Construction of a predictive model of deep vein thrombosis in elderly patients with hip fractures

LUO Jiehua; LIU Qipei; GONG Chunzhu

doi:10.12173/j.issn.1004-5511.202504005

Abstract：Objective To identify risk factors for preoperative lower limb deep vein thrombosis(DVT)in elderly patients with hip fractures and to develop a predictive nomogram model.Methods Elderly patients with hip fractures admitted to the Third Affiliated Hospital of Guangzhou University of Chinese Medicine from June 2021 to June 2024 was enrolled.Clinical data were collected and analyzed.Independent influencing factors for preoperative DVT were screened using Lasso regression and multivariate Logistic regression.A nomogram model was developed and validated internally and externally.Model performance was evaluated using Bootstrap resampling,calibration curves,receiver operating characteristic(ROC)curves with area under the curve(AUC),and decision curve analysis(DCA)to assess discrimination,calibration,and clinical utility.Results A total of 644 patients were included,with 451 in the training set and 193 in the validation set.Multivariate Logistic regression revealed that elevated D-dimer levels[OR=1.927,95％CI(1.675,2.257)],prolonged time from injury to admission[OR=1.023,95％CI(1.012,1.034)],increased fibrinogen[OR=1.348,95％CI(1.076,1.703)],higher total cholesterol[OR=1.256,95％CI(1.053,1.504)],and hypoalbuminemia[OR=0.936,95％CI(0.909,0.965)]were independent predictors of preoperative DVT.The AUCs were 0.871[95％CI(0.816,0.925)]for the training set and 0.879[95％CI(0.805,0.953)]for the validation set.Calibration curves showed that the predicted probabilities were consistent with the actuality.DCA demonstrated that the model yielded a higher net benefit than traditional strategies across a risk threshold range of 10％to 70％.Conclusion The nomogram-based predictive model developed in this study enables intuitive and accurate preoperative risk stratification for DVT in elderly patients with hip fractures.It serves as a robust scientific tool to guide early clinical screening and targeted interventions for thromboprophylaxis.