Data-Driven Estimated Glomerular Filtration Rate Strata Predict 90-Day Major Complications Following Total Knee Arthroplasty in Patients With Chronic Kidney Disease

Authors

Daniel A. Raftis, From the Department of Orthopaedic Surgery, The George Washington University School of Medicine and Health Sciences, Washington, DC (Raftis, Zhao and Gu), the Department of Orthopaedic Surgery, Johns Hopkins University School of Medicine, Baltimore, MD (Dr. Agarwal, Dr. Harris, Dr. Kurian, and Thakkar), and the Department of Orthopaedic Surgery, Virginia Commonwealth University School of Medicine, Richmond, VA (Golladay).
Amy Y. Zhao
Amil R. Agarwal
Alex Gu
Andrew B. Harris
Shyam Kurian
Savyasachi C. Thakkar
Gregory J. Golladay

Document Type

Journal Article

Publication Date

1-29-2025

Journal

The Journal of the American Academy of Orthopaedic Surgeons

DOI

10.5435/JAAOS-D-24-01029

Abstract

INTRODUCTION: Lower estimated glomerular filtration rate (eGFR) in patients who have chronic kidney disease (CKD) is associated with increased risk of complications following total knee arthroplasty (TKA). However, there is a lack of literature that identifies eGFR levels those are associated with notable differences in risk of these complications. The purpose of this study was to create eGFR strata for CKD patients that are associated with varying risks of 90-day major complications following TKA. METHODS: Nondialysis patients who have CKD and known eGFR levels one month before primary TKA were identified using a national database. Stratum-specific likelihood ratio (SSLR) analysis was used to construct data-driven eGFR strata associated with varying risks of 90-day major complications. The incidence rates were recorded for each stratum. Each stratum was propensity score matched to the highest eGFR stratum. The risk ratio with a corresponding 95% confidence interval for 90-day major complications was recorded for each stratum. RESULTS: A total of 24,895 patients with CKD who underwent TKA were included in this study. SSLR identified four data-driven eGFR strata: 15-31, 32-44, 45-54, and 55+. The unmatched 90-day major complication incidence increased sequentially as the eGFR strata decreased: 55+ (10.72%), 45-54 (13.87%), 32-44 (17.30%), and 15-31 (25.16%). When compared with the matched highest eGFR strata (55+), the risk of sustaining a 90-day major complication increased sequentially as the eGFR strata decreased (RR: 1.27, 1.56, 2.06, P < 0.001). The risk of death within 90 days was higher in the 15-31 stratum (RR: 3.08, P < 0.001) when compared with the matched 55+ stratum. CONCLUSION: Using SSLR analysis, four data-driven strata were identified with varying risks of 90-day major complications following TKA. These eGFR thresholds were created specifically for TKA and can therefore be appropriately used to risk-stratify CKD patients in the preoperative setting when discussing TKA.

APA Citation

Raftis, Daniel A.; Zhao, Amy Y.; Agarwal, Amil R.; Gu, Alex; Harris, Andrew B.; Kurian, Shyam; Thakkar, Savyasachi C.; and Golladay, Gregory J., "Data-Driven Estimated Glomerular Filtration Rate Strata Predict 90-Day Major Complications Following Total Knee Arthroplasty in Patients With Chronic Kidney Disease" (2025). GW Authored Works. Paper 6305.
https://hsrc.himmelfarb.gwu.edu/gwhpubs/6305

Department

School of Medicine and Health Sciences Resident Works