Construction and validation of a predictive model for lower extremity deep vein thrombosis after total knee arthroplasty

Abstract

The aim was to investigate the independent risk factors for lower extremity deep vein thrombosis (DVT) after total knee arthroplasty, and to establish a nomogram prediction model accordingly. Data were collected from total knee replacement patients from January 2022 to December 2023 in our hospital. Unifactorial and multifactorial logistic regression analyses were used to determine the independent risk factors for lower extremity DVT after total knee arthroplasty and to establish the corresponding nomogram. The receiver operating characteristic curves were plotted and the area under the curve was calculated, and the calibration curves and decision curves were plotted to evaluate the model performance. A total of 652 patients with total knee arthroplasty were included in the study, and 142 patients after total knee arthroplasty developed deep veins in the lower extremities, with an incidence rate of 21.78%. After univariate and multivariate logistic regression analyses, a total of 5 variables were identified as independent risk factors for lower extremity DVT after total knee arthroplasty: age > 60 years (OR: 1.70; 95% CI: 1.23–3.91), obesity (OR: 1.51; 95% CI: 1.10–1.96), diabetes mellitus (OR: 1.80; 95% CI: 1.23–2.46), D-dimer > 0.5 mg/L (OR: 1.47; 95% CI: 1.07–1.78), and prolonged postoperative bed rest (OR: 1.64; 95% CI: 1.15–3.44). the nomogram constructed in this study for lower extremity DVT after total knee arthroplasty has good predictive accuracy, which helps physicians to intervene in advance in patients at high risk of lower extremity DVT after total knee arthroplasty.