Downloaded from www.ajronline.org by 177.236.114.215 on 11/08/15 from IP address 177.236.114.215. Copyright ARRS. For personal use only; all rights reserved
131
Occult
Glove
Frequency Procedures
Margaret Donald George
E. Hansen1 D. Mclntire2 L. Miller 1111
Southwestern
Medical
Center,
Medical
Center,
0361 -803X/92/1 591-0131 C American
Roentgen
of Texas
5323 Harry Hines Address reprint re-
Blvd., Dallas, TX 75235-9071. quests to M. E. Hansen. 2 Computing Services, Texas Southwestern 75235-9071.
after revi-
Ray Society
University
of
Dallas, TX
During
lnterventional
Radiologic
Because the intact surgical glove is an important barrier against exposure to blood and other body fluids, occult glove perforations occurring during procedures may have implications for infection control and protection of health care workers. Although the frequency of occult glove perforations during surgery has been studied, no such data have been reported for interventional radiology. This study was designed to determine the frequency of occult glove perforations in interventional radiology. Gloves used during interventional radiologic procedures were collected and tested for occult perforations according to previously accepted methods. The frequency of occult defects in unused gloves was also determined. A form completed for each pair of gloves indicated length of use, awareness of perforations, and information about the procedure. The anatomic distribution of perforations was also studied. Occult perforations were found in 49 (10%) of 492 used gloves. The frequency of defects in unused gloves was 1% (2/ 160). Gloves worn fewer than 2 hr had a perforation rate of 7% (29/406), and those worn 2 hr or longer had a rate of 23% (20/86; p < .0001). Perforations were noticed by the wearer in only three (6%) of the 49 perforations. The perforation rate for inner gloves of a double-glove set was 3% (1/34; the corresponding outer glove also was perforated). Although the 10% rate of occult glove perforation during interventional radiologic procedures found in this series is lower than rates described with surgical procedures, it is not insignificant. The significant increase in the rate of perforation when gloves are worn more than 2 hr suggests that gloves should routinely be changed at or before 2 hr of wear, particularly in procedures involving high-risk patients. AJR
Received January 7, 1992; accepted sion February 13, 1992. Department of Radiology, University
Perforations:
159:131-135,
July 1992
Concern about transmission of infectious agents between health care workers and patients is rising, particularly with regard to HIV. This is due in part to widespread publicity surrounding a case in which an HIV-positive dentist is thought to have transmitted HIV to five of his patients [1]. Less well publicized is the number of health care workers who may have acquired HIV infection from occupational exposure, perhaps because in some such cases it is difficult to document how or when infection occurred [2-7]. Although the risk of occupational HIV infection is quite low, it is not zero. The overall risk of occupational hepatitis B virus transmission is of similar concern [8-10]. The Centers for Disease Control (CDC) have recommended the use of universal precautions for all health care workers whenever they may be exposed to patients’ blood or other body fluids [1 1 12]. Interventional radiohogic procedures probably pose little risk of physician to patient transmission of infection, but could result in physicians being exposed to infectious agents. Universal precautions as endorsed by the CDC include the use of gloves, masks, goggles, and impermeable gowns where appropriate. Surgical gloves are certainly the most universally used of these barriers in interventional radiology, serving to maintain asepsis and to protect both patient and operator from transmission of infectious agents. The frequency of glove perforationsoccurring during interventionalradiologicprocedures is therefore im,
132
HANSEN
portant.
Although the frequency of occult glove perforations during surgical and dental procedures has been studied [1 3-22], no such data have been reported for interventional radiology. This study was designed to determine the frequency of occult glove perforations during interventional radiologic procedures.
Downloaded from www.ajronline.org by 177.236.114.215 on 11/08/15 from IP address 177.236.114.215. Copyright ARRS. For personal use only; all rights reserved
occurring
ET AL.
AJR:159,
the second finger and one on the third finger. Used gloves were designated as “single” when only one glove was worn on each hand. “Double-outer” and “double-inner” designated outer and inner gloves when two gloves were worn on each hand. Single and double-outer gloves were considered as one group; double-inner gloves were considered separately. Forty-nine double-outer
Materials
and Methods
Surgical
gloves
used
during
interventional radiology procedures tJtraderm (brand A) and Triflex Healthcare Corp., Valencia, CA) and Spectra
were collected over a 4-week period. (brand
B) gloves
gloves
(brand
(Baxter
C) (Becton-Dickinson,
Franklin Lakes, NJ) were used.
Each pair of gloves was accompanied by a completed form indicating the length of time worn; the wearer’s awareness of leaks, perforations, or needle sticks; the type of procedure being done; the wearer’s experience level and dominant hand; and a subjective estimation of the patient’s risk for HIV infection. Gloves were inspected for gross perforations and then were filled with water to check for occult perforations (Fig. 1) according to the method described by Skaug [1 3]. One hundred sixty unused gloves (brand A), taken from the supply of surgical gloves available in the interventional radiology procedure rooms, were evaluated in the same manner. Data were analyzed by using the SAS statistical package (SAS Institute
Inc.,
Cary,
NC).
The
chi-squared
test
was
used
for
data
comparing perforation rates across categories. In some cases, such as the comparison across risk categories, Fisher’s exact test was used because the chi-squared approximation was not optimal for these small sample sizes.
Results Two occult perforations were found in 160 unused gloves (perforation rate, 1%). Both were in the right glove, one on
Fig. 1.-Method of glove testing. Holes are shown by fine jets of water (arrow) spraying from glove.
Fig. 2.-Drawing perforations were
July 1992
perforations (1 0%) were found in 492 single or gloves. Two were grossly visible before filling
the gloves with water; neither perforation was recognized by the wearer. Forty-seven perforations were detected only after the gloves were filled with water. Only one glove had more than one perforation; in this case, only one of the two perforations was tabulated. One perforation was found in 34 double-inner gloves (3%); the corresponding outer glove also had a perforation, and the surface of the inner glove was contaminated with blood. No perforations
were
found
in double-inner
gloves
when
the
corresponding outer gloves were intact (0/33). Only three of the 49 perforations were recognized by the wearers. In one case, a perforation resulted from nicking the heft thumb with a scalpel blade. No bleeding occurred, and the blade was discarded without further use. No specific cause could be identified for the other two recognized perforations. In both cases, the wearers noticed fluid leaking through the glove onto their skin during the procedure or after removal of the gloves. The anatomic distribution of glove perforations is shown in Figure 2. Dominant and nondominant hands were equally involved (53% and 47%, respectively; p = .668 [chi squared]). Gloves were grouped by length of time worn, and perforation rates were calculated for each group (Fig. 3): eight (6%) of 130 gloves in the 0-29 mm group had perforations, nine (11%) of 86 gloves in the 30-59 mm group had perforations,
shows anatomic in palm area.
distribution
of perforations.
Palmar
surface
is shown,
as most
AJR:159,
SURGICAL
July 1992
GLOVE
PERFORATIONS
cases. The frequency of perforations was 10% in both groups (22/21 8 and 27/274, respectively). No significant difference was found between these groups (p = .930, chi squared). Patients were considered at high risk for HIV infection in six cases, at uncertain risk in 1 36 cases, and at low risk in 1 04 cases. The frequencies of perforations in these three groups were 8% (1/1 2), 8% (22/272), and 13% (27/208),
z
Downloaded from www.ajronline.org by 177.236.114.215 on 11/08/15 from IP address 177.236.114.215. Copyright ARRS. For personal use only; all rights reserved
0
0 U. UJ 0.
respectively; I-.
z
UJ C.) UJ
a.
TIME WORN (MIN)
Fig. 3.-Graph shows relationship between rations and length of time gloves were worn.
frequency
of glove
perle-
six (6%) of 98 gloves in the 60-89 mm group had perforations, six (7%) of 92 gloves in the 90-1 19 mm group had perforations, 13 (22%) of 60 gloves in the 1 20-1 49 mm group had perforations, two (20%) of 10 gloves in the 1 50-1 79 mm group had perforations, and five (36%) of 14 gloves in the 180-209 mm or more group had perforations. No perforations were found in two gloves worn for more than 21 0 mm. Perforation rates increased significantly (p < .0001)for gloves worn 2 hr or more. The number of gloves worn 3 hr or more was too small to determine whether the perforation rate in this group (5/1 6, 31 %; p = .402 [chi squared]) was significantly higher than the rate for the 2-3 hr group. No significant difference was found in perforation rates for different types of procedures, but numbers in many of these subgroups were small. Twelve perforations (9%) were found in 1 36 gloves used in neuroangiography, 27 perforations (1 1 %) were found in 242 gloves used in visceral angiography, and 10 perforations (9%) interventional
133
IN RADIOLOGY
were found procedures.
in 1 1 4 gloves No significant
used in nonvascuhar difference was found
in perforation rates for different levels of wearer experience, but numbers in several of these subgroups were small. No perforations were found in 1 0 gloves worn by technologists, 30 perforations (1 0%) were found in 300 gloves worn by junior residents, 15 perforations (1 4%) were found in 108 gloves worn by senior residents, three perforations (6.5%) were found in 46 gloves worn by fellows, and one perforation (4%) was found
in 28 gloves
worn
by faculty
members.
In five cases, both right and heft gloves of a pair had perforations. The duration of the procedure in these cases ranged from 5 to 180 mm (mean, 1 14; median, 130). In four of these cases, the wearers commented on the survey form on the difficulty of the procedure. In the fifth case, the gloves were changed after 5 mm because the wearer nicked one finger with a scalpel blade. Of the 492 single or double-outer gloves, 218 (44%) were used in emergency procedures and 274 (56%) in elective
the
differences
among
the
groups
were
not
significant (p = .1 77, Fisher’s exact test). Operators chose to wear double gloves in three (50%) of six high-risk cases, 12 (9%) of 1 36 uncertain-risk cases, and two (2%) of 104 low-risk cases. Most of the used gloves were brand A (444, 90%); few B (36, 7%) and C (1 2, 3%) brand gloves were used. Glove selection was based on the wearer’s preference in most cases; very large (size 8.5) or very small (size 6 or smaller) gloves were only stocked in brand B. The frequencies of perforations were 9% (42/444), 14% (5/36), and 17% (2/12) for brands A-C, respectively. These numbers are too small to show statistical significance (p = .356, Fisher’s exact test; the probability of detecting a difference of three percentage points between the perforation rates for gloves B and C is only .08 with this sample size).
Discussion The overall frequency of surgical glove perforations found in this study is lower than rates previously reported for surgical and dental procedures, which have ranged from 11% to 37.5% [1 3-23]. Interventional radiohogic procedures involve much less handling of sharp instruments than surgical or dental procedures do, so this finding was not unexpected. Moreover, when sharp instruments are used during interventional radiologic procedures, both of the operator’s hands are generally in full view rather than positioned within a body cavity, where they may be at greater risk of injury. The occult perforation rate of 1 0% may be greater than many intervenradiologists would expect, however, given the relatively
tional
low frequency of reported needle sticks and other recognized injuries during interventional radiologic procedures. An explanation for these occult glove injuries may lie in the types of stresses placed on gloves during interventional radiologic procedures. Pinching glove material between catheter hub and the power-injector connector is a common occurrence, as is catching a glove on a stopcock. Handling of guidewires, some of which may have sharp back ends, may also head to occult perforations. Forceful hand injection of saline or contrast material may contribute to glove damage as well. Some authors [1 7, 24] have advocated routine double gloving as a means of reducing the risk of skin contamination during surgical procedures. The number of double-glove sets in our series was too small to assess the value of this approach for interventional radiologic procedures. In addition, only one double-outer perforation occurred in our series, and the corresponding inner glove was also perforated. In this case, the perforated inner glove was contaminated with blood on the outside, indicating that it did not protect against skin contamination. The how frequency of perforations in unused
Downloaded from www.ajronline.org by 177.236.114.215 on 11/08/15 from IP address 177.236.114.215. Copyright ARRS. For personal use only; all rights reserved
134
HANSEN
gloves (1 %) indicates that routine double gloving is not warranted, although this precaution may be indicated if the wearer has a recent cut, abrasion, or other area of nonintact skin on one or both hands, or if the procedure involves a high-risk patient. In our series, operators wore double or single pairs of gloves according to personal preference. Some indicated that they routinely double gloved, even for patients estimated to be at how risk for HIV infection; others used double gloving for high-risk patients only. Most of our cases were estimated to be of uncertain risk (55%); relatively few were considered high risk (2%). The anatomic distribution of perforations in our series, with almost equal involvement of dominant and nondominant hands, differs from that reported previously [6, 16, 21 22]. Surgical perforations are more common on the left-hand gloves of right-handed surgeons; the heft index finger is the most common site. This may be due to suturing techniques in which the heft hand is used to palpate or retract tissue while suture material is passed with the right, or the needle is grasped with the fingers of the left hand after passing suture material through the tissue. In interventional radiohogic procedures, both heft and right hands are used about equally in many maneuvers, and should be equally at risk for glove perforations. Two obvious exceptions are administration of local anesthetics and arterial (or venous) puncture; both are maneuvers in which one hand may be used to palpate the pulse while a needle is advanced with the other [25]. The increased number of perforations in gloves worn 2 hr or longer is in agreement with previous surgical reports, which found higher rates of glove perforation or accidental injury during operations that lasted longer than 1 hr [21 23]. Interventional radiohogic procedures that last 2 hr or more are often complex or difficult. Several catheter and guidewire manipulations or exchanges may be necessary. A larger number of angiographic runs may need to be performed, with more frequent connections of the catheter to the power injector. Similarly, several hand injections may be required to follow progress or determine the position of a catheter during such a case. All of these factors could contribute to the increased frequency of perforations. Only three of the 49 perforations in our series were recognized by the wearer of the gloves. This agrees with previous studies [1 4, 16, 1 8, 21 ]. Numbers in our series were too small to determine if different types of interventional radiohogic procedures carry greater or lesser risk of glove perforations. However, the perforation rates for neuroangiography, visceral angiography, and nonvascular interventional procedures were ,
,
fairly similar.
Operator experience may be a factor in the frequency of glove perforations, as suggested by surgical studies [1 3, 19, 23] in which more experienced operators incurred fewer glove perforations
or accidental
injuries
than
inexperienced
opera-
tors did. Numbers in several of our subgroups are too small to show significant differences, but they do suggest a trend in this direction. However, other factors must be considered. Most interventional procedures at our institution are performed by residents, who are supervised by fellows and faculty members. Direct participation by fellows or faculty members is necessary in a minority of cases, thus accounting
ET AL.
AJR:159,
July 1992
for the small number of used gloves from these two groups. When fellows or faculty members become directly involved, it is often for only a portion of the procedure, and they may change gloves more frequently as other tasks require attention. Thus, fellows and faculty members rarely wear a single pair of gloves for 2 hr or longer, and their gloves may not be exposed to the same stresses as those worn by residents. In addition, fellows or faculty members rarely perform catheterinjector hookups, arterial punctures, or suturing; these tasks are most often done by residents. This too may account for a reduced frequency of glove perforations for more experienced operators in our series. The lack of a significant difference in perforation rates for emergency and elective interventional radiologic procedures agrees with the results of surgical studies, in which differences between these groups were marginal or nonexistent [1 6, 23]. The number of brand B or C gloves in our series was too small for us to detect significant differences in perforation rates among the three brands, all of which meet current standards for surgical gloves [26]. Further study of various glove types would be necessary to determine whether significant differences exist. It is important to emphasize that the overall rate of glove perforation during interventional radiologic procedures is low compared with the rate for surgical procedures. It is of some concern, however, that most perforations were unrecognized by the wearer of the glove, raising the possibility of prolonged skin exposure to blood or other body fluids. Although interventional radiohogic procedures do pose some risk of occupational transmission of HIV, this risk is probably quite low. Studies of surgeons and dentists have shown very low occupational risk of infection [6, 7], and risk to interventional radiologists is unlikely to be significantly higher. The risk of occupational C) is greater
transmission of hepatitis B virus (and hepatitis for both patients and health care workers [8-
10]. It may be prudent, especially when high-risk patients are involved, to change gloves at or before 2 hr of wear, even if the wearer is not aware of any leaks or perforations. The number of glove perforations due to needle sticks or other sharp injuries is low in interventional radiologic procedures, but all efforts should be made to reduce such injuries. Interventional
radiologic
procedures
do
not
generally
in-
volve large open incisions or work within a confined body cavity, and exposure to blood or other body fluids is usually minimal. Opportunity for contamination of the patient’s wound with the operator’s blood following injury is likewise small. In addition, the rate of sharp injuries in interventionah radiologic procedures is low. In our series, only one injury occurred during 21 9 person-procedures (case exposure rate, 0.5%; 95% confidence interval, 0.01 -2.5%). This is much lower than the surgical injury rate, which can be as high as 21 % (Tokars J et al., presented at the VII International Conference on AIDS, June 1 991). In no case from our series was a patient known to be exposed to the blood or other body fluid of an operator, either directly or via a contaminated instrument. Data from other institutions would be useful to confirm the safety of interventional radiologic procedures for both operators and patients, so that a rational approach to infection control in interventional radiology may be taken.
AJR:159,
SURGICAL
July 1992
GLOVE
PERFORATIONS
1 1 . Recommendations
ACKNOWLEDGMENTS We thank collection.
our residents
and fellows
for help with
glove
and data
1 2.
Downloaded from www.ajronline.org by 177.236.114.215 on 11/08/15 from IP address 177.236.114.215. Copyright ARRS. For personal use only; all rights reserved
1 3.
14.
REFERENCES 1 . Update: transmission dure-Florida. MMWR
2. Update: acquired ciency
virus
IN RADIOLOGY
of HIV infection 1991;40:21-27,
proce-
1 5.
and human immunodefiMMWR 1988;37:
1 6.
JE, Gerberding J. Occupational infection with human virus (HIV): risks and risk reduction. Ann Intern Med
17.
immunodeficiency
infection
among
during 33
an invasive
syndrome
health-care
dental
workers.
229-239 3. Becker CE, Cone immunodeficiency
for prevention
of HIV transmission
in health-care
set-
tings. MMWR 1987;36(S2): 1S-18S Update: universal precautions for prevention of transmission of human immunodeficiency virus, hepatitis B virus, and other bloodborne pathogens in health-care settings. MMWR 1988;37:377-382, 387-388 Skaug N. Micropunctures of rubber gloves used in oral surgery. Int J Oral Maxillofac Surg 1976;5:220-225 Church J, Sanderson P. Surgical glove punctures (letter). J Hosp Infect 1980;1 :84 Brough Si, Hunt TM, Barrie WW. Surgical glove perforations. Br J Surg 1988;75:31 7 Matta H, Thompson AM, Rainey JB. Does wearing two pairs of gloves protect operating theatre staff from skin contamination? BMJ
1988;297:597-598
1989;110:653-656 4. Beekmann SE, Fahey BJ, Gerberding JL, Henderson DK. Risky business: using necessarily imprecise casualty counts to estimate occupational risks for HIV-1 infection. Infect Control Hosp Epidemiol 1990;11:371-379 5. Marcus R, CDC Cooperative Needlestick Surveillance Group. Surveillance of health care workers exposed to blood from patients infected with the human immunodeficiency virus. N EngI J Med 1988;319: 1118-1123 6. Lowenfels AB, Wormser GP, Jam R. Frequency of puncture injuries in surgeons and estimated risk of HIV infection. Arch Surg 1989;1 14:
1284-1286 7. Klein RS, Phelan JA, Freeman K, et al. Low occupational risk of human immunodeficiency virus infection among dental professionals. N EngI J Med 1988;318:86-90 8. Weber DJ, Hoffmann KK, Rutala WA. Management of the healthcare worker infected with human immunodeficiency virus: lessons from nosecomial transmission of hepatitis B virus. Infect Control Hosp Epidemiol
1991;12:625-63O
M, Tilzey AJ, Noah ND, Banatvala
infections after gynaecologic
JL, Littell
C, Tarkington
of surgical personnel
A, Brown
to patients’
A, Schecter
WP.
Risk
of
blood during surgery at San
Francisco General Hospital. N EngI J Med 1990;322: 1788-1 793 1 8. Devenish EA, Miles AA. Control of Staphylococcus aureus in an operatingtheatre. Lancet 1939;1 : 1088-1 094 19. Penikett EJK, Gorrill RH. The integrity of surgical gloves tested during use. Lancet 1958;2: 1042-1 043
20. Shouldice
EE, Martin CJ. Wound infections, surgical gloves and hands of Can Med Assoc J 1959;81 :636-640 21 . Russell RT, Roque FE, Miller FA. A new method for detection of the leaky glove. Arch Surg 1966;93:245-249 22. McCue SF, Berg EW, Saunders EA. Efficacy of double-gloving as a bamer operating
23. 24.
viral hepatitis,
AD. Hepatitis B surgery. Lancet 1989;1 :205-207
Gerberding
exposure
25.
9. Snydman DR, Bryan JA, Dixon RE. Prevention of nosocomial type B (hepatitis B). Ann Intern Med 1975;83:838-845
10. Welch M, Webster
135
26.
personnel.
to microbial contamination during total joint arthroplasty. J Bone Joint Surg [Am] 1981;63-A:811-813 Hussein SA, Latif ABA, Choudhary AAAA. Risk to surgeons: a survey of accidental injuries during operations. Br J Surg 1988;75 :314-316 Gerberding JL, Schecter WP. Surgery and AIDS: reducing the risk (editorial). JAMA 1991;265:1572-1573 Williams DM, Marx MV, Korobkin M. AIDS risk and risk reduction in the radiology department (commentary). AJR 1991;1 57:919-921 Standard specification for rubber surgical gloves (D3577). In: Annual book of ASTM standards, vol. 9.02. Philadelphia: American Society for Testing and Materials, 1991:344-346
131
Occult
Glove
Frequency Procedures
Margaret Donald George
E. Hansen1 D. Mclntire2 L. Miller 1111
Southwestern
Medical
Center,
Medical
Center,
0361 -803X/92/1 591-0131 C American
Roentgen
of Texas
5323 Harry Hines Address reprint re-
Blvd., Dallas, TX 75235-9071. quests to M. E. Hansen. 2 Computing Services, Texas Southwestern 75235-9071.
after revi-
Ray Society
University
of
Dallas, TX
During
lnterventional
Radiologic
Because the intact surgical glove is an important barrier against exposure to blood and other body fluids, occult glove perforations occurring during procedures may have implications for infection control and protection of health care workers. Although the frequency of occult glove perforations during surgery has been studied, no such data have been reported for interventional radiology. This study was designed to determine the frequency of occult glove perforations in interventional radiology. Gloves used during interventional radiologic procedures were collected and tested for occult perforations according to previously accepted methods. The frequency of occult defects in unused gloves was also determined. A form completed for each pair of gloves indicated length of use, awareness of perforations, and information about the procedure. The anatomic distribution of perforations was also studied. Occult perforations were found in 49 (10%) of 492 used gloves. The frequency of defects in unused gloves was 1% (2/ 160). Gloves worn fewer than 2 hr had a perforation rate of 7% (29/406), and those worn 2 hr or longer had a rate of 23% (20/86; p < .0001). Perforations were noticed by the wearer in only three (6%) of the 49 perforations. The perforation rate for inner gloves of a double-glove set was 3% (1/34; the corresponding outer glove also was perforated). Although the 10% rate of occult glove perforation during interventional radiologic procedures found in this series is lower than rates described with surgical procedures, it is not insignificant. The significant increase in the rate of perforation when gloves are worn more than 2 hr suggests that gloves should routinely be changed at or before 2 hr of wear, particularly in procedures involving high-risk patients. AJR
Received January 7, 1992; accepted sion February 13, 1992. Department of Radiology, University
Perforations:
159:131-135,
July 1992
Concern about transmission of infectious agents between health care workers and patients is rising, particularly with regard to HIV. This is due in part to widespread publicity surrounding a case in which an HIV-positive dentist is thought to have transmitted HIV to five of his patients [1]. Less well publicized is the number of health care workers who may have acquired HIV infection from occupational exposure, perhaps because in some such cases it is difficult to document how or when infection occurred [2-7]. Although the risk of occupational HIV infection is quite low, it is not zero. The overall risk of occupational hepatitis B virus transmission is of similar concern [8-10]. The Centers for Disease Control (CDC) have recommended the use of universal precautions for all health care workers whenever they may be exposed to patients’ blood or other body fluids [1 1 12]. Interventional radiohogic procedures probably pose little risk of physician to patient transmission of infection, but could result in physicians being exposed to infectious agents. Universal precautions as endorsed by the CDC include the use of gloves, masks, goggles, and impermeable gowns where appropriate. Surgical gloves are certainly the most universally used of these barriers in interventional radiology, serving to maintain asepsis and to protect both patient and operator from transmission of infectious agents. The frequency of glove perforationsoccurring during interventionalradiologicprocedures is therefore im,
132
HANSEN
portant.
Although the frequency of occult glove perforations during surgical and dental procedures has been studied [1 3-22], no such data have been reported for interventional radiology. This study was designed to determine the frequency of occult glove perforations during interventional radiologic procedures.
Downloaded from www.ajronline.org by 177.236.114.215 on 11/08/15 from IP address 177.236.114.215. Copyright ARRS. For personal use only; all rights reserved
occurring
ET AL.
AJR:159,
the second finger and one on the third finger. Used gloves were designated as “single” when only one glove was worn on each hand. “Double-outer” and “double-inner” designated outer and inner gloves when two gloves were worn on each hand. Single and double-outer gloves were considered as one group; double-inner gloves were considered separately. Forty-nine double-outer
Materials
and Methods
Surgical
gloves
used
during
interventional radiology procedures tJtraderm (brand A) and Triflex Healthcare Corp., Valencia, CA) and Spectra
were collected over a 4-week period. (brand
B) gloves
gloves
(brand
(Baxter
C) (Becton-Dickinson,
Franklin Lakes, NJ) were used.
Each pair of gloves was accompanied by a completed form indicating the length of time worn; the wearer’s awareness of leaks, perforations, or needle sticks; the type of procedure being done; the wearer’s experience level and dominant hand; and a subjective estimation of the patient’s risk for HIV infection. Gloves were inspected for gross perforations and then were filled with water to check for occult perforations (Fig. 1) according to the method described by Skaug [1 3]. One hundred sixty unused gloves (brand A), taken from the supply of surgical gloves available in the interventional radiology procedure rooms, were evaluated in the same manner. Data were analyzed by using the SAS statistical package (SAS Institute
Inc.,
Cary,
NC).
The
chi-squared
test
was
used
for
data
comparing perforation rates across categories. In some cases, such as the comparison across risk categories, Fisher’s exact test was used because the chi-squared approximation was not optimal for these small sample sizes.
Results Two occult perforations were found in 160 unused gloves (perforation rate, 1%). Both were in the right glove, one on
Fig. 1.-Method of glove testing. Holes are shown by fine jets of water (arrow) spraying from glove.
Fig. 2.-Drawing perforations were
July 1992
perforations (1 0%) were found in 492 single or gloves. Two were grossly visible before filling
the gloves with water; neither perforation was recognized by the wearer. Forty-seven perforations were detected only after the gloves were filled with water. Only one glove had more than one perforation; in this case, only one of the two perforations was tabulated. One perforation was found in 34 double-inner gloves (3%); the corresponding outer glove also had a perforation, and the surface of the inner glove was contaminated with blood. No perforations
were
found
in double-inner
gloves
when
the
corresponding outer gloves were intact (0/33). Only three of the 49 perforations were recognized by the wearers. In one case, a perforation resulted from nicking the heft thumb with a scalpel blade. No bleeding occurred, and the blade was discarded without further use. No specific cause could be identified for the other two recognized perforations. In both cases, the wearers noticed fluid leaking through the glove onto their skin during the procedure or after removal of the gloves. The anatomic distribution of glove perforations is shown in Figure 2. Dominant and nondominant hands were equally involved (53% and 47%, respectively; p = .668 [chi squared]). Gloves were grouped by length of time worn, and perforation rates were calculated for each group (Fig. 3): eight (6%) of 130 gloves in the 0-29 mm group had perforations, nine (11%) of 86 gloves in the 30-59 mm group had perforations,
shows anatomic in palm area.
distribution
of perforations.
Palmar
surface
is shown,
as most
AJR:159,
SURGICAL
July 1992
GLOVE
PERFORATIONS
cases. The frequency of perforations was 10% in both groups (22/21 8 and 27/274, respectively). No significant difference was found between these groups (p = .930, chi squared). Patients were considered at high risk for HIV infection in six cases, at uncertain risk in 1 36 cases, and at low risk in 1 04 cases. The frequencies of perforations in these three groups were 8% (1/1 2), 8% (22/272), and 13% (27/208),
z
Downloaded from www.ajronline.org by 177.236.114.215 on 11/08/15 from IP address 177.236.114.215. Copyright ARRS. For personal use only; all rights reserved
0
0 U. UJ 0.
respectively; I-.
z
UJ C.) UJ
a.
TIME WORN (MIN)
Fig. 3.-Graph shows relationship between rations and length of time gloves were worn.
frequency
of glove
perle-
six (6%) of 98 gloves in the 60-89 mm group had perforations, six (7%) of 92 gloves in the 90-1 19 mm group had perforations, 13 (22%) of 60 gloves in the 1 20-1 49 mm group had perforations, two (20%) of 10 gloves in the 1 50-1 79 mm group had perforations, and five (36%) of 14 gloves in the 180-209 mm or more group had perforations. No perforations were found in two gloves worn for more than 21 0 mm. Perforation rates increased significantly (p < .0001)for gloves worn 2 hr or more. The number of gloves worn 3 hr or more was too small to determine whether the perforation rate in this group (5/1 6, 31 %; p = .402 [chi squared]) was significantly higher than the rate for the 2-3 hr group. No significant difference was found in perforation rates for different types of procedures, but numbers in many of these subgroups were small. Twelve perforations (9%) were found in 1 36 gloves used in neuroangiography, 27 perforations (1 1 %) were found in 242 gloves used in visceral angiography, and 10 perforations (9%) interventional
133
IN RADIOLOGY
were found procedures.
in 1 1 4 gloves No significant
used in nonvascuhar difference was found
in perforation rates for different levels of wearer experience, but numbers in several of these subgroups were small. No perforations were found in 1 0 gloves worn by technologists, 30 perforations (1 0%) were found in 300 gloves worn by junior residents, 15 perforations (1 4%) were found in 108 gloves worn by senior residents, three perforations (6.5%) were found in 46 gloves worn by fellows, and one perforation (4%) was found
in 28 gloves
worn
by faculty
members.
In five cases, both right and heft gloves of a pair had perforations. The duration of the procedure in these cases ranged from 5 to 180 mm (mean, 1 14; median, 130). In four of these cases, the wearers commented on the survey form on the difficulty of the procedure. In the fifth case, the gloves were changed after 5 mm because the wearer nicked one finger with a scalpel blade. Of the 492 single or double-outer gloves, 218 (44%) were used in emergency procedures and 274 (56%) in elective
the
differences
among
the
groups
were
not
significant (p = .1 77, Fisher’s exact test). Operators chose to wear double gloves in three (50%) of six high-risk cases, 12 (9%) of 1 36 uncertain-risk cases, and two (2%) of 104 low-risk cases. Most of the used gloves were brand A (444, 90%); few B (36, 7%) and C (1 2, 3%) brand gloves were used. Glove selection was based on the wearer’s preference in most cases; very large (size 8.5) or very small (size 6 or smaller) gloves were only stocked in brand B. The frequencies of perforations were 9% (42/444), 14% (5/36), and 17% (2/12) for brands A-C, respectively. These numbers are too small to show statistical significance (p = .356, Fisher’s exact test; the probability of detecting a difference of three percentage points between the perforation rates for gloves B and C is only .08 with this sample size).
Discussion The overall frequency of surgical glove perforations found in this study is lower than rates previously reported for surgical and dental procedures, which have ranged from 11% to 37.5% [1 3-23]. Interventional radiohogic procedures involve much less handling of sharp instruments than surgical or dental procedures do, so this finding was not unexpected. Moreover, when sharp instruments are used during interventional radiologic procedures, both of the operator’s hands are generally in full view rather than positioned within a body cavity, where they may be at greater risk of injury. The occult perforation rate of 1 0% may be greater than many intervenradiologists would expect, however, given the relatively
tional
low frequency of reported needle sticks and other recognized injuries during interventional radiologic procedures. An explanation for these occult glove injuries may lie in the types of stresses placed on gloves during interventional radiologic procedures. Pinching glove material between catheter hub and the power-injector connector is a common occurrence, as is catching a glove on a stopcock. Handling of guidewires, some of which may have sharp back ends, may also head to occult perforations. Forceful hand injection of saline or contrast material may contribute to glove damage as well. Some authors [1 7, 24] have advocated routine double gloving as a means of reducing the risk of skin contamination during surgical procedures. The number of double-glove sets in our series was too small to assess the value of this approach for interventional radiologic procedures. In addition, only one double-outer perforation occurred in our series, and the corresponding inner glove was also perforated. In this case, the perforated inner glove was contaminated with blood on the outside, indicating that it did not protect against skin contamination. The how frequency of perforations in unused
Downloaded from www.ajronline.org by 177.236.114.215 on 11/08/15 from IP address 177.236.114.215. Copyright ARRS. For personal use only; all rights reserved
134
HANSEN
gloves (1 %) indicates that routine double gloving is not warranted, although this precaution may be indicated if the wearer has a recent cut, abrasion, or other area of nonintact skin on one or both hands, or if the procedure involves a high-risk patient. In our series, operators wore double or single pairs of gloves according to personal preference. Some indicated that they routinely double gloved, even for patients estimated to be at how risk for HIV infection; others used double gloving for high-risk patients only. Most of our cases were estimated to be of uncertain risk (55%); relatively few were considered high risk (2%). The anatomic distribution of perforations in our series, with almost equal involvement of dominant and nondominant hands, differs from that reported previously [6, 16, 21 22]. Surgical perforations are more common on the left-hand gloves of right-handed surgeons; the heft index finger is the most common site. This may be due to suturing techniques in which the heft hand is used to palpate or retract tissue while suture material is passed with the right, or the needle is grasped with the fingers of the left hand after passing suture material through the tissue. In interventional radiohogic procedures, both heft and right hands are used about equally in many maneuvers, and should be equally at risk for glove perforations. Two obvious exceptions are administration of local anesthetics and arterial (or venous) puncture; both are maneuvers in which one hand may be used to palpate the pulse while a needle is advanced with the other [25]. The increased number of perforations in gloves worn 2 hr or longer is in agreement with previous surgical reports, which found higher rates of glove perforation or accidental injury during operations that lasted longer than 1 hr [21 23]. Interventional radiohogic procedures that last 2 hr or more are often complex or difficult. Several catheter and guidewire manipulations or exchanges may be necessary. A larger number of angiographic runs may need to be performed, with more frequent connections of the catheter to the power injector. Similarly, several hand injections may be required to follow progress or determine the position of a catheter during such a case. All of these factors could contribute to the increased frequency of perforations. Only three of the 49 perforations in our series were recognized by the wearer of the gloves. This agrees with previous studies [1 4, 16, 1 8, 21 ]. Numbers in our series were too small to determine if different types of interventional radiohogic procedures carry greater or lesser risk of glove perforations. However, the perforation rates for neuroangiography, visceral angiography, and nonvascular interventional procedures were ,
,
fairly similar.
Operator experience may be a factor in the frequency of glove perforations, as suggested by surgical studies [1 3, 19, 23] in which more experienced operators incurred fewer glove perforations
or accidental
injuries
than
inexperienced
opera-
tors did. Numbers in several of our subgroups are too small to show significant differences, but they do suggest a trend in this direction. However, other factors must be considered. Most interventional procedures at our institution are performed by residents, who are supervised by fellows and faculty members. Direct participation by fellows or faculty members is necessary in a minority of cases, thus accounting
ET AL.
AJR:159,
July 1992
for the small number of used gloves from these two groups. When fellows or faculty members become directly involved, it is often for only a portion of the procedure, and they may change gloves more frequently as other tasks require attention. Thus, fellows and faculty members rarely wear a single pair of gloves for 2 hr or longer, and their gloves may not be exposed to the same stresses as those worn by residents. In addition, fellows or faculty members rarely perform catheterinjector hookups, arterial punctures, or suturing; these tasks are most often done by residents. This too may account for a reduced frequency of glove perforations for more experienced operators in our series. The lack of a significant difference in perforation rates for emergency and elective interventional radiologic procedures agrees with the results of surgical studies, in which differences between these groups were marginal or nonexistent [1 6, 23]. The number of brand B or C gloves in our series was too small for us to detect significant differences in perforation rates among the three brands, all of which meet current standards for surgical gloves [26]. Further study of various glove types would be necessary to determine whether significant differences exist. It is important to emphasize that the overall rate of glove perforation during interventional radiologic procedures is low compared with the rate for surgical procedures. It is of some concern, however, that most perforations were unrecognized by the wearer of the glove, raising the possibility of prolonged skin exposure to blood or other body fluids. Although interventional radiohogic procedures do pose some risk of occupational transmission of HIV, this risk is probably quite low. Studies of surgeons and dentists have shown very low occupational risk of infection [6, 7], and risk to interventional radiologists is unlikely to be significantly higher. The risk of occupational C) is greater
transmission of hepatitis B virus (and hepatitis for both patients and health care workers [8-
10]. It may be prudent, especially when high-risk patients are involved, to change gloves at or before 2 hr of wear, even if the wearer is not aware of any leaks or perforations. The number of glove perforations due to needle sticks or other sharp injuries is low in interventional radiologic procedures, but all efforts should be made to reduce such injuries. Interventional
radiologic
procedures
do
not
generally
in-
volve large open incisions or work within a confined body cavity, and exposure to blood or other body fluids is usually minimal. Opportunity for contamination of the patient’s wound with the operator’s blood following injury is likewise small. In addition, the rate of sharp injuries in interventionah radiologic procedures is low. In our series, only one injury occurred during 21 9 person-procedures (case exposure rate, 0.5%; 95% confidence interval, 0.01 -2.5%). This is much lower than the surgical injury rate, which can be as high as 21 % (Tokars J et al., presented at the VII International Conference on AIDS, June 1 991). In no case from our series was a patient known to be exposed to the blood or other body fluid of an operator, either directly or via a contaminated instrument. Data from other institutions would be useful to confirm the safety of interventional radiologic procedures for both operators and patients, so that a rational approach to infection control in interventional radiology may be taken.
AJR:159,
SURGICAL
July 1992
GLOVE
PERFORATIONS
1 1 . Recommendations
ACKNOWLEDGMENTS We thank collection.
our residents
and fellows
for help with
glove
and data
1 2.
Downloaded from www.ajronline.org by 177.236.114.215 on 11/08/15 from IP address 177.236.114.215. Copyright ARRS. For personal use only; all rights reserved
1 3.
14.
REFERENCES 1 . Update: transmission dure-Florida. MMWR
2. Update: acquired ciency
virus
IN RADIOLOGY
of HIV infection 1991;40:21-27,
proce-
1 5.
and human immunodefiMMWR 1988;37:
1 6.
JE, Gerberding J. Occupational infection with human virus (HIV): risks and risk reduction. Ann Intern Med
17.
immunodeficiency
infection
among
during 33
an invasive
syndrome
health-care
dental
workers.
229-239 3. Becker CE, Cone immunodeficiency
for prevention
of HIV transmission
in health-care
set-
tings. MMWR 1987;36(S2): 1S-18S Update: universal precautions for prevention of transmission of human immunodeficiency virus, hepatitis B virus, and other bloodborne pathogens in health-care settings. MMWR 1988;37:377-382, 387-388 Skaug N. Micropunctures of rubber gloves used in oral surgery. Int J Oral Maxillofac Surg 1976;5:220-225 Church J, Sanderson P. Surgical glove punctures (letter). J Hosp Infect 1980;1 :84 Brough Si, Hunt TM, Barrie WW. Surgical glove perforations. Br J Surg 1988;75:31 7 Matta H, Thompson AM, Rainey JB. Does wearing two pairs of gloves protect operating theatre staff from skin contamination? BMJ
1988;297:597-598
1989;110:653-656 4. Beekmann SE, Fahey BJ, Gerberding JL, Henderson DK. Risky business: using necessarily imprecise casualty counts to estimate occupational risks for HIV-1 infection. Infect Control Hosp Epidemiol 1990;11:371-379 5. Marcus R, CDC Cooperative Needlestick Surveillance Group. Surveillance of health care workers exposed to blood from patients infected with the human immunodeficiency virus. N EngI J Med 1988;319: 1118-1123 6. Lowenfels AB, Wormser GP, Jam R. Frequency of puncture injuries in surgeons and estimated risk of HIV infection. Arch Surg 1989;1 14:
1284-1286 7. Klein RS, Phelan JA, Freeman K, et al. Low occupational risk of human immunodeficiency virus infection among dental professionals. N EngI J Med 1988;318:86-90 8. Weber DJ, Hoffmann KK, Rutala WA. Management of the healthcare worker infected with human immunodeficiency virus: lessons from nosecomial transmission of hepatitis B virus. Infect Control Hosp Epidemiol
1991;12:625-63O
M, Tilzey AJ, Noah ND, Banatvala
infections after gynaecologic
JL, Littell
C, Tarkington
of surgical personnel
A, Brown
to patients’
A, Schecter
WP.
Risk
of
blood during surgery at San
Francisco General Hospital. N EngI J Med 1990;322: 1788-1 793 1 8. Devenish EA, Miles AA. Control of Staphylococcus aureus in an operatingtheatre. Lancet 1939;1 : 1088-1 094 19. Penikett EJK, Gorrill RH. The integrity of surgical gloves tested during use. Lancet 1958;2: 1042-1 043
20. Shouldice
EE, Martin CJ. Wound infections, surgical gloves and hands of Can Med Assoc J 1959;81 :636-640 21 . Russell RT, Roque FE, Miller FA. A new method for detection of the leaky glove. Arch Surg 1966;93:245-249 22. McCue SF, Berg EW, Saunders EA. Efficacy of double-gloving as a bamer operating
23. 24.
viral hepatitis,
AD. Hepatitis B surgery. Lancet 1989;1 :205-207
Gerberding
exposure
25.
9. Snydman DR, Bryan JA, Dixon RE. Prevention of nosocomial type B (hepatitis B). Ann Intern Med 1975;83:838-845
10. Welch M, Webster
135
26.
personnel.
to microbial contamination during total joint arthroplasty. J Bone Joint Surg [Am] 1981;63-A:811-813 Hussein SA, Latif ABA, Choudhary AAAA. Risk to surgeons: a survey of accidental injuries during operations. Br J Surg 1988;75 :314-316 Gerberding JL, Schecter WP. Surgery and AIDS: reducing the risk (editorial). JAMA 1991;265:1572-1573 Williams DM, Marx MV, Korobkin M. AIDS risk and risk reduction in the radiology department (commentary). AJR 1991;1 57:919-921 Standard specification for rubber surgical gloves (D3577). In: Annual book of ASTM standards, vol. 9.02. Philadelphia: American Society for Testing and Materials, 1991:344-346
