• BACKGROUND
    • Although prophylactic antibiotics are considered the standard of care, data with regard to the comparative efficacy of specific antibiotics in the prevention of periprosthetic joint infection (PJI) have remained limited. This study evaluated whether perioperative antibiotic choice affects rates of PJI development in shoulder arthroplasty.
  • METHODS
    • From 2000 to 2019, all primary shoulder arthroplasty types (hemiarthroplasty, anatomic total shoulder arthroplasty, reverse shoulder arthroplasty) performed for elective and trauma indications with perioperative antibiotic data and a minimum follow-up of 2 years were identified from a single institution. Demographic characteristics, PJI risk factors, and PJI-free survivorship data were retrieved. Multivariable analyses were conducted to determine the association between the antibiotic administered and the development of PJI.
  • RESULTS
    • Of 7,713 shoulder arthroplasties, cefazolin was administered in 6,879 procedures (89.2%) and non-cefazolin antibiotics consisting of vancomycin (465 procedures [6.0%]), clindamycin (345 procedures [4.5%]), and alternative regimens (24 procedures [0.31%]) were administered in 834 procedures (10.8%). PJIs occurred in 101 shoulder arthroplasties (1.3%), with Cutibacterium acnes as the most common pathogen (44 procedures [43.6%]). PJI-free survivorship was greater in shoulder arthroplasties in which cefazolin was administered compared with those in which non-cefazolin antibiotics were administered, with 0.91% greater survival free of PJI at 1 month, 1.4% at 1 year, and 2.7% at 15 years (p < 0.001). Cefazolin administration, compared with non-cefazolin administration, was associated with a 69% reduction in all-cause PJI risk and a 78% reduction in C. acnes PJI risk (p < 0.001). A higher risk of PJI for both groups was observed with vancomycin; the hazard ratio [HR] was 2.32 (95% confidence interval [CI], 1.22 to 4.40; p = 0.010) for all-cause PJI and 2.94 (95% CI, 1.12 to 7.49; p = 0.028) for C. acnes PJI. A higher risk of PJI was also observed for both groups for clindamycin; the HR was 5.07 (95% CI, 2.83 to 9.05; p < 0.001) for all-cause PJI and 8.01 (95% CI, 3.63 to 17.42; p < 0.001) for C. acnes PJI.
  • CONCLUSIONS
    • In primary shoulder arthroplasty, cefazolin administration was associated with a significantly lower rate of PJI compared with non-cefazolin alternatives, including both vancomycin and clindamycin. These risk discrepancies were observed across all infectious pathogens and may be considered even greater when C. acnes was the infecting bacterium.
  • LEVEL OF EVIDENCE
    • Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.