Identification of Traumatic Subarachnoid Hemorrhage Using Susceptibility Weighted Imaging

Introduction: The occurrence of subarachnoid hemorrhage (SAH) in traumatic brain injury (TBI) is 12% ~ 53% and related to poor outcome (1, 2). Early diagnosis of SAH and active intervention in removing hemorrhage from subarachnoid space is of great significance in the management of acuteTBI. Susceptibility weighted imaging (SWI) has been successfully applied in TBI and proved to be 6 times more sensitive than conventional gradient echo imaging in detecting intra-cerebral hemorrhage (3). To date, no study has been done using SWI to detect SAH. In this study, we explore the sensitivity of SWI in detecting SAH, compare it with CT and FLAIR, and summarize the characteristics of SAH on SWI. Materials and Methods: Both MRI and CT data from 20 acute TBI patients (16 males and 4 females, 39.1±19.6 yrs) with SAH were retrospectively reviewed by two neuroradiologists. The CT (Siemens, SOMATON) acquisition parameters were as follows: x-ray tube current = 360 mA, kvp = 120 kV, resolution = 0.5 x 0.5 x 6 mm. Four MRI sequences were collected at 3T (Siemens TRIO), including: T1W, T2W, FLAIR and SWI. According to the location, five regions of SAH were defined: interhemispheric fissure (IF), sylvian cistern (SC), sulci of cerebral convex (SCC), basilar cistern (BS) and intra-ventricle (IV). Basilar cistern includes pontine, interpeduncular, ambient, and quadrigeminal cisterns. SAH is defined as high density on CT in the subarachnoid space, hyper-intense signal on FLAIR, and hypo-intense signal on SWI with the normal veins excluded. The two neuroradiologists first independently reviewed CT, FLAIR and SWI separately and then discussed their results and came to agreement.