Predictive factors for traumatic cerebral contusion volume, expansion, and outcomes

OBJECTIVE: Traumatic hemorrhagic cerebral contusions are a well-established cause of morbidity and mortality in neurosurgery. This study aimed to determine prognostic factors for long-term functional outcomes and longitudinal contusion volume changes in traumatic brain injury (TBI) patients. METHODS: Data from 285 patients with traumatic cerebral contusions were retrospectively reviewed to identify variables predictive of initial contusion volume, contusion expansion on short-term follow-up imaging, and functional outcomes according to the modified Rankin Scale (mRS). Predictors of these variables were identified using a stepwise logistic regression analysis. RESULTS: Older age, larger initial contusion volumes, and lower presenting Glasgow Coma Scale (GCS) scores were associated with worse functional outcomes (mRS score ≥ 3). Patients with contusion volumes ≥ 15 ml at presentation had lower GCS scores and longer ICU stays (in days). Older age (OR 1.043, CI 1.024-1.063), need for a craniotomy or craniectomy (OR 2.562, CI 1.010-6.502), longer ICU stay (OR 1.092, CI 1.034-1.154), and lower total GCS score (OR 0.781, CI 0.729-0.836) were associated with worse functional outcomes. Additionally, lower admission GCS verbal score was a significant predictor of larger initial contusion volume (OR 0.779, 95% CI 0.667-0.911) and contusion expansion during hospitalization (OR 0.649, 95% CI 0.497-0.847). CONCLUSIONS: Functional outcomes in traumatic cerebral contusion patients may be associated with age and admission GCS score, and verbal GCS score may predict initial contusion volume and contusion expansion. These findings supplement an evolving understanding of factors that influence outcomes in patients with cerebral contusions, and further study into the utility of GCS to guide these decisions could help to guide the clinical management of these highly complex patients.