Comparison of Magnetic Resonance Imaging with Computed Tomography in Subarachnoid Hemorrhage

Kim, Tae Hoon; Kim, Gab Teug

Abstract

PURPOSE
Misdiagnosis of subarachnoid hemorrhage (SAH) can result in considerable mortality and morbidity. Computed tomography (CT) has high sensitivity for detection of acute SAH, but falls off rapidly over time, and approaches 0% at three weeks. The aim of this study was to conduct a comparison of magnetic resonance imaging (MRI) and CT in detection of SAH in acute and subacute, and chronic stages.
METHODS
This retrospective study included 62 patients with spontaneous SAH from January 2006 to December 2011. For each patient, we obtained non-enhanced CT scans, fluid-attenuated inversion recovery (FLAIR), and T2-weighted gradient-echo (T2*) MRI images. We defined SAH based on areas of high attenuation on non-enhanced CT scans, regions of hyperintensity on FLAIR images, and regions of hypointensity on T2* images in intracranial subarachnoid spaces. In order to investigate the superiority of tools for diagnosis of SAH, comparison of sensitivity of CT scans and MRI was performed.
RESULTS
Sensitivity of CT to SAH was 93.5% on the first day, but fell off rapidly with time, and approached 0% at 20 days. Sensitivity of MRI was not affected by stages and amounts of bleeding (p>0.05). Sensitivity of MRI was higher than that of CT in SAH of Fisher grade 0-1 of subacute stage of bleeding (p=0.001) and in all cases of chronic stage of bleeding (p=0.000). FLAIR images were superior to T2* images, but without statistical significance (p>0.1).
CONCLUSION
MRI was superior to CT in detection of subacute and chronic SAH, as well as a small amount of SAH.