Identification of
(Spender et al. 1988; Stallings et al. 1993a, b; Stevenson et al.
malnutrition in
children with
cerebral palsy: poor
performance of
weight-for-height
centiles
Lisa J Samson-Fang MD, Assistant Professor of Pediatrics,
University of Utah;
Richard D Stevenson* MD, Associate Professor of Pediatrics,
Department of Pediatrics, University of Virginia, 2270 Ivy
Road, Charlottesville VA 22903, USA.
*Correspondence to second author at address above.
E-mail: rds8z@virginia.edu
This study aims to evaluate the use of the United States
National Center for Health Statistics (NCHS) weight-for-
height centiles (WHC) in screening children with cerebral
palsy (CP) for depleted body fat and to identify an alternate
screening method. Growth data from 276 children aged from
3 to 12 years with CP were analyzed retrospectively. Height or
a proxy for height, mid-upper arm circumference, weight, and
skinfold thicknesses were recorded. Mid-upper arm fat area
was calculated for each participant. The sensitivities and
specificities of WHC and a number of alternative
anthropometric screening methods for identifying
participants with severely depleted fat stores were determined.
WHC <10th centile failed to identify 45% of children with
severely depleted fat stores. Triceps skinfold thickness <10th
centile identified 96% of malnourished children. WHC
standards lack adequate sensitivity for identification of
severely depleted fat stores in children with CP. Use of triceps
skinfold thickness, using cut-off value of <10th centile for
age and sex, is recommended to screen for suboptimal fat
stores in children with CP.
162 Developmental Medicine & Child Neurology 2000, 42: 162–168
Malnutrition is common in children with cerebral palsy (CP)
1994; Stallings et al. 1995). While clinicians are most aware
of the nutritional risks in severely affected children, those
with less severe CP frequently experience nutritional defi-
ciencies. For many years, small size and low weight were
accepted by clinicians as part of the child’s disability. It is now
clear that poor nutrition is not a necessary or unalterable
component of CP (Patrick et al. 1986, Sanders et al. 1990,
Amundson et al. 1994, Corwin et al. 1996).
The untoward consequences of malnutrition are numer-
ous and clinically significant (Viteri 1991). Malnourished
patients have decreased muscle strength, including respira-
tory musculature, with resultant impaired cough and pre-
disposition to pneumonia (Laaban 1997, Rigaud et al.
1997). Malnutrition results in increased circulation times
and diminished cardiac work capacity, clinically expressed
as a hypocirculatory state and a predisposition to develop
congestive heart failure when placed under cardiorespirato-
ry stress (Viart 1977, 1978). Malnutrition-related distur-
bances in immune function predispose to infection (Santos
1994). Cerebral growth, cognitive development, and motor
progress are impaired (Engsner et al. 1974, Grantham-
McGregor et al. 1991, Husaini et al. 1991, Viteri 1991, Smart
1993). Undernourished children show lower levels of
exploratory activity and attachment behavior which may
affect social–emotional development (Graves 1978).
Irritability and decreased activity have been ascribed clini-
cally to undernourished children (Viteri 1991) and docu-
mented in animal models of malnutrition (Graves 1978).
These characteristics may affect a child’s ability to partici-
pate in play, school, or rehabilitation.
Research documenting the effects of malnutrition specific
to CP has been limited. Linear growth failure, higher surgical
morbidity, delay in decubitus ulcer healing, and death have
been reported (Patrick et al. 1986, Shapiro et al. 1986, Jevsevar
and Karlin 1993, Stallings et al. 1993, Amundson et al. 1994,
Stevenson et al. 1994, Samson-Fang and Stevenson 1998).
In view of the high prevalence and negative conse-
quences, the identification and correction of malnutrition in
children with CP is paramount. However, assessing the nutri-
tional status of a child with CP is not straightforward. In the
general pediatric setting, clinicians frequently use weight-
for-height centiles (WHC), which are graphically represent-
ed on the USA’s National Center for Health Statistics (NCHS)
growth charts (Hamill et al. 1979). The recent development
of equations to estimate height from segmental measures
may lead to increased use of WHC in populations with dis-
abilities (Chumlea et al. 1994, Stevenson 1995).
WHC norms, based upon typically developing popula-
tions, may not be a valid nutritional indicator in children with
disabilities. Undernourished children with CP have changes
in body composition and proportion compared with normal-
ly developing peers. Alterations include increased total body
water, severely depleted fat stores, minimally depleted mus-
cle stores, severe short stature, and decreased bone density
(Berg and Isaksson 1970, Stallings et al. 1993, Stallings et al.
1995, Wilmshurst et al. 1996, Henderson et al. 1998). These
changes in body composition and proportion may affect the
extent WHC reflects body fat stores. Luckily there are other
anthropometric methods which can be used to assess the
nutritional status of a child with CP. Stallings et al. studied a
small group of children with CP and concluded that nutrition-
al status could be accurately assessed by measuring skinfolds
(Stallings et al. 1995). However, in the clinical setting, skin-
folds are rarely measured routinely. If WHC is in the normal
range, clinicians often judge the child to be nutritionally
sound and do not perform further nutritional assessment. In
our tertiary-care setting, it has been noted that many children
with normal WHC are malnourished when skinfolds are
assessed. On several occasions, a primary physician or health
department has discontinued a patient’s nutritional supple-
ments because of a normal WHC, despite skinfold thickness
being in the deficient range.
Based upon our clinical experience, we hypothesized that
NCHS WHC standards perform poorly when used to screen
for low fat stores in children with CP. The ability of WHC to
distinguish children with depleted fat stores was evaluated
using an anthropometric database and an anthropometrical-
ly derived indicator of body fat stores. Alternative anthropo-
metric measures were identified and their ability to identify
malnutrition in this population was assessed. Recommend-
ations are made regarding the best method of identifying
malnutrition in children with CP.
been reported previously (Stevenson et al. 1994). When pos-
sible, each measure was completed twice and the average
used for analysis. If scoliosis, joint contractures, or the child’s
level of cooperation precluded accurate measurement, the
measurement was not recorded. Demographic and clinical
information was collected by chart review and included type
of CP. Classifications for type of CP were spastic (quadriplegic,
diplegic, and hemiplegic), extrapyramidal, and mixed.
For this study, the anthropometric database was retro-
spectively reviewed. Data sets for children with CP aged from
3 to 12 years were included. This age-range restriction was
chosen because of the availability of validated equations to
predict height from segmental measures. Data sets with no
recorded tibia length, mid-upper arm circumference, triceps
skinfold, and subscapular skinfold thickness were excluded.
During the 6-year collection period, data sets of 276 children
met inclusion criteria. Some children had multiple data sets.
In these cases, one set was randomly chosen for analysis. The
study population was 48% female, 75% white, 24%
African–American, and 1% Asian–American. Mean age was
6.4 years. The distribution by type of CP was 36% spastic,
29% extrapyramidal, and 35% mixed.

Method INDEPENDENT VARIABLE : BODY FAT ASSESSMENT

An anthropometric database was established at the Kluge
Children’s Rehabilitation Center, Virginia, USA in 1991 for the
purpose of studying growth in children with CP. All children
with CP attending the outpatient clinic or admitted to the in-
patient service at the rehabilitation center were eligible to par-
ticipate. At each convenient contact, measurements were
recorded including: standing height (Harpenden digital sta-
diometer), recumbent length (digital supine measuring
table), arm circumference (steel tape measure), tibia length
(Harpenden anthropometer), head circumference (Ross
Inser-TapeTM), knee height (Ross knee height caliper), weight
(digital scale), triceps and subscapular skinfold thickness
(Holtain skinfold caliper). Children were weighed in a thin
layer of clothing and dry diaper or underwear. Trained anthro-
pometrists measured all children using published techniques
(Cameron 1986). All measures were performed on the child’s
right side unless the child’s CP was asymmetrical, in which
case, measures were taken on the less affected side. Excellent
intrarater and interrater reliability were established and have
Mid-upper arm fat area was decided as our ‘gold standard’
against which to compare WHC and other anthropometric
methods for screening for inadequate nutrition.
Determination of mid-upper arm fat area
Mid-upper arm fat area from triceps skinfold thickness and
mid-upper arm circumference were calculated using pub-
lished equations (Table I). For each participant, mid-upper
arm fat area was compared with published norms and per-
centage of mean for age/sex calculated (Frisancho 1981). For
study purposes, low fat area was defined as a mid-upper arm
fat area <5th centile for age and sex.
DEPENDENT VARIABLES
Weight-for-height centile
Using tibia length and published equations (see Table I), height
estimates were calculated for each participant (Stevenson
1995). WHC were calculated using Epi Info (version 6.04b)
software (Dean et al. 1995). Sensitivity and specificity of low

Table I: Anthropometric equations used in data analysis

Mid-upper arm fat and muscle areas (mm2)a

From arm circumference (AC in mm) and triceps skinfold thickness (TSF in mm)
Mid-upper arm area (MAA) = π/4 × d2 where d = AC/π
Mid-upper arm muscle area (MAMA) = [AC – πTSF)2]/4π
Mid-upper arm fat area (MAFA) = MAA – MAMA
Estimation of stature (S) from segmental measures (ages 2 to 12 years)b
Tibia length (TL in cm) S = (3.26 × TL) + 30.8 cm Standard error of estimation = 1.4 cm
Knee height (KH in cm) S = (2.69 × KH) + 24.2 cm Standard error of estimation = 1.1 cm
Alternative screening toolsc (Foman et al. 1982, Hammer et al. 1991)
Body mass index (BMI) = Wt/Ht2 where wt = weight (kg) and ht = height (m)
Lean body mass index (LBMI) = Ht2/Wt where wt = weight (kg) and ht = height (m)
Head circumference to arm circumference ratio (HC/AC) = arm circumference (cm)/head circumference (cm)
a Frisancho 1981.
b Stevenson 1995.
c Frisancho 1981, Hammer et al. 1991, Committee on Nutrition American Academy of Pediatrics 1998.

Malnutrition and Cerebral Palsy Lisa J Samson-Fang and Richard D Stevenson 163
WHC to identify children with low arm fat area were deter-
mined. These calculations were performed using WHC cut-off
values of <5th and 10th centile and χ2 analysis.
Alternative screening methods
A number of anthropometric nutritional indicators used to
screen for malnutrition in other settings were identified.
These included arm circumference for age, head circumfer-
ence to arm circumference ratio, body mass index
(weight/height2), lean body mass index (height2/weight), tri-
ceps and subscapular skinfold thickness (Frisancho 1981,
1990; Foman et al. 1982; Hammer et al. 1991; Committee on
Nutrition American Academy of Pediatrics 1998). Their ability
to identify malnutrition in our study population was calculat-
ed by using the same methodology described above for WHC.
Multiple cut-off values for each parameter were examined.
For several, cut-off values relative to height age (e.g. arm cir-
cumference for height age) were used because extreme short
stature was prevalent.
ADDITIONAL ANALYSIS
To assess the impact of altered body proportion and compo-
sition upon WHC, the height z score and mid-upper arm
muscle area were calculated for each study participant.
Height z scores were calculated using tibia length: height
estimate and Epi Info (version 6.04b) software (Dean et al.
1995). Mid-upper arm muscle area was calculated using a
published equation (see Table I). For each child, mid-upper
arm muscle was compared with published normative values
and calculated percentage of the median for age/sex and
height age/sex.
Results
The children had a high prevalence of depleted fat stores.
Mid-upper arm fat area was, overall, 77% of expected for
age/sex, with 38% of participants having arm fat area <5th
centile (Table II).
WEIGHT- FOR- HEIGHT CENTILE
The sensitivities and specificities of WHC and other
assessed anthropometric screening tools are presented in
Table III. Low WHC was a highly specific but not a sensitive
indicator of depleted body fat stores. Using WHC <5th
centile as a screening tool, the clinician will miss 57% of
children with a low arm fat area. Raising the screening cut-
off for WHC to <10th centile improved sensitivity minimal-
ly. Assessment of WHC in participants with different types
of CP revealed poor performance in all groups. To account
for possible error introduced when calculating height
from tibia length, the analysis was repeated using stature
and recumbent length in the subset of participants for
whom these measures were possible. The analysis was also
repeated using knee-height: height estimates. Sensitivities
remained poor in all analyses.

Table II: Nutritional status of the study population

Mid-upper arm fat area method (MAFA)

n (%) Sex ratio Mean % of median
(F:M) MAFA for age/sex

All participants 276 132:144 77 ± 45

Participants with MAFA >5%a 171 (62) 78:93 98 ± 46
Participants with MAFA <5%b 105 (38) 54:51 43 ± 10
a Mid-upper arm fat area >5th centile for age and sex.
b Mid-upper arm fat area <5th centile for age and sex.

Table III: Sensitivities and specificities of weight-for-height centile and other assessed screening
tools for identification of participants with mid-upper arm fat area <5th centile for age

Mid-upper arm fat area <5% for age/sex

N Sensitivity Specificity X p value

WHC <5% 272 44 91 45 <0.001

WHC <10% 272 54 88 56 <0.001
Arm circumference <10% for age/sex 276 60 88 74 <0.001
Arm circumference <10% for height age/sex 275 34 99 55 <0.001
Body mass index <10% for age/sex 276 65 82 61 <0.001
Body mass index <10% for height age/sex 275 68 80 62 <0.001
Head circumference/arm circumference <0.38 270 94 43 41 <0.001
Lean body mass/index >600 276 93 42 37 <0.001
Triceps skinfold <5% for age/sex 276 78 95 156 <0.001
Triceps skinfold <10% for age/sex 276 96 82 158 <0.001
Subscapular skinfold <5% for age/sex 276 25 98 34 <0.001
Subscapular skinfold <10% for age/sex 276 49 91 54 <0.001

164 Developmental Medicine & Child Neurology 2000, 42: 162–168

Alternative screening tools
Arm circumference, body mass index, and subscapular skin-
fold thickness performed poorly as screening tools. Arm cir-
cumference to head circumference ratio, lean body mass
index, and triceps skinfold thickness were more promising
(see Table IV). The positive predictive values of arm circum-
ference to head circumference ratio and lean body mass
index were low. Clinical use would result in extensive over
referral. The positive predictive value for triceps skinfold
thickness <10th centile was high. Statistical testing using it
in 100 participants with CP showed that the clinician would
miss two children with low arm fat area and would over refer
only 10 with normal arm fat area.
ALTERED BODY PROPORTIONS
To assess the contribution of altered body proportion and
composition to the poor sensitivity of WHC, the participant’s
stature and arm muscle area were assessed. The results are
presented in Table V. Short stature was common. The stature
of the children with low fat stores was, on average, 2.8 SDs
below that expected for age. Participants maintained muscle
areas near the average range, despite extremely depleted fat
stores. While fat-area percentage of the mean changed only
slightly when adjusted for the population’s height age, mus-
cle-area percentage of the mean was higher when adjusted
for height age.
Discussion
Malnutrition, defined as low mid-upper arm fat area, was
common in our population. The high prevalence likely
reflects referral bias. It is difficult to compare prevalence with
that of prior studies as different methods of assessing and
defining malnutrition in differing diagnostic subgroups have
been used. Stallings et al.’s study of children with diplegia
and hemiplegia showed that 29% of participants had a tri-
ceps skinfold thickness <15th centile (Stallings et al. 1993a).
In this study group, 27% of children with diplegia or hemi-
plegia had triceps skinfold <10th centile. This suggests
excellent agreement in similar tertiary-care populations.
WHC
WHC was not an adequate screen for malnutrition, for sev-
eral reasons. First, children with CP often experience mal-
nutrition of long duration. WHC is a better indicator of
acute malnutrition than chronic (Waterlow 1973, Waterlow
et al. 1977, Radheshyam 1987). Second, there was a high
prevalence of extreme short stature. The NCHS derived the
WHC from cross-sectional measures on children without
disabilities and with normally distributed stature. These
normative values may not be of use in evaluating children
with extremely short stature. Waterlow reviewed the litera-
ture and concluded that expected weight for height is rela-
tively independent of age (Waterlow 1973, Waterlow et al.
1977). However, his analysis generally focused upon
statures within the broad range of normal measurements
for age. Third, WHC may perform poorly as artificially
added weight may elevate WHC, thus reducing its ability to
reflect fat stores. Potential added weights in this study
included clothing and fecal mass, as severe chronic consti-
pation is common in our clinical setting. The use of tibia
length to estimate height could result in false elevation of
WHC if a child has undergrowth of the lower extremities in
association with motor impairment. A subset of our study
population was able to have a standing height or recum-
bent length measured. For these children, measured and
estimated stature were highly correlated, and WHC based

Table IV: Sensitivities, specificities, and predictive values of the more promising alternative
screening methods

Mid-upper arm fat area <5% for age and sex

Sensitivity Specificity PPV NPV

Head circumference/arm circumference <0.38 94 43 50 93

Lean body mass index >600 93 42 49 91
Triceps skinfold <10% for age/sex 96 82 77 97

PPV, positive predictive value; NPV, negative predictive value.

Table V: Mid-upper arm fat and muscle areas mean percents of the median for age and height age for participants with low and
normal fat areas

Category n (%) Height Mid-upper arm fat area Mid-upper arm muscle area
z score Average % Average % of Average % Average % of
of median median for of median median for
for age height age for age height age

All participants 276 –2.0 ± 1.7 77 ± 45 81 ± 41 106 ± 25 120 ± 24

Participants with normal fat storesa 171 (62) –1.6 ± 1.6 98 ± 46 100 ± 40 114 ± 24 125 ± 24
Participants with low fat storesb 105 (38) –2.8 ± 1.6 43 ± 10 47 ±10 93 ± 20 111 ±21
a Children with mid-upper arm fat area >5th centile for age.
b Children with mid-upper arm fat area <5th centile for age.

Malnutrition and Cerebral Palsy Lisa J Samson-Fang and Richard D Stevenson 165
upon standing height or recumbent length continued to
perform poorly as a screening tool for depleted fat stores.
The use of WHC to screen for malnutrition in other ter-
tiary referral populations has had limited study. It performed
poorly in a group of pediatric oncology patients and its use
has been questioned in children with chronic renal disease
(Chantler and Holliday 1973, Wilmet et al. 1995). This study
contributes to the growing body of evidence that WHC is an
inadequate screen for depleted fat stores in populations like-
ly to experience chronic malnutrition, extreme short stature,
or altered body proportions. Such populations might
include children with malabsorption, complex congenital
heart disease, chronic pulmonary or renal disease, and chil-
dren with chronic infections (e.g. HIV). Clinicians should
consider other methods of assessing nutritional status in
these populations.
BODY COMPOSITION ANALYSIS
This study suggests that alteration in body composition and
proportion in children with CP may alter the ability of WHC
to reflect fat stores. This might contribute to its poor perfor-
mance as a nutritional indicator. Our study was not designed
specifically to assess body composition. The available data
did allow for assessment of stature and muscle area. On aver-
age, malnourished children were extremely short. Muscle
area was maintained near expected values for age and sex
despite the extreme fat-store depletion. Combining the
maintained muscle area with extreme short stature, muscle
areas for height age were elevated above expected values.
This elevation of muscle area for stature may raise WHC.
Thus the child with severely depleted fat stores may have a
WHC within the normal range based upon the excessive con-
tribution of muscle tissue to body composition. The preser-
vation of the upper-arm muscle area was not an unexpected
finding (Stallings et al. 1993, 1995). It may reflect the highly
compensated state malnourished children are able to main-
tain (Viteri 1991). An alternative explanation is that abnormal
tone stimulates and maintains muscle bulk (Smith et al.
1991). To explore this theory, muscle area of children with
quadriplegic CP to that of children with diplegia or hemiple-
gia was compared in this study. The muscle areas were not
significantly different among these groups. However, chil-
dren with diplegia and hemiplegia may have had training
effects due to use of assistive devices or excessive use of unin-
volved extremities which may also stimulate maintenance of
muscle bulk.
ALTERNATE ANTHROPOMETRIC SCREENING TOOLS
Among the alternate screening devices, only measurements
of triceps skinfold thickness <10th centile performed to
standards thought to be clinically acceptable. While several
others had adequate sensitivities, their positive predictive
values resulted in significant overreferral. The higher sensi-
tivity of triceps skinfold thickness over subscapular skinfold
thickness reflects the high centripetal fat ratio previously
noted in malnourished populations including children with
CP (Spender et al. 1988, Stallings et al. 1993). The cut-off
point of triceps skinfold thickness <10th centile performed
well as a screen for depleted fat stores in our tertiary-care
referral population. In the primary-care setting, where mal-
nutrition may be less prevalent, the positive predictive value
may be lower. The usefulness of this screening method
166 Developmental Medicine & Child Neurology 2000, 42: 162–168
needs to be confirmed in a population-based sample of chil-
dren with CP.
STUDY LIMITATIONS
The performance of WHC and a number of alternative
anthropometric screening tools were assessed in compari-
son to determination of body fat stores using calculated mid-
upper arm fat area. Many assumptions are made when
calculating mid-upper arm fat area; bone area is assumed to
be a constant, the arm to be cylindrical in shape, triceps and
biceps skinfold thicknesses to be equivalent, the skinfold to
be twice the average fat rim diameter, the muscle compart-
ment to be cylindrical, the cross-sectional areas of the neuro-
vascular tissue and the humerus to be relatively small and
ignored, and bone atrophies in proportion to muscle
wastage during protein–energy malnutrition. These assump-
tions may not hold true in the child with CP (Van Den Berg-
Emons et al. 1998). However, at this time, calculation of
mid-upper arm fat area represents the best anthropometric
indicator of body fat stores, with published normative values
available for comparison. Further study, using a more direct
method of assessing fat stores, would be ideal but would
require significant funding for a large population.
The assumptions, listed above, also apply to calculation of
muscle area. Another assumption is that the muscle compart-
ment contains only muscle. In animal models of CP, examina-
tion of muscle by light- and electron-microscopy has
revealed normal structure except reduced fiber length
(Tardieu et al. 1977). This suggests that our calculations of
muscle area were not artificially elevated by fibrotic or fatty
tissue within the muscle compartment. It is important not to
equate maintenance of muscle area with maintenance of
muscle mass. Measurement of muscle area in one extremity
may not reflect the muscle mass in the rest of the body.
Furthermore, despite similar muscle area, the child with
extreme short stature will have less muscle mass than a taller
child with equivalent muscle area. Our methodology did not
allow for assessment of total-body muscle mass. Berg and
Isaksson, using isotope dilution, found low body cellular
mass in children with CP (Berg and Isaksson 1970). This sug-
gests that muscle mass is diminished. Further studies using
more direct methods to assess muscle area and overall mus-
cle mass are needed to confirm these findings.
The ability of WHC and other anthropometric tools to
identify malnutrition in this population was studied.
However, the definition of inadequate nutrition in this study
was a cut-off value for mid-upper arm fat area chosen arbi-
trarily and referenced against normative data in children
without disability. Ideally, values used to define malnutri-
tion should be related to clinical outcomes. Research has
documented numerous negative physiologic consequences
of malnutrition but has focused upon populations dissimi-
lar to ours. The bulk of study has been in developing
nations. Extreme poverty, recurrent and chronic infections,
and micronutrient deficiencies are confounding variables
usually not present in our patients with CP. Other studies
have focused on children with cystic fibrosis; the elderly;
and adults with trauma, burns, or HIV. Data obtained in
these populations may not be relevant. Future research
must examine the health consequences of malnutrition in
this population. For example, would nutritional improve-
ments result in better cognitive and functional outcomes,
less surgical morbidity, and improved quality of life? Without
outcome data, defining normal and abnormal nutritional
status remains arbitrary.
This study focused upon identifying children with deplet-
ed fat stores because, until demonstrated otherwise, it is
assumed that depleted fat stores are not good.
Unfortunately, once the child is identified, the clinician can
only be offered little scientifically based guidance on how to
intervene. Current understanding of prevention and treat-
ment of malnutrition in this population is in its infancy.
Intervention trials are limited. Gastric feeding studies have
assessed impact upon weight gain and linear growth with lit-
tle assessment of functional outcomes (Patrick et al. 1986,
Shapiro et al. 1986, Rempel et al. 1988, Sanders et al. 1990,
Corwin et al. 1996). Controversies regarding their impact
upon morbidity and mortality have been raised (Strauss et al.
1997). Other intervention trials have been small. Gisel docu-
mented no nutritional status change in response to oral sen-
sorimotor treatment (Gisel et al. 1995, Gisel 1996). Physical
activity and adjustment of oral diet had a modest positive
effect in a small number of children generally with milder CP
(Berg 1970). Clinical trials are needed to examine the life
and health impacts of nutritional interventions including
oral-motor therapies, oral caloric supplements, and feeding
gastrostomies.
Conclusion
Children with CP are at high risk for malnutrition and WHC is
not a reliable screen. Routine measurement of skinfold thick-
ness in this population is recommended. For clinical efficiency,
triceps skinfold thickness <10th centile is a good initial screen.
In the child with a low triceps skinfold thickness, assessment of
linear growth, state of health, and other skinfolds will clarify
nutritional status. These results need to be confirmed using
direct methods of assessing body composition. Methods of
screening in other pediatric populations likely to have a high
prevalence of chronic malnutrition need validating. The unto-
ward effects of malnutrition in CP and the best methods for
prevention and treatment need extensive study.
Accepted for publication 11th August 1999.
Acknowledgements
The authors wish to thank L Virgil Cater, Collen Amra, Judy
Jacobson, Carla Wheaton, Carmen Booker, and Vivienne Spauls who
were anthropometrists during the study period. We are grateful to
the children and families who consented to measurement. We are
also grateful to James A Blackman MD MPH, Professor of Pediatrics at
the University of Virginia and Director of Research at the Kluge
Children’s Rehabilitation Center for his guidance. This study was
supported by Genentech Foundation for Growth and Development,
United Cerebral Palsy Research and Educational Foundation, The
Kluge Research Fund of the University of Virginia, and The Research
Development Fund of the University of Virginia.
References
Amundson JA, Sherbondy A, Van Dyke DC, Alexander R. (1994)
Early identification and treatment necessary to prevent
malnutrition in children and adolescents with severe disabilities.
Journal of the American Dietetic Association 94: 880–3.
Berg K. (1970) Effect of physical activation and improved nutrition
on the body composition of school children with cerebral palsy.
Acta Paediatrica Scandinavica 204: (Suppl.) 53–69.
— Isaksson B. (1970) Body composition and nutrition of school
children with cerebral palsy. Acta Paediatrica Scandinavica
204: (Suppl.) 41–52.
Malnutrition and Cerebral Palsy Lisa J Samson-Fang and Richard D Stevenson 167
Cameron N. (1986) The methods of auxological anthropology. In:
Falkner F, Tanner JM, editors. Human Growth: A Comprehensive
Treatise. New York: Plenum Press.
Chantler C, Holliday MA. (1973) Growth in children with renal
disease with particular reference to the effects of calorie
malnutrition: a review. Clinical Nephrology 1: 230–42.
Chumlea WC, Guo SS, Steinbaugh ML. (1994) Prediction of
stature from knee height for black and white adults and
children with application to mobility-impaired or handicapped
persons. Journal of the American Dietetic Association
94: 1385–8.
Committee on Nutrition American Academy of Pediatrics. (1998)
Assessment of nutritional status. In: Barness LA, editor. Nutrition
Handbook. Elk Grove, IL: American Academy of Pediatrics.
p 193–201.
Corwin DS, Isaacs JS, Georgeson KE, Bartolucci AA, Cloud HH,
Craig CB. (1996) Weight and length increases in children after
gastrostomy placement. Journal of the American Dietetic
Association 96: 874–9.
Dean A, Dean J, Coulombier D. (1995) Epi Info. (6.04b). Atlanta,
Centers for Disease Control and Prevention.
Engsner G, Habre D, Sjogren I, Vahlquisr B. (1974) Brain growth in
children with Kwashiorkor. Acta Paediatrica Scandinavica
63: 687–94.
Foman SJ, Haschke F, Ziegler EE, Nelson SE. (1982) Body
composition of reference children from birth to age ten years.
American Journal of Clinical Nutrition 35: 1169–75.
Frisancho AR. (1981) New norms of upper limb fat and muscle areas
for assessment of nutritional status. American Journal of Clinical
Nutrition 34: 2540–5.
— (1990) Anthropometric Standards for the Assessment of Growth
and Nutritional Status. Ann Arbor, MI: University of Michigan
Press.
Gisel EG. (1996) Effect of oral sensorimotor treatment on measures
of growth and efficiency of eating in the moderately eating
impaired child with cerebral palsy. Dysphagia 11: 48–58.
— Applegate-Ferrante T, Benson JE, Bosma JF. (1995) Effect of oral
sensorimotor treatment on measures of growth, eating efficiency
and aspiration in the dysphasia child with cerebral palsy.
Developmental Medicine & Child Neurology 37: 528–43.
Grantham-McGregor S, Powell C, Walker S, Himes J. (1991)
Nutritional supplementation, psychosocial stimulation, and
mental development of stunted children: the Jamaican study.
Lancet 338: 1–5.
Graves P. (1978) Nutrition and infant behavior: a replication study in
the Katmandu Valley, Nepal. American Journal of Clinical
Nutrition 31: 541–51.
Hamill PVV, Drizd TA, Johnson CL, Reed RB, Roche AF, Moore WM.
(1979) Physical growth: National Center for Health Statistics
percentiles. American Journal of Clinical Nutrition
32: 607–29.
Hammer LD, Kraemer HC, Wilson DM, Titter PL, Dornbusch SM.
(1991) Standardized percentile curves of body mass index for
children and adolescents. American Journal of Disease of
Children 145: 259–63.
Henderson RC, Lin PP, Green WB. (1998) Bone-mineral density in
children and adolescents who have spastic cerebral palsy.
Journal of Bone and Joint Surgery 77A: 1671–81.
Husaini M, Karyadi L, Husaini Y, Karyadi S, Pollitt E. (1991)
Developmental effects of short-term supplementary feeding in
nutritionally-at-risk Indonesian infants. American Journal of
Clinical Nutrition 54: 799–804.
Jevsevar DS, Karlin LI. (1993) The relationship between
preoperative nutritional status and complications after operation
for scoliosis in patients who have cerebral palsy. Journal of Bone
and Joint Surgery 75A: 880–4.
Laaban J. (1997) Malnutrition, renutrition and respiratory function.
Pediatric Pulmonology 16: (Suppl.) 167–8.
Patrick J, Boland M, Stoski D, Murray GE. (1986) Rapid correction of
wasting in children with cerebral palsy. Developmental Medicine
& Child Neurology 29: 734–9.
Radheshyam B. (1987) A comparison of five anthropometric indices
for identifying factors of malnutriton. American Journal of
Epidemiology 126: 258–67.
Rempel GR, Colwell SO, Nelson RP. (1988) Growth in children with
cerebral palsy fed via gastrostomy. Pediatrics 82: 857–60.
Rigaud D, Moukaddem M, Cohen B, Malon D, Reveillard V, Mignon M.
(1997) Refeeding improves muscle performance without
normalization of muscle mass and oxygen consumption in
anorexia nervosa patients. American Journal of Clinical
Nutrition 65: 1845–51.
Samson-Fang LJ, Stevenson RD. (1998) Linear growth velocity in
children with cerebral palsy. Developmental Medicine & Child
Neurology 40: 689–92.
Sanders K, Cox K, Cannon R, Blanchard D, Pitcher J, Papathakis P,
Varella L, Maughan R. (1990) Growth responses to enteral
feeding in children with cerebral palsy. Journal of Parenteral and
Enteral Nutrition 14: 23–6.
Santos J. (1994) Nutrition, infection, and immunocompetence.
Infectious Disease Clinics of North America 8: 243–67.
Shapiro BK, Green P, Krick J, Allen D, Capute AJ. (1986) Growth of
severely impaired children: neurologic versus nutritional factors.
Developmental Medicine & Child Neurology 28: 729–33.
Smart J. (1993) Malnutrition, learning and behavior: 25 years on
from the MIT symposium. Proceedings of the Nutrition Society
52: 189–99.
Smith DE, Stevens MC, Booth IW. (1991) Malnutrition at diagnosis
of malignancy in childhood: common but mostly missed.
European Journal of Pediatrics 150: 318–22.
Spender QW, Cronk CE, Stallings VA, Hediger ML. (1988) Fat
distribution in children with cerebral palsy. Annals of Human
Biology 15: 191–6.
Stallings VA. (1993a) Nutritional status and growth of children with
diplegic or hemiplegic cerebral palsy. Developmental Medicine &
Child Neurology 35: 997–1006.
— Charney EB, Davies JC, Cronk CE. (1993b) Nutrition-related
growth failure of children with quadriplegic cerebral palsy.
Developmental Medicine & Child Neurology 35: 126–38.
— Cronk CE, Zemel BS, Charney EB. (1995) Body composition in
children with spastic quadriplegic cerebral palsy. Journal of
Pediatrics 126: 833–9.
Stevenson RD. (1995) Use of segmental measures to estimate
stature in children with cerebral palsy. Archives of Pediatric and
Adolescent Medicine 149: 658–62.
— Hayes RP, Carter LV, Blackman JA. (1994) Clinical correlates of
linear growth in children with cerebral palsy. Developmental
Medicine & Child Neurology 36: 135–42.
Strauss D, Kastner T, Ashwal S, White J. (1997) Tubefeeding and
mortality in children with severe disabilities and mental
retardation. Pediatrics 99: 358–62.
Tardieu C, Tabary JC, Tabary C, Huet de la Tou E. (1977)
Comparison of the sarcomere number adaption in young adult
animals. Journal de Physiologie 73:1045–55.
Van Den Berg-Emons RJ, Van Baak MA, Westerterp KR. (1998) Are
skinfold measurements suitable to compare body fat between
children with spastic cerebral palsy and healthy controls?
Developmental Medicine & Child Neurology 40: 335–9.
Viart P. (1977) Hemodynamic findings in severe protein-calorie
malnutrition. American Journal of Clinical Nutrition 30: 334–48.
— (1978) Hemodynamic findings during treatment of protein
calorie malnutrition. American Journal of Clinical Nutrition
31: 911–26.
Viteri FE. (1991) Protein energy malnutrition. In: Viteri FE, Walker AW,
Durie PR, Hamilton JR, Walker-Smith JA, Watkins JB, editors.
Pediatric Gastrointestinal Diseases. Philadelphia: BC Decker.
p 1596–611.
Waterlow JC. (1973) Note on the assessment and classification of
protein-energy malnutrition in children. Lancet 1: 87–9.
— Buzina R, Keller W, Lane JM, Nichaman MZ, Tanner JM. (1977)
The presentation and use of height and weight data for
comparing the nutritional status of groups of children under the
age of ten years. Bulletin of the World Health Organization
55: 489–98.
Wilmet E, Ismail AA, Heilporn A, Welraeds D, Bergmann P. (1995)
Longitudinal study of the bone mineral content and of soft tissue
composition after spinal cord section. Paraplegia 33: 674–7.
Wilmshurst S, Ward K, Adams JE, Langton CM, Mughal MZ. (1996)
Mobility status and bone density in cerebral palsy. Archives of
Disease in Childhood 75: 164–5.

168 Developmental Medicine & Child Neurology 2000, 42: 162–168