The American Journal of Surgery (2012) 204, 976 –980

The Southwestern Surgical Congress

Sterile gloves: do they make a difference?

Jennifer Creamer, M.D.*, Kurt Davis, M.D., William Rice, M.D.

Department of Surgery, William Beaumont Army Medical Center, El Paso, TX, USA

KEYWORDS: Abstract
Gloves; BACKGROUND: Multiple studies have demonstrated that ⬎105 organisms/mL are needed to cause
Infection a wound infection. The aim of this study was to determine if there was a difference in bacterial
colony-forming units (CFUs) on sterile gloves versus clean gloves in an outpatient clinical setting.
METHODS: Volunteers self-gloved with pairs of clean gloves, and culture swabs were obtained from
the palmar surface. Cultures were also obtained after volunteers self-donned sterile gloves and donned
sterile gloves with the assistance of a surgical technician.
RESULTS: Twenty-five volunteers participated. Mean growths were as follows: clean gloves, 14.08 ⫾
15.45 CFUs/mL (range, 0 – 44 CFUs/mL); self-donned sterile gloves, 1.28 ⫾ 4.28 CFUs/mL (range,
0 –20 CFUs/mL); and technician-assisted sterile gloves, 1 positive with 8 CFUs/mL.
CONCLUSIONS: There was a statistically significant difference in bacterial load on clean gloves
versus sterile gloves (P ⬍ .001). However, when comparing the bacterial contamination on clean gloves
with that required to cause an infection, it appeared that this statistically significant difference was
clinically irrelevant.
Published by Elsevier Inc.

Evidence-based medicine has become the standard by
which decisions are made and outcomes measured. Despite
this, there are many “traditional” practices that continue
through dogma rather than on the basis of scientific validity.
One such practice is the use of sterile gloves for minor
procedures performed in an outpatient setting. Within the
dental literature, studies have found no difference in post-
operative complications whether using sterile or clean non-
sterile gloves (hereafter, clean gloves).1,2 A multicenter
randomized controlled trial demonstrated no significant dif-
ference in infection rates when repairing lacerations in an
emergency department with sterile versus clean gloves.3
Similar results were found in studies of patients who un-
derwent minor procedures in both general practice and der-
matology clinics.4,5
* Corresponding author. Tel.: 915-742-2698; fax: 915-569-2903.
E-mail address: jennifer.creamer1@us.army.mil
Manuscript received March 12, 2012; revised manuscript May 29, 2012
In this study, we sought to determine whether a signifi-
cant difference in bacterial count was present on clean
gloves versus sterile gloves. Several studies in the literature
have shown that ⬎105 organisms/mL are needed to cause a
wound infection.6 –9 No studies have specifically investi-
gated the difference in bacterial load on sterile versus clean
gloves in an outpatient setting. We sought to identify if any
such difference was present to validate or refute the current
practice of using sterile gloves for minor office procedures.
Methods
After institutional review board approval was obtained,
25 volunteers with differing levels of experience were re-
cruited for participation in the study, and informed consent
was obtained. All gloves were nonlatex, to minimize the
risk for allergic reaction. Individuals entered the outpatient
clinic minor procedure room as if they were about to ex-
amine a patient. They obtained pairs of clean gloves from

0002-9610/$ - see front matter Published by Elsevier Inc.

http://dx.doi.org/10.1016/j.amjsurg.2012.06.003
J. Creamer et al. Do sterile gloves make a difference? 977

Table 1 CFUs per glove type

yses were conducted using SAS version 9.2 (SAS Institute
Inc, Cary, NC). Statistical significance was determined for
No Mean ⫾ SD P values ⬍ .05.
Glove type growth Growth CFUs/mL
Clean 7 18 14.08 ⫾ 15.45
Sterile self-donned 22 3 1.28 ⫾ 4.28
Sterile technician assisted 24 1 8 Results
Twenty-five volunteers were recruited and completed the
3 separate glove cultures. There were no adverse events or
the box of nonsterile gloves and put them on. Next, the allergic reactions. Volunteers consisted of 10 PA-S, 10
volunteers held out their dominant hands with fingers spread general surgery residents, and 5 surgical specialty staff
and extended for the microbiologist to obtain a culture. members. No other demographic factors were collected.
After cultures were obtained, the volunteers removed the Surgical residents were made up of postgraduate years 1
clean gloves and using a sterile technique opened an appro- through 5.
priately sized package of sterile gloves and self-gloved. Clean gloves had the highest percentage of colonization.
Cultures were obtained using the same technique, and these Twenty-two of 75 gloves cultured had bacterial growth
gloves were removed. Finally, the volunteers scrubbed their (Table 1). Clean glove cultures were positive for 18 volun-
hands with an approved preoperative scrub chemical (Ava- teers, with a mean growth of 14.08 ⫾ 15.45 colony-forming
gard, 3M, St Paul, MN; Betadine, Purdue Products, Stam- units (CFUs)/mL. Self-donned sterile glove cultures were
ford, CT; ChloraPrep, CareFusion Corporation, San Diego, positive in 3 volunteers, with a mean growth of 1.28 ⫾ 4.28
CA), and a fully gowned and gloved, certified scrub tech- CFUs/mL, and only 1 scrub technician–assisted sterile
nician assisted them in putting on sterile gloves, after which glove culture was positive, with 8 CFUs/mL.
the microbiologist obtained a final set of cultures. There was a statistically significant difference in the
mean bacterial growth between clean gloves and sterile
Culture details self-donned gloves (14.08 vs 1.28 CFUs/mL, P ⬍ .001).
There was also a significant difference between clean gloves
Cultures were obtained using the swab-rinse technique and technician-assisted sterile gloves (14.08 vs 0.32 CFUs/
with the following steps: (1) Sterile culture swabs were mL, P ⬍ .001). No significant difference was seen in the
premoistened with Trypticase Soy Broth (Becton, Dickin- mean bacterial colonization between the 2 different sterile
son and Company, Franklin Lakes, NJ). (2) A swab was gloves (1.28 vs 0.32 CFUs/mL, P ⫽ .75).
rolled over the gloved surface of the palm and fingers When comparing experience groups, PA-S had signifi-
(including the in-between areas) for 20 seconds. (3) The cantly more contamination on clean gloves than residents
swab was placed in a tube containing 250 ␮L Trypticase (22.4 vs 5.6 CFUs/mL, P ⫽ .002) but not staff members
Soy Broth, and the tube was placed on ice. (4) In the (22.4 vs 14.4 CFUs/mL, P ⫽ .73). No other significant
laboratory, tubes were vortexed for 10 seconds. After vor- effects were found between the experience groups (Table 2).
texing, sterile forceps were used to remove the swab, first Within the experience groups, clean gloves had significantly
rimming the swab against the side of the tube to expel more contamination than either type of sterile glove within
excess fluid from the cotton swab into the Trypticase Soy the PA-S as well as the staff members, but not the residents
Broth in the tube. (5) One hundred microliters of the fluid (Table 3).
was plated out onto a blood agar plate, which was incubated Although the highest rate and amount of colonization
at 35°C in 5% CO2. (6) At 24 and 48 hours, the colony existed for clean gloves, the amount of colonization was
count was recorded, and the organisms were identified by minimal for all glove types. The highest number of CFUs
gram stain. Results were converted from colony-forming
units/250␮L to colony-forming units/mL for analysis.
Table 2 Average CFUs between experience levels per glove
type
Statistical analysis
Glove type Comparison CFUs/mL P
Statistical analysis was performed using general linear Clean PA-S vs residents 22.4 vs 5.6 .002
mixed-model analyses for repeated measures to test the PA-S vs staff members 22.4 vs 14.4 .73
effects of glove type (clean, sterile self-donned, and sterile Residents vs staff members 5.6 vs 14.4 .62
technician assisted) and level of training (physician assistant Sterile self- PA-S vs residents 2.8 vs 0.4 1
donned PA-S vs staff members 2.8 vs .0 1
students [PA-S], general surgery residents, and surgical Residents vs staff members 0.4 vs .0 1
specialty staff members), as well as the interaction of glove Sterile PA-S vs residents 0.8 vs .0 1
type and level of training, on the colony counts. With the technician PA-S vs staff members 0.8 vs .0 1
significant interaction, the Tukey-Kramer post hoc proce- assisted Residents vs staff members .0 vs .0 1
dure showed where differences in the means lay. All anal-
978 The American Journal of Surgery, Vol 204, No 6, December 2012

They found that silk, silicone-treated silk, cotton, and Da-

Table 3 Average CFUs within experience levels per glove
type cron were the most effective in potentiating infection, with
only 101 to 102 CFUs necessary to produce an infection,
Experience whereas nylon required ⱖ102. Further studies supported
level Comparison CFUs/mL P these findings by showing that braided sutures have a higher
PA-S Clean vs self-sterile 22.4 vs 2.8 ⬍.001 adherence index compared with monofilaments.11,12 Of
Clean vs tech-sterile 22.4 vs .8 ⬍.001 note, the above studies used sutures directly inoculated with
Self-sterile vs tech-sterile 2.8 vs .8 1 predetermined amounts of bacteria before implantation into
Residents Clean vs self-sterile 5.6 vs 0.4 .9
Clean vs tech-sterile 5.6 vs .0 .85
humans or mice. In a modification of this model, Katz et al11
Self-sterile vs tech-sterile 0.4 vs .0 1 used rats and implanted clean sutures into subcutaneous
Staff Clean vs self-sterile 14.4 vs .0 .04 pockets, which were subsequently inoculated with predeter-
members Clean vs tech-sterile 14.4 vs .0 .04 mined amounts of bacteria. They found that the minimal
Self-sterile vs tech-sterile .0 vs .0 1 infective dose of staphylococci in the presence of any suture
Self-sterile ⫽ sterile gloves self-donned; tech-sterile ⫽ sterile material was 105 organisms/mL.11
gloves donned with technician assistance. Among clinical studies comparing clean versus sterile
gloves, only 1 study addressed additional clinical risk fac-
tors for wound infection. In their prospective study looking
at infections in dermatologic surgery, Rogues et al5 allowed
cultured was 44 CFUs/mL, in the clean glove culture group.
the participation of immunocompromised patients. Al-
Most positive cultures had ⱕ16 CFU/mL measured on final
though they found a significant difference in the incidence
count (Table 4).
of wound infection after excision for patients undergoing
immunosuppressant treatment, they found no difference in
the infectious complications in simple excisions when clean
Comments gloves were used.
There were several limitations to our study. Although we
In this study, we identified the bacterial count on clean had participants at all levels of training, our cohort was
versus sterile gloves before performing a minor procedure small at 25 participants. Additionally, gloves are not the
in an outpatient setting. It showed, with 25 different pro- only factor in creating a wound infection. We have focused
viders, that the maximum number of CFUs present on the our discussion on issues common to outpatient minor pro-
surface of a gloved hand was 44 CFUs/mL for clean gloves, cedures, such as the bacterial count necessary to cause a
20 CFUs/mL for self-donned sterile gloves, and 8 CFUs/mL wound infection and the impact of suture. Other factors such
for sterile gloves when assisted by a scrub technician. as organism virulence, foreign bodies other than suture,
Multiple studies within the literature have demonstrated devitalized tissue, and so on, can decrease the amount of
that ⬎105 organisms/mL are required to cause a wound organisms necessary to create a wound infection and were
infection. Elek6 first studied bacterial load in the 1950s, not addressed. This study was meant as a pilot study in
when he inoculated healthy volunteers with increasing preparation for a larger randomized controlled trial looking
doses of cocci and found 105 organisms/mL to be the at the use of clean versus sterile gloves in minor procedures
minimum required pus-forming dose, with all volunteers performed in an outpatient surgical clinic. The results are
requiring 106 organisms/mL to produce an infection. Since not meant to be applied in other surgical settings.
his study, several other studies have produced similar results Research has shown that ⬍105 CFUs are not enough to
in different patient populations. Robson et al7 observed that cause a wound infection.6 –9 With suture present, only 1
⬎105 organisms/mL were required to cause infection in study showed that a wound infection can be produced with
civilian trauma wounds. Krizek et al8 showed skin graft ⬍102 CFUs, and only if the suture was inoculated before
survival was ⬍20% in burn patients when the bacterial implantation.10 Our study helps explain why the previous
count was ⬎105 organisms/mL. Raahave et al9 found the clinical studies have not found differences in the wound
median infective dose of aerobic and anaerobic bacteria infection rate when using clean versus sterile gloves in
together in the postoperative wounds of healthy patients was minor procedures.1–5 Although there was a statistically sig-
4.6 ⫻ 105 CFUs. nificant difference in bacterial load on clean gloves versus
When foreign material is present, studies have shown sterile gloves, it appears that this statistically significant
that the CFUs necessary to produce infection are decreased.
Several studies have examined the effect of sutures on the
CFUs necessary to produce a wound infection. In a subset Table 4 CFUs per glove type
analysis of the previously mentioned study, Elek6 showed
that the minimum pus-forming dose was reduced from 106 CFU/mL Clean Self-Donned Technician assisted
to 102 CFUs when cocci were inoculated onto silk sutures ⱕ16 10 2 1
and implanted in the skin. James and MacLeod10 expanded ⬎17 8 1 0
this idea by looking at a variety of different sutures in mice.
J. Creamer et al. Do sterile gloves make a difference?
difference may be clinically irrelevant. In today’s economy,
it is important to be fiscally responsible. Although the sav-
ings of about $5 per procedure may seem trivial, with the
amount of minor procedures being performed daily, this
savings could quickly make a difference. Future random-
ized studies with larger cohorts will further define if there is
no difference in the rate of wound infections between the
gloves and either support or refute the use of clean gloves
for outpatient minor procedures.
References
1. Adeyemo WL, Ogunlewe MO, Ladeinde AL, et al. Are sterile gloves
necessary in nonsurgical dental extractions? J Oral Maxillofac Surg
2005;63:936 – 40.
2. Chiu WK, Cheung LK, Chan HC, et al. A comparison of post-
operative complications following wisdom tooth surgery performed
with sterile or clean gloves. Int J Oral Maxillofac Surg 2006;35:174 –9.
3. Perelman VS, Francis GJ, Rutledge T, et al. Sterile versus nonsterile
gloves for repair of uncomplicated lacerations in the emergency de-
partment: a randomized controlled trial. Ann Emerg Med 2004;43:
362–70.
4. Bruens ML, van den Berg PJ, Keijman, JM. Minor surgery in general
practice: are sterilised gloves necessary? Br J Gen Pract 2008;58:
277– 8.
5. Rogues AM, Lasheras A, Amici JM, et al. Infection control practices
and infectious complications in dermatological surgery. J Hosp Infect
2007;65:258 – 63.
6. Elek SD. Experimental staphylococcal infections in the skin of man.
Ann N Y Acad Sci 1956;65:85–90.
7. Robson MC, Duke WF, Krizek TJ. Rapid bacterial screening in the
treatment of civilian wounds. J Surg Res 1973;14:426 –30.
8. Krizek TJ, Robson MC, Kho E. Bacterial growth and skin graft
survival. Surg Forum 1967;18:518 –9.
9. Raahave D, Friis-Møller A, Bjerre-Jepsen K, et al. The infective dose
of aerobic and anaerobic bacteria in postoperative wound sepsis. Arch
Surg 1986;121:924 –9.
10. James RC, MacLeod CJ. Induction of staphylococcal infections in
mice with small inocula introduced on sutures. Br J Exp Pathol
1961;42:266 –77.
11. Katz S, Izhar M, Mirelman D. Bacterial adherence to surgical
sutures. A possible factor in suture induced infection. Ann Surg
1981;194:35– 41.
12. Alexander JW, Kaplan JZ, Altemeier WA. Role of suture materials in
the development of wound infection. Ann Surg 1967;165:192–9.
Discussion
Dr Ronald Stewart (San Antonio, TX): Thank you Dr
Creamer, thank you for getting me the well-written manu-
script well in advance. It seems to me that you and your
co-authors have shown that sterile gloves are sterile and
clean gloves are quite clean when they come from the clinic
and experienced people putting those gloves on do a bit
better job and avoid contamination than inexperienced peo-
ple. You focused on the 105 end-point in the manuscript and
the presentation, but I would question whether that is really
the best end-point in that clearly in biologic systems a single
organism or a pair of organisms in the right environment
979
can rapidly multiply. So I think there is, and you allude to
it, but for bacteria there is other factors—the virulence, the
susceptibility, the host, foreign body, so my question sort of
begs the question, what situations would you recommend
using sterile versus clean gloves, and isn’t the presence of
any bacteria on the glove the critical question for clean
procedures? Secondly, what is the cost difference between
sterile and clean gloves and specifically do you believe the
cost difference for using clean versus sterile gloves might be
negated with 1 wound infection in particularly high risk
patients? These data come from a clinic? A more worrisome
environment might be the Surgical Intensive Care Unit
where the bacteria are more likely to be virulent, you have
more compromised hosts. Did you examine clean gloves
from other settings? And, if not, do these data generalize to
other settings?
Before I close, I would like to also compliment you. This
came from a debate among surgeons in your institution and
instead of endlessly debating, you sought out to answer the
question, so congratulations on that effort.
Dr Jennifer Creamer (El Paso, TX): Thank you Dr
Stewart for your comments and questions. These all bring
about great points. On the first question, virulence can
potentially make a difference. In the studies that Elek per-
formed, he tried to keep this in mind. In his first study, he
compared several different nasal strains and also used bac-
teria from human lesions when calculating the minimal
pus-forming dose in man. He saw no difference regardless
of which strain he used. In subsequent studies, he looked at
multiple strains of bacteria and again found no difference.
But there are studies that show if you have a more virulent
strain, significantly less bacteria is needed—1 study by
Foster showed only 15 bacteria were needed to cause a wound
infection, demonstrating that virulence can be a factor. As for
a foreign body, studies clearly show that suture decreases the
rate of bacteria needed to cause a wound infection, and this can
be extrapolated to any type of foreign body. In terms of
immunocompromised patients, this was an exclusion criteria
for most studies comparing clean and sterile glove studies. It
was evaluated in the dermatology study with about 5% of their
patients being immunocompromised. Although they found no
difference in the infection rate between the 2 gloves after minor
procedures, they did find immunosuppressant treatment as an
independent risk factor for infection, showing host susceptibil-
ity should play a part in the decision to use clean or sterile
gloves.
In terms of the cost difference, we calculated at our
facility the cost difference between clean and sterile gloves
is about $2. That doesn’t seem like a lot, but if you calculate
that you probably have an assistant, and maybe some stu-
dents/residents also wearing sterile gloves, along with the
amount of minor procedures over time—that cost can add
up and you can have a significant cost savings. Treatment of
a wound infection potentially caused by wearing clean
gloves, depending on its severity can cost anywhere from
980
hundreds to thousands of dollars. Further studies need to be
performed to see if there would be an overall cost savings.
To answer your third question, yes this study was per-
formed in a minor surgical procedure clinic, and is intended
only to be applied to this setting. We need a randomized
controlled trial in a minor surgical procedure clinic, not just
retrospective studies, to see if there truly is a difference in
wound infection rates between the use of clean and sterile
gloves. I don’t believe you should put our results into other
situations; you need to look at each setting separately.
Dr Courtney Scaife (Salt Lake City, UT): I have 2
opposing questions. I have always obsessed about the fact that
we wash our hands and then rinse them in nonsterile water. So
I always use the heavy cleanse gel afterwards. There are 5
colony forming units on the sterile gloves. Should I forget that
practice and just go ahead and rinse my hands in nonsterile
water? And the opposite question is I can’t stand watching the
anesthesiologists put the A-line in with nonsterile gloves.
Should I go ahead and let them do that?
Dr Creamer: Those are good questions. It is very inter-
esting that in our study there was such a small difference
The American Journal of Surgery, Vol 204, No 6, December 2012
bacteria on clean versus sterile gloves and that raises the
question of: is that small difference in bacteria going to be
what sparks the infection that ultimately leads to your pa-
tient getting a wound infection. As for the anesthesiologists,
I think that again you have to take each situation as a
separate context, and I am unaware of any studies looking at
their use of nonsterile gloves. One of the studies evaluating
clean and sterile gloves showed a significant difference in
wound infections when more extensive procedures were
performed. In situations where there are more risk factors
involved, you should consider the use of sterile gloves over
clean gloves.
Dr Marie Allo (San Jose, CA): To turn Courtney’s
question to the other side, should we be scrubbing after we
put our gloves on?
Dr Creamer: It’s true, as we saw in our study, that if you
scrub and have your gloves put on by a scrub tech, chances
of having bacteria on your gloves are about zero. But to
apply that process to every procedure using gloves would
incur a significant cost, that appears is likely an unnecessary
cost to spend in order to prevent a wound infection.