Functional outcome in patients with traumatic or hemorrhagic brain injuries undergoing decompressive craniectomy versus craniotomy and 6-month rehabilitation

Abstract Decompressive craniectomy (DC) and craniotomy (CT) to treat increased intracranial pressure after brain injury are common but controversial choices in clinical practice. Studying a large cohort of patients with traumatic brain injury (TBI) and hemorrhagic stroke (HS) on rehabilitation pathways, we aimed to determine the impact of DC and CT on functional outcome/mortality, and on seizures occurrence. This observational retrospective study included patients with either TBI, or HS, who underwent DC or CT, consecutively admitted to our unit for 6-month neurorehabilitation programs between January 1, 2009 and December 31, 2018. Neurological status using Glasgow Coma Scale (GCS), and rehabilitation outcome with Functional Independence Measure, both assessed at baseline and on discharge, post-DC cranioplasty, prophylactic antiepileptic drug use, occurrence of early/late seizures, infectious complications, and death during hospitalization were evaluated and analyzed with linear and logistic regression models. Among 278 patients, DC was performed in 98 (66.2%) with HS, and in 98 (75.4%) with TBI, whilst CT in 50 (33.8%) with HS, and in 32 (24.6%) with TBI. On admission, GCS scores were lower in patients treated with CT than in those with DC (HS, p = 0.016; TBI, p = 0.024). Severity of brain injury and older age were the main factors affecting functional outcome, without between-group differences, but DC associated with worse functional outcome, independently from severity or type of brain injury. Unprovoked seizures occurred post-DC cranioplasty more frequently after HS (OR = 5.142, 95% CI 1.026–25.784, p = 0.047). DC and CT shared similar risk of mortality, which associated with sepsis (OR = 16.846, 95% CI 5.663–50.109, p < 0.0001), or acute symptomatic seizures (OR = 4.282, 95% CI 1.276–14.370, p = 0.019), independently from the neurosurgery procedures. Among CT and DC, the latter neurosurgical procedure is at major risk of worse functional outcome in patients with mild-to-severe TBI, or HS undergoing an intensive rehabilitation program. Complications with sepsis or acute symptomatic seizures increase the risk of death