Characteristics of Antibiotic Prophylaxis and Risk of Surgical Site Infections in Primary Total Hip and Knee Arthroplasty

Background: Despite numerous antibiotic prophylaxis options for total hip arthroplasty (THA) and total knee arthroplasty (TICS), an assessment of practice patterns and comparative effectiveness is lacking. We aimed to characterize antibiotic utilization patterns and associations with infection risk and hypothesized differences in infection risk based on regimen. Methods: A retrospective cohort study was performed using data from 436,724 THA and 862,918 TICS (Premier Healthcare Database; 2006-2016). Main exposures were antibiotic type and duration: day of surgery only (day 0) or through postoperative day 1 (day 1). The primary outcome was surgical site infection (SSI) <30 days postoperation. Mixed-effect models measured associations between prophylaxis regimen and SSI as odds ratios (ORs) with 95% confidence intervals (CIs). Results: SSI prevalence was 0.21% (n = 914) for THA and 0.22% (n = 1914) for TICS. Among THA procedures, the most commonly used antibiotics were cefazolin (74.1%), vancomycin (8.4%), "other" antibiotic combinations (7.1%), vancomycin + cefazolin (5.1%), and clindamycin (3.3%). Here, 51.8% received prophylaxis on day 0 only, whereas 48.2% received prophylaxis through day 1. Similar patterns existed for TICS. Relative to cefazolin, higher SSI odds were seen with vancomycin (OR = 1.36; CI 1.09-1.71) in THA and with vancomycin (OR = 1.29; CI = 1.10-1.52), vancomycin + cefazolin (OR = 1.35; CI = 1.12-1.64), clindamycin (OR = 1.38; CI = 1.11-1.71), and "other" antibiotic combinations (OR = 1.28; CI = 1.07-1.53) in TICS. Prophylaxis duration did not alter SSI odds. Results were corroborated in sensitivity analyses. Conclusion: Antibiotic prophylaxis regimens other than cefazolin were associated with increased SSI risk among THA/TICS patients. These findings emphasize a modifiable intervention to mitigate infection risk. (C) 2020 Elsevier Inc. All rights reserved.