BENZOYL PEROXIDE REDUCES INTRAOPERATIVE RE-EMERGENCE OF CUTIBACTERIUM ACNES COMPARED WITH HYDROGEN PEROXIDE IN SHOULDER ARTHROSCOPY: A PROSPECTIVE FACTORIAL STUDY

Aim Cutibacterium acnes is a common cause of surgical site infection in shoulder surgery. Although chlorhexidine gluconate (CHG) and hydrogen peroxide (H 2 O 2 ) reduce C. acnes, their ability to maintain this effect during surgery remains unclear. Benzoyl peroxide (BPO) may provide a more sustained antimicrobial effect; however, no studies have directly compared BPO and H 2 O 2 , particularly across different intraoperative time points. The purpose of this study was to compare the effects of BPO and H 2 O 2 on intraoperative C. acnes contamination. Methods In this prospective factorial study, 200 patients undergoing arthroscopic shoulder surgery between January and June 2026 were planned, and 120 patients were included in this interim analysis. Patients were allocated to four groups based on the use of benzoyl peroxide (BPO) and/or hydrogen peroxide (H 2 O 2 ). All patients received standard skin preparation with 1% CHG in alcohol. A 5% BPO was applied five times over 48 hours preoperatively, and 3% H 2 O 2 was applied immediately before surgery. Skin swabs were obtained at three time points: before skin preparation, after preparation, and at wound closure. The presence of Cutibacterium acnes and other bacteria was assessed. Analyses included comparisons across the four groups, factorial analyses based on BPO and H 2 O 2 use, and multivariable logistic regression to identify factors associated with C. acnes positivity at wound closure. In addition, patient-reported questionnaires were used to assess adherence to the BPO protocol and adverse events. Results At baseline, C. acnes positivity was lower in the BPO groups than in the non-BPO groups (12% vs. 73%; p < 0.001). Although skin preparation reduced positivity in all groups, C. acnes re-emerged at wound closure in the control and H 2 O 2 groups, whereas it remained low in the BPO groups. In factorial analyses, BPO use was associated with a significantly lower positivity rate at wound closure (OR, 0.15; p = 0.001), whereas H 2 O 2 showed no significant effect. Full compliance with BPO was 80% and was associated with lower C. acnes positivity. Multivariable analysis confirmed that BPO use was independently associated with reduced C. acnes positivity (OR 0.13; p = 0.002). No skin complications or symptomatic surgical site infections were observed. Conclusions Standard skin preparation with chlorhexidine does not provide sustained suppression of C. acnes throughout surgery, and hydrogen peroxide also failed to maintain suppression until wound closure. In contrast, the addition of benzoyl peroxide helps prevent intraoperative re-emergence and could be a more effective adjunct than hydrogen peroxide.