Does hydrogen peroxide application to the dermis following surgical incision affect Cutibacterium acnes cultures in total shoulder arthroplasty in male patients? A randomized controlled trial

Background: Periprosthetic joint infections occur in 1%-4% of primary total shoulder arthroplasties (TSAs). Cutibacterium acnes is the most commonly implicated organism and has been shown to persist in the dermis despite use of preoperative antibiotics and standard skin preparations. Studies have shown decreased rates of cultures positive for C acnes with use of preoperative benzoyl peroxide or hydrogen peroxide (H2O2), but even with this positive deep cultures remain common. We sought to determine whether an additional application of H2O2 directly to the dermis following skin incision would further decrease deep culture positivity rates. Methods: We performed a randomized controlled trial comparing tissue culture results in primary TSA in patients who received a standard skin preparation with H2O2, ethanol, and ChloraPrep (CareFusion, Leawood, KS, USA) vs. an additional application of H2O2 to the dermis immediately after skin incision. Given the sexual dimorphism seen in the shoulder microbiome regarding C acnes colonization rates, only male patients were included. Bivariable and multivariable analyses were performed to compare rates of positive cultures based on demographic and surgical factors. Results: Dermal cultures were found to be positive for C acnes at similar rates between the experimental and control cohorts for the initial (22% vs. 28%, P = .600) and final (61% vs. 50%, P > .999) dermal swabs. On bivariable analysis, the rate of positive deep cultures for C acnes was lower in the experimental group, but this difference was not statistically significant (28% vs. 44%, P = .130). However, patients who underwent anatomic TSA were found to have a significantly greater rate of deep cultures positive for C acnes (57% vs. 28%, P = .048); when controlling for this on multivariable analysis, the experimental cohort was found to be associated with significantly lower odds of having positive deep cultures (odds ratio, 0.37 [95% confidence interval, 0.16-0.90 ], P 1/4 .023). There were no wound complications in either cohort. Conclusions: An additional H2O2 application directly to the dermis following skin incision resulted in a small but statistically significant decrease in the odds of having deep cultures positive for C acnes without any obvious adverse effects on wound healing. Given its cost-effectiveness, use of a post-incisional dermal decontamination protocol may be considered as an adjuvant to preoperative use of benzoyl peroxide or H2O2 to decrease C acnes contamination.