Computed tomography predictors for in-hospital mortality in severe and moderate head injury patients

Abstract

Background:Traumatic brain injury is single leading cause of death and disability following injury. Clinical information collected at baseline and computed tomography (CT) brain scan findings may predict in-hospital mortality rate. The aim of this study was to evaluate which features on the admission CT scan might add significantly to other baseline clinical information for predicting mortality in patients with head injury.Methods: Baseline CT scans were reviewed for patients with moderate and severe head injuries from head injury registry from October 2007 to October 2009. Baseline demographic and injury status and outcome at discharge or death were recorded. Details from the CT scan using marshals grading was used along with other CT scan findings. CT characteristics like diffuse cerebral edema, mass lesions like extra dural haemotoma, subdural haemotoma, intra parenchymal haemorrhage, traumatic sub arachnoid haemorrhage, intra ventricular haemorrhage, basal cisterns, midline shift were evaluated. All the patients were classified into marshalls divisions. If required repeated scans were evaluated and the worst CT scan has been taken into consideration. Results:211 CT scans were read from patients with moderate severe head injuries. Out of the 24 patients in grade I, 4 died accounting for 16.666%. In grade II out of 47 patients 9 died accounting for 19.148%. In grade III out of 24 patients 5 died accounting for 20.833%. In grade IV out of 30 patients 11 died accounting for 36.666%. In grade V out of 76 patients 20 died accounting for 27.631%. In grade VI out of 10 patients 6 died accounting for 60%. In this study of 211 patients 55 died accounting for 26.06%.Conclusions:Marshall`s CT grading holds good for mortality prediction in moderate and severe head injury patients. Status of basal cisterns and midline shift serves as independent predictors for in-hospital mortality.