Despite the many advances in free breathing 3D coronary MR angiography techniques, it remains a challenging field with little penetration into routine clinical practice. Recently developed breath-hold 2D radial techniques minimize the impact of respiratory patterns on image quality and enable efficient evaluation of the proximal coronary arteries. We compared two such approaches, radial quiescent-interval slice-selective (QISS) and radial 2D balanced steady-state free-precession (bSSFP), in healthy subjects and patients with coronary artery disease (CAD). The study was approved by the Institutional Review Board and used written, informed consent. Imaging was performed using a six-element cardiac phased array coil at 1.5 Tesla (MAGNETOM Avanto, Siemens Healthcare, Erlangen, Germany) and 3 Tesla (MAGNETOM Verio). Healthy subjects (six male, age range 23-35 years) and two patients with CAD (male, ages 39 and 57 years) were imaged. Imaging was performed using prototype multi-shot QISS and T2-prepared (TE = 40ms) 2D bSSFP pulse sequences with a radial k-space trajectory and equidistant azimuthal view angles [1]. Slice thickness was 2.1-mm with in-plane resolution ranging from 1.0 to 1.4-mm before interpolation. Two to five shots were acquired depending on heart rate. For two shots, 10 contiguous or overlapping slices were obtained per breath-hold. Coronary vessel sharpness was measured as the inverse of the distance between the 20th and 80th percentile points of a signal profile through the left anterior descending artery (LAD). A fellowship-trained cardiovascular radiologist scored image quality for the left coronary circulation.

Compared with QISS, T2-prepared bSSFP generally showed greater fluid signal in the pericardial recesses and slightly greater streak artifact. Coronary sharpness was significantly better with QISS (0.66±0.09 mm-1 with QISS vs 0.55±0.12 mm-1 with T2-prepared bSSFP, P<0.01). QISS image quality was better for the left circumflex (mean values of 3.98 vs 3.29, P<0.05), but not significantly different for the left main (3.92 vs 3.54) or LAD (4.00 vs 3.73). Coronary-to-myocardium CNR values were 1.6-fold better for QISS than T2-prepared bSSFP (26.8±10.7 vs 16.6±5.2, P<0.05). The use of increased numbers of shots was sometimes helpful in subjects with fast heart rates and for imaging of the right coronary artery, but reduced multi-slice capability.

In healthy subjects, the coronary arteries showed a similar appearance with the two pulse sequences (Figure 1). However, in one patient with a chronic total occlusion of the LAD, QISS correctly depicted the coronary occlusion while T2-prepared bSSFP suggested the presence of flow beyond the occlusion due to backfilling from collateral vessels (Figure 2).

Both QISS and T2-prepared 2D bSSFP show promise for rapid breath-hold evaluation of the proximal coronary arteries. In healthy subjects, depiction of coronary arteries is similar aside from increased pericardial fluid signal and radial streak artifact with T2-prepared bSSFP. In patients with severe CAD and restricted or absent flow, the appearance of the coronary arteries with QISS and T2-prepared bSSFP can be expected to diverge due to the different mechanisms for flow contrast. Additional clinical study in patients with CAD is needed to determine the respective roles of these two coronary MR angiographic techniques.