Вы находитесь на странице: 1из 7
American Journal of Infection Control 41 (2013) 167-73 Contents lists available at ScienceDirect American Journal of

Contents lists available at ScienceDirect

American Journal of Infection Control

journal homepage: www.ajicjournal.org

American Journal of Infection Control
American Journal of
Infection Control

Review article

Preoperative chlorhexidine shower or bath for prevention of surgical site infection: A meta-analysis

Maciej Piotr Chlebicki MD a , Nasia Safdar MD, PhD b , c , d , * , John Charles O Horo MD e , Dennis G. Maki MD b , c

a Department of Infectious Diseases, Singapore General Hospital, Singapore

b Section of Infectious Diseases, Department of Medicine, University of Wisconsin Medical School, Madison, WI

c Infection Control Department, University of Wisconsin Hospital and Clinics, Madison, WI

d William S. Middleton Memorial Veterans Hospital, Madison, WI

e Department of Graduate Medical Education, Aurora Healthcare, Milwaukee, WI

Key Words:

Antiseptic

Systematic review

Clean

Clean-contaminated

Background: Chlorhexidine showering is frequently recommended as an important preoperative measure to prevent surgical site infection (SSI). However, the ef cacy of this approach is uncertain. Methods: A search of electronic databases was undertaken to identify prospective controlled trials evaluating whole-body preoperative bathing with chlorhexidine versus placebo or no bath for preven- tion of SSI. Summary risk ratios were calculated using a DerSimonian-Laird random effects model and a Mantel-Haenzel dichotomous effects model. Results: Sixteen trials met inclusion criteria with a total of 17,932 patients: 7,952 patients received a chlorhexidine bath, and 9,980 patients were allocated to various comparator groups. Overall, 6.8% of patients developed SSI in the chlorhexidine group compared with 7.2% of patients in the comparator groups. Chlorhexidine bathing did not signi cantly reduce overall incidence of SSI when compared with soap, placebo, or no shower or bath (relative risk, 0.90; 95% con dence interval: 0.77-1.05, P ¼ .19). Conclusions: Meta-analysis of available clinical trials suggests no appreciable bene t of preoperative whole-body chlorhexidine bathing for prevention of SSI. However, most studies omitted details of chlo- rhexidine application. Better designed trials with a speci ed duration and frequency of exposure to chlo- rhexidine are needed to determine whether preoperative whole-body chlorhexidine bathing reduces SSI. Copyright 2013 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.

Approximately 25.2 million inpatient surgical procedures were performed in the United States in 2002. 1 In the same year, almost 135,000 patients were discharged with a principal diagnosis of postoperative surgical site infection (SSI). The mean length of hospitalization for these patients was 7.5 days, and the mean hospital charges were $24,346, resulting in aggregate charges of almost $3.3 billion. 2 Patients who develop SSI have longer and costlier hospitalizations than patients who do not develop such infections, 3 are 60% more likely to spend time in an intensive care unit and twice as likely to die, and more than 5 times more likely to be readmitted to the hospital. 4 SSIs following clean surgery are caused in nearly all cases by endogenous organisms colonizing the patient s skin and are

* Address correspondence to Nasia Safdar, MD, PhD, Section of Infectious Diseases, MFCB University of Wisconsin Madison, 1685 Highland Avenue, Madison, WI 53705. E-mail address: ns2@medicine.wisc.edu (N. Safdar). Con icts of interest: None to report.

introduced into the surgical wound intraoperatively. 5 Most proce- dures for intraoperative surgical asepsis are designed to minimize cutaneous colonization of the surgical site and intraoperative contamination of the wound to prevent subsequent infection. Whole-body bathing or showering with an antiseptic agent, such as 2% to 4% chlorhexidine gluconate, has been shown to reduce bacterial colonization of the skin. 6 Studies have shown that the antibacterial effect of chlorhexidine is cumulative 7 and lasts longer than that produced by other antiseptic agents. 8 Whereas some studies have shown that preoperative chlorhexidine showers reduce the incidence of SSI, 9,10 others have found minimal or no clinically relevant bene t. 11-17 We have undertaken a meta-analysis to systematically review published, controlled, clinical trials eval- uating use of chlorhexidine baths or showers for prevention of SSI.

METHODS

We performed a search of MEDLINE, including PUBMED, and the COCHRANE NETWORK including the Cochrane Register of Clinical

0196-6553/$36.00 - Copyright 2013 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved. doi: 10.1016/j.ajic.2012.02.014

168

M.P. Chlebicki et al. / American Journal of Infection Control 41 (2013) 167-73

Trials from inception until July 30, 2011, using the following keywords: chlorhexidine, bath, shower, wash, scrub, disinfection, preoperative care, perioperative care, surgical site infection, and surgical wound infection. No language restrictions were applied. References from articles were reviewed to identify other potentially relevant articles. Our search strategy and analysis complied with the Preferred Reporting Items for Systematic Reviews and Meta- Analyses (PRISMA) statement. 18 Prospective randomized controlled trials and quasirandomized trials d those that allocated treatment by day of the week or by ward d as well as before-after trials comparing a preoperative whole-body chlorhexidine shower or bath with nonantiseptic soap or no bath and reporting SSI as an outcome were included in our review. We chose to include quasirandomized trials because these are common study designs that may yield useful information. Case reports, abstract-only, review articles, letters, and editorials were excluded. All authors independently reviewed each report identi- ed by the search strategy.

Study quality

We assessed study quality and assessed for potential bias using the recommendations of the Cochrane collaboration. 19 Selection bias was assessed by evaluations of subject allocation and blinding, especially relevant for nonrandomized studies. Performance bias d systematic differences in care provided apart from the intervention being evaluated d was evaluated by comparing the different bathing protocols and other interventions employed to reduce SSI in each study. Detection bias was evaluated by exam- ining the operational de nition for SSI. Reporting bias was analyzed using Egger s statistical test and a funnel plot to speci cally examine publication bias. 19

Statistical analysis

Summary risk ratios (RR) and their 95% con dence intervals (CI) were calculated using the DerSimonian and Laird random-effects model. 20 Heterogeneity was assessed with an I 2 statistic, where 0% indicates no heterogeneity, and 100% indicates the highest level of heterogeneity. 21 Sensitivity and subgroup analyses were performed to analyze sources of heterogeneity. Data analysis was performed using Cochrane Database s Review Manager 5.1.0 software. 19

RESULTS

Study selection

The literature search strategy yielded 857 potential articles. After screening and reviewing articles, 16 studies met our inclusion criteria. 9,10,12-17,22-29 The search strategy and reasons for exclusion are detailed in a PRISMA ow diagram in Figure 1. 30 In one of these trials, 26 the chlorhexidine group was compared with 2 control groups: 1 group that received only a limited, partial chlorhexidine body wash and another group that received no shower. We only included the chlorhexidine treatment group and the control group that received no shower in our statistical analysis.

Study characteristics

Eight randomized control trials and 8 quasiexperimental trials were identi ed. The 16 included trials enrolled 17,932 patients 6,10,12-17,22,24-29 : 7,952 patients received a chlorhexidine shower or bath, and 9,980 patients were allocated to various comparator groups of a preoperative shower or bath with

comparator groups of a preoperative shower or bath with Fig 1. PRISMA fl ow diagram of

Fig 1. PRISMA ow diagram of search strategy. Figure generated with Review Manager

5. 19

nongermicidal soap (n ¼ 5,696) or placebo (n ¼ 2,207) or no preoperative shower/noncompliance with regimen (n ¼ 2,077). Six studies 10,14,15,27-29 had patients bathe as outpatients prior to procedure, nine 11-13,16,17,22,24-26 had baths or showers as inpatients, and one 9 did not specify where the washes took place. There was considerable interstudy variation in the volume, timing, and/or number of applications as summarized in Table 1. In 7 trials, 12-14,16,17,22,26 patients had a single preoperative bath. Two baths were used in 5 trials, 10,15,27-29 and 3 trials 9,11,24 evaluated the effect of 3 or more preoperative chlorhexidine baths or showers. One did not specify the regimen. 25 In all trials using a liquid formulation of chlorhexidine, patients were required to apply the chlorhexidine solution or comparator

M.P. Chlebicki et al. / American Journal of Infection Control 41 (2013) 167-73

169

and then thoroughly rinse it off, without specifying the duration of continuous exposure of the skin to 4% chlorhexidine. Trials using chlorhexidine cloths instructed patients to use the product once the night before and once the morning of surgery without rinsing the wash off. Sponges or brushes were not used in any of the trials. Preoperative antibiotic prophylaxis was routinely administered in 6 trials 12,16,17,27-29 but was given to only in 12% of patients in the trial conducted by Rotter et al. 15 Ayliffe et al 22 and Lynch et al 24 mention that the frequency of antimicrobial prophylaxis was similar both in treatment and comparator groups, without providing data. The use of preoperative antimicrobial prophylaxis was not mentioned in the remaining 7 trials.

Details of randomization

The highest quality trials performed were a multicenter center, prospective, randomized, double-blind, and placebo-controlled trial by Rotter et al 15 and a single center, randomized controlled trial by Veiga et al. 16 In both studies, randomization was carried out in each participating surgical unit by means of computer-generated patient trial numbers, which were attached by manufacturer to the bottles containing either chlorhexidine or placebo. The trial by Lynch et al 24 was also prospective, randomized, double-blind, and placebo controlled, but no details of the randomization procedure were provided. Similarly, Randall et al 14 and Earnshaw et al 12 did not provide any details of the randomi- zation scheme. Murray et al 28 was randomized and used masked allocation but did not use an identical placebo. Wells et al 17 used a quasirandomized design in which patients were assigned to a chlorhexidine or soap bath by month. In a study conducted by Wihlborg, 26 participating patients were allocated to different wards by the chief surgeon. Finally, the 3 remaining trials used a design in which all patients admitted to selected surgical wards were allocated to a chlorhexidine bath or shower group, and patients admitted to other wards constituted the control group 10,22 with the ward allocations reversed after a speci ed period of time. Two studies 27,29 used a prospective intervention and historical controls.

De nitions of SSI

The investigators used varying de nitions of SSI in their trials ( Table 1 ). Six studies 10,14,15,22,27,29 used clinical criteria, such as wound in ammation, breakdown, or discharge, to de ne SSI. Although the trials did not explicitly state so, these criteria are similar to the CDC de nition for deep incisional SSI as follows:

Meet[ing] one of the following criteria: infection occurs

the operative procedure

the operative procedure and involves deep soft tissues (eg, fascial and muscle layers) of the incision and patient has at least 1 of the following: (1) purulent drainage from the deep incision but not from the organ/space component of the surgical site, (2) a deep incision spontaneously dehisces or is deliberately opened by a surgeon and is culture positive or not cultured and has 1 of the following signs or symptoms: fever, or localized pain, or tender- ness. A culture negative nding does not meet this criterion. (3) An abscess or other evidence of infection involving the deep incision is found on direct examination during reoperation or by histopatho- logic or radiologic examination. (4) Diagnosis of a deep incisional SSI by a surgeon or attending physician.31 Dimitrov et al, 25 Wihl- borg, 26 Earnshaw et al, 12 and Wells et al 17 de ned SSI as purulent wound discharge. Lynch et al 24 required the presence of purulent discharge and an ASEPSIS (Additional treatment, presence of Serous discharge, Erythema, Purulent exudate, Separation of the deep tissues, the Isolation of bacteria and the duration of inpatient

after

and infection appears to be related to

Stay) 32 score > 10 to de ne SSI. Murray et al used clinical diagnosis during surgical follow-up without a clear de nition of what constituted SSI. 28 Ten studies speci ed the duration of hospital follow-up to identify SSI. Four followed patients until discharge from the hospital, 12,13,17,22 1 followed patients for 7 days, 14 1 for 3 weeks, 15 1 for 1 month, 16 1 for 2 months, 28 and 2 for 6 weeks postoperatively. 10,11

Study quality

Nine of the included studies were determined to have a high risk of bias stemming from lack of randomization, use of historical controls, or lack of blinding when diagnosing SSI, and the remaining 7 studies were classi ed as low risk. The risk assess- ments of the individual studies are listed in Table 1. All studies had patients shower or bathe at home or in the hospital prior to surgery. Details of the protocols used are available in Table 1. All provided some instruction about proper bathing technique with chlorhexidine to subjects, but only one 29 performed an assessment of compliance, having patients remove stickers from the chlorhexidine package and af x them to a sheet af rming that the product was used.

Incidence of SSI

Three trials 9,10,26 showed a signi cant reduction in incidence of SSI among patients randomized to receive a preoperative chlo- rhexidine shower or bath. Table 2 shows the incidence of SSI in each study. Figure 2 shows the relative risk of SSI in the 16 included trials. Overall, 6.8% (543 of 7,952) of patients developed SSI in the chlorhexidine group compared with 7.2% (715 of 9,980) of patients in the comparator groups. The summary RR estimate of patients randomized to preoper- ative chlorhexidine showers/baths was 0.90 (95% CI: 0.77-1.05), failing to reject the null hypothesis ( P ¼ .19), shown in Figure 2 . The I 2 test for heterogeneity showed low heterogeneity (28%).

Subgroup and sensitivity analysis

Sensitivity analysis was performed, excluding studies sequen- tially to determine whether any individual study or subgroup had a disproportional impact on the outcome. Sensitivity analyses did not show signi cant impact from any study or subgroup. To determine whether there was a differential effect of chlo- rhexidine in patients undergoing clean surgery, we analyzed the ndings in the studies reporting the ef cacy of preoperative chlo- rhexidine showers or baths for prevention of SSI in patients undergoing clean surgery. This subgroup failed to reach the threshold for statistical signi cance, with whole-body preoperative chlorhexidine showering or shower demonstrating no clear bene t (RR, 0.88; 95% CI: 0.71-1.09, P ¼ .24). I 2 value was 38%, indicating moderate heterogeneity. The lack of bene t was seen also in contaminated and clean contaminated surgery (RR, 0.94; 95% CI:

0.76-1.16, P ¼ .55), with low heterogeneity (I 2 ¼ 0%). We conducted an analysis restricted to the high-quality, randomized controlled trials. 11,12,14-16,24,28 In these, no bene t of perioperative chlorhexidine washing or showering could be demonstrated (RR, 0.99; 95% CI: 0.85-1.16, P ¼ .90). Heterogeneity of these results was low with I 2 ¼ 0%. We also analyzed the 7 studies in which patients underwent 2 or 3 preoperative showers or baths. 9-11,15,24,27,29 The use of multiple chlorhexidine baths did not reach statistical signi cance, although there was a trend toward reduction of SSI (RR, 0.79; 95% CI: 0.60- 1.03, P ¼ .08).

M.P. Chlebicki et al. / American Journal of Infection Control 41 (2013) 167-73170

Table 1 Characteristics of the included studies

Chlorhexidine dosing

Authors, year

Study design

Type of surgery

Exclusion criteria

De nition of SSI

schedule

Comparator group (s)

Risk of bias

Brandberg et al, 9 1979 Prospective nonrandomized

Clean (vascular surgery)

Not speci ed

Pus collection at surgical wound site

Three to 8 chlorhexidine shower-baths prior to procedure Single bath or shower with undiluted 4% chlorhexidine (25 mL) on the day of or before operation Single bath or shower with undiluted 4% chlorhexidine (supplied in 250-mL bottle) a few hours before operation.

Soap and water wash

High

Ayliffe et al, 22 1983

Before/after, two 30-week periods

Clean, CL/CONT,

Trauma and emergency surgery

In ammation, breakdown or discharge from wound

Nonmedicated bar soap

High

contaminated

 

Leigh et al, 13 1983

Before/after, 2 wards, 4 months

Clean, CL/CONT,

Not stated

Not stated

Non-medicated soap

High

contaminated

 

Wells et al, 17 1983

Quasirandomization (by

Clean (cardiothoracic

Emergency surgery

Not stated

Single scrub with undiluted 4% chlorhexidine (30 mL)

Nonmedicated bar soap

High

month)

surgery)

 
 

1

day before operation

Randall et al, 14 1983

Randomized, (no details) controlled trial, 3 groups

Clean (vasectomy only)

Not stated

Discharge or wound breakdown

Single bath or shower 1 hour before surgery (no other details provided) Two baths or showers with undiluted 4% chlorhexidine (supplied in two 25-mL sachets) 1 day before and on the morning of operation Patients instructed to bath with chlorhexidine prior to surgery, regimen and concentration not speci ed Single shower (double application) on the afternoon of the day before surgery. Two baths or showers (2 applications each time) with undiluted 4% chlorhexidine (supplied in one 100-mL bottle) 1 day before and on the morning of operation

Two groups: nonmedicated bar soap and no shower

Low

Hayek et al, 10 1987

Before/after, 2-month

Clean, CL/CONT,

Emergency surgery,

High

periods

contaminated

concurrent infection

In ammation, breakdown or discharge from wound

Two groups: placebo and nonmedicated bar soap

Dimitrov et al, 25 1984

Prospective study, method of allocation not speci ed

CL/CONT (gastrointestinal

Not stated

Suppuraton at surgical site

Bar soap

High

and genitourinary

 

surgeries)

Wihlborg, 26 1987

Quasirandomization (by

Clean, CL/CONT (only biliary, hernia or breast surgery)

Not stated

De nite collection of pus.

No shower

High

ward)

   

Rotter et al, 15 1988

Randomized, multicenter,

Clean (various procedures) Fever, concurrent infection, recent antibiotics (7 days), incarcerated inguinal hernia or radical mastectomy

In ammation, or discharge from wound

Identical placebo

Low

double-blind, placebo

 

controlled

 

Earnshaw et al, 12 1989 Randomized (no details), controlled trial

Clean (vascular surgery)

Not stated

Purulent (but not serous) wound discharge

Single bath or shower with

Nonmedicated soap

Low

 

2

applications of

undiluted 4% chlorhexidine (no other details provided) Three showers with 4% chlorhexidine prior to surgery Three baths or showers with undiluted 4% chlorhexidine on admission, before going to bed and on the morning of operation

 

Byrne et al, 11 1991

Randomized controlled

Clean (hernia surgery)

Not stated

Not stated

Identical placebo

Low

trial

 

Lynch et al, 24 1992

Randomized (no details), double-blind, placebo controlled

Clean, CL/CONT,

Not stated

Nonmedicated liquid soap

Low

contaminated

Purulent discharge or ASEPSIS score > 10

 

ASEPSIS, Additional treatment, presence of Serous discharge, Erythema, Purulent exudate, Separation of the deep tissues, Isolation of bacteria, and the du ration ofinpatient Stay; CL, Clean; CONT, contaminated; SSI, surgical siteinfection.

High

High

Low

Low

Soap and water shower

Shower with identical placebo

Noncompliance with

Noncompliance with

regimen

regimen

Shower with liquid detergent based chlorhexidine 4% liquid detergent 2 hours before surgery Patients were instructed to apply 2% chlorhexidine cloths the night before and morning of surgery to the head and trunk, each arm and leg, and the surgical site. Patients were instructed to apply 2% chlorhexidine cloths the night before and morning of surgery within 2 hours of surgery to the surgical site and ipsilateral arm, axilla, chest and back. Patients were instructed to apply 2% chlorhexidine cloths the night before and morning of surgery for the neck, chest, abdomen, neck, each extremity, and the surgical site.

Deep infection of joint space as de ned by CDC

Deep infection of joint space as de ned by CDC

CDC de nition

Not stated

Chlorhexidine allergies, skin lesions, antibiotics at time of surgery, diabetes, heavy smoking and immunosuppression Not stated

Current infection, open wound, chronic immunosuppressive condition

Known active joint infection at time of surgery

Clean (knee arthrosplasty and revision)

Elective clean plastic surgical procedures of the trunk

Clean (shoulder surgery)

Clean (total hip arthroplasty)

Randomized control trial

Single blind randomized controlled trial

prospective cohort

prospective cohort

Nonrandomized

Nonrandomized

Johnson et al, 27 2010

Murray et al, 28 2011

Zywiel et al, 29 2011

Veiga et al, 16 2009

M.P. Chlebicki et al. / American Journal of Infection Control 41 (2013) 167-73

171

Three studies used a 2% chlorhexidine impregnated cloth product rather than a liquid preparation used in the other studies. Analysis restricted to studies that used the cloth product (n ¼ 3) did not show a reduction ( P ¼ .29) in incidence of SSI.

Publication bias

Publication bias was assessed with a funnel plot ( Fig 3 ). 33 This is consistent with some degree of publication bias.

Adverse effects of chlorhexidine

In only 1 trial 15 were adverse effects of preoperative showering or bathing with chlorhexidine reported. In that trial, itching or reddening of skin occurred in 5 patients (0.36%) of the placebo group and 5 (0.36%) in chlorhexidine group. No adverse effects attributed to the chlorhexidine shower or bath were mentioned in the remaining 15 trials.

DISCUSSION

Our analyses show that preoperative whole-body showering or bathing with chlorhexidine, in its current nonspeci ed manner, is of no bene t for prevention of postoperative SSI. Our ndings agree with and extend those of a Cochrane systematic review that included data from 9 trials from 1983 through 2005 and failed to show a bene t for chlorhexidine bathing or showering. 34 Our review extends the literature search to July 2011 and identi ed 4 additional trials since the Cochrane review, thus increasing the statistical power of the analysis to detect a clinically relevant reduction in SSI if one exists. Our study also included quasiexper- imental designs that were excluded in Cochrane s methodology. Given the known ef cacy of chlorhexidine in reducing bacterial burden, it is somewhat surprising that this meta-analysis demon- strated no clinical bene t. Several factors may have contributed to this. Many of the trials studied patients undergoing clean- contaminated or contaminated types of surgery, where the patient s cutaneous ora contributes negligibly if at all to the incidence of SSI. 35,36 Also, the authors of most of the studies did not provide speci c data on the proportion of patients in each group that received perioperative antimicrobial prophylaxis. 35,37 Major disparities between study groups in frequency of antimicrobial prophylaxis are unlikely in randomized studies; however, it is very possible that signi cant disparities in the use of antimicrobial prophylaxis could have occurred in studies using a time-sequence, before-after design and masked a true bene cial effect of preop- erative chlorhexidine showering or bathing. It is also worth noting that chlorhexidine may have been employed in a less than optimal fashion in several studies. In all the trials, patients were simply instructed to bathe or shower with the study agent in the usual fashion, not speci cally instructed to keep the 4% chlorhexidine on their skin for several minutes before rinsing. Moreover, the application of chlorhexidine was not supervised in the majority of trials. It is possible that the duration of chlorhexidine application was considerably less than adequate, effective cutaneous antisepsis. Studies evaluating the optimal duration of chlorhexidine hand scrubs provide insights into this potentially very important issue. Pereira et al showed that a 5-minute initial hand scrub followed by consecutive 3-minute scrubs resulted in far lower bacterial counts on hands than a 3-minute initial scrub followed by consecutive 30- second scrubs, 38 suggesting that application of 4% chlorhexidine to the skin for 3 to 5 minutes provides optimal antibacterial effect on the cutaneous micro ora. It is very plausible that the patients enrolled in the analyzed trials received far too brief application of

172

Table 2 Results of the included studies

M.P. Chlebicki et al. / American Journal of Infection Control 41 (2013) 167-73

Authors, year

Duration of follow up for determination of SSI

Number of patients in chlorhexidine arm

Number of SSI in chlorhexidine arm (%)

Number of patients in comparator arm

Number of SSI in comparator arm (%)

Brandberg et al, 9 1979 Ayliffe et al, 22 1983

Not speci ed Through hospital discharge Through hospital discharge Through postoperative day 10 or hospital discharge Seven days Six-week follow-up Not speci ed Not speci ed Three weeks Through hospital discharge Six-week follow-up Not speci ed Thirty days follow-up Not speci ed Two months Not speci ed

171

13 (7.6)

170

30 (17.6)

2,703

147 (5.4)

2,833

140 (4.9)

Leigh et al, 13 1983

109

12 (11.0)

115

13 (11.3)

Wells et al, 17 1983

209

11 (5.3)

245

14 (6.7)

Randall et al, 14 1983 Hayek et al, 10 1987 Dimitrov et al, 25 1984 Wihlborg, 26 1987 Rotter et al, 15 1988 Earnshaw et al, 12 1989

32

12 (37.5)

62

19 (30.6)

689

62 (9.0)

1,326

163 (12.3)

57

0 (0)

46

0 (0)

541

9 (1.7)

437

20 (4.6)

1,413

37 (2.6)

1,400

33 (2.4)

31

8 (25.8)

35

4 (11.4)

Byrne et al, 11 1991 Lynch et al, 24 1992 Veiga et al, 16 2009 Johnson et al, 27 2010 Murray et al, 28 2011 Zywiel et al, 29 2011

29

2 (6.9)

27

2 (7.4)

1,575

225 (14.3)

1,576

241 (15.3)

50

1 (2)

50

1 (2)

157

0 (0)

897

14 (1.6)

50

0 (0)

50

0 (0)

136

0 (0)

711

21 (3.0)

4% chlorhexidine, which might well explain the failure to detect any bene t in the pooled data. The antibacterial effect of chlorhexidine is cumulative and greatly enhanced by repeated applications. Paulson 39 evaluated the

effect of chlorhexidine showers daily over 5 days and found that, as the study progressed, ever greater microbial reductions from the baseline were achieved vis-à-vis a cumulative effect. It is very plausible that multiple applications of chlorhexidine over several

that multiple applications of chlorhexidine over several Fig 2. Number of infections in clean surgery and

Fig 2. Number of infections in clean surgery and clean contaminated/contaminated subgroups. When excluding higher risk of bias studies, OR, 0.98 (95% CI: 0.83-1.16) Figure generated with Review Manager 5 software. 19

M.P. Chlebicki et al. / American Journal of Infection Control 41 (2013) 167-73

173

/ American Journal of Infection Control 41 (2013) 167-73 173 Fig 3. Funnel plot. Figure generated

Fig 3. Funnel plot. Figure generated with Review Manager 5 software. The funnel plot functions as a measure of publication bias, graphing the effect size from relative risk versus the standard error of the log of the relative risk. Because large studies should have a smaller SE(log)OR, nding studies falling outside the funnel, as is the case here, makes publication bias more probable. 19

days prior to anticipated surgery may be far more ef cacious than

a single preoperative chlorhexidine shower or bath. There was

a trend toward fewer infections in individuals receiving multiple

applications compared with single applications in our study, although this failed to reach statistical signi cance. In conclusion, our analysis does not support routine use of preoperative, whole-body chlorhexidine showering or bathing for prevention of SSI. However, many of the trials done were suboptimal in design, and the manner in which the chlorhexidine solutions

were used in the trials was not clearly described. Although chlo- rhexidine bathing preoperatively is a low risk and relatively low cost intervention that may be employed even if the magnitude of bene t

is marginal, it is nonetheless important to de nitively answer this

question, given constrained resources for infection control. A cluster randomized trial to provide adequate statistical power in patients

undergoing clean surgery that employs standardized application techniques and a prolonged duration of chlorhexidine with rigorous blinded assessment of SSI outcomes is needed.

References

1. Darouiche RO, Wall MJ Jr, Itani KM, Otterson MF, Webb AL, Carrick MM, et al. Chlorhexidine-alcohol versus povidone-iodine for surgical site antisepsis. N Engl J Med 2010;362:18-26.

2. Anderson DJ, Kaye KS, Classen D, Arias KM, Podgorny K, Burstin H. Strategies to prevent surgical site infections in acute care hospitals. Infect Control Hosp Epidemiol 2008;29(Suppl 1):S51-61.

3. Whitehouse JD, Friedman ND, Kirkland KB, Richardson WJ, Sexton DJ. The impact of surgical-site infections following orthopedic surgery at a community hospital and a university hospital: adverse quality of life, excess length of stay, and extra cost. Infect Control Hosp Epidemiol 2002;23:183-9.

4. Alonso-Echanove J, Edwards JR, Richards MJ, Brennan P, Venezia RA, Keen J, et al. Effect of nurse staf ng and antimicrobial-impregnated central venous catheters on the risk for bloodstream infections in intensive care units. Infect Control Hosp Epidemiol 2003;24:916-25.

5. Jakob HG, Borneff-Lipp M, Bach A, von Pückler S, Windeler J, Sonntag H, et al. The endogenous pathway is a major route for deep sternal wound infection. Eur J Cardiothorac Surg 2000;17:154-60.

6. Brandberg A, Andersson I. Whole body disinfection by shower-bath with chlorhexidine soap. Paper presented at The Royal Society of Medicine Inter- national Congress and Symposium; 1979.

7. Byrne DJ, Napier A, Cuschieri A. Rationalizing whole body disinfection. J Hosp Infect 1990;15:183-7.

8. Aly R, Maibach HI. Comparative antibacterial ef cacy of a 2-minute surgical scrub with chlorhexidine gluconate, povidone-iodine, and chloroxylenol sponge-brushes. Am J Infect Control 1988;16:173-7.

9. Brandberg A, Holm J, Hammarsten J, Schersten T. Post-operative wound infections in vascular surgery: effect of pre-operative whole body disinfection by shower-bath with chlorhexidine soap. The Royal Society of Medicine International Congress and Symposium. Int Congr Symp 1979;23:71-5.

10. Hayek LJ, Emerson JM, Gardner AM. A placebo-controlled trial of the effect of two preoperative baths or showers with chlorhexidine detergent on post- operative wound infection rates. J Hosp Infect 1987;10:165-72.

11. Byrne DJ, Phillips G, Napier A, Cuschieri A. The effect of whole body disinfection on intraoperative wound contamination. J Hosp Infect 1991;18:145-8.

12. Earnshaw JJ, Berridge DC, Slack RC, Makin GS, Hopkinson BR. Do preoperative

chlorhexidine baths reduce the risk of infection after vascular reconstruction? Eur J Vasc Surg 1989;3:323-6.

13. Leigh DA, Stronge JL, Marriner J, Sedgwick J. Total body bathing with Hibis- crub (chlorhexidine) in surgical patients: a controlled trial. J Hosp Infect 1983;

4:229-35.

14. Randall PE, Ganguli L, Marcuson RW. Wound infection following vasectomy. Br

J Urol 1983;55:564-7.

15. Rotter ML, Larsen SO, Cooke EM, et al. A comparison of the effects of preop- erative whole-body bathing with detergent alone and with detergent con-

taining chlorhexidine gluconate on the frequency of wound infections after clean surgery. The European Working Party on Control of Hospital Infections.

J Hosp Infect 1988;11:310-20.

16. Veiga DF, Damasceno CA, Veiga-Filho J, et al. Randomized controlled trial of the effectiveness of chlorhexidine showers before elective plastic surgical proce- dures. Infect Control Hosp Epidemiol 2009;30:77-9.

17. Wells FC, Newsom SW, Rowlands C. Wound infection in cardiothoracic surgery. Lancet 1983;1:1209-10.

18. Moher D, Altman DG, Liberati A, Tetzlaff J. PRISMA statement. Epidemiology

2011;22:128.

19. Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 (updated March 2011). The Cochrane Collabora-

tion, 2011. Available from: http://www.cochrane-handbook.org .

20. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;

7:177-88.

21. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557-60.

22. Ayliffe GA, Noy MF, Babb JR, Davies JG, Jackson J. A comparison of pre-operative bathing with chlorhexidine-detergent and non-medicated soap in the prevention of wound infection. J Hosp Infect 1983;4:237-44.

23. Gould DJ, Moralejo D, Drey N, Chudleigh JH. Interventions to improve hand hygiene compliance in patient care. Cochrane Database Syst Rev 2010;(9):

CD005186.

24. Lynch W, Davey PG, Malek M, Byrne DJ, Napier A. Cost-effectiveness analysis of the use of chlorhexidine detergent in preoperative whole-body disinfection in wound infection prophylaxis. J Hosp Infect 1992;21:179-91.

25. Dimitrov I, Tashev P, Madzharova S. Importance of the preoperative bathing of patients with the Hibiscrub preparation: a bacteriological and clinical study. Khirurgiia (So ia) 1984;37:355-8.

26. Wihlborg O. The effect of washing with chlorhexidine soap on wound infection rate in general surgery: a controlled clinical study. Ann Chir Gynaecol 1987;76:

263-5.

27. Johnson AJ, Daley JA, Zywiel MG, Delanois RE, Mont MA. Preoperative chlo- rhexidine preparation and the incidence of surgical site infections after hip arthroplasty. J Arthroplasty 2010;25(6 Suppl):98-102.

28. Murray MR, Saltzman MD, Gryzlo SM, Terry MA, Woodward CC, Nuber GW.

Ef cacy of preoperative home use of 2% chlorhexidine gluconate cloth before

shoulder surgery. J Shoulder Elbow Surg 2011;20:928-33.

29. Zywiel MG, Daley JA, Delanois RE, Naziri Q, Johnson AJ, Mont MA. Advance pre- operative chlorhexidine reduces the incidence of surgical site infections in knee arthroplasty. Int Orthop 2011;35:1001-6.

30. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009;339:

b2535.

31. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection, 1999. Centers for Disease Control and

Prevention (CDC) Hospital Infection Control Practices Advisory Committee. Am

J Infect Control 1999;27:97 e 132; quiz 133-4; discussion 196.

32. Wilson AP, Treasure T, Sturridge MF, Gruneberg RN. A scoring method (ASEPSIS) for postoperative wound infections for use in clinical trials of anti- biotic prophylaxis. Lancet 1986;1:311-3.

33. Petitti DB. Meta-analysis, decision analysis, and cost-effectiveness analysis. Second Edition. New York [NY]: Oxford University Press; 2000.

34. Webster J, Osborne S. Preoperative bathing or showering with skin antiseptics to prevent surgical site infection. Cochrane Database Syst Rev 2007;(2):CD004985.

35. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline for preven- tion of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol 1999;20:250-78. quiz 279 e 80.

36. Popovich KJ, Hota B, Hayes R, Weinstein RA, Hayden MK. Daily skin cleansing with chlorhexidine did not reduce the rate of central-line associated blood- stream infection in a surgical intensive care unit. Intensive Care Med 2010;36:

854-8.

37. Bratzler DW, Houck PM. Antimicrobial prophylaxis for surgery: an advisory statement from the National Surgical Infection Prevention Project. Clin Infect Dis 2004;38:1706-15.

38. Maldonado J, Pereira T. Self-measurement of blood pressure in arterial hypertension: preliminary results from the AMPA study. Rev Port Cardiol 2009;

28:7-21.

39. Paulson DS. Ef cacy evaluation of a 4% chlorhexidine gluconate as a full-body shower wash. Am J Infect Control 1993;21:205-9.