Noninvasive Fractional Flow on MRA and Recurrent Stroke: SPS3 Trial

Background and Purpose: Noninvasive fractional flow reserve (FFR) on time-of-flight magnetic resonance angiography (TOF-MRA) may be used to identify high-risk intracranial lesions. We tested whether FFR was associated with vascular territory of the qualifying lacunar stroke in participants of the Secondary Prevention of Small Subcortical Strokes (SPS3) trial and the utility of FFR for predicting recurrent stroke during the trial.Methods: SPS3 was a randomized trial investigating optimal blood pressure target and antiplatelet regimen in patients with recent, symptomatic, MRI-confirmed lacunar stroke patients TOF-MRA proximate to study entry was adequate and available for 2169 of 3020 study patients. Signal intensity (SI) was measured in the background, and proximal and distal aspects of 7 intracranial arteries (internal carotid, middle cerebral, basilar, and vertebral). Adjusted FFR was then calculated in each artery: FFR = [distal SI - background SI] / [proximal SI - background SI] and divided into quartiles by artery. Associations between the vascular territory of the qualifying infarct and the FFR quartile of the relevant artery were investigated using contingency tables and chi-square tests. Risks for recurrent stroke associated with FFR quartiles were evaluated using Cox Proportional Hazards models (model adjusted for assigned treatment groups).Results: Mean age of the 2169 patients included was 63 yr with 63% male; hypertension, diabetes, and prior lacunar stroke were present in 75%, 36%, and 10% respectively. Median FFRs varied by artery with the lowest in the basilar (0.793) and highest in the middle cerebral arteries (left 1.154; right 1.176). A recurrent stroke occurred in 195 patients during a mean follow-up of 3.5 years (annualized rate 2.5% per patient-year).No significant association was found between the FFR tertiles and the vascular territory of the qualifying infarct. Quartiles of adjusted FFR in any of the 7 arteries were not found to be predictive of recurrent stroke.Conclusion: In this large well-characterized cohort of lacunar stroke patients, FFR was not associated with the location of the qualifying subcortical infarct and did not predict the risk of recurrent stroke.