Quantification of Exercised-Induced Myocardial Deformation with Exercise CMR

BACKGROUND: Physiological stress may reveal myocardial deformations not evident at rest. Single-beat myocardial tagging in exercise cardiovascular magnetic resonance (Ex-CMR) enables assessment of myocardial deformation at both rest and during exercise. In this study, we developed the myocardial torsional reserve (MTR), a quantitative metric of exercise-induced rotational deformation, and evaluated its potential as a marker of abnormal myocardial mechanics in patients with overt or exercise-induced heart failure with preserved ejection fraction (HFpEF). METHODS: Free-breathing balanced steady-state free precession single-beat tagging images were acquired at rest and post-exercise (stress). Apical and basal rotation angles were measured at end-systole in short-axis planes relative to an equilibrium configuration. Torsion was defined as the difference between basal and apical rotation normalized by the distance between slices. MTR was defined as the ratio of stress to rest torsion. We retrospectively analyzed data from 94 participants of a prospective multicenter study who underwent 3T Ex-CMR using a supine ergometer. All patients underwent invasive exercise testing before Ex-CMR. Established invasive thresholds defined patient cohorts. Intra- and inter-observer repeatability of image analysis was evaluated using the intraclass correlation coefficient (ICC). Group differences in MTR were analyzed using ANOVA with post hoc comparisons. RESULTS: After excluding 27 participants due to arrhythmias, rest and stress slice misalignment, or triggering issues, 67 participants were included: 24 healthy controls, 17 with non-cardiac dyspnea, 15 with exercise-induced HFpEF, and 11 with overt HFpEF. MTR measurement was successful in 76% of cases without arrhythmias and demonstrated good to excellent repeatability: inter-observer ICC=0.90 0.70, 0.97 and intra-observer ICC=0.95 0.83, 0.98. MTR was equally reduced in exercise-induced HFpEF (1.1±0.1) and overt HFpEF (1.2±0.2) compared to healthy controls (1.7±0.4, p<0.01 for each comparison) and to patients with non-cardiac dyspnea (1.6±0.4; p<0.001 and p<0.01, respectively). MTR was comparable between healthy controls and non-cardiac dyspnea patients (p=0.978). CONCLUSION: Ex-CMR-derived MTR enables quantification of exercise-induced rotational deformation and mechanical reserve through single-beat myocardial tagging. In a pilot clinical study, MTR values distinguished patients with exercise-induced or overt HFpEF from individuals with non-cardiac dyspnea or healthy controls, supporting its potential utility as a marker of abnormal myocardial response to exercise.