Abstract
To construct a precise and personalized nomogram model and assess the risk factors associated with deep vein thrombosis (DVT) in patients undergoing (traumatic brain injury) TBI. Clinical data from TBI patients between January 2015 and January 2020 were retrospectively gathered. Divided into the model training set and the model validation set in chronological order. The risk factors for DVT were analyzed using LASSO regression and multifactor logistic regression. Post-modeling assessments were conducted for differentiation, consistency, and clinical efficacy. LASSO regression results showed that Age, BMI, smoking history, balance of intake and output, interval between operation and injury, preoperative D-dimer, preoperative FIB, and preoperative PT were the risk factors of DVT in patients with TBI after surgery (P < 0.05). The nomograph model was constructed using the above 8 risk factors. The AUC of the training set and validation set models were 0.833 (0.790-0.876) and 0.815 (0.748-0.882) respectively, and the Brier values of the training set and verification set were 0.157 and 0.165 respectively, indicating that the calibration of the model was good. Clinical decision curves for both sets confirmed the model's high net benefit, indicating its effectiveness. Age, BMI, smoking history, balance of intake and output, interval between operation and injury, preoperative D-dimer, preoperative FIB, and preoperative PT are identified as significant risk factors for DVT development in TBI patients. The risk prediction model exhibits robust consistency and prediction efficiency, offering valuable insights for medical practitioners in early identification and targeted invervention for high-risk TBI patients prone to DVT.