Distinguishing between occluded and low-flow vessels with PROPELLER DWI sequence: a phantom study

Abstract

PURPOSE: To evaluate the ability of periodically rotated overlapping parallel lines with enhanced reconstruction diffusion-weighted imaging (PROPELLER DWI) to distinguish between vessel occlusion and slow flow. MATERIALS AND METHODS: Using a flow phantom with various velocities (1.37 to 11.1 cm/s), the signal-intensity ratios of the phantom, with the intensity of no flow as baseline, were measured using the following imaging sequences: PROPELLER DWI, spin-echo T1-weighted imaging (SE T1WI), fast-spin-echo T2-weighted imaging (FSE T2WI), two-dimensional phase-contrast imaging (2D PC), and two-dimensional time-of-flight imaging (2D TOF). The b-factor of PROPELLER DWI was varied from 0 to 1000 s/mm(2). The velocity encoding of 2D PC was varied from 2 to 30 cm/s. RESULTS: At the lowest flow velocity (1.37 cm/s) , the signal-intensity ratio was 0.0075 for PROPELLER DWI (b-factor=1000 s/mm(2)), 1.8 for SE T1WI, 0.67 for FSE T2WI, 11 for 2D PC (velocity encoding=2 cm/s), and 1.4 for 2D TOF. The signal-intensity ratio was smallest for PROPELLER DWI, even when the reciprocals of the signal-intensity ratio of 2D PC or 2D TOF were considered. CONCLUSION: The results indicate that PROPELLER DWI provides the best signal intensity-ratio between vessel occlusion and slow flow. Although DWI with single-shot echo-planar imaging (EPI) or multi-shot EPI may have similarly high sensitivity for slow flow, these sequences do not have high spatial resolution or robustness to susceptibility artifacts. PROPELLER DWI would be a better choice for distinguishing between occluded and low-velocity arteries in the skull base or parasellar regions.