When intravenous vancomycin prophylaxis is needed in shoulder arthroplasty, incomplete administration is associated with increased infectious complications.

Background: Vancomycin is often used as antimicrobial prophylaxis for shoulder arthroplasty (SA) either when first generation cephalosporins are contraindicated or colonization with resistant bacteria is anticipated. In general, vancomycin necessitates longer infusion times to mitigate potential side effects. When infusion is started too close to the time of the incision, administration may not be complete during surgery. This study evaluated whether incomplete administration of intravenous vancomycin prior to SA affects the rate of infectious complications.

Methods: Between 2000 and 2019, all primary SA types (hemiarthroplasty, anatomic total SA, reverse SA) performed at a single institution for elective and trauma indications using intravenous vancomycin as the primary antibiotic prophylaxis and a minimum follow-up of 2 yr were identified. The time between the initiation of vancomycin and skin incision was calculated. Complete administration was defined as at least 30 min of infusion prior to incision. Demographic characteristics and infectious complications including survival free of prosthetic joint infection (PJI) were generated. Multivariable analyses were conducted to evaluate the association between vancomycin timing and the development of PJI.

Results: A total of 461 primary SAs were included. Infusion was incomplete (< 30 minutes preoperatively) for 163 [35.4%] SA and complete (> 30 minutes preoperatively) for 298 [64.6%] SAs. The incomplete group demonstrated higher rates of any infectious complication (8% vs. 2.3%; P = .005), PJI (5.5% vs. 1%; P = .004), and reoperation inclusive of revision due to infectious complications (4.9% vs. 1%; P = .009). Survivorship free of PJI was worse in SA with incomplete compared to those with complete vancomycin administration. Survival rates for incomplete and complete administration were 97.6% and 99.3% at 1 mo, 95.7% and 99.0% at 2 yr, 95.1% and 99.0% at 5 yr, and 93.9% and 99.0% at 20 yr, respectively (P = .006). Multivariable analyses confirmed that incomplete vancomycin administration was an independent risk factor for PJI compared with complete administration (hazard ratio, 4.22 [95% confidence interval, 1.12-15.90]; P = .033), even when other independent predictors of PJI (age, male sex, prior surgery, methicillin-resistant Staphylococcus aureus colonization, and follow-up) were considered.

Conclusions: When vancomycin is the primary prophylactic agent used at the time of primary SA, incomplete administration (infusion to incision time under 30 min) seems to adversely increase the rates of infectious complications and PJI. Prophylaxis protocols should ensure that complete vancomycin administration is achieved to minimize infection after SA.