Periprosthetic Joint Infection Is Associated with an Increased Risk of Venous Thromboembolism Following Revision Total Knee Replacement: An Analysis of Administrative Discharge Data

Background: Little is known about the relationship between infection and the risk of risk of venous thromboembolism (VTE) following orthopaedic surgery. We assessed the 90-day risk of VTE following revision total knee replacement to measure the association between periprosthetic joint infection and the risk of postoperative VTE.

Methods: We used New York Statewide Planning and Research Cooperative System data to identify all New York State residents undergoing revision total knee replacement from 1998 to 2014. ICD-9-CM (International Classification of Diseases, Ninth Revision, Clinical Modification) codes were used to identify comorbidities and to classify the indication for revision total knee replacement as aseptic, infection, or fracture. The primary outcome was any diagnosis code for VTE recorded for the revision surgery and/or subsequent admissions within 90 days. A multivariable logistic regression model that included demographic characteristics and comorbidities was used to estimate the risk of VTE after revision for infection or fracture, with aseptic revision as the reference group.

Results: The present study included 25,441 patients who were managed with revision total knee replacement; the indication for revision was aseptic for 17,563 patients (69%), infection for 7,075 (28%), and fracture for 803 (3%). The mean age (and standard deviation) was 66 ± 12 years, 15,592 (61%) of the patients were female, 3,198 (13%) were Black, 1,192 (5%) were smokers, and 4,222 (17%) were obese. Seven hundred and nineteen patients (2.8%) had VTE within the 90 days after revision total knee replacement, including 387 (1.5%) during the admission for the revision procedure. The 90-day incidence of VTE was 2.1% after aseptic revision, 4.3% after revision for infection, and 5.9% after revision for fracture. The adjusted odds ratio (aOR) for VTE relative to aseptic revision was 2.01 (95% confidence interval [CI], 1.72 to 2.35) for septic revision total knee replacement and 2.62 (95% CI, 1.91 to 3.6) for fracture. A history of VTE was also a strong risk factor for VTE following revision total knee replacement (aOR, 2.01; 95% CI, 1.48 to 2.71).

Conclusions: We found that the odds of VTE after revision total knee replacement for infection were double those after aseptic revision total knee replacement. Although fracture accounts for a small percentage of revision total knee replacements, the risk of VTE was 2.6-fold higher after these procedures. The indication for revision total knee replacement should be considered when choosing postoperative VTE prophylaxis.

Level of evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.