Background Although intravoxel incoherent motion (IVIM) MRI has emerged as an in vivo marker of tissue diffusion and perfusion, its prognostic value in patients with hypertrophic cardiomyopathy (HCM) remains unclear. Purpose To investigate whether IVIM‐MRI derived parameters are associated with outcomes in patients with HCM. Study Type Prospective cohort. Subjects A total of 112 patients (51.72 ± 17.13 years) with suspected or known HCM. Field Strength/Sequence Single‐shot echo planar IVIM imaging, balanced steady‐state free precession, and phase‐sensitive inversion‐recovery late gadolinium enhancement (LGE) sequences at 3 T. Assessment All patients were followed up of 29.3 ± 12.3 months for combined major adverse cardiac events (MACE) including cardiac death, aborted sudden death, heart transplantation, and rehospitalization for heart failure. The CVI42 imaging platform was used to assess morphological and functional MRI indices and to quantify LGE. The Body Diffusion Toolbox was used to derive pseudo diffusion (D*), water molecular diffusion (D) and perfusion fraction (f). Statistical Tests Univariable and stepwise multivariable Cox model analyses were used to investigate the association between variables and composite endpoints. Kaplan–Meier curves were constructed to assess event‐free survival, and the event rates were compared by the log‐rank test. Results A total of 19 patients reached endpoints. Patients with MACE showed a significantly impaired D* value, lower f value, and more extensive LGE than those without MACE (all, P < 0.05), while there was no significant difference in D value (P = 0.285). In the Cox regression models, D* value (hazard ratio [HR] 0.93; 95% CI: 0.88–0.98) and f value (HR 0.65; 95% CI: 0.45–0.92) were independent predictors for MACE. Moreover, in Kaplan–Meier survival analysis, the incidence of MACE was significantly higher in patients with decreased D* value and f value. Conclusions Impaired D* and f values derived from IVIM‐MRI are associated with adverse outcomes in patients with HCM. Level of Evidence 2 Technical Efficacy Stage 2

No abstract

No abstract

To the Editor: We appreciate the authors for their thoughtful comments regarding our recent work and related articles. The authors raised an important point: As the diffusivity of water molecules is temperature driven and restricted and/or hindered by the tissue microstructure, the D value is subject to the theoretical limit of 2.9 Â 10 À3 mm 2 /s. We fully agree with the statement, and this limitation was omitted in our discussion.Previous studies have shown that the structures within tissues do not impede water molecules diffusion due to sufficient permeability. Despite estimates of the obstruction effect being possible, accurate evaluations are difficult. 1 In clinical practice, the D value is often substituted by the apparent diffusion coefficient (ADC) value for its inaccuracy. 2 As indicated by our findings, the D value had relatively poor repeatability and was discrepant with previously reported values. For this, we studied the present literature carefully and considered the differences in scan protocols, multiple b-value combinations, and image post-processing that may cause the deviation of the D value. [2][3][4] Most importantly, it should be noted that conventional single-shot echo-planar imaging is apt to have artifacts in vivo heart, resulting in apparent signal attenuation.The stimulated echo acquisition mode (STEAM) sequences or retrospective motion compensation strategies, as mentioned by the authors of the letter, is an elegant approach to assess diffusion in the intravoxel incoherent motion (IVIM) model. Theoretically, STEAM can be used for in-vivo assessment of myocardial microstructure. However, cardiac STEAM was also found to have the disadvantage that low signal-noise ratio (SNR) data yield parameter biases in the IVIM model. 3 It also suffers from the dependence upon a regular heart rate and the potential confounding effects of strain, 4 which makes studies difficult during arrhythmia. Besides, a limited number of studies in ex vivo tissues, animal models, or healthy subjects have used a related diffusion-prepared technique for diffusion-weighted imaging, indicating limited success. [3][4][5] So far, we regard this approach needs further verification to establish technique viability and fully assess the diagnostic potential in the evaluation of hypertrophic cardiomyopathy (HCM) patients, and is therefore not yet suitable for use in daily clinical practice.