Three-dimensional feature tracking cardiac magnetic resonance in hypertrophic cardiomyopathy: comparison with two-dimensional algorithm

Introduction Myocardial deformation is a highly sensitive marker for early-stage myocardial dysfunction with an independent predictive value in hypertrophic cardiomyopathy (HCM). Although speckle-tracking echocardiography (STE) is a widely available and powerful diagnostic tool to assess myocardial dysfunction even in the presence of preserved EF, an inadequate acoustic window can impede its utilisation. In these cases, cardiac magnetic resonance (CMR) imaging can be used as an alternative. While the conventional approach to myocardial strain measurement on CMR involves using 2-dimensional (2D) feature-tracking imaging (FTI), new algorithms have been created that allow three-dimensional (3D) tracking. To date, 2D and 3D FTI have not been compared in patients with HCM. The aim of this study is to determine whether 2D and 3D FTI are equivalent in patients with HCM and preserved LVEF. Methods Consecutive patients with HCM and LVEF>55% were included in the study. CMR FTI parameters included left ventricular (LV) global longitudinal (GLS), circumferential (GCS) and radial (GRS) strain, which were calculated using both 2D and 3D FTI CMR derived from multi-slice 2D cine image stacks in the short LV short axis plane. FTI was calculated using Circle Cvi42 software. Results Sixty HCM patients (40 male, mean age 50±16 years) were included (56.7% septal HCM, 25% apical HCM). FTI analysis was feasible in all individuals. Mean LV end-diastolic volumes was 80 ± 20 ml/m2, LV end-systolic volume 30 ±13 ml/m2 and LV ejection fraction 64.4% ± 9. Mean maximum wall thickness (MWT) was 17 ± 4.4mm and LV mass index 76 ± 24.4g/m2. Seventy-two percent of patients exhibited focal fibrosis in 4.2±2.4 segments. Mean GLS 2D and 3D values were -12.6% ± 2.5 and -9.9±3.5, mean GCS 2D and 3D were-16.6% ± 2.4 and -16.3% ± 6.6 and mean GRS 2D and 3D values were 27.9% ± 6 and 38.7% ± 13.9, respectively. As previously reported for healthy individuals, myocardial strain evaluated using 3D FT-CMR was consistently lower compared to 2D FT-CMR. The highest level of agreement was noted between the 2D and 3D values of GCS, with a mean difference of -0.3% (95% CI -1.8115 to 1.1971, p=0.7). Furthermore, among the 3D FTI measures, only GCS 3D showed a correlation with LVESVi (R=0.49) and LV mass index (R=0.44). Regarding GLS, the mean difference was -2.7% (95% CI -10.7924 to -7.7996, p<0.0001). GRS exhibited low agreement between 2D and 3D measurements, with a mean difference of -10.8% (95% CI -39.4193 to -28.8631, p-< 0.0001). Conclusions Three-dimensional CMR FTI is feasible in HCM patients. While 3D FTI confirms the results of 3D STE, which have shown that absolute values tend to be lower than those obtained using the 2D approach, 2D and 3D FTI by CMR are not fully interchangeable. This observation needs to be taken into consideration when applying the new 3D FTI methods in clinical practice and research. Agreement between 2D and 3D FTI.