Heart failure (HF) substantially impacts the health and financial security of an increasing proportion of the US population. It worsens debility and quality of life and may lead to hospitalization and death. HF is a clinical syndrome with diverse symptomatic presentations. Physicians generally divide patients with HF into 2 groups: those with a left ventricular ejection fraction (LVEF) greater than or equal to 50% and those with an LVEF less than 49%. This review focuses on the group of patients whose LVEF is greater than or equal to 50%. This classification of HF is referred to as HF with a preserved ejection fraction (HFpEF). Few beneficial therapies have been identified for this condition, possibly because of its heterogenous etiologies (eg, myocardial, vascular, metabolic, and other physiologic derangements). Clinicians should focus on diagnosing, treating, and preventing the etiologies that are known to cause HFpEF. Results from a small proportion of randomized controlled trials have shown therapeutic benefit for small molecules, although limited, if any, demonstrated mortality benefit has been noted. More research and investment are needed to decrease the burden of HFpEF and to discover lifesaving treatments for this growing population.
Am J Manag Care. 2023;29:S187-S194. https://doi.org/10.37765/ajmc.2023.89417
For author information and disclosures, see end of text.
Heart failure (HF) burdens a large and increasing proportion of the population in the United States and causes substantial clinical impact and financial expenditure. Here we review the epidemiology, diagnosis, management, and financial burden of HF with a left ventricular ejection fraction (LVEF) of 50% or more, also known as HF with a preserved ejection fraction (HFpEF). We aim to cover what is known, reveal current gaps in knowledge, and highlight where future evidence generation is needed.
As defined in the 2022 American College of Cardiology/American Heart Association/Heart Failure Society of America (ACC/AHA/HFSA) guideline for the management of HF, “HF is a complex clinical syndrome with symptoms and signs that result from any structural or functional impairment of ventricular filling or ejection of blood.”1 The clinical syndrome consists of symptoms such as dyspnea on exertion, reduced ability to exercise, fatigue, and lower extremity swelling. These symptoms limit physical activity, worsen debility, and impair employment and recreational engagement, and they may lead to hospitalization.
LVEF is used to broadly characterize HF, because most historical clinical trials that included LVEF as an enrollment criterion and a factor for clinical care and guidelines have since adopted it as a descriptor. However, because it is a label defined by historical context, there is variability to the LVEF that distinguishes HF subtypes, including HFpEF.2 Current US guidelines define HFpEF as HF with an LVEF of at least 50%.1,3 This point of LVEF delineation, recently ranging from 40% to 55%, has varied over time.4 LVEF is influenced by sex, age, body mass, and method of measurement, which may cause clinicians to question its accuracy, but it generally can be consistently estimated within a range of 5% to 10%.5 Although some patients with HF with improved LVEF (ie, patients whose LVEF has improved from < 40% to ≥ 50%) can have an LVEF of at least 50%, the treatment of these patients and their response is similar to those of patients with HF with a reduced ejection fraction (HFrEF) (LVEF ≤ 40%); this situation is not included in this discussion.
Superficially, the phenotype of HFpEF neatly categorizes patients; however, HFpEF appears to be a heterogenous and multifaceted syndrome with numerous potential underlying etiologies.6 Hemodynamically, the syndrome of HFpEF manifests with elevated left ventricular filling pressures either at rest or with exertion, but these elevations may be the final common pathway for many diseases. The archetypal HFpEF disease remains cardiac amyloidosis, which likely has confounded the study of other types of HFpEF repeatedly; however, myocardial, vascular, metabolic, and other physiologic derangements likely participate in some types of patients with HFpEF.7 Thus, although LVEF offers a simple way of classifying this disease, substantially more specific and detailed elucidation of the subcategories of HF with LVEF of 50% or more are required for accurate characterization.
The incidence of HFrEF has decreased, yet the incidence of HFpEF and its proportion of all HF continues to rise.8 It is estimated that currently over half of all patients with HF do not have a reduced LVEF.9
HFpEF more frequently occurs in patients with multiple comorbidities, and it has a higher prevalence in women and older patients.10 Among Black patients with HF, HFpEF appears to account for up to 70% of all HF and to be associated with worse outcomes.11,12
HF—and HFpEF specifically—has a deep economic impact on health care systems. Health care costs related to HF in the United States are expected to exceed $50 billion annually by 2030.10 Although designed as a quality improvement program, the Hospital Readmissions Reduction Program (HRRP) revealed a financial impact of HF that illustrates its burden on health systems. Implemented under the 2010 Patient Protection and Affordable Care Act, the HRRP is a mandatory, federal, pay-for-performance program to decrease 30-day hospital readmission rates; it financially penalizes hospitals with 30-day readmission rates that are higher than expected.13 The HRRP came into effect on October 1, 2012; it penalizes close to 80% of hospitals and earns Medicare more than $500 million annually.14
Along with health systems and the government, patients with HFpEF face a considerable financial burden. One report found that the rate of outpatient visits for this patient population was 3.6 visits per month. Of these patients, 27% were hospitalized within 30 days of their first urgent care HF visit; the mean total monthly health care cost per patient was $7482. The total monthly medication cost was higher for patients with an LVEF of 50% or more than for those with a reduced LVEF ($495 vs $429, respectively).15
Compounding this financial burden, patients with HFpEF also have decreased quality and quantity of life. Although tremendous progress has been made in the treatment of HFrEF, there remains a paucity of therapies that decrease mortality and morbidity for patients with HFpEF. These patients have persistently diminished exercise tolerance and ability to perform physical activity.16 With the explosion of obesity and diabetes, HFpEF also affects younger patients, and an analysis of results from 3 clinical trials (RELAX [NCT00763867], NEAT-HFpEF [NCT02053493], INDIE-HFpEF [NCT02742129]) showed that patients with HFpEF who are young, who have obesity, and who have diabetes were more likely to score lower on quality of life (QOL) measures.17
Few effective therapies have been identified for patients with HFpEF, perhaps because this medical condition can occur in response to a wide variety of underlying etiologies. Further, unlike HFrEF, neurohormonal antagonists have not yet been shown to improve the multiple varieties of HFpEF.6 The heterogeneity exists in part because LVEF is a poor differentiator among the various flavors of HFpEF. Because of these issues and the symptom and diagnostic variability of HFpEF, clinicians should focus on diagnosing, treating, and preventing the etiologies known to cause this clinical syndrome.
A large proportion of the US population has high risk factors for HF; these include hypertension, obesity, diabetes, and atherosclerotic cardiovascular disease.1,4 Common causes of HFpEF include metabolic heart disease, valvular heart disease, and ischemic heart disease. Other causes may include cardiomyopathies that are stress induced, genetic, or peripartum. Infiltrative and inflammation-related diseases (eg, sarcoidosis) can also contribute, and amyloidosis should be appropriately evaluated and treated separately if present. Cardiotoxicity related to use of alcohol, stimulants (eg, cocaine, methamphetamine), and oncologic therapeutics can cause this clinical syndrome. Right ventricular dysfunction and pulmonary hypertension may lead to left-sided cardiomyopathy, but they more frequently are inappropriately labeled as HFpEF; careful vigilance is required to diagnose these patients and refer them to specialty centers.1
Identification of the underlying etiologies is essential to patient management, because the most effective treatment is often determined by the etiology of presentation. Current guidelines recommend control of blood pressure in line with guideline-directed medical therapy for hypertension to decrease occurrence of HFpEF and encourage regular exercise and weight loss for patients with obesity.1,18 For those patients with HFpEF with underlying diseases, the ACC recommends coronary revascularization, valve replacement or repair, and medical ischemia management.18-22 Patients with HFpEF and obstructive sleep apnea may present with insomnia, daytime hypersomnolence, obesity, loud snoring, and witnessed apnea.23 Treatment is directed toward weight loss via methods such as bariatric surgery and prescription of continuous positive airway pressure.23 Arrhythmias such as atrial fibrillation can also exacerbate HFpEF symptoms, although HFpEF and increased left ventricular filling pressures also increase the risk of atrial fibrillation.1
Until recently, few medications or devices have demonstrated improvement in mortality or morbidity for patients with HFpEF (Table).24-36 Fortunately, several recent successes in pharmaceutical development have overcome the results of multiple prior programs that were unable to show treatment effectiveness in the setting of HF with an LVEF of 50% or more (Figure).24-26,32 Diuretic use has not been supported by the results of clinical trials in the setting of HFpEF, but it remains a central part of the standard of care due to near-universal associations of patient exacerbation with intravascular or extravascular volume overload. Regardless of the etiology, patients with HFpEF often present with fluid retention. To relieve congestion, clinicians typically opt for a loop diuretic—intravenously in the acute inpatient setting and orally in the outpatient setting. Increasingly, ambulatory diuretic infusion clinics are used to assuage the hospital burden of HF and keep patients in the ambulatory setting as much as possible.37
In the TRANSFORM-HF trial (NCT03296813), investigators compared the treatment strategy of torsemide versus furosemide for patients with HFrEF or HFpEF who were hospitalized.38 Dosing was left to the discretion of the treating physicians. With a median follow-up of 17.4 months, there was no difference in all-cause mortality between patients who received torsemide versus those who received furosemide (HR, 1.02; 95% CI, 0.89-1.18).39 The finding was present in patients with HF, regardless of whether their LVEF was reduced or preserved. For those patients whose disease is refractory to loop diuretic therapy alone, concomitant metolazone or chlorothiazide use may be considered to augment diuresis.
Outside of a case of acute decompensation, use of few medications and devices appear beneficial in improving outcomes for patients with HFpEF. These classes include the sodium-glucose cotransporter 2 inhibitors (SGLT2is), mineralocorticoid receptor antagonists (MRAs), and angiotensin receptor–neprilysin inhibitors (ANRIs) and pulmonary artery (PA) pressure monitoring.
In the EMPEROR-Preserved trial (NCT03057951), use of the SGLT2i empagliflozin demonstrated improvement on the primary composite end point of cardiovascular death and HF hospitalization for patients with HF and an LVEF of more than 40% (HR, 0.79; 95% CI, 0.69-0.90).24 Most of the treatment effect was due to a reduction in hospitalizations in the treated group (from 6.0 to 4.3 hospitalizations/100 patient-years). Of the enrolled patients, 67% had an LVEF of at least 50%; in those patients, there also appeared to be benefit in terms of the primary end point, although the treatment effect appeared to diminish at progressively higher values of baseline LVEF.2
The EMPEROR-Preserved data led to an expansion of the empagliflozin label to not restrict use to patients with HF and a “reduced ejection fraction.”40 For those patients, the addition of empagliflozin appeared to be a meaningful advance in care to reduce morbidity and improve QOL. In addition, the EMPULSE trial (NCT04157751) enrolled 169 patients with an LVEF of more than 40%.41 These patients appeared to have similar benefit in terms of the composite win ratio end point of time to death, HF event frequency, time to event, and QOL score measured by the Kansas City Cardiomyopathy Questionnaire as patients enrolled with lower LVEF values.42
Results of the DELIVER trial (NCT03619213) showed that use of another SGLT2i, dapagliflozin, decreased the risk of worsening HF (defined as an urgent visit or unplanned hospitalization for HF) and cardiovascular death in patients with an LVEF of more than 40% (HR, 0.82; 95% CI, 0.73-0.92).25 Dapagliflozin also reduced the risk of worsening HF and cardiovascular death for patients with an LVEF of 60% or more (HR 0.78; 95% CI, 0.62-0.98), which was similar to that of patients with an LVEF below 60% (HR 0.83; 95% CI, 0.73-0.95).
In the SOLOIST-WHF trial (NCT03521934) of patients with type 2 diabetes (T2D), sotagliflozin therapy reduced the rate of the primary end point of cardiovascular death and hospitalization or urgent visits for HF for patients with diabetes mellitus and recent worsening of HF without limitation by LVEF26; however, the trial was terminated early by the sponsor.43 Similarly, results from the SCORED trial (NCT03315143) of patients with T2D and chronic kidney disease (ie, estimated glomerular filtration rate [eGFR] 25-60 mL/min/1.73 m2 regardless of albuminuria status), and cardiovascular risk factors showed that sotagliflozin therapy significantly reduced the primary end point of cardiovascular death and hospitalization or urgent visits for HF.44 These data and the consistent results of a meta-analysis suggested that the beneficial effect for patients with HFpEF may be a class effect for the SGLT2is.45 As of May 2023 the SGLT2is empagliflozin, dapagliflozin, and sotagliflozin have received regulatory approval for HF regardless of ejection fraction.40,46,47
The evidentiary support for SGLT2 inhibitors is sufficiently strong that the ACC recent consensus statement suggests that an SGLT2 inhibitor should be considered at the time of HFpEF diagnosis, except in patients with certain comorbidities (orthostatic hypotension, eGFR < 20-25 mL/min/1.73 m2, very frequent yeast infections, or a history of severe genitourinary infections, including Fournier gangrene).48
In the PARAGON-HF trial (NCT01920711), patients with HF and an LVEF of at least 45% who received sacubitril-valsartan did not have significantly decreased rates of cardiovascular death or total HF hospital admissions when compared with findings in patients who received valsartan, based on the initial event adjudication (rate ratio, 0.87; 95% CI, 0.75-1.01).32 However, more total events were adjudicated by site investigators compared with those reviewed by the central adjudicators; further, when the data were analyzed by either the principal investigator adjudicated events or when the trial was readjudicated in a blinded fashion, there was a beneficial effect of sacubitril-valsartan on the primary end point (rate ratio, 0.83 [95% CI, 0.73-0.95] and 0.86 [95% CI, 0.75-1.00], respectively).32,33,49-51 Similar to results noted with use of empagliflozin, the benefit of sacubitril-valsartan was primarily on HF hospitalizations,52 and there was a decrement in treatment effectiveness at higher levels of LVEF. For this indication, the prescribing information for sacubitril-valsartan includes that “benefits are most clearly evident in patients with LVEF below normal,” referring to the diminishing treatment effect with respect to higher LVEF.52
Use of renin–angiotensin–aldosterone system inhibitors, apart from sacubitril-valsartan, has not produced a statistically significant reduction in mortality or hospitalization for patients with HFpEF. In the CHARM-Preserved trial, the decrease in cardiovascular death or HF hospitalization in patients with an LVEF of more than 40% and who received candesartan did not achieve statistical significance (unadjusted HR, 0.89; 95% CI, 0.77-1.03).27 In the I-PRESERVE trial (NCT00095238), patients with LVEF of at least 45% who received irbesartan did not have significantly decreased rates of mortality or hospitalization when compared with the placebo group (HR, 0.95; 95% CI, 0.86-1.05).28
Although the mineralocorticoid antagonist spironolactone did not achieve a regulatory indication for HF with a higher LVEF, it is often used by clinicians for these patients.53 Results of the TOPCAT trial (NCT00094302) did not show a statistically significant reduction in composite primary end point of cardiovascular death, cardiac arrest, or HF hospitalization for patients with HF and a LVEF of at least 45% (HR 0.89; 95% CI, 0.77-1.04).30 However, there were multiple regional irregularities found in the trial conduct, including enrollment of patients with very low event rates, unclear diagnoses of HF, and blood testing demonstrating an absence of spironolactone metabolites in patients intended to be taking the drug.54 In that light, post hoc analysis suggested efficacy in enrollees from the Americas (HR, 0.82; 95% CI, 0.69-0.98).55 Similar to therapy with empagliflozin and sacubitril-valsartan, use of spironolactone appeared primarily to reduce the risk of hospitalization for HF, which seemed to be greatest for patients with a lower LVEF.
The diminishing therapeutic benefit noted among patients with a higher LVEF appears to be similar in the EMPEROR-Preserved, PARAGON-HF, and TOPCAT results and does not appear unique to empagliflozin, sacubitril-valsartan, or spironolactone.50 A similar decrement in effectiveness for patients with HF and higher LVEF is apparent for most of the pharmacotherapies that have demonstrated a beneficial treatment effect for this population. Although the point estimate for effectiveness for dapagliflozin in patients with a higher LVEF from DELIVER is consistent with lower values of LVEF, and the statistical test for heterogeneity supports a benefit, at higher values of LVEF the confidence intervals are wide and cross no-benefit and the heterogeneity test is statistically weak, meaning that the evidence for effectiveness at elevated LVEF is substantially weaker than at lower LVEF.56 The diminishing effectiveness of these pharmacotherapies at higher ranges of LVEF in patients with HF may be due to the multiple comingled pathophysiologies in these patients, many of whom may not respond to these therapies.53 The trials of HF with higher LVEFs enrolled patients who could have had cardiac amyloidosis; it is likely that a proportion of patients did, as they were not systematically screened for this comorbidity, which may be unresponsive to the therapies tested above and may be better treated with more targeted therapies, as discussed below.7
The CardioMEMS HF System (Abbott) is a permanently implantable, battery-free, pressure-sensitive capacitor that is placed percutaneously in the left PA to measure PA pressure. Patients can communicate these measurements electronically, typically on a daily basis, to clinicians with their at-home portable communication units. Changes in PA pressure can then be appropriately treated by adjustment of diuretic or vasodilator dose or with other medical or device therapy.57 The CHAMPION trial (NCT00531661) results showed a significant decrease in HF-related hospitalizations after 6 months for patients with HF without LVEF exclusion who received a PA pressure sensor compared with those who did not.34 A post hoc analysis of patients with an LVEF of at least 50% showed a reduction in hospitalization that was similar to that seen in the overall trial.58 The GUIDE-HF trial (NCT03387813) expanded the evidence for patients with New York Heart Association class II to IV symptoms, although this was substantially impacted by the COVID-19 pandemic. Based on the overall data, the CardioMEMS system has regulatory approval for wireless measurement of PA pressure and heart rate for patients with HF,
irrespective of LVEF.59
Cardiac amyloidosis should be specifically evaluated as a cause of HFpEF, as it responds to targeted amyloid therapies that are distinct from the HF management discussed above. Amyloidosis is a restrictive cardiomyopathy caused most often by deposition of either immunoglobulin light chains or transthyretin. Amyloidosis is often present in undifferentiated patients with HF and an LVEF of at least 50%. Results of the ATTR-ACT trial (NCT01994889) showed that use of tafamidis, a transthyretin stabilizer, reduced all-cause mortality and cardiovascular-related hospitalizations and improved functional capacity and QOL when compared with placebo.35
Because patients with HFpEF frequently have multiple other comorbidities, recommended interventions include care given by multidisciplinary teams. This often includes arrhythmia specialists and staff who provide structured HF education and screening for mental illness, frailty, and health literacy.1
Although meaningful advances have been made in the development of effective therapeutics for patients with HF and an LVEF of at least 50%, barriers remain to delivering these evidence-based treatments. Therapeutic cost poses a substantial obstacle. Overall, prescription medication cost comprises the biggest portion of outpatient cardiovascular spending, and it is rising faster than is inflation.60 However, the complex relationship between drug makers, insurers, pharmaceutical companies, and distributors conceals the prices of medications as it influences the accessibility of these medications to patients.60,61 Patient assistance can facilitate access to sacubitril-valsartan and empagliflozin for some patients with HFpEF; however, other individuals are limited by high insurance co-payments and co-insurance or other utilization management procedures (Table).24-36
Tafamidis exemplifies the barriers to effectively treating patients with newly developed therapeutics. At an annual cost of $225,000, tafamidis is the most expensive cardiovascular drug ever launched in the United States. For tafamidis to be cost-effective, a 92.6% reduction in its price would be needed. Annual health care spending would increase by $32.3 billion if all eligible patients with HF were treated with tafamidis. Current guidelines classify tafamidis as a drug with low economic value (> $180,000/quality-adjusted life-years [QALYs] gained) due to its pricing.1
The EMPA-REG OUTCOME trial (NCT01131676) demonstrated that empagliflozin reduces cardiovascular mortality in patients with T2D.62 Several post hoc analyses have evaluated the cost-effectiveness of empagliflozin and found that, compared with standard of care, empagliflozin is a cost-effective treatment option for patients with HF and T2D.63-65 Results of a US-based, post hoc analysis showed that empagliflozin therapy added 0.67 QALYs and $17,322 per patient over a lifetime horizon for those patients, which is generally accepted to be cost-effective.66 However, the cost-effectiveness of empagliflozin for patients with HFpEF is highly dependent on cost.67 The cost of empagliflozin per QALY gained was $437,442 based on the EMPEROR-Preserved results and the list price at the time of publication. Dapagliflozin cost-effectiveness presumably is similarly sensitive to the same analyses.
Based on the available clinical trial data, the CardioMEMS device reduction in hospitalizations appears to be of good economic value (< $60,000/QALY gained). In patients similar to those in the CHAMPION trial cohort, the cost per QALY gained was $47,768 in patients with HFpEF.68
Given the cost and accessibility of HF therapies for patients with HFpEF, low-cost interventions could increase implementation of guideline-directed medical therapy. PROMPT-HF (NCT04514458) was a cluster-randomized, comparative effectiveness trial in which 100 clinicians managing patients with HFrEF were randomly assigned to receive an electronic health record (EHR) alert recommending guideline-directed medical therapy for HF or to provide usual care without an EHR best practice advisory alert. HF medical therapy prescriptions were significantly increased in the alert exposure group, and similar methodology could be used for the prescription of therapies for patients with HFpEF.69
HFpEF remains a challenging pathology for physicians and patients because of the ongoing evolution of a standardized definition of the disease state, results of few randomized controlled trials showing therapeutic benefit, and no therapies having a clearly demonstrated mortality benefit. In addition, substantial treatment barriers and financial burdens exist for patients, payors, and health systems. Because the population of patients with HFpEF continues to grow and impact a greater proportion of patients in the US medical systems, comprehensive strategies are needed to improve care, prevent disease worsening, and develop new treatments.
Acknowledgements
The authors meet criteria for authorship as recommended by the International Committee of Medical Journal Editors (ICMJE). The authors received no payment related to the development of the manuscript. Marissa Buttaro, MPH, and Jennifer Garrett, MBBS, of Envision Pharma Group provided editorial and formatting support, which was contracted and funded by Boehringer Ingelheim Pharmaceuticals, Inc (BIPI) and Lilly USA, LLC. BIPI and Lilly were given the opportunity to review the manuscript for medical and scientific accuracy as well as intellectual property considerations.
Author Affiliations: MedStar Georgetown University Hospital (SK), Washington, D.C.; Inova Heart and Vascular Institute, Inova Health System (MAP), Falls Church, VA.
Funding Source: Publication of this article was supported by Boehringer Ingelheim Pharmaceuticals, Inc. and Lilly USA, LLC.
Author Disclosures: Dr Kumar and Dr Psotka report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Analysis and interpretation of data (SK, MAP); drafting of the manuscript (SK, MAP); and critical revision of the manuscript for important intellectual content (SK, MAP).
Address Correspondence to: Mitchell A. Psotka, MD, PhD, Inova Heart and Vascular Institute, 3300 Gallows Rd, Falls Church, VA 22042.
Email: Mitchell.Psotka@inova.org