Free-Breathing 3D Whole Heart and Aorta Cine MRI Without Contrast Agent-Comparison to Clinical Standard.

BACKGROUND

The demand for cardiac MRI is increasing with the growing burden of cardiovascular disease. However, conventional protocols require sequential acquisitions for multi-breath-hold 2D cine and 3D MR angiography (MRA), which is time-consuming. In addition, breath-hold 2D cine can be challenging for patients with limited breath-hold capacity.

PURPOSE: To further develop and evaluate a free-breathing 3D whole-heart cine MRI technique for simultaneous assessment of cardiac function and aortic anatomy in a single non-contrast acquisition at 1.5 T.

STUDY TYPE: Prospective.

SUBJECTS: Twenty-four healthy volunteers (mean age 31.8 ± 9.9 years, 50% female).

FIELD STRENGTH/SEQUENCES: 1.5 T; A cartesian spiral (CASPR) bSSFP 3D cine, 2D cine and 3D mDixon MRA.

ASSESSMENT: Acquisition and reconstruction times were assessed for 3D cine at 2.5 and 2.0 mm. LV mass, LVEF, and RVEF were assessed by two observers (5 and 32 years of experience) and compared with 2D cine. Aortic root and ascending aortic areas were compared with 3D MRA. Image quality was evaluated using blood pool-to-myocardium contrast ratio and endocardial/epicardial edge sharpness. Qualitative image preference was assessed by three observers (5, 32, and 33 years of experience).

STATISTICAL TESTS: One-way ANOVA with Tukey post hoc tests and Bland-Altman analysis with paired t-tests were used. Intraclass correlation coefficient (ICC) assessed inter- and intra-observer agreement. p < 0.05 was considered significant.

RESULTS

Acquisition time was 5 min (2.5 mm) and 7 min (2 mm), versus 11 min for 2D cine and 8 min for 3D MRA. Reconstruction time was approximately 5 min. LV mass showed no differences versus 2D cine. LVEF showed small but significant bias (2.5 mm: 1.38%, p = 0.005; 2.0 mm: 1.39%, p < 0.001). RVEF showed no significant differences. Ascending aorta areas showed significant differences (0.28mm2 and 0.25mm2, p < 0.001), while aortic root areas showed no difference versus 3D MRA. Reproducibility was at least moderate (ICC 0.75-0.998). 3D cine showed lower contrast and edge sharpness than 2D cine (p < 0.001). Observers preferred 2.0 mm (ICC = 0.64).

DATA

CONCLUSION

Free-breathing 3D cine MRI enables operator-independent, time-efficient, and accurate assessment of cardiac function and aortic anatomy in healthy volunteers in a single non-contrast acquisition at 1.5 T, compared to conventional 2D cine and 3D MRA.

EVIDENCE LEVEL: 2.

TECHNICAL EFFICACY: Stage 1.