Left Atrial Dysfunction: The Next Key Target in Heart Failure with Preserved Ejection Fraction

This article refers to ‘Impaired left atrial strain predicts abnormal exercise haemodynamics in heart failure with preserved ejection fraction’ by F. Telles et al., published in this issue on pages 495–505. Left atrial (LA) remodelling and dysfunction are common in patients with heart failure (HF) with preserved ejection fraction (HFpEF).1-4 The distending pressure of the left atrium is principally determined by left ventricular (LV) filling pressure during diastole, and therefore LA remodelling has been considered to be a secondary consequence of LV diastolic dysfunction. However, recent data have shown that LA dysfunction, rather than simply remodelling, may be more than a bystander, but rather an important player in the pathophysiology of HFpEF. Recent studies have shown that LA dysfunction is associated with worse symptoms, more severe pulmonary vascular disease, greater right ventricular (RV) dysfunction, depressed exercise capacity, and increased mortality in HFpEF.1-6 Left atrial function may be assessed by conventional two-dimensional and Doppler echocardiography, but recent studies have shown potentially greater utility of LA deformation analysis using speckle-tracking echocardiography.2-7 Abnormalities in LA mechanics are correlated with resting haemodynamics and clinical status, but it is unknown whether deranged LA mechanics can help identify patients with more profound haemodynamic responses to exercise in HFpEF. In this issue of the Journal, Telles et al.8 present intriguing data that shed new light on the contribution of LA mechanics to abnormal exercise haemodynamics in patients with HFpEF. The investigators prospectively examined 71 subjects with ejection fraction ≥ 50% referred for exercise right heart catheterization because of exertional dyspnoea. Participants were diagnosed with HFpEF [pulmonary capillary wedge pressure (PCWP) ≥ 15 mmHg at rest and/or ≥ 25 mmHg with exercise) or non-cardiac dyspnoea (NCD) (PCWP < 15 mmHg at rest and < 25 mmHg with exercise) according to contemporary standards.9 Two-dimensional echocardiography was performed to evaluate resting LA mechanical properties just prior to the invasive measurements. Phases of LA function were assessed by volumetric measurements and speckle tracking-derived longitudinal strain and strain rate. LA stiffness was then calculated as the ratio of PCWP or E/e′ to LA reservoir strain.8 Consistent with prior studies,1-7 patients with HFpEF displayed reduced LA reservoir and pump function assessed by longitudinal strain, strain rate and LA ejection fraction, with increased LA stiffness as compared to NCD.8 LA conduit function was similar in the HFpEF and NCD groups. An extremely important observation was that neither LA volume nor LA volume index differed between groups, meaning that LA function was the greater correlate of abnormal haemodynamics rather than structure. Among patients with HFpEF, increasing burden of atrial fibrillation (AF) was associated with greater LA mechanical impairment.8 As compared with NCD subjects, HFpEF patients displayed higher left and right-sided filling pressures with higher pulmonary artery (PA) pressure at rest and with exercise, and lower cardiac output during exercise, in agreement with earlier studies evaluating exercise haemodynamics.9-11 The abnormalities in LA reservoir and pump strain assessed prior to exercise were strongly correlated with left heart filling pressure, PA pressure, PA elastance, and cardiac index at rest and, remarkably, during peak exercise. LA strain was also related to abnormal haemodynamic reserve responses with exercise, indicating that an abnormality in resting function could identify dynamic responses to stress. The authors importantly went on to show that the relationships between LA reservoir strain and exercise haemodynamics remained significant even after the exclusion of AF patients.8 Telles and colleagues then went on to evaluate the potential diagnostic ability of LA strain to discriminate HFpEF patients from NCD.8 In patients with B-type natriuretic peptide measurements (n = 63), LA reservoir strain showed a good diagnostic accuracy (area under the curve 0.76), and LA reservoir strain ≤ 33% demonstrated reasonable sensitivity (86%) and specificity (76%) to diagnose HFpEF, with significant net reclassification improvement in comparison to the 2016 European Society of Cardiology criteria.8 The investigators are to be commended on this important contribution, which confirms and extends upon earlier studies while providing important new insights regarding the role of LA dysfunction for understanding pathophysiology, and potentially, diagnosis in HFpEF.8 Why does the left atrium play such a key role in maintaining optimal cardiac performance in patients with HFpEF? The left atrium can be perceived as having two key duties: (i) it modulates LV filling through its reservoir, conduit, and pump functions to maintain LV preload and cardiac output, and (ii) it helps to keep the lungs free of congestion through its compliance properties to absorb venous return from the lungs without untoward increases in pressure. As LA function deteriorates, the organs upstream of the left heart are compromised, leading to worsening pulmonary capillary hypertension, vascular remodelling, pulmonary arterial hypertension, and RV dysfunction, which then contribute to even greater morbidity and mortality in HFpEF1-6 (Figure 1). This suggests that LA dysfunction may represent an important target for therapy, and that patients with HFpEF and greater burden of LA dysfunction might be viewed as part of a different sub-phenotype within the broader spectrum of patients, that is if a treatment could be applied targeting LA dysfunction.12 Left atrial dysfunction could be targeted in three fundamental ways. Reduction of LV filling pressures (through a variety of means) will lower LA afterload and can improve LA structure and function in a manner analogous to systemic afterload reduction. Alternatively, agents that directly target myocardial structure and function in the left atrium may be also useful, such as neurohormonal antagonists.13 Finally, maintenance or restoration of AF may also be important. A substantial proportion of patients with HFpEF develop AF, and they suffer from more LA dysfunction, worse functional capacity, RV dysfunction, and increased risk of death.12, 14 Catheter ablation could be effective to reverse or at least prevent the serious consequences of AF in HFrEF,15 but this has yet to be tested in HFpEF. Echocardiography plays a central role in the diagnosis of HFpEF.12 However, it has become clear that currently used echocardiographic parameters such as E/e', LA volume index, and LV hypertrophy are insensitive to identify HFpEF, especially when patients display no overt congestion at rest.9, 12 Even if LA volumes were similar between HFpEF and NCD in this study, LA function was markedly impaired as assessed by strain parameters, with incremental diagnostic value beyond current guidelines.8 LA reservoir strain ≤ 33% could identify HFpEF with reasonably high sensitivity (82%), but the cut-off used by the authors is based on previous reports of risk stratification and normal distribution, but not on diagnosis, and it seems to fall within normal range according to a recent European multicentre study (the lowest limits of normality: 26.1%).16 Like LV strain, there can be vendor variability of LA strain, which makes it more difficult to standardize absolute values. Other concerns include how LA strain is influenced by the presence of concomitant diseases such as mitral valve disease, prior history of ablation, or AF. Further study is required to establish optimal cut-offs for LA strain to distinguish HFpEF from unexplained dyspnoea. Ideally, such studies would involve LA strain assessment with the same vendor and definitive ascertainment of HFpEF by the invasive exercise haemodynamic test. In summary, Telles et al.8 have provided exciting new data that again emphasize the importance of LA dysfunction in HFpEF. By confirming the associations between LA dysfunction and haemodynamic derangements, the current data suggest a potential target for treatment in HFpEF, a syndrome for which there is no proven therapy. Further study is needed to determine how best to identify LA dysfunction, and how to optimally treat it to improve clinical outcomes. Barry A. Borlaug is supported by R01 HL128526, R01 HL 126638, U01 HL125205 and U10 HL110262, all from the National Institute of Health. Masaru Obokata is supported by a research fellowship from the Uehara Memorial Foundation, Japan. Conflict of interest: none declared.