HFpEF: The Heart Failure With a Normal Ejection Fraction That Is Twice as Common in Women

She had been telling her primary care physician for two years that she could not walk to the end of the block without stopping to catch her breath. She had gained eight pounds without changing her diet. Her ankles were swollen by evening. The echocardiogram report came back and the attending read the key line: ejection fraction 60%. Normal. She was told her heart was fine. Her ejection fraction was normal. Her heart was failing. These are not the same thing.

The Mechanism

Ejection fraction measures the proportion of blood the left ventricle pumps out with each contraction. An EF of 60% means the ventricle is squeezing 60% of its volume into the aorta with each beat. That number is normal, and that number is also the source of a pervasive diagnostic error.

Heart Failure with Preserved Ejection Fraction (HFpEF) is a failure of the other half of the cardiac cycle: the relaxation phase, called diastole. After each contraction, the left ventricle must actively relax and refill. That relaxation is not passive. It is an energy-dependent process that requires the cardiomyocyte to release calcium from the contractile apparatus and allow the sarcomere to lengthen again. When that process is impaired, the ventricle resists filling. The pressure required to push blood into a stiff ventricle rises. On echocardiography, diastolic dysfunction is graded using the E/A ratio, which reflects the ratio of early passive filling (E wave, driven by ventricular relaxation) to late filling from atrial contraction (A wave). In a healthy young heart, E exceeds A. In early diastolic dysfunction, the E/A ratio inverts as the ventricle depends increasingly on atrial contraction to fill adequately. In advanced diastolic dysfunction, the E/A ratio paradoxically normalizes but the E/e’ ratio rises, reflecting elevated filling pressures rather than normal physiology. Left atrial volume index increases as the atrium dilates to compensate for chronically elevated filling pressures. The left atrium, in a real clinical sense, is the scar tissue of poorly treated diastolic dysfunction.

Filling pressures elevate. Fluid backs up into the pulmonary vasculature and then into the lungs. The cardiac output is insufficient for demands above minimal exertion. The patient reports that she cannot climb stairs anymore. The echocardiogram report says the EF is normal.

Concentric hypertrophy from pressure overload. When the heart works against chronically elevated afterload, typically from hypertension, the left ventricle compensates by thickening its walls. This is called concentric hypertrophy. A thicker wall is not a stronger wall in the relevant sense: it is a stiffer wall. The cavity does not dilate; the wall thickens inward. A ventricle with concentric hypertrophy has less room to fill and greater resistance to filling, which compounds the diastolic dysfunction. The EF remains normal or even elevated because the thick-walled ventricle squeezes efficiently. The pump looks fine. The filling is the problem.

The role of titin. At the molecular level, much of the myocardial stiffness in HFpEF is mediated by titin, a giant sarcomeric protein that functions as a molecular spring governing the passive elasticity of the cardiomyocyte. In a compliant ventricle, titin allows the sarcomere to stretch during diastole and recoil during systole. With age, and particularly following the loss of estrogen at menopause, titin undergoes isoform shifts and reduced phosphorylation that make it stiffer. A 2013 framework published by Paulus and Tschope in the Journal of the American College of Cardiology described the specific pathway: systemic inflammation arising from metabolic syndrome impairs coronary microvascular endothelial function. Reduced nitric oxide bioavailability in the myocardial microvasculature creates oxidative stress within cardiomyocytes. That oxidative stress suppresses protein kinase G (PKG) activity. PKG normally phosphorylates titin, keeping it in a compliant, spring-like state. When PKG activity falls, titin becomes underphosphorylated and the sarcomere stiffens at rest. The ventricle resists filling not because of large-vessel disease or pressure overload alone, but because the molecular architecture of the cardiomyocyte has changed at a level that no coronary CT or stress test will detect.

This pathway explains a clinically important patient who is frequently missed: the woman in her early 60s, obese, with borderline blood pressure, a normal coronary calcium score, and no macrovascular disease, who nevertheless cannot walk two blocks without stopping. Her large vessels are intact. Her sarcomeres are not.

Why women are the predominant population. Women have 2.8 times higher odds of developing HFpEF relative to HFrEF when compared to men with heart failure. Several converging mechanisms explain this.

Post-menopausal arterial stiffening is the most established driver. Pulse wave velocity, the standard measure of aortic stiffness, increases substantially across the menopause transition as estrogen declines. Stiffer arteries impose greater pressure on the left ventricle with each beat. The ventricle responds with concentric hypertrophy, and the hypertrophied ventricle fills poorly. The pathway from estrogen loss to HFpEF runs through aortic stiffness and into the left ventricular wall.

Women also develop hypertension at higher rates than men after age 65, and poorly controlled hypertension is the single strongest modifiable risk factor for HFpEF. The path from uncontrolled blood pressure in the early 60s to symptomatic HFpEF in the 70s is both biologically plausible and epidemiologically documented.

The metabolic syndrome pathway disproportionately affects women in a way that is not simply explained by prevalence differences. Women appear to develop greater myocardial inflammatory burden per unit of metabolic risk relative to men, a sex-specific amplification of the Paulus-Tschope mechanism described above.

What the Evidence Shows

Part of why the HFpEF diagnostic gap is so persistent is historical: cardiology spent most of the twentieth century studying, treating, and building clinical intuition around heart failure with reduced ejection fraction (HFrEF), the variety where the pump is visibly weak. HFpEF received far less research attention because its defining feature is a normal-appearing echocardiogram.

A landmark 2006 paper by Owan and colleagues in the New England Journal of Medicine documented that HFpEF prevalence had been rising steadily in the Olmsted County population-based cohort while HFrEF prevalence remained stable, and that women comprised the majority of HFpEF cases. The implication was clinically important: the form of heart failure least studied and most frequently dismissed with a reassuring “your EF looks fine” was also the form that disproportionately affected women.

Mortality data reinforced the urgency. Five-year mortality in HFpEF runs approximately 50 to 60 percent, comparable to many malignancies. This is not a benign condition with a favorable trajectory. Most excess mortality is cardiovascular: sudden cardiac death, progression to reduced-EF failure, and cardiovascular events from comorbid conditions. HFpEF is not a diagnosis that warrants reassurance.

The failed trials and what they revealed. For most of cardiology’s history, HFpEF had no disease-modifying treatment. Therapies that dramatically improved outcomes in HFrEF, including ACE inhibitors, beta-blockers, and angiotensin receptor-neprilysin inhibitors, all failed to show mortality benefit in HFpEF trials. The TOPCAT trial evaluated spironolactone in HFpEF and returned a neutral overall result, though subgroup analyses of patients enrolled in the Americas showed a survival benefit while patients enrolled in Russia and Georgia did not. The Americas subgroup finding was hypothesis-generating and suggested the drug might benefit the right population if properly identified, but the overall trial did not support routine use. TOPCAT’s failure was both a clinical dead end and, in retrospect, a signpost: HFpEF is a heterogeneous syndrome, and the therapies that work in one phenotype may not generalize across the whole population. The path to a positive trial required either finding the right drug or identifying the right population.

SGLT2 inhibitors: the first drugs to work. Sodium-glucose cotransporter-2 inhibitors are now the first drug class with demonstrated outcome benefit in HFpEF across major clinical trials.

The EMPEROR-Preserved trial, published in 2021, evaluated empagliflozin versus placebo in 5,988 patients with HFpEF (EF above 40%). The primary composite endpoint of cardiovascular death and heart failure hospitalization was reduced by 21% relative risk with empagliflozin. This was the first positive major outcomes trial in HFpEF in the history of the disease.

The DELIVER trial, published in 2022, evaluated dapagliflozin in 6,263 patients with HFpEF (EF above 40%) and confirmed the class effect across a broad range of ejection fractions. The primary composite of worsening heart failure or cardiovascular death was significantly reduced. The benefit was consistent across subgroups including women, patients without diabetes, and patients across the spectrum of EF above 40%.

The mechanism of benefit is not simply diuresis. SGLT2 inhibitors appear to reduce myocardial inflammation, improve cardiomyocyte energetics, reduce cardiac fibrosis, and lower epicardial adipose tissue volume, all effects that bear directly on the mechanisms driving HFpEF. The benefit in patients without diabetes supports the conclusion that the cardiac effects are independent of glucose lowering.

GLP-1 agonists in obese HFpEF. The STEP-HFpEF trial, published in 2023, evaluated semaglutide in 529 patients with HFpEF and obesity (EF at or above 45%, BMI at or above 30). Semaglutide produced substantial improvements in symptoms measured by the Kansas City Cardiomyopathy Questionnaire, clinically meaningful increases in six-minute walk distance, and reductions in NT-proBNP and C-reactive protein. The trial was not powered to detect mortality differences, but the symptom and functional benefit was consistent and clinically meaningful. For a woman with HFpEF and a BMI above 30, the STEP-HFpEF data support a direct conversation about GLP-1 agonist therapy as part of the initial management plan.

Exercise training. Controlled trials of supervised exercise training in HFpEF show improvements in peak VO2 of approximately 2 ml/kg/min. This magnitude of change is clinically significant: it corresponds to a meaningful improvement in functional capacity and reduces the exercise intolerance that is the dominant symptom limiting quality of life. Cardiac rehabilitation programs that include HFpEF protocols are the appropriate referral for eligible patients.

The symptoms that precede the diagnosis. HFpEF does not announce itself with the sudden chest pain of a myocardial infarction. Its trajectory is gradual, and in women it is particularly likely to be attributed to deconditioning, aging, perimenopause-related fatigue, or anxiety before a cardiac diagnosis is pursued.

The signature symptom is exertional dyspnea disproportionate to the level of activity. A flight of stairs that previously required no particular effort now produces breathlessness that takes minutes to resolve. A walk that was routine now requires stopping. The breathlessness resolves with rest because the cardiac output at rest is adequate; the problem is a heart that cannot increase output to meet the demands of activity.

Orthopnea and paroxysmal nocturnal dyspnea, which is needing more pillows to breathe comfortably or waking from sleep with breathlessness that resolves when sitting up, reflects elevated pulmonary venous pressures when supine. Lower extremity edema by evening, which resolves partially overnight, reflects the elevated venous pressures that accompany diastolic dysfunction. Fatigue that is not explained by sleep deprivation, thyroid dysfunction, or iron deficiency reflects a heart that cannot increase its output adequately during the metabolic demands of daily activity.

Each of these individually is nonspecific. The pattern together, particularly in a woman over 55 with hypertension and metabolic risk factors, warrants echocardiography with diastolic function assessment, BNP or NT-proBNP measurement, and cardiologist referral without delay.

What to Do This Week

1. Request your NT-proBNP. This blood test is not routine at most primary care visits. Ask for it by name. A result above 125 pg/mL in the outpatient setting warrants cardiology referral. In obese patients, natriuretic peptide levels are suppressed by adipose tissue, so high-normal results still merit attention when symptoms are present.

2. Ask whether your echocardiogram included diastolic function grading. An echo report that mentions E/e’ ratio, left atrial volume index, and tissue Doppler imaging has assessed diastolic function. A report that mentions only ejection fraction has not. If your echo lacked diastolic assessment, ask for a repeat study that includes it.

3. Ask about exercise stress echocardiography if your resting echo was normal or mildly abnormal but your symptoms are significant. This test, conducted with graded exercise rather than at rest, reveals elevated filling pressures that resting studies miss entirely. It requires a specific order and is not performed routinely.

4. Track your weight daily. Weigh yourself each morning after voiding and before eating. A gain of two or more pounds overnight is a direct signal of elevated cardiac filling pressures. Keep a written log to bring to every appointment. The number, not how you feel, is the early warning system.

5. Ask your cardiologist directly about an SGLT2 inhibitor. If you have confirmed HFpEF and are not currently taking one, ask whether it is appropriate for you. The EMPEROR-Preserved and DELIVER trials support use regardless of diabetes status. If your provider is unfamiliar with these trials in the HFpEF context, a second opinion from a heart failure specialist is warranted.

The diagnosis of HFpEF is delayed, often for years, while patients are told their hearts are fine. The echocardiogram reporting a normal ejection fraction measures the right number for the wrong question. A woman who cannot walk to the end of the block deserves a workup that asks the right one.