Recovery Trajectories and Long-Term Outcomes in Traumatic Brain Injury: A Secondary Analysis of the Phase 3 Citicoline Brain Injury Treatment Clinical Trial

Abstract

Prospects for recovery after traumatic brain injury (TBI) are often underestimated, potentially leading to withdrawal of care in the comatose TBI patient who may ultimately have a favorable outcome with aggressive care. Outcomes and trajectories of recovery in a large series of patients with TBI were evaluated at 30, 90, and 180 days postinjury. A secondary analysis of the phase 3 Citicoline Brain Injury Treatment (COBRIT) trial was performed analyzing recovery trajectories and long-term outcomes at 30, 90, and 180 days postinjury. A Glasgow Outcome Scale-Extended (GOS-E) score of 5 or higher was considered favorable. Pearson χ2 analysis was used, and a P value of 0.05 was considered significant. A locally weighted, polynomial regression model was used to model recovery trajectories in a nonlinear fashion. Subjects with TBI in the COBRIT trial had high rates of favorable outcome (57% of severe TBI, 86% of moderate TBI, and 93% of complicated mild TBI) at 6-month follow-up. These favorable outcomes often converted from high rates of unfavorable outcome at initial 1-month follow-up (85% of severe TBI, 57% of moderate TBI, and 21% of complicated mild TBI). Recovery trajectories had not plateaued at 6 months, suggesting that further improvement occurs beyond 6 months postinjury. In this secondary analysis of the COBRIT trial, most patients had favorable outcomes by the GOS-E at 6 months postinjury in all complicated mild and moderate TBI groups, with over half of patients with severe TBI achieving a favorable outcome as well. Of subjects in a vegetative state (GOS-E score 2) at 1 month postinjury, 18% improved to a favorable outcome by 6 months postinjury. There was substantial improvement in all groups from 1 to 6 months, and this improvement may continue beyond 6 months. Clinical trials in TBI should consider recovery curves with repeated measures to assess outcomes because arbitrary single-moment outcome determination likely underestimates treatment effect in TBI care.