A nomogram for predicting the risk of venous thromboembolism in patients with solid cancers

Abstract

AbstractCancer patients with venous thromboembolism (VTE) are prone to poor prognoses. Thus, we aimed to develop a nomogram to predict the risk of VTE in these patients. We retrospectively analyzed 791 patients diagnosed with solid tumors between January 2017 and May 2021 at Tongji Hospital. Univariate logistic analysis and multivariate logistic regression were adopted in this study. Our results indicated that age ≥ 60 years, tumor stages III-IV, platelet distribution width (PDW) ≤ 12.6%, albumin concentration ≤ 38.8 g/L, lactate dehydrogenase (LDH) concentration ≥ 198 U/L, D-dimer concentration ≥ 1.72 µg/mL, blood hemoglobin concentration ≤ 100 g/dL or the use of erythropoiesis-stimulating agents and cancer types were independent risk factors. The nomogram prediction model was developed based on the regression coefficients of these variables. We assessed the performance of the nomogram by calibration plot and the area under the receiver operating characteristic curve and compared it with the Khorana score. The concordance index (C- index) of the nomogram was 0.852 [95% confidence interval (CI) 0.823 to 0.880], while the Khorana score was 0.681 (95% CI 0.639 to 0.723). Given its performance, this nomogram could be used to select cancer patients at high risk for VTE and guide thromboprophylaxis treatment in clinical practice, provided it is validated in an external cohort.