Neurologic Outcome After Decompressive Craniectomy: Predictors of Outcome in Different Pathologic Conditions

OBJECTIVE: Decompressive craniectomy (DC) has been proposed as a lifesaving treatment in patients with elevated intracranial pressure, but its effectiveness on reaching a favorable neurologic outcome remains unclear. We identified predictors of outcome in a large, single-center cohort of patients undergoing DC for different pathologic conditions. METHODS: This retrospective study included all patients undergoing DC from 2006 to 2014. The 1-year outcome, assessed using the Glasgow Outcome Scale (GOS), was dichotomized into favorable (GOS 4e5) and unfavorable (GOS 1e3) outcome. Predictors of outcome were identified by analyzing patient characteristics. RESULTS: DC was performed in 204 patients for ischemic stroke (n [57), traumatic brain injury (n [50), aneurysmal subarachnoid hemorrhage (aSAH) (n [44), intracerebral hemorrhage (ICH) (n [29), cerebral venous thrombosis (CVT) (n [14), or other indications (n [10). Overall, 69 (34%) patients survived favorably, 39 (19%) survived unfavorably, and 96 (47%) died. Higher age, poor Glasgow Coma Scale score, intubated status before DC, bilateral absence of pupillary light reflexes, DC for aSAH, and additional surgeries after DC (excluding cranioplasty) were significant predictors of unfavorable outcome. When patients were sorted for pathologic conditions and predictors of outcome, favorable outcome rates differed remarkably, ranging from 91% for CVT patients undergoing uncomplicated DC to 0% for aSAH patients undergoing DC for secondary infarction or ICH patients with unilateral or bilateral abnormal pupillary light reflexes upon admission. CONCLUSIONS: Long-term neurologic outcome after DC differed remarkably among subpopulations of patients, with favorable outcome rates ranging from 0% to >90%.