Incidence and risk factor analysis of deep venous thrombosis in patients with severe traumatic brain injury

Abstract

To estimate the incidence and risk factors for deep venous thrombosis (DVT) in patients with severe traumatic brain injury (TBI) treated in the intensive care unit (ICU). 105 patients with TBI admitted to the First Affiliated Hospital of University of Science and Technology of China from January 2016 to June 2018 were enrolled [Glasgow coma scale (GCS) 3-8; concise injury score for other parts ≤ 3]. All patients did not receive any medication or physical measures to prevent DVT during hospitalization. Bilateral compression Doppler ultrasounds of the double lower limbs and upper limbs were performed to clarify the occurrence of DVT on the first day of admission and twice a week until ICU discharge or the death of patient. The examination was performed by a senior ultrasound doctor. It was defined as DVT as long as any deep vein had thrombosis. Patients were divided into two groups according to whether DVT occurred or not during hospitalization. Clinical data such as body mass index (BMI), coagulation index, platelet count (PLT) and deep venous catheterization were obtained from the clinical chemistry laboratory database and patient files. Logistic regression was used to analyze the risk factors of DVT. Binary Logistic regression was used to calculate the predictive probability of risk factors. The predictive value of risk factors and predictive probability to the occurrence of DVT was analyzed by receiver operating characteristic (ROC) curve. In 105 patients with simple TBI, 42 patients developed DVT during hospitalization, and the incidence of DVT was 40%. Univariate Logistic regression showed that high BMI [odds ratio (OR) = 1.490, 95% confidence interval (95%CI) = 1.174-1.891, P = 0.001], high PLT (OR = 1.023, 95%CI = 1.006-1.040, P = 0.007), shorten activated partial thromboplastin time (APTT; OR = 0.913, 95%CI = 0.853-0.978, P = 0.010) and catheterization in deep vein (OR = 0.114, 95%CI = 0.026-0.493, P = 0.004) were risk factors for DVT. It was shown by multivariate regression analysis that high BMI (OR = 1.378, 95%CI = 1.411-1.665, P = 0.001), high PLT (OR = 1.017, 95%CI = 1.003-1.032, P = 0.020), low APTT (OR = 0.920, 95%CI = 0.860-0.982, P = 0.012) and catheterization in deep vein (OR = 0.113, 95%CI = 0.029-0.443, P = 0.002) were independent risk factors for DVT. The predictive probability formula: Logit P = -4.673+0.321×BMI-0.083×APTT+0.017×PLT-2.181×catheterization in deep vein. It was shown by ROC curve analysis that high BMI, high PLT, low APTT and catheterization in deep vein could predict the occurrence of DVT in severe TBI patients, and the area under ROC curve (AUC) was 0.775, 0.709, 0.709 and 0.680, respectively. The AUC of prediction probability was 0.890, and its sensitivity and specificity were respectively increased to 88.10% and 85.71%. The incidence of DVT was higher in patients with simple severe TBI. High coagulation, high BMI, high PLT and catheterization in deep vein are the independent risk factors for DVT.