Evidence-Based Bundles and Cesarean Delivery Surgical Site Infections A Systematic Review and Meta-analysis

OBJECTIVE: To estimate the association of implementation of evidence-based bundles with surgical site infection rates after cesarean delivery. DATA SOURCES: We searched MEDLINE through PubMed, EMBASE, Scopus, the Cochrane Database of Systematic Reviews, Google Scholar, and ClinicalTrials. gov. METHODS OF STUDY SELECTION: We searched electronic databases for randomized controlled trials and observational studies comparing evidence-based infection prevention bundles for cesarean delivery, defined as implementation of three or more processes proven to prevent surgical site infection such as chlorhexidine skin preparation, antibiotic prophylaxis, and hair clipping, with usual care. The primary outcome was overall surgical site infection, defined using Centers for Disease Control and Prevention's National Healthcare Safety Network criteria. Secondary outcomes were superficial or deep surgical site infection and endometritis. Quality of studies and heterogeneity were assessed using validated measures. Pooled relative risks (RRs) with 95% CIs were calculated using random-effects models. Numbers needed to treat were estimated for outcomes with significant reduction. TABULATIONS, INTEGRATION, AND RESULTS: We found no randomized controlled trials. Fourteen preintervention and postintervention studies met inclusion criteria. Eight were full-text articles, and six were published abstracts. Quality of most of the primary studies was adequate with regard to the intervention, but modest in terms of implementation. The rate of surgical site infection was significantly lower after implementing an evidence-based bundle (14 studies: pooled rates 6.2% baseline compared with 2.0% intervention, pooled RR 0.33, 95% CI 0.25-0.43, number needed to treat=24). Evidence-based bundles were also associated with a lower rate of superficial or deep surgical site infection (six studies: pooled rate 5.9% baseline compared with 1.1% intervention, pooled RR 0.19, 95% CI 0.12-0.32, number needed to treat=21). The rate of endometritis was low at baseline and not significantly different after intervention (six studies: pooled rate 1.3% baseline compared with 0.9% intervention, pooled RR 0.57, 95% CI 0.31-1.06). CONCLUSION: Evidence-based bundles are associated with a significant reduction in surgical site infection after cesarean delivery.