[Multi-parameter arterial spin labeling versus dynamic magnetic-sensitive enhanced cerebral perfusion imaging for diagnosis of transient cerebral ischemic attack]

Abstract

Objective: To evaluate the clinical value of three-dimensional pseudo-continuous arterial spin labeling (3D pCASL) perfusion magnetic resonance imaging (MRI) and dynamic susceptibility contrast (DSC) enhanced perfusion MRI in the diagnosis of transient ischemic attack (TIA).

Methods: Thirty-nine consecutive patients with suspected TIA underwent multi-modal MRI scans including DSC, magnetic resonance angiography (MRA), diffusion-weighted imaging (DWI) and 3D pCASL (post-labeling delay, PLD=1.5 s and 2.5 s) within 24 h of symptom onset. Cerebral blood flow (CBF) from ASL and the time to the maximum of tissue residual function (Tmax) map from DSC were calculated using AW workstation. DWI and MRA were applied to detect acute cerebral infarction and intracranial artery stenosis. Two neuroradilogists who were blinded to the patients' clinical data assessed the presence of perfusion deficit, ischemic lesion and the lesion sites both from 1.5 s, 2.5 s PLD ASL-CBF and DSC-Tmax independently, and then graded them. The differences in the ranking grades between 1.5 s, 2.5 s PLD ASL and DSC were analyzed, and the frequency of lesion detection was compared between ASL-CBF, Tmax and MRA combining DWI method.

Results: No significant differences was found in hypoperfusion grades detected by 3D pCASL (including PLD1.5 s and 2.5 s) CBF and Tmax maps, while significant differences were detected between 1.5 s PLD ASL-CBF and MRA combining DWI method; ASL with PLD 1.5 s CBF detected ischemic lesions and lesion site significantly more frequently than MRA combining DWI method.

Conclusion: s Three dimensional pCASL is a non-invasive perfusion method free of radiation exposure, and short PLD ASL is more sensitive than long PLD ASL for detecting ischemic lesions and lesion sites.