Microstructural and Microvascular Phenotype of Sarcomere Mutation Carriers and Overt Hypertrophic Cardiomyopathy

BACKGROUND: In hypertrophic cardiomyopathy (HCM), myocyte disarray and microvascular disease (MVD) have been implicated in adverse events, and recent evidence suggests that these may occur early. As novel therapy provides promise for disease modification, detection of phenotype development is an emerging priority. To evaluate their utility as early and disease-specific biomarkers, we measured myocardial microstructure and MVD in 3 HCM groups—overt, either genotype-positive (G+LVH+) or genotype-negative (G−LVH+), and subclinical (G+LVH−) HCM—exploring relationships with electrical changes and genetic substrate. METHODS: This was a multicenter collaboration to study 206 subjects: 101 patients with overt HCM (51 G+LVH+ and 50 G−LVH+), 77 patients with G+LVH−, and 28 matched healthy volunteers. All underwent 12-lead ECG, quantitative perfusion cardiac magnetic resonance imaging (measuring myocardial blood flow, myocardial perfusion reserve, and perfusion defects), and cardiac diffusion tensor imaging measuring fractional anisotropy (lower values expected with more disarray), mean diffusivity (reflecting myocyte packing/interstitial expansion), and second eigenvector angle (measuring sheetlet orientation). RESULTS: Compared with healthy volunteers, patients with overt HCM had evidence of altered microstructure (lower fractional anisotropy, higher mean diffusivity, and higher second eigenvector angle; all P<0.001) and MVD (lower stress myocardial blood flow and myocardial perfusion reserve; both P<0.001). Patients with G−LVH+ were similar to those with G+LVH+ but had elevated second eigenvector angle (P<0.001 after adjustment for left ventricular hypertrophy and fibrosis). In overt disease, perfusion defects were found in all G+ but not all G− patients (100% [51/51] versus 82% [41/50]; P=0.001). Patients with G+LVH− compared with healthy volunteers similarly had altered microstructure, although to a lesser extent (all diffusion tensor imaging parameters; P<0.001), and MVD (reduced stress myocardial blood flow [P=0.015] with perfusion defects in 28% versus 0 healthy volunteers [P=0.002]). Disarray and MVD were independently associated with pathological electrocardiographic abnormalities in both overt and subclinical disease after adjustment for fibrosis and left ventricular hypertrophy (overt: fractional anisotropy: odds ratio for an abnormal ECG, 3.3, P=0.01; stress myocardial blood flow: odds ratio, 2.8, P=0.015; subclinical: fractional anisotropy odds ratio, 4.0, P=0.001; myocardial perfusion reserve odds ratio, 2.2, P=0.049). CONCLUSIONS: Microstructural alteration and MVD occur in overt HCM and are different in G+ and G− patients. Both also occur in the absence of hypertrophy in sarcomeric mutation carriers, in whom changes are associated with electrocardiographic abnormalities. Measurable changes in myocardial microstructure and microvascular function are early-phenotype biomarkers in the emerging era of disease-modifying therapy.

H ypertrophic cardiomyopathy (HCM) is a leading cause of heart failure and sudden death. 1 Characterized clinically by hypertrophy, HCM has considerable heterogeneity of presentation and outcomes. 2,3 Cardiac function, extreme hypertrophy, and extent of fibrosis are known to be adverse markers, but these may be late in phenotype development and may only modestly predict adverse events. [2][3][4][5] Histologically, HCM is characterized by myocyte hypertrophy and fibrosis but also by 2 other features: small-vessel disease and disarray. 5,6 It is thought that these may be important in both phenotype development, from subclinical (in the absence of hypertrophy) to overt disease, and risk; but until recently, these have been hard to measure clinically, and no study has measured them concurrently. [7][8][9] Novel advances permit quantification of microstructural indices indicative of disarray and microvascular disease (MVD), bypassing the need for tissue specimens or ionizing radiation. 10,11 Cardiac diffusion tensor imaging (cDTI) measures the diffusion of water within an imaging voxel, thereby characterizing the myocardial microstructural environment. HCM is characterized by: (1) low fractional anisotropy (FA), a measure of the directional variability of water diffusion, with low values suggestive of myocyte disarray and collagen deposition; (2) high mean diffusivity (MD), measuring the magnitude of diffusion, thought to reflect myocyte packing, with high values reflecting increased interstitial fibrosis and changes in intracellular and extracellular volume; and (3) elevated absolute second eigenvector angle (|E2A|), reflecting sheetlet orientation (functional units of myocytes that dynamically reorientate throughout the cardiac cycle to facilitate wall thickening). Sheetlets conform to hypercontracted states (elevated |E2A|) in systole with failure to reorientate in diastole, implying roles in diastolic failure in HCM and increased cardiomyocyte tension [12][13][14][15] (Figure 1).
Patients with overt HCM without a detectable sarcomeric mutation, called genotype-negative (G−LVH+), represent approximately half of patients and are characterized by fewer events (heart failure or ventricular arrhythmia), less fibrosis (lower extracellular volume [ECV] or less scar), and differing morphology (more apical HCM or more isolated basal septal left ventricular [LV] hypertrophy [LVH]). 1,16,17 Individuals with pathogenic sarcomeric variants without hypertrophy (G+LVH−) identified on cascade genetic screening have architectural and functional abnormalities (crypts, trabeculation, hyperdynamic function, impaired diastolic function, and myocardial mechanics). Furthermore, G+LVH-with abnormal electrocardiographic findings have a 4-fold increased risk of progression to overt disease. [18][19][20] As novel sarcomere modulators provide promise for disease modification, early markers of disease are a

Clinical Perspective
What Is New?
• With the use of diffusion tensor imaging in hypertrophic cardiomyopathy, evidence of myocardial disarray is detectable in pathogenic sarcomere mutation carriers even in the absence of hypertrophy. • Diffusion tensor imaging parameters suggestive of myocardial disarray are related to electrocardiographic abnormalities in subclinical and overt hypertrophic cardiomyopathy. • In overt disease, the presence versus absence of sarcomeric mutation has differing effects on microstructure and microvasculature.
What Are the Clinical Implications?
• Diffusion tensor imaging may discriminate pathogenic sarcomere mutations from health and mutation status in overt disease. • Disarray and ischemia are associated with stage of phenotype evolution. rapidly emerging research priority. 21 Analyzing the effect of sarcomeric mutation on phenotype development will enhance precision medicine in the future. 1,5,22 We aimed to investigate the prevalence, extent, and interrelationships of microstructural alteration and MVD in both overt and subclinical HCM. We then studied the associations between these biomarkers and pathological electrocardiographic findings, nonsustained ventricular tachycardia (NSVT), and genetic substrate (genotype positive [G+LVH+] and genotype negative [G−LVH+]).

Study Population
The data that support the findings of this study are available from the corresponding author on reasonable request. We formed an academic collaboration consisting of 3  the subclinical HCM (G+LVH−) cohort under the same ethics protocol. A separate age-, sex-, and ethnicity-matched cohort of HVs was compared with the overt HCM group to assess whether any diffusion tensor imaging (DTI) changes were attributable to demographics variables (n=24; 11 overlapped with the younger HV group). HVs were not genotyped and were not relatives of those with HCM. To optimize conditions for cDTI and to reduce the risk of confounding, those with known obesity (body mass index ≥30 kg/m 2 ) and poorly controlled hypertension (poor blood pressure control despite 2 antihypertensives) were excluded. Presence of LV outflow obstruction and gradients was collected from clinical echocardiographic data at the time of recruitment. Other exclusion criteria were as follows: claustrophobia, unwillingness to consent, known coronary disease or significant pretest probability for coronary disease without previous coronary imaging, contraindications to adenosine, pulmonary hypertension, previous alcohol septal ablation or myectomy, previous cardiovascular surgery, congenital disease, and implantable cardiac devices.

CMR Image Acquisition and Analysis
All scans were acquired on a single 3-T Prisma (Siemens Healthineers, AG, Erlangen, Germany). Standard long-and short-axis cine imaging was performed, with segmentation for ventricular dimensions and function performed with fully automated algorithms that exceed human precision. 24,25 T1 mapping before and after contrast was performed with Modified Look-Locker Inversion Recovery (MOLLI 5s[3s]3s) with automated contouring using Circle CVI42 software (version 5.14.1) and manual adjustment when required for septal ECV (using sameday hematocrit) in line with recommendations. 26 Fully quantitative vasodilator stress perfusion was performed with a validated dual sequence approach as previously described. 10 In brief, adenosine was given intravenously at 140 to 210 µg•kg −1 •min −1 for a minimum of 4 minutes, until a minimum heart rate increase of 15 bpm and symptoms suggestive of an adequate physiological stress response. Gadolinium-based contrast (Dotarem, Gadoteric Acid, Guerbet, UK) was then administered intravenously at 0.05 mmol/kg. The same was repeated for rest imaging, which was acquired at a minimum of 7 minutes after adenosine. Automated in-line adenosine stress perfusion maps were acquired to obtain stress and rest myocardial blood flow (MBF; in milliliters per gram per minute), myocardial perfusion reserve (MPR; ratio of stress and rest MBF), and visual perfusion defects (defined according to clinically recommended task force criteria 27 ) assessed separately by 2 experienced operators (G.J. and J.C.M.) from conventional images and perfusion maps. 28 A second-order motion-compensated single-shot spinecho echo-planar imaging DTI sequence was performed for 3 short-axis slices at peak systole following previously described protocols. 12,13 This was free breathing without respiratory navigation, has been validated both ex vivo and in vivo, and has been demonstrated to detect microstructural changes in HCM. 12,13,29 cDTI analysis was performed according to previously described protocols; in brief, data processing was performed with custom-built Matlab software (MathWorks), with each image quality-controlled for misregistration or artifacts. Average magnitude images were generated from registered data by averaging cross-repetitions, and diffusion tensors were calculated. Contouring was performed on cDTI maps, with tensor eigenvalues, MD, FA, and absolute |E2A| obtained globally and segmentally. Segments containing artifacts were omitted from analysis. 12,13 Late gadolinium enhancement (LGE) imaging was performed 5 minutes after administration of gadolinium-based contrast agent for pixelwise quantification of focal fibrosis (defined as signal intensity 5 SDs from the mean of a remote region of interest and expressed henceforth as absolute grams). Circle CVI 42 (version 5.14.1) was used for LGE map analysis with automated contouring of endocardial and epicardial borders and manual adjustment when required.

Digital Electrocardiographic Analysis
A same-day standard resting digital 12-lead ECG was performed with a Mindray Beneheart R3 electrocardiograph (Shenzhen Mindray Bio-Medical Electronics Co), and automated analysis was performed by a core laboratory to identify abnormalities associated with disease progression: abnormal Q waves (defined as present in ≥2 contiguous leads and minimum amplitude of 0.3 mV, ≥25% of the subsequent R wave, or duration >40 milliseconds), Sokolow-Lyon index (SV1+RV5/6>3.5 mV) or Cornell criteria (RaVL+SV3>2.8 mV in men and >2.0 mV in women), and repolarization abnormalities (defined as T-wave inversion in ≥0.1 mV in ≥2 contiguous leads or ST-segment depression ≥0.1 mV in ≥2 contiguous leads). 20,30 For overt HCM, Holter recordings were available in 99% of patients (101/102) within 6 months of the CMR, with NSVT defined as 3 beats ≥120 bpm occurring on Holter monitoring.

Statistical Analysis
Statistical analysis was performed with SPSS (IBM SPSS Statistics, version 28). Continuous variables were described as mean±SD for normally distributed variables and compared with independent-samples t test. For nonnormally distributed variables (tested with the Shapiro-Wilk test), median and interquartile range with equivalent comparison tests were used. Categorical/binomial variables were expressed as absolute values and percentages and compared with χ 2 or Fisher exact test as appropriate. Univariable and multivariable linear regression models were used to determine associations of diffusion tensor, perfusion, fibrosis, and hypertrophy parameters. Betweengroup comparisons involving DTI were adjusted for age (see later discussion), fibrosis, and hypertrophy (septal ECV, LGE mass, and MWT) and expressed as standardized β coefficients and 95% CIs. A variance inflation factor <3 excluded multicollinearity. Logistic regression models with inclusion of fibrosis (ECV and LGE) and hypertrophy (MWT) variables as covariates were used to test for odds ratios (ORs) for independent risk of electrocardiographic abnormalities and NSVT. Continuous predictors were normalized before entering the logistic regression, and ORs are expressed per 1-SD increase in MD and |E2A| and 1-SD decrease for FA, stress MBF, and MPR. A 2-sided value of P<0.05 was considered significant. Multiple testing correction with Bonferroni was done for the primary end points (perfusion and DTI in patients with G+LVH− versus HVs and G+LVH+ versus G−LVH+) in a table-by-table basis (adjusted P threshold=0.05/[number of parameters analyzed multiplied by number of comparisons]). Age, sex, ethnicity, and body mass index were tested for associations with DTI parameters (Supplemental Results) in pooled HVs (n=41), subclinical HCM, and overt HCM. MD was found to be associated with age in overt HCM; therefore, age was included in all between-group regression analyses involving DTI. An exploratory analysis was performed to determine whether any gene-specific differences could be detected (Supplemental Results).
Overt HCM: All LVH+ Versus HVs Overt HCM was characterized by higher ejection fraction and more severe markers of hypertrophy (higher MWT and higher LV mass) and fibrosis (higher ECV and higher burden of LGE; Table S1).
Overt HCM: G+LVH+ Versus G−LVH+ There were no differences in ejection fraction and MWT. Patients with G−LVH+ had higher LV mass. G−LVH+ had similar LGE burden and lower ECV (but not after correction for multiple comparisons; Table 1).

Subclinical HCM (G+LVH−) Versus HVs
Patients with G+LVH− had reduced stress MBF but not after correction for multiple comparisons (  Figure 3B). Patients with subclinical HCM had a higher prevalence of perfusion defects (28% versus 0%; P=0.002) but no difference in MPR ( Figure S1).

Microstructural Indices Measured by DTI
Associations of DTI parameters, fibrosis, and MWT are presented in Supplemental Results.

Overt HCM (All LVH+) Versus HVs
Compared with HVs, patients with overt HCM had evidence of microstructural alteration: lower FA, suggestive of disarray and higher MD, and higher |E2A| suggestive of a more hypercontracted sheetlet configuration (  Figure 3A; Table 2).

Subclinical HCM (G+LVH−) Versus HV
Compared with HVs, patients with G+LVH− had evidence of microstructural alteration (lower FA, higher MD, and higher |E2A|), although this was less severe compared with patients with overt HCM (

Subclinical HCM (G+LVH−) ECG Overall
Of patients with subclinical HCM, 34% (26) had an abnormal ECG. An abnormal ECG was associated with all

Individual Electrocardiographic Abnormalities
Regarding individual electrocardiographic abnormalities, 21% (16) had abnormal Q waves; 9% (7) had T-wave inversion; 12% (9) met voltage criteria for LVH; and no participant had ST-segment depression. When we considered associations with individual electrocardiographic abnormalities in isolation, only FA and MD were predictive of abnormal Q waves after adjustment for MPR, fibrosis, and hypertrophy (FA: OR, 5.

DISCUSSION
In the era of cascade genetic screening and emerging novel therapy, detection of phenotype development in subclinical HCM is an emerging priority. Study findings show that changes in DTI and quantitative perfusion occur even in the absence of hypertrophy. DTI abnormalities in subclinical disease relate to electrocardiographic abnormalities and perfusion defects, showing the likely importance of disarray and MVD in phenotype development.
Recent evidence suggests that genotype-positive HCM and genotype-negative HCM are different in LV morphology and clinical outcomes. 1,16,17 In overt disease with LVH, all genotype-positive patients have perfusion defects, with 18% of genotype-negative patients having none. DTI abnormalities were unexpectedly more marked if the patient was gene-negative (more elevated |E2A|). This suggests that changes in DTI and quantitative perfusion are sensitive to mutation status.
This study is both the largest in vivo cDTI study performed to date and the largest prospective study of subclinical HCM. The study used a validated DTI sequence and consecutive recruitment from genetics databases of 3 referral cardiomyopathy centers where phenocopies are routinely screened. The exact mechanisms behind how disarray and MVD are present, even without hypertrophy, in those with pathogenic sarcomeric mutations are still elusive.

Microstructural Changes, MVD, and Overt HCM
According to several preclinical models, DTI abnormalities are likely related to microstructural alteration (myocyte disarray and abnormal sheetlet orientation), although further human histological validation is needed. 15,29,31 In line with other studies, overt disease was characterized by low FA and high MD. [12][13][14][15]22 Although others found an independent association of FA and MD with fibrosis, our study findings also demonstrate an independent relationship with hypertrophy, mirroring previous histological work. 12,22,32 Our study is the first to systematically examine the presence of microstructural abnormalities, including disarray, in patients with G+LVH+ and those with G− LVH+. Genotype-negative HCM has been hypothesized to result in part from polygenic inheritance. 17 |E2A| elevation reflects a hypercontracted microstructural state in systole with failure to reorientate in diastole. 12,14,15 Although G− LVH+ has been associated with less severe outcomes and fibrosis, here, it unexpectedly is related to elevated |E2A|, suggesting a more severe microstructural phenotype. Conversely, although both patients with G−LVH+ and those with G+LVH+ displayed considerable MVD compared with HVs, all G+LVH+ patients had visual perfusion defects compared with 82% of G−LVH+ patients. In overt disease, perfusion defects are associated with abnormal blood pressure response to exercise and, in apical HCM, aneurysm formation. 33,34 In a previous positron emission tomography study, global MBF was more impaired in patients with G+LVH+ compared with those with G−LVH+, and although this was not found here, our study also supports the hypothesis of a more direct deleterious impact of sarcomeric mutation on microvascular function. 35 Microstructural Changes and Subclinical HCM Subclinical HCM was characterized by LVH and fibrosis parameters similar to health, and despite this, DTI demonstrated lower FA, suggestive of more myocyte disarray. Although historically, disarray has been described in HCM for several decades, its presence in phenotype evolution was less understood. 8 Recent 3-dimensional histological analysis uncovered that disarray is detected ex vivo in fetal murine models without hypertrophy. 9 Our previous work and other investigations have modeled early disease by examining DTI parameters of nonhypertrophied myocardium in overt disease; however, this has limitations, as remote remodeling (in nonhypertrophied segments) occurs by the time LVH is detected. 18,22,36 Elevated |E2A| in patients with G+LVH− suggests that the hypercontracted microstructural state demonstrated in overt disease is also present before hypertrophy. [12][13][14] Elevated |E2A| may have a role in the hyperdynamic function, diastolic dysfunction, and impaired myocardial mechanics found in this cohort. 18,37 Overall, our findings support the hypothesis that altered myocardial microstructure is an early phenomenon in the disease pathophysiology, in keeping with animal model work. 9

Microstructural Changes and Perfusion Relationship
Twenty-eight percent of patients with subclinical HCM had perfusion defects, and these had evidence of low FA, suggestive of more disarray, and more elevated |E2A| than those without perfusion defects. This relationship also persisted into overt disease, with markers of MVD being independently associated with markers of microstructural integrity. A unifying explanation for the findings of abnormal microstructure, fibrosis, and abnormal microvasculature is the capillary:myocyte coupling hypothesis, whereby microvasculature and the matrix meshwork are abnormal even during organogenesis. 7,38 In overt HCM, compressive forces of hypertrophied myocardium and LV outflow tract obstruction also worsen ischemia. 39 An interrelationship of perfusion, disarray, and myocardial mechanics could explain the compounding of perfusion and microstructural changes with LVH and their continued association in overt disease.

MVD, Microstructural Changes, and Pathological Electrocardiographic Findings
In line with other studies in subclinical HCM, pathological electrocardiographic findings occurred in 34% and were more common in overt disease at 84%. 20,40 The study shows for the first time that microstructural alteration (including lower FA, suggestive of cardiomyocyte disarray) is independently associated with electrocardiographic abnormalities in subclinical disease. Prognostically, pathological electrocardiographic findings in subclinical HCM are associated with a 4-fold increased risk of progression to overt disease. 20 Mechanisms linking disarray to arrhythmia susceptibility include disruption of gap junctions, alteration of longitudinal to transverse conduction velocity ratios, and provision of different pathways for conduction. 8 A pioneering DTI study elicited FA of the thickest LV segment associated with NSVT; however, histological studies demonstrate large variations in disarray from segment to segment. 32 This relationship was not replicated here when global FA was used, accounting for the low number of individuals with NSVT in this relatively low-risk cohort (implantable devices excluded). 22 Others have hypothesized that ischemia results in scar, which, in turn, leads to reentry circuits. 6 We found an independent association of MVD parameters with abnormal ECG, with MPR associating with electrocardiographic abnormalities in subclinical HCM, suggesting that the arrhythmogenic potential of MVD could be more complex than scar formation itself. 20 Limitations DTI parameters are variably influenced by the presence of fibrosis and other factors and therefore are not direct measures of disarray. However, the best available techniques were used for measurement of the important confounders: fibrosis, perfusion, and hypertrophy. MBF using adenosine stress is also an indirect measure of MVD and does not physiologically represent exercise. We are unable to fully exclude coronary disease, as systematic invasive coronary imaging was not performed. HVs did not undergo genotyping or Holter monitoring. Subclinical genotypenegative disease was not assessed. As with all contemporaneous cDTI sequences, susceptibility artifact, particularly in the inferolateral wall, was commonly found, and patients with devices were excluded.

Conclusions
Microstructural alteration and MVD occur in overt HCM and are different in G+ and G− patients; both occur even in the absence of hypertrophy in sarcomeric mutation carriers, in whom changes are associated with electrocardiographic abnormalities. Microstructural alteration and MVD are associated in both subclinical and overt disease, suggesting relationships in phenotype development. Measurable changes in myocardial microstructure and microvascular function are early-phenotype biomarkers in the emerging era of disease-modifying therapy.