This study aimed to systematically review the relative effectiveness of preincision cefazolin with or without adjunctive prophylaxis (macrolides or metronidazole) vs cefazolin alone in decreasing the incidence of postcesarean delivery surgical site infections.

We performed a systematic search on PubMed, Ovid EMBASE, Google Scholar, ClinicalTrials.gov, and the Cochrane Central Register of Controlled Trials from October 25, 2020, to November 25, 2020, to identify studies comparing cefazolin with adjunctive macrolides or metronidazole with cefazolin alone. The reference lists were reviewed, and a manual search of articles published after the last database search was performed.

Overall, 3 randomized controlled trials and 1 prospective observational study of reproductive-age women undergoing cesarean deliveries were included in the study. We excluded studies of women who were immunocompromised (eg, patients who were HIV positive) or women with a diagnosis of chorioamnionitis before cesarean delivery. All patients received first-line cefazolin (either cefazolin 1 g or 2 g). We compared preincision cefazolin alone with preincision cefazolin plus adjunctive therapy (500 mg, oral or intravenous formulations of azithromycin, metronidazole, or clarithromycin).

A total of 6 review authors independently assessed the risk of bias for each study, using the Cochrane Risk of Bias criteria. Synthesis and further appraisal were done using the Grading of Recommendations, Assessment, Development, and Evaluation levels and the American College of Obstetricians and Gynecologists appraisal guidelines. Disagreements were resolved by discussion. Treatment effects were evaluated using meta-analysis, and pooled relative risks and 95% confidence intervals were generated using random-effects models using the Review Manager 5 software (version 5.4.1).

Overall, 3 randomized controlled trials and 1 prospective observational study representing 2613 women met the criteria for inclusion. Significant reductions in surgical site infections (relative risk, 0.46; 95% confidence interval, 0.34–0.63; 3 randomized controlled trials) and the duration of hospital stay (weighted mean difference, −1.46; 95% confidence interval, −2.21 to −0.71; 2 randomized controlled trials) were observed with preincision cefazolin and adjunctive prophylaxis compared with cefazolin alone. No significant difference was observed in maternal febrile morbidity (relative risk, 0.38; 95% confidence interval, 0.11–1.25; 2 randomized controlled trials).

Our findings have provided evidence for the use of preincision adjunctive extended-spectrum prophylaxis with cefazolin over cefazolin alone. However, future investigations are required to establish the relative efficacies of different adjunctive antibiotic options.