CEFAZOLIN VERSUS CEFAZOLIN PLUS VANCOMYCIN PROPHYLAXIS FOR PRIMARY TKA PATIENTS WITHOUT DOCUMENTED MRSA COLONIZATION: AN ASAP TRIAL-INFORMED PROPENSITY-MATCHED ANALYSIS

Background Routine addition of vancomycin to cefazolin for perioperative prophylaxis remains common in arthroplasty practice. In the ASAP randomized trial of 4,239 arthroplasty patients, cefazolin plus vancomycin failed to reduce surgical site infection compared with cefazolin alone among patients without known MRSA colonization. We evaluated whether this finding was reproducible among primary TKA patients in a large U.S. real-world cohort. Methods A retrospective cohort study was conducted using the TriNetX U.S. Collaborative. Adults ≥18 years undergoing primary TKA who received day-of-surgery cefazolin plus vancomycin were compared with those who received cefazolin alone. To approximate ASAP eligibility, we excluded patients with prior periprosthetic joint infection, any infection within 30 days pre-operatively, and any documented or suspected MRSA carriage, pre-operative mupirocin use, or pre-operative S. aureus nucleic acid testing. Cohorts were 1:1 propensity-score matched on age, sex, race, ethnicity, BMI, comorbidities, and antibiotic allergy status. Outcomes within 90 days of surgery included periprosthetic joint infection, surgical site infection (any, superficial, deep), sepsis, cellulitis, acute kidney injury, hypersensitivity reactions, all-cause readmission, and emergency department visits. Results After matching, 120,878 patients (60,439 per arm) were analyzed, with all standardized mean differences below 0.04. The cefazolin plus vancomycin group had a higher coded incidence of 90-day periprosthetic joint infection (0.67% vs 0.45%; RR 1.48; p < 0.001), any surgical site infection (0.71% vs 0.58%; RR 1.23; p = 0.003), superficial SSI (RR 1.38; p = 0.001), acute kidney injury (RR 1.24; p < 0.001), cellulitis (RR 1.25; p < 0.001), and 90-day readmission (13.2% vs 11.9%; RR 1.11; p < 0.001). Deep SSI showed a directionally similar association without reaching statistical significance. Sepsis, hypersensitivity, and emergency department visits did not differ between groups. Conclusion In this study of more than 120,000 primary TKA patients approximating ASAP eligibility, the cefazolin plus vancomycin regimen was associated with higher coded rates of periprosthetic joint infection, surgical site infection, acute kidney injury, cellulitis, and 90-day readmission compared with cefazolin alone. Viewed alongside the ASAP trial, these findings do not provide supportive evidence for routine dual prophylaxis in patients without known MRSA colonization and instead reinforce the need for more selective, risk-based use of vancomycin. These findings should be interpreted as hypothesis-generating given the limitations of federated EHR data, including inability to verify antibiotic dosing/timing or intra-articular vancomycin use, reliance on administrative codes, potential residual confounding, and site-level variation in data capture.