Cardiac MRI T2 mapping with heart phase acquisition and improved motion-sensitized driven equilibrium blood suppression

Abstract

Abstract Background Cardiovascular magnetic resonance (CMR) imaging relies on T2 mapping to assess myocardial tissue properties. However, T2 maps are subject to motion artifacts and signal interference, prompting exploration of techniques such as gradient-spin-echo (GraSE) sequences, heart phase acquisition, and dark blood (DB) imaging to improve image quality and reliability. Purpose This prospective study evaluates the T2 relaxation time (T2) variability and the image quality of T2 maps obtained in end-systole and end-diastole with and without fat saturation (FS), using GraSE sequence in CMR imaging. Furthermore, improved motion-sensitized driven-equilibrium (iMSDE) was compared with double inversion recovery (DIR) as an alternative DB technique. Materials and Methods Five variants of the DB GraSE sequence were developed and performed on a 1.5 Tesla MRI scanner. Forty-four healthy volunteers prospectively underwent the following sequences: GraSE in end-diastole (GraSE-ED), GraSE-ED with FS (GraSE-ED-FS), GraSE in end-systole (GraSE-ES), GraSE-ES with FS (GraSE-ES-FS), and commercial GraSE-ED (GraSE-CO). Four GraSE variants utilized iMSDE technique, while GraSE-CO used DIR for blood suppression. T2, image quality, and visual artifacts were measured. Results Ten volunteers were excluded due to image artifacts or missing datasets. Among the remaining 34 participants, the mean global T2 was measured. No significant differences were seen among all variants (P > .05 for all comparisons). Intra- and inter-reader agreement of global T2 values for all GraSE sequence were very good (r > 0.8 for both). Image quality was rated moderate or good for all variants of GraSE sequences. A lower incidence of artifacts was observed in end-systolic compared to end-diastolic imaging. Conclusion All variants of GraSE sequence are highly reproducible and myocardial T2 values did not significantly differ with heart phase. iMSDE is feasible as an alternative DB technique for T2 mapping enabling acquisition in systole, which shows a lower incidence of artifacts compared to diastole.