Rate-Adaptive Pacing for Heart Failure With Preserved Ejection Fraction

In this issue of JAMA, Borlaug and colleagues¹ report the results of a important clinical trial, which are likely to alter paradigms and clinical guidelines for heart failure with preserved ejection fraction (HFpEF).¹ The prevalence of this syndrome has progressively increased, and HFpEF has become the most common form of heart failure, particularly in older persons, women, and Black individuals.^2,3 It affects more than 3 million people in the US and is associated with severe symptoms of exertional dyspnea and fatigue (exercise intolerance), impaired health-related quality of life, severely reduced exercise capacity (the primary outcome in the current trial), and increased rates of rehospitalizations and death.^2,3

HFpEF is not only one of the most common cardiovascular disorders but also one of the most poorly understood. Many initial assumptions regarding the pathophysiology of HFpEF have been disproved by careful mechanistic studies.³ Collectively, these studies support that HFpEF is a systemic syndrome, initiated and promoted by undetermined circulating factors but likely related to systemic inflammation arising from excess, maladapted adipose tissue and resulting in widespread mitochondrial dysfunction and loss of tissue capillarity, which are critical to organ perfusion and energy metabolism.² These in turn impair the function not only of the heart but of other organs, such that HFpEF is now viewed as a quintessential systemic, multiorgan disorder.^2,3

Many initial assumptions regarding potential therapies for HFpEF have also been disproved. Nearly all the drugs definitively proven to reduce death and hospitalization in heart failure with reduced ejection fraction have been shown ineffective in HFpEF.²

Fewer clinical trials have focused on the equally important outcome of severely impaired exercise capacity, an objective, clinically meaningful outcome associated with patients’ impaired quality of life and severe symptoms that limit ability to perform and enjoy daily activities.² With rare exception, more than a dozen adequately powered clinical trials using many different drugs have been neutral on this important outcome.^2,3

The notable exception to the litany of neutral clinical trials of HFpEF has been the recent finding that the sodium-glucose cotransporter 2 (SGLT-2) inhibitors improve clinical events in HFpEF and likely increase exercise capacity and qualify of life as well.^4,5 The success of SGLT-2 inhibitors may be due to their broad, pleiotropic, metabolic activity that may uniquely address the systemic, multifactorial, multiorgan nature of HFpEF.²

Exercise capacity is objectively measured as peak oxygen consumption during exercise (peak $\dot{V}{O}_2$).² By the Fick equation, peak $\dot{V}{O}_2$ is the product of cardiac output and arteriovenous oxygen difference. Cardiac output is the product of heart rate and left ventricular stroke volume. The current trial by Borlaug et al¹ is predicated on the observation that patients with HFpEF often have significantly reduced heart rate responses to exercise (chronotropic incompetence), and this is correlated with their reduced peak $\dot{V}{O}_2$. They tested the hypothesis that rate-adaptive atrial pacing would increase heart rate during exercise, thereby increasing exercise cardiac output and improving peak $\dot{V}{O}_2$.¹

The well-designed, well-conducted, randomized clinical trial by Borlaug et al¹ is a monumental achievement. HFpEF trials are notoriously challenging, particularly those using invasive procedures and implanted devices, both of which were used in the present trial. Indeed, 2 other similar trial attempts, including one in the US,⁶ were abandoned. An additional challenge to conducting this type of trial is that current clinical guidelines can be interpreted to support use of rate-adaptive pacing in patients with chronotropic incompetence and symptoms of exercise intolerance, including those with heart failure.

Borlaug et al¹ found that although pacing predictably increased exercise heart rate, there was no improvement in exercise capacity or any other outcome. A likely cause was that the increase in exercise heart rate was counterbalanced by a reduction in exercise stroke volume, such that exercise cardiac output was unchanged. This finding is credible, since in healthy persons and patients with cardiovascular disease, including HFpEF, heart rate and stroke volume are inversely tightly related at rest, and pacing-induced heart rate increases result in inexorable, progressive declines in end-diastolic volume and stroke volume, resulting in unchanged cardiac output.^7,8 In the present trial, it was hoped that the adaptations invoked during exercise (such as autonomic withdrawal, $\beta$-adrenergic drive, and increased venous return) would overcome the dynamic, inverse relationship between heart rate and stroke volume, but such was not the case.¹ A pacing study during exercise in healthy persons found similar results.⁸ The finding of reduced exercise stroke volume that counterbalances increased heart rate resulting in unchanged cardiac output has also been observed with exercise training in patients with HFpEF. Exercise training in these patients produces a large increase in peak $\dot{V}{O}_2$, nearly all of which is due to increased arteriovenous oxygen difference.⁹

The reduced stroke volume may have been due to reduced left ventricular end-diastolic filling pressure (LVEDP). Silverman et al⁷ showed that pacing in at-rest patients with HFpEF strikingly decreases LVEDP, with resultant large declines in end-diastolic volume and stroke volume, and Munch et al⁸ showed the same phenomenon during exercise in healthy persons.⁸ Potentially reduced LVEDP with no change in exercise capacity would challenge the common belief that increased LVEDP limits exercise capacity in patients with HFpEF. Indeed, Sarma et al¹⁰ recently showed that reducing LVEDP with nitroglycerin in patients with HFpEF does not increase exercise capacity. A conclusion that elevated pulmonary capillary wedge pressure may not be the primary cause of exertional symptoms and reduced exercise capacity in HFpEF may seem heretical; however, similar findings were shown decades ago in patients with heart failure with reduced ejection fraction.¹¹

Despite careful patient selection and experienced operators, the pacing intervention was associated with a 20% rate of serious adverse events, including incision site reactions, pericardial effusion requiring drainage, and upper extremity deep vein thrombosis.¹ This complication rate is within the clinically expected range. There were also 5 excess cases of chest pain, which could have been due to higher myocardial oxygen demand from the pacing-driven increased heart rates. Thus, the pacing strategy was not only ineffective but was harmful.

The current report demonstrates how a moderately sized but well-executed trial can provide definitive results that disprove conventional paradigms and should prompt reconsideration of current clinical management guidelines. There are other ongoing pacing trials in HFpEF (NCT04546555; NCT03338374), and a recent trial suggested benefit from a higher backup pacing rate among patients with subclinical or overt HFpEF who already had pacemaker implantation for other indications.¹² However, a recent meta-analysis of pacing for heart failure supports the results of the presently reported trial,¹³ furthering the case for abandoning rate-adaptive pacing for most patients with HFpEF, despite their chronotropic incompetence.

The results of the present trial and the many other neutral trials that have focused primarily on cardiac dysfunction support that a multifactorial, multiorgan systemic syndrome such as HFpEF requires pleiotropic treatment strategies to address the broad deficits that underpin this disorder, including its primary manifestation, exercise intolerance.² The concept that treatment of exercise intolerance in HFpEF should extend beyond a singular focus on cardiac dysfunction is also supported by conclusions from recent National Institutes of Health workshops² and by the successes of pleiotropic interventions, such as exercise training, caloric restriction, and use of SGLT-2 inhbitors.^2–4,14

In contrast to the extensive focus on cardiac dysfunction, relatively little attention has been given to reduced exercise arteriovenous oxygen difference, which accounts for more than 50% of the severely reduced peak $\dot{V}{O}_2$ in patients with HFpEF¹⁵ and is due primarily to their severe skeletal muscle dysfunction.^2,16,17 In contrast to the myocardium, which is terminally differentiated, skeletal muscle has a high capacity for rapid, robust rejuvenation and remodeling.¹⁶ As reviewed elsewhere, a single bout of exercise in sedentary individuals evokes a nearly immediate, large increase in skeletal muscle gene transcription.¹⁶ In addition to exercise and use of SGLT-2 inhibitors, dietary weight loss also produces a robust increase in exercise capacity in patients with obese-metabolic HFpEF phenotype and appears to do so via extracardiac mechanisms.¹⁴ A benefit of improving exercise capacity by increasing arteriovenous oxygen difference is avoidance of the increased myocardial oxygen demand that results from strategies that increase heart rate and stroke volume.

The present trial is an example of how well-conducted trials with negative results can have an impact equal to or perhaps even greater than that of positive studies. Several important lessons arise from this important trial. In addition to potential guidelines revisions, these include embracing the broader pathophysiology of exercise intolerance in HFpEF.² Doing so could provide many novel therapeutic opportunities that could result from targeting the systemic and extracardiac abnormalities that are strong, modifiable contributors to symptoms and reduced exercise capacity in HFpEF, including inflammation; dysfunctional excess adipose tissue; and skeletal muscle, microvascular, and mitochondrial dysfunction.²