The interplay between functional mitral regurgitation and left atrial function

This article refers to ‘Quantified mitral regurgitation and left atrial function in heart failure with reduced ejection fraction: interplay and outcome implications’ by A. Malagoli et al., published in this issue on pages 694–702. In this issue of the Journal, Malagoli and colleagues report the results of a prospective study exploring the association of functional mitral regurgitation (FMR), and left atrial (LA) function, with cardiovascular outcomes among patients with heart failure with reduced ejection fraction (HFrEF).1 Furthermore, the authors establish an interesting interplay between FMR and LA function, and show that up to a certain degree, the left atrium may ‘accommodate’ FMR. To do that, they enrolled 286 patients with chronic HFrEF, and followed them for up to 5 years. The authors included only patients in sinus rhythm that did not have other significant valvular disease. FMR was quantitatively measured (using the effective regurgitant orifice area [EROA] method) and speckle strain echocardiography was used to evaluate LA function. The study shows that (i) even mild degree of FMR is associated with increased heart failure hospitalizations and mortality, with excess event risk starting at about the EROA value of 0.1 cm2; (ii) an EROA cutoff of 0.3 cm2 further increases cardiovascular outcomes, and, most interestingly, (iii) within a moderate EROA range, LA function may mitigate the clinical consequences of FMR. In fact, in patients with EROA between 0.10 cm2, and 0.30 cm2, dismal outcome was seen only in patients with reduced LA function (peak atrial longitudinal strain [PALS] <14%). The question of whether FMR or severe FMR is the cause, or should be considered just as a surrogate marker of adverse outcome is the focus of an ongoing debate. An EROA cutoff of ≥0.2 cm2 of FMR was suggested following an early study demonstrating that this cutoff was associated with poorer prognosis.2 Data that support the clinical relevance of FMR per se were driven from several retrospective studies demonstrating that the presence of FMR of any degree was associated with increased morbidity and mortality.3, 4 However, presenting FMR as a cause for adverse outcome was complicated because the presence of FMR was usually correlated with worse left ventricular (LV) dysfunction, implying that the left ventricle was the main driver for these results.5 Furthermore, conflicting data were presented as to the means by which FMR should be treated.6-9 A signal pointing towards a potential beneficial intervention came from a sub-study of the STICH trial. In this sub-study, mitral regurgitation (MR) severity along three strata (i.e. none/trace, mild and moderate to severe) was associated with increased mortality. Furthermore, the data showed that compared with either medical therapy or coronary artery bypass surgery, the addition of mitral valve repair was associated with improved survival.10 These initial results led to the surgical treatment of moderate ischaemic MR which showed just the opposite, no survival benefit in adding mitral valve repair to bypass surgery in coronary artery disease patients.6 Of note, though this study was randomized, participants had lower prevalence of heart failure, and milder MR compared with those recruited to the STICH trial. As knowledge regarding the unique mechanism of FMR accumulated, different cutoff values for FMR degrees were suggested.11, 12 Established quantitative measures of MR (e.g., regurgitant volume and fraction) were shown to underestimate FMR severity due to diminished stroke volume. Furthermore, the regurgitant orifice was shown to be not circular, affecting vena contracta measurement and, consequently, the size of EROA.13 Put together, these findings produced inconsistencies with the definition of FMR severity, the appropriate treatment and required strengthening of the data, with special emphasis on other echocardiographic measures to identify a heart failure phenotype where FMR is most likely a driving force and not a mere bystander of the disease. A previous attempt to assess the prognostic impact of FMR, in the context of other markers, was published by Goliasch and colleagues.14 They demonstrated that severe FMR had a prognostic impact mostly in patients with non-severe heart failure phenotype; meaning in those with smaller left ventricles and moderate LV dysfunction, but not in those with severely enlarged left ventricles and severely reduced LV function. Similarly, severe FMR did not have a prognostic effect in those with severe tricuspid regurgitation and significantly elevated natriuretic peptide levels. Another interestin1 g finding in this study, corresponding with the observations made by Malagoli et al., was the interplay between LA size and FMR. Goliasch and colleagues reported that severe FMR was associated with worse outcomes in those with smaller left atria. Nevertheless, the role of the left atrium in mitigating FMR is still an area for debate. For example, Carluccio and colleagues suggested that an increased, and not decreased, LA volume was found to be a marker for poor prognosis.15 Preliminary studies assessing the role of LA function usually required the use of magnetic resonance imaging.16, 17 Therefore, the emergence of LA reservoir speckle strain echocardiography, a reproducible simple measure which is less affected by loading conditions and can be implemented with the use of echocardiography alone, is promising. In a comprehensive echocardiographic study, researchers in Italy examined the association of several LA function parameters with outcomes in HFrEF patients and showed that LA reservoir function (as presented by PALS) was associated with reduced LV function and worse outcomes.18 This group suggested a cutoff value of <15% as a marker for increased risk. Soon thereafter, Malagoli and his group reported similar findings and further explored the potential association between LA function and FMR.19 Their preliminary results showed that high LA reservoir protects against the detrimental effect of FMR in HFrEF patients. However, in their current recent analysis, the group laid another piece in the structure of the association between LA function and FMR. They demonstrated that the protective effect of PALS against FMR is true only to a certain degree. Here, both FMR and low PALS were found to be independent predictors of morbidity and mortality in HFrEF.1 Furthermore, though severe FMR (i.e. EROA > 0.3 cm2) was shown to be associated with increased cardiovascular outcomes regardless of PALS values, high PALS was found to have a protective effect only against lower grades of FMR. The potential mechanism behind this effect could be attributed to the interplay between the left ventricle and the left atrium during the cardiac cycle.20 The constant volume within the heart is preserved through the reciprocal emptying and filling of the atria and ventricle, mostly through longitudinal stretching of both compartments. Nevertheless, atrial longitudinal strain maintain its prognostic effect due to its association with atrial compliance. Improved atrial compliance may allow the left atrium to mitigate increased LV filling pressures, and apparently also the detdrimental haemodynamic effects of FMR. Nevertheless, the present study has several limitations. First, the authors excluded patients with other valvular diseases and, importantly, patients with atrial fibrillation, factors which limit their findings in many patients encountered in clinical practice. Second, the authors did not provide indisputable proof that LA compliance is reduced in patients with reduced PALS because they did not use invasive haemodynamic measures. Third, the optimal echocardiographic surrogate marker of atrial compliance is PALS and E/e′, and not PALS alone.21 Fourth, they do not provide data on ventricular longitudinal function that can be associated with atrial PALS. Nevertheless, this new exciting study sheds new light and helps better refine the population of HFrEF patients with FMR and potentially present another explanation for the conflicting results of major interventional trials conducted in this field. Conflict of interest: none declared.