Conventional enhancement CT: a valuable tool for evaluating pial collateral flow in acute ischemic stroke.

Abstract

BACKGROUND: To establish an easy and rapid method for evaluating pial collateral flow, we compared the Alberta Stroke Program Early CT Score (ASPECTS) on nonenhanced CT (NECT), conventional contrast-enhanced CT (CECT), and CT angiography source images (CTA-SI) in patients with acute ischemic stroke.

METHODS: We reviewed 55 consecutive patients with acute ischemic stroke involving the anterior circulation who underwent thrombolysis within 6 h of referral to the stroke center. We evaluated axial images using NECT, CECT and CTA-SI. Pial collateral formation was graded as fair (1-2 points) or bad (3-5 points) based on 4-vessel angiography. The outcomes were dichotomized into good (modified Rankin Scale, mRS 0-2) or poor (mRS 3-6) using a 90-day mRS.

RESULTS: Demographics (age, sex, initial National Institutes of Health Stroke Scale score, time to CT acquisition and stroke subtypes) did not significantly differ between patients with fair or bad collateral formation. ASPECTS on CECT (r = -0.788, p < 0.0001) was more inversely correlated with pial collateral formation than ASPECTS on NECT (r = -0.557, p < 0.0001) or ASPECTS on CTA-SI (r = -0.662, p < 0.0001). Furthermore, ASPECTS on CECT demonstrated a high discriminative capability, with an area under the receiver operating characteristic curve of 0.885 for fair collateral circulation, compared to 0.790 for ASPECTS on NECT and 0.794 for ASPECTS on CTA-SI. Multiple regression analysis revealed that ASPECTS on CECT (≥8) was an independent predictor for fair collateral circulation (odds ratio = 23.00, p < 0.001) and a good prognosis (odds ratio = 17.81, p < 0.001).

CONCLUSION: ASPECTS on CECT is a feasible method for predicting pial collateral flow and overall outcomes in acute ischemic stroke.