Development of an Integrated Nomogram for Predicting Postoperative Deep Vein Thrombosis Risk in Trauma Patients: Combining Thrombosis Risk Assessment Profile Score and Thrombosis Biomarkers

AIM: This study aims to evaluate the effectiveness of combining the risk assessment profile for thromboembolism (RAPT) score with thrombotic biomarkers in predicting postoperative deep vein thrombosis (DVT) in patients with traumatic fractures and to create a nomogram model for risk assessment. METHODS: This retrospective cohort study recruited 329 traumatic fracture patients from Shouxiang Community Health Service Center of Yinhu Street between September 2021 and September 2024. Patient data were randomly assigned to a training set (n = 230, 70%) and a test set (n = 99, 30%) for model development and validation. In the training set, patients were stratified based on DVT state into a DVT group (n = 110) and a non-DVT group (n = 120). The RAPT score and thrombotic biomarker levels were compared between the two groups. Multivariate logistic regression analysis was conducted to identify independent risk factors for postoperative DVT. Based on these factors, a nomogram model was developed, and its diagnostic performance was assessed through receiver operating characteristic (ROC) curve analysis, calibration curve analysis, and clinical decision curve analysis. RESULTS: The DVT group exhibited significantly higher levels of RAPT score (7.00 [5.00, 9.00] vs. 4.00 [2.00, 7.00]), D-dimer (D-D) (874.12 ± 77.16 vs. 841.37 ± 86.94), fibrinogen (FIB; 4.00 [3.90, 4.30] vs. 4.00 [3.70, 4.20]), and thrombin-antithrombin complex (TAT; 16.60 [14.43, 18.38] vs. 15.40 [14.10, 16.90]) relative to non-DVT group (p < 0.05). Multivariate logistic regression analysis identified the RAPT score, D-D, FIB, and TAT as independent risk factors for postoperative DVT, with odds ratios (ORs) of 1.209, 1.006, 3.625, and 1.246, respectively (p < 0.05). Using these factors, a nomogram model was constructed. In both the training and test sets, the fitting degree of this nomogram model was good. ROC curve analysis revealed that the area under the curve (AUC) of 0.7714 (0.7107–0.832) and 0.7066 (0.603–0.8103) for predicting the occurrence of lower extremity DVT in the training set and the test set, respectively. The calibration curve demonstrated excellent agreement between the predicted probabilities and the observed outcomes. Decision curve analysis (DCA) demonstrated that the nomogram yielded a higher net benefit than the “treat all” or “treat none” strategies across a threshold probability range of 0.055–0.755 in the training set and 0.095–0.805 in the testing set. CONCLUSIONS: The integration of the RAPT score with thrombotic biomarkers (D-D, FIB, and TAT) offers a feasible and effective approach for predicting postoperative DVT in patients with traumatic fractures, guiding targeted prophylactic strategies and enhancing perioperative management and patient outcomes.