In search for a mechanism of poor outcome in patients with aortic valve sclerosis without aortic stenosis and left ventricular hypertrophy: results of a populational observational study with normal left ventricular geometry

This editorial refers to ‘Aortic valve sclerosis and subclinical LV dysfunction in the general population with normal LV geometry’, by Y. Yoshida et al., https://doi.org/10.1093/eurjpc/zwac279. Aortic valve stenosis (AS) is the most common degenerative valvular heart disease in the Western societies. Its prevalence increases with age with moderate-to-severe AS set to nearly triple by 2060.1 Most of the patients are and will be undiagnosed1 until symptomatic deterioration, unless populational screening programmes will be launched worldwide. Irrespective of its grade, AS is an independent predictor of all-cause and cardiovascular mortality compared with patients with no AS. The more severe the AS, the worse the outcome.2 The effect of severe AS on myocardial structure and function had been widely reported.3 The impact of the increased afterload on the left ventricular (LV) remodelling, e.g. increased LV mass, LV relative wall thickening, concentric Left ventricular hypertrophy (LVH), diffuse myocardial fibrosis,3,4 and LV systolic dysfunction (left ventricular ejection fraction [LVEF] < 55%), had been reported and is associated with poor outcome.5 The largest epidemiological study in AS using the Australian database (NADA) investigated the prevalence of all forms of aortic stenosis in 241 303 unselected participants and found that 6.7% of individuals exhibited mild aortic stenosis and had an ∼1.5-fold increase in risk for all-cause mortality.2 Previous studies suggested that the aortic valve sclerosis (AVS) is associated with worse outcome, independent of the traditional cardiovascular risk factors.6,7 To investigate the possible mechanism of poor outcome of early degenerative changes of the aortic valve, at the stage of AVS defined by fibrotic valve thickening or spot of calcification on the leaflet without impact on valvular mobility, is an interesting concept addressed in the recent publication by Yoshida and colleagues in the journal.8 They have recruited 962 patients (age range 50–75) with no AS (defined as peak velocity <2.6 m/s) from the Subclinical Cardiac Dysfunction in General Population database—from the region of Tokyo, Japan, as part of the promotion of health and prevention of cardiovascular disease from 2014 to 2018.9 Patients in this study were screened for cardiovascular risk factors and baseline anthropometric variables and underwent detailed transthoracic echocardiography and blood tests including cardiac chamber quantification, Doppler measurements including mitral valve annular velocities, and speckle tracking echocardiography as well as fasting serum glucose, total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, C-reactive protein, and B-type natriuretic peptide (BNP). Careful attention was paid to assess the aortic valve structure and to detect thickening, fibrotic changes, and calcific deposits on the cusps from the parasternal long- and short-axis views. Based on the aortic valve appearance, they have divided patients into four groups: (i) no valve disease, (ii) fibrotic thickening only, (iii) spot of calcium on a single cusp, and (iv) spot of calcium on more than two cusps of the aortic valve. For this analysis, they have excluded patients with known coronary artery disease, atrial flutter or fibrillation bicuspid aortic valve and LVEF <50%, and significant mitral valve disease, as well as concentric or eccentric LV remodelling. The prevalence of AVS was 20% (195/962). Severity, expressed as valve thickening (74/962), single leaflet calcification (87/962), or more than two leaflet calcifications (34/962), was associated with older age, diabetes mellitus, hypercholesterolaemia, higher systolic blood pressure, and serum BNP compared with the group with no AVS. With the more advanced AVS features, patients had higher LV mass index, peak aortic jet velocity, higher A and lower e′ hence lower E/A and higher E/e′ ratio, and lower LV global longitudinal strain (LVGLS) although all mean values were within the normal range (representative images in Figure 1).8 Using the cut-point for LV diastolic dysfunction as E/e′ ≥ 13 and systolic dysfunction as LVGLS < 17% for men and <18% for women (as absolute value), they found a greater prevalence of subclinical LV diastolic dysfunction of the AV sclerotic groups with the highest in the more extensive calcification group compared with the no AVS group, whereas the subclinical systolic dysfunction (reduced LVGLS) was more prevalent only in the most heavily calcified AVS group compared with the other groups. Interestingly, after adjustment for variables such as age, sex, hypertension, diabetes mellitus, hypercholesterolaemia, smoking status, and body mass index, subclinical LV diastolic dysfunction was independently associated with all three forms of AVS with an odds ratio of 7.57 (3.10–18.46) for the most calcific AV leaflet group. The association remained only significant for subclinical LV systolic dysfunction for the group with more extensive calcific deposit. Representative images of each AVS group.8 A, late diastolic transmitral flow velocity; AVS, aortic valve sclerosis; E, early diastolic transmitral flow velocity; e′, early diastolic mitral annular velocity; LV, left ventricle. This populational observational study is the first to raise the hypothesis that subclinical changes of LV function can be present at an early stage of AVS. Longitudinal follow-up of developing progressive AV stenosis and its clinical outcome would be essential to implement this observation into the daily clinical practice as a marker of disease progression and influence management of such patients. Since the association of AVS and LV dysfunction in this study was also independent of traditional cardiovascular risk factors, to find the appropriate target for intervention would be an upcoming challenge. There is a need to further explore mechanisms leading to such subclinical LV dysfunction for early intervention. Such unexplored mechanisms could be e.g. valvular or periaortic inflammation that may also affect myocardial function. Indeed, AVS bares the similar pathomechanism as atherosclerosis via endothelial damage and inflammatory processes involving macrophage mast cells and T lymphocytes.10,11 If causality were to be confirmed, novel technologies addressing anti-inflammatory pathways could be called upon. Echocardiographic parameters assessing diastolic and systolic dysfunction have their limitations especially within the near-normal ranges as acknowledged by the authors. Second major shortcoming of this study is due to the limitation by echocardiography related to its resolution and gain settings to classify AVS. Of course, the most accurate way to define AVS would be computed tomography (CT)-based imaging that may not be feasible on a populational basis. However, collaborative assessment of the aortic valve from databases where patients were recruited for atypical chest pain assessment with normal coronary arteries and without LVH could support the current findings. Attila Kardos, (MD. PhD. FRCP. FESC.) None declared.