Comparison of extracranial artery stenosis and cerebral blood flow, assessed by quantitative magnetic resonance, using digital subtraction angiography as the reference standard

Extracranial arteriosclerosis usually indicates a high risk of ischemic stroke. In the past, a clinical decision following diagnosis was dependent on the percentage of vessel stenosis determined by an invasive technique. We aimed to develop a quantitative magnetic resonance (QMR) technique to evaluate artery structure and cerebral hemodynamics noninvasively. QMR and digital subtraction angiography (DSA) were performed in 67 patients with suspected cerebral vascular disease at our hospital. Accuracy, sensitivity, positive predictive values (PPVs), negative predictive values (NPVs), and Pearson correlation coefficient of QMR were calculated and compared for the detection and measurement of vascular stenoses using DSA as a gold standard. For patients with unilateral artery stenosis, quantitative cerebral blood flow (CBF) was measured by QMR in ipsilateral and contralateral hemispheres. Among 67 subjects (male 54, female 12), 201 stenoses were detected by QMR and DSA. QMR measuring the degree of stenosis and lesion length was in good correlation with the results obtained by DSA (r(2) = 0.845, 0.721, respectively). As for artery stenosis, PPV and NPV of QMR were 89.55% and 95.71%, respectively. As for severe stenosis, sensitivity and specificity of QMR were 82.3% and 86.0% with DSA as a reference. For subjects with unilateral carotid stenosis, CBF in basal ganglia decreased significantly (P < 0.001) compared with the contralateral one in symptomatic and asymptomatic groups. For subjects with moderate stenosis (50–79%), CBF of temporal and basal ganglia was decreased compared with the contralateral ganglia. However, CBF in subjects with severe stenosis or occlusion in the basal ganglia was mildly elevated compared with the contralateral ganglia (P < 0.001). In our study, a good correlation was found between QMR and DSA when measuring artery stenosis and CBF. QMR may become an important method for measuring artery stenosis and cerebral hemodynamics in the future.