Noncontrast 3D Steady-State Free-Precession Magnetic Resonance Angiography of the Whole Chest Using Nonselective Radiofrequency Excitation over a Large Field of View: Comparison With Single-Phase 3D Contrast-Enhanced Magnetic Resonance Angiography

Abstract

To evaluate the feasibility of three-dimensional (3D) steady-state free-precession (SSFP) magnetic resonance angiography (MRA) using nonselective radiofrequency excitation in the assessment of cardiac morphology, thoracic aorta, main pulmonary, and proximal coronary arteries. Thirty consecutive patients (19 males; 11 females; age range, 20-74) with various cardiac and thoracic vascular diseases underwent free-breathing respiratory navigator-gated electrocardiogram-triggered noncontrast SSFP MRA and conventional high-resolution 3D contrast-enhanced MRA (CE-MRA) of the thorax at 1.5 T. Two readers evaluated both datasets for findings, vascular delineation and sharpness (from 0, not visualized to 3, excellent definition), artifacts, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) in 14 vascular segments including aorta, supra-aortic, pulmonary, and coronary arteries, and in cardiac chambers. Statistical analysis was performed using Wilcoxon test for vessel delineation, and [kappa] coefficient for interobserver variability. 3D SSFP and CE-MRA were successfully performed in all patients. Scan time for SSFP MRA ranged from 5 to 10 minutes (mean +/- standard deviation, 7 +/- 2 minutes). On SSFP MRA, readers 1 and 2 graded 233 (97.1%) and 234 (97.5%) coronary arterial segments and cardiac chambers, and 275 (91.7%) and 278 (92.7%) noncoronary arterial segments with diagnostic definition (grades 2 and 3) (k = 0.86). On conventional CE-MRA, readers 1 and 2 graded 10 (4.2%) and 12 (5%) coronary arterial segments and cardiac chambers, and 272 (90.7%) and 270 (90%) noncoronary arterial segments with diagnostic definition (grades 2 and 3) (k = 0.89). Segmental visibility was higher for aortic root, pulmonary trunk, proximal coronary arteries, and heart chambers (P < 0.001), and lower for supra-aortic arteries (P < 0.001) on SSFP MRA for each reader. SNR and CNR values were higher for aortic root and aorta on SSFP MRA (P < 0.001 for both). No significant difference existed between SNR and CNR values for the other vascular segments and cardiac chambers on SSFP and CE-MRA (P > 0.05 for all). The 2 readers demonstrated vascular stenosis and dilatation/aneurysm in 7 and 35 segments on both datasets, respectively. Noncontrast 3D SSFP MRA with nonselective radiofrequency excitation provides high image quality and sufficient SNR and CNR for confident assessment of cardiac and thoracic vascular diseases including congenital heart diseases. Our results suggest that noncontrast SSFP MRA outperforms CE-MRA in visualization of cardiac chambers, proximal coronary arteries, pulmonary trunk, and aortic root.