CT PERFUSION IN ACUTE ISCHAEMIC STROKE: DO WE COVER THE LESION AND WHAT DOES IT MEAN?

Introduction Identifying and quantifying infarct core and penumbra in acute ischaemic stroke patients may improve patient selection for thrombolysis. Although CT Perfusion (CTP) imaging has several potential advantages over multimodal MRI, 64 slice CT scanners are limited to 40 mm of z–directional coverage in a single perfusion sequence. The significance of this limited coverage–and in particular the significance of a lack of a perfusion deficit within a 40 mm slab–has not been directly explored. We aimed to: 1) Compare the proportion of subjects within a prospective hyperacute stroke cohort without a deficit on baseline CTP to on non–contrast CT (NCCT) 2) examine whether the CTP slab tended to be acquired too superiorly or too inferiorly in our cohort and 3) examine the clinical and radiological characteristics of patients without a perfusion deficit. Methodology Patients recruited to two recently completed multi–site prospective observational studies of sub–6 hour acute ischaemic stroke were included for analysis. Subjects underwent admission assessment including National Institute of Health Stroke Scale (NIHSS) prior to NCCT and CTP (with 40 mm slab coverage placed at the level of the basal ganglia) acquisition and follow up assessment using the modified Rankin Scale (mRS). Acute NCCT imaging was scored blinded to CTP, according to the Alberta Stroke Program Early CT Score (ASPECTS). CTP was post–processed for total volumes of penumbra, infarct core and total deficit (penumbra+core), according to published thresholds. These volumes were also recorded separately for the most superior and most inferior slices. Clinico–radiological characteristics of patients without perfusion deficits were identified and compared to those with perfusion deficits using Mann–Whitney U tests. Results 129 subjects with a final clinical diagnosis of ischaemic stroke were included for analysis. Median admission NIHSS was 11 (Range: 1–33). Median onset–CT time was 170 minutes (Range: 15–380). Sixty three subjects (49%) had no deficit on baseline NCCT ASPECTS and 37 (29%) had no acute CTP deficit. Of those with no deficit on NCCT, 31 (49%) had a CTP deficit and 7 (11%) had a deficit >20cm3. In contrast, of those without CTP deficits, only 5 (13.5%) had a deficit on NCCT ASPECTS. Of those with CTP deficits, 79 subjects (86%) had a deficit in the most superior slice and 60 (65%) had a deficit in the most inferior slice. Only four subjects (3%) had likely total lesion coverage, with a CTP deficit but no deficit in the outermost slices. Subjects without CTP deficits had significantly milder admission NIHSS deficits (p Conclusions Even with limited z–directional coverage CTP is more sensitive than NCCT ASPECTS scoring to detect evidence of acute stroke. In this cohort CTP acquisition at the basal ganglia was more often too inferior than it was too superior to maximise lesion coverage. Patients without perfusion deficits have better functional outcomes than those with perfusion deficits.