Comparison of bacterial counts in glove juice of surgeons wearing smooth band rings versus those without rings
Comparison of bacterial counts in glove juice of surgeons wearing smooth band rings versus those without rings
Tami R. Waterman, DVM,a Daniel D. Smeak, DVM, Diplomate ACVS,a Joseph Kowalski, DVM, MS, PhD, Diplomate ACVM,a and Erinn M. Hade, MSb Columbus, Ohio
Background: Removal of rings is recommended before surgeons aseptically prepare for surgical procedures. Objectives: This study was undertaken to determine whether there is a difference between bacterial counts under surgical gloves of ring wearers compared with nonring wearers after proper aseptic hand preparation and 3 hours of wear. Methods: Twenty volunteer veterinary medical students were divided into 2 groups: One group wore a smooth ring band (without projections or mounted stones) on their ring finger, and the other group did not wear a ring. A modified glove juice method was used to obtain bacterial counts (colony-forming units/mL) inside surgically gloved hands prior to a proper aseptic hand preparation and 3 hours after hand preparation and wear. Each of the pre- and postsurgical glove juice samples were inoculated onto Letheen agar plates, which were incubated aerobically under atmospheric conditions for 48 hours at 35°C. Gloves were tested for perfo rations using a water pressure test.
Results: No differences, or significant change, in bacterial counts were found before or after surgery between all ring hands and nonring hands or between ring and nonring hands for ringed participants. No differences in bacterial counts were found within perforated versus nonperforated gloves.
Conclusion: There is no compelling evidence to suggest that surgeons wearing rings possess higher bacterial counts under their gloves during surgery. (Am J Infect Control 2006:34:421-5.)
Removal of rings is recommended before surgeons aseptically prepare for surgical procedures.' Although most surgery team members remove their rings, some are not physically able to remove them or choose not to do so.2 There is reason for concern because ring wearers in the operating room are thought to have increased bacterial counts under their gloves coupled with an even greater risk of glove perforation at the ring site.' ·2 Although there are numerous factors associ ated with wound infection risk, surgical glove perfora tion alone during a procedure has not been shown to be an important risk factor, provided the surgeon's hands are aseptically prepared for surgery. Glove perfo rations commonly occur in up to 65 % of operations performed by nonring wearers without resulting in higher surgical site infection (SSI).35 However, if bacte rial counts under gloves are increased in ring-wearing surgeons, the patients' wound infection risk could be significant. It is known that, when the contaminating dose of bacteria reaches greater than 105 microorgan isms per gram of tissue in a wound, SSI ensues.6 When foreign material, or devitalized tissue, is present in a wound, the dose of contaminating organisms re quired for an SSI to occur is significantly reduced.6'7 Consequently, even nominal increases in wound con tamination could increase infection risk. Thus, it is important to determine whether ring wearers have in creased bacterial counts under their gloves during a procedure and if they indeed are an important source of wound contamination.
Previous literature has demonstrated that skin be neath rings can be more heavily colonized with bac teria than comparable areas of skin on fingers without rings. However, mean bacterial colony counts on hands immediately after handwashing have been shown to be similar among ring wearers and non-ring wearers.8'9'11 Surprisingly, bacterial counts under gloves at the end of a procedure have not been com pared between surgeons wearing rings versus those with no hand jewelry. This is particularly important be cause ring wearers pose a greater risk of perforating the glove at the ring site compared with nonring wearers, especially when handling heavy orthopedic instru ments with the ringed hand.1 This study was un dertaken to determine whether there is a difference between bacterial counts under surgical gloves of ring wearers compared with nonring wearers after a proper aseptic hand preparation and 3 hours of wear. Our hypothesis is that there is no difference in the postsur-gical bacterial counts in glove juice between ring and nonring wearers.
METHODS Participants
Twenty veterinary medical students from The Ohio State University volunteered to be included in the study. One group consisted of 12 students (4 male, 8 female) who continuously wore a smooth ring band without projections or mounted stones on their ring finger for at least 1 month prior to the study. One of the male ring-wearing participants dropped out of the study because of illness, leaving 11 ring-wearing partic ipants. All of the students with a smooth ring band wore it on her left hand, except for 1 individual who wore it on her right hand. The other group consisted of 8 students (2 male, 6 female) who did not wear a ring. All participants had a nail length less than 5 mm and had not washed their hands with antiseptic surgical scrub for at least 3 days prior to the study.
Prescrub glove juice sampling
To suspend bacteria on the participant's hands, a modified glove juice method was used.1215 All partici pants donned sterile gloves, and 20 mL glove juice so lution (0.075 mol/L phosphate-buffered normal saline solution with 0.05% Tween 80) was instilled inside each glove separately. The glove was sealed to the par ticipant's wrist with a rubber band. Participants rubbed the solution down the fingers of the glove using the op posite hand, clenched their fists, and shook their hand for 60 seconds. While maintaining sterile technique, 5 to 10 mL of the liquid (glove juice) was removed from the cupped palm of the hand inside the glove using a Tom Cat (open end) catheter (The Kendall Company, Mansfield, MA) attached to a 20-mL syringe The glove juice was then transferred to a sterile container for cul ture. Participants were asked to record any obvious leakage of fluid from their gloves during sampling.
Postsurgical glove juice sampling
After the prescrub samples were taken, all partici pants scrubbed their hands with a chlorhexidine gluco-nate scrub brush (BD E-2 Scrub 107; Becton, Dickinson and Company, Franklin Lakes, NJ) for 5 minutes using a standard handscrub method.13·1618 After the surgical scrub, all participants donned sterile gloves and gowns using a closed glove technique. The participants re mained gowned and gloved for 3 hours. During that time, participants practiced soft tissue dissection on dog cadavers, which had not been clipped or sterilely prepared for the surgical procedure. Postscrub glove juice samples were taken at the end of the 3 hours, re peating the method as described above for the prescrub glove juice sampling. Participants were asked to record a suspected glove puncture throughout the surgical procedure and any obvious leakage of solution from their gloves during sampling.
Microbiology and data handling
Each of the pre- and postsurgical glove juice sam ples were further diluted by 2 serial 10-fold dilutions using 0.075 mol/L phosphate-buffered normal saline with 0.05% Tween 80, yielding 2 additional dilutions: 101 and 10~2 Using Ά sterile bacteriologic micro-pipette, Letheen agar (Bacto Letheen Agar; Becton Dickinson Microbiology Systems, Becton Sparks Dick inson and Co., Sparks, MD) was inoculated in duplicate with 0.2 mL of undiluted glove juice and with each of the serial dilutions. The inoculum was spread evenly over the plates using sterile plastic spreaders. After in cubation of the plates at 35°C for 48 hours under atmo spheric conditions, the total number of colony-forming units (CPU) per dilution was determined using a lighted grid colony counter (New Brunswick Scientific, New Brunswick, NJ). Plates used for counts and calculations were those with >30 and <300 colonies.1215'19 Colony counts from the 2 plates were averaged and used for statistical analysis.
Glove perforation testing
Water pressure test5·12'2023: Gloves were tested for perforations by filling the gloves with 1 L water and sealing the cuff of the glove. The exterior surface was blotted dry, and gentle pressure was exerted to exam ine all surfaces for leaks. Leaks were recorded topo graphically according to whether they occurred on the hand or a finger (palmar and/or dorsal surface or in the interdigital crevice).
Statistics
Descriptive statistics, including the mean, median, minimum, and maximum were computed for each group, and bacteria counts were expressed as number per 102 The distribution of bacteria counts had an extremely skewed distribution; therefore, several rank-based methods were chosen for this analysis. Compar isons were made for ringed subjects only via the sign test to compare ringed and nonring hands' presurgery, postsurgery, and change measurements. The compari sons between pre- and postsurgery counts, ring hands and nonring hands, and perforated and non-perforated gloves for all participants were made on the ranked values of bacteria counts and were ad justed for the clustering present on subjects using a robust estimate of the variance. All reported Ρ values are 2-sided.
Table I. Comparison of bacterial counts* expressed as number Χ Ι02 for all-participants group
Table 2. Comparison of bacterial counts* expressed as number Χ Ι02 for ringed participants
*Colony-forming units/mL.
* Comparison made by a ranked analysis, accounting for clustering on subject.
RESULTS Microbiology testing
No differences were found before and after surgery between bacterial counts of all ring hands and nonring hands or between ring and nonring hands for ringed participants (Tables 1 and 2). As expected, for all hands, there was a significant decrease in bacteria counts comparing counts before scrubbing with those ob tained after surgery (Table 1). No difference was ob served in the change in bacteria counts before and after surgery between all ring hands and nonring hands or between ring and nonring hands for ringed participants (Tables 1 and 2). Bacterial counts listed are expressed in colony-forming units per milliliter (CFU/mL).
Glove perforation testing
No obvious leakage from the participants' gloves was noted during both the prescrub and postsurgical sampling procedure. There were no perforations found in any of the gloves after the prescrub glove juice sam pling procedure using the water pressure test. However, it was determined that 8 of the 19 participants perfo rated at least 1 glove after 3 hours of surgery (27%). Some gloves had more than 1 perforation. The majority of perforations was found in the nondominant hand (61 8 hands; 75%). The nondominant hand was also the left ringed hand in the majority of the participants who punctured their glove (4/6 hands; 67%). Most of perforations were located at the fingertips (7/11; 63.6%); 1 hole was a tear at the tip of the thumb, which
*Colony-forming units/mL. *Comparison made by the sign test.
was likely created during glove removal. Eight of the 11 perforations were made in the left ringed hand (73%), with only 1 perforation made at the ring site (0.09%) (Fig 1). No differences were found between bacterial counts in glove juice of perforated and nonperforated gloves 3 hours after wear (Table 3).
DISCUSSION Bacterial counts
As expected, there was a statistically significant re duction of bacterial counts in glove juice when com paring all prescrub to postsurgical bacterial counts. All prescrub bacterial counts had an elevated bacterial count (median, 47.5 Χ 102 CFU/mL) when compared with the postscrub bacterial counts (median, 0.13 X 102 CFU/mL) The quantity of bacteria we documented was similar to previous studies in which prescrub and postscrub bacterial counts were recorded using the glove juice sampling technique.1214 This suggests that the protocol and sampling methods used in this study appear to be reliable and repeatable.
Several sampling methods have been used to deter mine bacterial counts on the skin. Some researchers have used swabs2·10 to collect bacteria on the skin at the region of the ring site and other nonring regions of hand skin for bacterial quantification. Others have stamped hand prints on agar plates (hand-imprinting technique)9 or used a 60-second friction rinse using a phosphate-buffered solution (timed standardized hand-washing technique) to quantify bacterial numbers on hands.8'11 The glove juice method used in this study was first described by Larson et al15 in 1980 and is con sidered the gold standard for quantifying bacterial counts within gloved hands.1214 There appears to be less potential for sample contamination and a greater accuracy of obtaining aerobic bacterial counts within the gloved hand with this method. Because of the difficultly in
Ringed Participants
Non-ringed Participants
Table 3. Comparison of postsurgical bacterial counts* expressed as number Χ Ι02 between perforated and nonperforated gloves
Left Ringed Hand (8) Right Non-ringed Hand (2) Left Non-ringed Hand (1) Right Non-ringed Hand (0)
Fig I. Number of the perforations and their location found in the postsurgical gloves are shown
above.
sampling, no published studies have at tempted to quantify anaerobic bacteria. Although it is unlikely that significant anaerobic bacterial counts are present under gloves on properly prepared hands, it should be noted that we counted aerobic hand flora only in this study.
This study demonstrated no detectable or clinically significant difference in the prescrub bacterial counts in ringed versus nonringed hands for both the ringed group and the all-participant group (all: 46.0 vs 50.0 median CFU/mL, respectively; Ρ = .70; ringed only: 46.0 vs 21.2 median CFU/mL, respectively; Ρ = 1.00). Previous studies using bacterial swabs to sample the skin under a ring have shown that rings predispose the skin to higher bacterial counts when compared with a similar area on the same hand or the opposite hand.2'10 In another study, elevated bacterial counts were found on ringed hands compared with nonringed hands prior to handwashing using the timed standard ized handwashing technique. In their study, Jacobson et al sampled hands from 15 subjects containing 2 to 5 rings per hand, which may have contributed to the el evated bacterial counts.8 Other studies have shown an insignificant difference between ringed and nonringed hands prior to handwashing using the hand-imprinting technique and timed standardized handwashing technique.9'11
We found no detectable or statistically significant difference in the postscrub bacterial counts in ringed versus nonringed hands for both the ringed group and the all-participant group (all: 0.20 vs 0.05 median CFU/mL, respectively; Ρ = .58; ringed: 0.20 vs 0.05 me dian CFU/mL, respectively; Ρ = .45). Other studies have shown that thorough handwashing of ringed hands reduces bacterial counts immediately after scrubbing to counts similar to that obtained when rings were not worn.8'9'11 This was demonstrated in all studies in which immediate posthandwashing bacterial counts of ringed versus nonringed hands were compared. However, none of these studies used the gold standard glove juice technique or sampled gloved hands after a 3-hour surgical procedure as we did in this study. Because glove puncture rates and bacterial counts on hands increase
CFU/mL
Ν Mean Min Median Max Ρ value
Perforated gloves 8 6.01 0.00 0.53 39.50 .93f
Nonperforated gloves 30 7.90 0.00 0.13 133.00
*Colony-forming units/mL
^Comparison made by a ranked analysis, accounting for clustering on subject.
with increasing operating time, a 3-hour time frame was chosen as our postscrub sampling time to model more accurately the bacterial counts on hands during an operation.21'22
The prescrub bacterial count was compared with the postsurgical bacterial count to determine the degree of change for both the ringed group and the all-partici pant group. This information was used to demonstrate that there is not a significant difference in the change of bacterial counts between ringed and nonringed hands (all: 38.0 vs 40.5 median CFU/mL, respectively; Ρ = .74; ringed only: 38.0 vs 12.25 median CFU/mL, respectively; Ρ = 1.00). This further supports the lack of difference between ring and nonring hands when prescrub and postsurgical bacterial counts were compared.
Perforations
An important factor related to the frequency of glove perforation during surgery has been shown to be the type and length of procedure. Higher perforations rates are often seen in nonsoft tissue surgeries (ie, ortho pedic, neurologic, thoracic, and dental) because of the increased length of surgery time and the intense manipulation of bone, screws, wire, and teeth in some of these procedures.1'5'20'24 As a result, there is wide variation in the incidence of glove perforations during surgery reported in the literature ranging from 2.6% to 65 %: neurologic (23 % ),24 dental (52 %),20 orthopedic (32.2%-60%),1'22'24 thoracic (2.6%-65%),522 gastroin testinal (31 %),22 and soft tissue (18.8%).24
Different methods of testing glove integrity may also contribute to the variable glove perforation rates re ported in the literature. Currently, the 2 most widely used methods include the water pressure5·20·23 and electrical resistance tests.23'24 The water pressure method was chosen for this study because it is easy to perform, and it can determine the location of the perforation during the test. This method, however, is useful for detecting substantial glove perforations but may not detect minute tears. It could be argued that large tears are more likely to be clinically significant than microscopic ones. Because there may be a greater degree of false-negative findings when using the wa ter pressure testing method, the actual number of per forations in this study may have been higher than reported.20·23'24 We chose this test because we were primarily interested in determining the number of glove perforations in the region of the ring, and this cannot be ascertained by electrical resistance tests.
The overall incidence of glove perforations in this study was low and within the range reported during soft tissue surgery procedures performed on humans (27%). The majority of the perforations were found in the nondominant hand (75%) and the tips of the fin gers (64%). These are some of the more common areas of glove perforation documented in the literature.5'12 Interestingly, the left ringed hand was also the non-dominant hand for most of the participants with a per forated glove (67%). There is not a clear explanation as to why most of the perforations occurred in the left ringed hand, especially because only 1 of the 8 perfo rations found on the left ringed hand was found at the ring site. Higher perforations rates at the ring site have been previously demonstrated in orthopedic sur gery, in which a high degree of handling and frequent changing of large instruments were used.:
We did not have evidence to suggest that bacterial counts in glove juice were different between perforated and nonperforated gloves, suggesting that there was insignificant environmental microbial contamination of the hands through the perforations (0.53 vs 0.13 median CFU/mL, respectively; Ρ = .93). Even though perforations occur commonly in surgery, prospective studies have demonstrated that they are not an impor tant causative factor for SSI.3·4'25 This may be partially due to chlorhexidine's prolonged antibacterial action inside the glove15 and the self-sealing capabilities of the glove itself.23 Likewise, patient characteristics (ie, patient's age, altered immune function, length of preoperative stay, and others) and operating character istics (ie, preoperative skin preparation, antimicrobial prophylaxis, surgical technique, and others) each con tribute to development of a SSI.18 Because we could not find a difference in bacterial counts on ringed hands versus nonringed hands, and clinical studies have shown that glove perforations do not appear to in crease SSI, or pose a threat to the surgeon when hands are properly prepared, we do not have compelling evi dence to suggest that surgeons who wear rings (without projections) during surgery pose additional risk to the patient. Given the limited size and observational design of this study, further clinical studies are needed to determine whether surgeons who wear rings under gloves increase bacterial loads in wounds or increase the risk of SSI.
References
Tami R. Waterman, DVM,a Daniel D. Smeak, DVM, Diplomate ACVS,a Joseph Kowalski, DVM, MS, PhD, Diplomate ACVM,a and Erinn M. Hade, MSb Columbus, Ohio
Background: Removal of rings is recommended before surgeons aseptically prepare for surgical procedures. Objectives: This study was undertaken to determine whether there is a difference between bacterial counts under surgical gloves of ring wearers compared with nonring wearers after proper aseptic hand preparation and 3 hours of wear. Methods: Twenty volunteer veterinary medical students were divided into 2 groups: One group wore a smooth ring band (without projections or mounted stones) on their ring finger, and the other group did not wear a ring. A modified glove juice method was used to obtain bacterial counts (colony-forming units/mL) inside surgically gloved hands prior to a proper aseptic hand preparation and 3 hours after hand preparation and wear. Each of the pre- and postsurgical glove juice samples were inoculated onto Letheen agar plates, which were incubated aerobically under atmospheric conditions for 48 hours at 35°C. Gloves were tested for perfo rations using a water pressure test.
Results: No differences, or significant change, in bacterial counts were found before or after surgery between all ring hands and nonring hands or between ring and nonring hands for ringed participants. No differences in bacterial counts were found within perforated versus nonperforated gloves.
Conclusion: There is no compelling evidence to suggest that surgeons wearing rings possess higher bacterial counts under their gloves during surgery. (Am J Infect Control 2006:34:421-5.)
Removal of rings is recommended before surgeons aseptically prepare for surgical procedures.' Although most surgery team members remove their rings, some are not physically able to remove them or choose not to do so.2 There is reason for concern because ring wearers in the operating room are thought to have increased bacterial counts under their gloves coupled with an even greater risk of glove perforation at the ring site.' ·2 Although there are numerous factors associ ated with wound infection risk, surgical glove perfora tion alone during a procedure has not been shown to be an important risk factor, provided the surgeon's hands are aseptically prepared for surgery. Glove perfo rations commonly occur in up to 65 % of operations performed by nonring wearers without resulting in higher surgical site infection (SSI).35 However, if bacte rial counts under gloves are increased in ring-wearing surgeons, the patients' wound infection risk could be significant. It is known that, when the contaminating dose of bacteria reaches greater than 105 microorgan isms per gram of tissue in a wound, SSI ensues.6 When foreign material, or devitalized tissue, is present in a wound, the dose of contaminating organisms re quired for an SSI to occur is significantly reduced.6'7 Consequently, even nominal increases in wound con tamination could increase infection risk. Thus, it is important to determine whether ring wearers have in creased bacterial counts under their gloves during a procedure and if they indeed are an important source of wound contamination.
Previous literature has demonstrated that skin be neath rings can be more heavily colonized with bac teria than comparable areas of skin on fingers without rings. However, mean bacterial colony counts on hands immediately after handwashing have been shown to be similar among ring wearers and non-ring wearers.8'9'11 Surprisingly, bacterial counts under gloves at the end of a procedure have not been com pared between surgeons wearing rings versus those with no hand jewelry. This is particularly important be cause ring wearers pose a greater risk of perforating the glove at the ring site compared with nonring wearers, especially when handling heavy orthopedic instru ments with the ringed hand.1 This study was un dertaken to determine whether there is a difference between bacterial counts under surgical gloves of ring wearers compared with nonring wearers after a proper aseptic hand preparation and 3 hours of wear. Our hypothesis is that there is no difference in the postsur-gical bacterial counts in glove juice between ring and nonring wearers.
METHODS Participants
Twenty veterinary medical students from The Ohio State University volunteered to be included in the study. One group consisted of 12 students (4 male, 8 female) who continuously wore a smooth ring band without projections or mounted stones on their ring finger for at least 1 month prior to the study. One of the male ring-wearing participants dropped out of the study because of illness, leaving 11 ring-wearing partic ipants. All of the students with a smooth ring band wore it on her left hand, except for 1 individual who wore it on her right hand. The other group consisted of 8 students (2 male, 6 female) who did not wear a ring. All participants had a nail length less than 5 mm and had not washed their hands with antiseptic surgical scrub for at least 3 days prior to the study.
Prescrub glove juice sampling
To suspend bacteria on the participant's hands, a modified glove juice method was used.1215 All partici pants donned sterile gloves, and 20 mL glove juice so lution (0.075 mol/L phosphate-buffered normal saline solution with 0.05% Tween 80) was instilled inside each glove separately. The glove was sealed to the par ticipant's wrist with a rubber band. Participants rubbed the solution down the fingers of the glove using the op posite hand, clenched their fists, and shook their hand for 60 seconds. While maintaining sterile technique, 5 to 10 mL of the liquid (glove juice) was removed from the cupped palm of the hand inside the glove using a Tom Cat (open end) catheter (The Kendall Company, Mansfield, MA) attached to a 20-mL syringe The glove juice was then transferred to a sterile container for cul ture. Participants were asked to record any obvious leakage of fluid from their gloves during sampling.
Postsurgical glove juice sampling
After the prescrub samples were taken, all partici pants scrubbed their hands with a chlorhexidine gluco-nate scrub brush (BD E-2 Scrub 107; Becton, Dickinson and Company, Franklin Lakes, NJ) for 5 minutes using a standard handscrub method.13·1618 After the surgical scrub, all participants donned sterile gloves and gowns using a closed glove technique. The participants re mained gowned and gloved for 3 hours. During that time, participants practiced soft tissue dissection on dog cadavers, which had not been clipped or sterilely prepared for the surgical procedure. Postscrub glove juice samples were taken at the end of the 3 hours, re peating the method as described above for the prescrub glove juice sampling. Participants were asked to record a suspected glove puncture throughout the surgical procedure and any obvious leakage of solution from their gloves during sampling.
Microbiology and data handling
Each of the pre- and postsurgical glove juice sam ples were further diluted by 2 serial 10-fold dilutions using 0.075 mol/L phosphate-buffered normal saline with 0.05% Tween 80, yielding 2 additional dilutions: 101 and 10~2 Using Ά sterile bacteriologic micro-pipette, Letheen agar (Bacto Letheen Agar; Becton Dickinson Microbiology Systems, Becton Sparks Dick inson and Co., Sparks, MD) was inoculated in duplicate with 0.2 mL of undiluted glove juice and with each of the serial dilutions. The inoculum was spread evenly over the plates using sterile plastic spreaders. After in cubation of the plates at 35°C for 48 hours under atmo spheric conditions, the total number of colony-forming units (CPU) per dilution was determined using a lighted grid colony counter (New Brunswick Scientific, New Brunswick, NJ). Plates used for counts and calculations were those with >30 and <300 colonies.1215'19 Colony counts from the 2 plates were averaged and used for statistical analysis.
Glove perforation testing
Water pressure test5·12'2023: Gloves were tested for perforations by filling the gloves with 1 L water and sealing the cuff of the glove. The exterior surface was blotted dry, and gentle pressure was exerted to exam ine all surfaces for leaks. Leaks were recorded topo graphically according to whether they occurred on the hand or a finger (palmar and/or dorsal surface or in the interdigital crevice).
Statistics
Descriptive statistics, including the mean, median, minimum, and maximum were computed for each group, and bacteria counts were expressed as number per 102 The distribution of bacteria counts had an extremely skewed distribution; therefore, several rank-based methods were chosen for this analysis. Compar isons were made for ringed subjects only via the sign test to compare ringed and nonring hands' presurgery, postsurgery, and change measurements. The compari sons between pre- and postsurgery counts, ring hands and nonring hands, and perforated and non-perforated gloves for all participants were made on the ranked values of bacteria counts and were ad justed for the clustering present on subjects using a robust estimate of the variance. All reported Ρ values are 2-sided.
Table I. Comparison of bacterial counts* expressed as number Χ Ι02 for all-participants group
Table 2. Comparison of bacterial counts* expressed as number Χ Ι02 for ringed participants
| Ν | Mean | Min | Median | Max | Ρ value | |
| Prescrub | 38 | 298.82 | 0.25 | 47.50 | 2580.00 | <.OI* |
| Postsurgery | 38 | 7.50 | 0.00 | 0.13 | 133.00 | |
| Prescrub counts | ||||||
| King hands | II | 128.64 | 0.30 | 46.00 | 1020.00 | .70* |
| Nonring hands | 27 | 368.15 | 0.25 | 50.00 | 2580.00 | |
| Postsurgery counts | ||||||
| Ring hands | II | 5.06 | 0.00 | 0.20 | 33.00 | .58* |
| Nonring hands | 27 | 8.49 | 0.00 | 0.05 | 133.00 | |
| Change in counts | ||||||
| (presrub vs | ||||||
| postsurgery) | ||||||
| Ring hands | II | 123.58 | 0.10 | 38.00 | 1005.85 | .74* |
| Nonring hands | 27 | 359.66 | 0.25 | 40.50 | 2580.00 |
*Colony-forming units/mL.
* Comparison made by a ranked analysis, accounting for clustering on subject.
RESULTS Microbiology testing
No differences were found before and after surgery between bacterial counts of all ring hands and nonring hands or between ring and nonring hands for ringed participants (Tables 1 and 2). As expected, for all hands, there was a significant decrease in bacteria counts comparing counts before scrubbing with those ob tained after surgery (Table 1). No difference was ob served in the change in bacteria counts before and after surgery between all ring hands and nonring hands or between ring and nonring hands for ringed participants (Tables 1 and 2). Bacterial counts listed are expressed in colony-forming units per milliliter (CFU/mL).
Glove perforation testing
No obvious leakage from the participants' gloves was noted during both the prescrub and postsurgical sampling procedure. There were no perforations found in any of the gloves after the prescrub glove juice sam pling procedure using the water pressure test. However, it was determined that 8 of the 19 participants perfo rated at least 1 glove after 3 hours of surgery (27%). Some gloves had more than 1 perforation. The majority of perforations was found in the nondominant hand (61 8 hands; 75%). The nondominant hand was also the left ringed hand in the majority of the participants who punctured their glove (4/6 hands; 67%). Most of perforations were located at the fingertips (7/11; 63.6%); 1 hole was a tear at the tip of the thumb, which
| Ν | Mean | Min | Median | Max | Ρ value |
| Prescrub counts | |||||
| Ring hands 1 1 | 128.64 | 0.30 | 46.00 | 1020.00 | 1.00* |
| Nonring hands 1 1 | 227.74 | 0.25 | 21.20 | 1435.00 | |
| Postsurgery counts | |||||
| Ring hands 1 1 | 5.06 | 0.00 | 0.20 | 33.00 | .45* |
| Nonring hands 1 1 | 13.43 | 0.00 | 0.05 | 133.00 | |
| Change in counts | |||||
| (prescrub vs | |||||
| postsurgery) | |||||
| Ring hands 1 1 | 123.58 | 0.10 | 38.00 | 1005.85 | 1.00* |
| Nonring hands 1 1 | 214.31 | 0.25 | 12.25 | 1302.00 |
*Colony-forming units/mL. *Comparison made by the sign test.
was likely created during glove removal. Eight of the 11 perforations were made in the left ringed hand (73%), with only 1 perforation made at the ring site (0.09%) (Fig 1). No differences were found between bacterial counts in glove juice of perforated and nonperforated gloves 3 hours after wear (Table 3).
DISCUSSION Bacterial counts
As expected, there was a statistically significant re duction of bacterial counts in glove juice when com paring all prescrub to postsurgical bacterial counts. All prescrub bacterial counts had an elevated bacterial count (median, 47.5 Χ 102 CFU/mL) when compared with the postscrub bacterial counts (median, 0.13 X 102 CFU/mL) The quantity of bacteria we documented was similar to previous studies in which prescrub and postscrub bacterial counts were recorded using the glove juice sampling technique.1214 This suggests that the protocol and sampling methods used in this study appear to be reliable and repeatable.
Several sampling methods have been used to deter mine bacterial counts on the skin. Some researchers have used swabs2·10 to collect bacteria on the skin at the region of the ring site and other nonring regions of hand skin for bacterial quantification. Others have stamped hand prints on agar plates (hand-imprinting technique)9 or used a 60-second friction rinse using a phosphate-buffered solution (timed standardized hand-washing technique) to quantify bacterial numbers on hands.8'11 The glove juice method used in this study was first described by Larson et al15 in 1980 and is con sidered the gold standard for quantifying bacterial counts within gloved hands.1214 There appears to be less potential for sample contamination and a greater accuracy of obtaining aerobic bacterial counts within the gloved hand with this method. Because of the difficultly in
Ringed Participants
Non-ringed Participants
Table 3. Comparison of postsurgical bacterial counts* expressed as number Χ Ι02 between perforated and nonperforated gloves
Left Ringed Hand (8) Right Non-ringed Hand (2) Left Non-ringed Hand (1) Right Non-ringed Hand (0)
Fig I. Number of the perforations and their location found in the postsurgical gloves are shown
above.
sampling, no published studies have at tempted to quantify anaerobic bacteria. Although it is unlikely that significant anaerobic bacterial counts are present under gloves on properly prepared hands, it should be noted that we counted aerobic hand flora only in this study.
This study demonstrated no detectable or clinically significant difference in the prescrub bacterial counts in ringed versus nonringed hands for both the ringed group and the all-participant group (all: 46.0 vs 50.0 median CFU/mL, respectively; Ρ = .70; ringed only: 46.0 vs 21.2 median CFU/mL, respectively; Ρ = 1.00). Previous studies using bacterial swabs to sample the skin under a ring have shown that rings predispose the skin to higher bacterial counts when compared with a similar area on the same hand or the opposite hand.2'10 In another study, elevated bacterial counts were found on ringed hands compared with nonringed hands prior to handwashing using the timed standard ized handwashing technique. In their study, Jacobson et al sampled hands from 15 subjects containing 2 to 5 rings per hand, which may have contributed to the el evated bacterial counts.8 Other studies have shown an insignificant difference between ringed and nonringed hands prior to handwashing using the hand-imprinting technique and timed standardized handwashing technique.9'11
We found no detectable or statistically significant difference in the postscrub bacterial counts in ringed versus nonringed hands for both the ringed group and the all-participant group (all: 0.20 vs 0.05 median CFU/mL, respectively; Ρ = .58; ringed: 0.20 vs 0.05 me dian CFU/mL, respectively; Ρ = .45). Other studies have shown that thorough handwashing of ringed hands reduces bacterial counts immediately after scrubbing to counts similar to that obtained when rings were not worn.8'9'11 This was demonstrated in all studies in which immediate posthandwashing bacterial counts of ringed versus nonringed hands were compared. However, none of these studies used the gold standard glove juice technique or sampled gloved hands after a 3-hour surgical procedure as we did in this study. Because glove puncture rates and bacterial counts on hands increase
CFU/mL
Ν Mean Min Median Max Ρ value
Perforated gloves 8 6.01 0.00 0.53 39.50 .93f
Nonperforated gloves 30 7.90 0.00 0.13 133.00
*Colony-forming units/mL
^Comparison made by a ranked analysis, accounting for clustering on subject.
with increasing operating time, a 3-hour time frame was chosen as our postscrub sampling time to model more accurately the bacterial counts on hands during an operation.21'22
The prescrub bacterial count was compared with the postsurgical bacterial count to determine the degree of change for both the ringed group and the all-partici pant group. This information was used to demonstrate that there is not a significant difference in the change of bacterial counts between ringed and nonringed hands (all: 38.0 vs 40.5 median CFU/mL, respectively; Ρ = .74; ringed only: 38.0 vs 12.25 median CFU/mL, respectively; Ρ = 1.00). This further supports the lack of difference between ring and nonring hands when prescrub and postsurgical bacterial counts were compared.
Perforations
An important factor related to the frequency of glove perforation during surgery has been shown to be the type and length of procedure. Higher perforations rates are often seen in nonsoft tissue surgeries (ie, ortho pedic, neurologic, thoracic, and dental) because of the increased length of surgery time and the intense manipulation of bone, screws, wire, and teeth in some of these procedures.1'5'20'24 As a result, there is wide variation in the incidence of glove perforations during surgery reported in the literature ranging from 2.6% to 65 %: neurologic (23 % ),24 dental (52 %),20 orthopedic (32.2%-60%),1'22'24 thoracic (2.6%-65%),522 gastroin testinal (31 %),22 and soft tissue (18.8%).24
Different methods of testing glove integrity may also contribute to the variable glove perforation rates re ported in the literature. Currently, the 2 most widely used methods include the water pressure5·20·23 and electrical resistance tests.23'24 The water pressure method was chosen for this study because it is easy to perform, and it can determine the location of the perforation during the test. This method, however, is useful for detecting substantial glove perforations but may not detect minute tears. It could be argued that large tears are more likely to be clinically significant than microscopic ones. Because there may be a greater degree of false-negative findings when using the wa ter pressure testing method, the actual number of per forations in this study may have been higher than reported.20·23'24 We chose this test because we were primarily interested in determining the number of glove perforations in the region of the ring, and this cannot be ascertained by electrical resistance tests.
The overall incidence of glove perforations in this study was low and within the range reported during soft tissue surgery procedures performed on humans (27%). The majority of the perforations were found in the nondominant hand (75%) and the tips of the fin gers (64%). These are some of the more common areas of glove perforation documented in the literature.5'12 Interestingly, the left ringed hand was also the non-dominant hand for most of the participants with a per forated glove (67%). There is not a clear explanation as to why most of the perforations occurred in the left ringed hand, especially because only 1 of the 8 perfo rations found on the left ringed hand was found at the ring site. Higher perforations rates at the ring site have been previously demonstrated in orthopedic sur gery, in which a high degree of handling and frequent changing of large instruments were used.:
We did not have evidence to suggest that bacterial counts in glove juice were different between perforated and nonperforated gloves, suggesting that there was insignificant environmental microbial contamination of the hands through the perforations (0.53 vs 0.13 median CFU/mL, respectively; Ρ = .93). Even though perforations occur commonly in surgery, prospective studies have demonstrated that they are not an impor tant causative factor for SSI.3·4'25 This may be partially due to chlorhexidine's prolonged antibacterial action inside the glove15 and the self-sealing capabilities of the glove itself.23 Likewise, patient characteristics (ie, patient's age, altered immune function, length of preoperative stay, and others) and operating character istics (ie, preoperative skin preparation, antimicrobial prophylaxis, surgical technique, and others) each con tribute to development of a SSI.18 Because we could not find a difference in bacterial counts on ringed hands versus nonringed hands, and clinical studies have shown that glove perforations do not appear to in crease SSI, or pose a threat to the surgeon when hands are properly prepared, we do not have compelling evi dence to suggest that surgeons who wear rings (without projections) during surgery pose additional risk to the patient. Given the limited size and observational design of this study, further clinical studies are needed to determine whether surgeons who wear rings under gloves increase bacterial loads in wounds or increase the risk of SSI.
References
-
Nicholai P, Aldam CH, Allen PW. Increased awareness of glove perfo ration in major joint replacement. J Bone Joint Surg Ι997;79-Β:37Ι-3.
-
2. Field EA, McGowan P, Pearce PK, Martin MV. Rings and watches: should they be removed prior to operative dental procedures? J Dent 1996:24:65-9.
-
Howe CW, Marston AT. A study on sources of postoperative staphy-lococcal infection. Surg Gynecol Obstet 1962;! 15:266.
-
Davidson AIG, Smylie HG. A bacteriological study of the immediate environment of a surgical wound. Br J Surg I97I;58:326.
-
Eklund AM, Ojajarvi J, Laitinen K, Valtonen M, Werkkala KA. Glove punctures and postoperative skin flora of hands in cardiac surgery. Ann Thorac Surg 2002;74:l49-53.
-
Krizek TJ, Robson MC. Evolution of quantitative bacteriology wound management. Am J Surg 1975; 130:579-84.
-
James RC, Macleod CJ. Induction of staphylococcal infections in mice with small inocula introduced on sutures. BrJ Exp Pathol 1961 ;42:266-77.
-
Jacobson G, Thiele JE, McCune JH, Farrell LD. Handwashing: ring-wearing and number of microorganisms. Nurs Res 1984;34:186-8.
-
Nicholson-Pegg A. The wearing of wedding rings in the operating department. NATNews 1982;Apr: 19-22.
-
Hoffman PN, Cooke EM. Microorganisms isolated from skin under wedding rings worn by hospital staff. Br Med J 1985;290:206-7.
-
I I. Salisbury DM, Hutfilz P, Treen LM, Bollin GE, Gautam S. The effect of rings on microbial load of health care workers' hands. Am J Infect Control 1997;25:24-7.
-
Olsen RJ, Lynch P, Coyle MB, Cummings J, Bokete T, Stamm WE. Examination gloves as barriers to hand contamination in clinical prac tice. JAMA 1993:270:350-3.
-
Wheelock SM, Lookinland S. Effect of surgical hand scrub time on subsequent bacterial growth. AORN J 1997;65:1087-97.
-
Wan PY, Blackford JT, Bemis DA, Rohrbach BW, Knoll DE, Provenza MK. Evaluation of surgical scrub methods for large animal surgeons. Vet Surg 1997:26:382-5.
-
Larson El, Strom MS, Evans CA. Analysis of three variables in sampling solutions used to assay bacteria of hands: type of solution, used of an tiseptic neutralizers, and solution temperature. J Clin Microbiol 1980; 12:355-60.
-
Lowbury EJL, Lilly HA, Bull JP. Methods for disinfection of hands and operation sites. Br Med J 1964;2:531.
-
Larson EL. APIC guideline for hand washing and hand antisepsis in health-care settings. Am J Infect Control 1995:23:251-69.
-
Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Hospital Infection Control Practices Advisory Committee. Guideline for pre vention of surgical site infection. Infect Control Hosp Epidemiol 1999:20:247-78.
-
Kwochka K, Kowalski JJ. Prophylactic efficacy of four antibacterial shampoos against Staphy/ococcus intermedius in dogs. Am J Vet Res 1991:52:115-8.
-
Albin MS, Bunegin L, Duke ES, Ritter RR, Page CR Anatomy of a defective barrier: sequential glove leak detection in a surgical and dental environment. Crit Care Med 1992;20:170-84.
-
Jamal A, Wilkinson S. The mechanical and microbiological integrity of surgical gloves. ANZ J Surg 2003:73:140-3.
-
Laine T, Aarnio P. How often does glove perforation occur in surgery? Comparison between single gloves and a double-gloving system. AmJ Surg 2001; 181:564-6.
-
Sohn RL, Murray MT, Franko A, Hwang PK, Dulchavsky SA, Grimm MJ. Detection of surgical glove integrity. Am Surg 2000:66:302-6.
-
Character BJ, McLauglin RM, Hedlund CS, Boyle CR, Elder SH. Post operative integrity of veterinary surgical gloves. J Am Anim Hosp Assoc 2003:39:311-20.
-
Cruse PJ. Surgical wound infection. In: Gorbach SL, BartlettJG, Black-low NR, editors. Infectious diseases. Philadelphia: W.B. Saunders Co; 1992. p. 759-64.
