Predicting Stroke Risk in Symptomatic Intracranial Atherosclerotic Disease

Introduction: Intracranial atherosclerotic disease (ICAD) is a leading cause of stroke and risk factor for recurrence. Regional hypoperfusion was previously shown to predict subsequent stroke risk in posterior circulation ICAD patients enrolled in the prospective observational VERiTAS and MYRIAD studies, using quantitative MRA (QMRA) to assess large vessel flow. We examined whether a similar approach to regional flow assessment predicted stroke risk in anterior circulation ICAD.

Methods: MYRIAD enrolled patients with symptomatic proximal intracranial artery stenosis of 50-99%; primary outcome was ischemic stroke in the stenotic artery territory within one year. Baseline flow was measured in the major intracranial arteries by QMRA. Using an algorithm assessing distal flow and collateral capacity based on age-normalized middle cerebral artery (MCA) and hemispheric flows, patients were classified as low or normal flow status. The optimal algorithm for stroke risk prediction was determined after testing thresholds for flow status categorization.

Results: 7 of 73 (9.6%) enrolled subjects with symptomatic anterior circulation ICAD had recurrent stroke. Z score thresholds for age normalized flow were evaluated, and optimal thresholds of -1 and -0.75 were identified for the MCA and hemispheric flow, respectively. Based on these thresholds, 24 (33%) patients were categorized as low flow; recurrent stroke occurred in 21% of low flow vs 4% of normal flow patients (age adjusted OR 7.2 , 95% CI 1.2-43.2). In the full cohort of 99 subjects with anterior and posterior circulation ICAD, 11 (11.1%) had recurrent stroke, with low flow patients having a similarly high recurrent stroke risk (21.4% vs 7%, age adjusted OR 3.8, 95% CI 1.02-14.2).

Conclusions: QMRA regional flow measurement can assess distal flow status and predict recurrent clinical stroke in both the anterior and posterior ciruclation. The ability to identify high-risk patients has important implications for selection of patients for endovascular and surgical interventions.