Comparative Study of Antemortem Computed Tomography (CT) Findings With Autopsy Findings in Head Injury Cases

Head injury is a major cause of morbidity and mortality worldwide, and accurate assessment of traumatic head injuries is important for both clinical management and medicolegal investigations. Computed tomography (CT) is routinely used for the evaluation of head injuries in living patients, whereas medicolegal autopsy remains the standard method for postmortem assessment. This prospective comparative observational study was conducted to compare antemortem CT scan findings with autopsy findings in fatal head injury cases.

A total of 34 cases that underwent both antemortem CT examination and subsequent medicolegal autopsy were included. Findings related to scalp injuries, skull fractures, extradural hemorrhages (EDHs), subdural hemorrhages, subarachnoid hemorrhage (SAH), brain contusions, brain lacerations, intraventricular hemorrhage (IVH), and cerebellar herniation were compared. Agreement between CT scan and autopsy findings was assessed using percentage agreement, Cohen's kappa coefficient, and McNemar's test.

Substantial agreement was observed for skull fractures, EDH, IVH, and cerebellar herniation, whereas lower agreement was observed for scalp injuries, SAH, and brain lacerations.

Antemortem CT scan demonstrated good concordance with autopsy for skull fractures and major intracranial hemorrhages and proved to be a valuable adjunct in the evaluation of fatal head injury cases. However, a CT scan could not completely replace conventional medicolegal autopsy. These findings support the complementary role of CT imaging in forensic practice and its potential contribution to the development of CT-based virtual autopsy in head injury cases. Furthermore, by directly comparing antemortem CT findings with subsequent autopsy findings, this study provides evidence that may facilitate future integration of antemortem CT data into medicolegal practice and guide further research on CT-based virtual autopsy, particularly in resource-limited settings.