The Combination of Feature Tracking and Late Gadolinium Enhancement for Identification Between Hypertrophic Cardiomyopathy and Hypertensive Heart Disease

BACKGROUND: The differentiation between hypertrophic cardiomyopathy (HCM) and hypertensive heart disease (HHD) is challenging due to similar myocardial hypertrophic phenotype. The purpose of this study is to evaluate the feasibility of cardiovascular magnetic resonance feature tracking (CMR-FT) and late gadolinium enhancement (LGE) to distinguish between HCM and HHD and the potential relationship between myocardial strain and cardiac functional parameters. METHODS: One hundred and seventy subjects (57 HCM, 45 HHD, and 68 controls) underwent 3.0 T CMR, including steady-state free precession cines and LGE images. Global and segmental (basal, mid, and apical) analyses of myocardial radial, circumferential, longitudinal strain, and left ventricular (LV) torsion, as well as global and 16 segments of LGE were assessed. The multivariate analysis was used to predict the diagnostic ability by combining comprehensive myocardial strain parameters and LGE. RESULTS: Global radial strain (GRS), global circumferential strain (GCS), and LV torsion were significantly higher in the HCM group than in the HHD group (GRS, 21.18 ± 7.52 vs. 14.56 ± 7.46%; GCS, −13.34 ± 3.52 vs. −10.11 ± 4.13%; torsion, 1.79 ± 0.69 vs. 1.23 ± 0.65 deg/cm, all P < 0.001). A similar trend was also seen in the corresponding strain rate. As for segmental strain analysis, basal radial strain (BRS), basal circumferential strain (BCS), basal longitudinal strain (BLS), mid-radial strain (MRS), and mid-circumferential strain (MCS) were higher in the HCM group than in the HHD group (all P < 0.001). The receiver operating characteristic (ROC) results showed that the area under the curve (AUC) of LGE in the mid-interventricular septum (mIVS) was the highest among global and segmental LGE analyses. On the multivariate regression analysis, a combined model of LGE (mIVS) with GRS obtained the highest AUC value, which was 0.835 with 88.89% sensitivity and 70.18% specificity, respectively. In addition, for patients with HCM, GRS, GCS, and global longitudinal strain had correlations with LV ejection fraction (LVEF), maximum interventricular septum thickness (IVST max), and left ventricular mass index (LVMi). Torsion was mildly associated with LVEF. CONCLUSION: CMR-FT-derived myocardial strain and torsion provided valuable methods for evaluation of HCM and HHD. In addition, the combination of GRS and LGE (mIVS) achieved the highest diagnostic value.