Relationship between visible branch arteries distal to the stenosis on magnetic resonance angiography and stroke recurrence in patients with severe middle cerebral artery trunk stenosis: a one-year follow up study

BACKGROUND To evaluate the relationship between the flow signal intensity of branch arteries distal to the stenosis on 3-dimensional (3D) time-of-flight (TOF) magnetic resonance angiography (MRA) and the risk of stroke recurrence in patients with severe middle cerebral artery (MCA) trunk stenosis. METHODS We prospectively recruited 153 patients (mean age 62.9 ± 13.0 years, 106 males) with a first ischemic stroke or transient ischemic attack caused by a severe MCA trunk stenosis (70 % to 99 %) confirmed by 3D TOF MRA and followed them for one year to determine the stroke recurrence. The MCA branch signal intensity distal to the site of stenosis on 3D TOF MRA was classified as either good (grade A) or poor [mild reduction (grade B) or severe reduction (grade C)] according to the extent to which the MCA could be visualized. The patients were divided into groups A (35), B (58), or C (60) based on the MRA grading of the MCA branch signal intensity distal to the site of stenosis. RESULTS Poor MCA branch signal intensity was associated with internal border-zone infarction (p < 0.05). The risk of stroke recurrence in the ipsilateral MCA in the first year was 18.3 %. The 1-year cumulative incidence of recurrence was higher in the patients in group C (30 %) than in groups B (12.1 %) or A (8.6 %) (Log rank, p = 0.007). Multivariate analyses via Cox proportional hazard regression demonstrated that only a grade C classification of the signal intensity of the MCA branches was an independent predictor of stroke recurrence in the ipsilateral MCA (hazard ratio = 3.0, 95 % confidence interval = 1.3-7.4, p = 0.014). CONCLUSIONS This study demonstrated that MCA branch signal intensity as assessed via 3D TOF MRA may be a useful and simple tool to stratify the risk of stroke recurrence in patients with severe MCA trunk stenosis.