Detection of cerebral aneurysms in nonenhanced CT images of patients with subarachnoid hemorrhage using the filling defect sign

Abstract

Purpose: Non- traumatic subarachnoid hemorrhage (SAH) constitutes a life-threatening event, which is caused in 85% by ruptured aneurysms. Identifying those aneurysms as the bleeding cause is essential for further therapy. This study explores whether relatively hypodense sparing of aneurysms in the blood-filled subarachnoid space (filling defect sign = FD) on nonenhanced CT scans (NECT) can help estimate aneurysm location and size. Methods: NECT of fifty patients with aneurysmal SAH, who were also examined with a catheter angiography, were retrospectively analyzed by three neuroradiologically experienced radiologists. They rated the aneurysm location and size by applying the FD. Logistic regression analysis was performed to determine if a correlation between aneurysm location, aneurysm size, Fisher score and the detectability of a FD exists. For quantitative analysis the average densities of the aneurysm were compared to that of surrounding structures. Results: In 75% of the cases, the aneurysm was identified correctly by using the FD. The highest detection rate was given for MCA (90%) and BA aneurysms (90%). Aneurysm location (p = .019), size (p = .020), and the Fisher score (p = .008) significantly influenced aneurysm detection using the FD. The aneurysm size measured in correctly identified aneurysms correlated significantly with the sizes measured in the catheter angiography (p < .001). A cut-off point of 51 HU could be determined to distinguish aneurysm from surrounding SAH with a specificity of 92% and a sensitivity of 86%. Conclusions: In three-fourths aneurysm location and size can be determined on NECT using the FD. An additional quantitative analysis can support the diagnosis with a high specificity and sensitivity.