Risk factors for preoperative deep vein thrombosis in patients with fractures around the knee and construction of a predictive model

TANG Zhuo-dong; LI Ting-yan; WANG Ming-you; WANG Chuan; LIU Shao-jiang

doi:10.13820/j.cnki.gdyx.20241241

Abstract：Objective To explore the risk factors for preoperative deep vein thrombosis(DVT)in patients with fractures around the knee(FAK)and to construct a predictive risk model.Methods A retrospective analysis of the clinical data of 239 FAK patients admitted to our hospital from January 2020 to December 2023.According to preoperative lower limb Doppler ultrasound results,patients were divided into two groups:the DVT group(n=64)and the non-DVT group(n=175).Univariate and multivariate logistic regression analyses were performed to identify the independent risk factors for preoperative DVT.Based on these independent factors,a nomogram prediction model was constructed.The model's predictive performance for preoperative DVT was evaluated by the receiver operating characteristic(ROC)curve,and its consistency and effectiveness were validated using calibration and decision curve analysis(DCA)curves.Results Among the 239 patients included in the study,64 had preoperative DVT,with an incidence rate of 26.77％.Univariate analysis showed significant differences between the two groups in terms of age,injury severity,fracture site,preoperative immobilization time,preoperative hemoglobin(Hb),preoperative hematocrit(HCT),and preoperative D-dimer levels(P＜0.05).Multivariate logistic regression analysis identified age,injury severity,preoperative immobilization time,and preoperative D-dimer levels as independent risk factors for preoperative DVT in FAK patients.The ROC curve indicated that the nomogram model based on these independent risk factors had good predictive value,with an area under the curve(AUC)of 0.863(95％CI:0.805-0.920),sensitivity of 64.0％,and specificity of 97.1％.The calibration curve showed good consistency of the model.The DCA curve indicated high clinical net benefit across threshold probabilities ranging from 0.1 to 0.8.Conclusion Age,injury severity,preoperative immobilization time,and preoperative D-di-mer levels are closely related to the occurrence of preoperative DVT in FAK patients.The nomogram prediction model based on these factors demonstrates good predictive performance and is worthy of clinical application.