journal of Hospital Infection (1992) 22, 279-286

The mechanisms and risks of surgical glove

perforation

J. D. Palmer and J. W. S. Rickett

Department of Surgery, Torbay Hospital, Torquay, Devon, UK

Accepted for publication 22 September 1992

Summary: Intact surgical gloves are a barrier to hepatitis B virus and

human immunodeficiency virus (HIV) but once perforated during surgery
they cannot sustain adequate defence. This study examines the rate of glove
perforations during surgery at a District General Hospital. In total, 275 pairs
of gloves were collected from 100 consecutive operations. In the 43% of
gloves that had been damaged 200 perforations were recorded. The mean rate
per operation in the surgeon’s gloves was 1.18. Injuries to the non-dominant
index finger were significantly higher than injuries to other parts of the hand.
Injuries occurred particularly during manipulation of the needles and at
wound closure. Consultants were more likely to have glove perforation than
juniors. Operations requiring manipulation of instruments deep within the
wound had a higher rate than those on the surface. The results of the study
indicate that a surgeon risks more than one hepatitis B infection per lifetime
and that at least one in 1500 surgeons is likely to be infected by HIV during
the next 35 years.

Keywords: Hepatitis B; HIV; surgical gloves; glove perforations.

Introduction

Surgical gloves protect against hepatitis B virus (HBV) and human

immunodeficiency virus (HIV) infection.‘,* HBV transmission from
surgeon to patient and vice versa is well documented.3*4 Seroconversion to
HIV has been seen in isolated cases of needlestick injury.’ Double gloving
may reduce the perforation rate6 and has been shown to reduce the risk of
HBV transmission from surgeon to patient.7 Glove perforation with
inoculation, or contact of blood with a skin defect under the glove, may
increase the risk of HBV and possibly of HIV. The incidence of HBV
among homosexuals and intravenous drug abusers is falling but in surgeons
the incidence in 1980-84 (2.5 new cases per 100 000) was twice that in
1975-79 (12 new cases per 100 OOO).3 As HBV can be transmitted in
dilutions as low as 10e7 to lo-‘, there is no doubt that surgeons are at risk.
To investigate the risks to surgical staff we performed a prospective study

Correspondence to: Mr J. D. Palmer, Research Neurosurgery Registrar, Wessex Neurological Centre,

Southampton General Hospital, Southampton SO9 4XY, UK.
0195-6701/92j120279+08 $OS.OO/O 0 1992 The Hosptal lnfectmn Society

279
280 J. D. Palmer and J. W. S. Rickett

of glove perforation rates with special reference to the mechanism and site of
injury.

Methods
Gloves (‘Regent Dispo’ or ‘Rekent Biogel’, LRC Products Ltd) were
collected from all members of the surgical team, including the scrub nurse,
following 100 consecutive operations classified as intermediate and major.
Three surgical teams were involved in the study, two from general and one
from gynaecological surgery. Each glove was tested by one of three
observers, who distended the gloves in an iden ‘cal way with water. It was
found that filling the glove with water did not etect all the punctures;
rather each finger had to be distended with water \ o demonstrate these.
Perforations were detected by observing fine jets of water and the number
and site were recorded. A control group of 100 pairs of unused gloves was
also tested in an identical fashion.
A questionnaire about the details of the operation and the factors possibly
leading to an increased risk of injury was completed by each member of the
scrub team. The operations were subdivided into three groups, superficial,
intermediate and deep, as defined in Table I. The status (emergency or
planned) and the type of operation, the time of day, whether the subject was
right- or left-handed, his grade, the presence of skin damage on the hands
before surgery, the awareness of glove damage, and any changes of gloves
were recorded. When the subject was aware of damage the mechanism of
injury was noted. Statistical analysis of the data was made by the Student’s
t-test.

Results
In 275 pairs of gloves collected from 100 operations 200 perforations were
detected; 43% of gloves had one or more perforations. The majority of the
gloves (89%) were ‘Regent Biogel’ so comparison between glove types could
not be made.
The glove perforation rate for the surgeon was 1.18 perforations per
operation overall and varied according to the type of procedure from O-3

Table I. Dejinition of types of operation

Superficial Operations that do not require the

operator’s hand to enter the wound
cavity
Intermediate Operations that may require part of
the operator’s hand to enter the
wound cavity
Deep Operations that may require the
operator’s whole hand to enter the
wound cavity
 Surgical glove perforation 281

perforations per operation for herniorrhaphy to 2.2 per operation for colonic
resection (N= 3) and six per operation for the only thyroidectomy (Table
II). The rate for the deep group of operations (0.85 perforations per
operation) was greater than that for the superficial group (0.41) (P~0.05).
Only a small number of intraabdominal gynaecological procedures were
included in the study, so that no comparisons could be made with general
surgery.
The difference of 1.6 perforations per operation occurring in the
emergency group in comparison with 2.1 in the elective group was not
significant. The mean rate of perforation was at its lowest between midnight
and 9 am and highest in the afternoon but this difference was not significant
(P> 0.05).
The consultant (1.71 perforations per operation) was more likely to
perforate his gloves than the registrar (O-89 perforations per operation) or
the senior house officer (0.74 perforations per operation) (P value for both
< 0.05). There was no significant difference in the perforation rate between
the grades of junior staff or the grades of nurses (Table III). There was no
significant difference between the surgical teams.
In 13% of the operations the subject had damage to his hands before the
start of the procedure. In 42% of the operations where glove damage
occurred the subject was aware of the needle perforation. Gloves were
changed in only 28% of procedures where the subject was aware of the
damage. Most perforations were associated with closure of the surgical
wound (Table IV).
Of the perforations in the surgeon’s gloves 73% involved the
non-dominant hand; 50% were limited to the index finger and 25% to the

Table II. Perforations in the surgeon’s gloves per operation

according to operation type

Operation Rate No. of

operations

Thyroidectomy 6.0 1
Colposuspension 3.3 3
Adrenalectomy 3-o 1
Colorectal surgery 2.2 13
Aortic surgery 2.0 2
Oesophagogastric surgery 1.8 12
Varicose veins 1.8
Cholecystectomy 0.9 1;
Breast surgery 0.9 7
Small bowel surgery 0.8 5
Herniorrhaphy 13
Appendicectomy 0.6 5
Anal surgery 0.5
Laparotomy-no resection 0.4 :
Other 0.3
Total 1.18 10:
282 J. D. Palmer and J. W. S. Rickett

Table III. Perforations per operation according to grade of staff

Grade Rate Number

Surgeon
Consultant 1.71 41
Registrar 0.89 28
Senior House Officer 0.74 31
Total (all grades) I.18 100
Assistant
Consultant 0.67
Registrar 0.24 2;
Senior House Officer 0.33 42
House Officer 0.08 12
Total (all grades) 0.29 81
Nurse
Sister 0.59 32
Staff Nurse 0.65 43
Enrolled Nurse 0.63 16
Operating Department Assistant 0.50 2
Student Nurse 0.00 1
Total (all grades) 0.62 94

Table IV. Perforations detected by the subject: mechanisms of injury

Activity Surgeon Assistant Nurse

Wound closure 14 1 4
Instrument use in wound
Opening wound 6 : :
Manipulating needle i 0 2
Other instrument use 0 1 5
Drain insertion
Preparing patient : : 01
Injured by surgeon 3 1

middle finger (Figure 1). Perforations on the surgeon’s non-dominant index

finger were significantly more frequent than those on any other finger
(P<O*OS). Perforations in the gloves of the scrub nurse and assistant were
more evenly distributed (Figures 2 and 3). Perforations occurred over 10
times more often at the tips of the fingers than at the bases (PC 0.05).
In the control group of 100 unused gloves of varying sizes no perforations
were detected when these were tested in an identical manner.

Discussion
In this study perforation occurred in 43% of gloves. Church and
Sanderson’ described a rate of 11.5% in 130 pairs; Brough, Hunt and
Barrie” 37.5% in 339 pairs, O’Connor” 43% in 68 pairs, and Matta,
Thompson and Rainey’ 11% in 728 pairs of gloves. The type of surgery
 Surgical glove perforation 283

Non-dominant A

Figure 1. Distribution of glove perforations: surgeon’s palms n , > 20% of perforation; E$,
lo-20%; Ed <IO%. (Dorsa: <lO%.)

Non-dominant

Figure 2. Distribution of glove perforations: assistant’s palms. For key see Figure 1. (Dorsa:
<lo%.)
284 J. D. Palmer and J. W. S. Rickett

Figure 3. Distribution of glove perforations: nurse’s palms. For key see Figure 1. (Dorsa:
<IO%.)

greatly affects the perforation rate. The risk to the surgeon is proportional to
the number of perforations per operation (1 a18 in our investigation). In one
operation seven perforations were detected in the surgeon’s gloves.
Not all perforations are associated with needlestick injury. Such injuries
are grossly under-reported; indeed during the study period not one
needlestick injury was reported. Hussain, Latif and Choudhary’* described
a 5.6% incidence of intra-operative injury in a series of 2016 operations of
all types; 95% of these incidents were due to needlestick. We found that in
13 % of our operations surgeons had skin lesions such as cuts and abrasions
preoperatively which might provide a route of infection without needlestick
injury. In 42% of incidents the person was aware of the glove injury and in
20% there was evidence of superficial injury to the skin that had caused
bleeding or that could be made to bleed.
A general surgeon may perform 1’8 000 operations during his lifetime (10
intermediate and major operations per week during 35 years). The general
population carrier rate for HBV e-antigen is 0.1%13 so the surgeon may
operate on 18 such patients in his lifetime. At an average 1.18 perforations
per operation he will receive 21 glove perforations while operating on HBV
carriers. We estimate that 25% of perforations are associated with a
superficial needlestick injury or pre-existi?ng dermal damage. So the surgeon
may have five superficial injuries whilst operating on this group of high-risk
patients. At an estimated 3 5 % risk of transmission from a single needlestick
injury,14 when operating on HBV e-positike carriers, the unimmunized
surgeon risks l-75 HBV infections per lifetime.
 Surgical glove perforation 285

Indeed, the rate of new cases of HBV among surgeons is four times that of
the general population, 3 but most infections are subclinical and unreported.
In a study of 507 District General Hospital employees 10 of 29 junior
doctors had evidence of HBV exposure.” In a sample of 224 Danish
surgeons 23% were also found to have serological evidence of infection.”
HBV infection has a significant risk of chronic active hepatitis and remains a
worrying occupational hazard for the surgeon, for the dangers to his health,
and, because if he becomes HBV e-positive he will not be able to continue to
operate.
HIV can be transmitted by a superficial needlestick injury17 but the
infectivity rate is 100 times lower than that for HBV: a single needlestick
injury of any type has an estimated 0.35% risk of seroconversion.” The
majority of seroconversion incidents in health care workers has been
associated with injury from hollow needles with inoculation. In the United
Kingdom 29 of 800 000 first-time blood donors were found to be HIV
positive, equivalent to 0*0036% of the general population.” Using this
incidence and the same earlier risks of injury we calculate that the lifetime
risk of HIV infection is 0*00067. As this rate over the next 35 years one in
1492 surgeons may be infected with HIV, without calculated consideration
of the increase in reported cases of HIV or the observation that patients with
HIV are more likely than the general population to require surgery.
HBV vaccine has eliminated infection in many high risk settings2” but as
infection can occur despite prophylaxis, 3 identification of carriers and
prevention of transmission remain important. The presence of HBV surface
antigen is associated with HIV carriage. At least 26 health care workers
worldwide have been infected with HIV.21,22 Double gloving has been
shown to decrease perforation rate but is poorly tolerated.6 We have shown
that the non-dominant index and middle fingers require special protection.
There is a need for a special glove with increased thickness in the high risk
areas leaving the sensitive dominant fingers free. An alternative is the use of
a thimble23 or similar device to protect the finger tips at risk.
The most dangerous part of the operation for needlestick injury is the
wound closure. We recommend the use of stapling devices where possible in
high risk patients. Hands are frequently used by general surgeons in the
‘deep’ group of operations. It is essential that a ‘no touch technique’ be
adopted in surgical practice.
Defining ‘at-risk’ groups for surgery would require screening of all
patients for HBV surface antigen, and an extensive social and sexual history,
or screening for HIV antibodies with counselling of patients. Such
screening is impractical in most general surgical settings with a high
proportion of emergency work. It is therefore prudent to treat every patient
as ‘at-risk’ and use all the techniques available to reduce disease
transmission.
286 J. D. Palmer and J. W. S. Rickett

References
1. Rimland D, Parkin WE, Grayson PH, Miller B, Schrack WD. Hepatitis B outbreak
traced to an oral surgeon. N Engl J Med 1977; 296: 953-958.
2. Dalgleish AG, Malkovsky M. Surgical gloves as a mechanical barrier against human
immunodeficiency viruses. BrJ Surg 1988; 75: 171-172.
3. Polakoff S. Acute viral hepatitis B: laboratory reports 1980-84. Br Med J 1986; 293:
37-38.
4. Anon. Collaborative study. Acute hepatitis B associated with gynaecological surgery.
Lancet 1980; i: l-6.
5. Anon. Needlestick transmission of HTLV-III from a patient infected in Africa. Lancet
1984; 2: 1376-1377.
6. Matta H, Thompson AM, Rainey JB. Does wearing two pairs of gloves protect
operating theatre staff from skin contamination? Br MedJ 1988; 297: 597-598.
7. Carl M, Francis DP, Blakey DL, Maynard JE. Interruption of hepatitis B transmission
by modification of a gynaecologist’s surgical technique. Lancet 1982; 1: 731-733.
8. Barker LF, Murray R. Relationship of virus dose to incubation time of clinical hepatitis
and time of appearance of hepatitis-associated antigen. AmJ Med Sci 1972; 263: 27-33.
Church J, Sanderson P. Surgical glove punctures. J Hasp Infect 1980; 1: 84.
1:: Brough SA, Hunt TM, Barrie WW. Surgical glove perforations. Br J Surg 1988; 75:
317.
11. O’Connor AG. Glove puncture during operation. Nurs Times 1984; 8O(Suppl.) 5-6.
12. Hussain SA, Latif ABA, Choudhary AAAA. Risk to surgeons: a survey of accidental
injuries during operations. Br J Surg 1988; 75: 316316.
13. Finch R. Time for action on hepatitis B immunisation. Br Medr 1987; 294: 197-198.
14. Bradbeer C. AIDS-epidemiology and screening. Med Int 1986; 2: 1241-1246.
15. Abbas AMA, Denton MD, Francis RA. Prevalence of hepatitis B markers among
district general hospital staff. Br Med J 1985; 290: 1212.
16. Hardt F, Aldershville J, Dietrichson 0. et al. Hepatitis B virus infection among Danish
surgeons. J Infect Dis 1979; 140: 972-974.
17. Oskendler E, Harzic M, Le Roux J-M, Rabian C, Clauvel JP. HIV infection with
seroconversion after a superficial needlestick injury to the finger. N Engl J Med 1986;
315: 582.
18. Wormser GP, Rabkin CS, Joline C. Frequency of nosocomial transmission of HIV
infections among health care workers. N Engl J Med 1988; 319: 307-308.
19. Gunson HH. HIV infection and the blood supply. Royal Society of Medicine-The
AIDS Letter 1988; ,8: l-3.
20. Polakoff S. Hepatitis in dialysis units in the UK. r Hyg 1981; 87: 4433451.
21. AIDS: HIV Infected Health Care Workers. DHSS Report of the recommendations of the
expert advisory group on AIDS. London: Department of Health and Social Security
1988.
22. Waste11 C. Aids and the surgeon: an update. In: Risks and Complications. The Patient and
Surgeon in Theatre. Oxford: The Medicine Group (UK) Ltd 1989; 9-10.
23. Miles AJG, Waste11 C, Allen-Mersh TG. Protection for the left index finger whilst
operating on HIV positive patients. Ann R Co11 Surg Engl 1989; 71: 225.