Transthoracic Electromechanical and Mechanical Cardiac Function Imaging in Acute and Chronic Myocardial Infarction Progression

Accurate and precise quantitative functional echocardiographic assessment remains a clinical challenge. Myocardial elastography (ME) and electromechanical wave imaging (EWI) are ultrasound-based elastographic myocardial strain estimation methods that, in tandem, aim to aid in the diagnosis and monitoring of cardiac diseases. Until now, these modalities have been used separately. In this study, we explore their compatibility to localize and detect both the electromechanical and mechanical effects of ischemia. A longitudinal myocardial infarction model was studied in canine subjects (n = 9) for either 4 days (4-d; n = 6) or 5 week (5 wk; n = 3) to investigate the ability of ME and EWI to detect, locate and monitor collateral perfusion recovery from infarction generated via ligation of the left anterior descending coronary artery (LAD). Canine subjects were imaged with both a 300 frames/s and a 2000 frames/s ultrasound radiofrequency acquisition scheme for ME and EWI, respectively, at baseline, daily (n = 9) for 1 week and on a weekly basis (n = 3) for 4 weeks. Both techniques were performed on the acquired data and then segmented by coronary perfusion region. ME displacements were calculated with 1-D normalized cross-correlation in the axial and lateral directions (4.6 mm window, 80% window overlap, 3× lateral re-correlation), and EWI displacements were calculated with the 1-D normalized cross-correlation (5 mm window, 90% window overlap). Strains were computed using a least-squares estimator, resulting in ME cumulative radial strains and EWI interframe axial strains. The means and standard deviations were calculated by day for each technique and observed for the duration of the study. The differences in the cumulative systolic strain and the systolic activation timings between each treatment day and baseline were calculated between each perfusion segment. In the 4-d experiments, the difference between baseline and treatment days was significantly greater in the LAD segment than in the right coronary artery or left circumflex coronary artery (LCX) for both ME (p < 0.005, mixed-effects analysis with Geisser-Greenhouse correction and Tukey's multiple comparisons) and EWI (p < 0.001, mixed-effects analysis with Geisser-Greenhouse correction and Tukey's multiple comparisons). Over the duration of the 4 d study, the LAD showed the most significant electromechanical delay from baseline imaging. The same trend was observed in the 4 wk studies, where the LAD segment showed the most significant difference (ME: -25.8% ± 15.2%; EWI: 58.2 ± 39.5 ms) from baseline for both ME (LCX: -10.0% ± 22.2%, p < 0.01, Kruskal-Wallis test with Dunn's multiple comparisons) and EWI (right coronary artery: 31.4 ± 36.6 ms, p < 0.01, Kruskal-Wallis test with Dunn's multiple comparisons; LCX: 33.6 ± 40.1 ms, p < 0.01, Kruskal-Wallis test with Dunn's multiple comparisons). The LAD segment showed a more significant electromechanical delay between baseline and each treatment day for EWI over the 4 wk studies. This was also observed with an increased EWI systolic activation time delay between baseline and week 4. Both ME cumulative radial strain and EWI systolic activation timing showed the ability to monitor myocardial infarction recovery over both an acute (4 d) and a chronic (4 wk) period. ME also showed substantial collateral reperfusion of the canine myocardium in the recovery of radial strain magnitude, while an electromechanical delay remained throughout both phases of the study over the LAD perfusion segment. Taken together, these findings indicate that even after apparent recovery of cumulative radial strain to baseline values, there is a remaining functional deficit in the form of electromechanical activation. EWI showed the ability to complement ME by providing insight into the electromechanical recovery of the myocardium in addition to the radial strain magnitude.