Cefazolin Alone Versus Cefazolin With Tobramycin or Gentamicin as Intraoperative Antibiotic Prophylaxis for Total Joint Arthroplasty

Abstract

Background

Prosthetic joint infection is a serious complication of total knee arthroplasty (TKA) and total hip arthroplasty (THA). While cefazolin is the standard prophylactic antibiotic, some groups have considered adding aminoglycosides (gentamycin or tobramycin) to this regimen.

Methods

Adult primary THA and TKA patients were identified from 2010-2021 Q1 PearlDiver M151 database. Inclusion criteria were activity in the database ≥ 90 days postoperative and no infectious, neoplastic, or traumatic diagnoses within 90 days preoperative. Intraoperative antibiotic prophylaxis regimens were determined using Current Procedural Terminology J codes. For both THA and TKA, 2 subcohorts were created: (1) Patients who received cefazolin alone and (2) cefazolin and tobramycin or gentamicin. Differences in 90-day postoperative adverse outcomes were assessed with multivariable logistic regression controlling for age, sex, and Elixhauser Comorbidity Index. Five-year implant survival was assessed using Kaplan-Meier analysis and log-rank tests.

Results

For THA, 32,882 patients were identified, prophylactic cefazolin alone was given to 30,527 (92.8%), and cefazolin + gentamicin/tobramycin was given to 2355 (7.2%). For TKA, 119,611 patients were identified, prophylactic cefazolin alone was given to 110,469 (92.4%), and cefazolin + gentamicin/tobramycin was given to 9142 (7.6%). Overall SSI rate for THA was 1.1% and for TKA 0.8%. For both THA and TKA, antibiotic subgroups were clinically similar with regard to age, sex, and Elixhauser Comorbidity Index. On univariable and multivariable analysis, no 90-day outcomes varied significantly. Five-year implant survivals were not significantly different.

Conclusions

For THA and TKA, cefazolin alone vs cefazolin + gentamicin/tobramycin were not found to have differences in rates of perioperative adverse outcomes (including SSI/prosthetic joint infection) or 5-year revision rates.