RAPID INACTIVATION OF INFECTIOUS PATHOGENS CHLORHEXIDINE-COATED GLOVES

BY

Shanta Modak, PhD; Lester Sampath, BA; Harvey S.S. Miller, BA; Irving Millman, PhD

A B S T R A C T

OBJECTIVE: Gloves containing chlorhexidine gluconate in an instant-release matrix on their inner surface (CHG gloves) were tested to determine their ability to rapidly inactivate infectious pathogens that may permeate or leak through the latex surface. DESIGN: CHG gloves were exposed for 1 to 10 minutes to blood or media containing infectious pathogens (e.g., bacteria, fungi, parasites, and viruses) as well as to lymphocytes and macrophages that are known to be the primary carriers of human immunodeficiency virus (HIV). Inactivation of pathogens was determined either by in vitro assay or in vivo infectivity. Stressed control and CHG glove fingers were submerged in a viral pool (retrovirus or bacteriophage) and after a set time, the glove interiors were checked for presence of permeated viz-ions. INTRODUCTION The use of latex gloves as a barrier to protect healthcare workers from contact with blood or other

RESULTS: CHG gloves rapidly inactivate all the pathogens tested including retrovirus and hepatitis B virus (90% to 100%). In the stressed glove fingers, live virus was detected in 26% of the control group but not in any of the CHG group. CONCLUSIONS: The use of CHG gloves may reduce the risk of exposure to infectious fluidborne pathogens should the integrity of the latex barrier be compromised by overt failure or by permeation of viruses. Rapid destruction of lymphocytes and macrophages may facilitate inactivation of HIV associated with these cells. Tests have shown that CHG coating does not alter physical properties of the glove, and, furthermore, CHG gloves do not show potential for dermal irritation or sensitization. (Infect Control Hosp Epidemiol. 1992;13:463-471.)

body fluids containing pathogens (e.g., human immunodeficiency virus [HIV] and hepatitis B virus [HBVI) has been well recognized. However, it has been

From the Department of Surgery, Columbia University College of Physicians and Surgeons, New York, New York (D1: Modak and Mx Sampath); Daltex Medical Sciences, Inc., Fairfield, New Jersey (D1: Miller); and Fox Chase Cancer Center, Divisions of fipulation, Oncology, and Clinical Research, Philadelphia, Annsylvania (D1: Millman). HBV/WHVand RLVstudies were conducted at the Fox Chase Cancer Center, Philadelphia, Annsylvania. The authors thank Drs. A. Sampath, Harlem Hospital Center, New York, New York, P Bocarsly, University of Medicine and Dentistry of New Jersey, Newark, New Jersey, and G. Fernandes, The University of T&as Health Center, San Antonio, Texas, for their assistance and advice in some of the experiments and Baruch S. Blumberg, MD, PhD, Fox Chase Cancer Center, for his contribution to the research involving HBT( Wm and RLK as well as his counsel on the overall testing program. Address reprint requests to Shanta Modak, PhD, Dept. of Surgery, Black Bldg. Rm. 1734, Columbia University, College of Physicians and Surgeons, 630 W 168th St., New York, NY10032. Modak S, Sampath L, Miller HSS, Millman I. Rapid inactivation of infectious pathogens by chlorhexidine-coated gloves. Infect Control Hosp Epidemiol. 1992;13:463-471.

464

INFECTION CONTROL

AND

reported that a significant number of vinyl and latex gloves allow the seepage of blood and other body fluids through minute pinholes or damaged glove surfaces caused by chemical and mechanical insults.1-5 The protective effect of gloves often can be compromised by inadvertent puncture from fingernails and medical instruments or stretching during use, which causes permeabilityG7 Moreover, breaches in the glove barrier increase with the time worn and fre quently are unnoticed. 4,6 Individuals who have cuts, nicks, or abrasions on the surface of their hands are especially at risk in the event of glove failure. Clearly, in spite of glove use, healthcare workers are still at risk of exposure to fluidborne infectious agents. The commonly practiced method of antiseptic handwashing prior to glove donning8 cannot provide adequate protection, because the necessary antiseptic concentration needed for rapid kill of subsequently intruding pathogens is not available from a washed skin surface. Coating the inner surface of the glove with an instant-release matrix containing a potent, broad-spectrum antiseptic agent may alleviate this problem. Such a coating provides sufficient antiseptic in situ for rapid action against fluidborne pathogens. Chlorhexidine gluconate (CHG) appears to be best suited for this purpose because it is a widely known antiseptic that has been used to decontaminate skin for over three decades.g10 Its other applications include use in oral prophylaxis11-14 and in surgical and bladder irrigation fluids. 15-16 We evaluated the rapid microbicidal efficacy of latex gloves coated with a proprietary CHGcontaining matrix on the inner surface of the glove. MATERIALS AND METHODS

Control and CHGcoated latex gloves (CHG glove) were manufactured by Daltex Medical Sciences, Inc, Fairfield, New Jersey. All the experiments were carried out using examination gloves. However, some of the experiments were repeated using hypoallergenie surgical gloves (made by the same manufacturer) . Each Hibistat towelette (Stuart Pharmaceuticals, Wilmington, Delaware) contained 5 ml of 0.5% chlorhexidine gluconate as the active ingredient in 70% isopropanol. Ficoll hypaque (Histopaque-1077) was purchased from Sigma Chemical Company, St. Louis, Missouri. TISO Medium for CHG Neutralization

Trypticase soy broth (TSB) containing the following ingredients was used as the CHG neutraliiing growth medium (TLSO):17 5% Tween 80, 2% lecithin, 0.6% sodium oleate, 0.5% sodium trisulfate, 0.1% proteose peptone, and 0.1% tryptone. A preliminary exper-

HOSPITAL EPIDEMIOLOGY

August 1992

iment to determine the neutralization capacity of the TISO medium was carried out by exposing Staflhylococcus aureus cultures to CHG (well above the levels in the glove) in TSB and in TLSO. The results showed that the bacterial growth was not inhibited by CHG in TLSO medium, indicating its effectiveness as a rapid neutralizer of CHG. Pathogens

The pathogens used in this experiment were obtained as follows: Staphylococcus aureus (strain # 10390) from American Type Culture Collection, Rockwell, Maryland; Pseudomonas aeruginosa (strain VA134) from the US Army Institute of Surgical Research, Fort Sam Houston, Texas; Neisseria gonorrhoeae and Trichomonas vaginalis from the Clinical Microbiology Diagnostic Laboratory, Harlem Hospital Center, New York, New York; and Candida a&cans (strain #11651) from American Type Culture Collection. A virulant strain of herpes simplex virus (HSV) (strain #2931) was used in this study HBV used in this study was isolated from the serum of infected woodchucks. Woodchuck hepatitis virus (WHV) is considered an acceptable model for the human hepatitis viru~.~~ Rauscher murine leukemia virus @LV) (strain RVB,) was obtained from Dr. Strand, Johns Hopkins University, Baltimore, Maryland. Pseudomonas phaseolicola and 06 Bacteriophage

06 Bacteriophage and Pseudomonas phaseolicola HBlOY were obtained from Dr. Leonard Mindich, Department of Microbiology, The Public Health Research Institute, New York, New York. 06 is a lipophilic phage, comparable in size to HIV (06 = 90 nm, HIV = approximately 100 nm) and is susceptible to the action of chlorhexidine. P phaseolicola is the host bacterium for 06 bacteriophage and was used for evaluating phage infectivity.1g-21 Whole Blood

Human blood was collected from laboratory volunteers in tubes containing EDTA as the anticoagulant. Studies involving human subjects were conducted at Columbia University, New York, New York. Columbia-presbyterian Medical Center Institutional Review Board approval was obtained for this study. In vitro Evaluation of the Rapidity of Kill of Viuious Pathogens by CHG Gloves

S aureus was used as the model for bloodborne pathogens in this and other experiments where indicated. Earlier, we had determined the concentration of CHG needed to completely inactivate S aureus and found it to be nearly identical to that reported for HlV22 ‘lb determine the rate of inactivation of S

Vol. 13 No. 8

CHLORHEXIDINE GLOVES

aureus by CHG glove fingers, 0.1 ml of S aureus culture (either in saline or blood) containing 5 x lo5 colony forming units (CFU)/ml was added to each glove finger, which was draped over a test tube. The bacterial culture was spread over the entire inner surface of the glove finger using a stirring rod and was then continuously manipulated with the same rod for one and five minutes at room temperature to mimic the movement of fingers inside the glove. Si fingers were used in the control and CHG glove groups. To inactivate the CHG in the glove extracts, 0.9 ml of TLSO was added to each finger at the desired time, mixed well, and the contents recovered. A total of 0.2 ml was plated on a blood agar plate and incubated at 37” C for 24 hours, and the number of colonies was counted. The experiment also was repeated using surgical gloves. Similar protocols were used to determine the inactivation of other pathogens by CHG gloves. Another experiment was carried out to evaluate the in vivo efficacy of CHG gloves in inactivating pathogens that may continuously seep in during glove use. In this experiment, the volunteers wore a control glove on one hand and a CHG glove on the other. An 18 g needle was used to make a pinhole in each glove finger prior to donning. A second control glove containing 0.25 ml of the volunteers’ own blood inoculated with S aureus (lo5 CFU/ml) in each finger was donned over the first glove. After an interval of either 15 minutes or two hours, the outer glove was removed, the blood on the outer surface of the inner glove was blotted off, and 0.9 ml TLSO was added to the inside of each glove finger while the glove was still donned using the procedure described below. The glove was rolled back over itself and was then pulled away from the palm to gain access to the inside of the glove fingers. After adding TLSO to completely wash out the blood from the wearer’s fingers and the inner side of the glove fingers, the fluid was removed carefully using a sterile Pasteur pipette and an aliquot was subcultured on a blood agar plate to determine the colony counts. Effect of Hibistat Towelette and CHG Gloves on Hand Contamination

The purpose of this experiment was to determine whether an effective concentration of CHG would be available to inactivate the pathogens that may intrude at any time during glove use. This experiment was performed as follows using laboratory volunteers. Two volunteers participated in this study for three consecutive days. On Day 1, the volunteers wore test and control gloves on the right and left hands, respectively, for ten minutes to represent the exposure time for examination gloves. On Day 2, the same

465

volunteers wore the test and control gloves for three hours to represent the average exposure time for surgical gloves. On Day 3 control gloves were worn after wiping the hands with Hibistat towelettes as per the manufacturer’s directions, which state that both hands should be rubbed with the towelette vigorously for approximately 15 seconds, paying particular attention to nails and interdigital spaces. The gloves on one hand were worn for ten minutes and those on the other for three hours. After these procedures, once the wearing time had elapsed, the gloves were challenged with S aureus culture as follows: log phase culture was mixed with the volunteers’ own blood (lo5 CFU/ml) and 0.1 ml was injected into each of the glove fingers. The wearer then manipulated his or her fingers to simulate actual use conditions for ten minutes. While the glove was still donned, 0.9 ml of TLSO was added as described earlier to wash out the blood from both the wearer’s fingers and the glove surface. The fluid was carefully removed and an aliquot was subcultured on a blood agar plate to determine the colony counts. The three-hour exposure experiment was repeated with surgical gloves. Effect of CHG Gloves on HSV

This experiment was carried out in the Virology Laboratory of Dr. P Bocarsly, Department of Pathology, New Jersey Medical School, Newark, New Jersey. In order to avoid questionable results, a preliminary experiment was carried out to evaluate the toxic effect of CHG on the host cells of the virus. A total of 20 ~1 of solution containing various concentrations of CHG (12-500 pg/ml) or extracts from CHG glove fingers of known drug level were placed in sets of four replicate wells containing confluent monolayers of Vero monkey kidney cells. A total of 200 pl of Dulbecco’s Minimal Essential medium containing 10% fetal calf serum was then added to the wells and serially diluted. Toxicity to the monolayer was scored after 48 hours of incubation at 37°C in a 4% CO, atmosphere. Toxicity was scored from 0 to + + + + , where 0 represents an intact monolayer and + + + + represent complete destruction of the monolayer. At concentrations of 25 Fg/ml and below, toxicity was 0. ‘Ib determine the antiviral activity of the coated gloves, the viruses exposed to the gloves were added to monolayers of Vero monkey kidney cells in a g&well microtitre plate. A 1 ml aliquot of HSV stock (3 x lo6 plaque forming units [PFU] per ml of Hank’s buffered salt solution) was placed in each of the CHG and control glove fingers (six glove fingers in each group) and incubated at room temperature. After ten minutes, four 20 ~1 aliquots of the virus suspension from each glove finger were removed, serially diluted on monolayers of Vero cells, and processed as

466

INFECTION CONTROL

AND

described above. After 48 hours of incubation, the virus titers were read from the dilution that had a CHG concentration less than 10 pg/ml, which is below the lethal concentration for the host cells and that reported for the viru~.~~ Effect of CHG Glove on HBV

The effect of the glove against HBV was evaluated by assessing the decrease in the virus-associated DNA polymerase activity. A total of 200 ~1 of the viral suspension was exposed to CHG and control glove fingers (six fingers in each group) for five minutes at 37°C. The viral suspension was removed, passed through a Sephadex G-200 column to remove CHG from the fluid and the resulting void volume was used as the source of the viral enzyme. As a control, 200 p,l of viral suspension was incubated at 37°C for five minutes and was passed through a Sephadex column. Aliquots from all of the above samples were assayed for DNA polymerase activity using the method described by Modak and Ma~-cus.~~ In brief, 20 ~1 aliquots of the samples were added to a standard reaction mixture containing tritiated deoxythymidine triphosphate (3Hd’ITP) and incubated at 37°C for 30 minutes. The amount of DNA synthesized was then estimated by measuring the radioactivity in the trichloroacetic acid insoluble precipitate. The percentage inhibition was calculated using the DNA polymerase activity of the control virus (unexposed to the glove) as 100%. Effect of CHG Gloves on the Infectitity of a Model Retrovirus (RLI

Because in vivo experiments with HIV are diicult to perform and no established in vivo model exists, RLV was used as a model to study the effect of CHG glove fingers on the retroviral infectivity in vivo. This retrovirus is routinely used as a model for HIV because RLV induces clear disease indicators in a relatively short period of time.25-27 Mouse splenomegaly is produced when RLV is injected in the tail vein. Viral infectivity can then be measured by determining the relative weight increase of the mouse spleen 20 days after the injection. In this experiment four groups of six, six-week old female BALB/C ICR mice were injected with aliquots of viral suspension prepared from the following groups (six fingers in each group): CHG glove finger + 0.1 ml RLV stock (240 foci units/ml); control glove finger + 0.1 ml RLV stock; negative control (0.1 ml phosphate buffered saline [PBS]); positive control: (0.1 ml RLV stock). The groups were incubated for ten minutes at 37°C and the viral suspension from each glove finger was collected. All six samples from each group were

HOSPITAL EPIDEMIOLOGY

August 1992

pooled and clarified by centrifugation for five minutes. The supematant was applied to a Sephadex G-200 column, and the CHGfree void volume peak containing the virus was collected and diluted to 2 ml with PBS. Each of six mice were then injected with 0.25 ml of the diluted sample, and after 20 days they were sacrificed and their spleens weighed. To evaluate the effect of virus-free extracts from CHG and control glove fingers on the spleen, the glove fingers were incubated with 0.4 ml of PBS at 37°C and 0.25 ml was directly injected into the mice. Spleen weights were determined after 20 days. Rapidity of the Virucidal Action of CHG Gloves in the Presence of Blood

To determine the rapid disinfecting capability of CHG gloves on viruses in body fluids, 0.1 ml of blood containing 06 bacteriophage (lo5 PFU/ml) was added to each glove finger, which was draped over a test tube (12 control glove fingers and 12 CHG glove fingers). A glass rod was used to continuously spread the blood over the latex surface for two minutes. Then, 0.9 ml saline was then added to the finger and all the fluid was removed. After a lOOfold dilution, 0.5 ml aliquots were passed through a Sephadex G-200 column, and the CHGfree void volumes were collected. The void volume containing the virus was added to 2.5 ml of Luria-Bortani (LB) semi-solid medium inoculated with 0.2 ml of P phaseolicola (lo8 CFU/ml) and poured onto an LB solid agar plate. After incubation at 27°C for 18 hours, the viral plaques were counted. Effect of CHG Gloves on Lymphocytes in BJood

A total of 0.1 ml human blood was spread on each control and CHG glove finger with a stirring rod (ten fingers in each group) for five minutes. Each sample was diluted with 0.5 ml PBS and pooled. The pooled blood PBS mixture was layered onto 3 ml of Histopaque gradient, and the lymphocytes were separated by a standard procedure given by the manufacturer (Sigma). Lymphocyte viability was determined by exclusion of trypan blue. Effect of CHG Gloves on Macrophages

Macrophages isolated from the peritoneal cavity of mice were adjusted to a cell concentration of approximately 106/ml. A total of 0.1 ml aliquots of this suspension were plated in microtitre plates, to which 10 ~1 of the extracts (0.1 ml saline/finger) from control and CHG glove fingers (six fingers in each group) were added. The control plates had only PBS. After ten minutes of incubation at 37°C in a CO, incubator, the cell viability was determined by exclusion of trypan blue.

Vol. 13 No. 8

CHLOKHEXIDINEGLOVES

Infectivity of virus Permeated Tbrou& Stressed Control Gloves and CHG GJoves

The efficacy of CHG gloves in preventing infectivity of viruses that may permeate through stressed glove surfaces was evaluated using bacteriophage 06 and RLiV as the viral models, as described below. 06 bacteriophage as the viral model. The

glove fingers were stressed using a standardized set of manipulations designed to mimic the wear and tear experienced during patient care. Control and CHG glove tinger tips (1.5 cm length) were stressed by stretching them ten times. This was accomplished by inserting the bottom of a glass test tube inside each glove finger and distending the finger tip a distance equal to three times its length. After removing the test tube, 0.1 ml normal saline was added to the glove finger and the test tube was reinserted. The finger tip was stretched once more, fastened in place on the test tube with a rubber-band, and partially submerged (1 cm) in a beaker containing 25 ml of a viral suspension (lo6 PFU/ml). After ten minutes, the test tubes were removed and 0.5 ml saline was added to each finger to recover any phage that might be present. All the fluid was immediately withdrawn, centrifuged for two minutes and the supernatant passed through a 1 ml Sephadex G-200 column. The CHGfree void volumes containing the virus were collected and used to carry out the infectivity test as described earlier. After the above experiment, all the glove fingers were subjected to an air bubble test to determine the number of leaky fingers. This was accomplished by inflating the fingers with 60 cc of air followed by immersing them under water. The appearance of air bubbles was an indication of leakage. Only those phage-positive fingers that had no leakage as determined by the air bubble test were deemed to be permeable to the virus. Statistical evaluation of the results was carried out using Student’s t test. RLY as the viral model. A 10 ml RWinfected mouse spleen homogenate containing 400 foci units of RIV/ml was placed in a large glass test tube and a glove finger was stretched over its mouth. Sii control glove fingers and seven CHG glove fingers were used in this experiment. A total of 200 ~1 of PBS was placed in the tip of each glove finger. A glass pestle was used to distend the glove into the viral pool for 20 stretch cycles. Each stretch cycle extended the glove finger 3.5 times its length. After the stretching cycles, the glove fingers were distended and held in the viral pool for an additional 20 minutes. All the fluid from the interior of each glove finger was removed, and the inner surface was rinsed with 2 ml of buffer and pooled with the initial extract. The extract was further diluted to 4.8 ml and centrifuged first at a low speed (5000 x g) to remove the particulate matter. The

467

supernatant was further centrifuged for one hour at 108,000 x g to sediment the viral particles (all the CHG released from the glove fingers was found in the supernatant). The sediment was suspended in 250 pl of PBS and was injected into each mouse by the tail vein. The mice were divided into the following groups: control glove fingers exposed to virus; CHG glove fingers exposed to virus; and buffer with no virus (negative control). After 20 days, all the animals were sacrificed and their spleens were weighed. Statistical analysis of the results was carried out by Student’s t test. RESULTS

In Vitro Evaluation of the Rapidity of Kill of Vhrious Pathogens by CHG Gloves

When CHG glove fingers were exposed to S aureus, 100% kill was observed in 1 minute. C albicans, P aeruginosa, N gonorrhoeae, and T vaginalis showed >99.7% inactivation in 5 minutes. Similar results also were obtained with surgical gloves. When blood containing S aureus (105 CFU/ml) was allowed to seep continuously through perforated glove fingers, the following results were obtained. After 10 minutes, 99% inactivation of the intruding pathogens was noted in the CHG glove fingers (compared with the control glove group), and after 2 hours, these gloves inactivated 88% of the intruding pathogens. Effect of Hibistat ToweIette and CHG Gloves on Hand Contamination

The anti-infective efficacy of the conventional method of Hibistat towelette use prior to glove donning was compared with that of CHG gloves. This was tested in volunteers using S aureus as the challenging organism. In the Hibistat group, 60% of the glove fingers showed positive cultures at the lOminute and 3-hour intervals. In contrast, none of the CHG glove fingers was positive. One hundred percent of the glove fingers were contaminated in the glove group, which were donned over antiseptic-free hands. Similar results were also obtained with surgical gloves. Effect of CHG Gloves on HBV and HSV

Table 1 shows the antiviral activity of CHG gloves against HBV and HSV. Complete inactivation of HSV occurred within 10 minutes of exposure to CHG gloves, while 91% inactivation was noted within five minutes in the case of HBV Effect of CHG Gloves on the Infectivity of a Model Retrovhs (RLV)

When the control glove extracts containing the virus were passed through a Sephadex G-200 column,

A68

TNFECTION

TABLE 1 EFFECT OF CHG GLOVES Time of Virus*

HSV HBV

ON

CONTROL

HSV AND HBV

August 1992

HOSPITAL EPIDEMIOLOGY

AND

TABLE 2 EFFECT OF CHG G~VFJS

ON

Splenomegaly in Mice*

lnactivatii of Virus (%)

Exposure

Control Glove

CHG Glove

(Minutes)

Fingers

Ringers

10 5

0 30

s9.9 91

+ The virus stocks exposed to the glove fingers were tested for viability by plaque assay for HSV and by determination of virus associated DNA polymerase activity for HBV. The results a~ the mean of 6 fingers.

a high molecular weight void volume peak was observed, indicating the presence of intact virus. However, no such peak could be found with the viral cultures exposed to the CHG glove group. This indicated that CHG glove fingers completely disintegrated the RIY in 10 minutes. As shown in Table 2, the mice in the control glove and the virus control groups developed splenomegaly, as expected. However, no increase in spleen weight was noted in the CHG glove group. Rapidity of the WucidaJ Action of CHG Gloves in the Presence of Blood

When the virus (06 bacteriophage) suspended in blood was exposed to the glove fingers for 2 minutes, 83% of the control fingers showed the presence of infective phage, compared with only 8% in the CHG glove fingers (103-lo4 PFU/fmger in both groups). Effect of CHG Gloves on Lymphocytes in Human Blood

In the control glove group, 88% of the lymphocytes were viable, compared with 0% in the CHG glove group. Microscopic examination revealed fragments of disintegrated lymphocytes in the latter group. Macrophages exposed to CHG glove finger extracts also showed 100% inactivation. Infectivity of Virus Permeated Through Stressed Control and CHG Gloves

In the stressed control glove group, live phage was detected even in the fingers that had no apparent breaks in the latex barrier. In contrast, none was detected in the CHG glove fingers, including those that were perforated (Table 3). Table 4 shows the infectivity of RLJI particles that were permeated through stressed glove fingers. The small but consistent increase in spleen weight of all the mice in the control glove group shows the permeation of a few viral particles through the stressed glove surfaces. However, there was no

RLV

Group

PBS control Virusalone Virus exposed to control glove Viis exposed to CHG glove l

Spleen Weight (mg)

98 1,627 1,486 111

Tbe results are the mean spleen weights from 6 mice. High spleen weight (splenomegaly) indicates viral infection.

increase in the spleen weight in the CHG glove group, indicating that the small number of viral particles permeated through these glove fingers were completely inactivated. Statistical analysis (Student’s t test) of both bacteriophage and RLV results clearly demonstrated a significant difference between control and CHG gloves with respect to viral presence because of either permeability or leakage through the latex barrier (# = .Ol for permeated RLY or phage and p = .05 for leaked phage). DISCUSSION

Routine use of gloves by healthcare workers has been recommended to minimize the risk of contracting infections. The increase in the incidence of infections in healthcare workers and surgeons has been attributed to defective gloves, glove perforation, and permeation of pathogens through stressed but appar ently intact gloves.2833 A recent study34 to determine the frequency of glove leakage after dental and surgical procedures indicates that 33% of the surgical gloves had leakage, 17% of which occurred within 75 minutes after donning. After dental procedures, 32% of the gloves leaked, 24% of which occurred within 30 minutes; the leakage was somewhat decreased with double-gloving. The novel approach described in this study employs a glove whose inner surface is coated with CHG in an instant-release matrix for the rapid inactivation of fluidborne pathogens. The chemical barrier of CHG supplements the physical latex barrier, providing additional protection against pathogens, should the latter fail. The CHG coating does not alter the physical properties of the glove such as tensile strength, durability, flexibility, or lubricity and is not affected by sterilization (unpublished data). Furthermore, the CHG matrix also inactivates infectious viral particles that may permeate through stressed but apparently intact latex surfaces. This report describes the results of challenging

TABLE 3 INFECTIVITY

OF

469

C HLORHEXIDINE GLOVES

Vol. 13 No. 8

06 BACTERIOPHAGEIN STRESSED GLOVES Control Glove CHG Glove

Percentage of fingers perforated with leaks Percentage of fingers with infective phages

Fingers

Fingers

10

10

5

0

TABLE 4 INFECTMTY OF RLV PERMEATED THROUGHSTRESSED GLOVEFINGERS Splenomegaly In Mice Spleen Weight

(mg)

No. of (p= .05)

Percentage of fingers intact but permeable

26

26*

Percentage of fingers with infective phages

26

0

Group

Negative control (no virus; PBS only) Control glove fingers CHG glove fingers

6

95

1.0

6

124

2.2

7

89

0.9

(p= .Ol) l

cp= .Ol)

Fingers Tested (Mean) (S.D.)*

Standard deviation.

* Permeability is indicated by the presence of infective phages. Even though there were no infective phages in the CHG glove groups despite being subjected to the identical stressing procedure as the control, it was presumed that the same number of fingers were permeable.

CHG glove fingers with various pathogens, human lymphocytes, and macrophages. Because the glove finger is the most common point of fluid entry35 and accumulation, we used it as a model. The studies were based on an empirical finding that a typical glove finger would accumulate approximately 0.1 ml blood or other body fluids before being noticed by the glove wearer and removed. In an informal survey, surgeons at Columbia-Presbyterian Medical Center and elsewhere made the visual estimation that this is the usual maximum volume commonly found postprocedurally in a perforated glove finger. The CHG glove fingers were found to rapidly inactivate the pathogens regardless of the carrier fluid (e.g., blood, sweat, etc.). Of particular concern to healthcare workers today is the exposure to HIY HIV has been isolated from body fluids such as blood and saliva,36,37 either as a free virus or harbored in lymphocytes and macrophages. Our studies indicate that CHG released from a single glove finger can completely disintegrate the lymphocytes and macrophages present in 0.1 ml blood within one to ten minutes. Their disintegration, in turn, may result in the destruction of HIV virions associated with these cells, because the lethal concentration of CHG for HIV is much lower than that for lymphocytes present in 0.1 ml blood. It has been reported that 2,000 kg/ml of CHG completely inactivates HIV in 15 seconds.38 When the glove finger is breached by 0.1 to 0.2 ml of body fluid, the concentration of CHG released greatly exceeds this amount. The activity of CHG gloves is stable because neither the physical act of donning nor perspiration during

three hours of glove use had any significant effect on the activity. In addition, these gloves were found to be efficacious against fluidborne pathogens that were continuously seeping in through perforated gloves during two hours of use. Use of chlorhexidine in the form of a Hibistat towelette (0.5% CHG) or Hibiclens (4% CHG) for decontamination of hands has been recommended and is widely practiced by medical professionals between patients and prior to surgery. Our studies show that 76% of the CHG applied on the hand from the towelette binds to the skin within 30 minutes. This was determined by measuring the CHG levels in sequential hand washings after towelette use. The lower efficacy of the towelette relative to the CHG glove may be due to the unavailability of free CHG on the hand for the rapid inactivation of subsequently intruding pathogens. In addition to HTV, HBV also remains a serious threat to healthcare workers.3g CHG gloves were also found to be effective against this virus. The evaluation of the CHG gloves on HBV was carried out using an earlier prototype that contained only one-third of the CHG available with subsequently developed antimicrobial gloves. Even with this small amount, 91% inactivation of HBV was found. It has been reportedz3 that, in the presence of a 1:lO dilution of 0.12% chlorhexidine gluconate mouth rinse, 94% to 99% inactivation of HBV (as determined by the inhibition of virus-associated DNA polymerase activity) occurs in five to 15 minutes. Because the drug released from the CHG glove is much higher than this amount, similar, if not better, results can be expected with the use of CHG gloves. It is also important to note that the maximum amount of CHG potentially released from the interior

470

INFF,CTION CONTROL

AND

of the glove is well within the established dosage limitations for safe and effective topical use of this compound. Because the amount of CHG delivered to the hand by the CHG glove is significantly lower than that from a towelette, allergic reactions from the routine use of these gloves is unlikely. In fact, a repeat insult patch test of these gloves on 204 volunteers indicates that these gloves do not have the potential for dermal irritation or sensitization (Personal communication. Diane Fraitz, Daltex Medical Inc., Fairchild, New Jersey, 1991). CHG gloves are also effective against bacterial and fungal pathogens to which healthcare workers may be exposed that can result in infection and cross-contamination.2p40 Another concern with the routine use of any antimicrobial agents is the possible emergence of resistant pathogens. Despite the worldwide use of chlorhexidine for almost four decades, only a few incidences of the possible emergence of CHG-induced resistant bacteria have been reported.41 However, there have been a number of reports in the literature of intrinsically resistant bacterial strains.42~43 Our attempts to develop resistant strains of S aureus and Escherichia coli by repeated exposure of subinhibitory concentrations of CHG to these pathogens in vitro were unsuccessful. Based on these results, the emergence of chlorhexidine-resistant mutants because of routine use of CHG gloves appears unlikely. The use of CHG gloves by healthcare workers may reduce the risk of their exposure to fluidborne pathogens, including HIV and HBV, in the event the integrity of the glove barrier is compromised. REFERENCES 1. Kotilainen HR, Brinker JP Avato JL, Gantz NM. Latex and vinyl examination gloves: quality control procedures and implications for health care workers. Arch Intern Med. 1989:149:2749-2753. 2. Russell TR, Roque FE, Miller FA. A new method for detection of the leaky glove: a study on incidence of defective gloves and bacterial growth from surgeon’s hands. Arch Surg. 1966;93:245249.

3. Church J, Sanderson I? Surgical gloves puncture. /Hosp Infect. 198O;1:84. 4. Paulsen J, Eidem T, Kristiansen R. Perforations in surgeons’ gloves. J Hosp Infect. 1988;11:82-85. 5. Dodds RDA, Guy PJ. Surgical glove perforation. BY J Surg. 1988;75:96&968.

6. Miller JM, Collier CS, Griffith NM. Permeability of surgical rubber gloves. Am J Surg. 1972;124:57-59. 7. Hochreiter MC, Barton LL. Epidemiology of needlestick injury in emergency medical service personnel. J Emerg Med. 1988;6:912.

8. Peterson AF; Rosenberg A, Alatary SD. Comparative evaluation of surgical scrub preparations. Sung Gynecol O&et. 1978;146:6365.

9. Rosenberg A, Alatary SD, Peterson AI? Safety and efficacy of the antiseptic chlorhexidine gluconate. Surg Gynecol O&et. 1976;143:789-792. 10. Denton GW. Chlorhexidine. In: SS Block ed. D&infection, Sterilization, and F7eservation. Philadelphia, Pa: Lea & Febiger; 1991:274-289. 11. Schiott CR Effect of chlorhexidine on the microflora of the oral

HOSPITAL EPII)EMIOIX)C;Y

August 1992

cavity. JPeriodont Z&s. 1973;8:7-10. 12. Loe H. Chlorhexidine in the prophylaxis of dental diseases. Introduction. JFbiodont Z&s. 1973;8:5-6. 13. Cumming BR, Loe H. Optimal dosage and method of delivering chlorhexidine solutions for the inhibition of dental plaque. J Periodont Res. 1973;8:57-62. 14. Schiott, Loe H. The sensitivity of oral streptococci to chlorhexidine. J Periodont &s. 1972:7:192-194. 15. Hannah WJ. Gevention of urinary tract infections after vaginal surgery. Can Med Assocj 1963;88:802-805. 16. Bastable JRG, Peel RN, Birch DM, Richards B. Continuous irrigation of the bladder after prostatectomy: its effect on post-prostatectomy infection. BrJ Lkol. 1977;49:689-693. 17. Benson L, Bush L, LeBlanc D. Importance of neutralizers in the stripping fluid in a simulated healthcare personnel handwash. Infect Control Hosp Epidemiol. 1990;11:595599. 18. Millman I, Southam L, Halbherr T, Simmons H. Kang CM. Woodchuck hepatitis virus; experimental infection and natural occurrence. Hepatology 1984;4:817-823. 19. Sinclair JE Tzagoloff A, Levine D, Mindich L. Proteins of bacteriophage 06. J Viral. 1975;16:685695. 20. Olkkonen VM. Gottlieb l? Strassman 1. Qiao X. Bamford DH. Mindich L. In vitro assembly of infectious nucleocapsids of bacteriophage 06: formation of a recombinant double-stranded RNA virus. Proc Nut1 Acad Sci. 1990;87:9173-9177. 21. Gottlieb P Strassman J, Qiao X, Frucht A, Mindich L. In vitro replication, packaging, and transcription of the segmented double-stranded RNA genome of bacteriophage 06 studies with procapsids assembled from plasmid-encoded proteins. 1 Butt. 1990;172:57745782. 22. Montefiori DC, Robinson WE, Modliszewski A, Mitchell WM. Effective inactivation of human immunodeficiency virus with chlorhexidine antiseptics containing detergents and alcohol. J Hosp Infect. 1990;15:279-282. 23. Bernstein D, S&ii G, Echler G, et al. In vitro virucidal effectiveness of a 0.1236chlorhexidine gluconate mouthrinse. J Dent Res. 1990;69:874-876. 24. Modak MJ, Marcus S. Purification and properties of Rauscher leukemia virus DNApolymerase. JBiol Gem. 1977;252:11-19. 25. Ruprecht RM, Rossoni LD, Haseltine WA, Broder S. Suppression of retroviral propagation and disease by suramin in murine system. Z+oc Nat1 Acud Sci. 1985;82:7733-7737. 26. Ruprecht RM, O’Brien LG. Rossoni LD, Nusinoff-Lehrman S. Suppression of mouse viraemia and retroviral disease by S’azido3’-deoxythymidine. Nature. 1986;323:467469. 27. Murine lentivirus model concept and others ok’d by NIAID advisors. AIDS &date. 1989;2:1-2. 28. Gallo RC, Salahuddin SZ, Popovic M, et al. Human lymphotropic virus. HTLVIII. isolated from AIDS oatients and donors at risk for AIDS. Science. 1984;224:500503. . 29. Geddes AM. Risks of AIDS to health care workers. Br Med / 1986;292:711-712. 30. Weiss SH, Goedert JJ, Gartner S, et al. Risk of human immunodeficiency virus (HIV1). Infection among laboratory workers. Science. 1988;239:68. 31. Newsom SWB, Rowland C, Wells FC. What is in the surgeon’s glove? J Hosp Infect. 1988;ll (suppl A) :244-259. 32. Dalgleish AG, Malkousky M. Latex gloves not enough to exclude viruses. BrJSurg. 1988;76:171-172. 33. Reingold AL, Kane MA, Hightower AW. Failure of gloves and other protective devices to prevent transmission of hepatitis B virus to oral surgeons. JAMA. 1988;259:25582560. 34. Albin MS, Bunegin L, Duke ES, Ritter RR, Page CI? Anatomy of a defective barrier: sequential glove leak detection in a surgical and dental environment. Crit Cure Med. 1992;20:170. 35. Quebbeman EJ, ‘Rlford GL, Hubbard S, et al. Risk of blood contamination and iniurv to operating room personnel. Ann Suvg. 1991;214:614-620. 36. Groooman IE. Salahuddin SZ. Sarngadharan MG. Markham PD. Gonda M, &ski A, Gallo Rd. HTLiIII in saliva’of people with AIDS related complex and healthy homosexual men at risk for AIDS. Science. 1984;226:447-449. 37. Ho DD, MoudgilT, Alam M. Quantitation of human immunodeti-

Vol. 13 No. 8

C HLORHEXIDINE Gr.ov~s

ciencyvirustype 1 in the blood of infected persons. NEnglJMed. 1989;321:1621-1625.

38. Harbison MA, Hammer SM. Inactivation of human immunodeficiency virus by betadine products and chlorhexidine. J Acquired Immune Deficiency Syndromes. 1989;2:1&20. 39. Dienstag JL, Ryan DM. Occupational exposure to hepatitis B virus in hospital personnel. Am J Epidemiol. 1982;115:26-39. 40. Burnie JP Candida and hands. J Hosp Infect. 1986;8:1-4.

471

41. Stickler DJ. Chlohexidine resistance to PLateus mirabilis. J Clin Rzth. 1974;27:284-287. 42. Pitt TL, Gaston MA, Hoffman PN. In vitro susceptibility of hospital isolates of various bacterial genera to chlorhexidine. / Hosp Infect. 1983;4:173-176. 43. Russell AD. Chlorhexidine: antibacterial action and bacterial resistance. Comment. Infection. 1986;14:212-215.