Efficacy of different preoperative skin antiseptics on the incidence of surgical site infections: a systematic review, GRADE assessment, and network meta-analysis

Surgical site infection (SSI) is the most common postoperative complication and substantially increases health-care costs. Published meta-analyses and international guidelines differ with regard to which preoperative skin antiseptic solution and concentration has the highest efficacy. We aimed to compare the efficacy of different skin preparation solutions and concentrations for the prevention of SSIs, and to provide an overview of current guidelines. This systematic review and network meta-analysis compared different preoperative skin antiseptics in the prevention of SSIs in adult patients undergoing surgery of any wound classification. We searched for randomised controlled trials (RCTs) in MEDLINE, Embase, and Cochrane CENTRAL, published up to Nov 23, 2021, that directly compared two or more antiseptic agents (ie, chlorhexidine, iodine, or olanexidine) or concentrations in aqueous and alcohol-based solutions. We excluded paediatric, animal, and non-randomised studies, and studies not providing standard preoperative intravenous antibiotic prophylaxis. Studies with no SSIs in both groups were excluded from the quantitative analysis. Two reviewers screened and reviewed eligible full texts and extracted data. The primary outcome was the occurrence of SSI (ie, superficial, deep, and organ space). We conducted a frequentist random effects network meta-analysis to estimate the network effects of the skin preparation solutions on the prevention of SSIs. A risk-of-bias and Grading of Recommendations, Assessment, Development, and Evaluation assessment were done to determine the certainty of the evidence. This study is registered with PROSPERO, CRD42021293554. Overall, 2326 articles were identified, 33 studies were eligible for the systematic review, and 27 studies with 17 735 patients reporting 2144 SSIs (overall incidence of 12·1%) were included in the quantitative analysis. Only 2·0–2·5% chlorhexidine in alcohol (relative risk 0·75, 95% CI 0·61–0·92) and 1·5% olanexidine (0·49, 0·26–0·92) significantly reduced the rate of SSIs compared with aqueous iodine. For clean surgery, we found no difference in efficacy between different concentrations of chlorhexidine in alcohol. Seven RCTs were at high risk of bias, 24 had some concerns, and two had low risk of bias. Heterogeneity across the studies was moderate (I2=27·5%), and netsplitting did not show inconsistencies between direct and indirect comparisons. Five of ten studies that mentioned adverse events related to t