Intrawound vancomycin for prevention of surgical site infection: meta-analysis of randomized clinical trials.

Abstract

Background: Surgical site infection remains a major cause of postoperative morbidity. Intrawound vancomycin has been proposed as an adjunct to standard prophylaxis, but its effectiveness across surgical settings remains uncertain. This systematic review and meta-analysis of randomized clinical trials aimed to evaluate the association between intraoperative intrawound vancomycin and surgical site infection in adults undergoing surgery.

Methods: MEDLINE, Embase, and CENTRAL were searched from database inception to March 2026 for randomized clinical trials comparing intrawound vancomycin with standard care in adults undergoing operative procedures. Two reviewers independently performed screening, data extraction, and risk-of-bias assessment using the Cochrane Risk of Bias 2 tool. Random-effects meta-analyses were undertaken, with certainty of evidence rated using the Grading of Recommendations Assessment, Development and Evaluation framework. Primary outcomes were overall surgical site infection, deep surgical site infection, superficial surgical site infection, and Gram-positive surgical site infection.

Results: Twenty-two trials enrolling 8424 patients (vancomycin 4228, control 4196) were included. Intrawound vancomycin was associated with a lower risk of overall surgical site infection (risk ratio 0.79, 95% confidence interval 0.64 to 0.99) and Gram-positive surgical site infection (risk ratio 0.53, 0.35 to 0.79), corresponding to numbers needed to treat of 82 and 78, respectively. No significant differences were observed for deep or superficial surgical site infection, Gram-negative infection, reoperation, wound complications, mortality, or length of hospital stay. The certainty of evidence was moderate.

Conclusion: Intrawound vancomycin was associated with a modest reduction in surgical site infection, driven by reductions in Gram-positive infections. These findings support a selective, context-specific use rather than routine application across all procedures.