Introduction
Hypertrophic cardiomyopathy (HCM) is the most common inherited cardiac disease and remains a leading cause of heart failure, atrial fibrillation and sudden cardiac death.12 Although contemporary management has substantially reduced disease-related mortality, progressive exercise intolerance and heart failure have become the dominant determinants of quality of life and long-term outcome. Yet the mechanisms responsible for symptom progression remain incompletely understood. Identifying why some patients evolve from compensated hypertrophy to disabling heart failure whereas others remain minimally symptomatic despite similar ventricular morphology has become one of the major challenges in contemporary HCM.3–5
Current evaluation of HCM remains predominantly ventricle-centred. Diagnosis, risk stratification and therapeutic decision-making rely largely on left ventricular hypertrophy, outflow tract obstruction, fibrosis and systolic function. While these parameters are essential for defining the disease and estimating arrhythmic risk, they correlate only modestly with exercise capacity and heart failure severity.
Patients with comparable wall thickness, obstruction and ejection fraction frequently exhibit markedly different functional status, suggesting that ventricular morphology alone does not fully explain the heart failure phenotype of HCM.2367
The left atrium may represent this missing biological link. Rather than being a passive chamber exposed to chronically elevated filling pressures, the left atrium integrates the cumulative consequences of impaired ventricular relaxation, myocardial fibrosis, elevated filling pressures and abnormal atrioventricular interaction. Progressive atrial remodelling ultimately evolves into atrial mechanical failure, characterised by increased chamber stiffness, impaired reservoir function, reduced contractile reserve and loss of atrioventricular coupling.
Once this transition occurs, the atrium can no longer adequately augment ventricular filling during exercise, limiting stroke-volume reserve, forward cardiac output and functional capacity. In this perspective, atrial failure represents not merely a consequence of ventricular disease but an integrated manifestation of disease progression and a potential determinant of the heart failure phenotype in HCM.8–12
Conventional markers of atrial remodelling, including left atrial volume and reservoir strain, have consistently demonstrated prognostic value in HCM. More recently, composite indices integrating atrial mechanics with ventricular filling pressure or chamber interaction—including the Left Atrial Stiffness Index (LASI), Left Atrial Coupling Index (LACI) and Left Atrioventricular Coupling Index (LAVCI)—have been proposed as quantitative markers of atrial mechanical failure. Whether these indices better explain functional limitation and identify patients at risk of heart failure progression than conventional ventricular assessment remains unknown.13–15
We therefore hypothesised that progressive atrial failure, rather than ventricular morphology alone, is a central determinant of the heart failure phenotype in HCM. In a large consecutive cohort of patients undergoing comprehensive echocardiography and cardiopulmonary exercise testing (CPET), we investigated the relationship between novel markers of atrial mechanical failure, objectively measured exercise capacity and subsequent clinical outcomes. Rather than comparing imaging indices, our objective was to determine whether assessment of atrial failure provides a more clinically relevant framework for understanding symptom progression and heart failure evolution in HCM.
Material and methods
Study population
All patients consecutively diagnosed and genotyped for HCM at Rennes University Hospital between January 2014 and January 2024 were retrospectively included.
The diagnosis of HCM was made in accordance with current recommendations: left ventricle (LV) hypertrophy ≥15 mm (≥13 mm in the presence of family history of HCM) not sufficiently explained by loading conditions such as arterial hypertension, aortic stenosis and athlete’s heart.151617
Exclusion criteria were: age <18 years, significant coronary artery disease, previous myocardial infarction, moderate to severe valvular disease unrelated to HCM, patients with prosthetic valves, patients with atrial fibrillation on baseline echocardiography and patients with incomplete or poor-quality echocardiography (figure 1).
Data collection
Clinical data, including anthropometrics, vital signs, cardiovascular symptoms, personal and family history, and medication, were collected from electronic medical records.
Baseline evaluation included a 12-lead ECG, NT-proBNP assay (standard <125 pg/mL), 48-hour Holter ECG, genetic analysis (panel of 5 or 16 genes according to the recommendations in force at the time of sampling, online supplemental appendix 1) and resting transthoracic echocardiography (TTE).
Stress echocardiography was not systematically performed and was guided by the patient’s symptoms and the presence of obstruction at rest.
Most patients underwent CPET at baseline. Those who could not undergo the test were only included in analyses of follow-up events.
Most patients also underwent cardiac magnetic resonance (CMR) within 1 year of their HCM diagnosis. Left ventricular ejection fraction (LVEF) on CMR was determined using a semi-automatic measurement of left ventricular volumes, and the presence of late gadolinium enhancement was visually and qualitatively assessed on dedicated sequences.
For all patients under 80 years, the 5-year risk of sudden death was calculated according to the HCM Risk-SCD, recommended by the European Society of Cardiology.5
Echocardiographic evaluation
TTE was performed on a Vivid 7, E9 or E95 (GE Healthcare, Horten, Norway). All recorded loops (2D, colour Doppler, pulsed Doppler, continuous Doppler, tissue Doppler and 2D-speckle tracking) were stored on a dedicated workstation (EchoPAC V.204; GE Healthcare) and all measurements were conducted by an independent echocardiographer blinded to clinical, exercise and outcome data.
Standard measurements, focusing on the structural and functional assessment of the four cardiac chambers and the mitral valve, were performed as recommended.
We classified HCM into six subtypes based on hypertrophy morphology: localised basal septal, septal reverse curvature, apical, concentric, mid-cavity with or without apical aneurysm and others.17
Left intraventricular obstruction was defined as a subvalvular aortic gradient ≥30 mm Hg at rest and/or ≥50 mm Hg during exercise or provocation tests (ie, Valsalva manoeuvres).
Left atrial strain (LAS) and left ventricle global longitudinal strain (LV-GLS) were obtained by speckle tracking echocardiography from dedicated 2D apical views (2 and 4 for atrium and 2, 4 and 3 for ventricle). Endocardial borders were semi-automatically defined and divided into six segments, then manually adjusted if needed. The region of interest was adjusted to fit the myocardial thickness.
Visual assessment of wall tracking was done and patients were excluded in case of insufficient tracking. R-waves on ECG were set as the baseline point and measurements were obtained between two consecutive R-waves. The final values corresponded to the average of each endocardial segment collected from all the previously mentioned views.
Atrial assessment parameter
The anteroposterior diameter measurements were obtained in end-systole using a parasternal long-axis view. The Left Atrial Volume Index (LAVI) was measured in end-systole from apical four-chamber and two-chamber views using the area-length method and indexed to body surface area.
From the LAS curves, we collected three deformation markers (figure 2). An expansion phase, the LA reservoir function, estimated by the peak strain during ventricular systole which corresponds to LA filling during LV systole. Two shortening phases contribute to left ventricular filling: a passive phase between mitral valve opening and atrial contraction corresponding to the conduit function (LAScd) and an active phase due to atrial contraction function (LASct) which occurs until mitral valve closure only in patients with sinus rhythm.15
Endocardial limits were also used to semi-automatically collect the left atrial ejection fraction (LAEF) and the minimum atrial volume at end-diastole (LAVmin).
In addition to these conventional measures, three composite indices of LA remodelling were calculated.
The LASI (=E/e′/LASr) is an indicator inversely related to atrial compliance and associated with left atrial fibrosis, developed to assess diastolic dysfunction.18 It is the ratio of a marker of left atrial pressure in early diastole (E/e′) to a marker reflecting the left atrial volume reached before the onset of early diastole (LASr).
The LACI (=LAVI/LASct) assesses the structural and functional coupling of the left atrium. Here, the LAVI, a structural marker of the left atrium, is related to LASct rather than the usually described a′ wave. LASct is a valuable parameter of left atrial function due to its independence from acquisition angle and its previously established positive correlation with exercise capacity in patients with HCM.14
Finally, the LAVCI (=LAVmin/LVEDV) assesses the relationship between end-diastolic left heart chamber volumes: LAVmin and LVEDV.14 At this cardiac phase, the two chambers are interconnected, and left atrial volume is directly linked to left ventricular remodelling and function. Moreover, LAVmin, more recently studied than its systolic counterpart, is not influenced by the apical displacement of the mitral annulus associated with left ventricular longitudinal function. As a result, it appears to have a stronger association than LAVI with left ventricular filling pressures and clinical outcomes in HCM (figure 3).
Cardiopulmonary exercise testing
CPET was performed on a cycle ergometer using a progressive load protocol with continuous ECG monitoring. Exercise capacity was assessed based on peak VO₂ and expressed as a percentage of the theoretical peak, calculated according to age, sex and weight using the Cooper and Storer formula.19
To explore the underlying mechanisms of exercise limitation, several parameters were analysed: maximum power output (W), maximum heart rate, percentage of theoretical maximum heart rate (calculated using the 220−age formula), exercise blood pressure profile, ventilatory efficiency (VE/VCO₂ slope), first ventilatory threshold (VT1), O₂ pulse, percentage of predicted O₂ pulse and the presence of an O₂ pulse rebound (online supplemental materials).117
The reference group included participants with a peak VO₂ greater than 80% of the predicted value. Exercise intolerance was defined by the commonly used threshold of peak VO₂ ≤80%, which has been associated with increased morbidity and mortality in HCM. Values below 60% indicated severe functional limitation, a level linked to higher mortality and a greater risk of developing severe symptoms within a few years.
Follow-up outcomes
Patients underwent follow-up in accordance with current recommendations,17 including annual cardiology consultations, repeated 24–48-hour Holter monitoring, and when applicable implantable cardioverter-defibrillator (ICD) interrogation to assess for ventricular arrhythmias and appropriate shocks.
Event data were collected from electronic patient records and, when available, from reports provided by the referring cardiologists.
Several groups of events of interest were retrospectively defined and independently adjudicated by an HCM specialist, blinded to VO₂ and left atrial remodelling data. ‘Heart failure events’ included HF-related hospitalisations, septal reduction therapy, cardiac transplantation and cardiac-related death. ‘Atrial arrhythmia and thromboembolic events’ comprised recorded asymptomatic supraventricular arrhythmias, clinically documented supraventricular arrhythmias and embolic events.
‘Ventricular arrhythmic events’ included unexplained syncope, ventricular tachycardia, aborted sudden cardiac arrest, appropriate ICD therapy for sudden cardiac death prevention and sudden cardiac death. Finally, all-cause mortality was evaluated.
Statistical analysis
Continuous variables were expressed as mean±SD, and categorical variables as counts and percentages. Comparisons between the three groups, defined according to the percentage of predicted VO₂ (>80%, ≤80%, <60%) were performed using logistic regression models for both continuous and categorical variables. A focus on gender was conducted.
Results are reported as ORs with 95% CIs. A p value<0.05 was considered statistically significant.
Receiver operating characteristic (ROC) curves with calculation of the area under the curve (AUC) and Youden’s index were used to assess the diagnostic performance of atrial remodelling variables of interest, and to identify clinically relevant cut-off values.
In the multivariate logistic regression analysis, we included only relevant variables that were significantly associated in both univariable subgroup comparisons, as these may represent more robust prognostic markers. To avoid collinearity, only the atrial parameter with the higher AUC was retained. Parameters directly explaining peak VO₂, such as exercise-derived haemodynamic or ventilatory measures, were excluded from this analysis, as they are descriptive rather than explanatory in nature.
Interobserver and intraobserver variability were assessed in a random sample of 20 patients, based on measurements by two echocardiographers and repeated after 1 week. Agreement and reproducibility were quantified using the intra-class correlation coefficient (ICC) and Bland-Altman analysis.
Kaplan-Meier survival curves were used to evaluate the association between a predefined threshold of the most relevant atrial remodelling marker and follow-up outcomes in patients with a follow-up duration of at least 1 year. Differences between groups were assessed using the log-rank test. Patients lost to follow-up were censored at the date of their last known contact.
All statistical analyses were performed using SAS V.9.4 (SAS Institute, Cary, NC, USA) and R V.4.5.0 (R Foundation for Statistical Computing, Vienna, Austria).
Results
Population description
Among the 336 patients included, 67.3% were male, with a mean age of 55.6±14.2 years (online supplemental figure 1). Their clinical and biological characteristics are detailed in table 1. A sarcomeric gene mutation (pathogenic or likely pathogenic) was identified in 34.5% of patients, most frequently involving MYBPC3, followed by MYH7 and TNNT2 (online supplemental appendix 2).
Women comprised 61/174 (35.1%) of patients with peak VO₂ >80% predicted, 37/120 (30.8%) with peak VO₂ >60% to ≤80% and 16/54 (29.6%) with peak VO₂ ≤60%.
During CPET, hypertension was recorded in 74/174 (42.5%), 48/120 (40.0%) and 20/54 (37.0%) patients across the >80%, >60% to ≤80% and ≤60% groups, respectively; coronary artery disease in 3/174 (1.7%), 3/120 (2.5%) and 2/54 (3.7%); and a history of heart failure in 3/174 (1.7%), 7/120 (5.8%) and 7/54 (13.0%). Beta-blocker use was recorded in 87/173 (50.3%), 73/119 (61.3%) and 40/54 (74.1%); diuretic use in 14/173 (8.1%), 17/119 (14.3%) and 10/54 (18.5%).
Table 2 presents the imaging data, including TTE and CMR. The two predominant hypertrophy phenotypes were basal septal and reverse septal curvature, with a mean maximal left ventricular wall thickness of 18.2±4.0 mm on CMR. The LA were mildly dilated (LAVI: 39.3±17.3 mL/m²) with LASr at 23.5±8.9%, LAScd at 12.5±5.9%, LASct at 11.1±5.9% and LAEF at 51.6±14.2%.
Remodelling indices were as follows: LASI was 0.6±0.7, LACI 6.3±8.9 and LAVCI 0.4±0.3. A significant left ventricular obstruction, either at rest or inducible, was observed in 25% of patients.
Patients with reduced peak VO₂ (tables 1 and 2) were significantly younger and more often treated with beta-blockers. They had more severe and advanced disease, with greater LV remodelling (wall thickening, alterations in LVEF, LV-GLS and CMR-detected fibrosis) and progression to heart failure (higher NYHA class, elevated NT-proBNP, heart failure history, diuretic use). Left atrial remodelling was also more pronounced, with larger volumes and impaired mechanics, lower LASr, LASct and LAEF, while remodelling indices (LASI, LACI and LAVCI) were significantly higher.
In contrast, the prevalence of left ventricular outflow tract obstruction and systolic anterior motion did not differ between groups.
Regarding CPET parameters (table 3), the average peak oxygen uptake (VO₂ peak) was 21.5±7.4 mL/min/kg, corresponding to 82.9±21.4% of the predicted value. Patients with reduced peak VO₂ exhibited a significant deterioration across multiple parameters: a reduced percentage of predicted maximal heart rate, a lower peak systolic blood pressure and an earlier VT1. Oxygen pulse also showed a significant stepwise reduction with a non-significant trend toward a higher frequency of oxygen pulse rebound.
Atrial remodelling parameters associated with peak VO₂
All LA remodelling indices, except LAScd, were significantly more impaired in patients with reduced peak VO₂ (table 4). In univariate logistic regression, LAVCI demonstrated the strongest association with reduced functional capacity, with an OR of 4.36 (95% CI 1.94 to 9.77; p<0.001) for VO₂ ≤80% and 5.44 (95% CI 2.02 to 14.67; p<0.001) for VO₂ <60%, compared with VO₂ >80%. These findings were consistent across subgroup comparisons.
ORs for other significantly associated parameters are provided in online supplemental appendix 3.
In exploratory unadjusted analyses of the source data, the OR for peak VO₂ ≤80% predicted per one-unit increase in LASI was 1.72 (95% CI 1.03 to 2.87) in women (n=111; 53 events) and 2.37 (95% CI 1.28 to 4.41) in men (n=229; 117 events). The sex-by-LASI interaction was not statistically significant (p=0.431).
ROC analysis identified LASI ≥0.51 as the cut-off with the highest discriminative performance for identifying patients with exercise intolerance (VO₂ ≤80%) (AUC 0.623, sensitivity 0.46, specificity 0.76), slightly outperforming LAVCI (AUC 0.603) and other indices (table 5; online supplemental appendix 4).
In the multivariable analysis (online supplemental appendix 2), which retained only variables significant in both subgroup comparisons (see the Statistical analysis section), LASI ≥0.51 remained independently associated with severe functional limitation (VO₂ <60%; OR 3.29, 95% CI 1.04 to 11.30; p<0.05). Younger age was also an independent correlate, whereas beta-blocker therapy was associated only with exercise intolerance (VO₂ ≤80%).
Intraobserver agreement was excellent across all parameters, with intraclass correlation coefficients (ICC(3,1) of 0.95 for LASI, 0.89 for LACI and 0.89 for LAVCI. Bland-Altman analysis (online supplemental appendices 5 and 6) demonstrated minimal mean bias and narrow 95% limits of agreement, indicating high temporal stability and consistency of repeated measurements by the same observer. Interobserver reproducibility was also good to excellent, with ICC(2,1) values of 0.95 for LASI, 0.90 for LACI and 0.76 for LAVCI.
Corresponding Bland-Altman plots (online supplemental appendices 5 and 6) showed small mean biases without evidence of proportional or systematic bias between observers. Overall, the reproducibility of the three indices was robust, with LASI demonstrating the highest interobserver and intraobserver agreement, while LAVCI exhibited slightly greater inter-observer variability.
Prognosis value of the cut-off in follow-up outcomes
The previously identified LASI threshold of 0.51 was then applied to follow-up outcomes in 311 patients with at least 1 year of follow-up (mean 1121±622 days). Online supplemental appendices 5 and 6 shows the distribution of clinical events according to this threshold.
In this cohort, the overall mortality is 3.8%. There were 76 (24.4%) cases of atrial arrhythmias and thromboembolic events, 58 (18.6%) cases of ventricular arrhythmic events. Additionally, 48 cases (15.4%) of heart failure events, including 5 patients who underwent heart transplantation, 23 patients (7.4%) were referred for septal reduction therapy, mostly alcohol ablation and 12 patients (3.9%) were hospitalised for heart failure.
Peak VO₂ was normalised to values predicted from age, sex and weight, but this adjustment cannot account for every sex-related determinant of exercise performance. Younger age remained associated with severe functional limitation, and beta-blocker use with peak VO₂ ≤80% predicted in the multivariable model. These associations are observational.
Beta-blocker prescription may identify patients with more symptomatic disease; dose, heart-rate response and indication for treatment may also influence the observed relationship. A causal effect of beta-blockers on exercise capacity cannot be inferred.
Figure 4 displays the Kaplan-Meier survival curves for each event category (see the Follow-up outcomes section), stratified by the LASI threshold. Patients were censored after the first occurrence of an event. Those with LASI ≥0.51 experienced significantly more HF events (33.0% vs 5.1%; log-rank p<0.0001) and higher all-cause mortality (9.6% vs 0.5%; p=0.0002).
In contrast, this cut-off did not discriminate the risk of atrial arrhythmia and thromboembolic events (p=0.071) or ventricular arrhythmic events (p=0.61).
Discussion
The principal finding of this study is that progressive atrial mechanical failure, rather than conventional left ventricular morphology, is closely associated with functional limitation and heart failure progression in HCM. Among all echocardiographic parameters evaluated, the LASI emerged as the strongest marker of impaired exercise capacity and independently identified patients at increased risk of heart failure hospitalisation and death. More importantly, our findings suggest that atrial failure is not simply a consequence of HCM but a central biological determinant of its heart failure phenotype.91320–22
Current phenotyping of HCM remains predominantly ventricle-centred.23 Left ventricular hypertrophy, outflow tract obstruction, systolic function and myocardial fibrosis remain fundamental for diagnosis and sudden cardiac death risk stratification, yet they explain only a limited proportion of symptom burden and exercise intolerance. Patients with comparable ventricular morphology frequently exhibit markedly different functional capacity, indicating that ventricular disease alone does not account for clinical progression. Our results support the concept that the left atrium integrates the cumulative haemodynamic consequences of chronic diastolic dysfunction, impaired ventricular compliance, myocardial fibrosis and abnormal atrioventricular interaction.
Progressive atrial remodelling ultimately evolves into atrial mechanical failure, characterised by increased chamber stiffness, impaired reservoir function, reduced contractile reserve and loss of atrioventricular coupling. In this context, atrial failure represents an integrated manifestation of disease progression rather than a passive consequence of ventricular dysfunction.24
Residual confounding from comorbidities, beta-blocker dose and indication, other treatments, and non-cardiac determinants of peak VO₂ remains possible. Exploratory sex-stratified source-data estimates showed no evidence of interaction, but their precision and consistency with the final analytic cohort require confirmation.
This biological model also provides a coherent explanation for exercise intolerance in HCM. During exercise, augmentation of stroke volume depends on efficient atrioventricular coupling despite a hypertrophied and relatively non-compliant ventricle. Progressive atrial failure compromises this adaptive mechanism.
Increased atrial stiffness limits reservoir function, reduced contractile reserve impairs late ventricular filling, and loss of atrioventricular coupling restricts stroke-volume augmentation during exertion. The associated reduction in oxygen pulse, lower peak systolic blood pressure and worsening ventilatory efficiency observed in our cohort are consistent with impaired haemodynamic reserve rather than isolated ventricular dysfunction. Importantly, resting stroke volume showed little relationship with exercise capacity, emphasising that the inability to augment forward cardiac output during exercise—not resting ventricular performance—is a defining characteristic of symptomatic HCM.
These observations resonate strongly with the emerging pathophysiological model of heart failure with preserved ejection fraction (HFpEF).25–28 Over the past decade, HFpEF has increasingly been recognised as a disease of progressive atrial myopathy, in which atrial fibrosis, increased atrial stiffness and impaired atrioventricular coupling contribute directly to exercise intolerance and heart failure progression. In this setting, atrial stiffness has emerged as an integrated marker of chronic filling pressure, atrial fibrosis and cardiovascular reserve, outperforming isolated measurements of atrial size or function. Our findings extend this emerging concept of atrial failure from HFpEF to HCM.
Despite distinct initiating mechanisms—sarcomeric disease in HCM and systemic myocardial dysfunction in HFpEF—both conditions appear to converge toward a common final pathway characterised by atrial mechanical failure, impaired stroke-volume reserve and progressive heart failure. This convergence suggests that atrial failure may represent a fundamental mechanism of heart failure beyond disease-specific phenotypes.25–28
Within this framework, LASI should not be viewed as simply another echocardiographic parameter. Rather, it represents a quantitative expression of atrial mechanical failure. By integrating atrial compliance with ventricular filling pressure, LASI captures the haemodynamic consequences of progressive atrial dysfunction more comprehensively than isolated measurements of atrial size or strain.
A LASI ≥0.51 independently identified patients with severe exercise intolerance and, importantly, predicted heart failure hospitalisation and all-cause mortality, whereas no association was observed with atrial or ventricular arrhythmic events. This distinction is biologically important. If LASI primarily reflected electrical remodelling or atrial enlargement alone, an association with atrial fibrillation would be expected.
Instead, its preferential association with heart failure outcomes indicates that LASI predominantly reflects haemodynamic deterioration and loss of cardiovascular reserve rather than electrical instability.914
These findings have important clinical implications. CPET remains the reference standard for evaluating functional limitation but is not universally available. In contrast, LASI can be obtained from routine echocardiography without additional image acquisition and may identify patients entering the transition from compensated hypertrophy to progressive heart failure before overt clinical deterioration occurs.
More broadly, our findings support a conceptual shift in HCM from a predominantly ventricle-centred disease towards an integrated model in which atrial mechanical failure represents the biological bridge between ventricular pathology and clinical heart failure. This framework may improve disease phenotyping, facilitate earlier identification of patients at risk of heart failure progression, and help select individuals most likely to benefit from therapies targeting diastolic dysfunction or disease-modifying treatments, including cardiac myosin inhibitors.23 Whether serial assessment of LASI can monitor disease progression or therapeutic response now deserves prospective evaluation.
Limitations
This study has several limitations. Its retrospective single-centre design may limit generalisability, although consecutive patient inclusion and blinded echocardiographic analysis reduced selection and measurement bias. Peak oxygen consumption, while the accepted reference standard for functional assessment, may be influenced by extracardiac factors including pulmonary disease, musculoskeletal limitations and patient motivation.
Event rates were relatively low, limiting statistical power for individual clinical endpoints, particularly arrhythmic events. Patients with baseline atrial fibrillation were excluded to ensure reliable assessment of atrial mechanics, potentially underestimating the full spectrum of atrial failure in advanced HCM. CMR data reflected routine clinical practice and did not include standardised quantitative assessment of myocardial fibrosis.
Finally, these findings should be considered hypothesis-generating and require confirmation in prospective multicentre studies.
Conclusions
Progressive atrial mechanical failure appears to be a central determinant of the heart failure phenotype in HCM. LASI, a simple quantitative expression of atrial failure, identifies patients with impaired exercise capacity and adverse heart failure outcomes beyond conventional ventricular assessment. More broadly, our findings suggest that atrial failure may represent a common biological pathway linking ventricular disease to symptomatic heart failure across diverse cardiovascular disorders.
This supports a shift from a ventricle-centred to an atrium-centred approach to HCM phenotyping and provides a new framework for risk stratification and future therapeutic strategies.
Supplementary material