Collateral circulation supports brain tissues to maintain blood perfusion in cerebral ischemic stroke and are of great benefit for a better outcome. A non-invasive approach relative to currently widely used digital subtraction angiography (DSA) is needed. ASL (arterial spin labeling) is a novel perfusion method without contrast agent injection and features both temporal and cerebral blood flow(CBF) information. The investigators applied multiple post labeled delay(PLD) time to pseudo-continuous Arterial Spin Labeling (3D pCASL) MRI and subtraction images were obtained to evaluate the collateral robustness and quantitatively assess the collateral perfusion in patients with unilateral middle cerebral artery atherosclerotic stenosis and the ability to predict future stroke recurrence.

Patients with unilateral middle cerebral artery (MCA) moderate to severe stenosis were consecutively enrolled. MRI protocols including diffusion weighted image, magnetic resonance angiography , 3D pCASL with two PLD of 1.5s and 2.5s were performed.

Cerebral blood flow(CBF) map of 3D pCASL with PLD 1.5s and 2.5s were postprocessed on workstation. Subtraction images were obtained by CBF 2.5s subtracted CBF 1.5s. Slices involving MCA downstream territory were equally separated as upper slices,middle slices and inferior slices corresponding to CIS system referring to previous study of Al-Ali F. Slices with residual signal areas more than 1/3 MCA territory involved would count 1, or would be 0. Total ASL collateral score(ASLCS) was calculated from 0-3. Then ,favorable collaterals were graded as total ASLCS 2-3 and poor collaterals were graded as total ASLCS 0-1.

Early-arriving flow perfusion proportion was defined as [CBF 1.5s at lesion side/CBF 2.5s of the contralateral side]×100%. Late-arriving perfusion proportion was defined as [(CBF 2.5s minus CBF 1.5s) at lesion side minus (CBF 2.5s minus CBF 1.5s) at normal side]/CBF 2.5s at the contralateral side ×100%. Antegrade scales and collateral grades were evaluated in patients with conventional angiography. Spearman correlation was analyzed between early-arriving and late-arriving flow and angiographic antegrade and collateral scales.

Baseline characteristics of patients were recorded including age, gender, hypertension,hypercholesterolemia,diabetes mellitus,smoking,obesity,qualifying stroke event, admission National Institute of Health stroke scale (NIHSS), diffusion-weighted image-ASPECTS.

One year stroke event recurrence and three months modified Rankin Scale (mRS) were followed up by neurologist by phone call or face to face.

Multi-variants Logistic Regression is performed to exam whether ASL collateral score and collateral perfusion quantification is significantly correlated with future ischemic event and functional outcome.

Inclusion Criteria:

• Ischemic stroke or TIA in anterior circulation within 90 days
• MCA atherosclerotic moderate to severe stenosis(50%-99%) confirmed by conventional angiography or magnetic resonance angiography
• Age >18
• 2 or more atherosclerotic risk factors including hypertension, hypercholesterolemia, diabetes mellitus, cigarette smoking, and obesity
• Medical treatment
• Not receiving stent therapy
• Ability to comply with all studies

Exclusion Criteria:

• Multiple intracranial arteries stenosis (> 50%) or occlusion
• Less than 2 atherosclerotic risk factors
• Pregnancy and other contraindication to MRI scan
• Informed consent not obtained

Symptomatic and asymptomatic patients with MCA atherosclerotic moderate to severe stenosis(500%-99%) confirmed by conventional angiography or magnetic resonance angiography.