Locally applied vancomycin is increasingly being used in primary hip and knee arthroplasty to reduce the risk of infection. Despite encouraging initial results, considerable debate remains on the basis of the data currently available. In particular, it has been unclear up to now whether local vancomycin is suitable to further reduce the risk of infection even if the rate of infection is already low (< 1%). In this monocentric retrospective cohort study, all primary total hip and knee arthroplasties performed between 2013 and 2018 were included. After a change in procedure at the hospital, 1 g vancomycin powder was applied intraarticularly before wound closure. The remaining perioperative procedure was constant over the investigation period. The follow-up was one year. The presence of an infection according to the currently valid MSIS criteria was defined as the endpoint. In patients with TKA two infections (0.3%) were observed under vancomycin prophylaxis in contrast to 44 infections (1.3%) in the control group (p = 0.033). In patients with THA two infections (0.5%) were observed under vancomycin prophylaxis and 48 infections (1.1%) in the control group without local vancomycin but this difference was statistically not significant. No wound complications requiring revision were observed as a result of the vancomycin. On the basis of the results of this study, intraarticular application of vancomycin powder in total hip and knee arthroplasty may be considered. Prospective randomized studies have to confirm this promising results prior a common recommendation. Level of Evidence III Retrospective cohort study. Infections after endoprosthetic hip or knee replacement, in spite of all perioperative improvements, are observed with an incidence of 1-2% 1-4. Despite the apparently low risk, efforts to further reduce the risk are sensible in light of the consequences for the patients concerned, with the potential for several follow-up operations and the often limited functional outcome. In spinal surgery, the reduction in the rate of infection through locally applied vancomycin was established more than 10 years ago and has been well documented by meta-analyses and animal experiments 5-8. The positive effect is set against minor complications caused in 0.3% of cases, mostly culture-negative seroma formation 9, so that it is even used successfully in children 10. Otte et al. were able to show that intrawound vancomycin powder lowers the infection rate after knee and hip arthroplasty from 1.6 to 0.5%, compared with a control group 11. In a subgroup analysis, the effect in revision operations was even greater, with a reduction of the infection rate from 3.9 to 0%. Patel et al. showed a reduction in the infection rate through vancomycin of 2.7-0.3% and determined a number needed to treat (NNT) of 47.5 12. Systemic side effects were not observed. The local application of vancomycin was assessed to be safe and cost-effective. Dial et al. observed a similar effect in primary hip arthroplasty, with a lowering of the infection rate...