Abstract
This study aims to develop and validate a nomogram for predicting therapeutic efficacy in elderly hip fracture patients with preoperative lower extremity deep vein thrombosis (DVT). This retrospective study analyzed data from 172 elderly hip fracture patients with preoperative DVT treated between March 2021 and October 2023. Patients were randomly divided into training (n = 120) and validation (n = 52) sets. Univariate and multivariate logistic regression identified independent predictors of DVT treatment efficacy. A nomogram was constructed and evaluated using the concordance index, calibration curves, and decision curve analysis. A total of 172 elderly hip fracture patients with preoperative lower extremity DVT were randomly allocated into training (n = 120) and validation (n = 52) sets. The rates of ineffective DVT treatment were 25.83% (31/120) and 28.84% (15/52), respectively. Multivariate analysis identified 6 independent predictors: age (odds ratio [OR] = 1.282, 95% confidence interval [CI]: 1.151-1.428), bedridden duration (OR = 2.231, 95% CI: 1.622-3.069), fasting blood glucose (OR = 1.898, 95% CI: 1.369-2.631), fibrinogen (OR = 2.255, 95% CI: 1.547-3.288), lactate dehydrogenase (OR = 1.040, 95% CI: 1.021-1.059), and D-dimer (OR = 4.217, 95% CI: 1.783-9.972).The nomogram demonstrated excellent discrimination, with concordance index values of 0.955 (training) and 0.954 (validation). Calibration was satisfactory (mean absolute errors: 0.076 and 0.061), and the Hosmer-Lemeshow test indicated good fit (both P > .05). Receiver operating characteristic analysis yielded area under the curves of 0.955 (95% CI: 0.865-1.000) and 0.954 (95% CI: 0.914-0.994), respectively. Sensitivity and specificity were 0.714/0.875 (training) and 0.770/1.000 (validation), confirming robust predictive performance and clinical utility. The nomogram accurately predicts therapeutic efficacy for preoperative DVT in elderly hip fracture patients, offering a valuable tool for clinical decision-making.