Treatment of Heart Failure with Preserved Ejection Fraction

Development of Treatment Strategies Based on The Pathophysiology of Heart Failure with Preserved Ejection Fraction

The development of treatment strategies for HFpEF is based on our evolving understanding of the pathophysiology of this condition (reviewed in detail in Chapter 11 ). HFpEF commonly afflicts elderly patients with comorbidities of hypertension, left ventricular (LV) hypertrophy, diabetes mellitus, myocardial ischemia, and obesity. Of these risk factors, hypertension and subsequent concentric LV hypertrophy are the most prevalent and highly associated with HFpEF. Less commonly, HFpEF may occur as a result of restrictive and infiltrative cardiomyopathies and transplant rejection. In the presence of the aforementioned conditions, clinical symptoms and signs of HF are commonly precipitated by concomitant anemia, pulmonary disease, renal insufficiency, atrial fibrillation, infection, and uncontrolled hypertension. In patients with HFpEF, in the absence of significant valvular or pericardial disease, diastolic dysfunction consisting of abnormalities of LV relaxation and increased LV stiffness have long been thought to be the central pathophysiologic abnormality contributing to the development of HF. This led to the term diastolic HF to describe this condition. Even though mechanistic studies demonstrate that abnormalities of diastolic function are invariably present in HFpEF, there is some disagreement concerning the relative contribution of diastolic dysfunction to clinical HF in an elderly comorbid population in whom there is a high prevalence of diastolic dysfunction even in the absence of clinical HF. It is likely that these elderly patients frequently experience clinical decompensation when the precipitants listed previously occur in the presence of the underlying substrate of diastolic dysfunction. Decompensated HF would be unlikely when the same precipitants occur in patients without underlying diastolic dysfunction.

Furthermore, factors other than diastolic dysfunction have also been suggested as contributors to the development of HFpEF and include both central (cardiac) and peripheral (skeletal muscle and vasculature) abnormalities. These include increased vascular and LV systolic stiffness, early systolic dysfunction, endothelial dysfunction, volume overload secondary to renal disease with abnormal renal sodium handling, atrial dysfunction, neurohumoral (specifically renin-angiotensin-aldosterone system [RAAS]) activation, reduced vasodilator reserve, chronotropic incompetence during exercise, and impaired right ventricular-pulmonary artery coupling and have all been related to the HFpEF syndrome. Cardiometabolic diseases, including obesity, hypertension, and diabetes are thought to induce a systemic proinflammatory state, which in turn may trigger systemic and coronary microvascular inflammation. This results in reduction of nitric oxide (NO) bioavailability downstream, which promotes myocyte and myocardial hypertrophy, cardiomyocyte stiffness, and interstitial fibrosis. Aging is associated with a reduction in the elastic properties of the heart and vasculature associated with an increase in systolic blood pressure. Aging and the higher prevalence of comorbidities likely contribute to the much higher prevalence of HFpEF in the elderly.

The following sections in this chapter will provide an overview of therapeutic modalities that could be effective in patients with HFpEF, based either on symptomatic benefit or on targeting pathophysiologic mechanisms. The clinical approach to management of these patients with HFpEF will then be summarized based on current evidence and consensus opinion.

Treatment of Volume Overload and Congestion

In HFpEF, diuretics reduce ventricular filling pressures and are therefore useful in providing symptomatic benefit in patients presenting with pulmonary vascular congestion and peripheral edema. It should be noted that some patients who fit the classic profile of “diastolic heart failure” with significant concentric left hypertrophy and small LV volumes may exhibit a fall in cardiac output with rapid diuresis, resulting in hypotension and prerenal azotemia. This results from the fact that these patients have a very steep LV diastolic pressure-volume curve such that a small change in diastolic volume causes a large change in pressure and cardiac output. In addition, right ventricular characteristics may affect this relationship. In one small study of HFpEF patients, worsening renal function during acute HF hospitalization was associated with reduced right ventricular function and increased right ventricular free wall thickness compared with matched patients with no worsening renal function. It is known that the right ventricle forms the external pressure for approximately one-third of the surface area of the left ventricle. Therefore elevation of right-heart diastolic pressures can constrain the filling of the left ventricle. In some patients, reduction of right-sided diastolic pressures by diuretics may unload the interventricular septum, improving LV distensibility, and may therefore be associated with a reduction in pulmonary venous pressures while maintaining LV filling and cardiac output. More recently, Maurer and colleagues have demonstrated that in a subgroup of patients with hypertensive HFpEF, the LV end-diastolic pressure-volume relationship may be shifted rightward with somewhat increased end-diastolic volumes (in contrast to the classic paradigm of diastolic HF with a leftward- and upward-shifted end-diastolic pressure volume relationship and smaller LV volume). This may be a result of a volume overload state contributed to by extracardiac factors such as renal dysfunction with abnormal renal sodium handling, obesity, and anemia. This could also represent a group of patients who would also respond more favorably to diuretic therapy. In addition, low-dose diuretics, especially thiazide diuretics, are useful in the treatment of hypertension, a key pathophysiologic factor in HFpEF. Some patients with HFpEF and severe volume overload may also be candidates for ultrafiltration (discussed in Chapter 36 ).

Furthermore, renal dysfunction is a key pathogenic process leading to HFpEF. Abnormal renal function is also likely required to produce salt-sensitive hypertension, seen very frequently in HFpEF in older patients. Inappropriate salt and water retention with sodium and volume overload contribute to renal and cardiovascular (CV) dysfunction in a vicious cycle. In animal models, a high-salt diet results in hypertension, severe ventricular hypertrophy, and cardiac fibrosis, as well as proteinuria, glomerulosclerosis, and renal inflammation. Importantly, these effects were attenuated with the use of a thiazide diuretic. Fluid overload may be even more important than the hypertension, in that in patients with chronic kidney disease, CV events are more likely in those with modest degrees of fluid overload, with or without concomitant hypertension. Therefore adequate treatment of patients with volume-optimizing strategies may offer more than just symptomatic benefit.

Nitrates and Nitrites

Therapy with nitrates may provide symptomatic benefit in patients with HFpEF with pulmonary vascular congestion. Nitrates are primarily venodilators, with some arterial vasodilating action. They may benefit patients with HFpEF by reducing preload, thus leading to a reduction in ventricular filling pressures and pulmonary congestion. In acute decompensated HF, they can be used intravenously and may improve symptoms by reducing filling pressures, as well as by controlling systemic hypertension. Again, as with diuretics, caution is required when nitrates are used in patients without hypertension or with severe diastolic dysfunction with monitoring for a significant reduction in cardiac output and blood pressure as a result of preload reduction. Theoretically, by releasing NO, nitrates may also improve the diastolic distensibility of the ventricle.

The N itrate’s E ffect on A ctivity T olerance in H eart F ailure with P reserved E jection F raction (NEAT-HFpEF) trial evaluated the effects of nitrates on daily activity in 110 ambulatory patients with HFpEF ( Table 39.1 ). This trial was a 6-week dose-escalation regimen of isosorbide mononitrate (from 30 mg to 60 mg to 120 mg once daily) or placebo, with subsequent crossover to the other group for 6 weeks. The primary end point was the daily activity level, quantified as the average daily accelerometer units. In the group receiving the 120-mg dose of isosorbide mononitrate, as compared with the placebo group, there was a nonsignificant trend toward lower daily activity and a significant decrease in hours of activity per day. During all dose regimens, activity in the nitrate group was lower than that in the placebo group. There were no significant between-group differences in the secondary end points of 6-minute walk distance, quality-of-life scores, or N-terminal pro-B type natriuretic peptide (NT-proBNP) levels. Although this trial did not demonstrate benefit of nitrates across patients with chronic HFpEF, selective use of nitrates during acute hospitalization for HFpEF, especially in the setting of uncontrolled hypertension, or the use of nitrates chronically as antianginal therapy in HFpEF with concomitant coronary artery disease may still be beneficial.

Although nitrates improve hemodynamics with vasodilation and thus reduce systemic vascular resistance, they are subject to tolerance over a period of time. In contrast, nitrites can continue to produce sustained vasodilation with long-term exposure. Omar et al. explored the pharmacologic properties of nitrites in the human circulation, reporting that nitrite vasodilates not only the arteriolar and venous circulation but also the conduit blood vessels, an effect similar to that observed with nitroglycerin. This effect was associated with a reduction in central systolic blood pressure, augmentation index, and pulsed wave velocity, which represent hemodynamic effects that might show therapeutic promise in the setting of HFpEF.

Importantly, because nitrite is a stable metabolic product of NO and is readily reduced back to NO in hypoxic and acidic environments, its use in HFpEF to improve exercise capacity is a potentially promising. Experiments in mice show that nitrite therapy increased NO levels, and it prevented cardiac dysfunction and improved LV dimensions in HF models. A study done in humans using inhaled sodium nitrite via a nebulizer showed that acute administration of inhaled sodium nitrite reduced biventricular filling pressures and pulmonary artery pressures at rest and during exercise in HFpEF.

The I norganic N itrite D elivery to I mprove E xercise Capacity in H eart F ailure with P reserved E jection F raction (INDIE-HFpEF) trial was designed to study the effectiveness of inhaled inorganic nitrite on exercise capacity in patients with HFpEF (see Table 39.1 ). Inorganic nitrite functions as an important in vivo reservoir for NO generation, particularly under hypoxic and acidotic conditions. As such, inorganic nitrite becomes most active at times of greater need for NO signaling, as during exercise when LV filling pressures and pulmonary artery pressures increase. The trial enrolled 105 HFpEF patients with a median age of 68 years, LVEF ≥50% with New York Heart Association (NYHA) class II to IV with objective evidence of HF. Patients were randomly assigned to receive a placebo or the inorganic nitrite administered with a nebulizer three times a day for 4 weeks and then crossed over. Preliminary results showed that inhaled nitrite was not effective in improving exercise capacity measured by peak oxygen consumption (VO ₂ ) or other indices of clinical status of HFpEF including levels of NT-proBNP, nor did it improve quality-of-life scores. Thus currently the data do not support the use of inhaled nitrite for symptomatic relief in HFpEF. Further studies looking at different dosing, methods of delivery, or length of therapy using nitrites will be needed to fully evaluate if nitrites will be useful in the treatment of HFpEF.

Treatment of Hypertension (See Also Chapter 25 )

Of the various risk factors for the development of HFpEF, hypertension, and subsequent LV hypertrophy are the most prevalent and highly associated with the condition. The significant contribution of hypertensive heart disease to the development of diastolic dysfunction and HFpEF (reviewed in Hoit and Walsh) implies that treating hypertension should be beneficial not only for the treatment but also for the prevention of HFpEF. In addition, the increased afterload imposed by significant arterial hypertension reduces LV relaxation and filling rates. Stiffening of the aorta and the left ventricle, as occurs in elderly patients with HFpEF, increases the tightness of the coupling of arterial systolic and left atrial pressures, with an increase in systolic arterial pressure resulting in elevation of left atrial pressures. Controlling systolic hypertension could allow the left ventricle to eject to a smaller end-systolic volume, thus allowing the ventricle to operate with a smaller diastolic volume and reduced left atrial pressure. Lowering the systolic pressure allows the left ventricle to relax more rapidly, enhancing early filling. In addition, concentrically hypertrophied hearts demonstrate increased passive stiffness and impaired relaxation independent of hemodynamic loads and have limited coronary vascular reserve that can contribute to myocardial ischemia even in the absence of epicardial coronary artery disease. Adequate control of hypertension should benefit patients with HFpEF by favorably altering loading conditions in the short term and, in the long term, by leading to regression of LV hypertrophy. Although there may be some additional benefits of using one class of drugs versus others, the most important goal is achieving an adequate reduction in blood pressure. Several trials evaluating reduction of blood pressure using angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs), compared with other agents have suggested that ultimately blood pressure control rather than the specific class of antihypertensive agents used may be the major determinant of regression of hypertrophy and improvement in diastolic function. Moreover, trials have definitively demonstrated approximately 50% reduction in the incidence of HF in patients treated for hypertension, especially in the elderly population. Based on the totality of the data available, the most recent update of the American College of Cardiology and American Heart Association (ACC/AHA) HF guidelines recommend that in patients at increased risk (stage A HF) the optimal blood pressure in those with hypertension should be less than 130/80 mm Hg. In addition, patients with established HFpEF and persistent hypertension after management of volume overload should be prescribed guideline directed therapy titrated to attain a systolic blood pressure less than 130 mm Hg.

Drugs such as beta-blockers and calcium channel blockers that reduce blood pressure and heart rate and thus indirectly improve diastolic function, as well as increase diastolic filling time, may be beneficial in patients with HFpEF. In contrast, the direct myocardial effects of slowing the relaxation rate of the ventricle and the negative inotropic actions of these drugs may be detrimental with respect to diastolic function. In addition, a recent study demonstrated that during exercise, patients with HFpEF achieved less of an increase in heart rate (inadequate chronotropic response) and thus a blunted cardiac output despite a similar rise in end-diastolic volume, stroke volume, and contractility compared with matched subjects with hypertensive cardiac hypertrophy. These data suggest possible deleterious effects of heart rate–reducing drugs, such as beta-blockers and certain calcium channel blockers, in patients who may already have reduced exercise capacity resulting from reduced chronotropic reserve.

Over the years, a number of small studies evaluating calcium channel blockers, beta-blockers, ACE inhibitors, and ARBs variably suggested modest benefit in exercise capacity, NYHA class, quality of life, and diastolic function in patients with HFpEF. Since then, large, randomized multicenter trials evaluating the benefit on longer-term outcomes have been performed.

Given that several patients with HFpEF have uncontrolled, drug-resistant hypertension and that hypertension is a major driver of incidence and exacerbation of HFpEF, strategies other than medications that could aid in better blood pressure control are attractive in the prevention and treatment of HFpEF. Although the role of the baroreflex system in long-term blood pressure control has been debated, studies suggest that the system is important in chronic hypertension and that renal sympathoinhibition with a resultant increase in natriuresis may be one of the mechanisms by which the baroreflex participates in long-term blood pressure control. Animal and human studies have demonstrated a safe and effective lowering of blood pressure with chronic electrical stimulation of the carotid sinus. The postulated mechanism is that activation of the baroreceptors is interpreted by the brain as elevation in blood pressure, with resultant activation of the cardiac parasympathetic tone and diminished sympathetic outflow to the heart, kidneys, and peripheral vasculature. A systematic review/meta-analysis examined the efficacy of baroreflex activation therapy (BAT) (through implantable baroreceptor stimulating devices) in patients with resistant hypertension. Twelve studies, including 1 randomized clinical trial and 11 prospective studies, were evaluated, of which 5 prospective studies were selected for meta-analysis. The data analysis showed that office systolic blood pressure and diastolic blood pressure decreased by BAT treatment and that the effect on systolic blood pressure was significant in both the Barostim neo device and the Rheos System. However, the available evidence was limited by risk of bias, small sample size, and few randomized trials, with the conclusion that there is presently insufficient evidence to fully evaluate the efficacy and safety of BAT for patients with resistant hypertension. If BAT is proven to be successful in treating resistant hypertension, the resultant treatment option contributing to a sustained improvement of blood pressure in patients with HFpEF and resistant hypertension could translate into improvement in HFpEF outcomes.

Renin-Angiotensin-Aldosterone Blockade

Angiotensin-Converting Enzyme Inhibitors and Angiotensin Receptor Blockers

As in HFrEF, preclinical and clinical evidence suggests that the activation of the RAAS is a contributing factor in the development of HFpEF, principally through the trophic effects of angiotensin II on the vasculature and myocardium, but perhaps also through myocardial fibrosis mediated by aldosterone ( see also Chapter 5 ). In addition, angiotensin II slows LV relaxation resulting in elevation of LV diastolic pressure. Therefore agents such as ACE inhibitors and ARBs, with their antihypertensive and angiotensin II–attenuating effects, were attractive options in the treatment of HFpEF. Furthermore, clinical trials have shown ACE inhibitors and ARBs to be effective in improving CV outcomes in populations with diabetes, coronary artery disease, vascular disease, and hypertension, comorbidities that are frequently present in and contribute to the development of HFpEF.

On the basis of a strong theoretical rationale for RAAS blockade in patients with HFpEF, three large randomized clinical trials were designed specifically to evaluate ACE inhibitors and ARBs in patients with HFpEF. These include the Candesartan in Heart failure Assessment of Reduction in Mortality and morbidity-Preserved (CHARM-Preserved) trial, the Perindopril in Elderly People with Chronic Heart Failure (PEP-CHF) trial, and the Irbesartan in Heart Failure with PRESERVED Ejection Fraction (I-PRESERVE) trial. The characteristics of the patient populations evaluated in these trials are summarized in Table 39.2 .

TABLE 39.2

Study Characteristics of Large Multicenter Clinical Trials Evaluating Morbidity and Mortality in Heart Failure with Preserved Ejection Fraction

	CHARM-Preserved	PEP-CHF	I-PRESERVE	TOP-CAT	PARAGON-HF
HF trial population	N = 3023 LVEF >40% NYHA class II–IV NYHA III/IV: 39% Mean age: 67 years >75 years: 23% Male: 60%	N = 850 LVEF >40% NYHA class I–IV NYHA III/IV: 24% Mean age: 75 years ≥70 years: 100% Male: 45% Diastolic dysfunction and/or LVH or LA dilation	N = 4128 LVEF ≥45% NYHA class II–IV NYHA III/IV: 79% Mean age: 72 years ≥60 years: 100% Male 40% HF hospitalization or substrate for HF (LVH or LA dilation)	N = 3445 LVEF ≥45% NYHA class II–IV NYHA III/IV: 34% Mean age: 69 years Male: 48% HF hospitalization or BNP >100 pg/mL or NT-proBNP ≥360 pg/mL	Estimated N = 4600 LVEF ≥45% NYHA class II–IV Age ≥50 years Structural heart disease: LVH or LA dilation NT-proBNP ≥200 pg/mL with HF hospitalization within 9 months otherwise NT-proBNP >360 pg/mL
Major exclusion criteria	Significant hypotension Creatinine >3 mg/dL Serum K >5.5 meq/L	Systolic bp <100 mmHg Creatinine >2.3 mg/dL Serum K >5.4 meq/L	Systolic bp <100 or >160 mmHg Creatinine >2.5 mg/dL	GFR <30 mL/min/1.73 m 2 or creatinine ≥2.5 mg/dL Serum K ≥5.0 meq/L Uncontrolled hypertension	GFR <30 mL/min/1.73 m 2 Serum K ≥5.2 meq/L Uncontrolled hypertension History of angioedema
Use of ACE inhibitors/ARBs at baseline	ACE inhibitor use allowed	No concomitant ACE inhibitor/ARB	ACE inhibitors allowed in ⅓ patients when indicated for diabetes or vascular disease	ACE inhibitor/ARB allowed	No concomitant ACE inhibitor/ARB
Protocol: a Double-blind randomized	a Candesartan versus placebo, uptitrated to target 32 mg/day Median follow-up: 36.6 months	a Perindopril versus placebo, uptitrated to target 4 mg/day Median follow-up: 25.2 months	a Irbesartan versus placebo, uptitrated to target 300 mg/day Mean follow-up: 49.5 months	a Spironolactone versus placebo uptitrated to target 45 mg/day Mean follow-up:	a Valsartan target 160 mg twice daily or sacubitril/valsartan target 97/103 mg twice daily. All patients with sequential single-blind run-in periods to ensure tolerability of both drugs at half target doses (i.e., valsartan titrated to 80 mg twice daily followed by sacubitril/valsartan 49/51 mg twice daily) Results not yet reported
Primary end point	Composite of CV mortality or HF hospitalization	Composite of all-cause mortality or HF hospitalization	Composite of all-cause mortality or CV hospitalization	Composite of CV mortality, aborted cardiac arrest or HF hospitalization	Composite of CV death, or total (first and recurrent) HF hospitalizations

ACE , Angiotensin-converting enzyme; ARB , angiotensin receptor blockers; bp , blood pressure; CV , cardiovascular; HF , heart failure; LA , left atrial; LVEF , left ventricular ejection fraction; LVH , left ventricular hypertrophy; NYHA , New York Heart Association.

a Randomized control trial.

Of the 3023 patients enrolled in the CHARM-Preserved trial, almost 20% were on ACE inhibitors and 56% on beta-blockers at the time of randomization to candesartan or placebo. After a median follow-up of approximately 37 months, the primary end point (CV or HF hospitalization) occurred in 22% of the candesartan and 24% of the placebo group (hazard ratio [HR] 0.89, 95% confidence interval [CI] 0.77–1.03, P = .118; covariate-adjusted HR 0.86, 95% CI 0.74–1.00, P = .051) ( Fig. 39.1 ). The difference, which was of borderline statistical significance only for the adjusted hazard ratios, was driven mostly by a difference in HF hospitalizations between the candesartan- and placebo-treated groups (HR 0.85, 95% CI 0.72, 1.01, P = .072), with almost identical CV mortality rates between the two groups ( Fig. 39.2 ). In addition, the total number of HF hospitalizations was noted to be significantly lower in the candesartan group. Of note, at 6 months into the trial, the blood pressure was significantly more reduced in the candesartan group (6.9 mm Hg systolic and 2.9 mm Hg diastolic) compared with the placebo group ( P < .0001). It is therefore difficult to tease out whether the modest 15% relative reduction in HF hospitalizations with candesartan compared with a placebo in patients with HFpEF was mostly a result of blood pressure reduction versus any other more specific angiotensin-blocking effects of candesartan and whether any other drugs causing similar blood pressure reduction would have led to similar effects. The lack of a greater benefit from the use of the ARB in CHARM-Preserved may also have been contributed to by the fact that ACE inhibitors were allowed for patients in the trial, with 20% of patients being on ACE inhibitors even at the beginning of the trial, and the relatively lower than expected annual event rates of the primary composite outcome of 9.1% in the placebo group.

The PEP-CHF trial of perindopril in HFpEF was conducted in a more elderly patient population compared with the CHARM-Preserved trial (see Table 39.2 ). However, the results of the study were neutral with respect to the primary outcome (composite of all-cause mortality and unplanned HF-related hospitalization) and demonstrated only a trend toward modest benefit in other end points. As compared with the CHARM-Preserved trial, the PEP-CHF study enrolled only 850 patients with HFpEF, who, in addition to having relatively preserved LVEF, also had objective evidence of diastolic dysfunction at enrollment. The mean follow-up of 26 months was shorter than the CHARM-Preserved trial. The event rate was much lower than expected, and, despite much longer follow-up than originally intended, only 46% of the expected events occurred, giving the study only 25% power to show a difference in the primary end point. Furthermore, a large number of patients stopped their assigned treatment after 1 year, and most of them started taking open-label ACE inhibitors (discontinuation rate of 38% at 18 months, approximately 36% of patients on open-label ACE inhibitor treatment by the end of the study). Over the total duration of follow-up, perindopril was not associated with an improvement in the primary composite end point of death or HF hospitalization (HR 0.92, 95% CI 0.70–1.21, P = .55; Fig. 39.3A ). However, if analysis was confined to 1 year of follow-up, at which point most patients were still taking the assigned medication, perindopril was associated with a statistically borderline significant 31% relative reduction in the primary end point (HR 0.69, 95% CI 0.47–1.01, P = .055). Similarly, although perindopril was not associated with a benefit in HF hospitalization over the entire duration of the trial (see Fig. 39.3B ), at 1 year of follow-up, the perindopril group had a lower rate of HF hospitalizations (HR 0.63, 95% CI 0.41–0.97, P = .03). Thus, although perindopril did not have a beneficial effect on the overall primary outcome, there was a suggestion of reduction in HF hospitalizations at 1 year when patients were on assigned treatment. In addition, significant improvements were observed compared with the placebo group in some other secondary end points, including the proportion of patients in NYHA functional class I and change in 6-minute-walk distance at 1 year. Similar to the observation in the CHARM-preserved trial, the active arm of perindopril had a significantly greater reduction in systolic blood pressure (mean difference = 3 mm Hg, P = .03) compared with the placebo arm. Furthermore, it was noted that patients with a higher baseline blood pressure appeared to have a greater benefit with perindopril. Therefore, similar to the CHARM-preserved trial, it is not possible to rule out a significant contribution of the blood pressure–lowering effect of the ACE inhibitor as opposed to other RAAS-blocking actions of perindopril toward the observed short-term beneficial effect on HF hospitalizations. In addition, both trials illustrate the encountered difficulty of lower event rates of patients enrolled in clinical trials compared with those seen in the general population. The switch over to open-label ACE inhibitors illustrates the many existing indications for the use of ACE inhibitors or ARBs in patients with vascular disease, diabetes, and hypertension, the same group of comorbidities that are frequently encountered in patients with HFpEF, thus potentially limiting the opportunities to add ACE inhibitors or ARBs specifically for HfpEF.

The largest trial, I-PRESERVE, enrolled 4128 patients with HfpEF (see Table 39.2 ). The patients were randomly assigned to ARB, irbesartan, or placebo. During a mean follow-up of 49.5 months, there was no significant difference in the occurrence of the primary outcome (death from any cause or hospitalization for a CV cause [i.e., HF, myocardial infarction, unstable angina, arrhythmia, or stroke]) between irbesartan and the placebo (HR 0.95; 95% CI 0.86–1.05; P = .35; Fig. 39.4 ). Overall rates of death were also similar (HR 1.00; 95% CI 0.88–1.14; P = .98), as were rates of CV hospitalization, HF hospitalization, and other secondary outcomes. Although, compared with the CHARM-preserved trial, the I-PRESERVE trial studied a greater number of patients, with slightly better preserved EF (≥45% vs. >40% in CHARM-Preserved), greater specificity of the substrate for HfpEF, and a greater proportion of older patients and women (i.e., a study cohort more representative of the “real world” HfpEF population), this trial did not provide any evidence for overall benefit of ARBs on CV outcomes in HfpEF patients.

Thus apart from the modest signal of 15% relative reduction in HF hospitalizations with candesartan compared with a placebo in HfpEF patients observed in the CHARM-Preserved trial and a possible benefit on HF hospitalizations and functional class at an intermediate timepoint in PEP-CHF (in the absence of benefit over the entire duration of follow-up), neither ACE inhibitors nor ARBs has been proven to have a convincing beneficial effect on clinical outcomes in HfpEF.

In addition, of note, RAAS inhibitors can induce renal dysfunction in both HfrEF and HfpEF. However, in contrast to patients with HfrEF, where mortality increase with worsening renal function is small, HfpEF patients with RAAS inhibitor-induced worsening renal function may have an increased mortality risk, without experiencing improved outcome with RAAS inhibition.

Aldosterone Receptor Antagonists

Mineralocorticoid receptor activation by aldosterone contributes to the pathophysiology of HF through several mechanisms, including sodium retention, potassium loss, endothelial dysfunction, vascular inflammation, fibrosis, and hypertrophy. Smaller trials of aldosterone receptor antagonists suggested benefit on diastolic function without improvement in functional capacity. The Aldosterone Receptor Blockade in Diastolic Heart Failure (ALDO-DHF) trial was a prospective, randomized, double-blind, placebo-controlled trial evaluating the effect of spironolactone on diastolic function and exercise capacity in 422 patients with HfpEF (see Table 39.1 ). At the end of 12 months of treatment, compared with a placebo, spironolactone use resulted in improvement in echocardiographic parameters of diastolic dysfunction (mitral annular E/e′ adjusted mean difference between the placebo and spironolactone −1.5; 95% CI −2.0 to −0.9, P < .001), neuroendocrine activation (NT-proBNP geometric ratio 0.86, 95% CI 0.75–0.99, P = .03), and induced reverse remodeling (decline in LV mass index; difference −6 g/m ² , 95% CI −10 to −1 g/m ² ; P = .009) but did not improve HF symptoms or quality of life and slightly reduced the 6-minute walking distance, with no effect on hospitalizations. The lack of clinical improvement despite improvement in echocardiographic and laboratory parameters may have been caused by the relatively younger and healthier patient cohort (86% NYHA class II, ≈50% on diuretics and baseline median NT-proBNP 158 ng/mL), a low event rate, and shorter follow-up. Similarly, another smaller study, the Randomized Aldosterone Antagonism in Heart Failure with Preserved Ejection Fraction (RAAM-PEF) trial, evaluated the effects of the more selective aldosterone receptor blocker, eplerenone, compared with a placebo on 6-minute walk distance, diastolic function, and markers of collagen turnover after 6 months of treatment. The use of eplerenone was associated with improvement in diastolic dysfunction (E/e′, P ≤ .01) and reduced collagen turnover but no change in 6-minute walk distance ( P = .91) as compared with a placebo.

The large multicenter double-blind, placebo-controlled trial, Treatment of Preserved Cardiac function heart failure with an Aldosterone Antagonist (TOPCAT), evaluated the effects of spironolactone on morbidity and mortality in 3445 patients with HfpEF (LVEF ≥45%; see Table 39.2 ). Overall, the primary end point of CV death, HF hospitalization, or resuscitated cardiac arrest was similar between the spironolactone and the placebo arms (18.6% vs. 20.4%, HR 0.89, 95% CI 0.77–1.04, P = .14; Fig. 39.5A ), as were the individual components of CV mortality and aborted cardiac arrest. However, HF hospitalizations were lower (12.0% vs. 14.2%, P = .042; see Fig. 39.5B ); all-cause hospitalizations were similar ( P = .25). Hyperkalemia (18.7% vs. 9.1%, P < .001) and renal failure, defined as serum creatinine ≥ times the baseline value and above the upper limit of the normal range, were both significantly higher in the spironolactone arm. Therefore, in the TOPCAT trial, spironolactone was not overall superior to placebo in improving CV outcomes in patients with HFpEF, with a significantly higher rate of hyperkalemia and renal failure in patients treated with spironolactone. Most of these patients were already on an ACE inhibitor/ARB. The reduction in HF hospitalizations with spironolactone was encouraging, but this finding should be approached with caution because all-cause hospitalizations were similar between the placebo and spironolactone arms.

An interesting post hoc analysis revealed a disparity in outcomes between the centers in North and South America and those in Eastern Europe. In the United States, Canada, Brazil, and Argentina, where the rate of the primary composite outcome was 31.8% in the placebo group, spironolactone had a significant benefit on the primary end point (HR 0.82, 95% CI 0.69–0.98). However, in Russia and the Republic of Georgia, where the primary outcome occurred in only 8.4% of patients taking the placebo, spironolactone did not have any effect (HR 1.10, 95% CI 0.79–1.51; Fig. 39.6 ). Given that there are no current therapies that have been demonstrated to significantly improve clinical outcomes in HFpEF populations, it has been suggested that higher-risk HFpEF populations similar to those enrolled in the TOPCAT trial in the Americas, aldosterone receptor antagonists might be considered to reduce the risk of hospitalizations (see Table 39G.1 ).

Angiotensin Receptor/Neprilysin Inhibitors

Natriuretic peptides have been associated with potent natriuresis, vasodilation, and inhibition of the RAAS and may have antifibrotic and antisympathetic effects (see also Chapter 9 ). Neprilysin (neutral endopeptidase 24–11) inhibition leads to increased circulating levels of natriuretic peptides by preventing their breakdown with consequent cardiac, vascular, and renal protective effects. In addition, the augmentation of active natriuretic peptides increases generation of myocardial cyclic guanosine monophosphate (cGMP), which improves myocardial relaxation and decreases hypertrophy. However, neprilysin inhibition also prevents breakdown of vasoconstrictors like angiotensin II, thus diminishing the beneficial effects of neprilysin inhibition. Concomitant inhibition of the RAAS, along with neprilysin inhibition, helps to overcome this limitation and is the basis of dual-acting vasopeptidase inhibitors ( Fig. 39.7 ). Previous studies showed promising effects of a combination drug with neutral endopeptidase and an ACE inhibitor (omapatrilat) for treatment of patients with hypertension and HF. However, further development of this agent was stopped due to increased occurrence of angioedema, likely from increased bradykinin levels from neprilysin and ACE inhibition. ARBs are associated with a lower incidence of angioedema as compared with ACE inhibitors and were substituted for ACE inhibitors in the development of a new dual vasopeptidase inhibitor, LCZ696 (combination of an ARB, valsartan, and a neprilysin inhibitor, sacubitril), with higher antihypertensive potency as compared with valsartan and similar tolerability. The use of this ARB-neprilysn inhibitor or angiotensin receptor/neprilysin inhibitor (ARNI) was associated with a significant morbidity and mortality benefit in patients with HFrEF.

This combination of sacubitril-valsartan was also evaluated in a phase II double-blind, randomized, parallel group trial, the Prospective Comparison of ARNI with ARB on Management Of heart failUre with preserved ejectioN fracTion (PARAMOUNT) trial. The trial evaluated 301 patients with NYHA class II to III HFpEF (LVEF ≥ 45%) and NT-proBNP ≥400 pg/mL and randomized patients to LCZ696 titrated to 200 mg twice daily or valsartan titrated to 160 mg twice daily and treated for 36 weeks (see Table 39.1 ). The primary end point, a decrease in NT-proBNP from baseline to 12 weeks, was significantly greater in the LCZ696 group compared with the valsartan group (ratio of change LCZ696/valsartan at 0.77, 95% CI 0.64–0.92, P = .005; Fig. 39.8 ). Although, NT-proBNP remained significantly reduced at 36 weeks compared with baseline, the difference between the two treatment groups was no longer statistically significant ( P = .2). However, left atrial volume ( P = .003) and left atrial dimension ( P = .034) were significantly reduced and NYHA class improved at 36 weeks in patients in the LCZ696 group. No other echocardiographic parameters (including measures of diastolic function) differed between the treatment groups. The effect of LCZ696 on NT-proBNP appeared greater in subgroups of patients with systolic blood pressure higher than 140 mm Hg (possibly related to a greater blood pressure–lowering effect of this agent compared with ARBs and ACE inhibitors) and in diabetic patients.

The ongoing phase III trial will determine the effects of LCZ696 on clinical outcomes. The Prospective Comparison of ARNI with ARB on Management of Heart Failure with Preserved Ejection Fraction (PARAGON-HF) has completed enrollment of nearly 5000 patients with chronic HFpEF who are being followed for the cumulative number of primary composite events of CV death and total (first and recurrent) HF hospitalizations (see Table 39.2 ; Fig. 39.7 ). The trial has enrolled subjects with symptoms of HF on diuretics with LVEF ≥45%, evidence of structural heart disease (left atrial enlargement or LV hypertrophy), and elevated natriuretic peptides (NT-proBNP).

Neprilysin is also one of several proteases that plays a role in the degradation of amyloid-β proteins, and concern has therefore been raised about the theoretical possibility that long-term neprilysin inhibition could influence the development of Alzheimer disease. However, the specific role of neprilysin in the pathogenesis of Alzheimer disease remains unclear. In addition to the likely redundancy in the enzymes that break down amyloid-β, individuals with truncating mutations in the gene encoding for neprilysin, who have total neprilysin deficiency, appear to lack any phenotypic consequence. A multicenter randomized double-blind study to evaluate the effects of LCZ696 compared with Valsartan on cognitive function in patients with chronic HFpEF, Prospective Evaluation of Cognitive Function in Heart Failure: A Randomized Double-blind Study in Patients with Preserved Ejection Fraction Cardiac Failure Treated with Valsartan or Entresto (PERSPECTIVE) trial, which started enrollment of approximately 500 participants in 2016, will provide data on possible effects of LCZ696 on cognitive function due to possible amyloid-β deposition in the brain. This trial will assess the overall impact of long-term LCZ696 therapy on cognitive function through a comprehensive battery of neuropsychological tests and will assess amyloid-β deposition in the brain by using positron emission tomography amyloid imaging.

Beta-Blockade

No large definitive trials of beta-blockers in HFpEF have been performed. A smaller trial of 95 patients, the Swedish Doppler-echocardiographic study (SWEDIC), demonstrated no significant benefit of the beta-blocker, carvedilol, on diastolic function patients with mild HFpEF (LVEF >45%) who also met conventional Doppler criteria for diastolic dysfunction. The Japanese Diastolic Heart Failure Study (J-DHF) trial was a multicenter, randomized, placebo-controlled, open, blinded–end point trial, which evaluated the effects of carvedilol in 245 patients with HFpEF (LVEF >40%) during a median follow-up of 3.2 years. There was no difference in the primary composite outcome of CV death or HF hospitalization or in the secondary composite outcome of CV death or any CV hospitalization. However, only 245 of the planned 800 patients were enrolled, resulting in the trial being underpowered. Another smaller trial, the Effects of the Long-term Administration of Nebivolol on the clinical symptoms, exercise capacity, and LV function of patients with Diastolic Dysfunction: (ELANDD) study evaluated 116 patients with NYHA class II to III symptoms, LVEF greater than 45%, and echocardiographic signs of LV diastolic dysfunction. The patients were randomized to 6 months of treatment with nebivolol or a placebo. Nebivolol is a beta ₁ selective blocker that also has vasodilating and antioxidative properties and improves endothelial dysfunction via its effects on the endothelial NO synthase. However, in the trial, no differences were demonstrated in the 6-minute walk distance or peak oxygen consumption between the nebivolol or placebo groups. The resting and peak blood pressure decreased in patients on nebivolol; there was a significant interaction between change in peak exercise heart rate and peak oxygen consumption (r = 0.391, P = .003), and it was postulated that the negative chronotropic effect of the beta-blocker may have contributed to the neutral results of the study.

A larger multicenter trial evaluated the effects of nebivolol in elderly patients with HF, irrespective of LVEF. In the Study of the Effects of Nebivolol Intervention on Outcomes and Rehospitalizations in Seniors with heart failure (SENIORS), 2128 patients aged 70 years or older with a history of HF were randomized to receive nebivolol or the placebo. Of these, 35% of patients had LVEF greater than 35%. In the overall study, nebivolol was associated with a modest 14% reduction in the primary composite end point of death or CV hospitalization (HR 0.84, 95% CI 0.74–0.99), contributed to by a reduction in both mortality and CV hospitalization. A similar magnitude of reduction was noted in the subgroup with LVEF greater than 35%; however, it did not reach statistical significance, possibly a result of fewer patients in the subgroup (HR 0.82, 95% CI 0.63–1.05). Although labeled as patients with preserved ejection fraction or HFpEF, this subgroup of patients with EF greater than 35% clearly consisted of a heterogeneous patient population with both preserved and reduced LVEF. The proportion of patients with truly preserved LVEF (≥50%) was small (<15% of all patients in the trial). Therefore, although the overall trial suggests a modest CV benefit of the beta-blocker, nebivolol, in elderly patients with HF, the results did not convincingly demonstrate a benefit in elderly patients with HFpEF.

In summary, these trials of the beta-blockers carvedilol and nebivolol, although limited by sample size, have not demonstrated definitive benefits, or harm, on exercise and functional capacity, on diastolic dysfunction, or on clinical outcomes in patients with HFpEF.