A Novel Automated Calculation of Basal Cistern Effacement Status on Computed Tomographic Imaging in Traumatic Brain Injury.

Introduction To predict patient outcomes in traumatic brain injury (TBI) lesions, various scores have been proposed, which use objective assessments. These scores, however, rely on the observer's ability to determine them. This study presents a comprehensive, reproducible, and more anatomically stratified objective measurement of the degree of basal cistern effacement in brain computed tomographic (CT) scan images. Methods Patients with TBI admitted from August 2015 to February 2016 were included. The control group consisted of non-trauma patients, who had normal brain CT scans. The images were analyzed by an automated volumetric compression ratio (CR) defined as the volume ratio between the parenchymal tissue and the cerebrospinal fluid (CSF) in the basal cisterns. This value was compared with the TBI severity recorded at each patient's admission and a consensus score of the basal cisterns' degree of effacement by manual analysis. Results Seventy-three TBI patients were admitted. The mean admission Glasow Coma Scale (GCS) score was 9. In the non-TBI control group, 29 patients were enrolled. The average kappa value for the inter-observer agreement was 0.583. The CR had an inverse linear relationship with the severity of the TBI and the degree of effacement of the basal cisterns. The correlation between the CR value in the midbrain and the specialists' consensus determination was statistically significant (p < 0.01). The CR also showed a difference between the TBI and the control groups (p 0.0001). Conclusions The automated CR is a useful objective variable to determine the degree of basal cistern effacement. The proposed ratio has a good correlation with the classical basal cistern effacement classification and TBI severity.