Objective: To demonstrate that the prognosis for patients with traumatic brain injury (TBI) admitted to rehabilitation can be established with use of principled neurologic diagnosis and predictor variables of established value in neurosurgical populations.
Design: A cohort of patients with TBI accumulated at rehabilitation admission were followed up for 1 year. Severity measures (Glasgow Coma Scale score, length of coma, and duration of posttraumatic amnesia) and information to generate neuropathologic profiles were gathered retrospectively and prospectively; outcome measures were obtained prospectively.
Setting: The TBI rehabilitation unit in a freestanding rehabilitation hospital.
Patients: A consecutive sample of 243 patients with TBI admitted to a rehabilitation unit (age range, 8 through 89 years).
Main outcome measures: Functional outcome measured by the Glasgow Outcome Scale at 6 and 12 months after injury.
Results: Posttraumatic amnesia had a clear, predictable relationship to length of coma in patients with diffuse axonal injury (R2 = .58, P < .0001). Severity measures, particularly duration of posttraumatic amnesia, correlated with the Glasgow Outcome Scale score at 6 and 12 months after injury (R2 = .45, P < .0001, R2 = .48, P < .0001), strongly in patients with diffuse axonal injury but poorly in patients with primarily focal brain injury. Age was an important factor in recovery, beginning at age 40 years; older patients had significantly longer posttraumatic amnesia and worse functional outcome at any severity.
Conclusions: The early course of recovery and functional outcome in TBI can be characterized in neurorehabilitation populations and is highly dependent on specific neuropathologic diagnosis, severity, and age. Predictions that employ traditional measures of severity are most relevant in patients with diffuse axonal injury. Age has a potent, complex effect on recovery, particularly beyond age 40 years.