Jumping Down the Rabbit Hole: Unravelling the Right Ventricle in Heart Failure

This article refers to ‘Right ventricular dysfunction in left-sided heart failure with preserved versus reduced ejection fraction’ by L. Bosch et al., published in this issue on pages 1664–1671. Clinicians recognize that right ventricular dysfunction (RVD) often complicates the management of heart failure and when present it can profoundly affect patient outcomes.1, 2 Until recently, however, there has been scant attention paid to the pathogenesis of RVD. In this issue of the European Journal of Heart Failure, Bosch and colleagues provide information that adds to the knowledge base about the prevalence of RVD in heart failure patients and increases our understanding of its prognostic significance.3 They also offer some provocative insights into why RVD develops in this population. The patients described in this report were identified prospectively in a multi-ethnic Asian population based cohort study in Singapore that includes both heart failure with preserved ejection fraction (HFpEF) and heart failure with reduced ejection fraction (HFrEF) patients.4 A population of age- and sex-matched control subjects without obvious cardiovascular disease was included for comparison. Using a variety of echocardiographic measurements including right ventricular longitudinal strain (RVLS) and tricuspid annular plane systolic excursion (TAPSE) to assess right ventricular (RV) function, the investigators detected RVD in >60% of the HFrEF and 30–40% of the HFpEF patients. Previous studies have reported differing estimates of the prevalence of RVD in various populations. In patients with HFpEF for instance, the prevalence of RVD has ranged from 4% to 48%.5 The investigators also found that as in other cohorts of heart failure patients6 pulmonary hypertension was common in this population. Overall, there was a stepwise increase in pulmonary artery systolic pressure (PASP) from control subjects to HFpEF patients to HFrEF patients with pulmonary hypertension, defined as PASP >35 mmHg, being found in 39% of the HFpEF and 52% of the HFrEF patients. Differences in underlying cardiovascular diseases, duration and severity of heart failure, criteria for defining HFpEF and in assessing RVD all likely have contributed to the differences in prevalence of RVD and pulmonary hypertension reported in the medical literature. Nonetheless, both RVD and pulmonary hypertension were clearly quite common in this Asian population and the prevalence of each of them was generally consistent with that seen in patients from the United States and Europe. A strength of the present report is that it includes both HFrEF and HFpEF patients from the same population and the findings provide information about the prevalence of RVD across the heart failure continuum in a well-defined and carefully studied cohort of Asian patients. The investigators' decision to report several measures of RVD is also helpful, as the complex anatomy of the right ventricle does not lend itself to the use of a single measure to describe its function. Both the HFpEF and HFrEF patients followed in this study were at increased risk for events compared to the control population. Over a median follow-up period of 715 days, HFrEF patients had a slightly higher likelihood of experiencing either all-cause death or heart failure hospitalization than did HFpEF patients (44.7% vs. 34.7%). Both PASP and RVLS (but not TAPSE) were related to outcomes in adjusted analysis. The results of the present study confirm those previously reported showing that elevations in pulmonary artery pressures are related to prognosis in both HFpEF and HFrEF patients.6, 7 Whether or not RVD is an independent predictor of the clinical course of heart failure patients, however, has been less certain. The right ventricle is a thin walled structure that is highly sensitive to changes in afterload. In healthy individuals, the right ventricle operates at relatively low levels of pressure. As left ventricular dysfunction develops, filling pressures rise and are transmitted back through the pulmonary capillary bed to the pulmonary artery. The resultant increase in load imposed on the right ventricle, particularly when sustained over time, predisposes to the development of RVD. This scenario raises the question of whether the relationship between RVD and prognosis in heart failure is an epiphenomenon with causality due primarily to the adverse effects of elevated pulmonary artery pressure. Sorting this out has potentially important clinical implications as the direction of new therapeutic strategies depends to a certain extent on whether RVD or pulmonary hypertension drive poor outcomes in heart failure. To better assess the independent role of RVD in predicting prognosis, the investigators normalized values for RVLS and TAPSE, the two key measures of RV function in this study, for PASP. Their results show that even after adjusting for pulmonary artery pressures, both measures of RV function remained significantly associated with patient outcomes. Thus, RVD was both common in this heart failure population and an independent predictor of the clinical course in both HFrEF and HFpEF patients. This report adds to a growing body of literature demonstrating that RVD is an independent predictor of outcomes in heart failure2, 5, 8 and it should serve to further focus attention on the right ventricle as a target for future therapeutic interventions. Bosch and colleagues also provide some clues about the development of RVD in their patients. Numerous factors influence RV function in heart failure including conditions that affect both ventricles, those that affect the right ventricle alone and those that influence RV function by virtue of their effects on pulmonary artery pressures. The observation that RVD is related to left ventricular function in both HFpEF as well as HFrEF patients in this study is consistent with the possibility that disease processes affecting both ventricles played an important role in the development of RVD. Anatomical support for this comes from the fact that the right and left ventricles share a common wall in the interventricular septum and by the presence of transverse muscle bundles that are part of the complex interweave that make up the walls of both ventricles.9, 10 Exposure of the right ventricle to many of the same neurohormonal and inflammatory mediators that are known to adversely affect left ventricular structure and function further supports biventricular involvement. When stressed, the right ventricle also appears to be more vulnerable than the left ventricle to the deleterious effects of a variety of factors involved in the pathogenesis of heart failure, with greater susceptibility to oxidative stress, a reduced angiogenic response and greater likelihood of activating cell death pathways than the left ventricle.11 Determining the role of pulmonary artery pressures in causing RVD is not a straightforward exercise. Although the investigators found that RVD was not significantly correlated with greater PASP, these findings should not be interpreted as showing that the development of RVD was entirely free from the influence of the pressure load imposed on it. As noted above, RV function is quite sensitive to the ambient pressure in the pulmonary arterial bed. Assessing RV function as a function of PASP at a single point in time with patients at rest provides only a limited view of dynamic interactions between PASP and RV function that occur over time. The patients included in the report had at least stage C heart failure and elevations in PASP would have started in many cases well before enrollment into the present study. The results presented here do not take into account the possibility that RV function was more strongly influenced by pulmonary artery pressures during earlier stages of disease and that the association between greater PASP and RVD weakened as RV function progressively deteriorated and RV stroke volume was reduced. In addition, measurements obtained with the patient at rest do not recognize dynamic changes in pulmonary artery pressures that occur throughout the course of a patient's daily activities. Both exercise and conditions that result in hypoxia (including sleep apnoea which is common in heart failure patients) can increase pulmonary artery pressures12, 13 and their impact on RV function would not be captured when the patient is only studied at rest. Finally, the investigators assessed RV to pulmonary artery coupling by employing the ratio of either TAPSE or RVLS to PASP. Although previous studies have shown a highly significant correlation between TAPSE/PASP and RV systolic elastance/arterial elastance, the confidence intervals for this relationship were notably broad introducing a degree of uncertainty into the use echo values alone.8 The findings reported in this study clearly show that RVD is common in both HFrEF and HFpEF patients and that when present is an independent predictor of outcomes. They also provide impetus for further study of how RVD develops in heart failure in order to better understand the mechanisms involved. Addressing RVD has emerged as an important focus of therapeutic strategies.14 The observations made by Bosch and colleagues in this study emphasize the point that greater attention needs to be paid to RV function in the future and they suggest that approaches targeting processes that are common to both ventricles may be more likely to achieve success in improving outcomes than treatments aimed at reducing pulmonary artery pressures alone. Interestingly, the importance of RVD has been identified in patients with advanced left ventricular function receiving mechanical circulatory support either as bridge to transplant or as destination therapy.15 In this population, the presence of RV failure is a major therapeutic challenge. While that challenge may not be so acute or great in patients with chronic heart failure, the findings of Bosch and colleagues suggest that finding effective therapies for treating RVD may lead to better outcomes in both the HFpEF and HFrEF populations. Conflict of interest: none declared.