Antimicrobial dressings for the prevention of surgical-site infection: systematic review and meta-analysis

BACKGROUND: Surgical-site infections (SSIs) are a leading cause of postoperative morbidity, prolonged hospitalization, and significant healthcare costs. Antimicrobial dressings may mitigate SSI risk by minimizing local microbial burden, but their effectiveness remains uncertain. METHODS: MEDLINE, Embase, CINAHL, and Cochrane CENTRAL were searched from inception up to 5 July 2024, supplemented by searches of the grey literature and citation chasing. RCTs comparing antimicrobial dressings with any other inert dressing for closed incisional wounds after elective or emergency surgery were included. Data were extracted in duplicate, risk of bias (RoB) was assessed using the Cochrane RoB 2 tool, and certainty of evidence was rated using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) tool. Pairwise random-effects meta-analyses were performed. The primary outcome was SSI within 30 days (90 days if prosthesis used); secondary outcomes included adverse events and costs. RESULTS: A total of 35 RCTs involving 8718 participants were included. Meta-analyses were conducted for silver (21 studies, 5504 participants), dialkylcarbamoyl chloride (DACC) (3 studies, 892 participants), and mupirocin (3 studies, 1320 participants). Compared with standard dressings, silver dressings reduced the risk of SSI by 22% (risk ratio (RR) 0.78 (95% c.i. 0.62 to 0.97); I2 = 43%; moderate-certainty evidence). DACC dressings halved the risk of SSI (RR 0.49 (95% c.i. 0.29 to 0.83); I2 = 0%; moderate-certainty evidence) compared with standard dressings. Mupirocin dressings were not associated with SSI prevention (RR 0.62 (95% c.i. 0.15 to 2.63); I2 = 67%; very low-certainty evidence). Evidence for other agents was insufficient. CONCLUSION: Silver and DACC dressings probably reduce the risk of SSI in closed surgical wounds. There is residual uncertainty over the clinical effectiveness of other antimicrobial dressings.