β-Lactam vs Non–β-Lactam Antimicrobial Prophylaxis and Surgical Site Infection

Importance: β-lactam-based surgical antimicrobial prophylaxis (SAP) is the standard for most surgical procedures. Alternatively, and mostly due to allergies, non-β-lactam-based prophylaxis is used. The association between the use of non-β-lactam SAP and an increased risk of surgical site infection (SSI) rate has not yet been conclusively described. Objective: To assess whether non-β-lactam vs β-lactam prophylaxis is associated with the occurrence of SSI. Design, Setting, and Participants: This cohort study was based on the Swissnoso SSI surveillance system of 175 hospitals, including adult patients who underwent a major surgical procedure with SAP administration within 120 minutes prior to incision and postdischarge follow-up from January 2009 to December 2020. Patients with wound contamination class IV were excluded. Data analysis was conducted from July to December 2024. Exposures: β-Lactam SAP vs non-β-lactam SAP. Main Outcomes and Measures: The main outcome was occurrence of SSI, according to Centers for Disease Control and Prevention definitions. Mixed-effects logistic regression models were used to adjust for institutional, patient, and perioperative confounding factors. Results: Of 538 976 surveilled patients, 348 885 (196 411 [56.3%] female; median [IQR] age, 63.2 [47.0-73.3] years) fulfilled eligibility criteria. β-Lactam SAP was administered in 342 936 patients (98.3%) and non-β-lactam SAP in 5949 patients (1.7%). SSI was diagnosed in 9507 patients (2.8%) exposed to β-lactam SAP vs 364 patients (6.1%) who received non-β-lactam SAP (P < .001). Non-β-lactam SAP was significantly associated with a higher SSI rate (adjusted odds ratio [aOR], 1.78; 95% CI, 1.59-1.99; P < .001) compared with β-lactam SAP. A higher SSI rate for non-β-lactam SAP was found across all procedure types. Secondary analyses found a higher risk of SSI for ciprofloxacin (aOR, 1.57; 95% CI, 1.33-1.87; P < .001), vancomycin (aOR, 1.38; 95% CI, 1.03-1.86; P = .04), and clindamycin (aOR, 2.12; 95% CI, 1.82-2.47; P < .001) compared with β-lactam SAP. Conclusions and Relevance: In this large cohort study, administration of non-β-lactam SAP was associated with 1.8-fold higher odds of SSI. These findings suggest that non-β-lactam SAP should be avoided whenever possible. A careful evaluation of patients with reported β-lactam allergy should be performed before administering a second-choice antibiotic.