Abstract
Introduction
Antibiotic-impregnated bone cement (AIBC) has been used for decades to treat and prevent post-operative infections in joint arthroplasty. Local delivery of antibiotics may theoretically have a bactericidal effect, however evidence supporting this is controversial and literature suggests its prophylactic use in primary total knee arthroplasty (TKA) is seldom justified. With evolving standards of care, historical data is no longer relevant in addressing the efficacy of AIBC in the contemporary TKA. The purpose of this study was to evaluate outcomes following primary TKA using AIBC and regular non-AIBC by comparing rates of surgical site infection (SSI) and prosthetic joint infection (PJI).
Methods
A retrospective review was conducted of all cemented primary TKA procedures from a large institutional database between January 1, 2015 and December 31st, 2016. This identified 6,073 cases, n=2,613 in which AIBC was used and n=3,460 cases using bone cement without antibiotics. Patients were stratified into low risk and high-risk groups based on age (>65 years), BMI (>40), and Charlson Comorbidity Index (CCI; >3). Medical records were reviewed for diagnoses of SSI (skin and superficial wound infections) and PJI (deep joint infections requiring surgery) over a 2-year postoperative period. Univariate analysis and multivariate regression models were used to ascertain the effects of cement type, patient factors (age, gender, BMI, CCI), operative time, and length of stay on infection rates. Additionally, mixed models (adjusted for gender, age, race, BMI, and CCI) were built to account for surgeon variability.
Results
The use of AIBC and risk group distributions were equal across the study period and no collinearity was found between the study variables. The SSI rate was 3.0% and the PJI rate was 0.8% in the total study population. Univariate analysis showed there was no significant difference in SSI rates with AIBC compared to non-AIBC (3.3% vs. 2.8%, p=0.278) or in PJI rates (1.0% vs. 0.7%, p=0.203). Multivariate logistic regression analysis adjusted for patient factors, operative time, and length of stay showed no significant difference in SSI rates with a procedure using AIBC compared to non-AIBC (OR=0.90; 95% CI, 0.66–1.23; p=0.515) and no significant difference in PJI rates (OR=1.01; 95% CI, 0.55–1.84; p=0.984). Mixed models also showed no difference in PJI rates with AIBC use after adjusting for surgeon variability as well as patient factors (gender, race, age, BMI, and CCI).
Discussion
Prophylactic use of AIBC in primary TKA is not without consequence when considering the significant increase in cost and its potential side effects, namely organism specific antibiotic resistance and mechanical loosening. This study shows that even when adjusted for patient factors, procedure-related factors, and length of stay, there is no clinically significant decrease in infection rates with the use of AIBC in primary TKAs.
