Introduction
Coronary MR Angiography is a valuable tool for non-invasive assessment of coronary arteries. Presently, contrast-enhanced, fat-saturated, ECG-triggered and navigator-gated 3D spoiled gradient-echo sequence is employed for whole-heart Coronary MRA at 3 T[1]. However, large static field variations at 3 T frequently result in non-uniform fat-suppression over the field-of-view (FOV), obscuring the delineation of coronary arteries. Multi-echo Dixon approaches utilizing iterative decomposition have been shown to provide robust fat-water separation even in the presence of large field inhomogeneities. In this study, an ECG-triggered navigator-gated 3D spoiled gradient-echo multi-interleaved multi-echo (GRE-MEMI) pulse sequence is introduced which utilizes VARPRO[2] fat-water separation to achieve reliable fat-suppression and provides enhanced visualization of coronary arteries.
Methods
A 3D GRE-MEMI sequence (Fig. 1) was implemented on a 3 T whole-body MR scanner (MAGNETOM Trio, Siemens AG) with support for navigator-gating and ECG-triggering. Water-only and fat-only images were reconstructed using VARPRO. Four healthy volunteers were imaged pre- and during contrast agent administration targeting right coronary artery (RCA). Typical imaging parameters for pre-contrast GRE-MEMI scan are listed in Table 1. Additionally, a conventional single-echo fat-saturated GRE scan was acquired for comparison. Thereafter, 0.2 mmol/kg Gd-DTPA (Magnevist®, Bayer Healthcare) was slowly injected at a rate of 0.3 ml/s followed by 20 ml of saline solution injected at the same rate. GRE-MEMI acquisition with inversion preparation (TI = 300 ms) was started 30 s after injection.
Table 1. Typical imaging parameters for conventional 3D fat-saturated GRE and 3D GRE-MEMI measurements.
Figure 1. Pule Sequence Diagram for 3D ECG-triggered, navigator gated, spoiled gradient echo
sequence with multi-echo multi-interleave readout (GRE-MEMI). Multiple echoes are used during reconstruction by VARPRO for iterative water-fat
decomposition. No fat-saturation prepulse is applied separately. Multi-interleaved
scheme achieves shorter echo time increments between multiple echoes, which improves
the fat-water separation.
Results
Targeted RCA images were successfully acquired in all volunteers with effective fat-water separation. The average total imaging time was 8.93 ± 1.2 min with navigator efficiency of 33.8 ± 4.6%. Fig. 2 shows pre-contrast coronary artery images from a healthy volunteer. Conventional fat-saturation yields suboptimal fat-suppression whereas robust fat-suppression is evident in water-only images which provide clear depiction of coronary artery. Fig. 3 illustrates enhanced contrast-to-noise with the use of contrast agent.
Figure 2. Conventional fat saturation (chemical selective saturation) image (a) and fat-water
separated images (b, c) from a targeted right coronary artery (RCA) measurement at
3 T in a healthy volunteer without any contrast agent administration. Conventional fat saturation yields suboptimal results in some areas (a - red arrows),
however, robust fat suppression is achieved over the entire FOV using the proposed
technique. Moreover, the use of multiple echoes increases the signal-to-noise ratio
(SNR). Water-only (d) and fat-only (e) pre-contrast images from another health subject
demonstrate excellent fat suppression and clearly depict RCA (red arrows).
Figure 3. Water-only (a) and fat-only (b) images acquired from a healthy subject during slow
infusion of contrast media. Note that water and fat signals are effectively separated and the RCA (a - red arrows)
is sharply delineated. Compared to pre-contrast GRE measurements, use of contrast
agent increases contrast-to-noise ratio between blood and background tissues.
Conclusion
3D GRE-MEMI sequence was successfully utilized for targeted fat-water separated coronary artery imaging in healthy volunteers. VARPRO fat-water separation provides reliable fat-suppression at 3 T and improves the delineation of coronary arteries. Moreover, without the use of a fat-saturation prepulse, readout duration within a heartbeat can be extended to cover the entire quiescent period without any degradation in fat-suppression. Multi-echo acquisition results in increased acquisition time, however, the resulting water-only image provides the benefit of increased SNR due to intrinsic averaging effect of fat-water separation. Further improvement in acquisition speed using higher parallel imaging factors is required to achieve 3D whole-heart coverage.
