Purpose: Correlation of arterial stenosis with cortical and subcortical cerebral blood flow (CBF) in the middle cerebral artery (MCA) territory.
Methods: 126 patients with acute cerebrovascular symptoms from March to June 2015 underwent MRI and MR Angiography (MRA) in a University hospital using a 3.0 Tesla scanner. Sequences included T1W, T2W, FLAIR, DWI, MRA, Pseudocontinuous Arterial Spin Labeling (pcASL, post-labeling delay 1.525 s). 13 patients (corrupted pcASL data) were excluded, with 113 patients (mean age: 67.74±14.19) evaluated (61 acute ischemic stroke, 52 patients transient ischemic attack). Institutionally developed software was used to determine CBF. MCA stenosis was graded into 4 categories by a neuroradiologist: 0 (no stenosis), 1 (mild <50%), 2 (moderate 50-70%) and 3 (severe >70%). Mean and standard deviation of MCA categories (leptomeningeal and perforating) CBF and corresponding degree of MCA stenosis were measured. Spearman correlation coefficients between CBF of cortical and subcortical regions and degree of MCA stenoses were calculated using SPSS (version 23.0).
Results: The table showed the descriptive statistics. There was significant correlation between CBF of cortical region of MCA vascular territory and degree of stenosis of MCA in both left (rs= -0.296, p=0.001) and right (rs= -0.306, p=0.001) side. In the contrary, there was no correlation between subcortical CBF of MCA vascular territory and degree of stenosis of MCA in both sides.
Conclusion: pcASL is a feasible non-invasive method to measure CBF in clinical setting. In MCA territory, the cortical blood flow correlated (fairly) with large vessel stenosis but not subcortical flow. We conclude that cortical CBF correlated with large artery stenosis, though being attenuated by collateral blood supply. No such relationship in subcortical CBF might be due to differential grey and white matter CBF flow, variable MCA stenotic location, and perforators originating from other territories.