Establishment of a predictive nomogram for in-hospital mortality risk in deep vein thrombosis patients in intensive care units

Background Deep vein thrombosis (DVT) is a common complication among patients in intensive care unit (ICU) and may lead to pulmonary embolism and other severe complications, with a significant impact on patient outcomes. The purpose of this study is to establish and assess a nomogram for estimating the probability of in-hospital mortality in DVT patients in ICU. Methods The clinical information of 187 patients who experienced DVT while in the ICU at the First Affiliated Hospital of Jinan University between January 2017 and July 2024 was examined in this retrospective cohort analysis. Independent risk factors that were significantly linked to in-hospital death from DVT in ICU patients were analyzed using univariate and multivariate logistic regression analysis. To predict the risk of in-hospital death in ICU patients with DVT, a nomogram model was developed and assessed using decision curve analysis (DCA), integrated discrimination improvement (IDI), net reclassification improvement (NRI), and the area under the receiver operating characteristic (ROC) curve (AUC). Results Univariate and multivariate logistic regression analyses identified 9 independent risk factors strongly associated with in-hospital mortality in ICU patients with DVT. These factors included BMI (OR = 0.08, 95% CI = 0.01–0.90, P = 0.04), Acute Physiology Score II (OR = 1.17, 95% CI = 1.07–1.30, P = 0.001), blood carbon dioxide concentration (OR = 1.02, 95% CI = 1.00–1.04, P = 0.01), blood calcium level (OR = 1.86, 95% CI = 1.31–5.50, P = 0.02), platelet count (OR = 0.98, 95% CI = 0.97–0.99, P = 0.005), diastolic blood pressure (OR = 1.08, 95% CI = 1.01–1.15, P = 0.02) and peripheral oxygen saturation (OR = 0.92, 95% CI = 0.84-1.00, P = 0.04). Based on these independent risk factors, a nomogram was established to estimate in-hospital mortality in ICU patients with DVT. The area under the ROC curve (AUC = 0.8471, 95% CI = 0.769–0.925) and DCA demonstrated excellent predictive accuracy and meaningful net clinical benefit of the model. Similarly, NRI and IDI results confirmed the robust predictive performance of the model. Conclusion This study discovered independent risk factors linked to in-hospital mortality in DVT patients in ICU and created a prediction nomogram model. In clinical practice, systematically screening ICU patients for these relevant risk factors with this nomogram can predict the DVT mortality, leading to better intensive care.