Clinical application of a full free-breathing cardiac MRI protocol: feasibility, efficiency, and diagnostic accuracy

Conventional cardiac magnetic resonance imaging (MRI) protocols require multiple breath-holds (BHs) for each sequence, limiting their widespread clinical application. This study sought to develop a full free-breathing (FB) cardiac MRI protocol, including cine imaging, T2 weighted imaging (T2WI), and late gadolinium enhancement (LGE) imaging, and to assess its clinical feasibility.

A total of 94 patients who underwent paired BH- and FB-acquired cine imaging, T2WI, and LGE imaging at 3.0T were prospectively included in the study. The primary analysis included image quality (IQ), qualitative assessment of myocardial edema and enhancement, and quantification of biventricular function and LGE. The overall IQ of all sequences was visually evaluated using a five-point Likert scale. Blood to myocardial contrast (BMC) was measured to quantitatively evaluate the IQ of the short-axis cine sequence. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated to evaluate the IQ of the T2WI and LGE sequences. Function parameters, including left ventricular ejection fraction (LVEF) and right ventricular ejection fraction (RVEF), end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), and left ventricular (LV) mass, were measured and compared.

The proposed full FB protocol achieved identical diagnoses in all patients. The total scan time of all FB sequences was significantly shorter than that of all BH sequences (551±194 vs. 900±148 sec, P<0.001). The overall IQ did not differ significantly between the BH and FB imaging (mean scores 4.19±0.87 vs. 4.10±0.57, P=0.121 for cine images; 3.98±0.84 vs. 4.05±0.90, P=0.468 for T2WI images; and 4.29±0.73 vs. 4.27±0.69, P=0.385 for LGE images; respectively). There were no significant differences in the BMC of the cine images, or the SNR and CNR of the T2WI and LGE images between the BH- and FB-acquired scans. The qualitative assessment of myocardial edema and enhancement was identical across all images. The BH and FB images achieved similar quantitative results for all ventricular function parameters [mean difference (MD) with 95% confidence interval (CI) of 0.83 (-1.48, 3.14) mL, P=0.943 for LVEDV; -1.28 (-3.17, 0.02) mL, P=0.803 for LVESV; 2.47 (0.97, 3.97) mL, P=0.475 for LVSV; 1.05% (0.31%, 1.75%), P=0.528 for LVEF; -0.96 (-3.08, 1.16) g, P=0.821 for LV mass; 0.31 (-1.89, 2.52) mL, P=0.874 for RVEDV; -1.73 (-3.51, 0.05) mL, P=0.740 for RVESV; 1.94 (0.42, 3.47) mL, P=0.542 for RVSV; and 1.13% (0.23%, 2.02%), P=0.459 for RVEF; respectively] and LGE size [MD with 95% CI of -0.43% (-1.42%, 0.55%), P=0.850 for LGE%].

A FB cardiac MRI protocol is feasible in clinical practice, providing shorter acquisition times, comparable IQ, and highly accurate qualitative and quantitative results.