Aspirin Is Associated With Improved Thromboprophylaxis Versus Low-Molecular-Weight Heparin Following Revision Knee Arthroplasty for Periprosthetic Joint Infection.

BACKGROUND

There is no consensus regarding the optimal venous thromboembolism (VTE) chemoprophylaxis agent following surgery for periprosthetic joint infection (PJI). Aspirin (ASA) and low-molecular-weight heparin (LMWH) are two commonly used agents; however, efficacy data in this population is lacking. This study compared ASA to LMWH among patients undergoing first-stage revision total knee arthroplasty for PJI.

METHODS

A large national database was queried from 2016 to 2023. Patients undergoing antibiotic spacer placement for a knee PJI were identified and divided into those who received either ASA or LMWH on postoperative day one. Patients taking other prophylactic agents or those who had a VTE history were excluded. Propensity score matching was used to balance cohorts. There were 10,472 patients 1:1 matched (5,236:5,236) based on VTE prophylaxis type and were well balanced for covariates (standardized mean difference < 0.10). Both cohorts had an average age of 66 years and an average Charlson comorbidity index of 3.2. Tranexamic acid was used in 61.2% of the ASA cohort and 60.3% of the LMWH cohort. To account for residual confounding, multivariable logistic regression models were used to assess our outcomes of VTE and postoperative transfusion.

RESULTS

The ASA patients had lower rates of VTE (1.0 versus 1.6%, P = 0.002; adjusted odds ratio = 0.6, 95% confidence interval = 0.4 to 0.9). Postoperative transfusion rates were significantly lower in the ASA cohort (6.0 versus 8.2%, P < 0.001; adjusted odds ratio = 0.8, 95% confidence interval = 0.6 to 0.9).

DISCUSSION: The use of ASA was associated with lower rates of VTE and transfusion compared to LMWH among patients who did not have a VTE history undergoing revision total knee arthroplasty for PJI. The use of ASA appears to be a safe and effective option for VTE chemoprophylaxis in this patient population.