Comparative Radiographic Factors Predicting Functional Outcome After Decompressive Craniectomy in Severe Traumatic Brain Injury.

Abstract

Decompressive craniectomy (DC) is a last-tier therapy in the treatment of raised intracranial pressure after traumatic brain injury (TBI). We report the association of comparative radiographic factors in predicting functional outcomes after DC in patients with severe TBI. A retrospective analysis of a prospectively maintained database of cases between 2015 and 2018 at an academic tertiary care hospital was carried out. Univariate and multivariable regression analyses were performed for an array of comparative radiographic variables (pre- and post-DC) in relationship to functional outcome according to Glasgow Outcome Scale Extended (GOSE) at 180 days. GOSE was further dichotomized into favorable (GOSE:5-8) and unfavorable (GOSE:0-4) functional outcomes. All associations were reported as odds ratio (OR) with 95% confidence interval (CI). Statistical analysis included a cohort of 43 patients with a median age of 30.5 years (range: 18-62 years). The median GOSE at 180 days was 7. Multivariable regression analysis after adjusting for confounding variables (age, sex, comorbidities, site of surgery and size of decompression) showed that comparative radiographic findings of midline shift (MLS) > 10 mm (OR 3.2 (95% CI 1.25-8.04); P = 0.01); external cerebral herniation (ECH) > 2.5 cm (OR 2.5 [95% CI 1.18-5.2]; P = 0.02); and effacement of basal cisterns (OR 3.9 [95%CI 1.1-13.9]; P = 0.03), were significant independent predictors of poor functional outcome at 180 days after DC for severe TBI. However, the presence of infarction (OR 2.7 [95%CI 0.43-17.2]; P = 0.28) and absence of gray-white matter differentiation (OR 0.18 [95%CI 0.03-1.2]; P = 0.07) did not reach statistical significance. The comparative radiographic findings that include MLS > 10mm, ECH > 2.5cm, and effacement of basal cisterns are predictive of poor functional outcome in severe TBI.