489 Time to Following Commands and Favorable Recovery After Severe Traumatic Brain Injury

Abstract

INTRODUCTION: Accurate prognostication after severe traumatic brain injury (TBI) is crucial to goals-of-care discussions and assisting with difficult medical decisions. The capacity to follow commands indicates clinical arousal from unconsciousness and higher level of cognitive function. While it is a predictor of good outcome, time to following commands after injury can widely vary. METHODS: Consecutive participants from 2003-2018 were recruited at the Brain Trauma Research Center (BTRC) using IRB-approved prospective observational study design. Inclusion criteria were age 16-80 years, Glasgow Coma Scale (GCS) score =8 and motor GCS score <6, and Glasgow Outcome Scale - Extended (GOS-E) measure =4 at 2 years post injury. Interquartile range (IQR), mean increases/decreases (B), and 95% confidence intervals (CIs) were reported. Statistical significance was assessed at p=0.05. RESULTS: A total of 580 patients were enrolled in the BTRC from 2003-2018. There were 229 (39.5%) deaths, and 140 (24.1%) patients had favorable outcomes on the GOS-E at 24 months. Average age was 33.7±14.5 years old, median GCS was 7 (IQR 6-7), and median Injury Severity Score (ISS) was 30 (IQR 26-38). Mean time to following commands was 12.7±11.8 days, median GOS-E at 2 years was 6 (IQR 5-7). On multivariate regression, evidence of diffuse axonal injury (DAI) (B=9.2 days [4.8,13.7], p<0.0001) or intraventricular hemorrhage (IVH) (B=6.4 days [0.5,12.3], p<0.035) was associated with longer time to following commands, and patients who developed nosocomial infections (B=6.5 days [1.6-11.4], p<0.01). CONCLUSION: The majority of patients with favorable recovery after severe TBI began to follow commands by 2 weeks. Evidence of DAI, IVH, or nosocomial infections delay cognitive improvement in the acute period during hospitalization. These patients make considerable recovery after discharge and continue to benefit from long-term tailored rehabilitation.