Adding Preoperative Oral Antibiotics to Mechanical Bowel Preparation Reduces Surgical Site Infections in Elective Colorectal Surgery: A Meta-Analysis of Randomized Controlled Trials.

内容摘要

Background and Objectives: Surgical site infections (SSIs) remain common after elective colorectal surgery. This systematic review and meta-analysis evaluated whether adding oral antibiotic bowel preparation (OAB) to mechanical bowel preparation (MBP) reduces SSIs compared with MBP alone. Materials and Methods: PubMed, the Cochrane Library, Scopus, and ClinicalTrials.gov were searched for English-language randomized controlled trials published from January 2005 to January 2025. Eligible trials enrolled adults undergoing elective colorectal surgery and compared MBP+OAB versus MBP alone, with standard intravenous prophylaxis in both groups. The primary outcome was overall SSI; secondary outcomes were incisional SSI and organ-space SSI. Risk of bias was assessed with RoB 2, certainty with GRADE, and odds ratios (ORs) were pooled using DerSimonian-Laird random-effects models. The protocol was prespecified but not prospectively registered. Results: Twelve trials including 4073 patients were included (MBP+OAB, n = 2069; MBP, n = 2004). MBP+OAB reduced overall SSI (OR 0.53, 95% CI 0.37-0.75; p < 0.001; I2 = 62.5%; 95% prediction interval 0.17-1.66), incisional SSI (OR 0.52, 95% CI 0.34-0.80; p = 0.003; I2 = 57.5%), and organ-space SSI (OR 0.63, 95% CI 0.45-0.88; p = 0.007; I2 = 8.3%). The effect was preserved in metronidazole-containing regimens (OR 0.46, 95% CI 0.33-0.65), but this subgroup was exploratory. Excluding high-risk-of-bias studies supported the primary result. Publication-bias assessment was underpowered. Overall and organ-space SSI were moderate-certainty outcomes; incisional SSI was low-certainty, and anastomotic leak was very low-certainty. Conclusions: In contemporary elective colorectal surgery when MBP is used, adding preoperative OAB probably reduces SSIs. Findings do not establish whether OAB alone is sufficient or whether MBP is necessary; stewardship-relevant outcomes remain insufficiently reported. Funding was provided by ISCIII grant PI25/01285.