Clinical value of predicting lower extremity deep vein thrombosis based on nomogram combined with four-item thrombosis assay in transfused ICU patients

This study aims to develop and evaluate a nomogram incorporating four thrombosis biomarkers—plasmin-α₂-plasmin inhibitor complex (PIC), tissue-type plasminogen activator-inhibitor complex (t-PAIC), thrombin-antithrombin complex (TAT), and thrombomodulin (TM)—for the prediction of lower extremity deep vein thrombosis (DVT) in critically ill patients receiving blood transfusions. Fifty-six eligible patients, enrolled between February 2020 and February 2022, were prospectively followed for 12 months and subsequently stratified into a DVT group (n = 44, incidence 78.57%) and a non-DVT group (n = 12). Comparative analysis at baseline revealed significantly elevated plasma levels of PIC, t-PAIC, TAT, and TM in the DVT group (P < 0.05). Multivariate logistic regression confirmed the four markers as independent DVT risk factors. The constructed nomogram exhibited favorable calibration, as evidenced by consistent calibration curves and a nonsignificant Hosmer–Lemeshow goodness-of-fit test (χ² = 0.367, P = 0.478). Receiver operating characteristic (ROC) curve analysis produced an area under the curve (AUC) of 0.787 (95% confidence interval CI: 0.643–0.832), with a sensitivity of 86.76% and a specificity of 70.54%. Collectively, these findings indicate that PIC, t-PAIC, TAT, and TM serve as independent predictors of DVT in transfused ICU patients, and that the proposed nomogram offers a robust and clinically applicable tool for individualised risk stratification.