Document resource
Background Hypertrophic cardiomyopathy (HCM) is a genetic disease with a prevalence of 1 in 500, characterised by left ventricle hypertrophy (LVH). Identifying subclinical HCM, defined as genotype positive phenotype negative (G+LVH-), is clinically important for patient risk stratification and management. We previously demonstrated the role of cardiac diffusion tensor imaging (cDTI) cardiac magnetic resonance (CMR), in identifying microstructural changes in HCM (G+LVH-) patients. Myocardial strain with CMR feature-tracking (FT) has prognostic value in overt HCM, defined as HCM (G+LVH+). We sought to determine whether cDTI is a more sensitive marker than 3D-FT strain in the risk stratification of HCM (G+LVH-) patients.Methods This multi-centre collaboration included 127 subjects: 23 healthy volunteers (HV), 67 HCM (G+LVH-) and 37 HCM (G+LVH+) participants. All patients underwent 3T CMR imaging with the following protocol: cine imaging; second order motion compensated, free-breathing spin echo cDTI (b-values of 100 s/mm2 (3 DW directions, 12 repetitions), and 450 s/mm2 (30 DW directions, 6 repetitions); pre and post contrast T1 mapping and late gadolinium enhancement (LGE). Global cDTI analysis was undertaken to derive: mean diffusivity (MD) (a measure of the magnitude of diffusion, high values thought to reflect interstitial fibrosis), fractional anisotropy (FA) (directional variability of water diffusion with low levels thought to be related to collagen infiltration and cardiomyocyte disorganisation), secondary eigenvector angle (E2A) (a marker of sheetlet orientation). cDTI postprocessing was performed using an inhouse Matlab based software with automatic registration. Using Circle Cvi42 software, 3D feature tracking (FT) was derived for global radial strain (GRS), global circumferential strain (GCS), and global longitudinal strain (GLS).Results Whilst the HCM G+LVH- patients showed significantly different LVEF, mass and ECV compared to HCM G+LVH+, no difference in mass and ECV was found between HV and HCM G+LVH- (see table 1). GLS was significantly different only between HCM G+LVH- and HCM G+LVH+, while no other strain parameters differed significantly between groups.All cDTI biomarkers were significantly different between HV, HCM G+LVH- and HCM G+LVH+ [See table 1]. ROC curves showed a better diagnostic performance for cDTI biomarkers compared to strain in identifying subclinical HCM with AUC values of: MD 0.769, FA 0.828, E2A 0.731, GLS 0.579, GCS 0.547 and GRS 0.568. ROC curves analysis showed stronger diagnostic performance for cDTI biomarkers compared to strain in identifying overt HCM (G+LVH+) with AUC values of: MD 0.878, FA 0.939, E2A 0.971, GLS 0.805, GCS 0.595 and GRS 0.582.Abstract 6-003 Table 1Results [mean (SD)] HV (N=23) HCM (G+LVH-) (N=67) HCM (G+LVH+) (N=37) p-values HV vs HCM (G+LVH-) HCM (G+LVH-) vs HCM (G+LVH+) Age (yrs) 35 (9) 33 (11) 51 (14) 0.523 <0.001 LVEF (%) 66 (4) 71 (6) 77 (6) <0.001 <0.001 LV mass indexed (g/m2) 55 (15) 54 (22) 84 (22) 0.560 <0.001 Global ECV (%) 26 (3) 27 (3) 29 (3) 0.319 0.003 Septal ECV (%) 26 (3) 27 (3) 31 (5) 0.190 <0.001 MD (x 10–3mm2/s) 1.46 (0.04) 1.51 (0.05) 1.56 (0.08) <0.001 <0.001 FA 0.33 (0.02) 0.31 (0.02) 0.28 (0.04) <0.001 <0.001 |E2A|° 40.8 (8.6) 48.9 (8.5) 58.8 (5.6) <0.001 <0.001 GRS 39.9 (7.9) 42.4 (10.9) 44.1 (13.2) 0.334 0.752 GCS -21.6 (1.8) -21.1 (2.7) -20.8 (3.2) 0.505 0.417 GLS -16.4 (1.6) -15.9 (1.9) -13.5 (2.8) 0.263 <0.001 Conclusion cDTI has better diagnostic performance to detect adverse myocardial change associated with subclinical HCM than 3D-FT strain. Especially in the HCM G+LVH- population, where ECV showed no significant difference compared to HVs, cDTI may provide further utility in detecting subclinical HCM.