Circulation Journal
Online ISSN : 1347-4820
Print ISSN : 1346-9843
ISSN-L : 1346-9843

This article has now been updated. Please use the final version.

Diagnostic Approaches to Heart Failure With Preserved Ejection Fraction (HFpEF) ― Identification of Unrecognized HFpEF ―
Takahiro OkunoKazuki KagamiTomonari HaradaMasaaki HoshigaHideki IshiiMasaru Obokata
Author information
JOURNAL OPEN ACCESS FULL-TEXT HTML Advance online publication

Article ID: CJ-25-0214

Details
Abstract

The development of life-saving pharmacotherapies such as the sodium-glucose cotransporter-2 inhibitors has changed heart failure with preserved ejection fraction (HFpEF) into a treatable disease. This paradigm shift in treatment has made the diagnosis of HFpEF more important. However, HFpEF is underdiagnosed in primary and secondary/tertiary care settings due to its diagnostic difficulties. Particularly, HFpEF is often missed in patients with obesity or atrial fibrillation. This review describes the reasons for the difficulty in diagnosing HFpEF and proposes a 5-step approach to identifying HFpEF in patients with unexplained dyspnea. Primary care physicians play a key role in the early identification of HFpEF in the community. We also discuss potential approaches to enhancing community referral and thus improving the rate of HFpEF diagnosis.

Heart failure (HF) is a substantial public health problem, with a worldwide prevalence of over 64 million.1,2 In Japan alone, there are an estimated 1.2 million HF patients (prevalence rate ≈1%), and that number is projected to increase to 1.3 million by 2035, despite a declining overall population.3,4 The incidence and prevalence of HF with preserved ejection fraction (HFpEF) relative to HF with reduced EF (HFrEF) has been increasing, possibly due to the aging of the general population and the increasing burden of comorbidities such as diabetes mellitus, obesity, metabolic syndrome, atrial fibrillation (AF) and chronic kidney disease.2,58 A series of large epidemiologic studies in Japan have shown that nearly 70% of patients with HF have a preserved EF, suggesting that HFpEF is a major driver of the increase in HF, the so-called HF pandemic.9

With the emergence of disease-modifying therapies such as the sodium-glucose cotransporter-2 (SGLT2) inhibitors, HFpEF has now become a treatable disease,1015 which has made the accurate and timely diagnosis of HFpEF much more important than ever before.16 However, HFpEF remains often underdiagnosed or even neglected in practice, leading to undertreatment.1618 Underdiagnosis can occur in both the primary and secondary/tertiary care settings. In this review, we summarize the current understanding of the diagnosis of HFpEF and discuss potential approaches to preventing HFpEF underdiagnosis.

Paradigm Shift in the Diagnosis of HFpEF

The lack of effective treatment has been the primary problem in HFpEF clinical practice. Most clinical trials evaluating the efficacy of treatments for HFpEF to date have shown neutral results,19,20 and the management of HFpEF has focused on targeting congestion and comorbidities.2123 The development of effective treatments has dramatically changed this disease into a treatable condition.24 SGLT2 inhibitors, such as empagliflozin and dapagliflozin reduced cardiovascular deaths and HF hospitalization in the EMPEROR-Preserved and DELIVER trials.10,11 In the STEP-HFpEF trial, the glucagon-like peptide-1 (GLP-1) receptor agonist semaglutide reduced symptoms and physical limitations, and improved exercise capacity in patients with HFpEF and BMI ≥30 kg/m2.13 In addition, the dual agonist of the glucose-dependent insulinotropic polypeptide (GIP)/GLP-1 receptors tirzepatide has been recently shown to reduce cardiovascular death and worsening HF in patients with the HFpEF obesity phenotype.14

Because HF is a progressive syndrome, any delay in appropriate diagnosis and treatment leads to worse clinical outcomes.23,25 Guidelines have advocated early screening and identification for HF in high-risk patients, such as those with type II diabetes and hypertension.26,27 The clinical need for improved HFpEF identification is emphasized by the development of life-saving pharmacotherapies, but despite this, HFpEF remains largely underdiagnosed.17,18 Recent data showed that up to one-quarter of elderly patients with unexplained dyspnea in the community may have unrecognized HFpEF.17,18 Importantly, these patients experienced worse clinical outcomes.17 Given that unexplained dyspnea is a common presentation in clinical practice, this suggests that there are a substantial number of patients with unrecognized HFpEF in the community. Lack of diagnosis may lead to serious delays or even failure to deliver the evidence-based treatments to patients, eventually leading to poor clinical outcomes.25

Key 5 Steps in the Diagnostic Approach to HFpEF

The underdiagnosis of HFpEF may be related to its diagnostic difficulties (Figure 1). Here, we propose a 5-step approach to identify HFpEF among individuals with dyspnea. This approach is primarily intended to be applied in secondary or tertiary hospitals where echocardiography is available, but the first step may be helpful for primary care physicians.

Figure 1.

The 5-step approach to diagnosing HFpEF. Echo, echocardiography; EF, ejection fraction; HFpEF, heart failure with preserved ejection fraction; LVFP, left ventricular filling pressure.

1. Rule Out Noncardiac Mimics The diagnosis of HFpEF is straightforward when patients with symptoms or physical signs of HF (e.g., dyspnea, fatigue or peripheral edema), have objective evidence of congestion (e.g., pulmonary congestion or bilateral pleural effusion on chest radiography, elevated natriuretic peptide [NP] levels, or echocardiographic signs of congestion [e.g., elevated E/e′]).28 However, the diagnosis of HFpEF is challenging in patients with less congestion.29 This is because the symptoms or signs of HF are not specific to HFpEF and may be seen in patients with noncardiac diseases, such as lung diseases, anemia, severe obesity, renal disease, liver disease or even deconditioning.30 In addition, EF is normal in patients with HFpEF (by definition) and therefore similar to that of patients with noncardiac diseases. It can be difficult to differentiate these conditions. Therefore, the first action to diagnose HFpEF is exclusion of noncardiac diseases, which requires detailed medical interviews, blood and urinary tests (NPs, blood cell counts, liver and kidney function, D-dimer, proteinuria, albuminuria), chest X-rays, electrocardiography (ECG), pulmonary function test, and standard echocardiography. Specifically, the measurement of NPs is the most important screening tool to help rule in or rule out HFpEF.31,32 Given their high sensitivity, normal NP levels are associated with a low probability of HFpEF, but NP levels are often low or even normal in patients with obesity.3236 It should be noted that noncardiac diseases and HFpEF often coexist in patients, such as those with chronic obstructive lung disease, anemia, and chronic kidney disease.

2. Identify Elevated LV Filling Pressures After the exclusion of noncardiac mimics, echocardiography plays a key role in the diagnosis of HFpEF.37,38 The primary role of echocardiography is to objectively identify signs of elevated LV filling pressures through markers of LV diastolic function, including the transmitral flow pattern, early diastolic mitral tissue (e′) velocity, the ratio of early diastolic mitral inflow velocity to e′ velocity (E/e′ ratio), tricuspid regurgitation velocity, left atrial (LA) volume, and pulmonary venous flow pattern.3941 It is noteworthy that most of these indices are highly specific to identify elevated LV filling pressures and thus can be used to rule-in HFpEF.38 There are emerging echocardiographic parameters for the diagnosis of HFpEF, such the LA reservoir strain.42,43 Previous studies have reported a high correlation between LA reservoir strain and LV filling pressure.44,45 In the setting of worsening HF, point-of-care ultrasound is particularly useful in quickly identifying surrogates of elevated LV filling pressures,46 such as diffuse ultrasound B-lines, inferior vena cava morphology or pleural effusion.4750 The recently reported time difference between mitral and tricuspid valve opening may also be a useful marker.51

3. Don’t Use Echocardiography to Rule Out HFpEF The primary limitation of echocardiographic markers of LV diastolic function is low sensitivity for identifying elevated LV filling pressure.29,38 The E/e′ ratio, one of the most robust echocardiographic markers for HFpEF, has been reported to be poorly sensitive for detecting elevated LV filling pressures in patients with normal EF (reported sensitivity 0–70%).38 The current guidelines from the American Society of Echocardiography and the European Association of Cardiovascular Imaging (ASE/EACVI) recommend a combination of echocardiographic parameters for evaluating LA pressure to complement the limitations of low sensitivities of individual components of LV diastolic function, but even this has been reported to have low sensitivity.29,39,52,53 Therefore, it is important to be aware that negative findings for these parameters are insufficient to exclude HFpEF in patients with intermediate probability. For example, a normal E/e′ ratio in a 75-year-old patient with dyspnea and AF (i.e., H2FPEF score 4 points) would not be sufficient to rule out the possibility of having HFpEF.

4. Exclude HFpEF Mimics The next step is to rule out cardiac HFpEF mimics (or masqueraders) with specific etiologies such as valvular heart disease, cardiac amyloidosis, cardiac sarcoidosis, pericardial disease, hypertrophic cardiomyopathy, ischemic heart disease and pulmonary artery hypertension.36,54,55 This step is critical because HFpEF mimics may have disease-specific therapies.37,56 We propose the mnemonic “VASCHIP” to remind clinicians of the typical and common HFpEF mimics (Figure 2). Echocardiography has an important role in identifying these HFpEF mimics and there are key clues (Figure 2).37,38,5759 Further workup may be required, including right heart catheterization, transesophageal echocardiography, cardiac magnetic resonance imaging, coronary angiography, scintigraphy or cardiac computed tomography.60

Figure 2.

“VASCHIP” to exclude HFpEF mimics and their typical echocardiographic clues. IVC, inferior vena cava; LV, left ventricular; LVFP, left ventricular filling pressure; LVOTO, left ventricular outflow obstruction; SAM, systolic anterior motion of the mitral valve.

Cardiac amyloidosis is increasingly recognized as an important HFpEF mimic in practice due to its high prevalence and the development of disease-specific pharmacotherapies.56,61 Previous studies reported that 8–14% of patients originally considered to have typical HFpEF had evidence of cardiac amyloidosis based on scintigraphy or histopathology.61,62 In an autopsy report from Japan, transthyretin amyloid deposition was found in 11.5% of older patients (>80 years).63 Careful evaluation should be taken in patients with “red-flag” symptoms or signs of cardiac amyloidosis.64 A simpler diagnostic tool based on high-sensitivity troponin T, QRS duration, and LV posterior wall thickness is also available.65

5. Consider Exercise Stress Testing Patients with HFpEF often have normal LV filling pressure at rest and only develop abnormalities during exertion.29,6669 If cardiac filling pressures are normal at rest in these patients, there will be no evidence of congestion on physical examination or noninvasive surrogates such as chest X-ray, NP levels or echocardiography, making the identification of HFpEF more challenging.29 Exercise testing is necessary to detect the abnormalities that develop only during exercise, thereby improving the diagnosis.29,67,70 In a previous study involving 414 consecutive patients with unexplained dyspnea, 45% of patients were found to have elevated LV filling pressures only during exercise, suggesting that almost half of patients will be missed when relying solely on resting assessments.71 Invasive hemodynamic exercise testing (or exercise right heart catheterization) allows for direct measurements of intracardiac pressures at rest and during exercise and now serves as the gold standard test to diagnose HFpEF.40,71 The diagnosis of HFpEF is hemodynamically defined by pulmonary capillary wedge pressure (PCWP) ≥15 at rest and/or ≥25 mmHg during exercise.40,71

Exercise stress echocardiography is an alternative approach to invasive hemodynamic exercise testing by estimating the elevation in intracardiac pressures during exercise, and has been reviewed in detail.29,69 Being noninvasive, exercise stress echocardiography is increasingly performed in clinical practice.7275 Exercise stress echocardiography is most useful in patients with an intermediate probability of HFpEF based on clinical history, physical examination, NP levels, chest X-rays, ECG, and resting echocardiography. Two methods are also available to estimate the probability of HFpEF in ambulatory patients with symptoms or signs of HF: the H2FPEF score (age, body mass index, hypertensive drugs, AF, E/e′ ratio, and PASP) and HFA-PEFF algorithm (functional, morphological, and biomarker domains).40,71,76,77 Patients with an H2FPEF score of 2–5 or an HFA-PEFF score of 2–4 are considered reasonable candidates for exercise stress echocardiography. On the other hand, exercise stress echocardiography is not required for patients with either very low or high probability of HFpEF. Diagnosis of HFpEF is challenging if exercise stress echocardiography is not available. Given a reasonably high correlation between resting LA reservoir strain and PCWP during ergometry exercise,45 reduced LA reservoir strain may be used to identify HFpEF under such circumstances.

Two Comorbidities That Complicate HFpEF Diagnosis

Obesity and AF are 2 common comorbidities that complicate the diagnosis of HFpEF (Figure 3). Diagnosis of HFpEF requires objective evidence of congestion, which is most commonly identified by NP levels and echocardiography. Obesity is associated with lower NP levels and they are low or often normal in patients with obese HFpEF.33,78,79 Importantly, NP levels underestimate the severity of HF and clinical outcomes in obese HFpEF (i.e., clinical outcomes are worse in obese HFpEF than in non-obese HFpEF patients at the same NP levels).80 Recent data have shown that echocardiographic markers such as the E/e′ ratio and LA volume index will underestimate circulatory congestion in obese patients with HFpEF.34,78 For these reasons, it is often challenging to diagnose HFpEF in obese patients.

Figure 3.

Obesity and atrial fibrillation (AF) are 2 comorbidities that make the identification of HFpEF challenging. Natriuretic peptide (NP) levels and even echocardiographic markers often underestimate the presence or severity of HF. In contrast, NP levels may be elevated and LA may be dilated in many patients with AF, even in the absence of HF. Echo, echocardiography; HFpEF, heart failure with preserved ejection fraction; LA, left atrial; NP, natriuretic peptide; pts, patients.

In contrast to obesity, AF is well known to be associated with elevated NP levels even in the absence of HFpEF.31,81,82 Some clinicians assume that the NP levels will be elevated due to AF itself, not HFpEF. Similarly, patients with AF often have a dilated LA, making it difficult to determine whether it is due to AF or HFpEF. Clinical trials commonly use higher NP levels and larger LA volume index to establish the diagnosis of HFpEF in patients with AF.10,11 However, recent data revealed that >90% of patients presenting with unexplained dyspnea and AF had HFpEF.83 These findings suggest that the presence of AF is sufficiently diagnostic for HFpEF among individuals presenting with unexplained dyspnea.84

Appropriate Referrals and Diagnosis in the Community

Primary care physicians play an important role in identifying patients with or at risk for HFpEF in the community.85,86 As discussed before, the diagnosis of HFpEF is particularly challenging in primary care settings due to limited medical resources, especially the availability of echocardiography. The H2FPEF score and the HFA-PEFF algorithm have been developed to aid in diagnosing HFpEF,40,71 the need for definitive echocardiographic assessment may limit widespread use by primary care physicians. There are racial or regional variations in the clinical characteristics and comorbidities in HFpEF, particularly obesity.7 Obesity is much less common in Japanese patients with HFpEF (~6.5%) than in Americans (~80%).2,7,33,36,87 The H2FPEF score emphasizes the presence of obesity (2 points for BMI >30 kg/m2),71,76 and its diagnostic performance may be negatively affected in less obese populations such as the Japanese.7 Potentially, this will also be a problem for the newly developed HFpEF-ABA score, which is based solely on clinical variables (age, BMI, and history of AF).88

Diagnostic uncertainty in primary care affects referral to secondary/tertiary hospitals for further evaluation. Several approaches could help solve this problem. First, a collaborative approach between primary care physicians and secondary or tertiary hospitals may enhance referral and identification of HFpEF in the local community. In this regard, we recently reported that the rate of community referrals nearly doubled and the prevalence of HFpEF diagnosis trended up from 32% to 39% in a collaborative project between primary care physicians and a dyspnea clinic.86

Second, primary care tools to aid screen HFpEF that do not require echocardiography could facilitate referral and thus early identification of HFpEF in the community. Ideally, such tools should fit the clinical characteristics of the patients in the regions where they are used. Our group has recently developed and validated a scoring system that does not include echocardiographic variables for HFpEF screening of patients with shortness of breath undergoing exercise stress echocardiography (281 HFpEF and 341 controls). It is based on 7 clinical characteristics (age ≥65 years, coronary artery disease, elevated NP levels, anemia, cardiomegaly on chest X-ray, LV high-voltage on ECG, and AF; termed the BREATH2 score, Figure 4).89 In our study, the BREATH2 score accurately discriminated HFpEF from controls (area under the curve [AUC] 0.83, P<0.0001) and classified each patient into different risk categories of having HFpEF, ranging from 4% to 93%. Importantly, this score consists of widely available parameters in clinical practice and thus will be readily applicable to primary care physicians. The BREATH2 score allows them to estimate the pretest probability of HFpEF and consider referral to a secondary or tertiary hospital.

Figure 4.

The BREATH2 score to screen HFpEF. This score can be applied to patients with shortness of breath or unexplained dyspnea. It assesses the probability of having HFpEF in each individual. Using BREATH2, primary care physicians can assess which patients should be referred to a secondary or tertiary hospital. Patients with a high BREATH2 score (6–9 points) should be considered for referral to confirm the diagnosis of HFpEF and additional workup, given the higher probability of having HFpEF (77–93%). Patients with an intermediate score (4–5 points) would also be appropriate for referral, as they have a reasonable probability of having HFpEF (50%). Patients with a low score (0–3 points) are unlikely to have HFpEF (4–19%) and referral may not be required. AF, atrial fibrillation; BNP, B-type natriuretic peptide; CAD, coronary artery disease; HFpEF, heart failure with preserved ejection fraction; NT-proBNP, N-terminal pro-B-type natriuretic peptide.

There are other potential strategies to improve HFpEF referrals from the community. Dedicated dyspnea clinics, providing a multidisciplinary workup including exercise stress testing, are gaining attention for the identification of HFpEF and investigating other cardiac and noncardiac causes of unexplained dyspnea.90,91 The implementation of artificial intelligence may allow earlier detection of HFpEF through biological information obtained from a wearable device, chest X-ray or ECG.9297

Conclusions and Future Direction

HFpEF is the dominant form of HF worldwide, contributing to “the HF pandemic”. Despite increasing awareness and the development of life-saving pharmacotherapies, HFpEF remains underdiagnosed or often neglected in both the primary and secondary/tertiary care settings because of its diagnostic difficulties (Central Figure). Our novel 5-step approach will help clinicians identify HFpEF among individuals presenting exertional dyspnea. It is important to exclude HFpEF mimics that may have specific treatment and the “VASCHIP” approach may be useful. In the community, primary care physicians are often at the forefront of patients at risk for HFpEF and thus play a key role in identifying HFpEF. The H2FPEF and HFA-PEFF algorithms may be helpful for primary care physicians for this purpose, but the requirement of echocardiography and concerns about their applicability generated in Western populations may limit its broad application in practice. The BREATH2 score will be readily applicable for primary care physicians to consider referral to a secondary or tertiary hospital. Continued efforts are required to provide better diagnosis and treatment for people with HFpEF to improve their quality of life and clinical outcomes.

Acknowledgments

M.O. received research grants from the Japanese Circulation Society, Japanese College of Cardiology, AMI Inc, Nippon Boehringer-Ingelheim, Eli Lilly, Janssen Pharmaceutical K.K., JSPS KAKENHI (21K16078), and AMED (23jm0210104 h0002).

Funding

None.

Disclosures

M.O. received speaker honoraria from Novartis, Otsuka Pharmaceutical, AstraZeneca, Eli Lilly, and Nippon Boehringer-Ingelheim. H.I. received lecture fees from AstraZeneca Inc., Bayer Pharmaceutical Co., Ltd., Boehringer-Ingelheim Japan, Bristol-Myers Squibb Inc., Daiichi-Sankyo Pharma Inc., MSD K. K., Mitsubishi Tanabe Pharma Co., Ltd., Mochida Pharmaceutical Co., Ltd., Novartis Japan, and Pfizer Japan Inc. H.I. is an Associate Editor of Circulation Journal.

References
 
© 2025, THE JAPANESE CIRCULATION SOCIETY

This article is licensed under a Creative Commons [Attribution-NonCommercial-NoDerivatives 4.0 International] license.
https://creativecommons.org/licenses/by-nc-nd/4.0/
feedback
Top