Background and purpose: Alberta Stroke Programme Early CT Score (ASPECTS) is widely used for assessment of early ischemic changes in acute stroke. We hypothesized that CT angiography source image (CTA-SI) ASPECTS correlates better with baseline National Institutes of Health Stroke Scale score, final ASPECTS and neurological outcomes when compared with noncontrast CT ASPECTS.
Methods: We studied patients presenting with acute ischemic stroke and identified proximal arterial occlusions (internal carotid artery, middle cerebral artery M1, and proximal middle cerebral artery M2) from the Calgary CT Angiography database. CT scans were independently read by 3 observers for baseline noncontrast CT ASPECTS, CT angiography source image ASPECTS, and follow-up ASPECTS. Details of demographics and risk factors were noted. A modified Rankin Scale score ≤2 at 3 months was considered a favorable outcome.
Results: We identified 261 patients with proximal occlusions for analysis. We found a better correlation between CT angiography source image ASPECTS and follow-up ASPECTS (Spearman correlation coefficient r=0.65; 95% CI, 0.58 to 0.72; P<0.001) than between noncontrast CT ASPECTS and follow-up CT ASPECTS (r=0.46; 95% CI, 0.36 to 0.55; P<0.001). CT angiography source image ASPECTS correlated better with baseline National Institutes of Health Stroke Scale and 24-hour National Institutes of Health Stroke Scale when compared with noncontrast CT ASPECTS (P<0.001). In an adjusted model including both CT angiography source image ASPECTS and noncontrast CT ASPECTS, CT angiography source image ASPECTS was associated with good outcome (OR, 2.30; 95%, CI, 1.16 to 4.53), whereas noncontrast CT ASPECTS was not (OR, 1.54; 95% CI, 0.84 to 2.82). Among imaging parameters, CT angiography source image ASPECTS was the only independent predictor of good outcome (OR, 2.29; 95% CI, 1.16 to 4.53).
Conclusions: CT angiography source image ASPECTS correlates better with baseline stroke severity, is a better predictor of final infarct extension, and independently predicts neurological outcome than noncontrast CT ASPECTS.