A phantom study to determine the theoretical accuracy and precision of radial MRI to measure cross-sectional area differences for the application of coronary endothelial function assessment.

MRI has been used to noninvasively assess coronary endothelial function by measuring the vasoreactivity in response to handgrip exercise. However, the spatial resolution of MRI is limited relative to the expected vasodilation response of healthy coronary arteries (10%-20%), and the sensitivity of MRI to detect such small cross-sectional area differences has yet to be quantitatively examined. Holes of different diameters were drilled in a phantom to simulate a range of physiological responses of coronary arteries to stress. Radial cine MR images with different spatial resolutions were acquired under moving conditions, and different noise levels were simulated. Cross-sectional areas were automatically measured and statistically analyzed to quantify the smallest detectable area difference. Statistical analyses suggest that radial MRI is capable of distinguishing area differences of 0.2 to 0.3 mm2 for high signal-to-noise ratio images, which correspond to a percentage coronary area difference of 3% to 4% for a 3-mm baseline diameter. Furthermore, the smallest detectable area difference was largely independent of the pixel size for the sequence and range of diameters investigated in this study. Radial MRI is capable of reliably detecting small differences in cross-sectional areas that are well within the expected physiological range of stress-induced area changes in of healthy coronary arteries. Magn Reson Med 79:108-120, 2018. © 2017 International Society for Magnetic Resonance in Medicine