Abstract: Objective To analyze the risk factors for lower extremity deep vein thrombosis (LEDVT) formation in ICU polytrauma patients and to evaluate the predictive value of combined inflammatory biomarkers for LEDVT development in this population. Methods A retrospective study was conducted to collect clinical data from polytrauma patients admitted to the Surgical ICU and Emergency ICU of the Second Hospital of Lanzhou University between May 2021 and December 2024. Patients were categorized into thrombus and non-thrombus groups based on lower extremity vascular color Doppler ultrasound findings during hospitalization. Multivariable Logistic regression analysis was employed to identify independent risk factors for lower extremity deep vein thrombosis (LEDVT) and construct a predictive model. The model's performance was evaluated using receiver operating characteristic (ROC) curves and calibration curves. Results A total of 401 polytrauma patients were included, comprising 279 in the thrombosis group and 122 in the non-thrombosis group. Multivariate logistic regression analysis revealed that the following were independent risk factors for LEDVT in ICU polytrauma patients: Age (OR=1.09), Injury Severity Score (ISS) (OR=1.18), C-reactive protein (CRP) (OR=1.01), platelet-to-lymphocyte ratio (PLR) (OR=0.99), albumin (ALB) (OR=0.93), antithrombin III (ATⅢ ) (OR=1.03), and central venous catheterization (OR=3.97). An LEDVT risk assessment model for ICU polytrauma patients was constructed using the inflammatory biomarker PLR combined with age, ISS, ALB, CRP, and ATⅢ. The area under the ROC curve (AUC) for this model was 0.888 (95% CI: 0.85–0.91). This AUC was significantly higher than those of the individual independent risk factors: Age (AUC=0.734, 95% CI: 0.66–0.77), ISS (AUC=0.773, 95% CI: 0.72–0.81), CRP (AUC=0.657, 95% CI: 0.60–0.70), ALB (AUC=0.592, 95% CI: 0.54–0.64), ATⅢ (AUC=0.774, 95% CI: 0.73–0.81), and central venous catheterization (AUC=0.590, 95% CI: 0.54–0.63). Conclusion PLR is an independent risk factor for LEDVT in ICU polytrauma patients. The prediction model constructed by combining PLR with age, ISS, CRP, ALB, and ATⅢ demonstrates high predictive value and clinical significance.