Arterial spin labeling in patients with chronic cerebral artery steno-occlusive disease: Correlation with 15O-PET

Abstract

Heterogeneity of arterial transit time due to cerebral artery steno-occlusive lesions hampers accurate regional cerebral blood flow measurement by arterial spin labeling (ASL). To assess the feasibility of regional cerebral blood flow measurement by ASL with multiple-delay time sampling in patients with steno-occlusive diseases by comparing with positron emission tomography (PET), and to determine whether regional arterial transit time measured by this ASL technique is correlated with regional mean transit time, a PET index of perfusion pressure. Sixteen patients with steno-occlusive diseases received both ASL and (15)O-PET. The mean regional cerebral blood flow measured by ASL and PET, regional arterial transit time by ASL, and regional mean transit time by PET were obtained by a region-of-interest analysis. Correlation between regional cerebral blood flow by ASL and that by PET, and correlation between regional arterial transit time by ASL and regional mean transit time by PET were tested using Pearson's correlation coefficient for both absolute and relative values. A multivariate regression analysis was performed to test whether regional arterial transit time by ASL was a significant contributor in modeling regional mean transit time by PET after controlling the effect of regional cerebral blood flow by ASL. A significant positive correlation was found between regional cerebral blood flow by ASL and that by PET for both absolute (r = 0.520, P < 0.0001) and relative (r = 0.691, P < 0.0001) values. A significant positive correlation was found between regional arterial transit time by ASL and regional mean transit time by PET both for absolute (r = 0.369, P = 0.0002) and relative (r = 0.443, P < 0.0001) values. The regression analysis revealed that regional arterial transit time by ASL was a significant contributor in modeling regional mean transit time by PET after controlling regional cerebral blood flow by ASL (P = 0.0011). The feasibility of regional cerebral blood flow measurement using ASL with multiple-delay time sampling was confirmed in patients with cerebral artery steno-occlusive diseases. Moreover, it was suggested that mapping of regional arterial transit time has the potential to detect hemodynamic impairment.