Venous thromboembolic events in critically ill traumatic brain injury patients

Abstract

PURPOSE: To estimate the prevalence, risk factors, prophylactic treatment and impact on mortality for venous thromboembolism (VTE) in patients with moderate to severe traumatic brain injury (TBI) treated in the intensive care unit. METHODS: A post hoc analysis of the erythropoietin in traumatic brain injury (EPO-TBI) trial that included twice-weekly lower limb ultrasound screening. Venous thrombotic events were defined as ultrasound-proven proximal deep venous thrombosis (DVT) or clinically detected pulmonary embolism (PE). Results are reported as events, percentages or medians and interquartile range (IQR). Cox regression analysis was used to calculate adjusted hazard ratios (HR) with 95% confidence intervals (CI) for time to VTE and death. RESULTS: Of 603 patients, 119 (19.7%) developed VTE, mostly comprising DVT (102 patients, 16.9%) with a smaller number of PE events (24 patients, 4.0%). Median time to DVT diagnosis was 6 days (IQR 2-11) and to PE diagnosis 6.5 days (IQR 2-16.5). Mechanical prophylaxis (MP) was used in 91% of patients on day 1, 97% of patients on day 3 and 98% of patients on day 7. Pharmacological prophylaxis was given in 5% of patients on day 1, 30% of patients on day 3 and 57% of patients on day 7. Factors associated with time to VTE were age (HR per year 1.02, 95% CI 1.01-1.03), patient weight (HR per kg 1.01, 95% CI 1-1.02) and TBI severity according to the International Mission for Prognosis and Analysis of Clinical Trials risk of poor outcome (HR per 10% increase 1.12, 95% CI 1.01-1.25). The development of VTE was not associated with mortality (HR 0.92, 95% CI 0.51-1.65). CONCLUSIONS: Despite mechanical and pharmacological prophylaxis, VTE occurs in one out of every five patients with TBI treated in the ICU. Higher age, greater weight and greater severity of TBI increase the risk. The development of VTE was not associated with excess mortality.