Establishing a Nomogram to Predict the Risk of Pulmonary Embolism in Oncology Patients

Abstract

Abstract Objective Pulmonary embolism (PE) is a serious disease that can result in high morbidity and mortality among cancer patients. The aim of this study was to create a nomogram to accurately predict PE risk in oncology patients to enhance their medical treatment and management. Methods This study was designed as a retrospective analysis; information on medical history, complications, specific clinical characteristics, and laboratory biomarker results was collected for suspected PE patients admitted to the oncology department at the Affiliated Dongyang Hospital of Wenzhou Medical University between January 2012 and December 2021. A total of 512 patients were randomly divided into training and validation groups based on a 6:4 ratio. LASSO and multivariate logistic regressions were used to develop a nomogram-based scoring model. Model performance was evaluated using receiver operating characteristic (AUC), calibration, and clinical decision curves. Results In our study, over 50 features from 512 patients were analyzed. The nomogram-based scoring model was established using five predictive features, including the neutrophil count, sex, systolic blood pressure, surgical status, and D-dimer levels, which achieved AUC values of 0.758 and 0.702 in the training (95% CI 0.695–0.804) and validation cohorts (95% CI 0.630–0.776), respectively. For our model, the sensitivity was 85.58%, the specificity was 35.78%, the positive predictive value was 72.44%, and the negative predictive value was 55.71%. The calibration curve results showed a strong consistency between the probability predicted by the nomogram and the actual probability. Decision curve analysis (DCA) also demonstrated that the nomogram-based scoring model produced a favorable net clinical benefit. Conclusions In this study, we successfully developed a novel numerical model that can predict PE risk in oncology patients, enabling appropriate selection of PE prevention strategies and reducing unnecessary computed tomography pulmonary angiography (CTPA) scans and their associated adverse effects.