Search

Objective: To construct a new thrombus risk assessment model and evaluate its predictive ability for venous thromboembolism (VTE) in the patients with malignant tumors, and to provide the basis for the early predition of the malignant tumor patients with high risk for VTE. Methods: A total of 128 untreated malignant tumor patients were included, of which 40 were diagnosed with VTE within 2 months of malignant tumor diagnosis and categorized as VTE group. A total of 88 patients who did not develop VTE were categorized as non-VTE group. The clinical risk factors and laboratory indicators of the patients in two groups were compared and analyzed; the types of thrombotic events of the patients were analyzed; the diagnostic values of thrombin-antithrombin-complex (TAT), α2-plasmin inhibitor-plasmin complex (PIC), D-dimer(D-dimer), and fibrin degradation products(FDP) in malignant tumors complicated by VTE were assessed using receiver operating characteristic (ROC) curve analysis; Multivariate Logistic regression analysis was used to analyze the correlations of the clinical risk factors and biomarkers with the malignant tumors complicated with VTE. A new thrombus risk assessment model was constructed, consisting of TAT≥0. 70 μg·L-1, poor differentiation, and cardiovascular risk factors. The predictive probability of the model for malignant tumors complicated by VTE was evaluated based on the significance, goodness of fit, calibration curve, and C value of the model. The clinical application value of the new thrombus risk assessment model, COMPASS-CAT risk score (CRS), and Khorana risk score (KRS) in assessing malignant tumor patients complicated by VTE was compared using the C value and decision curve analysis (DCA). Results: The plasma levels of TAT (P<0. 001), PIC (P<0. 001), D-dimer (P<0. 05), and FDP (P<0. 01) of the patients in VTE group were higher than those in nonVTE group. Compared with the patients without cardiovascular risk factors, poor differentiation, and lymphatic metastasis, the malignant tumor patients with cardiovascular risk factors (P<0. 001), poor differentiation (P<0. 001), and lymphatic metastasis (P<0. 05) were more likely to develop VTE. Most VTE events (65%) were isolated deep vein thromboembolism(DVT). The ROC curve analysis showed that the area under the curve (AUC), sensitivity, and specificity of TAT and PIC were higher than those of D-dimer and FDP. TAT≥0. 70 μg·L-1 (P<0. 05), poor differentiation (P<0. 01), and cardiovascular risk factors(P<0. 01) were the independent risk factors for VTE in the malignant tumor patients. A new thrombus risk assessment model consisting of TAT≥0. 70 μg·L-1, poor differentiation, and cardiovascular risk factors was constructed. The new risk assessment model had a high goodness of fit (P=0. 805) and good predictive ability during internal validation (χ² =75. 266, P<0. 001). The ROC curve analysis results showed that the C values for the new thrombus risk prediction model, CRS, and KRS were 0. 908, 0. 676, and 0. 541, respectively. The DCA curve analysis results showed that the new thrombus risk assessment model had a higher net benefit rate compared with CRS and KRS. Conclusion: TAT and PIC have greater diagnostic efficiency than D-dimer in the early prediction of the malignant tumor patients with high-risk VTE. For the patients included in this study, the new thrombus risk assessment model, constructed from TAT≥0. 70 μg·L-1, poor differentiation, and cardiovascular risk factors, has superior diagnostic efficiency and clinical predictive value compared with CRS and KRS. [ABSTRACT FROM AUTHOR]