377 Craniectomy Versus Craniotomy in Traumatic Brain Injury: A Propensity-Matched Analysis of Long-Term Functional and Quality of Life Outcomes. (August 2016)

• Record Type: Journal Article
• Authors: Kelly, Michael L.; Shammassian, Berje Haroutuon; Roach, Mary Joan; Thomas, Charles; Wagner, Amy K.
• Is Part Of: Clinical neurosurgery. Volume 63(2016)Supplement 1
• Publication Date: 2016-08
• Subjects: Nervous system -- Surgery -- Congresses; Neurosurgery; Nervous system -- Surgery; Neurologie; Congresses; Conference papers and proceedings; 617.48
• DOI: 10.1227/01.neu.0000489865.02721.2a
• Languages: English
• ISSNs: 0069-4827

Abstract

Abstract : INTRODUCTION: Surgery for patients with traumatic brain injury (TBI) remains controversial. Studies suggest that craniectomy (CE) may be superior to craniotomy (CO) by reducing intracranial pressure and limiting postoperative brain swelling. Few studies report comprehensive long-term functional and quality-of-life outcomes. METHODS: All patients with TBI who underwent CE or CO were extracted from the TBI Model Systems database from 2000 to 2012. A 1:1 propensity matching with replacement technique was used to match baseline characteristics including age, Glasgow Coma Score, Marshall CT score, TBI subtype, and intracranial hypertension across groups. The matched sample was analyzed for outcomes during hospitalization, acute rehabilitation, and up to 2 years of follow-up. RESULTS: We identified 1470 patients in both CE and CO groups. Baseline characteristics were well-matched between groups (standardized mean difference <10). CE patients demonstrated a longer length of stay (LOS) in the hospital (median days: 22 vs 18; P < .001) and acute rehabilitation (26 vs 21; P < .001). CE patients were more likely to be hospitalized at 1-year follow-up (39% vs 25%; P < .001) for reasons other than cranioplasty including seizures (12% vs 8%; P < .001), neurological events (ie hydrocephalus) (9% vs 4%; P < .001), and infections (10% vs 6%; P < .001). CE patients were significantly more impaired on the Extended Glasgow Outcome Scale, required more supervision, and were less likely to be employed or living at home at 1-year postinjury. No difference was observed in Satisfaction with Life Scale (SWL) scores at 1 year. Kaplan-Meier estimates for mortality at 1- and 2-year follow-up showed no difference between CE and CO groups (hazard ratio: 0.57; P = .4). CONCLUSION: Patients who underwent CE vs CO after TBI had longer LOS, decreased functional status, and more rehospitalizations. Survival at 2 years and SWL scores remained similar. CE for TBI is associated with worse functional outcomes.