Vaginal antiseptic preparation at the time of hysterectomy: a systematic review and meta-analysis

Abstract

Objective

This study aimed to compare the rates of surgical site infection after hysterectomy using vaginal antisepsis with chlorhexidine gluconate vs povidone-iodine.

Data Sources

PubMed, Embase, and ClinicalTrials.gov databases were queried from January 1, 1985 through December 7, 2023.

Study Eligibility Criteria

Randomized controlled trials and nonrandomized comparisons of interventions of chlorhexidine gluconate and povidone-iodine vaginal antiseptic preparation before hysterectomy were included. The primary outcome was surgical site infection as defined by the Centers for Disease Control and Prevention. Secondary outcomes included postoperative urinary tract infections and vaginal culture specimens. Large single-group studies of vaginal chlorhexidine gluconate reporting adverse events and case reports of desquamation were also included.

Methods

The methodologic quality of each study was assessed using the original Cochrane RoB (Risk of Bias) and ROBINS-I (Risk of Bias in Non-Randomized Studies - of Interventions) tools. Restricted maximum likelihood meta-analyses of odds ratios were conducted.

Results

A total of 5289 abstracts were screened; 10 met the inclusion criteria, with a total of 9618 participants. The studies included 4 randomized controlled trials (n=306), 3 prospective nonrandomized comparisons of interventions (n=1089), and 3 retrospective nonrandomized comparisons of interventions (n=8223). Povidone-iodine was compared with 4% chlorhexidine gluconate in 4 studies (n=2491), 2% chlorhexidine gluconate in 2 studies (n=236), 0.1% chlorhexidine gluconate in 1 study (n=50), and both 2% and 4% chlorhexidine gluconate in 1 study (n=49). Meta-analysis revealed no statistically significant difference in surgical site infections, although they were somewhat more common with chlorhexidine gluconate (summary odds ratio, 1.22; 95% confidence interval, 0.91–1.63). The relative effect of antiseptic preparations on urinary tract infections was unclear, with an imprecise summary odds ratio (1.18; 95% confidence interval, 0.65–2.12). Positive vaginal cultures were less common with chlorhexidine gluconate preparation (summary odds ratio, 0.10; 95% confidence interval, 0.04–0.27). Two studies reported no adverse events with chlorhexidine gluconate, and 2 found no difference in vaginal irritation when compared with povidone-iodine. Two case reports described vaginal desquamation or hypersensitivity with chlorhexidine gluconate.

Conclusion

There is no evidence of a difference in postoperative infection with chlorhexidine gluconate use compared with povidone-iodine, but chlorhexidine gluconate vaginal preparation results in lower rates of positive intraoperative vaginal cultures. Despite inadequate reporting, risk of adverse events appears low.