ABSTRACT
PURPOSE: Comparison of a free breathing steady-state free precession (SSFP), a spoiled gradient-echo (GRE) and a turbo spin-echo sequence (TSE) for imaging of the coronary arteries (MRCA) in healthy volunteers. MATERIALS AND METHODS: Twenty-two healthy volunteers were imaged with a standard clinical scanner (1.5 T, Intera, Philips), with the right coronary system imaged in 11 and the left coronary system in the other 11 volunteers. Images were obtained with a 3D-SSFP (balanced TFE, TR 6.2 ms, TE 3.1 ms, alpha 65 degrees ), a 3D-GRE (TFE, TR 7.2 ms, TE 2.2 ms, alpha 30 degrees ) and a 2D-TSE (Dual-IR, TR 2RR, TE 25 ms) sequence. The in plane resolution was 0.7 x 0.8 mm for both the SSFP and GRE sequence with an effective slice thickness of 1.5 mm. For the TSE sequence, an in-plane resolution of 0.7 x 0.9 mm and a slice thickness of 3.0 mm were used. All investigations were performed using prospective navigator gating and slice-following technique. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) for the blood pool to myocardium and blood pool to epicardial fat were calculated. Image quality and measurement artifacts were assessed for all sequences by 5 independent investigators using a 4- and 5-point grading scale. RESULTS: CNR was significantly higher for the GRE sequence compared with the SSFP sequence and TSE sequence (mean 20.8 +/- 4.8 vs. 14.6 +/- 5.0 and 10.1 +/- 3.7 for blood pool to myocardium; mean 27.5 +/- 6.3 vs. 16.4 +/- 5.4 and 18.1 +/- 5.7 for blood pool to fat). The SNR revealed no significant differences between the SSFP and GRE sequences. The SSFP and the TSE sequences showed significantly more artefacts than the spoiled GRE sequence. Image quality was graded slightly higher for the GRE than for the SSFP sequence for the right coronary system, while there was no substantial difference in the left coronary system (median 2.1 +/- 0.6 and 2.5 +/- 0.6 vs. 2.5 +/- 0.8 and 2.6 +/- 0.7 for the right and left coronary system). In comparison, image quality was lower with the TSE sequence (median 2.9 +/- 0.5 for the right coronary system with p < 0.05 vs. GRE sequence and 3.0 +/- 0.3 for the left coronary system). CONCLUSION: For the scan parameters chosen in this study, the GRE-sequence represents the most robust technique for imaging of the coronary arteries. Currently, the TSE sequence is no alternative.
