We report that there is moderate-quality evidence supporting the use of chlorhexidine over iodine for preoperative skin antisepsis to prevent SSI. Additionally, there is moderate quality evidence that use of chlorhexidine is associated with fewer positive skin culture results after application.
There was a 36% reduction in the number of SSIs among patients who received preoperative skin antisepsis with chlorhexidine, compared with those who received iodine. This estimate was similar for most studies, regardless of the surgical procedure involved, the concentration of chlorhexidine used, and whether the chlorhexidine preparation included alcohol. Three studies had point estimates differing from the pooled estimate.20,21,23
These studies had very few events: there were a total of 4 SSIs in the study by Veiga and colleagues,21
1 SSI in the study by Ostrander and colleagues,23
and no SSIs in the study by Saltzmann and colleagues.20
This dramatic reduction associated with chlorhexidine use has potentially significant implications. One drawback of preoperative skin antisepsis with chlorhexidine has been cost-related. However, given the high costs and increased length of hospitalization associated with SSI, preventing these infections would likely result in significantly decreased lengths of stay after surgery and overall cost savings.
In this era of cost containment, demonstrating the economic value of an infection control intervention is pivotal to driving its adoption. Our study suggests that switching from iodine to the more expensive chlorhexidine can actually provide net cost savings for a hospital or healthcare system—compelling evidence to make such a switch. These findings were fairly robust with respect to changes in key variables, such as the efficacy of chlorhexidine in preventing SSIs, the incremental cost of SSI, and the incremental cost of chlorhexidine.
Use of chlorhexidine was also associated with an overall 56% reduction in positive skin culture results after skin preparation. This reduction was likely greater than the reduction in number of SSIs, because preoperative skin antisepsis is typically used to sterilize the surgical field in hopes of preventing surgical wound contamination, which can then result in SSI. Despite the adequate efficacy of skin antiseptics, SSI may still occur in several ways. Wounds can become contaminated from another source during surgery (eg, by gastrointestinal flora during abdominal surgery) or can become contaminated after surgery, after the effects of the skin antiseptic have worn off.
Our findings are comparable to those reported in studies evaluating the use of skin antiseptics for decreasing the incidence of catheter-associated bloodstream infection. The purpose of using chlorhexidine and iodine for catheter-site care is similar to the purpose for their use preoperatively: to decrease the level of bacterial contamination of the skin. Chaiyakunapruk and colleagues26
conducted a meta-analysis of 8 RCTs and reported that catheter-site care with chlorhexidine was associated with significantly decreased risk for catheter-related bloodstream infection, compared with use of iodine (RR, 0.49 [95% CI, 0.28–0.88]). They hypothesized that chlorhexidine may have higher efficacy because protein-rich biomaterials found in blood and on skin may decrease the antimicrobial effects of povidone-iodine but not those of chlorhexidine,27,28
and chlorhexidine may have a longer duration of activity.29
There are several potential limitations to our study. First, this systematic review did not include studies published in languages other than English and did not include “gray” literature. The effect of excluding trials published in languages other than English in systematic reviews and meta-analyses remains uncertain, with conflicting reports as to whether it affects overall results or effect sizes.11,30-33
Additionally, there is conflicting evidence regarding whether the quality of studies published in languages other than English may differ on the basis of the intervention studied.34,35
Given these uncertainties, coupled with the difficulties associated with obtaining and accurately translating manuscripts published in languages other than English, we elected to limit our systematic review to studies published in English. We also excluded “gray” literature (such as conference abstracts, unpublished studies, or data obtained from personal communications) because these have not undergone peer review and thus the validity of their results may be less certain.
Second, there were 3 studies that contributed more than 75% of the patients in the meta-analysis. However, almost all of the studies favored use of chlorhexidine, and there was no significant study heterogeneity identified using the I2 test and the χ2 test. Third, only a single reviewer screened the potential articles for inclusion. However, we verified that all relevant articles were captured by reviewing the reference lists of the systematic review captured by the search of the Cochrane Library, and the reference lists of all included RCTs. These additional measures did not yield any additional studies for inclusion. Last, the base case estimates for our cost analysis were derived from our institution (HUP). However, to ensure generalizability of our results to other institutions, we performed 1-way and 2-way sensitivity analyses using ranges derived from the existing literature, and obtained similar results.
Use of chlorhexidine for preoperative skin antisepsis is associated with a 36% reduction in the number of SSIs, compared with use of iodine. Although chlorhexidine is more costly than iodine, this dramatic reduction in the number of SSIs will likely result in greater overall cost savings with chlorhexidine use. Given the clinical and economic benefits, use of chlorhexidine should be considered over use of iodine for standard preoperative skin antisepsis. However, further studies are needed to evaluate what preparation of chlorhexidine (eg, what concentration and whether the preparation includes alcohol) is most effective in decreasing the incidence of SSI.