Early Identification of Myocardial Microstructural Alterations in Hypertrophic Cardiomyopathy with in Vivo Cardiac Diffusion-Tensor Imaging

Purpose To explore the diffusion characteristics of hypertrophic cardiomyopathy (HCM) using in vivo cardiac diffusion-tensor imaging (cDTI) and to determine whether cDTI could help identify abnormal myocardium beyond cardiac MRI findings of fibrosis and hypertrophy. Materials and Methods In this prospective study conducted from April to August 2023, participants with HCM and healthy volunteers were enrolled for cardiac MRI evaluation, including cine, late gadolinium enhancement (LGE), T1 mapping, and cDT imaging, using a 3.0-T scanner. LGE was performed only in participants with HCM. All cardiac MR images of the midventricular section were divided into six American Heart Association segments for segmental analysis. Repeated measure analysis of variance and Friedman test were used to compare cDTI parameters among different segments, and a linear mixed-effects model was fitted to account for within-participant clustering. Diagnostic performance of cDTI parameters in discriminating HCM segments from normal segments was assessed using receiver operating characteristic (ROC) analysis. Results Thirty-five participants with HCM (mean age, 46 years ± 13 [SD]; 19 [54%] male) and 15 healthy volunteers (mean age, 43 years ± 17; nine [60%] male) were included. Compared with controls, the HCM group showed significantly reduced fractional anisotropy (FA) (mean, 0.33 ± 0.05 vs 0.46 ± 0.04; P < .001) and increased diastolic second eigenvector angle (E2A) (mean, 48.85° ± 7.13 vs 35.05° ± 5.06; P < .001). In segmental analysis, FA and diastolic E2A of HCM segments with no hypertrophy or LGE also significantly differed from controls (estimated marginal mean, FA: 0.39 [95% CI: 0.38, 0.41] vs 0.45 [95% CI: 0.43, 0.47]; E2A: 46.03° [95% CI: 43.45, 48.62] vs 35.69° [95% CI: 31.69, 39.69]; both P < .001), with no evidence of a difference in native T1 values (estimated marginal mean, 1282.43 msec [95% CI: 1267.55, 1297.30] vs 1265.57 msec [95% CI: 1242.86, 1288.28]; P = .47). FA (area under the ROC curve [AUC], 0.81 [95% CI: 0.76, 0.86]) and diastolic E2A (AUC, 0.79 [95% CI: 0.74, 0.84]) showed higher performance than native T1 (AUC, 0.67 [95% CI: 0.61, 0.72]) in identifying HCM segments from normal segments (P < .001 and P = .007, respectively). Conclusion Myocardial disarray and abnormal microstructural dynamics of HCM can be identified by using in vivo cDTI even in the absence of hypertrophy or fibrosis, suggesting the potential value of cDTI in tissue characterization of HCM. Keywords: Cardiovascular MR Imaging, Hypertrophic Cardiomyopathy, Diffusion-Tensor Imaging, Myocardial Microstructure Supplemental material is available for this article. Published under a CC BY 4.0 license.