 1998; 28 :829-68. 80 Kellogg ND, Baill argeon J, Lukefahr JL et al. Comparison
59 Scientifi c Testimo ny, an onlin e journal: DNA tes ting of nucleic acid amplification tests a nd culture techniques
problems. Avail abl e online: www.scientific.org/ in the detect ion of Neisseria gonorriloea e and Chlamydia
DNAProblems/DNA-Problems.htm (acc essed April 2007) . trachoma tis in victims of suspected child sexual abuse.
60 Toledano T, Quarino L, Leung S et al. An assessment of DNA J Pediatr Ado/esc Gynecol2004; 17 :33 1-9.
contamination risks in New York City medical ex amin er 81 Tho mas A, Forster G, Robinson A, .Rogstad K. National
facilities. J Forensic Sci 1997 ; 42:721-4. Guidelines on th e Management of Suspected Sexually
61 Rutty GN. Human DNA contamination of mortuaries: does it Transmitted Infectio ns in Children and Young people.
matter? J Palh al 2000; 190:410-1 I. Clinical Effectiven ess Group (Assoc iation of Genitourinary
62 Rutty GN, Watson S, Davison 1. DNA con tamin atio n of Medicine and th e Medical Society for the Study of Ven ereal
mOltuary instruments and work surfaces: a significant problem Diseases), 2002. Availabl e on lin e:
in for ensic practice? In l J Leg Med 2000; 114:56-60. www.bashh.org/guidelin es/2002/
63 Lowe A, Murray C, Whitaker J et at. The propensity of ado lescencfi nal_0903.pdf (accessed April 2007).
individuals to deposit DNA and second ary transfer of low-level 82 2001 Nati onal Guidelines on the Management of Adult
DNA from individuals to inert surfaces. Forensic Sci lilt 2002; Victims of Sexual Assault. Clinical Effectiveness Group
129:25- 34. (Ass ociation of Genitourinary Medicine and (he Medica l
64 Rutty GN. An investi gati on into the transference and Society for the Study of Venereal Diseases). Avail able online:
survi vabili ty of human DNA follo win g s imul ated manual www.bas hh .o rg/gu idelines/2002/sexassauIC0601.pd f
strangulation with consideration of the problem of third (accessed April 2007).
party con taminat ion. lnt J Leg Med 2002; 1J 6:1 70-3. 83 Csete K, Beer Z, Varga T. Prenatal and newborn paternity
65 Sullivan K, Johnson P, Rowlands D, Allen H. New testing with DNA analysis. Forensic Sci Int 2005 ;
devel op ments and challenges in the use of the UK DNA 147:S57-60.
database: addressing the issue of contaminated consumables. 84 Robino C, Barilaro MR, Gino Set al. Inces tuous paternity
Forensic Sci Int 2004; 146:5175-6. detected by STR-typing of chorionic villi iso lated from
66 Peel C, Gill P. Attribution of DNA profiles to body fluid archival formalin-fixed paraffin-embedd ed abortion
stains. lnl Congress Ser 2004; 1261 :53-5. mate rial using laser microdissection. J Forensic Sci 2006;
67 Bond JW. Value of DNA evid ence in detectin g crime. 51 :90-2.
J Forensic Sci 2007; 52:128-36. 85 Christian CW, Lavelle JM, De Jong AR et a!. Forensic
68 Gill P, Kirkham A. Development of a simulation mod el to evidence findin gs in prepubertal victims of sexua l assault.
ass ess the im pa ct of co ntamination in casework using STRs. Pediatrics 2000; 106: 100-4.
J Forensic Sci 2004; 49:485-91. 86 American Academy of Pediatrics (Committee on Child Abuse
69 Rutty GN, Hopwood A. Tucker V The effectiveness of protective and Neglect). Guidelines for the eva lu ation of sex ual ab use
clothing in the reduction of potential DNA contaminati on of of children: subj ect revi ew. Pediatrics 1999; 103:186-91.
the sce ne of Clime. ln t J Leg Med 2003; II 7: 170-4. 87 Allard JE. The collecti on of data from findings in cases
70 Butler JM. Appendix N: DNA Advisory Board Quality of sexual assau lt and the sig nificance of spermatozoa
Assuran ce Standards. In Forensic DNA Typing, 2nd edn. on va gina l, an al and oral swabs. Sci Ju sti ce 1997;
Ams(erdam: Elsevier. 2005. 37:99 - 108.
---
418 I Forensic DNA profiling
131 Tully LA, Parso ns TJ, Steighner RJ et aJ. A sensitive denaturing 151 Ottenberg R. Medicoleg al application of human bl oo d
gradi ent-gel electro phoresis assay reveals a high frequ ency of grouping.] Am Med Assoc 192 1; 77:682-3.
heteroplasmy in hyp erva liabl e region I of the human mtDNA \ 52 Shaw MW. Paterni ty determination 1921 to 1983 and
co ntrol region. Am j Hum Gel1et 2000; 67:432-43. bey ond . jAm Med Assoc 1983; 250:2 53 6-7.
132 Paneto GG, Martins JA, Lon go LVG et a!. Heteroplasmy in 153 Albrecht K, Sch ulth eiss D. Proof of paternity: historical
hair: differences am ong hair and bl oo d From the same refl ect ions on an andrological-forensic ch al lenge.
individuals are still a matter of debate. Forel1s ic Sci lilt 2007; Andrologia 2004 ; 36:3 1-7.
173: lJ 7-21. 154 BBe. Embryo mix-up at IVF hospital. Monday 28 October
133 Gill P, Ivanov PL, Kimpton C et aJ. Id en tifi cation of th e 2002. http://ne ws. bbc. co.u k/ l/hi/h ea lth/2 36770 5.st m
remains of the Romanov family by DNA analysis. Nat Gel1et (accessed Ap ril 2007).
1994; 6:130- 5. 155 BBe. $1111 pay-outJol' IVF embryo mix-up. Thursday 5
134 Ivan ov PL, Wa dham s MJ, Roby RK et al. Mitochondrial DNA August 2004. Available onl ine: ht tp://news.bbc.co.uk/
sequence hetero plasmy in the Grand Duke of Russia Georgij l/hi/he alth /353 7950.stm (accessed Apri l 2007).
Romanov establishes the authenticity of the remains of Tsar 156 BBe. NF mix-ups 'occur regularly'. Wedn esday 24 July
Nich olas II. Nat Gellet 1996; 12:417-20. 2002. Ava il ab le online:
135 C6li ca MY, Rodriguez Card ozo MB , Abovi ch MA et aJ. http://news.bbc.co.uk/ l/hi/health/21484 23 .
Paterni ty cases when the alleged father is missing. still (accessed April 2007).
Int Co ngress Ser 2004; 1261 :520- 2. 157 Dy er O. Black twins are born to white parents after infertility
13 6 Mayor LR, Bald ing DJ. Discrim in ation of halF-sib lings when treat ment. BM] 2002; 325:64.
maternal genotypes are known. Forel1sic Sci lilt 2006; 158 Spriggs M. lVF mixu p: white couple have bla ck ba bi es, ]
159: 141-7. Med Ethics 2003; 29:65.
137 Tra cey M. Sho l1 tandem repeat-based identification of 159 BBe. ORA: Legal minefi eld of IVF 'm istake '. Monday 8 July
indi vid ual s an d parents. Croatian Med j 2001; 42 :23 3-8. 2002 . Avai lable on line: http://news.bbc.co.uk/ I/hi /health/
138 Linacre A, Goo dwin W. Role of and techni ques in fo rensic 2116589.stl11 (accessed April 2007).
hem atogentics - paterni ty testing. In Payne-James J, Busutti l J 60 BBe. Court rules on fath er of IVF mix- up twins. Wedn esday
A, Smock W (eds) Fo rensic Medicille: Clinical al1d 26 Febru ary 2003. Avail ab le onlin e:
Pathological Aspects. London: Greenwich Med ica l Media Ltd, http:// news.bbc.co. uk/ I/hi/
2003, pp. 677-88. hea lth /2797J69.stlll (a ccessed Apri l 2007 ).
139 Fun g WK, Hu Y-Q , Ch un g Y-K. On statistica l analysis of 161 McKonkey E. Lectu re 2 - the mechanisms of mutation: how
forens ic DNA: th eory, methods and co mputer programs. the hum an ge nome cha nges. In How tiJe Hum an Genome
Forel1s ic Sci lilt 2006 ; 16 2:17-23. Works. Sud bury, MA: Jones and Bartl ett Publishers, 2004,
140 Coleman H, Swenso n E. DNA ill the Courtroom: a Tmil pp. 19-33.
Watcher's Gu ide. Washin gto n, DC: GeneLex Press Seattle, 162 Brin km ann B, KJintscher M, Neuhuber F et al. Mutati on rate
2000. Avai lab le on line: www.genelex.com/ in human microsatellites: influence of the structure and
pate rnitytesting/pate rnityboo k.ht ml (accessed Ap ri l 2007). length of the tand em repea t. Am j Hum Gel1 1998;
141 Relationship Testin g Program Unit 2004. Annual Report 62:1408-15.
Summary for Testi ng in 2004. Ava il abl e online: 163 Saja ntil a A. Lukka M, Syvanen A- e. Experimentally
www.aabb. 0 rglD ocu men ts/ Accred ita tio n/Pa ren tage_Testin g_ observed germ line mutatio ns at hum an micro- and
Accreditation]rogram/rtannrpt04.pdf (accessed April 2007). minisateJlite Jo cl. Eur] Hum Genet 1999 ; 7:2 63 -6.
14 2 vo n Wurmb- Schwa rk N, M<ilyusz V, Simeoni E et aJ. Possible 164 Kayser M, Roewe r L, Hed man M et al. Characteristic and
pitfalls in motherless paternity ana lysis with related pu ta tive frequency of germline mutations at microsatellite lo ci from
fa thers. Forensic Sci Int 2006; 159:92-7. the human Y chromosome, as revealed by direct observation
143 Beren t J, Jacewicz R, Szram S. Effect on paterni ty index of in father/son pairs . Alii j Hu m Genet 2000; 66:1580- 8.
substituting the biological fath er by hi s brother 165 STR DNA Interne t Data base Mutation Rates for Common
preliminary repol1. Int COl1gl'ess SrI' 2004; 1261 :476-8. Loci. Ava ilabl e onlin e: www.cstl.n ist.gov/biotec h/strbase/
144 Hartl DL, Jones EW. Population ge neti cs and evo lution. In mutati on.htm (accessed Apri l 2007).
Essential Genetics: a Genomics Perspecrive, 3rd edn. 166 Yamamoto T, Tam aki K, Hu ang XL et al. The application of
SudbUly, MA: Jones and Bartlett Publishers, 2002 . minisatellite variant repea t mapping of PCR (iVIVR-PCR) in a
145 McKo nkey E. Lecture 3 - Gen etic diseases : th e co nsequ ences paternity case showin g fals e exclu sion du e to STR mutation.
of mutation. In How fiJ e Human Genome Works. Sud bury, ] Forel1sic Sci 2001; 46:374-8.
MA: Jones and Bartlett Pu blishers, 2004, pp. 49-50. 167 Brandt- Casad eva ll C, Gene M, Piqu e E et al. Presence of two
146 Sasaki H, lshino F. Gmomic Imprintil1g. Basel: Karger, 2006; mutations between fathe r and child in two cases of paternity
11J[ 1-4). tes ting. 1111 Congress Ser 2003; 12 39:S 15-1 8.
147 Naito E, Dewa K, Fukuda M et al. Novel paternity test ing 168 Nutini Ai, Mariottini A, Giunti L et al. Double
by distin gui shing parema l alleles at a VNTR locus in the incompa tibili ty at human alpha fibrinogen and penta E loc i
differentially methylated region upstrea m of tile hum an in pa te rnity testing. Croatian iVIed] 2003; 44:34 2- 6.
H 19 gene. ] Forensic Sci 2003; 48: 1275-9. 169 Thangaraj K, Reddy AG, Si ngh L. Mu ta tion in th e STR locus
148 Zhao G, Yang Q, Huan g D et al. Study on the app li cation of D2 1S 11 of father caus ing all ele mi smatch in the child.
parent-of-origin specific DNA methylation markers to ] Forel1s ic Sci 2004; 49:99-103.
forensic geneti cs. Forensic Sci In t 2005; 154: 122-7. 170 Jun ge A, Brinkmann B, Fimmers R, Madea B. Mutations or
149 Fraga MF, Ballesta r E, Paz MF et al. Ep igenetic differe nces exclusion: an unusu al case in paterni ty testing. Int] Leg
arise durin g th e lifetime of monozygotic twins. Proc Natl Med 2006; 120:360-3.
Acad Sci 2005; 102:1 0604-9. 171 Crouse CA, Rogers S, Ami ott E et a1. Analysis and
150 Co rach D, Pen ac ino G, Marino Met al. Incestuo us offspring interpretation of sh ort tand em rep eat microvariants and
detection inference by VNTR homozygosity in crement. lilt three-ba nded allele patterns usin g multiple allele detection
Co ngress Sel' 2003; 12 39:861-4. systems. j Fore nsic Sc i 19 99 ; 44 :87 -94.
References I 419
172 Clayto n TM , Guest JL, Urq uhart AJ, Gill PD. A ge netic bas is 193 Meye r HJ. The Kaprun cab le ca r fire disaster - aspecrs of
for ano malous band patterns enco untered during DNA STR forensic organ isation t'ollowing a ma ss fata lity with 155
profiling. J Forensic Sci 2004; 49: 1207-14. victims. Fore nsic Sci fJI t 2003 ; 138: 1- 7.
173 Sudbery P. Hu man genetic disease. In Hum lln Molecu lllr 194 Clayton TM. Whitaker JP, Maguire CN. Id en tificati on of
Genetics, 2nd edn. Harlo w, UK: Pea rso n Ed ucat ion. 2002 . bodies from the sce ne of a ma ss disaster lI sin g DN A
174 Dauber EM, Wenda S, Glock Bet aJ. Mosa icism as a poss ibl e amplification of sholt ta ndem repeat (STR) loci. Forel1sic Sci
reason for poor amp lification of ame logenin-Y in three fl1t J995: 76:7 -15
hum an male indiviciua ls. fl1l Congress SCI' 2004: 195 Hsu CM, Hu ang NE, Tsai LC et al. id entificatio n of victims of
1261 :508-10. the 1998 Taoyuan Airbus cras h accid ent using DN A analys is.
175 Farber CM. Georges M, De Bock G et aJ. Dem ons tration of JJIt J Leglll Mer! 1999; 11 3:43 - 6.
spo ntaneous XX/XY chimeri sm by DN A fin ge rprinting. J96 Pres iden t's DNA initiative. Lessons learn ed from 9/ll: DNA
Hum Genet 1989; 82:19 7-8. identification in mass fatality inciden ts, 200 6. Avail able
176 Yu N, Krus kall MS, Yun is JJ et al. Disp uted maternity online: www.ncj rs.gov/pdftil es I/110/214 78 1.pd f (acc essed
leading to identificatio n of tetra ga metic chimerism. N Eilgi J Ap ril 2007).
Med 2002; 346:1545. 197 Bre nner CH. We ir BS. Issues a nd strateg ies in the DN A
177 Mayne E, Borland S. The moth er with three chi ldren who identification of World Trade Ce nter victims. Theor Popul
don' t share her DNA. Ma il all SUlidav. 5 March 2006. Bio i 2003; 63:173 -8.
17 8 Mild e A, KJ1I1-Burm eiste r R. Ritz-Timme S. Ka atsch HJ. DN A 198 Coble 1\110, Butler J1\I1. Ch'lracterization of new miniSTR loci
typin g in cases of blood chimeri sm. flit J Leg Med 1999; to aid ana lysis of deg raded DNA. J Foreusic Sei 2005;
11 2:333 - 5. 50:43 -53.
179 Rub ock i RJ, McCue BJ, Kelly MT et aJ. Natural DNA 199 Butler J iVI. Single nucleotid e polymorphis\11s ami oth er
mixtu res ge nerated in fraternal twin s in utero. J ForellSic Sci bi -allelic mark ers. In Forellsic DNA T)'pin g, 2nd edn.
200 1; 46: 120-5. Amste rdam: Elsevier, 2005.
180 Dau ber EM, Fat' i, Stad lbache r S et al. STR typ in g in a pair 200 Sob rin o B, Brion lVI, Carracedo A. SNPs ill fore nsic genetics:
of chimeric tw ins. In( Congress Ser 2003: 1239 :569 - 71. a review on SNP typ ing methodologies. Forens ic Sci fni
181 Dau ber EM, Dorner G, Mitte rbau er Met al. Discrepant results 2005 : 154:1 8[ - 94.
of samples taken from diffe rent ti ss ues of a singl e ind ividual. 201 Fore nsic Science Se rvice Casefiles : Craig Harman - Family
jilt COllgress Ser 2004; 1261 :48- 9. DNA link offers crime breakthrough. Tile Forensic Science
182 Dauber EM, Muller CJ. Schii niger-H ekele M et al. Artificial Sen'ice Website, www. forensic.gov. uk / forensic_t/inside/
blood chimerism and grafl-versus-host disease after liver news/lis c casefiles.php?case=24 (accessed April 2007).
transpl anta tion. ll1 i Congress Su 2006; 1288:840- 2. 202 Beiber FR. Guilt by associa tion. Ne/U Scientist 2004;
183 Pop e S, Chap man I-I, Lam bert J. The effec t of bon e marrow 2470:20.
transp lants on DN A profi les : a case exam ple. Sci Justice 203 Gree ly HT, Riordan DP. Garrison NA , Mou ntain JL. Family
2007; 46:23 1-7. ties: the use of offen der da ta bases to catch offen der's kill.
184 von Wurl1lb-Schwark N. Bos inski H. Ritz-Timme S. What do J Law M ed Erliics 2006; 34: 248- 62 .
the X and Y chromosomes tell us about sex and gende r in 204 Haimes E. So cia l and ethical issues in the USe of familial
foren sic case analysis? J ForeJIsic Legal Mer! 2007: searchi ng in forensic investigations: insi ght from famil y and
14 :27-30. kin ship stu dies. J Law Med Ethics 2006; 34: 263 - 76.
185 Aldhous P. Bone marrow donors risk DNA id enti ty mix-up. 205 Grimes EA, Noake PJ, Dixon L. UrlJ uart A. Sequ~nce
New Scientist 2005; 2523: 11. polymorphism in the human l11 elanocoltin I rece ptor geJl e
186 Rodri guez Cardozo MB , Co lica MV, Abov ich fVIA et aJ. as an indicator of the red hair phenotype. Forensic Sci flit
Maternity testing in a ch imerical child . jJIt COll(Jress SCI' 200 1; 122: 124- 9.
2004; 126 1 :502-4. 206 Rees JL. Geneti cs of hair and sk in co lor. AIIII Re/J Genet
18 7 Foster EA. Job ling MA. Tay lor PG et aJ. Jefferson fathered 2003; 37 :67 -90.
slave's child. Nature 1998; 39 6:27-8. 207 Uil11aso n RL, MohicJeen MPK, Mes t JR et al. SLC24A5, a
188 Marshall E. Wh ich Jefferso n was th e father? ScieJIce 1999; pu tative cation excha nger, affects pigmentation in zebra fi sh
283: 153- 5. and huma ns. Sciellce 2005: 310: 17 82-6.
189 Check E. Jefferso n's desce nd ants continue to deny slave link. 208 Cho MK. Sa nk a r P. Fore nsic genet ics and ethica l legal and
Nature 2002; 41 7: 213. social implicati ons beyon d th e clinic. Nat Genet 2004;
190 Pa abo S. Molecular cl on in g of anci ent Egyptian mumm y 36:S8 - J 2.
DNA. Naiure 1985; 314 :644- 5. 209 Ossorio PN. About face : fo ren sic genetic test ing for race and
19 1 Ca laca l G, De lfin F, T'111 M et aJ. Id entificatio n of ex hum ed v isible trai ts. J Law Med Ethics 2006; 34:277-92.
rema in s of fire tragedy victims using conve ntio na l method s 210 Will iams R. Johnson P. Inclusi veness, eITectiveness and
and autosomallY-chromosomal short tandem rep eat DNA intrusiveness: issues in the develop ing uses of DNA profilin g
pro filin g. Am J Forensi c ivied Patlw12 005; 26:285- 9 1. in supp ort of cri minal investiga tions. J Law /Iiled Eth ics
192 Ca lacal G, De Ungria MC, Delfin F et al. Iden titlca tion of two 2006; 34:234-47.
fire vict im s by co mp arat ive nuclear DNA ty ping of skeleta l 211 Alva rez M, Ballantyne J. Th e ide ntification of newbo rn s
remains and stored umbili ca l tissues. AI1I J Forensic Med usin g messenger RNA profiling analys is. Anlll Biochem
Patilol 2003 ; 24 : 148- 52. 2006; 357:21-34.
I CHAPTER 22 I
THE DENTIST'S ROLE IN CHILD ABUSE
AND NEGLECT
David Whittaker
INTRODUCTION
it is not the responsibility of general dental practitioners
(GDPsl to diagnose child abuse. B Dental surgeons are, how
Damage to a child, other than by accidental means, has been ever, in a good position to diagnose NAI to the face and
recognized for centuries I but app ears to have firs t been oral structures and they should be aware of the possibility
described in a formal manner in 1962. 2 The authors used of abuse and have some knowledge of the key factors
the term 'batte red child syndrome' but this forms part of involved in its diagnOSis. The diagnosis of NAI is difficu lt
what is more commonly called 'child abuse'. Ch ild abuse to make with certainty. and va ri ous pathological or post
may include non-accidental injury (NAIl but also emotional mortem changes have been mistakenly identified as child
ab use, sexual abuse and neglect. It has been increasingly abuse.9 Conversely, child abuse or neglect has gone unde
recognized that the denta l surgeon may be the first profes tected in some unexp ected deaths in children 3 or in re po rts
sional to suspect physical abuse, especially in relation to of facial or intraoral injuries. 3 The incidence of NAJ to the
injuries involving the orofacial structures. 3 A child is consid head and neck region is high, v arying between 11 and 50
ered to have been abused if he or she is treated in a way that per cent of cases. 10,11 Therefore, the dental profession com
is unacceptable in a given culture at a given time. 4 For the prises an important part of health professionals involved
purposes of this review, only physical abuse in the form of in identifying and reporting child abuse or neglect. 12,13 A
NAI and neglect will be considered . Neglect occurs when an number of key indicators of NAt have been identified . 14
adult knowingly allows a child to endure pain or suffering or The more of these that are present the more likely it
fails to provide the basic requisites for proper maturation. becomes a defini tive dia gn osis:
Since the syndrome was specifically named it has • delay in seeking help;
become possible to add fu rther injuries to the pattern of • a vague history or a history that v aries between
injuries described. These further injuries include bite interviewees;
marks,s which may be found anywhere on a child's b;dy. • a history incompatible with the injury seen;
Bite marks in children rep resent child ab use until proven • abnorma l behaviour of the parents;
otherwise. They are rarely acci denta l and are good indica • the child's app earance, e.g . withdrawn, frightened or
t ors of genuine child abuse. 6 There have been ap proxi sad ;
mately 50 articles in the international dental li terature • other signs of abuse, such as neglect or deprivation;
relating to child abuse and bite marks. The dentist's rol e in • dis closure of an incident by the child.
diagnosis and management was described in 1986. 7 How
ever, the problem is still not well appreciated by the profes The diagnosis of NAI is discussed in greater detail in
sion at large. Apart from the specific injury of bite marks, Chapter I.
Introduction I 421
The prevalence of child abuse is reported differently in out that the relationship between parent and child, the child's
different countries, presumably reflecting social and cultura l reaction to other people, the child's reaction to a dental or
differences. 15 The true extent in a given population is diffi medical examination or the general demeanour of the child
cult to determine, but in 1967 in London it was believed that may have a bearing on possible child abuse.
at least 300 cases occurred per year. 16 In terms of mortality, It is important that the dental practi tioner without direct
in the 1980s at least 4 children in Britain 17 and 80 children extensive experience in referral techniques refer cases to
in the USA 18 died weekly as a result of abuse or neglect. It the local consultant in paediatric dentistry, the local con
was reported in 1988 in Britain that at least one child in sultant in dental public health or an appropriate consultant
every thousand under 4 years of age suffers severe physical paediatrician. It may be necessary and wise to discuss any
abuse each year. This might include fractures, brain haemor potential referral with the general medical practitioner of
rhages, severe internal injUli es or mutilation. 19 In 1991, an the child concerned. Under the 1989 Children Act and
average of 4.2 children out of every thousand in the UK were Working Together (1991) the local authority has a legal
known to have been physically abused , the highest rate duty 'to make or cause to be made' such enquiries (by
being seen in children less than 1 year of age. 20 It is thought another agency, such as the National Society for the Pre
that one in 10 000 children in Britain will die as a result of vention of Cruelty to Children [NSPCC]) as t hey consider
abuse, although many workers consider mortality rates to be necessary, to enable them to decide whether t hey should
considerably higher. It is generally believed that proven take any further action to safeguard the child 's welfare.25 ,26
cases of child abuse may be four or fi ve times as There are a number of reasons why general dental practi
frequent as t hey were some 10 years ago. 4 In 1974, 97 loca l tioners or their staff may fail to report potential child abuse.
authorities in the UK were approached by questionnaire in a Therefore, each local area should have a simple system,
survey covering 90 per cent of the population. Some 5700 preferably referrin g t he case through the normal referral sys
cases came to light in the last 9 months of that year with a tem that dentists might use, such as to their local paediatric
mortality of 0.7 per cent and a significant risk of re-injury.21 dentist or paediatric consultant. Failure to refer a child may,
In the USA, it is mandatory for health-care workers to of course, have serious consequences. Referral rates by GDPs
report suspected cases of child abuse. The second National are not known, but it is known that healt h professionals in
Incidence a nd Prevalence Study of Child Abuse and general refer 17 per cent of suspected abuse cases, whereas
Neglect22 reported that 4.95 children in every thousand were teachers and school nurses refer 23 per cent of cases,u
physically abused. Girls were more li kely to be sexually Reporting cases of child abuse is mandatory in the USA,
abused than boys, although there was no gender difference where 10 subtypes of child abuse and neglec t have been
for physical abuse. Low family incom e is significant, and classified 28 The most frequent of these is physical abuse,
children of Afro-Caribbean origins were more likely to be with an incidence of a lmost 32 per cent, and this is closely
physically abused than other racial groups. Sexual abuse is followed by educational neglect (28 per cent) and byemo
more common in Caucasian children. Whether or not the tional a buse (26 per cent).
true occurrence is increasi ng, it is clear that increased aware Suggestions as to how t he dentist should examine a child
ness results in many more cases being reported and, on the thought to be suffeling from child abuse or child neglect have
basis of current statistics, an individu al dental surgeon may been made. 29 These include the relationships between the
be expected to encounter a case at least once every 3 years. child and his or her parents; the cleanliness, stature and
In spite of this, British dentists have not formally been nutritional status of the child; the quality of his or her cloth
requested to develop an awareness of NAls to children, and ing; and the presence of physical problems, such as a limp or
many health authority child abuse procedure documents difficulty in climbing into the dental chair. It is recommended
make no mention of the possible role of the dentist. 23 that the body be systematically examined for traumatic
Since the Children Act (1989) and Working Together injury, bruises, previous scars, burn marks or injury caused by
(1991), child welfare services in the UK are required to have foreign bodies. If the dentist's suspicions are aroused, the par
procedural guidelines to deal with suspected child abuse ents should be informed that a n injury has been observed; the
cases. These app ly to aU persons working with children, dentist must then contact the approp riate child abuse or child
and general denta l practitioners are includ ed in this group. neglect authority. In the UK, it has been said that it is not the
National guidelines do not exist for GDPS,8 and the document responsibility of GDPs to make the diagnosis of child ab use,
Working Together (1991) gives very little information about nor should they ever do SO.8 However, the same authors state
the role of the GDP. The British Dental Association in its t hat no GDP should feel gui lty abou t referring children with
booklet Ethics in Dentistl)! does produce guidelines for child suspected child abuse. This does not mean that they are
protection 24 but it suggests that each individual GDP should accusing either parent or carer but they are simply seeking a
meet the guidelines laid down through their local district second opinion on a condit ion that is outside their expertise.
health authority. It is important that GDPs keep adequate and Failure to refer a child in need may result in more severe
accurate up-to-d ate dental records of all their patients; in the injuri es in t he future. The failure to follow up a suspicion of
case of child abuse, these may need to be referred to in case child abuse by a professional health-care worker may be
conferences or even court procedures. The guidelines point construed as a form of professional negligence. Accidentally
422 I Dentist's role in child abuse and neglect
acquired bruises or pseudobntises must be differentiated from neglect. 32 Betvveen 1991 and 1992, 2274 paediatric dentists
inflicted injuries. It is self-evident that infa nts who ca nnot and orthodon tists were surveyed in relation to reporting child
crawl cannot cause a self-inflicted accident. It follows that abuse. Forty-two per cent of pa ediatri c dentists reported hav
severe bruising or fractures in a child below th e ages of 6-9 ing seen a case of suspected child abuse in the previous year,
months are almost always inflicted non- accidentally by a whilst only 15 per cent of the orthodontists did. More signif
second party. icantly, only 11 per cent of the olthodontists actually reported
Alth ough not directly involved, general dental practition their suspected cases, whereas 48 per cent of the paediatric
ers should be aware of the problem of child sexua l abuse, dentists did. Among those paediatric dentists, denta l neglect
although this requires specialist and expert intervention was a frequent reason for suspicion?1 Dental neglect, a com
from those health-care workers trained specifIca lly in this mon form of child maltreatment, should be suspected if ram
area. 30 There is certainly in adequ ate in for mation both in pant caries and oral infection, bleeding and trauma persist
training and in th e dental literature to assist the paediatri c despite the elimination of financial and transportation obsta
dentist in examining these patients. Any physical examina cles. Questioning the child and parents separately may help
ti on of a child should normally be confmed to those parts un cover an obvious discrepancy ben,yeen the clinical findings
of the body considered to be within the expertise of the and the histo ry33 There are no published repolis of child
dentist. Suspicion of abuse lying in the purely medical fi eld abuse or criminal neglect in th e UK relating to advice or con
should be referred to a consultant paediatrician for complete trol of gross caries. Guidelines from the USA 34 indicate that
examination. dentists an d dental personnel have an obligation to repoli
occurrences of dental neglect. Dental neglect is 'wilful failure
of parent or guardian to seek and follow through with treat
DENTAL NEGLECT ment necess ary to ensure a level of oral health essential for
adequate function and freedom from pain and infection'.
Neglect of a child 's dental treatment needs could be con Dentists and dental staff should be trained in recognizing and
strued as a form of abuse. In this insta nce it is best described reporting this type of ab use and neglect. There is evidence
as an act of omission rather than an act of commission. that dental neglect is misunderstood an d under-reported.
Gross dental caries, exacerbated by poor diet, use of car Certain fami lies appear to be more at risk of such neglect than
cinogenic drinks and poo r ora l hygiene procedures, clearly others 35 Dentists, particularly paediatric dentists, are in a
lie within the resp onsibility of the child's parents. However, unique position to eva luate not only the dental needs of chil
the rate of progression and extent of dental ca ries may also dren but also famil y situations and circumstances. Greater
be genetically influenced and may dep end to som e ex tent collaboration between paediatricians and dentists in the eval
on the 'q ua li ty' of the tooth ena mel, on the microbiological uation and management of ab use an d neglect is in the best
flora of the oral cavity an d on the flow and buffering cap ac interests of the child 36 It is clear that further research is
ity of sa li va. These are clearly not the responsibility of the required to develop effective methods, both for eva lu ation
parents. Th e issu e is therefore co mp lex an d sho uld be and for the education of at-risk individu als.
app ro ached with care.
In the USA, state laws have defined chi ld neglect as
negligent tre atm ent, including a failure to provide ade FACIAL AND ORAL PATHOLOGY
quate care, su pport, nutrition or med ica l or surg ica l care so
as to threaten a child's health or welfare. In Ca lifornia, den The facial region is said to be a common site of NAls in chi l
tal neglect is listed as a component of physical neglect. 28 dren, and it has been claimed that almost 50 per cent of chil
In 1998, 15000 denta l surgeons in California were ques dren suffer injUly to this part of the body.5 Of 260 victims of
tioned concerning their knowledge, attitudes and practices ch ild ab use surveyed in Boston, 49 per cent had sustained
regarding child ab use and neglect. 31 Seventy-one per cent orofa cia l trauma 3 ? and a to ta l of 65 per cent sustained head
had never seen a case of child abuse or neglect. Twen ty-eight and facial tra um a. In fact, head or facial trauma was the prin
per cent received no forlllal training in the recogn ition of cipal reason for ad mission to the hospital in 45 per cent of all
child abuse or neglect whilst at dental schoo l. However, 84 cases. Despite this, few of the 537 dentists surveyed at that
per cent had read postgraduate literature on the subject s~nce time knew of their legal responsibility to repOli these injuries.
qualifYing. Sixty-four per ce nt were aware that they were Eleven per cent of them had seen orofacia l trauma cases of a
required by la w to report ab use and neglect, but alm ost the suspicio us nature. Ho wever, only 22 confirmed cases of child
sa me percentage did not kn ow whether failure to report was abuse had been noted by th e den tist ; of these, only four were
a felony. Approximately the same percentage were aware that repolted to the social agencies. Other in vestigato rs have
fa ilure of parents or gu ardians to foll ow through in dental reported an incidence of 60 per cent of cases involving injury
treatment once they had been infoillled about a chi ld 's ram to the head and neck region 38 The incidence in the UK of
pant caries could be considered as child neglect. AJ.most hal f fa ci al injuries related to child abuse has been cited as 43 per
of those questioned were aware that there is a strong con'ela ce nt. 39 Whilst it may be open to discussion as to whether
tion between dental neglect and the presence of physical dentists should become involved in the diagnosis of child
Facial and oral pathology I 423
abuse, 15 it seems clear that he or she would have an imp or seriously. Grip marks on the cheeks or chin can arise dur
tant role to play in the recognition, diagnosis and treatment ing restraint during feeding. Hand marks may also be in the
of injuries to the face and oral cavity. The largest survey form of a slap or a punch. Slaps may leave an imprint of
ava il able appears to be one from 1992 which investigated the the abuser's hand, whereas punches to th e forehea d, nose
denta l aspects of 1248 cases of child abuse in a major Amer and eye region may result in unilateral or bilateral peri
ican hospital over a 4-year period. 12 There were an equa l orbital ecchymosis. Other reasons should be eliminated ,
number of girls and boys, except where sexual abuse was such as infective, ren a l or haematological conditions but, if
involved. Fifty-three per cent of children were in the 0- to 4 the dentist has any doubts concerning unexplained or sus
year age group, and when all cases were considered together picious lesions on the face, the child should be referred to
38 per cent presented with injuries to the head, face and a consultant in paedi at ric dentistry or in paediatrics.2,27
mouth. In a smaller group surveyed the same yea r, 56 per Approximately 10 per cent of NAIs to the child involve
cent of children had abnOlmalities within the head and neck thermal injUly 4 5 These may be burns (73 per cent) or sca lds
region. In a larger studlOover a 5-yea r period, 4340 patients (27 per cent). Thirteen per cent of non-accidental thermal
were studied. Sixty-eight per cent of these were victims of injuries occur on the face , head and neck;4Gthe rol e of the
sex ual abuse and 32 per cent of physical abuse. Of these dental surgeon is clear in recognizing, diagnosing and
physically abused children, 49 per cent had evidence of reporting such injuries. Between 2 and 12 per cent of all
injury to the head and neck region, whereas in the sexually inpatient burns are NAls infl icted with the purpose of pun
a bused children only one per cent had injuries in this area. ishing or controlling the child's behaviour. Burns to t he
There were 11 deaths in this particular series. face have been found to be the most frequent injury in
so me studies,J9 and those from hot, solid objects a re the
easiest to diagnose, and are usually of second degree with
Extraoral Injuries out blister formation. The shape of the burn may resemble
the hot object producing it. For example, cigarette burns
The head and facial region is a common site for NAI not produce a circular, punched-out lesion of uniform size, but
only in t he West, 12 but also in t he Indian continent. 41 It is these are relatively rarely seen on the face. However, they
not known why the face often seems to be the focus ofvio are pathognomonic for child abuse. The use of hot hair
lent attack in children, but it may be because it represents dryers to inflict facial injury has been reported. 47 Friction
the persona of the individu al. burns have been described , and these present with broken
The types of injury have been classified J4 and include blisters over bony prominences, including those of the
contusions and ecchymoses (66 per cent), abrasions and lac face, and can be caused by being dragged across a carpet.
erations (28 per cent), burns and bites (four per cent) and Scalds can be on any part of the body but, in so far as the
fractures (two per cent). Bruises, lace rations, burns, cuts, face and head are concerned, the most common injUly is
scars and black eyes will suggest NAI.4 2 Extensive bruising due to hot liquid being thrown at a child. 2o
of the face with a history of limite d trau ma, especially if both Non-accidentally produced fractures of the facial bones
old a nd n ew bruises are present , is almo st diagnostic. High are uncommon in children and feature in only two per cent of
velocity slaps or lashes may cause an unbnlised, negative cases. 42 Several patterns of skull fracture that are indicative of
image of the object that caused the injury. 2o The margin of NAI have been described. 48 These are multiple fractures with
the injury may often be demarcated by petechial haemor complex configurations - fractures that a re depressed, wide
rhage from capiliaries. 4J This type of injury is sometimes or more than 5 cm long ; involvem ent of more than one cra
known as the tattoo bruise. Bruising to the soft tissues of the nia l bone and non-parietal fractures ; or fractures associated
fac e a nd head may occur in many different ways, but acci with intrac ranial haemorrhage. Multiple injury and inade
dental falls rarely produce bruises in the areas of the cheek; quate explanation in the history are the best indicators of a
they are more likely to involve skin overlying bony promi skull fracture being of non-accidental origin. Most fractures
nences, such as the forehead or zygomatic bone. Bruises in physically abused children occur up to the age of three. 49
change colour as they age,44 but the rate of change from red Accid ental fractures occur more commonly in older children
to blue, to purple, to green, to yellow cannot be used as a of schoo l age. The most common facial fracture is nasal (45
reliabl e indicator of the age of the injury. per cent), mandibular (32 per cent), and zygomatic and orbit
Many injuri es to the head and neck are caused by the complex (20 per cent).50 Specialized treatment is required for
adult human hand . This can leave press ure bruises in the facial fractures and their definitive management has been dis
form of grab marks by fingers and thumb, and these are cussed in the oral surgery literature. J7
velY commonly seen follo w ing pinching or pulling of the
ear. In these circumstances, there wiH usually be a match
ing bruise on the posterior surface of the ear. They are seen Intraoral Injuries
more commonly in babies and infants and may be indica
tors of potenti ally more serious injuries, such as subdura l A review of 1248 cases of child abuse 12 desc ribes five tooth
haemorrhage from shaking, and should always be taken fractures , three tongue and tongue frenulum lacerations,
424 I Dentist's role in child abuse and neglect
two lip frenulum lacerations, 11 injuries to the ora l mucosa , nature may become devitalized, especially if the root apices
eight palatal lesions, loose a nd missing teeth a nd ev idence have completed their development. The crowns of the teeth
of den ta l neglect. It has also been suggested 51 that sexu ally are likely to become discoloured in the months fo llowing
trans mitted diseases in children's mou t hs are a strong indi such trauma. A grey hu e usually develops ow ing to the pres
cation of sexual abuse. There are claims that child re n ence of blood breakdown produ cts in the dentinal tubules,
exposed to oral sexual activities over a co nside ra ble perio d a nd the presence of discoloured anterior teeth for which a
of tim e may develop higher ca ries rates a nd erosion on the sa tisfactory ex pl anation is not forthcoming may suggest
palatal surface of the maxill ary teeth. Den tal neglect in t hat NAI has occurred.
these patients may be pal1 of gen eral neglect. The report of the au topsy findings in 29 fatal cases seen
Although some reports in the literature have included durin g a 2-year period at the London Hospital Medical Col
the term 'neglect' in t he title, only a small number of these lege included o ne of the first assessments of soft-tissue
have discussed dental neglect as distinct from gen eral neg intraoral injury fo llowing child ab use. 5 Fifty per cent of vic
lect of the chi ld. An extraordinary case was reporied in tims had injuries to the head and neck, and 45 per cent of
the UK in 1990, whi ch escaped detection for several years. these had suffered damage to the upper lip region. This con
Three children in a family of six were found to ha ve miss sisted of laceratio ns to the mucosa of the inner aspect of the
ing permanent incisors. It tra nsp ired that their parents were upper lip, close to the labi al frenulum ; these were so metim es
extracting teeth as a punishment for bad beh aviour. 52 so severe that th e reflected mucosa of the lip was completely
A 6-year-old Hi spa nic girl was reporied to have avulsed tom away from that attached to the g ing iva. Tea ring of the
a tooth during physi cal abu se and the sw allowed tooth labial frenulum is not uncommon in very small children,
caused an oesophageal perforation resulting in a retropha who may fall accidentally onto the mout h whilst learning to
ryn gea l abscess a nd a medi ast inal abscess co nta inin g the walk, but this lesion should arouse suspicion as to a possible
toot h. This ap pears to be a unique ma nifestation of dental non- accidental cause when it is found in children less tha n
neg lect. 5) about 1 year old and more than ab out 2 years old. These
The mou th is a frequent site of NAl in children, possi bly injuri es can be produced by agg ressive bottle feeding or by
because of its psych ological signifi cance; it is the orga n of forcing a spo on into the mouth, as is sometimes done by a
phonation , a nd v iolence may be used aga inst it in order to frustrated parent whose child is a slow eater. The labial
silence the child. 54 Contusi ons and bruises, which occurred frenulum may also be damaged by blows to the mouth or by
in 43 per cent of a series of 14 cases reviewed , are the most tearin g the upper lip away fro m the gingival attachment
common injuries to the mouth;42 lace ra tions and trauma to (Fig. 22. 1) (see a lso Fig. 8.6, p. 152). Following this report, the
the t eeth eac h account for 28 per cent of injuries. tom labial frenulum has been regarded as pa thognomonic of
Examination of the mouth should commence with the child abuse, but it must be noted t hat no later reports of intra
lips; the lips usually heal well follo wing min or injuries a nd oral da mage in cases of child abuse have indicated such a
scarring should alert the dentist to previous severe trauma. high incidence. Many paediatric dentists of considera ble
Bruised and swo llen lips or ab rasions at th e corners of the experience worki ng in the UK have never seen a case of labial
mouth are suggestive of bl ows from a fis t; NAls are espe fren al damage that t hey could unequivocally relate to NAI.
cially likely if the bruises or injuri es are at varying stages When it does occur, the condit ion is often associated with
of healing. subluxation or fra cture of the upper centra l inciso rs, damage
The teeth themselves are vulnerable to blows to the to the teeth being caused by the same blow.
mouth, from either a fist, a hand slap or a weapon. It must,
of course, be remembered t hat children frequ ently avulse or
fracture teeth in the course of normal play activities, but an
unsa tisfacto ry history from the child or parent, or repeated
atte ndance for treatment of injuries, should arous e the sus
picion of the dent ist. The roo ts of the anterior teeth may be
incompletely formed in children and, as a result, blows to th e
face often result in complete avulsion and loss of the tooth
rather than in fra ctu re of the roots. When teeth are missing,
full radiographs of the jaw should be taken to rule out some
abnormality of development or conditions of partial ano
dontia; retained frac tured roots should be sought as these
will require surgical removal. Mult iple healed micro frac tures
of tooth roots may sometimes be v isible in intraora l radio
graphs - these are al most always indicative of trauma by
repeated abuse. Slight movements of the teeth may rupture Figure 22.1 Dam ag e to the labia l frenum may be caused by
delicate blood vessels pass ing through the apical fo ramina forcible feeding or, as in this ca se, by a blow to the mouth.
and, as a result, anterior teeth that survive trauma of this Teeth may be displa ced.
-~~~ -
Bite marks I 425
Penetrating injuries to the palate, the vestibule and the two of which involved teeth marks in battered babies. Trag
floor of the mouth resulting from forced feeding or forcible ically, both children died. 58 Shortly afterwards, a case of a
introduction of spoons and other implements into the bite on a 17-month-old child was reported ;59 bite marks
mouth have been described. Burns of the oral mucosa fol and a torn labial frenum, associated with battered child
lowing enforced ingestion of hot or caustic liquids are seen syndrome, were described . 5 The incidence of bites in victims
quite commonly,27 as are intraoral cigarette burns and tear was given as 12 per 100000 in the population of New York
ing of the oral mucosa by a parent's finger placed in the in 1977. 60
mouth. Dentists who are unfamiliar with the range of oral However, there are few studies on the frequency of bit
pathologies of a non-accidental nature that may occur in ing associated with child abuse. In one study 61 nurses work
the mouth should refer patients to a specialist if they are in ing in facilities that admitted child abuse patients were
doubt as to the cause of a particular lesion. trained by forensic dentists to recognize bite marks on their
patients. Bites were divided into incised bites without bruis
ing, those incised with bruising or discolouration, marks
THE DENTIST ACCUSED OF CHILD ABUSE with sucking bruises and marks in which the teeth had been
d ragged over the tissue. Of 1100 children examined, 17 had
It is known that child abuse or NAIs to children are fre evidence of bite mark abuse - an incidence of 1545 per
quently caused by an adult in close relationship to the child. 100000 sheltered chi ldren. Most of the bites had been
It has been pointed out 55 that this may include a ch ild 's pro inflicted on children between the ages of II and 15 years,
fessional adviser, such as a dentist. In the state of Tennessee but two were in ch ildren less than 3 years of age and four
in the USA malpractice insurance rates have increased by in chi ldren between 16 and 18. Most of the male victims
1500 per cent between 1981 and 1987, and cases of child were in the 4- to 10-year age group, whilst most females
ab use/criminal assault have been reported against several were in the 11- to 15-year-old age group. The most common
dentists performing procedures on patients. Dental practi site of bites was the head and neck region (43 per cent). The
tioners have been accused of child abuse by using excessive limbs and trunk were also common sites of attack. In an
restraint, for example by placing a hand over the mouth in unpublished studyG2 of 99 cases of bitten children there
order to deal with a difficult or noisy patient. This exercise were 169 bites. Thirty-three per cent were in children aged
was singled out by the Virginia Board of Dentistry as lead between 6 and J 8 months, 23 per cent in children aged
ing to reports of child abuse against dentists. between 18 months and 3 years but there were 17.2 per cent
Another risk is the non-consensual touching of a of bites in children under 6 months of age. There were no
patient, even if this is aimed at improving a child's health. statistical differences in this study between the girls and the
Questionnaires in the USA are currently being designed to boys. Few studies have been carried out on the distribution
discover how many dentists have been accused of battery of bite marks speCifically in chi ld abuse cases, but the
or non-consensua l touching. The information is not easy to anatomical distribution has been investigated in a series of
acquire, and until 1987 only one repol1 was available cov 67 victims of varying ages. 63 Among this group, 13 were
ering three cases. 55 These three incidents occurred during aged below 15 years. Female victims were most commonly
the course of normal t reatment and all three resulted in an bitten on the breasts, arms and legs, while the arms and
arrest. As child abuse is a criminal offence, if it occurs in shoulders were the most common sites in males. These
the dental surgery it is not covered by malpractice insur results differ somewhat from those reported in the UK,
ance. 56 There is a danger of dentists becoming overdefen where bite marks on the breasts of females were found to be
sive in these matters, and this may eventually be to the more common than in the American sample. Bite marks on
detriment of child patients. the arms were seen less frequently.64
Other studies on the anatomical distribution of bites
include th ose on 122 injuries observed in a New York hos
BITE MARKS pital 65 and on 114 cases reported from Kansas.66 In both of
these studies, bites were most often seen on the hands and
Bite marks have come to mean any injury or mark produced fingers. It appears that predilection for the different parts
in flesh, foodstuffs or other material by the teeth and the of the body has changed in more recent times. In NAIs in
surrounding soft tissues. 57 In non-living materials and in children, the entire surface of the child's body may be
foodstuffs, the mark left is usually produced entirely by the attacked, resulting in bite marks on the cheeks, shoulders,
teeth, but in bites left on human skin, the injury may be chest, abdomen, arms, legs and buttocks. 67 Multiple bites at
contributed to by pressure and suction and/or pressure from the same site have been described. In the Cardiff series,52
the tongue as well as from the teeth themselves. Recogni 37 per cent were on the shoulder or arm, 31 per cent on the
tion of bite marks as part of the multiple injuries sometimes leg or buttock, 18 per cent on the head or neck region (of
seen in NAIs to children is a relatively new development. which 15 per cent were on the face) and 12 per cent were
Attention appears to have been first drawn to the incidence on the ab domen or back. Less than three per cent were on
of bite marks in NAls in a report of 13 forensic dental cases, the chest region. There were no significant differences in
426 I Dentist's role in child abuse and neglect
- - --- - - - -
- - -
Bite marks I 427
Self-inflicted Bites
animals the child has been younger than 2 years old and
usually less than 1 year of age.
Dogs are probably the most common animal to bite chi l The Investigation of the Bite Marks
dren,88 and most of the injuries involve the extremities. As
a result of their small size, 33 per cent of bites in children Where bite marks are present in a case of NAl, they are usu
under 5 years of age are to the head and neck, 88 and in ally to be found on the child victim. However, it should be
some studies up to 67 per cent have been to the head and rememb ered t hat occasionally they may be on an assai lant,
neck. 82 There have been instances of damage to the central if the victim has attempted to fight back or is attacking in
nervous system caused by an English bull terrier, a Dober self-defence. The same principles of investigation apply in
man pinscher and a Bengal t iger89 In nearly all the both cases and, apart from the difficulties of examining a
reported cases of injury to the head from dogs or similar conscious sma ll child, the methods llsed to investigate bite
- - - -- - - ----
-- --
Bite marks I 429
marks on an adult can be applied routinely in most ch ild produced by the canine to canine of the upper arch opposed
cases. Injuries caused by biting can be thought of as 'tool to canine to canine of the lower arch. Incisor teeth tend to
marks', the tools in this case being the biting edges of the leave rectangular bruises or incisions whe reas the canines
teeth. In many ways they are a unique injUly in the sense leave puncture or triangular marks. A decision as to whether
that not only do they demonstrate injury which is difficult to an injury is, in fact, a human bite mark depends upon its
explain away as an accident, but they are also one of the few shape, colour and size and on the impressions made by indi
injuries that can be related directly to the assailant. Bite vidual teeth. Bruising may be present not only rel ated to the
marks in human flesh can be extremely difficult to interpret biting edges of the teeth but also in the central area of the bite
because of the dynamic nature of the injury. Distortion may beca use of tongue thrust or suction by the assailant. This usu
be caused by the bite itself, by subsequent changes and by ally presents as petechial haemorrhage in the central area of
many other factors. It is important, therefore, that an expe the bite. If the marks of individual teeth in their correct align
rienced forensic dentist should be included in the team ment can be satisfactorily identified then it is possible to say
investigating cases of childhood abuse. This is still not the at this stage, that the victim has suffered a human bite.
case in many child protection procedures produced by health Before the bite is touched, measured or photographed it is
authorities in the UK. As a general rule, bite mark injuries important to obtain evidence of saliva traces left on the skin
should always be carefully recorded and described before surface by the assailant. These swabs may be used for subse
the teeth of a putative suspect are examined. This is because quent blood grouping, usually limited to the ABO system
even experienced forensic odontologists may introduce an only. A, Band/or H substances are secreted by some 75 per
unconscious bias in to their interpretation if they already cent of the population. A negative result may therefore be
have a particular dentition in mind. The investigator should due to the absence of saliva to a non-secretor status of the
be attem pting to fit 'the injuries seen in the bite' to one of a biter or to poor sampling. Nowadays swabs are taken for
number of suspects and should not be attempting to fit 'o ne DNA analysis. Control swabs must be taken from swabs in
particular dentition' to a specific injury. other parts of the body where no bite marks are suspected 100
and it is often convenient to use the contralateral area of the
body. A saline control swab should also be supplied. These
The Bite Marks specimens are needed to exclude contamination by secre
tio ns from either the victim or the investigator, who should,
There are no detailed protocols for examinations of bite of course, wear gloves throughout the procedure. The use of
mark injuries that are acceptable throughout the world, stelile cotton swabs moistened with steriJe water followed by
and individual dentists have developed their own methods. dly swabs, the so-called double swab technique, is now rec
Guidelines have been laid down by the American Board of ommended. 101 The importance of a standard and satisfactory
Forensic Odontology.98 tech niqu e cannot be over emphasized.
Once recognized as such, a bite mark on a child should
always be investigated by a forensic odontologist if abuse
is suspected. Children receive bite marks from adults, other The Detail of the Bite
children or animals, and they are a common problem in
children presenting for medical attention. 99 When child Once an injury has been recognized as being caused by a
abuse is suspected, the recognition of bite marks may pose human bite and saliva samples have been taken, the details
significant problems for a medical clinician. It may be of the mark should be recorded. A human arch mark may
difficult to determine whether bruising or lacerations are be identified when four or five individual marks of adja
caused by a bite because the marks are often incomplete, cent teeth are present. If the tips of the canines in both
distorted and change over time. They may, on occasions, be upper and lower arches ca n be recognized, then the inter
confused with skin eruptions .6 ) Collaboration between pae canine width should be measu red. In a human bite, the
diatricians and forensic dentists is a satisfactory solution to intercanine width will lie between 2. 5 and 4.5 cm.70 If the
the assessment of bite marks where abuse is suspected. 71 intercanin e width is less than 3 cm, the bite will probably
After recording the usual details of the victim, the location have been inflicted by a child. In this context, a child refers
of the bite mark on the body, be it living or dead, is carefully to any biter below the age of 12 whose anterior teeth are
desClibed . At this stage it is important to note whether the either deciduous teeth or mixed dentition. The two arcades
bite mark is on a flat or a curved surface and whether it over of upper and lower teeth marks should be described and
lies soft tissue, fat or bone. Bite mark identification entails measured in terms of their curvature and overall size; in
several steps: recognition of the wound, documentation, and addition, marks left by individual teeth should be individ
interpretation . A human bite mark is usually identified ually measured and described in an attempt to reconstruct
because of the double arcade shape of the marks. If both the type of denti tion that could have made the mark. Miss
upper and lower arches have left marks, the characteristics ing teeth or areas where no bruises have occurred must be
of the resulting bite include an elliptical or ovoid pattern noted. The alignment of each bruise and the possibility of
containing tooth and arch marks,57 which are almost always them being made by rotated teeth should also be examined.
430 I Dentist's role in chi ld abuse and neglect
In general, the mark caused by the upper maxillary arch majority of cases, good-qu ality co lour and bl ack and white
is often more diffuse and wider than that produced by the photographs will be used. Some fo rensic dentists argue that
mandibular teeth. This results from the greate r surface area the photography should be carried out by a professional
of the incisal edges of the upper teeth, but also because the police photographer under the direction of the dentist, so
maxilla is stable in relation to the skull. By contrast, the that questions in court regarding scale and reproduction,
mandible is mobil e and acts as a cutting instnlment once colour balance and other photographic questions may be
the skin has been stabi lized against the upper teeth. Before answered with authority. However, exp erienced forensic
the bite mark is recorded the forensic dentist should dentists prefer to take their own photographs in a standard
attempt to establish the circumstances surrou ndin g the ized manner as better results are often obtained. A 35-mm
bite. A slowly produced bite usua lly results in stretching of camera with a focal len gth lens of abo ut 100 mm and a ring
the skin over the cutting edges of the teeth and pressure or flash and/or side flash is the most commonly used equip
suction from the tongue. This will be reflected in the extent ment and the techniques involved have been described in
of bruising, both in the too th marks and in the surroundin g detail. lOG Whatever equipmen t is used, photograp hs should
and central areas of the bite. It is said that a bite mark include low-power distance shots of the bite mark and sur
show ing these characteristics was probably produced under rounding tissues, which allow for orientation. A standard
amorous circumstances. However, even so-called ' love colour chart is often in clud ed along with the subject when
bites' require considerable force to produce and can be colour film is used. In the case of black and white, ultra
painful to the recipient. A bite mark that consists only of violet filters may be used, which may bring out details that
tooth marks is of an aggressive nature and has probably are not obvious when viewed by conventional lighting. 107
been delivered rapidly. This is especially so if teeth have Ultraviolet light requires a grea t deal of careful and expe
penetrated the sk in. The observer sho uld be wary of rien ced interpretation as injuries made many years before
attributing a particular force or intent, such as aggression, the bite may be imaged using this technique a nd may con
in any particular bite mark case. flict with the interpretation of a recent injUly.
It may sometimes be possibl e to offer a general opinion Whatever method of photography is used, photographs of
as to the timing of the bite. Too th marks that do not break the injury must be taken with a standard, rigid, L-shaped
the skin usually last from several minutes to 24 hours. 102 In centimetre scale, arranged as close as possible to the injury,
cases where the skin is broken, the borders or edges wi ll last but not overlapping it. Because injuries are often on curved
several days depending on the thickness of the tissue. Thin surfaces, it may be necessary to reposition the sca le a num
ner areas retain the marks 10ngeL IOJ It is clearly important ber of times in relation to different parts of the bite mark. In
that the forensic dentist be called in the early stages of the principle, the scale should be arranged in the same plan e as
investigation of a bite mark case. Many investigators have the portion of the injury being photographed and the plane
recommended that bite marks be photographed at 24-hour of film of the camera should also be parallel to both scale
intervals because their detail may improve in the first 2-3 and the portion of the injury being photographed. Inevitably,
days as swelling decreases and brui sing increases. 57 In add i on a sharply curved surface, a number of photographs will
tion, the outline and shape of a bite may change because of be required around the curvature of the injury (Fig. 22.6).
infection, oedema and discoloratio n of the skin. Accurate These procedures are essential to avoid introducing photo
determination of the time of a laceration or brui se is diffi graphic distortion into a situation that is already distorted by
cult and complicated by individual valiation.
The history of a bruise can be correlated to sequential
colour changes, but is frau ght with difficulty, and only a
gen eral comment should be made on these matters . The
current position has been reviewed recently. 104.105 Quite
different appearam:es will result if the bite was inflicted
immediately prior to or immediately after death . If the bite
mark was inflicted prior to death, providing that sufficient
time has elapsed to allow extravasation of blood into the
tissues, a clear bite mark may be seen that changes velY lit
tle until the onset of putrefaction. Bites inflicted after death
rarely result in bruising. Very considerable forces indeed are
required to produce an appearance of bruising post mortem.
Recording the Bite Mark Figure 22.6 Bite marks are often on cu rved su rfaces, such as
the breast, and each portion of the injury should be photographed
It is important to record the injury for posterity, especially individually. The mark on the right is from the upper arch with
when legal proceedin gs may subsequen tly occur. In the only three incisors present.
---
Bite marks I 431
possible. The technician making the castings must record distOIiion and size discrepancies, eliminate examiner subjec
the details of the procedure. The base of the study models tivity, provide better control of image visualization and pro
must be marked with the individual's name and also signed v ide standardization of procedures. However, it should be
and dated by the technician. Some dentists invite a suspect remembered that these imaging methods are only as good as
to produce a samp le bite into a wax recording medium at the original evidence and should not be used to enhance
the time of the examination. poor-qu ali ty evidence. 11 2 The final opinion is a very serious
issue for possible suspects as well as for the victim and fam
ily; in few areas of medical evidence is such a respons ibility
Comparison of Evidence from the Bite placed upon the objectivity and honesty of the expert. No
and the Suspect jury should be led to believe that bite mark analysis is sim
ple and accurate and the expert must clearly indicate the
The final comparison between the details of the bite mark pros and cons of each case.
injury and the dentition of a possible suspect is not a simple
matter. Some a uthorities claim that transparent overlays of
the biting edges of a suspect's teeth can be produced and REFERENCES
directly compared with the bite mark. An accurate fit then
warrants an opinion of guilt by a putative biter. However, Knigh t B. The hi story of chil d abu se. Forensic Sci I11t 1986;
30: 13 5- 41.
transparencies produced by whatever meth od rarely, if ever,
2 Kempe CH, Silverman FN, Steele BF, Droegemuell er W,
exactly fit the size and shape of a human bite mark. This is Silver HK. The battered child syndrome. lAMA 1962;
because of the distortion factors, which have been discussed 181:17- 24.
earlier in this chapter. There is a danger in ass uming that bite 3 Welbu ry RR. Ch ild physical abuse. In Andreason JO,
marks are equivalent to fingerprints and that a reasona ble Andreason FM reds) Textbook alld C%ur Atlas of Traumatic
Injuries to the Teeth, 3rd edn. Copenha gen: Munksgaard,
number of matching points wi ll prove the id entity of the
1994.
biter. A more satisfactOlY ap proach is to describe the bite 4 Mea do w RC. The ABC of Child Abuse. London: BMJ
mark in detai l before looking at the details of a suspect's Publis hing Group, 1997.
dentition a nd then to describe the suspect's dentition and 5 Cameron JM , J oh nso n HR , Camps FE. The battered ch ild
to note the points of correspondence or deviation. The bite sy ndrome. Med Sci Law 1966 ; 6: 2-2 1.
mark inju ry and the dentition of the suspects are then exam
Pediatr Dent 1906; 8 :96 -100.
ined together, and an iterative process should enable the 7 Needl eman HL Orofacial trauma in child abuse: types,
observer to determine whether or not the details of the injury prevalence, manageme nt, and the dental profession 's
can be explained satisfactorily by relating th em to the den in volvement. Pediatr Dent 1986; 8:71-80.
tition of the suspect. If they can be so related, the forensic 8 Wei bury RR, Murphy JM. The dental practitioner's role in
protecting chi ldren from abuse. 3. Reporting a nd subsequent
dentist shou ld base the strength of his or her final opinion on
management of abuse. Br Dent 1 1998; 184:115 - 19.
the si milarity of pattern between the injury and the dentition 9 Kirschner RH, Stein RJ. The mistaken diagnosis of chi ld
a nd should tl1en finally illustrate a nd confirn1 his or her abuse. A form of medical abuse 7 Am 1 Dis Child 1985;
scale. The non-scientific nature of this methodology has 10 John so n CF. Inflicted inju ry ve rsus accid e nta l injury. Pediatr
been emphasized, but it can work sat isfactorily in most Clill N Am 1990 ; 37:791-814.
11 Whittaker OK. The principles of forens ic denti stry. 2. Non
cases. 98 Unusual features of bo th the mark and a suspect's accidental injury, bite marks and arc haeo logy. Dent Upda te
dentition that would tend either to elimin ate or incriminate 1990; 17:386-90.
a suspect in the case should be noted. Thus, a single feature, 12 Fonseca MA, Feigal RJ. Ten Bensel RW. Dental aspects of
such as a missing, rotated or fractured tooth leaving a com 1248 cases of child maltreatment on fi le at a major country
hospital. Paediatr Dent 199 2; 14: 152-7.
parable mark in the .bite, may be a much more useful pointer
13 Rasm ussen P. Tannlegens forhoJd tilsykdo m hos barn,
than any number of matching features in an arch of normal barnhemish andling og omsorgssvikt. In Odontoiogi.
size and arrange ment (Fig. 22.6). Comparisons of dentitions Copenhagen : Munksgaard Publishers, 1992, pp. 81-92.
and bite marks have used superimposition of colour slides, 14 Speig ht N. Bruises and burns: accidental or non -accid enta l?
denta l articulators and the recording of bites in suitable In Accidents and Emergencies in Childhood. London: Royal
CoJl ege of Phys ic ians of London , 1992, pp. 97 -103 .
media. Sophisticated scientific analysis such as micro
15 Welbu ry RR, Murphy JM. The denta l prac titi oner's ro le in
co ntour mapping, scanning electron microscopy a nd reflex protecting chil dre n from abuse. 2. The orofacia l sig ns of
microscopy have been used. 11 6 In practice, the simpler and abuse. Br Dent 1 1998; 184:61 - 5.
more obvious the method, the more likely the jury memb ers 16 Simpson K. A Doctor's Guide to Court. London : Butterworth,
are to understand and appreciate the significance. The 1967.
17 Creighton SJ, Gallagh er B. Child Abuse Deaths: Informatio n
current vogue is to use computer analysis of the bite mark
Briefing, No.5 (revised). London: The National Society for the
photograph and to produce transparent overlays digitally Prevention of Cru e lty to Children, 1988.
from sca nned images of the dentition.III, ll l These methods 18 Schm itt BD. Physical abuse. Specifics of clinical diagnosis.
are sa id to be able to correct for common photographic Paediatr Dent 1986; 8:83.
References I 433
19 Creighton SJ. The incidence of child abuse and neglect 45 Lenosky EF, Hunter KA. Specific patterns of inflicted burn
(Chapter 3). In Browne K, Davies C, Stratton P (eds) Early injuries. J Tra uma 1977; 17:842-6.
Prediction and Prevention of Child Abuse. Chichester: 46 Kumar P. Child abuse by thermal injury - a retrospective
John Wiley Et Sons, 1988. survey. Bums 1984; 10:344-8.
20 Golder M. Non -accidental injury in children. Dellt Update 47 Prescott PRo Hair dlyers and non-accidental injury. Pediatrics
1995; 22:75-80. 1990; 86:692-7.
21 Fearn 1. Dental aspects of child abuse (Letter). Aust Dellt J 1987 ; 48 Hobbs CJ. Skull fracture and the diagnosi s of abuse. Arch
32:227. Dis Child 1984; 59:246-52.
22 Cappelleri JC, Eckenrode J, Powers .fL. The epidemiology of 49 Worlock p. Stower M, Barbor P. Patterns of fractures in
child abuse: findings from the second National Incidence and accidental and non -accidental injUly in cbildren: a
Prevalence Study of Child Abuse and Neglect. Am J Publ Hlth comparative study. Br Med J 1986 ; 293: 100-2.
199 3 ; 83: 1622. 50 Kaban LB, Mulliken JB, Murray JE. Facial fractures in
23 Hobson P. Child abuse: we can help (Letter). Br Dellt J 1987; children. Plost Recollstr Surg 1977; 59: 15-20.
162:53. 51 Kvaal 51. Oral injuries in cases of child abuse. Rev Beige [,vIed
24 Child protection in ethics ill dentistry. BDA Advisory Semices, Dentaire 1993; 48:49-53.
1995, pp. 45-6. 52 Carrotte pv. An unusual case of child abuse. Br Dent J 1990;
25 Department of Health. Working Together under the Children 168 :444-5.
Act (1989), 48-49, London, HMSO, 1991. 53 Ablin DS. Reinhart MA. Esoph agea l perforation by a tooth in
26 Depart ment of Health. Children alld Youllg People on Child child abuse. Pedia tr Radiol 1992; 22:339-41.
Protection Registers, Year Ending Jist March 1995, England. 54 Sperber N. The dual responsibility of dentistry in child abuse.
London: Gov ernment Statistical Service, 1995, p. 19. Mt J Orthod 1981; 19:21-8.
27 Macintyre DR, Jones GM, Pinkney RCN. The rol e of the dental 55 Schuman NJ. Child abuse and the dental practitioner:
practitioner in the management of non-accidental injury to discu ssio n and case rep0l1s. Quintessence 111t 1987;
children. Br Dellt J 1986; 161:108-10. 18:619-22.
28 Kittle PE, Richardson DS. Parker JW. Examining for child 56 Troutman KC. Non-consensu al touching: your input needed.
abuse and child neglect. Pediatr Dellt 1986 ; 8:80-2. Am Acad Pediatr Dent 1987; 22:3.
29 Casamassimo PS. Child sexual abuse and the paediatric 57 MacDonald DG. Bite mark recognition an d interpretation.
dentist. Paediatr Dent 1986; 8: 102-6. J Forens Sci Soc 1974; 14:299-233.
30 Anasavice K1. Treatment regimes in preventive and 58 Hod son JJ. Forensic odontology in cases of battered baby
resto rative dentistry. J Am Dent Assoc 1995 ; 126 :727-43. syndrome. I Den. Res 1971; 50:6 56.
31 Ramos-Gomez F. Rothman D, Blain S. Knowledge and 59 Sims BG, Grant JH, Cameron JM. Bite marks in the ' battered
attitudes among California dental ca re providers regarding baby' sy ndrom e. Med Sci Law 1973; 13:207-10.
chi ld abuse and negl ect. J Am Dent Assoc 1998; 129:340-8. 60 Marr JS, Beck AM. Lugo JA. An epidemiological study of the
32 Beck JD , Weintraub JA, Disney JA. University of North human bite. Public Health Rep 1979; 94:514-21.
Carolina. Caries Risk Assessment Study. Community Dellt Oral 61 Rawson RD, Koot A, Martin C et al. In cidence of bite marks in
Epidemiol 1992; 20:313-21. a selected juvenil e population: a preliminary report. J Foren s
33 Jessee SA. Physical manifestations of child abuse to the head. Sci 1984; 29:254-9.
face and mouth: a hospital survey. ASDC J Dent Child 1995; 62 Liddiard P. An analysis of bite mark cases in foren sic dental
62:245-9. practice, with reference to dental anomalies affecting case
34 Rupp RP. Conditions to be consid ered in the differential outco mes. MS c, University of Wales, 2002.
diagnosis of child abuse and neglect. Gen Dellt 1998; 63 Val e GL, Nogu ch i IT. Anatomical ciistribution of human bites
46:96-100. in a series of 67 cases. J Forens Sci 1983; 28:61-9.
35 Jessee SA. Risk factors as determinants of dental neglect in 64 Harvey W. Denf'al Identification and Forensic Odontology.
children. ASDC J Den. Child 1998; 65:17-20. London: Henry Kinpton, 1976. p. 91.
36 Kellogg N and the Committee on Child Abuse and Neglect. 65 Lowry TMcC. The surgical treatm ent of human bites. Al1nals
Oral and dental aspects of child abuse and neglect. Pediatrics Surg 1936 ; 104: 1103-6.
200 5; 116: 1565-8. 66 Speirs RF. Prevention of human bite infection. Surg Gynec
37 Sanders B, Brady FA, Johnson R. Injuries. In Saunders B (ed.) Obstet 1941 ; 72:619.
Pediatric Oml and Ma.xilloJacia1 Surgery. St Louis: Mosby, 67 Sims AP. Non-accidental injury in the child presenting as a
1979, pp. 330-400. suspected fracture of the zygo matic arch. Br Dent J 1985;
38 Fotana VJ, Dqnovan MD, Wong RJ. The maltreatment 158 :292-3.
syndrome in children. N Eng I J iVIed 1963; 269:1389-94. 68 Liston PN, Tong DC, Firth NA, Ki ese r JA. Bite injuries:
39 Skinner AE, Castle RL. A Retrospecti1Je Study. London: pathophysiology, forensic analysi s and management.
NSPCC, 1969. N Zealand Dent J 2001; 97:58-63.
40 Willging JP, Bower CM, Cotton RT. Physi cal abuse of 69 Pretty IA. Sweet D. Anatomical lo catio n of bite marks and
children. A retrospective review and an otolaryngology associated findings in 101 cases from the United States.
perspective. Arch OtolarYl1gol - Head Neck Surg 1992; J Forens Sci 2002; 45:812-14.
118 :584-90. 70 Simper RD, Whittaker OK. An analysis of bite marks
41 Mehta MN. Physical abuse of abandoned ch ildren in India. presenting in two centres in the United Kingdo m. J Dent Res
Child Abuse Neg11982; 6:171-5. 1994; 7J :791.
42 Becker DB. Needleman HL , Kotelchuck M. Child abuse and 7J Whittaker OK. Aitken M, Burfitt E, Sibert JR. Assessing bite
dentistry: orofacial trauma and its recognition by dentists. marks in children: wo rking with a forensic dentist.
J Am Dent Assoc 1978; 97 :24 -8. Ambulatory Child Hlth 1997; 3:225.
43 Feldman KW. Patterned abusiv e bruises of the buttocks and 72 Levine U. Bite mark ev idence. Dent Ciin North Am 1977;
pinn a. Pediatrics 1992; 90:633-6. 21: 145- 58.
44 Wilson EF. Estimation of the age of cutaneous contusion in 7J Whittaker DK. BrickJey MR. Evans L. A comparison of the
child abuse. Pediatrics 1977; 60:750-2. ability of experts and non-experts to different iate between
434 I Dentist's role in child abuse and neglect
adult and child human bite marks usin g receiver operating 95 Tsoukas CM, Hadjis T, Shuster J et al. Lack of transm iss io n
characteristic (ROC) ana lysis. Forflls Sc i Tilt 1998; 92 : 11-20. of HIV throug h h uman bi tes and sc ratches. J ACl{uir 11I1JI1!llie
74 Ky dd WL. Quantitati ve analysis of fo rces o n the tongue. Defic S.\'lId 1988; 1 :505-7 .
J Dellt Res 1956 ; 35:17 1- 4. 96 Anony mous. To tra nsmit HIV, biter mu st have blood in the
75 Harvey W, Millington PF, Evans , JH et a l. Bite ma rk s - the mouth. AIDS Policy Law 1996 ; 11:5.
clinical picture; physica l fea tures of skin a nd tong ue. 97 Pre tty lAo And e rso n GS, Sweet OJ. Hum a n bites and the risk
Sta nd a rd and scan nin g d ec o'on m icrosco py. mt MicroJorlll of human immunodeficiency virus transmiss ion . A m J
J Leg Med 1973; 8:3. Forells Med Porliol 1999; 20:232-9.
76 Baker MD, Moore SE. Human bites in chi ldren. A six -yea r 98 Arheart KL, Pretty IA. Results of th e 4th ABFO Bitema rk
expe rience. 1987; 141: 1285-90. Workshop - 1999. Forells Sci 1111 200 I; 124: 104- 11.
77 Leung AK, Robson WL. Human bites in chi ldren. Pedilltr 99 Sc hweich P. Fleisher G. Human bite marks in children.
fJ-nerg Cllre 1992; 8:255-7. Pediatr Emerg Cll re 1985; 1:51-3.
78 Levine LJ. The solution of a battered ch ild hom icide by dental 100 Pereira ilil. Possibilities and limitations of sa liva tests in
ev id ence : report of case. J Am Dellt Assoc 1973; 87:1234-6. forensic odonto logy. Br Dellt J 1971; 130:161.
79 Walllick AJ, Biedrzycki L, Russanow C. Not all bite marks are 101 Sweet 0, Lorente M, Lorente JA et al. An improved meth od
associated with abuse, sexual activities or h om icid es : a case to recover sa li va from hu man skin: th e double swab
study of a self-inflicted bite mark. J Fo rflls Sci 1987; 32:788-92. tec hnique. ) Forells Sci 1997: 42:320- 2.
80 And e rson W R, Hudson RP. Self- infli c ted bite marks in 102 Dinkel HI. th e use o f bite mark ev iden ce as a n in vesti gative
battered child syndrome. Forell s Sci 1976; 7 :7 1-4. a id. J Forells Sci 1974; 19:535 -47.
81 Ca lkins CM, Bensard DD . Patrick DA. Ka rre r FIVI. Life 103 Sper be r ND . Bite ma rk s, o ral and racial injuries - h arb ingers
threa tening dog attacks: a devas tating comb in at ion of of seve re c hild abuse 7 Pacdiatriciall 1989 ; 16: 207 - 11.
penetrati ng a nd blunt injuri es. J Pedilltl· Su rg 200 1: 104 Dail ey JC. Bowers CM. Ageing of bite marks: a lite rature
36 :1115-17. revi ew. J Forell5 Sci 1997; 42:792-5.
82 Gandhi RR. Liebman IVIA. Stafford BL, Stafford pw. Dog bite 105 Van ez is P. Inte rpreting bruises at necropsy. J Oill Pllthol
injuri es in c hildren: a preliminary survey. AIlI Surg 1999; 2000; 54:348-55.
65:863-4. 106 Summers R. Lewi n D. Forensic dental photography. In Cla rk
83 App lega te JA, Walhout MF. Childh oo d risks from the ferret. DH (ed.) Practica l forellsic Odoiliology. Oxrord: Wright,
J Emerg Med 1998; 16:425-7. J 992, p. 188.
84 Triratana 1. Bite marks. J Dent Assoc Thllilllilil 1970; 107 Krauss TC, Warl e n Sc. The forensic science use of re fl ec ti ve
20:259 -7 2. ultravi o let photography. J forells Sci 1985; 30:262-8.
85 Class RT, Jordan FB, Andrews EE. Multipl e animal bite 108 W h ittaker OK. MacDona ld DG. A Colour Atlas oj Forell sic
wounds: a case report. J forells Sci 19 75; 20: 305- 14. DentiSTry. London: Wolfe Publications, 1989, p. 108.
86 Rhe ul J , Bratz ke H, Fedderse n-Perterso n DU et al. Dea th 109 Whit taker DK. Fo rensic dentistry in the ident ificatio n of
ca used by 'attack' dog bites, Arch Kril1lillol 1998; victims a nd assaila nts. J Oill Forells IVIed 1995; 2: 145 -5/.
202: 140- 51. 110 Sheasby DR, Mac Donald DG. A forensic classification of
87 Sacks JJ , Sinclair L, Gilchrist Jet al. Breed s of dogs in vo lved distortion in h uman bite marks. fore lls Sci II/rem 200 1;
in fa tal human attacks in the United States betwee n 1979 a nd 122:75-8.
1998, J Am Vet Med Assoc 2000; 217:836 - 40. 111 Bowers Cilil. J o ha nse n RJ. Dig ital analysis of bite marks and
88 Belardi FG, Pascoe JM, Beegle ED, Pasteurell a l11ultocicla human id e ntifi cat ion. Dfllr Clill NAill 2001; 45:327-4 2.
menin g itis in an infant following occipital dog bite, J fam 112 Kara za lu s KP. Palm bach IT, Lee He. Digital enhancement of
P/'Gct 1982; 778:780-2. subquality bite mark photographs. J forells Sci 2001;
89 Ste inb ok P, Flodma rk 0, Scheifele Ow. Animal bites caus ing 46:954 - 58.
ce ntral nervous sys tem injury in children. A repolt of three 113 Millington PF. Hi stological stucii es of skin carrying bite
cases. Per/ iat!· Neurosci 1985; 12:96-100. marks. J Forms Sc i Soc 1973: 14:239-40,
90 Callaham ML. Treatment of cOl11mon dog bites: infect ion risk 114 Raekallio J. Histological estimation of the age of injuries, In
factors. JACEP 1978; 7 :83-7. Perper JA, Wec ht CH (eds) Microscopic Dillgnosis ill Forensic
91 Marcy SM, Infectio ns d ue to dog and ca t bites. Pediatr IIIJecr Parliolog.\'. Springfield, IL: Cha rl es C Thomas, 1980, pp. 3-16.
Dis 1982; 1:35 1-6. 115 Rothwell BR , Thie n AV Analys is of d isto rtion in preserveci
92 Brook I. Mic robio logy of human and anima l bite wounds in bite mark sk in. J Forel1s Sci 200 1; 46:573.
childre n. Pediotr IIIJecr Dis J 1987; 6 :29-32. 116 Lightellll AJ, Coetzee WJC, Va ll Ni eke rk PJ, The identifica tion
93 Griego RD , Rosen T, Orengo IF, Wol f JE. Dog, cat and human o f bite marks us ing th e re fl ex mi c ro sco pe. J Forms
bites: a rev iew. JAm Acad Dermawl 1995: 33:1019- 29 . Odolltosrollliltol 1987; 5: 1- 8.
94 FUOItes L, Me lso n E. Prima lY and recurrent herpes sim plex 117 Sweet 0, Parhar 1Vl. Wood RE. Computer- based production
infection in a paediatric nurse resul ting from a human bite of bite mark comparison overlays. J Forel1s Sc i 1998;
(Comment). TnJecr COlli Hasp Epidemiol 1989; 120: 295. 43: 1050-5.
-_.
- - - -- -
I CHAPTER 23 I
PAEDIATRIC DENTAL IDENTIFICATION
G Howard Moody
II'JTRODUCTION
radiographs, with ante-mortem dental reco rds to
corroborate or deny an identity. This is the most
Accurate identification of the deceased by the relevant common procedure. 5
authorities is a human right enshrined in Article 6 of the • Profiling: the process whereby a forensic dentist,
United Nations Declaration of Human Rights (1948) - as in co llaboration with colleagues, who may include
this, for most people, will be crucial in coming to terms with forensic pathologists, anthropologists, archaeologists
the loss of a relative or friend, for the resolution of legal or and biochemists, attempts to 'reconstruct' from human
insurance disputes and for attaining 'closure', enabling the remains the age, sex, height and pOSSibly the ethnic
mourning process and due ritual to proceed. origin of the deceased with a view to creating a
The identification of individuals from their teeth is not possible liken ess that may cause so mebo dy to come
a recent development. In AD49, Agrippina, mother of the forward and offer a possible id entification when the
future Roman Emperor Nero, ordered the murder of Lollia deceas ed is completely unknown to the authorities 6
Paulina, whom she believed wanted to become the wife of
Claudius, as this would have prevented Nero from becom
ing Emperor. On being presented with Paulina 's head, COMPARISON
Agrippina, to confirm its identity, 'op ened the mouth with
her own hands and inspected the teeth which had certain While generally regarded as a common method of identi
peculiarities'.J In mediaeval times, illiterate apprentices sig fication, the number of persons, which includes children,
nified thei r bondage to a master by biting into a piece of in the UK who are identified by their dental characteristics
wax, leaving a unique mark and becoming 'indentured'. in anyone year is not known. There is no central data bank
Worldwide, the task of identification is the responsibility of holding this information, and there is no evidence to
either the judiciary or the police; in Scandinavia, the police suggest that all forensic pathologists refer evelY dental
are responsible,2 while in the USA id entification is the role identification to a dental surgeon, let alo ne a forensic den
of the medical examiner. J In England, identification is the tist. Corroboration involves the comparison of ante- and
responsibility of the coron er and in Scotland of the procura post-mortem records; the certainty of the identification is
tor fiscal. 4 The forensic dentist (odontologist) often has to related to the degree of concordance between the two
provide these legal authorities with an opinion, assisting the sets of records and whetl1er any discrepancies that occur are
identification using the techniques of comparison or profiling. explicable. 7 To be va lid , comparative identification requires
• Comparison: the procedure involved is the comparison identifiable dental characteristics, the accurate recording of
of post-mortem dental chartings, including these characteristics and accurate ante- mortem data .
436 I Paediatric dental identification
Dental Characteristics decayed, missing and filled teeth (dmfl. The effect of this is
that the dentition of individual children is becoming clini
Dental characteristics are understood to be those features cally less characteristic (Fig. 23.2).
that contribute to a unique charting (Fig. 23.1). Such fea Dental disease, as measured by the DMF, decreased in all
tures include decayed, missing, or restored teeth and fixed social classes in the UK during the period 1978-98. 15 In
and removable prostheses (dentures). Restorations range 1996, Kas te et al 16 reported the assessment of dental caries in
from small white fillings through large amalgam and porce the USA as part of the Third National Health and Nutrition
lain restorations, crowns, fillings, root canal fillings and Examination Survey (1998-91). This survey yielded esti
implants. Given the complete adult dentition of 32 teeth, mates for over 58 million children aged 1-17 years. It was
each with five surfaces, the wide range of dental materials, found that, among infants aged 12-23 months, only 0.8 per
design and construction variations available with respect to cent scored positively for early childhood caries while 62.1
prostheses and the variations in root patterns discernible on per cent of children aged 2-9 years were caries-free in their
radiographs, the number of permutations is enormous. deciduous dentition. One-quarter of the children and adoles
The statistical probability of concordant points on ante cents aged 5-17 years with at least one permanent tooth
and post-mortem records has been calculated for extrac accounted for about 80 per cent of the caries identified in
tions,a amalgam restorations,9 root morphologiO and digital permanent teeth. White et al 17 also describes how, in the
radiographic scanning. ll ,1 2 A single unique dental feature USA, the number of children free from dental caries
may, on occasion, suffice,I3 but this is rare, and the number increased dramatically between 1963 and 1987. By 1987,
of concordant points that are considered sufficient to collec
tively ascertain an identification remains controversial. 14
[n children, however, the situation is very different. In
much of the industrialized world, dental decay is in sharp
decline with a corresponding reduction in the number of
---
Comparison I 437
-- ~ -
IDEN TAL Re f No POSTMORTEM Ref No is known as 'toothprints'. This procedure, devised and
SURNAM E FO RENAMES 0. 0 .8 developed by Tesini et aJ25.26 in conjunction with Kerr
AD D RESS Dental, utilizes an arch-shaped thermopl astic warm wafer
Tela,
into which the child bites for 50 seconds. After 2-3 min
() ~Hv ) D:.~) (Ill})
100
~
~ » 10:' ,I
"'"
". utes' cooling, the wafer is sea led into a pl astic bag, thereby
preserving both a very accu rate impression of the teeth and
DC(~ 0 C~ 0) (v q~ ~ ~ ~D'O (0 (J)~) CD (~··D
a suffi cient numb er of shed ep ithelial cells for fut ure DNA
analysis. Good a nalytica l results after 3 yea rs appear to be
~D (0 0 CDdD G achieved if the wafer is a ll owed to dry prior to sea ling in
the zipl ock bag 27 Longitudinal trials will be necessary to
,-
,'A
determine how long the wafer remains a valuable source of
DN A. This is important sin ce the de nta l record in the child's
impression will beco me of decreasing value as teeth are
shed, replaced by permanent teeth or restored.
Figure 23.5 Asimple plastic lam inated dental char t fo r use
at the locu s. An indeli ble pencil/pen should be used to prevent Post-mortem Procedures
damp and dirty conditions fr om spoiling 'first sourc e' evidence.
This record should be retained even after a 'cl ea n' copy for filin g,
The Illajority of denta l identification s are carried out on sin
and later ante -mortem matching, has been created. Close
gle bodies. It is good practice for a forensic dentist to be
examin ation , includin g rad iographs of the teeth and ja ws, should
called to the site (locus) where the body has been fou nd since
be undertaken in th e labora tory once the remains have been
the teeth necessary for identification may be loose or, follow
removed from the locus. Any laboratory findings, includ ing
ing a fire, brittle, and requi re expert handing to avoid dam
radiograp hs, may be added to the clean post-mortem record.
age or loss th ereof. When remains are found in a shallow
grave, dogs or foxes Illay have disturbed them, resul ting in
loose teeth or dentures being displaced. A forensic dentist will
be aware of these possibili ties and ensure that the site is
meticulously inspected. If teeth or dentures are found in pl ace
within the j aws, they should be photographed ;11 situ, ca re
full y bagged, labelled and removed from the scene only with
receipt of permi ssion from the investigating poli ce officer.
In the mortuary, a detailed dental examinati on is
usually best carried out afte r the completion of the post
mo rtem examin ation, using a forens ic dental kit comp ris
ing a full range of dental hand instruments, a portab le
water pick to flu sh away debris and fas t-setting adhesives,
-_ su ch as cyan acrylate, to stab ilize tooth fragments. If den
~
.........
,
'
. -~ - ~, -
- ....
~-
..
~. tures are present, they shou ld be photographed in position.
If the deceased is beli eved to have had den tures but these
Figure 23.6 An almost illeg ible ante-mortem dental chart. are not immediately availab le then impressions of the jaws
shou ld be taken in ord er to create plaster models upon
which a later retrieved dentu re may be fitted.
illegib ility may m-ake interpreta tion almost imposs ible (Fi g. Making photographic records is importa nt. Thi s may
23.6). Co mputerized records should red uce this problem. best be achi eved by utilizing the servi ces of police photog
Among the common ante-mo rtem charting erro rs are raphers under the directio n and in the presence of the
transposit ion of left and right, misreco rdin g of a bu cca lly forensic dentists. However, where th is is not possible, a
extended restoratio n as a ling ual extens ion, a tlrst bicuspid 35-mm single-lens reflex (SLR) digita l camera (10 mega
charted as a second bi cuspid tooth, and first an d sec ond pi xels or greater) with a dedicated flash facili ty will nor
permanent molar teeth cha rted as seco nd and th ird molars mally enable the dentist to obtain satisfactory illustrations.
when on e molar is absent from the arch. Other errors Accurate dental charting in the mOliualY is best achieved
includ e adding a surface to a resto ration or incorrect with the assista nce of a dental nurse or a seco nd forensic
record ing of su rfaces fill ed, listing a tooth as restored when dentist. The author has a preference for Llsing pencil at this
it is not and om itting to record a unique fe ature such as a stage as charts may become soiled and ink may run .
sup ernumera ry too th or an apicectomy.7 Whenever possible, the jaws should be radiographed.
A novel ante-mortem development designed to assist in Small whi te (composite) fillings may be very hard to see in
the identification of child victim s of fires or mass di sasters the mortuary but readily identified on radiographs, as are
--- - ----~---~-----------
- ---
Comparison I 439
1r ,-#
Figure 23.8 Fire has destroyed the anterior teeth but the extent
to which the soft tissues around the mouth protect the teeth is
evidenced by the preservation of most of a plastic denture. There
is no carbon trail across the palate. The victim died before the fire
engu lfed him.
--
~- ---
Age estimation I 441
Deciduous dentition
~
2yearS
5 months
in utero
~ (+/-6mOS)
o
7 months
in utero
Prenatal
~ O QwJ
~
-oooo dY1 Birth ~~
~0 .' U .
~
o
'0"1'e =
moo", '" lc
'0000= (+/- 2 mas)
,
~
' 4 years
[./- 9 m"i
~
In the author's view, dental age estimation in the peri
and prenatal periods is not a routine procedure and should 18 months
be carried out by two forensic dentists working in close
collaboration with expert histologists and radiographers, ~~o
(+/- 3 mas)
Early childhood
initially examining the material independently but ulti Infancy (pre-school age)
mately presenting their findings in a joint report.
PFMP Figure 23.13
may be very hard to sex from their skulls. 7o The cellular In addition , neither of these studies makes a distinction
material in toot h pulp can be analysed for X and Y sex between eruption and emergence. For a highly detailed and
chromosomes/ I •n although vVhittaker et al n also showed robust scientific study of odontometrics relating to compar
that the determination becomes less reliable as putrefaction ative and sexual dimorphism the reader is referred to Human
proceeds. Intact cell nuclei are needed in most methods, and Adult Odon tom etrics by JA Keisser. 8o
this is clearly of limited use in burnt remains. 73 However, Teeth have more recently caught the imagination of the
the protected dental pulp is an exploitable source of DNA Department of Anthropology of the University of Durham,
because of its resistance towards physical a nd chemical where Buck and Vidarsdottir81 of the Evolution ary Anthro
exterior conditions. 74 Work by Urbani et al 75 has demon pology Research Group have investigated differences in
strated that dental pulp is an excellent source of DNA for the mandibular morphology of 'subadults' (i.e. not adults)
gender determination, being retrievable and capable of using geometric analyses. One hundred and seventy-four
an alysis from unprotected teeth heated up to 200°C for 15 mandibles from five morphologically distinct samples were
minutes. With teeth embedded in bone and soft tissue a 100 digitized and subjected to Procruste's analysis. The results
per cent success rate was obtained from teeth heated up to showed significa nt morphological differences between the
350°C for 15 minutes. samples and obtained cross-validation results of over 70 per
Recently, Karaman 76 reported achieving 83 per cent cent accuracy in identification of unknown individuals
accuracy in predicting sex usin g diagonal , as distinct from using the complete mandibl e. They suggest that these tech
mesiobuccal, measurements of anterior teeth. However, this niques could provide a method for the identification of race
degree of accuracy could only suggest, not prove, the sex in a subadult individual. This appears to be, at first sight, a
of skeletonized remains. useful tool in demographic research, but application to pro
Nevertheless, more sophisticated methods are not with filin g an individual skeleton can only result in a suggestion
out their problems. Kumar and Hegde,n using polymerase or, a t best, increased probability of gender determination.
chain reaction techniques, were able to identify sex with
100 per cent accuracy from the pulps of deciduous teeth,
but only up to 1 month after ex traction. After 6 months, the REFERENCES
method was no longer effective.
Carey E. Dio's Romall History, vol. 8, book 61. London:
William Heinem a nn, 192 5, p. 17.
2 Solheim T, Lorentsen M, Sundnes P et a l. Th e 'S candinavian
Height Star' feny disaster 1990 - a challenge to forensic odontology.
lnf J Legal Med 199 2; 104:339-45.
Fazekas and Kosa 78 showed that all the bones of the fetus 3 Delattre V. Forensic denta l id entification in the greater
can be measured and that, by constructing regression curves Houston area . J Forensic Sci 200 1; 46:1379 - 84.
for each bone, the size of the bone may be related to the
In Clark D (ed.) Pracrical Forens ic Odontology. Oxfo rd:
height at death of the deceased. Clement and Kosa 43 claim Butterwo n h-Hein emann, 199 2.
that the basilar part of the occipital bone, the zygomatic and 5 Hill IR. Id entification , the dentist and the coroner in Engl a nd
the temporal bones are excellent for this purpose, but the and Wal es . J Foren sic Odontostomatol 1988; 6:77-82.
best and easiest bone to identify is the mandible. They state 6 Jones DG. Odontology often is the final piece in a grim
puzzle. J Calif Dent Assoc 19 90; 26:650-1.
that the length in millimetres of the mandible is equal to the
7 Dailey J. Charting errors in ma ss disa ster denta l reco rds. In
total crown-heel body length in centimetres, while the max Bouers CM, Bel l GL (eds) Th e Manual oj Forensic Odontology,
illa is 1/20th the length of the body. At birth, with all the 3rd edn. Saratoga Springs, NY: Am erican Society of Forensic
bones of the body much further developed, direct measure Odontology, 1995, pp. 250-8 .
ments are possible_and these methods are superfluous. 8 Kiesser- Neilson. Person Idelltijicatioll by Mea ns oJThe Teeth:
A Practical Guide. Bristol: John Wright, 1977.
9 Philips VM. Th e uniqu eness of amalgam restorations for
identifi ca tion. J Foren sic Odol1tostol11atol 1983; 1 :33-8.
Ethnicity 10 Smith B. Recons truction of root morphology in ske letonised
remains with postmortem dental lo ss. J Forensic Sc i 1992 ;
Prior to the advent of DNA analysis, the determination of 37: 176-84.
ethnic origins from dental remains was unreliable because II Wood RE, Tai CE, Blenkins op B, Johns ton D. Digiti sed slice
interpOSition in forens ic dental radiographic identification.
the dental a nd orofacial indicators were very imprecise and
A m J Forensic Med Pathol 1994; 15:70-0.
the overlap betvveen racial characteristics was so consider 12 Wood RE, Kirk NJ, Sweet DJ. Digita l dental radiographic
able that, realistically, they were of little value in determin identifica tion in the paediatric, mi xed and permanent
ing the ethnic origin of anyone set of remains.7o Hurme 79 dentitions. J Forensic Sci 1999; 44: 9 10-16.
states that the eruptive process occurs 6 months to 1.5 years 13 de Villi ers CJ, Philip s VM. Pe rson identification by me ans of a
single unique dental feature. J Forens ic Odonrostomotol 1990 ;
earlier in blacks than in whites, but this trend is of little
16:17-19.
value in profiling and it remains thus unless the circumstan 14 Nonje CJ. Harris AJIiIP. Maxi llo-facia l radiology in forens ic
tial evidence indicating a specific ethnic origin is strong. dentistry: a review. J Forensic Odon tostol11a tol 1986; 4:2 9-38.
- -~-
-- - - ---- -
References I 445
15 Watt R, Sheiham A. In equalities in oral health: a review of the 37 Tyrell A, Eviso n, MP, Chamberlain AT, Green MA. Forensic
evidence and recommendation for action. Br Dellt J 1999; three-dimensiona l facial reconst ruction: historical review and
187 :6- 12. contemporary developments. J Forensic Sci 1997 ; 42 :653-61.
16 Kaste LM , Selwitz RH, Oldakowski RJ et al. Coronal caries in 38 BrockJebank L, Holmgren CJ. Development of equipment for
the primary and permanent dentition of children and the standardization of skull photographs in perso nal
adolescents 1-17 years of age: United States, 1988- 1991. identification. J Forensic Sci 1989; 31: 1373-9.
J Den t Res 1996; 75:631-41. 39 Simpson JDR. The investigation of an unknown skull with
17 White BA, Caplan OJ, Weintraub JA. A quarter century of superimposition of known faces. Dissertation for a Master of
changes in oral hea lth in the United States. J Dent Educ Science Degree. Edinburgh: University of Edinburgh, 1998.
1995 ; 59:19-57. 40 Malaver PC, Yunis YY. Different dental tissues as source of
18 Flink A, Kallestal C, Holm AK et al. Distribution of caries in DNA for human identification in forensic cases. Croat Med J
12-yea r-old children in Sweden. Social and oral hea lth 2003; 44 :306-9.
related behavioural patterns. Coml11l1ni~y De/If Hea ltft 41 Calonius PEB, Lun in M, Stout F. Histologic criteria for age
1999; 16:160-5. estimation of the developing human dentition. Oral Surg
19 La wrence HP, Sheiham A. Caries progression in 12- to 16 1970 ; 29:869-76.
year-old school children in tluoridated and deficient areas in 42 Kraus BS, Jordan RE. The Human Dentition Bejore Birth.
Brazil. Communiry Dent Oral Epidemiol 1997; 25:402-11. Philadelphia, PA: Lea and Febiger, 1965.
20 Moody GH, Busuttil A. Identification in the Lockerbie air 43 Clement JG, Kosa F. The fetal skeleton. In Clark DH (ed.)
disaster. Am J Forensic Med Pathol 1994; 15:63-9. Practical Forensic Odontology. London: Wright, 1992,
21 Sholl SA, Moody GH. Evaluation of dental radiographic pp. 43-52.
identification: an experim ental study. Forensic Sci fnt 44 Ashley KF. Identification of children in a mass disaster by
2001; 115:165-9. estimation of dental age. Br Dent J J 970; 129: 167 -9.
22 Ahlberg JE. We must get our numbers right. FDI Newslerrer 45 Ciapparelli L. The chronology of dental development and age
1987; 158:8-9. assessment. In Clark DH (ed.) Practical Forensic Odontology.
23 Ayton FE, Hill CM, Parfitt HN. The dental rol e in the London: Wright, 1992, pp. 22-42.
identification of the victims of the Bradford City football 46 Schour L, Hoffman MM. Studies in tooth development, 2. The
ground fire. Br Dent J 1985; 159:262- 4. rate of ap position of enamel and dentine in man and other
24 Ireland RS, Harris RV, Pealing R. Clinical reco rd keeping by mammals. J Dent Res 1939; 18:161-75.
General Dental Practitioners piloting the Denplan 'Excel' 47 Noble HW. The estimation of age from the dentition. In
accreditation programme. Br Dent J 2001; 191 :260-3. Harvey W (ed.) Dental Identification and Forensic Odontology.
25 Tesini DA. O'Malley KD, Schwartz S. Development of bite London: Kimpton, 1976, pp. 28 - 35.
impression technique for use in identification of unknown and 48 Moorrees CFA Fann ing EA, Hunt EE. Formation and
missing children. J Mass Dent Soc 1985; 34:61, 63 , 69, passim. resorption of three deciduous teeth in children. Am J Phys
26 Tesini DA, Harte D, Crowley K. Dentistry's role in the A l1thropol 1963; 21 :205-13.
identification of missing and unknown children: upd ate on 49 Moorrees CFA. Fanning EA, Hunt EE. Age variation of
the dental bite impression technique. J Mass Dent Soc 1999 ; formation stages for ten permanent teeth. J Dent Res 1963;
48:29-34. 42: 1490-502.
27 Tesini DA. Comments on the Toothprints Bite Impression jo r 50 Demerjian A. Postnatal grow th. In Falkner F, Tanner JM (eds)
Search and Identification oj Missing and Unknowl1 Children. Dentition in HU1110n Growth. London: Bailliere Tindall , 1978,
2003. http: //www.kerrdental.com/Toothprints/ pp. 413 -44.
Toothprints0f020Position20Paper 51 Koski M, Gam SM. Tooth eruption sequence in fo ssil and
28 Whittaker DK, MacDonald D. A Colour Atlas oj Forell sic modern man. Am J Phys Anthropol 1957 ; 15:469-88.
Dentistry. London: Wolfe Medical Publishers, 1989. 52 Demiljian A. Goldstein H, Tanner JM. A new system of dental
29 Aws G. An Overview and CompaTison ojCAPMI and DAVID age assessment. Hum Bioi J973 ; 45:211-27.
Programmes. Dissertation for the Diploma in Forensic 53 Demerjian A, Goldstein H. New systems for dental maturity
Odontology. Hatfield, UK: Uni ve rsity of Hertfordshire, 1999. based upon seven and four teeth. Ann Hum Bioi 1976 ;
30 American Board of Forensic Odontology. Body identification 3:411-21.
guidelines. J Am Dent Assoc 1994 ; 125:1244-52. 54 Schour L, Massier Iv!. The development of the human
31 Koe1myer TO. Video ca mera superimposition and facial dentition. JAm Dellt Assoc J941; 28: 11 53 - 60.
reco nstru ction as an aid to identification. Am J Forensic Med 55 Maber M, Liversidge HM, Hector MP. Accuracy of age
Patho/1982; 3:45-8. estimation of radiographic methods using developing teeth.
32 Perpe r JA, Patterson GT. Face imaging reconstructive Forensic Sci Int 2006; 159:S68-73.
morphography. A new method for physiognomic 56 Willems G, Van OImen A, Spiessens B, Carels C, Dental age in
reconstruction. Am J Forensic Med Patho11988; 9:126-38. Belgian children: Demirjian's technique revisited. J Forensic
33 Hill B, Macleod RM, Crothers A. Rebuilding the face of Sci 2001; 46:893-5.
George Buchanan (1506 - 1582). J Audioli Media Med 1996; 57 De Salvia A. Calzetta C, Orrico M, De Leo D. Third mandibular
19: 11-15. molar radiological development as an indi cator of
34 Philips VM, Smuts NA. Facial reconstruction: utilization of chronological age in a European population. Foren.sic
computerized tomography to measure facial tissue thickness Sci Int 2004; 146:S9-1 2.
in a mixed race population. Forensic Sci Int 1996; 83:51. 58 Garamendi PM, Landa M1, Ballesteros J, Solano MA .
35 Law CA, Bowers CM. Radiographic reconstruction of root Reliability of the methods applied to assess age minority in
morphology in skeletonised remains: a case study. J Forensic living subjects around 18 years old. A survey on a Moroccan
Sci 1996 ; 41: 514-17. origin population. Forensic Sci Illt 2005; 154 :3 -12.
36 Shahrom AW, Vanezis P, Chapman RC et aJ. Techniques in 59 Ciapparelli L. The assessment of dental age in Essex school
facial reconstruction: computer-aided facial reconstruction children using panoral radiographs with forensic app li cations.
using a laser scanner and video superimposition. Int J Legal Dissertation for the Diploma in Forensic Odontology. Lond on :
Med 1996; 108:194-200. Hospital Medical College, 1985.
446 I Paediatric dental identification
60 Monstrad H, Staaf Y, Welander U. Age es timation with the aid 71 SenD M. Ishi zu H. Sex iden tification of a human tooth.
of tooth developm ent : a new method based on objecti ve illl J Forel/sic Dml 1973 ; 1:8 - 1I.
measurements. SCll1l d J DCIII Res 1994; 102: 137 - 43. 72 Whittaker DK, Ll ewelyn DR , Jones RW. Sex determin ation
61 Liversiclge Hrvl, Lyons F, Hector MP. The accuracy of three from necrotic pu lp tissue. Br Dellt J 1975; 139:403- 5.
methods of age estimation usin g rad iog raphic meas urements 73 Murakami H, Yamam oto Y, Yosh ito me K et al. Forensic study
of developing teeth. Forellsic Sci il1t 2003; 131 :22- 9. of sex determination using PCR on tooth sampl es. Acta Med
62 Foti B, La lys L, Adalian P et al. New forensic approach to age OkaYII/JI1I 2000 : 54:2 1-3 2.
es timation in ch ild re n based o n too th erup ti on. Forel1sic Sci 74 Ga ill a rd H. Lucl es B, Kaess B, Ma ngin P. Th e use of the teeth
ill I 2003; 132:49- 56. in genetic fin gerprinting. Bull GrouJl i nt Reclr Sci SI011111tal
63 Ritz -Timme S, Ca ttaneo C. Collins rvlJ et al. Age est im ation: OdO lltO / 1994; 37 :65-70.
the state of the art in relation to th e spe cific demand s of 75 Urbani C, lastrucci RD, Krame r B. The effec t of te mp era ture
forensic pra cti ce.illl J Lego! M ed 2000 ; 113:1 29 - 36. on sex determination usin g DNA-P CR analysis of dental pulp.
64 Arany S, Ohtani S, Yoshioka N. Gonmori K. Age esti mation J Forel1 sic OdOI/IOS I0111I1to!1999: 17:35-9.
from asparti c acid racemi za ri on of root dent in by internal 76 Karaman F. Use of diagonal tee th measurem ents in predicting
standard meth od. Forellsic Sci 111 1 2004: 141: 127-30. gender in a Turk ish population. J Forensic Sci 2006;
65 Yekkal R. Meers C. Van Schepdae l A et al. Rclc emi za rion of 51 :630-5.
aspartic acid from human dentin in the estimation of 77 Kumar MG, Heg de AM. Sex ide ntification from exfoliated
chronologica l age. Fomlsi c Sc i i1l12006: 159:S89 - 94. primary reeth - a PCR study. J Ciil1 Pedilltr Den12005 :
66 Sajdok J, Pilin A, Pudil F et al. A new method of age 30:19 - 21.
est im ation based on the changes in human non -col lageno us 78 Fazekas lG, Kossa F. Determ in ation of the body leng th and
prote ins from dentin. ForeHsi(' Sci i ll l 2006; 156:245-9. age of fe tuses on the basis of th e dimen sions of the facial
67 Aitchison J. Sex differences in teeth, jaws and skull s. Delli bones. Chapte r X . In Fa zeka s IG, Kossa F (eds) Forensic
Prerel 1963 ; 14:52-7. Fetll! Os re% m'. Budapest: Akade miai Ki ado, 1978,
68 Biggerstaff RH. Craniofacial characteristics as dete rminants of pp. 16 5-96.
age, sex and race in foren sic dentistry. Del/r Clil/ North Am 79 Hurm e va. Time and sequ ence ot- tooth el1lption. J Forel/s
J 997 ; 21 :85- 97. Sci 1957 ; 2: 371 -88.
69 Borha CT. Cra ni o facia l characteri stics as determ inants of age. 80 Ki esser JA. HUIIIan Adulr OdOIIlollIclrics. Studies in
race and sex in forensic dentistry - a hand s on guid e. Biol og ica l An thro pology. Camb rid ge: Cambridge Uni ve rsity
J Forel/ sic Od OI110S(0))/lIto/ 1991; 9:47-61. Press, 1990.
70 Harvey W. Effec ts of sex, race, hereditary and systematic 81 Buck TJ, Vidarsdottir US. A proposed method for the
disea se on oral tissues. In Harvey W [ed.) Del/ta! Jdel/rijicotiol/ id ent ifi ca tion of race in sub- adult skeletons: geo metric
III/d Forellsic Odol/to!ogy. Lond on: Henry Kimpron. 1976. morpho metr ic analysis of ma ndibular morph ology.
pp.36-4 3. J Forel/ sic- Sci 2004; 49 : 11 59 -64.
I CHAPTER 24 I
THE EXPERT WITNESS AND EXPERT
TESTIMONY
Anthony Busuttil
Physicians are not ordinary witnesses, but give judg in that they are assisting the court through (Latin ex.) their
ment rather than testimony. acquired specialized skill, craft or profession (Latin periNa).
WG Dickson stated that: 'Another exception to the general
Digest - Emperor Justinian (AD483-565)
rule against examining witnesses on matters of opinion
occurs wherever the issue involves scientific knowledge, or
acquaintance with the rules of any trade, manufacture or
INTRODUCTION business with which men of ordinary intelligence are not
likely to be familiar. A foundation for such an examination
The courts, by the very diversity of cases that are brought must be laid by ascertaining whether the witness is a person
to their attention, wi ll, from time to time, invariably require of skill or an "expert" [the English term], under which is
assistance from those who possess intimate and profound included those who have a theoretical acquain tan ce with the
knowledge of specific and-specialized subjects, of which the subject, as well as men who speak from practical know ledge:]
court is not expected to have direct first-hand knowledge. Medical practitioners who are summon ed to the courts,
This is particularly the case in the context of an adversarial criminal or civil, in order to assist with medical matters can
system of law, as practised in Great Britain - and in most therefore appear as expert witnesses. In criminal matters,
Commonwealth countries - and the USA, whether the case is their ex peliise is mainly required for the interpretation of
being tried in the civil or criminal courts. For example, if the traumatic changes and causes or modes of death; in civil
matter under consideration by the court is the safe construc cases, medical expertise is requ ired for cases of alleged
tion of a bridge, it is unlikely that the court would know medical or industrial negl igence and in li tigation relating
much about the specific matters related to this, unless to compensation for injury and disability.
mechanical engineers, architects, steel erectors and quantity As with all other evidence given to the courts, the testi
surveyors ass ist them by providing them with basic details of mony is taken under solemn oath or affirmation; thereby, it is
constructional requirements, good practice and building empowered with a further, more robust, sea l of assurance as
techniques. Such persons are referred to as eJ.pert witnesses to its veracity, totality, professional integrity and impartiality.
*The mal e gender that is used throughout th is chapter also refers to the female gender in all instances.
448 I Expert witness/expert testimony
As Cicero stated, 'the court is the temple of truth, and the position to give excellent expert advice on
search for truth is critical towards the conduct and conclu herniorrhaphy and its complications. The expert must
sion of a fair trial'; this should be the tenet that underlines appreciate the boundaries of his expertise and
the involvement of the expert witness. knowledge, and always ensure that he is not making
Similarly, in the preparation of cases for trials, be these pronouncements on subjects outside his experience.
criminal or civil cases, it is often useful for both the prose 3. The delivery of the evidence has to be such as to
cuting auth orities and legal defence team (criminal) or, for enable the court to appreciate and understand the
the legal team putting the case together for one or other specialized information that is being imparted to them;
of the litigants involved (civil) , to acquire information on if this evidence is written down in the form of the
subjects with which they are not familiar. This would guide report, it must be user-friendly, devoid of long,
them with respect to the strength or weaknesses of the case rambling sentences, and bereft of a surfeit of jargon
and in coming to considered decisions as to whether to and statistical calculations. It should be laid out
plead guilty to the charges, to settle out of court, to instruct logically, in sequentia l chapters, and it cannot be seen
further investigations to be carried out prior to the trial to be muddled, perhaps focally inconsistent, with
or to contest the trial. A very onerous duty and a heavy conclus ions in it that are not fully justified and
responsibility are placed on the expert. Initially he has to substantiated by wi?at appears earlier in the evidence.
consider very carefully whether or not he possesses the The information that is being imparted, albeit often
qualities appropriate to fulfilling the expectations placed quite complex and difficult, would somehow have to
on him by the courts or by legal professionals in the partic be titrated to a level that is readily comprehensible by
ular matter under review. the jury. The lawyers - and failing them the judge
himself - will usually steer the expel1 in the correct
direction while on the witness stand, and may ask him
MISSION STATEMENT OF THE EXPERT to elaborate on certain passages or chapters of his
evidence, perhaps by giving analogies and examples
By its very nature, the criteria that expert evidence has to that will assist the jury's comprehension.
satisfy are the following: 4. This evidence has to be, as much as possible, f1awlessly
1. It refers solely to specialized medical matters and dispassionate, totally unbi ased, non-partisan, carefully
themes, and not simply to matters that can be deduced balanced and comprehensive. It must not contain
from a common-sense or ordinary knowledge approach views tailored to the advantage of the side instructing
or by recourse to simple logic. It is on such specific the expe11. In an adversarial system, whichever side
issues, that the court - namely the ordinary 'ladies and calls on the services of the expert, the evide nce given
gentlemen of the jUly' and/or the presiding judge(s) to the court should be unsullied by leanings in favour
may require to be enlightened in order to reach the of one or other side. Experts - including medical
more appropriate verdict. expe11 witnesses - who fail to achieve this level of
2. It has to be delivered only by those who are probity have been variously referred to in a derogatory
appropriately qualified to do so, both by virtue of an fashion as 'hired guns' and, worse, as 'gobs for hire'.
appropriate duration of first-hand specific experience The expert must always keep in mind that he is not
in the matter under review and by having acquired there to win cases or to ensure that his 'side' triumphs
their expel1ise by periods of study - these academic against all odds; he is there to assist the court, to
degrees and diplomas shou ld be related to the subject the best of his capabilities, in reaching the most
areas being addressed as, an expert. The direct expertise appropriate decision or verdict, and no more. On
which is being sought should be kept in mind by the many an occasion, the temptation may be very
expert who is instructed, and, if it is felt that the pronounced to veer away from this rule - but on the
subject is not within his direct professional knowledge expert's head be it if he is found to have taken sides in
and specific experience, then he should decline to act this manner.
in the capacity of an exp e11 in the particular case. In 5. The data adduced in the expert's evidence must be
the recent past, a trend developed in civil court rooms accurate and up to date, both in terms of listing all the
in which persons with a sound general expeliise spoke information given to him in his instructions by the
on more specific matters of which they had little first legal personnel and in quoting the current peer
hand knowledge, and thereby brought the role of the reviewed literature. Using stale and superseded
expert witness into a certain level of disrepute. For references in either books or journals that are later
example, a general surgeon cannot, under ordinary discovered to be so by the courts discredits the expert's
circumstances, be expected to speak from a vantage testimony and his expertise. If the information given
point of expertise on intricate stereotactic neurological originally was somehow incomplete or in accurate, at
procedures; if he does, he is quite likely to be least in the light of the evidence actually stated in
overstretching his expertise. However, he is in a court, and the conclusions reached are no longer
Professional witnesses I 449
tenable in the face of what was actually heard from the indeed be present when the testimony is actually taken. The
witness box. then the expert should honestly inform expert is never immune to criticism and to adverse com
the instructing side that, as a direct consequence of ment; he can, and most certainly will, be challenged in open
this, his opinion has had to change. The courts, and court through searching quest ions and cross-examination.
eventually the instructing laYlryers , will accept Evidence from other experts may be adduced to demonstrate
sometimes quite reluctantly - that honesty is indeed that in the evidence given to the court, th ere were short
the best policy. This implies that there may be comings, inadequacy of knowledge, flawed interpretations, a
occas io ns when it is expedient and indeed essential failure to consider all po ten tial options, and so on. All these
that, with the COUlt's permission. the expert actually matters should be considered prior to agreeing to prepare an
sits in the court room and listens in to the factual expert report.
evidence as it is being delivered. There is no scope for arrogance or ex cathedra dogmatic
6. All the data referred to by the expert must be properly pronouncements of the ipse dixit style; the court will be
acquired at clinical examination or necropsy or from sure to query and to probe, and the expeli has to cultivate
available documents. To do otherwise would be an unperturbed style that enables him to parry such ques
fraudulent and. if the case proceeds to court, may be tions and indicate why he disagrees with the propositions or
tanta mount to perjury. In this respect, the infamous scenarios that are being put to him , a nd to remain cool and
instance of an expelt clinical neurologist who gave collected, even under what may appear to be serious provo
evidence on the findings of a clinical examination that cation and verbal pressure from those who have been
he had never act ually personally conducted is perhaps trained so to do.
a major sa lutary and admonitory lesson 2 Thus, the opening gambit in couli is for th e expert to
establish, to the satisfaction and probing of the court, the
The genera l judicial requirements of the expeli were breadth and depth of his specific expertise, in order to pro
summarized in the English civil case, National Justice Com vide reassurance. This is always the case, whether or not spe
pania Naviera SA v. Prudential Assurance Co. Ltd, tile cific criteria are actually laid down by the courts delineating
Tk.arian Reefer (1993) 2 Lloyds Reports [2] 68, 81-82, and, who can and who cannot be an ex pen. In continental juris
after appeal, in (1995) 1 Lloyds Reports 455. These are par dictions, the court has access to a list of persons who have
aphrased as fo llows: been considered, vetted and accepted by the courts as pos
• Expert evidence must be an independent product of the sessing the appropriate standing to qu ali fy them as experts.
expert uninfluenced by the exigencies of the litigation. In the USA, the Federal Rules of Evidence (Rules 104,40\-3,
• Expert evidence must be of independen t assistance to 702-6 - the Committee on the Judiciary, House of Represen
the court by way of objective unbiased opinion within the tatives, US Government publication) establish the authority
expert's expertise, wi th the expert never assuming the of the court to determine whether a person can be considered
role of an advocate. as an expert, whether his testimony can be heard, the type
• The expert witn ess must state the Facts or assumptions of evidence that can be admitted in court and the permitted
for the opinion expressed, never omitting material scope of such testimony.] In the light of the Woolf Report,
facts. the position in Britain may already have changed in some
• The expen witness must clearly indica te when a matter respects towards the continental system.
falls outside his ex pertise.
• The expert witness must say if the data available are
insufficient or if he cannot assert that the report PROFESSIONAL WITNESSES
contains the whole truth.,
• If the expert witness changes his opinion, he must A fine line of distinction needs to be drawn between the
co mmunicate this to the court and to the parties so-called professional witnesses and expert witnesses.4 Any
involved in the action. professional person, including a medical doctor, who is
• If any photographs or other material are used to required to give testimony a bout professional matters is
compile an expeli repoli, all of this must be made considered as a professional witness . Thus, for example, a
available to other parties. young ARE department doctor w ho gives evidence about
when he saw a patient wh o self-referred to his department
- describing the nature of the injuries that he observed ,
CLAIM TO EXPERTISE with their measurements and locations, the number of
sutures that he inserted, the presence and absence of frac
Within the forensic arena, no matter how eminent the expert tures - is giving professional evidence. Yet he should not
witness, aU aspects of the claimed expeliise and the testi be expected to be able to inform the court, from a position
mony given will be challenged by the legal parties involved of expertise, whether the injuries he described were likely
in the trial - quite often themselves aided by 'experts' whom to have been caused in any particular manner or by any
they have specifically hired to assist them and who may specific weapon, as this is unlikel y to be within his direct
450 I Expert witness/ expert testi many
that the courts adm itted expert evidence deduced fro m continued to cite cases, altho ugh they had been t he subject
weJl-recognized scientifi c principles or discoveries: ' the of decisi ve judicial findings of non-accidental injUlY inflic
thing from whic h the ded uction is made must be suffi tion. In summary, the judge considered this evidence to be
ciently establ ished to have ga ined general acceptance in unrel ia ble and inexpeli and he threw out his 'expert'
the particul ar field in which it belongs'. The legal grounds evidence.
of acceptab ili ty of expert evidence are even h igher a nd In the context of child ab use, particularly sexual abuse,
more demand ing than those of the expert's discipline 6 This a considerabl e debate has arisen, particularly in the USA,
'general acceptance test' changed after the Daubert v. concerning the weig ht tha t should be attribu ted to testi
Merrell Dow Pharmaceuticals Inc (1993) 113 S Ct 2786 7 case mony of such historical ab use because the subj ect could
and was replaced by a num ber of more flexible guidelines : not recall spontaneously a ny such occurrences; expert psy
• Can the theory or techni que on which t he evidence is cholo gists and psychotherap ists were required to gather
based be tested independ ently? Has it been tested? information through repet itive interviewing and sessions
• Can a determination be made of its fa lsifiab ili ty, i.e. under hyp nosis (sometimes combined with use of medica
can it be proved to be wrong if it is in fact wrong? tion). These in te rviews we re carried out in alleged attempts
• Has the theory or technique been subj ected to peer to tap the so-called ' rep ressed-memory', ' recovered mem
review and pub lic atio n? ory' or 'false-memory syndrome'. 10.11 The coulis a nd , per
• What is currently known of its potential error rate? haps more teJlingly, the insurance companies who refused
• Is the theo lY or technique generally accepted? to pay compensatio n to those alleged ly stressed and trau
matiz ed by suc h repressed occurren ces have not taken a
The Kumbo Tire Co Ltd v. Carmichael (1999) 526 US 137 very favourab le view of the entire mat te r.
case perhaps enhanced further the flexibility of the test by In civi l cases, in parts of Great Britain other than Scot
stating that 'Daubert's list of specific factors neither neces land, there has been a move to ration alize the use of ex perts,
sarily nor exclusively applies to all experts in all cases'. The particularly in personal injury compensation, and in a con
US supreme co urt in Daubert's perhaps sum s up the 'proof sideration of the moves towards cond itiona l fees by law
of the pudding' in relation to expeli ev idence, namely firms, namely the 'no win, no fee ' situation. In this field, the
'vigoro us cross-exam ination, presentat ion of contrary evi lawyers act ing for the claimant under the pre-action proto
dence and carefu l instruction on the burden of proof as col set-up prepare the case, obtain a ll the relevant medical
being the t raditional means of attacking shaky but admis documentation abou t their client and submit this to the firm
sib le evid ence'. of so licitors acting for the person/company who will be
In Brita in, detailed tests of admissibility of expert evi defending the act ion. If possible, the matter is resolved at
dence have not exercised the judiciary to any major extent; this point, before court proceedings are ini tiated. If there is
admissibility, to a large extent, has been based on a prag no agreement reached at this stage, then the Civil Procedure
matic approach from the bench, erring on the si de of accept Rules (SI 1998/3132 Part 35) and Practice Direction [1991]
ance and acquiescence rather than otherwise, with evidence 1 WlR 1124, apply. Basically, these infer that the ex pert evi
being ruled to be admissible if it is considered to be relevant dence must be reasonably necessary (Rule 35.1), with the
and its probative va lue is considered to outweig h its prejudi overriding duty o f the expert being to the co uli.) The court's
cial effect. s Thus, in R v. Robb 93 Cr App R 161, the COLlli of permission is sought for the expert evidence4 a nd is then
Appea l upheld the trial judge's admiss ion of evidence given obtained as a w ritten report. 5 At this stage, the opposing
by a phonetician on voice recognition as expert, even party may put written questions to the expert. 6 The cOUli
though misgivings were expressed about the scientific basis may direct that evidence will be required from only one
for such ex pertise. , expert,6 a single, joint expert who may be given instructions
In the English family division, in the Re AB (child abuse; by both parties in th e litigation. s If the court believes that
expert witnesses) (1995) 1 FLR 18 1 case the judge had to certain documentation should be further disclosed to the
weigh the evidence before him in terms of fractures sus expert then it may direct SO.9 Such disclosed evidence may
ta ined by a young child. An expeli for the defence claimed be adduced in court, I I but not undisclosed evidence. 12 Th e
that a form of brittl e bone disease akin, but not identical , expeli may seek direction from the court. 13 The Daniels v.
to the va rious types of genetica Lly documented brittl e bone Walker [2000] 1 WlR 1382, CA, Lord WoolfMR case lai d out
disease existed that did no t have simila r hereditalY a nd the proper approach for the cOUli .
familial characteristics of osteogenesis imperfecta but could In the criminal cOUlis, the prosecution and the defence
account for the spontaneous occu rrence and accidenta l wi ll usually instruct their own experts, and it is here, perhaps,
infliction of injuries, and thus exclude criminal damage and that the ogres of bi as and partisanship become manifest and
assau lt. 9 This 'temporary' brittle bone disease had not been have to be guarded against in the cut-and-thrust of the foren
substantiated by standard research studies or by pUbli ca sic courtroom atmosphere and the persistent questioning by
tion s in peer-reviewed journals. In particular, the judge, in barristers and advocates. In some instances, th e judge wi/ I
his criticism of this expert, indicated that he lacked obj ec have seen, prior to the trial, the written repOlis from other
tivity, omitted factors that fai led to support his theory and experts, and he may decide that there is a sufficient agreement
452 I Expert witness/expert testimony
and similarity between the two that perhaps a joint statemen t sexual abuse in which he has examined the patient with a
or minute of the points agreed to by both experts should be colposcope.
drafted with only the points of disagreement coming to the The report will then have to cite what documentary
jury under their direction for assessment. and other evidence (photographs, histology slides, sketches)
In this instance, there should not be any disagreement have been available for the compilation of the repOlt. If other
about facts and the information collected as evidence. If this inform ation has been used as background, then this should
is shown to exist, there is a fundamental evidential problem be included in the list. The various findings, if relevan t, in the
that has to be resolved. Thus, for example, if there is a dis clinical a nd any other reco rds ava ilable are then pn~c i sed or
pute about the substantive clinical findin gs in a patient who cited in fulJ as the case dictates. Obvious errors and inconsis
has been examined separately, albeit on different occasions, tencies have to be referenced, for example if, in a particular,
by two do ctors or missed limb fractures in a utopsy findin gs the nurse's notes refer to a left-sided pneumothorax but the
then th ese matters require earl y and definitive resolution . doctor's record mentions a right-sided pneumothorax. It may
Another fundamental problem relates to the breadth of be necessalY to indicate that the evidential bundle available
the informatio n collected. A skeletal surveyor a blood clot is inco mplete, for example clinical photographs alluded to are
ting scree n that has not been carried out by the prosecu tion not included, a sample sent for alcohol analysis has not been
expert when it is thou ght to be essentia l, omission of co l reported on or X-ray reports are not included. It is important
poscopic examination in a child abuse case, lack of histo to remember that those with legal training will have already
logical examination on a utopsy blocks, failure to instruct gone through the notes page by page; it is not co nfidence
toxicology or microbiol og ica l investi gations are all funda inspiring if the la wyers point out the omission in yo ur repOlt
mental flaws that never come before a jUly but should be of what turns out to be a vital bit of information that you
resolved prior to the court proceedings commencing. have missed from you r perusal of the clinical notes.
Having laid out all the facts , the co nclusion s should be
derived logically and consistently. Of various co nclusions
COMMUNICATIONS FROM THE EXPERT that are possibl e, there should be an indication of which
WITNESS one is the more likely. It should be kept in mind th at when
referring to wounds one must be alert to the fact that
By and large the expert will communicate w ith th e instruct lawyers may wish to know how muc h force was required,
ing la wyers and with the courts in two ways: through a whether the wo und could be considered as life-threa tening
report and , eventually, in person (when evidence is given in or disfiguring, or whether the wo und could have been acci
court). In all instances, there should always be absolute clar dentally inflicted or indeed se lf-inflicted. The options
ity in such communicatio ns, to the extent of using very ele should be ranked and not mixed up in random order. One
mentary co mprehensibl e language. will be doing a major disservice to the instructing solicitors
The ex pert 's reports should be dated, properly typed and a nd to oneself - if the conclusio ns are jumbled up in any
form atted, with numbered pages and, when approp riate, order and one has to admit in open court that the probability
numbered paragraphs. It should start by telling the readers of one scenario is 1 in 1000 while tha t of another is 1 in 5.
of th e report'in whi ch matter this report is being produced, In coming to conclusions, one may have to cite the lit
and who has instructed it and w hen. The reader would then erature - joumals and books (Jatest edition) - to bolster up
want to be introduced to the report wliter in terms of how the arguments that are being made. These have to be cited
he claims expertise in the matter under revie w, and all the in full. Ph otocopies of releva nt articles and chapters in
releva nt qualifications and experience should be listed; books should be appended to yo ur report.
some prefer to attach t his a~ a mini-curriculum lJitae, which The report has to be si gned, and it is a useful practice, to
is entirely acceptable, and some suggest that it is preferable. indicate that one has read the report carefully before sub
If the a uthor has been instructed previously in simil a r cases miss ion and cOITected any typing and other errors in it, to
and has also given evidence in court, this should be stated. sig n each p age individually. In spite of this - but obviou sly
If there are relev ant publications in the name of the expert much more so without a check - a numb er of howlers do
these should be cited and, when relevant, copies of these survive, even after the most meticulous trawling for errors
publications should be attached to the report. in the rep ort, and one has to accept these candidly and
Date of qualifi cation , General Medical Council (GMC) apologetically.
number and th e memberships a nd fello ws hips of va rious
royal colleges, institutes and learned societies sho uld all be
cited with the date of entry. The cunent post(s) held and DECLARATION BY THE EXPERT IN THE REPORT
previous relevant posts, with dates of appointments, should
be given. (In Scotland , the date of bilth of the ex pert is a It is essential that the ex pelt witness shows that he has
useful, but not essential, adjunct.) The report writer shou ld accepted his responsibilities and sbict duties to the COUlt full y
give some deta ils of his expertise in de al ing with th e and, as such, it is important that a decl aration is appended at
specific case, for example the number of cases of child the end of th e report to enunciate these principles.
--- - ~.
Pre-trial communication I . 453
This ' declaration' should include a statement to the Think carefully before you a nswer a question and indeed
effect that: allow an appropriate pause. Speak clearly and slowly
• The report submitted, and any subsequent oral enough that the judge's pen can keep up with you if he is
testimony which may have to be given later under recording the evidence in long hand. If you were un ab le to
oath or affirmation, is being given solely to assist the hear the question for one or other reason, or you wish the
court and that such an overriding obligation takes full qu estion to be repeated or rephrased, do not hesitate to say
precedence over any obligations and loyalties the so. Be careful to a nsw er the specific questions asked and not
expert may have to those who have engaged him. the qu estions that you would rather had been asked. The
• The report is a tru e and complete re port, and any judge may intervene if you do not play by his rules.
opinions expressed in it are correct. Whatever happens, keep your 'cool' and do not exhibit
• The report has been produced with all reasonable care bad temper: do not scowl, sneer or sulk. You may be pro
to ensure that it meets the high standards expected voked and tried ; thus, this advice is more easily' imparted
by the courts. than heeded. At the risk of being too nice, do not press your
• In further reference to the completeness of the report, advantage by chea pening the opinions of others out of hand
th e expert should indicate that he has cited all the or by making cheap points against them, even if they deserve
sources of inform ation on which his report is based, this. If, however, the question is objective and one is required
whether it favours his arguments and opinions or to be critical of others, be fair in expressing your opinion, and
othelwise. give cogent an d competen t reasons for such criticisms.
• The report is independent, unbiased and uninfluenced There is no place for arrogance and overconfidence ; be
by the views of others. patient and courteous. Lord Avonside, in Assessor for Lothian
• It is ap preciated that th ere could be cross-examination Region v. Wilson, states that 'It can't be overemphasized that
on it. no committee or, for that matter, no court is in any way bound
• Finally, it has to indicate that any professional fees due to accept expert evidence or to be dictated to by experts'.
for expert services rendered by the witness, i.e. for There is rarely scope for humour; the old adage that the
compiling the report and subsequently presenting judge is the only person in the court room who can make
evidence, does not depend in any way on the final jokes, and that it is indeed mandatolY to laugh at them, is
outcom e of the cas e. There have been regrettable well made. The following two utterances by learned judges
instances in North America of experts producing may illustrate this matter.
favourable reports in return for large remuneration, In Rolland v. Lothian Health Board, Lord Ross stated: 'I am
particularly in high-profile cases, or in oth er instances not disposed to place much weight on the evidence of Or H. He
where very large monetary settlements were in the struck me as a witness who in this instance was unable to
sights of the lawyers. 12 - 14 approach any question obj ectively; he was intent on defend
ing his hospital and his staff and I found it difficult to accept
some of the answers which he gave. Professor F gave his evi
IN THE WITNESS STAND OR BOX dence in what might not unfairly be described as a didactic
manner; most judges, I imagine, dislike being lectured by w it
It is said that lawyers require experts who can articulate nesses, and in tills respect I count myself among the majority'.
opinio ns clearly and pe rsuasively in court, can produce a In McVey v. Central Regional Council, RG McEwan sit
polished but not patronizing presentation and who do not ti ng as a tempo rary judge stated: '... He is a very experi
simply spout out raw technical detail. A speaker informs, enced surgeon and experienced witness. Whether this
and indeed persuades, by his, words, by his body language makes him unable or unwilling to accept that he could ever
and through the modulations of his voice at the time in be wrong I know not. However, I find his at ti tud e less tha n
which he is speaking, with the words actually spoken per helpful. He was patronizing almost to the point of discour
haps being the least effective communication. Body lan tesy with the counsel for the pursuer and at times sarcastic
guage, in terms of facial expressions, posture, eye contact ad personam over certain proper questions. Wisely counsel
and hand movements, is important and conveys signa ls to did not allo w herself to be provoked and responded only
the jurors and to the lawyers. with courtesy. I might not have been so patient. For these
To be able to do this well, you have to be very familiar and other reasons I did not have any co nfiden ce in his tes
with your script, and that includes all the other documen timony and where it differed from the other medical wit
tary productions that may have to be referred to in your nesses, I had no hesitation in preferring them.'
evidence. It impresses the jUly tha t you ac tually find imme
diately the relevant paragraph in your report that refers to
a particular question put to y ou , or th e page in the case PRE-TRIAL COMMUNICATION
records on which a particularly significant item has been
recorded and, what is more, to be able to assist the court by The expet1, to a much lesser extent but increasingly so - and
helping court officials find the right place. long may the trend continue - will engage in pre-trial
.. ~ -
454 I Expert witness/expe rt testimony
conferences with the side which has instructed him. There, as unfairness can be avoided; this will also prevent the judge or
in other communication, he should be absolutely dis jury, especially the latter, from falling into error. Over the
passionate and ruthless but, without fear or favour, should years, the rules of evidence have been relaxed; in Scotland, it
give all the points in favour and those against. Holding back was only in 1898 that the accused could give evidence before
or camouflaging the problems at this stage will not do his thejury trying him. In this respect, doctors often utilize infor
side, or the case, any good; when the question is asked in mation obtained at second hand from others in coming to
court, he will have to concede (and blurt out what he may their decisions, i.e. hearsay evidence. As a generality the
have politely, but reluctantly, tried to gloss over during a court will accept only the 'best evidence'. What constitutes
chummy cup of tea in chambers). The certainty, or otherwise, hearsay is a legal matter to be determin ed by the judge.
of the opinion given should be fully aired with counsel at this
stage; this will go a long way in saving on expensive court
time and, even better, preventing miscarriages ofjustice. CONCLUSION
Yet counsel should not assist the expert in any way in
the preparation of his report. Lord Wilberforce at 256 in Perhaps the classic statement of Lord President Cooper as
Whitehouse v. Jordan stated: 'While some degree of consul found in Davie v. Magistrates of Edinburgh sums up best
tation between experts and legal advisers is entirely proper, the scope of expert evidence:
is it not necessary that expert evidence presented should
be, and should be seen to be, the independ ent product of Expert witnesses, however skilled or eminent, can give
the expert, uninfluenced as to form or content by the exi no more than evidence. They cannot usurp the func
gencies of li tigation. To the extent it is not, the evi dence is tions of the jury or judge sitting as a jury, any more
likely to be not only incorrect but also self-defeating'. than a technical assessor can substitute his advice for
the judgement of the court ... Their duty is to furnish
the judge or jury with the necessary scientific criteria
CONFLICT OF INTEREST for testing the accuracy of their conclusions, so as to
enable the judge or jury to form their own independent
The 'other' side in any forensic case may also wish to speak judgement by application of these criteria to the facts
with you before the trial. Answer all the questions asked proven in evidence. The scientific opinion evidence, if
by them truthfully and honestly, and be willing to admit, if intelligibl e, convincing and tested, becomes a fa ctor
required , that certain aspects of the case had not been pre (and often an important factor) for consideration along
viously familiar to you and that on the basis of this fresh with the whole other evidence in the case, but the deci
information there may be scope for revising the opinions sion is for the judge or jury. In particular, the bare ipse
already offered. The expert may find that he may thu s dixit of a scientist, however eminent, upon a contro
inadvertently be sucked into a conflict of interests l 5 . 16 versial issue, will normally carry very little weight, for
between the parties involved. Perhaps the earthy statement it cannot be tested by cross exa mination nor independ
of the Unites States Supreme court in Williams v. Florida ently appraised, and the parties have invited the deci
should be kept in mind in this respect: 'The adversary sys sion of a judicial tribunal and not an oracular
tem of trial is hardly an end in itself; it is not a poker game pronouncement of an expert.
in which the players enjoy an absolute right always to con
ceal their cards until played '.
In some other instances, there could be a more genuine RECENT DEVELOPMENTS
conflict of interests in the m,ind of the expert, for example
when giving evidence in a child abuse case knowing In Kariing v. Purdue (2004 SLT 1067), the Scottish courts were
fuJJ well that the evidence may point to a member of the asked to consider a claim for damages against an expert wit
child's family, also a patient of the practice, and that there ness. Mr Karling had been convicted of murder in 1995. Dr
may be a disruption of that family unit or an unfavourable Basil Purdue was the defence expert asked to carry out a
divorce settlement. The expert, thus, comes to lie on the horns post-mortem examination of the victim. Dr Purdue's findings
of a clilemma; should his allegiance to the child override loy were consistent with those of the prosecution case. Mr Kar
alty to other patients - the parents? The child's welfare may ling was convicted and sentenced to life imprisonment. In
have to override obligations of medical confidentiality to the 200], fresh scientific evidence came to light indicating that
child (and his family) in the interests of public safety. there was no sound evidence on which to conclude that the
victim had been suffocated. In an appeal, Mr KarJing's con
viction was quashed. Subsequently, Mr Karling issued civil
RULES OF EVIDENCE proceedings against the expert. He sought damages of
£75000 on the grounds that, had the expert performed the
In the courtroom certain evidence can be excluded in post-mortem examination with sufficient care and compe
certain well-defined circumstances. The policy reasons for tence, there would have been a reasonable chance that he
such exclusions are basically ones in which confusion and wo uld have been acquitted of the charge. Purdue defended
References I 455
the actio n on the grounds of his abso lute immunity from suit. 2 Lav in JH. Ev elyon e believed the pl ainti ff except his doctor.
He argued that, in the absence of malice, no li ability could be Med Ecol/ 1991; 68:34-4 1.
3 Kaufmann HH. Th e expert witn ess. Neither Frye nor Daubert
attached to him, regardless of how incompetently th e post
so lved th e prob lem: what ca n be don e? Sci Jmtice 200 1;
mOliem or the subsequent report was carried out. 41 :7-20.
Mr Karling's respon se to this pl eading was that, by indi 4 Calvert-Smith D. Medica l evidence in criminal prosecutions.
cating in his rep Oli that he was attaching his conclusions l'vI ed Leg J 2001 : 68 :11 7-2 9.
sepa rately, Dr Purdue was effectively adm itting that part of 5 Ca lvert- Smi th D. Medi cal evidence in criminal prosecutions.
Med Leg J 2000: 68:117.
his report was intended for Mr Karling's le ga l team only. He
6 Starrs JE. In the lan d of agog : an allegory for the exp ert
argued further that this suggested that the expeli was act witn ess. J forellsic Sci 1985; 30:289-307.
ing as an advisor, therefore owing Mr Karling a duty of 7 Annas GJ. Scientific ev iden ce in the court room; the death of
care that was capable of standing separately from Dr Pur the FI")'e rule . N £lI gl J Med 1994: 330 : 101 8- 2 1.
due's role as expert witness. In rejecting these arguments, 8 Mi tch ell J. Scientific evi den ce: pri nciple or pragmatism .
Aied Leg J 200 1: 69: 117- 19.
the co urt said that the link in tim e and function with the
9 Ed itorial - Osteo genesis imperfecta, non-accidental injury and
criminal proceedings was more than suffici ently close to temporary brittle bo ne disease . Arch Dis Child 1995: 70:169.
conclude that, in relation to his engagement to perform 10 Brandon S. Recovered memories; some aspects of the
professional or expert services, Dr Purdue was immune controversy. Med Leg 199 9; 67: 25 -34.
from suit in Mr Karling's instance. The test imposed by the 1J Brahams D. 'Repressed memories' and the law. La ncet 2000;
356:358.
courts was whether the expert was preparing to give evi
12 Katz J. The fallacy of the impart expert rev isi ted . Bull Am
dence in proceedings. It was the proximity of th e criminal Aca d PS),ciliatr Law J9 92 ; 20:141-5 2.
proceedings that, in effect, made Dr Purdu e's 'advisory' 13 FacUo D. Buccia relli R. Peer review of the expe rt wi tness: an
capacity indistinguishable from the work done in the pro opportunity to improve our medical liability system. J Child
vision of services as an expert witness. Neu rol 1995: 10:403-4.
14 Fisher CWo Dombrowski MP, Jaszczak SE et al. Th e expert
Another more recent impoltant development in the field
witness: reo l issues and suggest ions. Am J Ob ster GYl1ecol
of expert evidence was the decision of the Court of Appeal 1995: J 72: 1792-800.
in ThE GEn eral Medical Council v. ProfEssor Sir Roy 15 Dona ldso n L. Kap lon C, Leung W-c, The medica l expert
Meadow (Her Majesty's Attorney Generol iJ1ter[Jel1ing) witness : time to regulate conflicts of interes t. Med Sci Law
(GMC v. Meadow) . In November 1999 , Sally Cl ark was tried 1999 : 39 :1 1- 16.
16 Ma narin B. Assess in g the expen:: a ca ll for reciprocal
for the murder of her two sons. Th e Crown relied. in palt,
di sc lo sure in Canada. Med Sci Law 1999 ; 39: 17 -22.
upon Professor Meadow's evidence to refute the proposi
ti on that Mrs Cl ark 's children may have died fro m sudden
infant dea th syndrome (SIDS), or co t death. Professor Sir
Roy Meadow ventured into the field of statistics in givin g Further Reading
hi s opinion on the likelihood of occurrence of a second cot
dea th in the same family; this invo lvement, sa id to have CASES
been erroneous and outs ide his field of expertise. resulted • Nariollal Justice Campania Nau iera SA V. Prudelltial Assurance
in proceedings before the fi tn ess to practise panel (FPP) of Co. Ltd, rlre Ikar iall Reefer (1993) 2. L1 0yds Reports 68, 8 1-82
the GMC, with Professor Mead ow being found gUilty of and (19 95) 1. L1 0yds Reports 45 5.
serious professional misconduct and res ultin g in his era • R V. Srockwell (1993) 97 CR App R 260.
• R v. Uljoh Et Orhers (2001) EWCA Crim 1381.
sure from the medical regis ter. Professor Meadow appealed • Re H an d E (Minors) CO Ult of Ap pea l (1996), 21 May.
to the High Court, whe re his appea l was a llo wed by Collins • R V. BO II,\'rllOli (1984) 38 SASR 45.
J. The finding of serious professional misconduct was • flye V. United Srates. 54 App. D.D. 46, 293 F. 1013 . 1923.
qu ashed . Additionally, the judge commented that expel1 • Dauberr v. iVlelTeli Do /U Phollllll ceu ricilis Inc (1993)
113 S Ct 2786.
witnesses, by their status, enjoyed not merely immunity
• KUlllbo Tire Co Lrd V. Carmichael (1999) 526 US 13 7.
from civil suit, but also a wide, albei t not absolute, immu • R v. Robb 93 Cr App R 161.
nity from disciplinary, regulatory or fitness to practi se pro • Re AB (cliild abuse; expert lVili/esses) (1995) 1 FLR 181.
ceed ings in relation to statements or ev id ence given by • Daniels v. Walker (2 000) 1 WlR 1382, CA.
them in or for the purpose of legal proceedings. These com • A ssess or fo r Lothian Region v. Wilso11 (1979) SC 34 1 lord
Avonside at 349.
ments resulted in the Attorney General's interventi on to
• Rollalld v. Lothian Health Board, 27th August 1981 - unreported.
the COUlt of Appea l; it held unanimously that the immu • McVey v. Central Regiona l Co ull cil, 4th Ma rch 19 93 - unreported.
nity identified by the judge did not exist. • White/lOuse v. Jordan (198 1) 1 WlR 246.
• Williams V. Florida (1970) 399, US 78.
• Davie V. Magistrates of Edillburgh (1953) SC 34 at 40.
• Karling v. Purdue (2004) SLT 1067 Coun of Session, Edinburgh.
REFERENCES • Tire Gelleral iVIed ical Council V. Professor Sir Roy Meadow (Her
Majes t."·s Artomey General ilitemel1iIJg) (GMC v. Meadow)
Di ckson WG. A Treiltise all the Law of Eliidel1ce ill Scorlaml, Neutral Orar ion Nu mb er: (2006) EWCA Civ 13 90 Case No:
3rd edn. Edinburgh, UK: Scottish Law Commiss ion, No. 1887, CO/5763/2005.
137 .4, paras 397 el sel].
I
I
APPENDIXA
CHILD PROTECTION
EXAMINATION FORMS
CONSENT FORM
Child's name ..... .... ......... ... .. ............ ... .. .. ..... ..... ....... ........ .... .. ...... .. ...... .... ..... .. ...... .... .OOB ...... .. ... .... .. ...... .. ........... .
Address ... ............ ... ... ......................... .. .......... ... ... ......... ... ........... .. ................ ............ ... ... ....... .. ... ............... ...... ..... .
Unit no ..... ..... .... .... .. ..... ...... ... ...... ...... .. .... ......... ....... .. .. ... ....... ......... .. .... ..... .. .............. ... ....... .. ...... .. ........... .. .......... .
Permission must be obtained from parent(s) or other(s) with responsibility for the child, and from the child where
appropriate.
Photographs and CDs will be stored as part of the clinical records. They may be used to support clinical evidence
of injury and may need to be shared with another Doctor involved in any court proceedings.
I give permission for photographs and CDs to be used to support clinical evidence in court proceedings.
Yes No N/A
Photographs and CD recordings can be used for teaching and training of other professionals working in Child
Protection proceedings. Photographs and CDs used for this purpose are anonymized.
I give permission for anonymized photographs and CD recordings of my child to be used for teaching
and training of other professionals. Yes No N/A
I understand that the information from the medical examination will be shared with:
The procedure has been fully explained to me and I understand that I have the right to withdraw my con
sent at any stage during the procedure.
Signed .......... ........ ....... ..... ........ ....... ..... .. ... ...... ....... ..... .. .. ........ .. ...... .. ... ... ............ Date ........ ... ...... .. ... .. .. .. .. ......... ... ..
Name ............. ...... .... ................ ..... ..... ..... .... ....... ...... ... ....... .. .... .... ....... ..... .......... ..Parent/carer/professional
Examining doctor(s) ... .... ........ .. ..... .. ......... .. .............................................. .. .. ......... ... ............ .. ... .. ...... ... ............. .... ..
Signature .... ....... ..... ....... ... ............ ... ......... ...... .. ......... .... .... ...... ............... ...... ....... Oate .......... .... ....................... .... .. .
Known as
Sex
CHI no.
GP Date of examination
Address
Time of examination
BACKGROUND INFORMATION
(e.g. previous concerns reo developmental delay, poor growth, possible episodes of NAI)
Family/social history
N
o
o
N
<D
.c
o
U
o
;?
~--------------------------------------------------------------------~
Page 1 of 12 1471
Appe ndix A: Child protection exam ination forms I 459
Scrotum
CONSENT
(source, i.e . parent, young person, person holding parental righ ts)
Date: N
o
o
N
Witnessed by: Q)
.D
.8
Name Position u
o
L _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _~~
o
Page 2 of 12 1471
460 I Append ix A: Child protect ion exa mination forms
PRESENTING SYMPTOMS
SYSTEMATIC ENQUIRY
N
o
o
N
Q)
.0
(Genitourinary enquiry) o
t5
see over o
~--------------------------------------------------------------------------------~~
Page 3 of 12 1471
Appendix A: Child protection examination forms I 461
GASTROINTESTINAL/GENITOURINARY ENQUIRY
Comment
Bowels Frequency
Soiling Yes/No
Mucus Yes/No
Other Yes/No
Frequency Yes/No
Dysuria Yes/No
Urgency Yes/No
Other Yes/No
Di scharge Yes/No
Itch Yes/No
Rash/inflammation Yes/No
Circumcised Yes/No
-
Menarche age
Menstrual cycle
N
o
o
N
Q;
.0
o
ti
o
o
~
Page 4 of 12 1471
462 I Appendix A: Child protection examination forms
CLINICAL EXAMINATION
General physical appearance of child (note especially any evidence of infection, neglect or injury)
Measurements
Weight kg centile
Height cm centile
(b) Teeth
(c) Eyes
(e)
- Cardiovascular system
(i) Genitalia/testes
(k) Locomotion/posture Ig
(I) Other I~
Q;
.c
0
0
0
0
~
Page 5 of 12 1471
--~.
Appendix A: Child protection examination forms I 463
.....-- ,,----
®' I@
~
l.\
.....c:::::'..: "4
,---,
C\l
o
o
C\l
Q;
.0
-----------------------------------p~~~~~--------------------------------~~
...
Page 6 of 12 1471 .
464 I Appendix A: Child protection examination forms
(\J
o
o(\J
Qj
.0
o
o
o
o
~--------------------------------------------------------------------~
Page 7 of 12 1471
-----
- -- - -------------------------
Appen dix A: Ch ild prot ecti on examinati on forms I 465
-
'"-'
N
o
o
N
Q;
.0
o
t5
o
~------------------------------------------------------------------------------~~
Page 8 of 12 1471
466 I Appendix A: Child protection examination forms
Mons pubis
Prepuce of clitoris
Clitoris
Frenulum of clitoris
Orifice of urethra
\ Labium majus (labia majora)
\
I
I
Vestibule (area between
labia minora)
Fourchette
Posterior commissure
Genital examination
Position
Method
Magnification __________
Video Yes/No
Identification:
Video no. I
Serial no.
.'.~:- '
J
C\J
o
o
C\J
OJ
.D
o
tl
o
o
Page 9 of 12 1471
-- - ---
Appendix A: Chi ld protection examination forms I 467
u\
Genital examination
Position
Method
Magnification
Video Yes/No
Identification:
Video no.
Serial no.
Examination of anus
Position
Method
Magnification
u
Video Yes/No
Identification:
Video no.
Serial no.
(\J
o
o
(\J
Q;
.D
.2
u
o
o
~--------------------------------------------------------------------------------~~
Page 10 of 12 1471
468 I Appendi x A: Child protection examin ation forms
INVESTIGATIONS
D Photography
Taken by:
Police:
Processed at:
Medical staff:
Identifier:
D Radiographs
D
Coagulation profile
Routine specialist:
D Serum HIV/hepatitis
D Swabs
Site:
D Urine
D
Pregnancy test
D Other
CONCLUSION/OPINION
TREATMENT/ACTION REQUIRED
C\J
o
o
C\J
Q;
.D
o
t5
o
o
~------------------------------------------------------------------------------~~
Page 11 of 12 1471
Appendix A: Child protection examination forms I 469
INFORMATION SHARING
AGREED PLAN
appropriate
5. Discuss with Procurator Fiscal
6. Contact GP
7. Other (specify)
as
Signed Date
Signed Date
AUDIT
(\J
o
o
(\J
Q)
.a
o
tl
o
o
~--------------------------------------------------------------------------------~~
Page 12 of 12 1471
Table 1 Fetal autopsy standards at 72-27 weeks of gestation'
Age (GW) FL CR CH HOC Body Brain Liver Lungs Heart Thymus Spleen Kidneys Adrenals
Maceration" 0-3 0-3 0-3 0-3 0-3 0-3 0-1 2 3 0-1 2-3 0-3 0-1 2 3 0-1 2-3 0-1 2-3 0-1 2-3
(mm) (em) (em) (em) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g)
~
12 Mean 9 7.4 9.8 7. 1 29.6 4.8 1.5 1.4 1.3 0.6 0.9 0. 10 0.D3 0.01 0 .25 0.19 0.04 0.11
SO 3 7.7 7.7 7.7 74.9 7.4 7.2 7.2 7.2 0.9 0.9 0. 14 0.06 0.02 0.15 0.75 0.78 0.18
13 Mean 12 8.7 11.8 8.5 37.4 6.5 2.0 1.7 1.2
to
SO
1.7 1.2 0.20 0.04 0.02 0.08 0.3 0.2 0.17 0.17
r
3 7.2 7.8 7.2 74.9 7.4 7.2 7.2 7.2 0.9 0.9 014 006 003 0.03 01 0.1 0.18 0.18 m
14 Mean 15 9.9 13.7 9.8 53.0 9.1 2.9 2.4 2.3 2.0 1.5 0.3 0.05 007 0.05 0.Q4 0.14 0.4 0.3 0.3 0.2 Vl
15
SO
Mean
3
18
7.2
11 .1
7.8
15.6
7.2
11 .1
74.9
76.5
2.5
12.7
7.2
4.2
7.2
3.3
7.2
3.2
0.9
2.9
0.9
2.1
0.1
0.5
0.06
0.07
0.06
0.08
0.06
0.06
0.04
0.06
0.04
0.17
0.7
0.6
0.7
0.5
02
0.5
0.2
0.3
o
'"T"j
SO 3 7.2 7.8 7.2 78.5 3.9 7.2 7.2 7.2 0.9 0.9 07 0.06 0.06 0.06 0.06 0.06 0.3 0.3 02 0.2 Vl
16 Mean
~~
21 12.4 17.5 12.4 108 17.3 5.9 4.5 4.2 3.9 2.7 0.8 0.11 0. 12 0.09 0.09 0.17 0.9 0.8 0.6 0.4
SO 3 7.3 7.8 7.3 47 5.4 7.5 7.5 7.5 7.2 1.2 0.2 0.06 0.06 0.06 0.08 0.08 0.4 04 0.3 0.3
17 Mean 24 13.5 19.3 13.6 147 22.9 8. 1 6.1 5.4 5.1 3.5 1.0 0.18 0.18 0.12 0.13 0.16 1.3 1.1 0.8 0.5
SO 3 7.3 7.9 7.3 53 6.9 3.0 3.0 tJ~
18 Mean 27 14.7 21.1 14.8 194 29.4 10.7 7.9
3.0
6.8
7.7
6.4
1.7
4.4
0.4
1.4
0.06
0.3
0.06
0.3
0.06
0.2
0.12
0. 19
0.72
0.15
06
1.8
06
1.5
04
1.0
04
0.7 »m
19
SO
Mean
SO
3
30
3
7.3
15.9
7.3
7.9
22.9
7.9
7.3
16.0
7.3
65
249
78
8.4
37.0
9.8
4.5
13.8
6.0
4.5
10. 1
6.0
4.5
8.4
6.0
2.3
7.9
2.8
2.3
5.4
2.8
0.5
1.7
0.7
0.2
0.4
0.3
0.2
0.4
0.3
02
0.3
0.3
0.17
0.3
0.2
0.17
0.15
0.22
0.8
2.4
1.0
0.8
2.0
1.0
0.4
1.2
0.5
0.4
0.8
05
b6 I ---'
20 Mean 33 17.0 24.6 17.2 312 45.5 17.2 12.5 10.2 9.5 6.5 2.1 0.6 0.5 0.3 0.4 0.17 3.0 2.5 1.4 1.0 ~><
SO 3 7.4 7.9 7.4 92 77.3 7.5 7.5 7.5 3.4 mto
»
3.4 08 0.4 0.4 04 0.3 0.29 1.2 1.2 0.6 0.6
21 Mean 36 18.2 26.3 18.3 382 55.0 21.1 15.2 12.3 11 .2 7.8 2.6 0.8 0.7 0.4 0.5 0.22 3.8 3.1 1.7 1.2
SO Vl
22 Mean
3
39
7.4
19.3
20
28.0
7.4
19.4
707
461
728
65.4
9.0
25.5
9.0
18.2
9.0
14.5
4.0
13.1
4.0
9.2
1.0
3.1
0.5
1.0
0.5
0.9
0.5
0.6
0.4
0.7
0.36
0.3
7.4
4.6
7.4
3.8
0.7
1.9
0.7
1.4 c
23
SO
Mean
3
41
7.4
20.4
2.0
29.6
7.4
20.5
722
547
74.3
76.9
70.4
30.2
70.4
21.6
70.4
16.9
4.6
15. 1
4.6
10.7
1.7
3.6
0.6
1.3
0.6
1.1
0.6
0.7
0.4
0.9
0.4
0.4
1.6
5.5
1.6
4.6
08
2.2
08
1.6
~
SO 4 7.5 2.0 7.4 722 75.8 77.9 77.9 77.9 5.3 5.3 1.3 0.8 0.8 08 0.5 0.5 7.9 1.9 08 0.8 ~
24 Mean 44 21.5 31.2 21.6 641 89.3 35.4 25.2 19.5 17.3 12.4 4.2 1.6 1.3 0.8 1.1 0.6 6.5 5.5 2.5 1.8
m
SO 4 7.5 2.0 7.5 737 77. 2 73.4 73.4 73.4 5.9 5.9 7.4 09 0.9 0.9 0.6 0.6 2.1 2.1 09 09 Z
25 Mean 47 22.6 32.8 22.6 743 103 41.1 29.1 22.3 19.6 14.1 4.9 1.9 1.6 1.0 1.4 0.8 7.6 6.4 2.8 2.0 --l
Vl
SO 4 7.5 2.7 7.5 754 79 74.9 74.9 74.9 6.6 6.6 1.6 1.1 1.1 1.1 0.7 0.7 2.4 2.4 7.0 1.0
26 Mean 50 23.6 34.3 23.6 853 11 7 47.1 33.4 25.3 22.0 16.0 5.6 2.3 1.9 1.2 1.7 1.1 8.8 7.4 3.1 2.3
SO 4 7.5 2.7 7.5 77l 20 76.4 76.4 76.4 7. 3 7.3 1.7 7.2 7.2 7.2 0.9 0.9 2.7 2.7 1.1 1.1
27 Mean 52 24.7 35.8 24.5 971 133 53.6 37.9 28.6 24.6 18.0 6.3 2.6 2.2 1.4 2. 1 1.4 10.1 8.4 3.4 2.5
SO 4 7.6 2.7 7.5 788 22 77.9 77.9 77.9 8.0 8.0 1.8 1.4 1.4 1.4 1.0 1.0 3.0 3.0 1.2 7.2
CH, crown-heel lengt h; CR, crow n-rump length ; FL, foot length ; GW, weeks of gestation; HOC, head circumference; SO, standard devi ation.
"0, none; 1, mild; 2, moderate; 3, marked.
""
-.j
I\)
Maceration' 0-3 0-3 0-3 0-3 0-3 0-3 0-1 2 3 0-1 2-3 0-3 0-1 2 3 0-1 2-3 0-1 2-3 0-1 2-3 x'
rn
(mm) (em) (em) (em) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g) (g)
n;i
IT
C1l
(fl
28 Mean 55 25.7 37.3 25.5 1096 149 60.6 42.7 32.0 27.4 20.2 7.1 3.1 2.5 1.6 2.5 1.8 11.4 9.6 3.7 2.8 S
en
SO 4 1.6 2.2 1.6 206 23 19.3 19.3 19.3 8.7 8.7 2.0 1.6 1.6 1.6 1.1 1.1 3.3 3.3 1.3 1.3 ill
:::J
29 Mean 57 26.7 38.7 26.4 1230 166 67.9 47.8 35.6 30.2 22.5 7.9 3.5 2.9 1.8 3.0 2.2 12.9 10.8 4.1 3.1 Q
Cll
SO 4 1.6 2.2 1.6 225 25 20.8 20.8 20.8 9.5 9.5 2.1 1.8 7.8 1.8 1.3 1.3 3.6 3.6 1.4 1.4 0.
30 Mean 60 27.7 40.1 27.2 1371 185 75.7 53.3 39.4 33.2 24.9 8.7 4.0 3.3 2.1 3.6 2.7 14.4 12.1 4.5 3.4 :3
C1l
Cll
SO 4 1.6 2.2 7.6 244 26 22.3 22.3 22.3 10.2 10.2 2.3 2.1 2.1 2.1 1.4 1.4 3.9 3.9 1.4 1.4 (fl
C
31 Mean 62 28.7 41.4 28.1 1520 204 83.9 59.0 43.4 36.3 27.4 9.6 4.5 3.7 2.3 4.2 3.3 16.0 13.4 4.8 3.8 CD
:3
SO 4 1.7 2.2 1.7 264 28 23.8 23.8 23.8 lJ.0 lJ.0 2.4 2.3 2.3 2.3 1.6 1.6 4.3 4.3 1.5 1.5 C1l
:::J
32 Mean 64 29.7 42.8 28.9 1677 224 92.6 65.0 47.6 39.6 30.0 10.6 5.0 4.2 2.6 4.8 3.9 17.7 14.9 5.2 4.1 u;
SO 4 1.7 2.3 1.7 285 29 25.3 25.3 25.3 lJ.8 ll.8 2.6 2.5 2.5 2.5 1.8 1.8 4.6 4.6 1.6 1.6
33 Mean 67 30.6 44.0 29.7 1842 245 102 71.3 52.1 43.0 32.8 11.6 5.6 4.6 2.9 5.5 4.5 19.5 16.4 5.6 4.5
SO 4 1.7 2.3 1.7 306 31 27 26.7 26.7 12.6 12.6 2.7 2.8 2.8 2.8 1.9 7.9 5.0 5.0 1.7 1.7
34 Mean 69 31.6 45.3 30.5 2015 268 111 77.9 56.7 46.6 35.7 12.6 6.2 5.1 3.2 6.3 5.2 21.4 18.0 6.0 4.8
SO 4 1.8 2.3 1.7 328 32 28 28.2 28.2 13.5 13.5 2.9 3.1 3.1 3.1 2.1 2.1 5.4 5.4 1.8 1.8
35 Mean 71 32.5 46.5 31.2 2195 291 121 84.8 61.5 50.3 38.7 13.7 6.9 5.7 3.5 7.2 6.0 23.3 19.6 6.5 5.2
SO 5 1.8 2.3 1.8 350 33 30 29.7 29.7 14.3 14.3 3.0 3.3 3.3 3.3 2.3 2.3 5.8 5.8 1.9 1.9
36 Mean 73 33.4 47.7 31.9 2383 315 132 92.1 66.5 54.1 41.9 14.8 7.5 6.2 3.8 8.1 6.7 25.4 21.4 6.9 5.6
SO 5 1.8 2.4 1.8 373 35 31 31.2 37.2 15.2 15.2 3.2 3.6 3.6 3.6 2.5 2.5 6.2 6.2 2.0 2.0
37 Mean 76 34.3 48.9 32.6 2580 340 142 100 71.7 58.1 45.1 16.0 8.2 6.8 4.2 9.1 7.5 27.5 23.2 7.4 6.0
SO 5 1.8 2.4 7.8 397 36 33 33 32.7 16.1 16.1 3.3 3.9 3.9 3.9 2.7 2.7 6.6 6.6 2.1 2.1
38 Mean 78 35.2 50.0 33.2 2784 366 154 107 77.2 62.2 48.5 17.2 8.9 7.4 3.9 10.1 8.3 29.8 25.0 7.8 6.5
SO 5 1.9 2.4 7.8 421 38 34 34 34.2 17.0 17.0 3.4 4.2 4.2 4.2 3.0 3.0 7.1 7.J 2.2 2.2
39 Mean 80 36.1 51.1 33.8 2996 394 165 116 82.8 66.5 52.1 18.5 9.7 8.0 5.0 11.2 9.1 32.1 27.0 8.3 6.9
SO 5 1.9 2.4 7.9 446 39 36 36 35.6 18.0 18.0 3.6 4.6 4.6 4.6 3.2 3.2 7.5 7.5 2.3 2.3
40 Mean 82 37.0 52.1 34.4 3215 422 177 124 88.6 70.9 55.7 19.8 10.5 8.6 5.4 12.4 9.9 34.5 29.0 8.8 7.4
SO 5 1.9 2.5 7.9 471 41 37 37 37.1 18.9 18.9 3.7 4.9 4.9 4.9 3.4 3.4 8.0 8.0 2.4 2.4
41 Mean 84 37.8 53.1 35.0 3443 451 190 133 94.6 75.4 59.5 21.2 11.3 9.3 5.8 13.7 10.7 37.0 31.1 9.3 7.9
SO 5 7.9 2.5 1.9 497 42 39 39 38.6 19.9 19.9 3.9 5.3 5.3 5.3 3.7 3.7 8.4 8.4 2.5 2.5
42 Mean 86 38.6 54.1 35.5 3678 481 203 142 101 80.1 63.4 22.5 12.2 10.0 6.2 15.0 11.5 39.6 33.3 9.9 8.4
SO 5 2.0 2.5 2.0 524 44 40 40 40 20.9 20.9 4.0 5.6 5.6 5.6 4.0 4.0 8.9 8.9 2.6 2.6
43 Mean 88 39.4 55.0 36.0 3922 512 216 151 107 84.9 67.4 24.0 13.1 10.7 6.6 16.4 12.2 42.2 35.5 10.4 8.9
SO 5 2.0 2.5 2.0 551 45 42 42 42 27.9 21.9 4.2 6.0 6.0 6.0 4.2 4.2 9.4 9.4 2.7 2.7
CH, crown-heel length; CR, crown-rump length; FL, foot length; GW, weeks of gestation; HDC, head circumference; SD, standard deviation.
'0, none; 1, mild; 2, moderate; 3, marked.
Appendix B: Tables of sta ndard measu rements I 473
700,-----r----,-----,----,-----r----.----~----r_-- __----~----~
90th percentile
600~----r----+----~----+---~~---+----~----~--_4~~~~--~ 75th percentile
Mean singleton
500 ~----r----+----~----+---~-----+----~~_=~--~~~~~--~ 25th percentile
10th percentile
200~--~~~~~~~~~q-----~---+----~----+_--~----_+-----
100~----+----1----~----+-----~---+----~----+_--~----~----~
22 24 26 28 30 32 34 36 38 40
Ges tat ion [weeks)
Table 5 Body weight, height and head circumference, boys, birth to 1 year. (Courtesy of the Child Growth
Foundation. Printed by Harlow Printing Ltd, Maxwell St, South Shields, NE33 4PU.j
I 51
NAME ..... ..................... ................ ..
98 th -
.1SI -1
50
49
Correct by......... ..... for prematurity , I
---l1:=.: -
:~: -, :~
46
45 u
. 3'2 34 3,6 3'8
. 1
2
J
6
l'
)-
I 1.-- -1
. I
I· I .•
.' I
I
f i'
( I ;
~.
'~_ 2:: -I'
2nd -
46
45
:; 111
41 pre-term ,
~/ I ;--<~1 ::r:l> --.
11 j / .
, ; . -......1"
< . "
40
39 II 1~ . ·
'/l /!.y m '
" f/1)1-~' I
I
99 .• ' b
84
82
38 I / / t IYI/ I II, "I _1~1~1--' 9;1h
:: t~ /, 1 /{1 ;1/ :i ::1-:1 ~;:~ ~ ;~:~ ::
I I I !I : I .-1---1
80
:: IVt)jfl ./r~,~ rl~: ' ' Iii J J.3 1:1 ~ -i:~~f __ ~~;:
~ L1- -~ - ___~-
32
31
30
I' "
I'
t
j/! / I :,
/1 /'
. /V'1/ ' I
'I
I t;1.<
, , '
;.-- ~ ""......,
1/1; t.-V: .: : ; . _> >-_:'--/--1-= - .-
1 t /-:--1-::'//1: ,'''',::_ --,=1--'-' -
. !.-' --I -J -] _' J --
. ' ---- . --1- --1- 0.4," 68
66
64
29
28
27
11 1 / I /t/
.(1
1/ Cf/'
f/. «( ~
I
j y l,.
-'
.
'
62
60
58
26 ' I, / / I , / f ()--,
5lt 1· / / -. 1/ '/ (" --
5~;S1'
):" ./ /~
25 1 14
24 I ;,--- , , I
-.--l 9 9.6 1b. 13
I
l4~ ' /1/
I
'1/ y
..(
j , 98 tb
12
~ 1;-( Slit·
:~ I
11
7Slh
10 50 t b - 10
~5th
9 9
9th
-~-
81' .- .
2 nd
8
71 -' O.4tb
_i ...-'
61 pre-term 6
5 5
3~ 32 3 4 36 38 40 42~4 46 + 8 50 52
8 9
28
D a Le Ag t.: ~ tcasllrem("n1 ~ ;J. m t·
3 *
:
2
:
~eeks /? months :
, 1
-'-~8·~'
~, I O 1~~8 ! :
.-
11 1'4 20 1 2 2'4 :
476 I Appendix B: Tables of standard measurements
Table 6 Body weight and height, boys, 7-5 years. (Courtesy of the Child Growth Foundation. Printed by
Harlow Printing Ltd, Maxwell St, South Shields, Nf334PUj
I
,
1 2 2'/, 3 3'/, 4 4'/, 5
130
1-5yrs years
--l- 125
996th
120
98th
115
gist 115
75tb
11 0
50th - 110
105
25tb 105
Sib
100
2.d -- 100
95
-- a.4th
95
90
85
/'
80
---~
I
- 9 9,6th 26
./
75
.. /
25
70
./
'"
T '" 24
98th
65
I
L /~
./
23
60 I 22
~--
I "'" /"
9 l st
r
1... _ _ _._---+-_ _ _ _ _ _ _
--t 21
20 20
75t h
19 19
18 ~ - .-+ ---- 18
5 0th
17 25 th
17
16 16
9t h
15 15
2nd
14 , 14
OAth
13
12
11 11
10 10
-~---+- 7
6 f---.
years
-- -+
-......J
1 1'/, 2 2'/, 3
3'/, 4 4 ' /, 5
Table 7 Body weight ond height, boys, 5-78 yeors. (Courtesy of the Child Growth Foundation. Printed by Harlow
Printing Ltd, Maxwell St, South Shields, NE33 4PU.;
20 5 2 05
5 6 7 8 9 10 11 12 13 14 15 16 17
200 - - - + - - 200
1
195
1 75
h / /' . .. 5 0th 17 5
/- : 1/ / 25 . h
170 77 170
16 5 4/ . ~-
9th
2n d
165
/ /
160
/
+7
v'
....--::i - -
./ r-. __ 0.4'h --'
160
/
155 .L-/ / /-I~ 155
/ /' / ;- /,V
150 L--/' /- 7'
,( / // / - y
/ /
145 /~ .4- /' 105
/ .-/, / 1-/
.L __/ --'
140 ~ .., 99.6' h - 100
/' /
/.-
/'
/
/ .-.
135 /t-/
././~ .
/ 4 _.- - + - - 95
/ ,/
,Lo/ ./ ./ /'
130 r- ~ 17' / 98th -- 90
/ / ./ ./
L/ /
125 85
/ /
120
L. /
/ '7
115
/ /
/ 75
75th
110 . 70
105 50t h
--- 65
100 25th 60
95 9t h ---- 55
90 2nd
50
O.4th
45
40 40
35 35
30 30
25 ---
.....--. 25
20 20
15 15
years
10 10
5 6 7 8 9 10 11 12 13 14 15 16 17
4 78 I Appendix B: Tables of standard measurements
Table 8 Body weight, height and head circumference, girls, birth to 7 year. (Courtesy of the Child
Growth Foundation. Printed by Harlow Printing Ltd, Maxwell St, South Shields, NE334PU.j
~8 h 3~
O-lyr 14 16
Q)
20
0
241' 28 3b 32
C!; 0 ®
36 38140 42 h4 46 ,f8 50 52
@ @ 6
50
91sl 48
3
50th
25th
- 47
46
45 · 45
9tl1
44
20d 44
43 , O.-1lh
43
42
42
41
40
39
pre-t e rm
.// 84
82
38
/ .'. 99 .6 1h 80
37
/ BSIII
78
36
/ 9Ist
75th
'76
35
74
50th
34 /
2Gth 72
,
33 /
1 Olll
70
2 tl d
32 / O.4t h
68
31 f 6
30 ,
29 /
f
I·
64
62
28 / ; 60
27
58
26
25
24 52
/ 56
54
52
50
48
13
46/
9fU;:h
.;.4
12
42
9~l h
4()
11
!J ls.t
10
75a; 10
50tb
9
9
25 th
~){ h
8
8
2nl!
7
O.·l lh
7
6 pre-term 6
4
/
/ 85052
,/
3
,. / 3
r
,
;'
r
2
/ 2
t
i
weeks / j> m onths
~ 8~ 12~14 ~8
1
32 34 36 38 EO D 2 6
10 16 20 l2 2 4 t-,- ---+---I--+-----+
o ___ T o o' ____ __
Tabl e 9 Body weight and height, girls 7-5 years. (Courtesy of the Child Growth Foundation. Printed by
Harlo w Printing Ltd, Maxwell St, South Shields, NE33 4PU.j
, ,
1 2 2 '/, 3 3'/, 4 4 '/, 5
1-
1-5yrs years
99.6 tb 120
25
105 28 th
105
9t h
2nd 100
OAth
95
90
27
9 O.6t h
26
25
98tb 24
60 23
55 22
9Is t
21
20 20
75t h
19 19
18 50t h 18
17 .-'/
17
16 ----- ..... --- 2:,l h
~
16
9 th
15 15
14 2nd
14
13 OA t h
13
12 12
11 11
10 10
,.. ,..
9 9
8
-- ;;-
8
7 7
6 6
years *Measurement: H i e =H c~ d Circtlmfcrt'llc('. L = LCIl!!t1 l. W = W cl~ hl. H = liCl,~il!
1 1 '/,
2 2 '/, 3 3 '/, 4 4 '/, 5
480 I Appendix B: Tables of standard measurements
Tabl e 10 Body weight ond height, girls, 5-78 yeors. (Courtesy of the Child Growth Foundotion. Printed by Harlow
Printing Ltd, MaxwellSt, South Shields, NE334PU.j
5
6 7 8 9 10 11 12 13 14 15 16 17
190
. 190
I
,
185
5-1Syrs
With provision for school
-; 185
180
f reception class 99 .6th .=, 180
175
NAME .. ...... ... ... ... ...... ... ... .. ... .. . _ --' 98th
175
165
-- __ ~ g lst _
76th
170
165
_ - - - -- 50th _
160
25th 160
9t h
155
155
/ /
2nd
150
150
/ /
- O.4th
145
145
/'
/'
140
/ .7
/ /
135
130
95
125
L 90
/'
_99.6 t l>
120
. 85
115
80
98th
110
75
105
..-:._ 91st - 70
/'
100
,; 65
75 th
95
/ 60
50t h
90
55
/ 50
9th
,;
2nd _ 45
"
40
_ 0 .4th
40
./
35
35
/' Y
/' /' ,,/,,/
30
30
.-----.-
/' /' /"
/' //' /
~ -- ------...- --
25
/ // ,.'
25
~
~ ~ .' ~-
• / __ - - - - - - -- _lo- -
20 - 20
. ..' . .....
~ -~--::-
15 :.:.:--:--:::- -::: .5. - . . 15
-~-
,...
years
10
10
5 6 7 8 9 10 11 12 13 14 15 16 17
~ ---
Appendix B: Tables of standard measurements I 481
Newborn
Brain 13 325
Heart 16 18
Right lung 12 29
Left lung 12 24
Liver 16 109
Spleen 16 8
Right kidney 14 11
Left kidney 14 11
Thymus 16 11
1 month
Brain 69 489
Heart 65 27
Right lung 63 55
Left lung 63 46
Liver 74 176
Spleen 74 16
Ri ght kidney 74 17
Left kidney 74 17
Thymus 68 21
2 months
Brain 102 569
Heart 100 31
Right lung 94 61
Left lung 94 52
Liver 102 193
Spleen 103 17
Right kidney 104 19
Left kidney 104 19
Thymus 90 28
3 months
Brain 81 651
Heart 80 33
Right lung 83 67
Left lu ng 83 57
Liver 85 223
Spleen 85 19
Right kidney 84 21
Left kid ney 83 22
Thym us 76 31
4 months
Brain 44 691
Heart 40 33
Right lung 37 73
Left lung 37 60
Li ver 44 232
Spleen 44 22
Right kidney 43 21
Left kid ney 43 21
Thym us 38 29
(Continued)
482 I Appendix B: Tables of standard measurements
Tabl e 11 (Continued)
5 months
Bra in 26 682
Heart 24 34
Right lung 23 69
Left lung 23 56
Liver 25 246
Spleen 26 22
Right kidney 26 23
Left kidney 26 26
Thymus 22 25
6 months
Bra in 17 766
Heart 15 37
Right lung 16 72
Left lung 16 61
Liver 15 280
Spleen 17 24
Right kidney 18 22
Left kidney 18 22
Thymus 16 32
7 months
Brain 12 763
Heart 11 37
Right lung 10 70
Left lung 10 57
Liver 12 285
Spleen 12 26
Rig ht kidn ey 12 24
Left kidney 12 27
Thymus 9 32
8 months
Brain 8 852
Heart 8 45
Rig ht lung 7 74
Left lung 7 63
Liver 8 340
Spleen 8 31
Right kidney 8 30
Left kidney 8 29
Thymus 6 32
9 month s
Bra in 12 925
Heart 12 45
Right lung 11 87
Left lung 11 79
Liver 12 342
spleen 12 45
Right kidney 12 28
Left kidney 12 28
Thymus 10 26
(Continued)
Appendix B: Tables of standard measurements I 483
Table 11 (Continued)
10 months
Brain 10 986
Heart 10 47
Right lung 9 89
Left lung 9 76
Liver 10 369
Spleen 9 36
Right kidney 10 30
Left kidney 10 35
Thymus 8 28
11 months
Brain 8 935
Heart 9 52
Right lung 8 104
Left lung 8 104
Liver 9 388
Spleen 9 39
Right kidney 9 35
Left kidney 9 35
Th ymus 9 33
12 months
Brain 3 920
Heart 3 51
Right lung 3 95
Left lung 3 86
Liver 3 405
Spleen 3 42
Right kidney 2 30
Left kidney 2 33
Thymus 2 16
Oa ta from ref. 4.
Sudden infant death syndrome (SIOS) and non-SIOS aggregated; no significant difference found between
M F M F M F M F M F M F
F, female; M, male.
--
I INDEX I
Note: For purposes of conciseness, the terms 'abuse' and 'non-accidental' imply suspected and/or actual abuse. Figures and tables
are comprehensively referred to from th e text. Therefore, significant material in figures and tables have only been given a page
reference in the absence of their concomitant mention in the text referring to that figure. Abbreviations: SIDS, sudden infant
death syndrome; SUD], sudden unexpected death in infancy.
identifying 413
ac ute metabolic autopsy 119-20
infections 240, 241
death
acyl-CoA deficiency )16
post-mortem exa mi na tio n, live
penetrating 192
medium chain see medium chain
see also inhalation; respirato ry tract
160
short chain 244
alcohol (ethanol)
abortion
Addison's disease 113, 128
fetal exposure to 267,274
definition 181
adnexa, ocular, non-accidental injury
fire deaths relating to 318, 321
induced/deliberate 181
125-8
immersion victims, testing 347
abortuses 141
drug ab use 256-7 , 341-2
road traffic accident
abuse
testing 272 -3
testing driver 388
death due to
fetal weight 451
Alport's syndrome 95
epidemiology 336-7
lipid dep letion and SIDS 212
in immersion 350
prevention 341
dental estim ation 441-3
am niotic fluid toxicology 267
proneness 33 7-8
air bags 386
beta-amyloid precursor protein
accidents airways
anaemia, aplastic 99
acromion fracture 59
imposed 332-3
anaphylaxis 109-10, 239-40
(aPTI/PTIK) 80,81
peripheral, resistance, increase with
ana to mical variants (normal)
abnormalities 89
upper 240-1
skeletal 70
486 I Index
X-linked 235
abuse see child abuse
immersion victims 347
animal bites 42 7- 8
asthma 342
infants/young children 149-50
anticoagulants 98
146
concealment of delivery or 181, 182,
243
249
live, determination 183-7
antiplasmin deficiency 93
attention deficit hyperactivity disorder
mechanisms of brain injury 304
a ntiplatel et drugs 99
(ADHD), methylphenidate in
sp inal cord 176,302,312
11-12,36-7,39-40
autopsy see post-mortem examinatio n 188-90
girls 26
nerve root 176
SUD Tand 218
in sum 2 17
babies see infan ts and babies
detail of 429-30
aplastic anaemia 99
sudden death (infancy/early
pattern/distribution 18, 425-6
167
bites 428
comparison evidence from bite and
246
basal gangl ia, hypoxic-ischaemic
231
132
bath tub and spa pool drownings
bruising 76-7
wi th 232, 365
accidental 353-4
evaluation/m easurements 82-94
in hospital 365
battered child syndrome see physical fundus haemorrhages 132
immersion 350
Bayley's Motor Scale 17
see also blood, loss; haemo rrh age
arrhythmoge nic right ventri cu lar bedding and SUDl/STD S 140, 203
blood
cardiomyopathy 229- 30
bed-sharing see co-sleeping
clotting see coagulation
~ ~r:.
- --
- - -
Index I 487
measurements (inc!.
genetic factors influencing
capilialY reactions, bra in 168
anthropometlY) 187
recovery 312
caput succedan eum 302
150
damage 172
carbon monoxide
bone breastfeeding
carboxyhaemoglobin 320, 322-3
facial, fetal hei ght estimation from infant poiso nin g 257
w ith 233
444
breech presentation 188
cardiomyop athy, su dden death 155,
injury
brittle bone disease
156, 228, 228-9, 229 -30
70,71
imperfecta
cardiopulmonary resuscitation see
99-100
bruises
perpetrating abuse or death 274,
boys
assessment for 17 - 18, 19
carnitine palmitoyltransferase type 11
technique 27
infants 214-15
cell death, brain (in early life) 167
brain
observed at scene 140
central venous catheter-related deaths
neurological lesions
bucket-handle fracture 53 , 54, 57
cephalic presentation 188
herniations 311
bullous impetigo 20
cepha1(o)haematoma 218, 301
infarction 310-11
pathological changes 324-5
middle (MCA), in head injury
malformations 169-71,237
skin conditions simulating 20
occlusion 290
findings 156-7
calcification, infantile coronary 232
29 1, 29 2, 29 2-3
neonatal 160
callosal agenesis 170
cerebral function an alys ing monitor,
SlDS 211
callus formation 52
head injury 292
weight 484
Camino fibreoptic pressure monitoring
sudden death 236-7
fetal 471
291
cerebral oedema see oedema
488 I Index
neurotrauma 309
351,352- 3 deficien cies 80, 83, 84,90, 92 - 3,
monitoring 291
see aIso filicid e; homicide; 96-7
cerebrovascular autoregulatory
Child Protection Orders 7
89
mechanisms 291
Child Protection Services 2
coca ine, environmental exposure 268,
311-12
chime rism 410
codeine, breastfeeding mother 257
injury) 393
chloride leve ls, post-mortem 110, III
video 7, 8, 11 , 27
hyperacute 295
liquid see liquid chromato graphy
computer-assis ted dental identification
non-accidental 326
see also X chromosome;
accidental injury 293
dissection 154
burns 326
conception, prod ucts of 141
neonatal 160
smoking see smoking
confi dentiali ty 8
traum a in abus e 68
Clauss fibrinogen assay 81, 82
confl ict of in terest, expert witnesses
alerting signs 2- 3
at birth 71 , 161,218 congenital infections
J-23
abuse syphilis 71
rol e 7
in hospital , road traffic accident 388
cyanotic 83 - 4, 101
definition 2
at scene of sudden death
drowning and 3 56
witnesses
coagulation (clotting) 79-80
in-hospital neonatal death 372
investigations 12
activation 100
upper airway 193, 240
medical conditions
pe tech ial haemo rrhages and 329
conj unctival petechiae, SUD! 213 -14
prevalence 2
101
conse nt (inci. informed .consent) in
types 3
inhibitors, tests showing 86, 87, 88,
abuse to examination 8, 457, 459
definition 181
tests 77
contusions
law 181
coagulation fa ctors 79-80
brain 294
Index I 489
birth-related 302
cyanide (fires) 320, 322, 323
degradation, drug, post-mortem 270-1
imaging 63
cyanotic congenital heart disease
dehydration, ante-mortem 110-11
copper deficiency 71
cyclists/bicyclists 388, 392
delivery
copy number
helmets 338-9, 392-3
concea lment of birth or 18], 182,
227,230-1,232-3,364
cystic fibrosis 397-8
unattended 182-3
embolism 364
cytochrome P450 (CYP) enzymes in
comparison technique 435, 435-49
damage 172
skull 60-1
14
204-5
'dead in bed' syndrome and insulin
depression, post-natal 352
cot 203
adequa te cause of, in SU D! 200-1
diabetes mellitus 112-13,243-4
140, 198
certifi cation in SIDS/SUDI 218-19
diaphragmatic hernia, co ngenital
bedside 204
head injury 283, 284-8
at birth 71
characteristics 202
heat-induced see fire fatalities;
dia to ms, immersion victims 347
syndrome
immersion see drowning
diffuse axonal injury, road tra ffi c
260
time of see time
identification 440
creatinine 114
see also cell death; ch ild-killing; dilute Russe ll's viper venom time 87
(1995) 7
body see body 152-6
crimping bruising 19
clothing see clothing disseminated intravascu lar coagulation
471-2
decomposition, electrolyte
diving reflex 349-50
autopsy 471-2
decongestants, nasal, affecting
DNA 396,397
490 I Index
persons) 411-13
therapeu tic use
295-6
mixed 406-7
see also toxic substances
endocrine disorders 112-14
documentatio nl records
haemorrhage
England, road traffic casual ties and
in abuse 16-17
in sudden death 157
fatalities 390
babies/infants/young children
spinal cord, in sudden death 158
enterocolitis, necrotizing 212
145-6
dyadic deaths (homicide-suicide)
entrapment asphyxia 330
145-6
dysrhythmias see arrhythmias sample contamination see
dental
contamination
post-mortem 438-40
ecchymoses 88, 89
sleeping
452-3
eczema (dermatitis), napkin 217
in fatty acid oxidation defects, liver
292-3
electrolytes, post-mortem 110-11
technique for drugs 259
aftermath 358-9
electrophoresis, ca pillary 1ca pillary
epiphyseal plate injury, radiology 56
causes 345-6
zone 261
epistaxis (nasal haemorrhage)
classification 352-9
electrophysiological monitoring, head
sum and history of 203
by site 352-9
injury 292
SUD! and presence of 214
351-9
hospital fatalities 364
ethanol see alcohol
investigations 349
pulmonary 235, 364
ethnicity, dental remains 444
medium 346-9
embryo transfer mix-ups 410
euthanasia by bath tub drowning 354,
pathophysiology 349-50
emotional abuse, assessment 15-16
355-6
-
~-~
_.
~ -:, " .- I , - _.
Index I 491
sex ual abuse 27- 8 reco nstructio n 440 age see gestati ona l age
non- accidental inju ry 125- 36 fat embolism 364 dental id entifica tion 439
post- mo rtem examination 157-8 fat her, alleged see paterni ty testing pathologiSt's role 3 19-27
eye witness, roa d traffi c co llision fa tty acid f)-o xidatio n defec ts 11 4-17 , scene 141 , 3 19- 20
387 244-6
firearm deaths 337
eyelids, non-acciden ta l inj ury 125 sudd en death 114-1 7, 244- 6
fi reworks 32 4
infa nt 114, 208 fi rst-degree burns 32 3
fa bri cated/in d uced/sim ula ted ill ness fatty acid ethyl esters 26 7 fixation, organs 163
(FlI ; Munchausen's sy ndrome by fatty change, liver, SID S 211-12 FLAIR MRI, head 64
proxy) 3 Fechtner's synd ro me 95 fl as h injuri es 326
assessment 16 fe male children see girls flotation test 185
492 I Index
fluid administration
types 128-9
glutaric aciduria 70
burns 325
see also retinal haemorrhages
glycoproteins, platelet 78
187,471-2
sudden death relating to 238
growth
haemorrhage 174
genes 397-8
growth plate (epiphyseal plate), in
death 330-1
bleeding due to 84, 90-3, 94, 94-5,
haemangioma, cardiac 234-5
6-7,8-9, 11,458-60
genetic (inherited) factors, recovery bruised child 77
298
461,466-7
HbAlc 112-13
non-accidental
girls see girls
haemophagocytic Iymphohistiocytosis
155, 163
Germany, child car restraints 386-7
type A (factor VIII deficiency) 80,
radiology 50,51-60
germinal matrix haemorrhage 172-3
88,96-7
long-bone 52-6
autopsy standard measurements by type B (factor IX deficiency) 80,
childhood) 146
dental estimation 441-2
sex and 83
SIDS 215,218
facial bone estimation 444
type C (factor XI deficiency) 80, 84,
424
466
haemorrhage
head-on 393
Glanzmann 's thrombasthenia 78, 84,
intracranial see intracranial
haemorrhage 128 - 34
glucose
pulmonary, sudden death and 240
mechanisms 129-30
see also hypoglycaemia
173-4, 301
- - ----- -
~ - -.=
Index I 493
subconjunctival 126
health professionals see carers and high-performance liquid
subperiosteal 161
health profess ion a ls chromatography 260-1
98
encephalopathy 172
histiocytoid cardiomyopathy
syndrom e 238
arrhythmias see arrhythmi as
histology (microscopic ap pearance)
haemosiderin 168
cond uction abnormalities 232
of brain injury in early life, timing
215
examination in sudden death
immersion victims 347
haemostasis 77 - 8,78-81
external 154
lungs in fetal / perina ta l death 185-6
primalY 78-80
his tological, for conduction
prenatal dental , age estimation via
measurements 82
abnormalities 162
441
secondary 80-1
internal 155-6
subdural haem ato m a 153
applications 272-3
surgery, death 370
SUm/SIDS 210-11,218
273-4
fetal 471
abuse 9-10, 25-6
268-9
heart block 232
death (perinatal /infant/you ng child)
head
congenital see congenital SUDr see subheading below
circumference (He)
malformations
sum 200,201-2
charts 475-80
drowning an d 354, 356
adverse historical factors 202-3
occipitofrontal 150
non-congenital/in general, sudden
infection 42
392-3
loss with burns 325
in hospital 373-5
101
abuse 12-13
n eonaticide; serial killing
death, scene 141
charts 475-80
homicide-s uicide see dyadic deaths
definition 282
post-mortem 150, 444
homocysti n uria, thromboembolism
differential diagnosiS 69-70
falls from 338 , 339- 40
235
epidemiology 283-94
see also crown-heel length ; horse-riding 341
294-300
helmets see headwear
defin ition and frequency 362-3
diagnosis 296-8
94
hot fluid burns see scalds
investigations 299-300
heparin 86, 86-7, 98-9
house fires see fire fatalities
mechanisms of brain injury 305
hepatic injury/problems see liver
HPV 41 ,42,404
severity, classification 283
heredity see entries under genetic
human immunodeficiency virus see
see also brain injury; neurolo gical
Hermansky-Pudlak syndrome 81, 82,
HIV
lesions
94-5
human papillomavirus (HPV) 41,42,
head-on collisions 393
herniations, brain 3 11
404
260, 262
272
hyaloid, posterior 129
headwear/helmets
high-molecular weight kininogens
hydroc ephalus 237
cycling 338-9,392-3
(HMWKs) 80
shun ts 369
3-hydroxyacyl CoA dehydrogen ase immers ion into hot water 326 infections 247-8
deficiency, long chain 244 see also drowning and near ante-moliem \09, 110
hymen exa mination 26 drowning bite injuries 428
measurements 35-6 immune thrombocytopenia 96 brain, young children 177
normal appearance/variants 29 -31, neonatal 98 fetal see congenital infec tions
32-5 immunoassays for drugs 259-60 fundus haemorrhages seconda ry to
configuration 32 false-positives 268 132
ga pi ng/ narro w i ng/ a tten ua tion oral fluids 266 sexually-transmitted, tests for 41-2,
34-5 imp act injuries 404
notches/ c1efts/ tra nsectio ns to head 302 sudden death 247-8
ridges/bumps/tags/bands 32-3 road traffic accident 388, 390, 391, in hospital 372
hyp ernatraemic dehydration III speed of impact see speed see also specific pathogens/diseases
hypersensitivity/allergy 342 see also shaken imp act syndrome inflamm ation
see also anaphyl ax is impac tion fractures 53 ante-m ortem \09
hypeliension, pulmonary arterial impetigo, bullous 20 SUD! 2 10-11
242 impress ions (w ith bite marks) inflammatory mediators, burns 325
hype li hermia, malignant 371 bite 431 informatio n
hypertonic vitreou s humour III suspect 431 from abused child 10-11
hyp ertrophic cardiomyopathy 228-9 in vitro fertilization mix-up 4\0 death (peI"inatal/infant/young child)
familial 228
inborn errors of met a bo lism see 146
s udd en death 156, 228-9
metabo lic disease post-mortem exc hange of (between
infant 206-7,228 indu ced illness see fabricated illness professionals) 163-4
hypofibIinogenemiaseefibrinogen infant(s) and babies (up to I year) recording see documentation
hypo g lycaemia 112, 113 death
from road traffic accident involvees
neonatal/infant 177 adequacy of cause of, in SUDI
388-9
hy pohidrotic ectodermal dys plasia , 200-1 see also histOly-taking; in terv iew ing
X- linked 235 bath tub drownings 354 informed consent see consent
hyp otensio n, matern a l, due to injUly bucket/pa il drownings 353 inhalation (aspiration)
192 insta ntaneo us, path ology 205-7 foreign body, asphyxial death
hypo thermia post-moliem examination, 330-1
immers ion 349 babies/infants/young c hildren of gastIic contents see sto mac h
hypotonic vitreous humour 111 rapid (in recognised illn ess), solvents see vo latile substances
hypovol ae mic shock with burns 325 path o logy 206, 207-8 water 332
hypoxaemia, head injury 290 sudden unexpected see sudde n inhelitance see entries under genetic
hypoxa nthine, time of death infant death syndrome; sudden injury/ traum a
estim ation 111-1 2 natural death; sudd en unexpected ch ild
hypoxi a (inc!. brain) \09 death in infancy accidental see ac cidental injury
drowning 349, 350 swimmin g pool drownin gs 356 asphyxia due to 330
hypoxic-ischaemic injury 167-8, 172 see also in fa nticide birth see birth trauma
non- accidenta l causes 297 genetic metabol ic disea se 114, 115 non-accidental see physical abuse
in shaken baby syndrome 307 growth chalis relative rankin g order of deaths due
boys 475
to 346
ICD see International Classificat ion of girls 478
sce ne of death from 141
Diseases neurological abno rmaliti es see
maternal , fet a l death due to 190-3
ice, spOliS injuIies 340-1 neurological les ions
see also specifIc sites and types of
identification of body/body remains newborn see neonates
injury
411-13
organ weigh ts 481-3
inqu ests and inquiri es 143
dental see dental identifica tion
infanticide 352-3 instrumental delivery
DNA techniques 411-13
definition
intracranial ha emo rrhage 174
fire fatality 320,4 12
England and Wales 181
sk ull fractures 30 1
missing persons 411-13
USA/elsewhere 181
insulin, levels 244
road traffic fatality 389
drowning 351,352-3,355
insulin-dependent diabetes melli tus
imaging see rad iology scene 141
244, 245
Index I 495
(ICD)
karyotype 396
leucomalacia, periventricular 169, 171,
198
keratin matrix, drug isolation from
fundus haemorrhages 132
accident 388
ketamine 272
life support systems, infant deaths
intestine (bowel)
ketones 113
following over 12 hours on
enquiry on protection examination
kidney
199-200
form 461
disease and failure
ligature, hanging by 332
injury 68
ante-mortem 113-14
lighting, scene of sudden death 139
in abuse 61-3,296,297,306-7
SIDS 211
linea vestibularis 41
imaging 71,289
kininogens see high-molecular weight
overload in intravenous feeding
191-2
labia minora and majora, examination lipid storage disorders 100, 177
perinatal 172-3
26
liquid chromatography
156
biochemical tests on post-mortem
260
infant 218
specimens see biochemical tests
liver
traumatic 293-4
haematological tests
injury
237-8
42-3
findings 154
birth-related 174
cyst 207
weight 484
fetus
spasm with immersion 350
neonatal and infant 481-3
290
389
long-bone fractures, radiology 52-6
lun gs
Meadow, Professor Roy 16, 199, 202,
exposure 274
examination/assessment 185-6
staining 159, 187-8, 193
chromatography (lVlECC) 261
maturity 187
in stomach 185
microbiological samples, sudden death
histology 162
abuse
mineralization, brain 168
weight 484
death due to 225-55
411-13
Fetal 471 - 2
bath tub drownings 356
mission statement of expert 448-9
haemorrhages 132
69-70, 71-3,76-105, 13 2
diseases of 244-7
lymphohistiocytosis, haemophagocytic
haematological 20, 76-105, 132
babies 177
84-5
sexual abuse 20,40- 1
cardiomyopathy 230
maceration
medical devices/procedures 368-71
molecular methods, post-mortem 150
macrocyto pathies 95
perinatal/inFant/young child death
toxicology) 260
macrophage response,
146
monitoring (physiological)
299-300
coning 369
mother
100
infection 247
infanticide 352
366, 367
genetic (inborn errors of metabolism)
injury causing fetal death 190-3
mass disaster
Reye-like 239
assessmen t 14, 17
fire 321
sudden unexpected death motor vehicle accidents see road traffic
mass spectrophotometry
post-mortem investigations 150
accidents; vehicles
261-2
inFants/yo un g children 150
mucocutaneo us bleeding 88
262
Fractures 53-6
dimen s ions
time-oF-flight 262
at birth 71
abuse 5-6
Index I 497
multicystic leucoencephalopathy
drowning 351,352
in hospital 364- 5
171-2
in hospital 372
imaging 66-7,310
mutations 410-11
neu roblastoma 84
ongoing health care, abuse 21
tumours 233-5
accidental head injLllY 289-90, 293
examination in sudden death 150-1
117, 208
neuroin tensive care, head injulY
injuries in abuse, dentists' role
neck
death 167
types I-IV 72
dissection 154
migration disorders 170, 217
osteomyelitis 71
injury
newborns see neonates
ostial ste nosis 230
ex amination
non - a cci denta 1/ in te n ti 0 n a 1/ deli bera te
~ - ox id at ion see fatty acid ~ -o x idation
neglect
pOisoning by health professionals
oxygenation
meningococcal infection
noxious gases in fires 322
pacemaker
neonates/newborns
nucleic acid amplification tech niques
failure 369
bod y
joint working with forensic
causes 209
observing the scene of sudden death
specialists 6-7,8-9,11,458-69
definition 180
139-40
paedophilia-related de aths 142
in hospital 371-2
obstetric events, death related to 248
pail, drowning in 353
498 I Index
pancreas
J t scene of sudden dea t h 138
mother, ca usin g fetal death 191
history from 10
in sudd en death in infan ts
monitoring
32 1
Pha rmChek'" 266-7
exa min at ion 160-1
Y chromosome 411
bite injury
platelets 78-9
38 7,388,390,391.392
susp ec t 431
adhesio n 78
death 19 2
sudd en death 147-8
neonatal 97-8
v iscera 68-9
scene of 138
sto ra ge po ol 92
penile injury 15
physical abuse/violence (non enhancem en t of coagul ation 79
perfusion. cerebral see cerebral acc ide nta l/in te ntional inj ury ; secretion/ rel ease 78-9
a re a
assessing pattern s of injury 17-l[l
fa lls 340
abnormalities 41
drownin g 354, 354-5
264
physiological 51-2. 70
medica l co nditions simulating see
hae morrha ge 236
171 ,2 11
SID S and p ath ol ogy raiSing
polymorphisms 399
perso nnel
co ncerns 212-17
restri ction fragment length 398-9
professionals
ther ma l injuries see th erm al injury
Sin g le nucl eotide (SNP) 405, 41 3
Index I 499
porencephaly 170
sudden death 210, 248
head see neuroimaging
post-mortem
professional witness 449-50
head injury see neuroim aging
CT scan 50
proforma 11, 17,458-69
immersion victims 347
106-26
prothrombin deficiency 89, 97
sudden death 146-7
necropsy)
prolonged 89-90
radiolluclid e bone scan, abuse 51
injuries 188-90
trauma
recovery room, death in 371
117-20
pulmonary arterial hypertension 242
reflex pulmonary vasc ular
111
pulmonary vascular vasoco nstriction, reporting of abuse
108
pulp, dental, DNA from 444
see also docum e ntation
pre-excitation (Wolff-Parkinson
pul se oximetry, head injury 290
respiratory chain (electron transport
pregnancy
Henoch-Schonlein 20 , 84, 85, 94
defects 116, 244, 246-7
events
pyruvate dehydrogenase deficiency
inh a lation into see inh alat ion
prekaJikrein 80, 81
177
in sudden death 240-2 , 243
deficiency 88, 97
SUDI and respiratory symp toms
pre-school children
related 176
restricted-access material s (RAMs) 260
9-10,460
prenatal age estimation 441
resuscitation (cardiopulmonary; CPR)
from parent 10
see also digital dental radiographic
retinal haemorrhages 129, 132
circumference charts
259
road traffic acc ident, unsuccessful
boys 475
radiology 47-76
388
retina)
laboratory tests) 12
drugs 259-61
238-9
biochem ica l tests
by drowning 354, 35 5, 35 7
Scale 14
scalds (hot fluids/ liquids) 324, 355
self-inflicted bite marks 19, 427
CPR-related 58 , 212
non-accidental 294
septic shock 363-4
155, 161-2
159
H. injluenzae 248
childhood) 146
fire fatalities 141,319-20 profeSSionals 274, 373 - 5
rickets 71 , 215
asphyxia l deaths 330
sex
338, 385-94
management 137-8
Y chromosome
193,393
sequence of events 139
conditions simulating 20, 40-1
fatal 385-90
school-age children
DNA tests 403 - 7
investigation 38 7-90
girls 480
skills and experience for 28-9
366
Scotland
sex ually -transm itted in fectio ns , tests
RSV 242
abuse
41-2, 404
legal dimensions 2, 3, 7
death scen e 141
429
road traffi c casualties and fatalities
hypoxic ischaemi a in 307
. ..
_...l.~_
--
~,:'- _I .
Index I 501
shaking 303-6
skin
speed (vehicle)
cervical spine injury see
burns see burns
of impact
cervicomedullary syndrome
disorders 20
deaths related to 386
mechanism 303
skull (cranium)
sperm/spermatozoa 395
pathophysiology 306
fractures 60-1, 294, 298
samples 28
unilateral 134
non-accidental 60-1 , 294-5
birth-related 176, 302, 312
imaging 65-6
in sudden death (infants/early
spinal cord trauma
shock
childhood) 146
wit hout radiographic abnormalities
syndrome 238
post-mo rtem exa mination, neon atal
post-mortem examination 157-8
septi c 363-4
birth-related 161
spin al trauma 58-9
spinal 311-12
post-mortem opening 156
cervical see cervical spine
413
asphyxial deaths 204, 205, 331
weight 484
mutations in 410
small bowel, sudden dea th relating to
neonatal and infant 471-3
technique 399-401
detectors/alarms 318-19
win ter, injuries 340-1
siderophages
passive
aspiration (of foreign material) fro m
significant harm 4
smothering
live birth determination, examination
signs of life, definition 181
accid ental (=overlaying) 203 , 215,
185
skeleton
150
stonefish 358
birth trauma 71
sodium val pro ate 99
stools, live birth determina tion
injuries 51-60
soft-tissue injury 69
strangulation
normal variants 70
soot inhalation 320,321,322
invasive 248
#
- -------------------- ~- --------------
502 I Index
subaponeurotic (subgaJeaJ)
in hospital 363, 363-5
technetium-99m scan, abuse 51
subdural haematoma/haemorrhage
diabetes mellitus 113
environmental, sleeping infant 203
in abuse 61-3,71,296,297,306
genetic metabolic disorders 114-17,
water, drowning and 348-9
302
in hospital see hospital
tet racyclin e and dental age estimatio n
location 175
nervou s system examination 166
443
191-2
sudden unexpected death in infancy
366-71
histology 153
epidemiology 201-3
non-accidenta l 325-7
infant 218
genetic metabolic disease 114, 208
dentist's role 423
traumatic 294
pathology 205-18
patterns 20
(subheading above)
209-12
th ird-d egree burns 323-4
173, 301
findings raising possibility of
93-4
198-224
dyadic death thrombocytopenic purpura, idiopathic
epidemiology 201-3
in hospital 373
94, 96
208
surgery, deaths assoc iated wi th
thromboplastin time
hypoxia 109
369-71
thrombosis, cerebral sinovenous 177
non-SIDS or 140
investigation 375
thromophilias, inherited 364
predicting 201
swimming pool drownings 356- 7
infant 210,211
terminology/definitions 140,
syphilis, congenital 71
weight
198-203
systematic enquiry in abuse 460
fetal 471-2
death in pinfancy
systemic lupus erythematosus, AV
time (determination)
fetus 193
systemic respon se to burns 325
of death 111-12
- - :, ._ . .
Index I 503
infant 199
monozygoti c 396
vehi cles
262
road see car; ro ad traffic acc idents
year olds)
ultrape rform ance liquid
358
boys 476
insertion/s tump 184- 5, J8 8
air 364
g irls 479
length 474
thrombotic 364
touch in g, non-consensua l 42 5
road traffic fatal ity in hospital,
129
fi res 322
uraemi a, vitreous humour a nalysis III
ventricles (brain), 3rd, colloid
techniqu es 259-62
collection 118- 19
vestibule, vaginal
387
USA, rep0l1ing of ab use 42 1
sce ne of sudde n death 138
tracheostomy 368
vaginal examination 27
see also physical abuse
292 -3
29 - 30
bites 428
390
see also vulvovag initis
neo natal, causi ng death 209
tumours
oc ular, damage in non-accide nta l
road traffic acc ident 389- 90
cardiac 233-5
injury 129
SJDS 215
intracranial 23 7
vascular 350
arres t 369
504 I Index
fetal 471-2
Wiskott-Aldrich syndrome 81,92,94
neonatal/infant 481-3
Wales, road traffic casualties and witness
127-8
warfarin 89, 90
eye, road traffic collision 387
134
loss with burns 324
232, 236
94
near-drowning; immersion
World Trade Center disaster (2001),
vitamin K
waterbeds 331
identifying remains 412
deficiency 98
watersiled damage 172, 311
malabsorption 101
wedging 204, 331
X chromosome 395
electrolytes 110-11
abuse 13
X-ray see radiography; skeleton, survey
inhalation/abuse 262,333,341-2
whiplash syndrome (whiplash shaking
sexual abuse 404, 405-6
family history 84
white matter
young children see infants; neonates;
neonatal 97
non-traumatic lesions 171-2
toddlers and pre-school children
- .
-~