Development of a Predictive Model for Pulmonary Embolism in Gastrointestinal Cancer Patients with Isolated Distal Deep Vein Thrombosis

ObjectiveTo identify factors associated with pulmonary embolism (PE) in gastrointestinal cancer patients with isolated distal deep vein thrombosis (IDDVT) and to develop a nomogram-based predictive model for PE.MethodsFrom January 2021 to January 2024, a total of 204 patients with gastrointestinal malignancies diagnosed with IDDVT were retrospectively analyzed. Pulmonary embolism was confirmed by computed tomography pulmonary angiography. Logistic regression analysis was performed to identify relevant risk factors, and a nomogram-based risk prediction model was subsequently established.ResultsAmong the 204 patients with IDDVT, 86 (42.2%) developed PE. Univariate analysis showed that age ≥60 years, diabetes mellitus, coronary heart disease, tumor stage, and glucocorticoid use were significantly associated with PE (P < .05). Multivariate logistic regression identified age ≥60 years, coronary heart disease, and glucocorticoid use as factors independently associated with PE (P < 0.05). A nomogram was constructed based on these variables to predict PE occurrence. The area under the receiver operating characteristic (ROC) curve was 0.77 (95% CI 0.71-0.83), with a sensitivity of 64.2% and specificity of 76.6%.ConclusionThis study observed a substantial frequency of PE among gastrointestinal cancer patients with IDDVT. Age ≥ 60 years, coronary artery disease, and lack of glucocorticoids were independently associated with PE in patients with IDDVT. With these risk factors in mind, we developed a nomogram risk prediction model to help detect concurrent PE early.