A predictive model incorporating TAT, PIC, t-PAIC, and D-dimer demonstrated good discriminative ability for perioperative deep vein thrombosis in hip fracture patients (AUC 0.829).
Cohort (n=210)
No
Can a predictive model incorporating TAT, PIC, t-PAIC, and D-dimer accurately predict perioperative deep vein thrombosis in patients with hip fractures?
A predictive model using coagulation and fibrinolysis biomarkers (TAT, PIC, t-PAIC, and D-dimer) shows good discriminative ability for identifying perioperative DVT risk in hip fracture patients.
Effect estimate: AUC 0.829
Objective This study aimed to evaluate the clinical utility of a perioperative deep vein thrombosis (DVT) prediction model based on combined coagulation–fibrinolysis biomarkers in patients with hip fractures, with the goal of improving the identification of individuals at high risk for DVT. Methods A retrospective cohort study was conducted including 210 patients with hip fractures admitted to the Second Affiliated Hospital of Soochow University between September 2023 and January 2025. Based on perioperative Doppler ultrasound findings, patients were divided into a DVT group (n = 79) and a non-DVT group (n = 131). Perioperative coagulation and fibrinolysis biomarkers including thrombin–antithrombin complex (TAT), plasmin–α2 inhibitor complex (PIC), thrombomodulin (TM), and tissue-type plasminogen activator inhibitor complex (t-PAIC), along with relevant clinical data, were collected. Independent risk factors for DVT were identified using multivariable logistic regression analysis and incorporated into a predictive model. The predictive performance of the model was evaluated using receiver operating characteristic (ROC) curve analysis, and a nomogram was constructed for risk visualization and clinical application. Results Logistic regression analysis identified TAT (OR = 1.048), PIC (OR = 1.682), t-PAIC (OR = 1.089), and D-dimer (OR = 1.094) as independent risk factors for DVT ( P < 0.05). The combined biomarker-based model demonstrated good discriminative ability, with an area under the ROC curve (AUC) of 0.829. The prediction model demonstrated good calibration, and decision curve analysis indicated a favorable clinical net benefit. Conclusion A predictive model incorporating TAT, PIC, t-PAIC, and D-dimer provides effective risk stratification for perioperative DVT in patients with hip fractures. Dynamic monitoring of these biomarkers may further guide individualized anticoagulation strategies and support clinical decision-making.
Zhu et al. (2026) conducted a cohort in hip fractures (n=210). Predictive model incorporating TAT, PIC, t-PAIC, and D-dimer was evaluated on Deep vein thrombosis (DVT) prediction (AUC 0.829). A predictive model incorporating TAT, PIC, t-PAIC, and D-dimer demonstrated good discriminative ability for perioperative deep vein thrombosis in hip fracture patients (AUC 0.829).