Single Subcortical Infarct: Pathomechanism Assessed by Thin-Section Computed Tomography Perfusion

Abstract

The pathomechanism of a single subcortical infarct (SSI) may be better determined by assessing the perfusion status between parent artery and ischemic lesion. We aimed to compare the classifications into branch atheromatous disease (BAD) versus non-BAD based on diffusion-weighted imaging (DWI) or computed tomography perfusion (CTP), and to test whether a CTP-based classification improves the predicting power for progression in SSI (PSSI) compared to that by DWI. We enrolled 109 consecutive patients with SSI examined by whole-supratentorial brain CTP and follow-up DWI. Time-to-drain (TTD) maps were calculated from 1-mm dynamic CTP data. BAD was assumed when either the ischemic lesion extended to the basal surface of the parent artery on axial DWI or the hypoperfused area (TTD ≥ 5 seconds) was <5 mm apart from the cerebrospinal fluid perforators interface on both coronal and sagittal CTPs. We tested the relationship between DWI and CTP for determining BAD, and compared demographics, imaging, and the frequency of PSSI between the BAD and non-BAD based on CTP. Multivariable regression analysis was performed to determine predicting factors for PSSI. On DWI, 66 of 109 patients (60.6%) were classified as BAD; on CTP, 32 patients were classified as BAD (29.4%), showing significant difference (P = .047). PSSI was significantly different between BAD versus non-BAD by CTP (40.6% versus 11.7%, P = .002), but not different by DWI (21.2% versus 18.6%, P = .930). BAD-type perfusion was the only independent predictor for PSSI (OR, 5.209; 95% CI, 1.745-15.555; P = .003). The classifications of SSI with and without BAD by CTP and DWI are significantly different. CTP may help to predict PSSI.