Abstract
Purpose
To develop high temporal resolution coronary artery spiral phase velocity mapping sequences and to compare the results obtained with those from FLASH sequences.
Materials and Methods
Velocity curves were obtained in eight left and eight right coronary arteries using breath‐hold interleaved spiral (BH_SP), free‐breathing interleaved spiral (FB_SP), breath‐hold segmented FLASH (BH_FL), and free‐breathing FLASH (FB_FL) sequences. Spatial resolution, temporal resolution, and acquisition durations (cardiac cycles) were as follows—BH_SP: 0.9 mm × 0.9 mm, 30 msec, 20 cycles; FB_SP: 0.9 mm × 0.9 mm, 42 msec, 100 cycles; BH_FL: 0.9 mm × 1.8 mm, 70 msec (effective), 20 cycles; FB_FL: 0.9 mm × 1.8 mm, 30 msec, 480 cycles. Peak systolic, peak diastolic, and mean velocities were compared between sequences.
Results
For left and right arteries, the FB_SP velocity profiles closely followed those from the FB_FL sequence. By comparison, the BH_FL sequence failed to resolve the sharp peaks in the temporal velocity profiles of the right coronary artery, significantly underestimating the peak systolic (88 mm/second vs. 252 mm/second, P < 0.001), peak diastolic (114 mm/second vs. 153 mm/second, P < 0.01), and mean (56 mm/second vs. 93 mm/second, P < 0.001) velocities. For the less mobile left artery, the peak systolic, peak diastolic, and mean velocities were also underestimated by the BH_FL sequence, although this only reached statistical significance for the systolic peak (80 mm/second vs. 135 mm/second, P < 0.01), 142 mm/second vs. 168 mm/second, (P = ns), and 87 mm/second vs. 101 mm/second, (P = ns) respectively.
Conclusion
We have shown that the FB_SP sequence developed agrees well with the FB_FL sequence, while the study duration is reduced by a factor of 10 for the same spatial resolution. By comparison, the BH_FL sequence underestimates flow velocities, particularly in the more mobile right coronary artery. J. Magn. Reson. Imaging 2004;19:40–49. © 2003 Wiley‐Liss, Inc.