Calcification in aortic stenosis

There is growing evidence that the degree of calcification of the aortic valve (AV) is prognostically significant, independent of the degree of imposed haemodynamic load.

The concept that aortic valve calcification (AVC) is an active biological process and that AVC load provides information incremental to haemodynamic measures of aortic stenosis (AS) severity is not completely intuitive, as the haemodynamic load has been considered the ultimate cause of death. Furthermore, calcification is a non-linear process that increases exponentially1 and accelerates disease progression.

However, calcification by echocardiography has been diagnosed on the basis of video qualitative assessment, which grossly correlates with AVC quantified by CT but with wide overlap between echo ‘calcification’ grades.1 Until now, the imprecise AV ‘calcium’ scoring on echocardiography may have impaired the detection of crucial differences between men and women in relatively small observational cohorts. In recent years, multidetector CT, taking advantage of the unequalled value of X-ray calcium assessment, provided definitions of sex-specific severe AVC load in relation to haemodynamic AS severity, with lower thresholds in women.2 3 Previous studies with CT1 linked valve calcification to cardiovascular events following diagnosis, dominated by performance AVAi, and the pitfalls of this endpoint may explain why this measure has not yet gained wide acceptance. Furthermore, different thresholds were used, no sex-specific threshold was defined, and due to the strong association between AVC and valve haemodynamic data, the incremental value of AVC remained unclear.

In their Heart manuscript, Thomassen et al 4 demonstrated that AVC in AS scored by echocardiography has sex-specific risk factors, and that the presence of moderate to severe AVC at baseline predicts higher rates of major cardiovascular events and combined aortic valve replacement (AVR) and death both in women and men.

AVC and AS management

In current US clinical guidelines5 severe or moderate to severe AVC is mentioned as a parameter to be considered in AS management, although there are no recommendations on specific cut-offs or implications. This recommendation is made on the basis of echocardiographic data regarding AS progression and cardiovascular events.

Although symptoms remain the main indication for AVR in patients with AS, the interpretation of symptoms can be challenging, especially in an elderly population. Specifically measuring AVC load may be crucial in patients with discordant Doppler echocardiographic markers of AS severity.

Furthermore, patients with severe AS may have a low gradient if they have a decrease in left ventricular ejection fraction or may paradoxically have a low transvalvular flow rate despite the presence of a preserved ejection fraction (EF). These disease patterns may lead to underestimation of AS severity and thus inappropriate delay of AVR, which may have impact on patient outcome. Although in patients with low EF, dobutamine echocardiography is valuable to detect pseudosevere AS,6 it is unclear how useful the test is in patients with preserved EF and low flow and its utility is even less likely in patients with normal flow. In the context of the uncertainty of haemodynamic status, obtaining an independent and direct measure of AV lesion severity such as AVC load is crucial.

A similar issue occurs in patients with extremes of body habitus. The haemodynamic consequences of AV area may be better reflected by indexing the value to body surface area; whereas an AV area of 1.0 cm² may not be associated with symptoms or left ventricular hypertrophy (LVH) in a small patient, a similar area may functionally compromise a large patient. Moreover, measurement errors can contribute to discrepancies in relationship of AV area and gradients.

AVC and gender

In AS, many studies have reported different remodelling responses between men and women, particularly at older ages, with similar left ventricular mass index but higher prevalence of LVH in women, according to sex-specific criteria. Women have smaller left ventricular (LV) chamber size, higher relative wall thickness, supernormal LV function and less wall stress, while men exhibit ventricular dilatation and earlier ventricular dysfunction. At histological level, some authors have described a higher gene expression of collagens I and III in men, and it has been suggested that oestrogens could have a protective role in the fibrotic response to pressure overload. Given that older patients have relative hypogonadal hormone concentrations, with a decrease in oestrogens and ovarian production of androgens in postmenopausal women, and decrease in androgens in men, it is expected that this protective effect is lost with ageing in women, and this could explain why elderly women have more hypertrophic ventricles. Notably, these differences translate into different outcomes according to sex, where a maladaptive LVH before surgery and the presence of residual LVH after surgery are associated with a worse outcome in women but not in men.

But these sex-specific changes in AS may not be restricted to LV remodelling. It seems that the same process is also taking place at valvular level. In animal models, valvular interstitial cells (VIC) behave differently according to sex. VICs have shown greater proliferation in women but more osteogenic differentiation in men. This may help to explain why men, for the same AS severity, present higher AVC load compared with women on CT evaluation, even after normalisation for body surface area and cross-sectional aortic annulus.2 More recently, Simard et al have confirmed these differences in explanted valves, showing that the there was an increased amount of non-calcified tissue in women in multidetector computed tomography, which correlated in histology with increased levels of fibrosis in stenotic valves from women relative to men.7

This substudy of -Intensive Lipid Lowering with Simvastatin and Ezetimibe in Aortic Stenosis (SEAS trial) confirms these CT findings in a large AS population, now using echocardiography, a cheap and widely available imaging method. Again, one can speculate that oestrogens may be the missing link in this story. Pvull polymorphisms in the oestrogen receptor alpha gene are related to both the presence of AS in postmenopausal women and to lipid levels in adolescent women, suggesting that this polymorphism may influence the risk of AS by affecting gender and lipid levels.8

AVC and prognosis in AS

In this study by Thomassen et al, AVC load evaluated by echocardiography has shown prognostic value both in men and women, reinforcing previous findings from a multicentre registry that found an excess mortality in patients with severe AVC, independent of clinical and echocardiographic parameters.9 There were, however, sex differences, with higher hazard rate of ischaemic cardiovascular events only in women, and higher all-cause mortality only in men. The meaning of these differences is uncertain. The relationship between AV sclerosis and cardiovascular events is well established but the possible explanation for a higher all-cause mortality in men is elusive. One can speculate that it can be due to higher inflammatory response in men, which may have contributed to higher non-cardiovascular mortality. In this analysis, there was an independent association between the systemic inflammatory marker high-sensitivity C-reactive protein and AVC only in men. This could be related to other emerging risk factors in AS, like epicardial adipose tissue (EAT), which seems to contribute to the inflammatory and profibrotic response to load and has been associated with LVH. Considering that EAT increases with ageing and that EAT volume index has also been independently associated with frailty in a small observational study of patients with severe AS, we have to raise the question if it could also contribute for the higher inflammation and worse prognosis in AS, although the SEAS trial included a younger and selected population of patients compared with the ‘real world’ clinical practice. New research is also needed in this field.

Clinical implications

The findings from this SEAS substudy underline the important pathophysiological differences between women and men and strongly support the development of future research in disease mechanisms specific to each sex and potential individualised interventions. But the question remains: Are we ready to introduce AVC in our recommendation for evaluation and management of patients in AS?