One hundred and six patients scheduled for a primary, elective open shoulder surgery were randomized to prepare the shoulder before surgery by using 5% benzoyl peroxide topical solution (n=50) or soap to shower according to local guidelines (n=56). Outcomes of interest included the following: bacterial colonization in the skin and dermis, and the incidence of deep surgical site infection (SSI). All outcomes were measured pre-operatively at the following timepoints: in the pre-operative area (T1), immediately before skin incision (T2), after skin incision in a sterile field (T3), and immediately after skin incision (T4). Results revealed a significantly lower number of patients with detectable Cutibacterium acnes (C acnes) using skin swabs in the benzoyl peroxide group vs control at T1 and T2 (p<0.05). As well, the total viable count of C acnes was significantly lower in the benzoyl peroxide group vs control at all timepoints (p<0.05 for all). With respect to dermal skin swabs, no significant difference was observed between the two groups which was measured immediately after skin incision (p=0.25).