Abstract
Background: Heart failure with preserved ejection fraction (HFpEF) and chronic kidney disease (CKD) have a strong pathophysiological interrelationship, and their combination worsens prognosis. Summary: This article briefly reviews the bidirectional epidemiological burden and the pathophysiological interplay between HFpEF and CKD. It also discusses some of the controversial aspects regarding the diagnosis and screening of HFpEF in CKD patients and focuses on the most effective therapeutic approaches to improve symptoms and prognosis in this high-risk population. Key Messages: Due to its prevalence and prognostic significance, HFpEF screening should be considered in patients with CKD, with careful use of traditional diagnostic tools in this population. Optimal medical therapy has seen major recent advances in patients with both HFpEF and CKD. SGLT2 inhibitors, finerenone, and semaglutide have consistently demonstrated cardio- and renoprotective effects in both conditions.
Introduction
Heart failure with preserved ejection fraction (HFpEF) and chronic kidney disease (CKD) share common risk factors and comorbidities, have a strong pathophysiological interrelationship, and thus frequently coexist [1]. This growing population with concomitant HFpEF and CKD has poorer functional capacity, worse quality of life, and a higher risk of adverse outcomes [2, 3]. However, several important knowledge gaps remain. How do the common risk factors of both conditions contribute to prognosis and disease progression? How should HFpEF be diagnosed in CKD patients? Are natriuretic peptides (NPs) and imaging techniques as useful in this population as they are in patients without renal dysfunction? How can life-saving therapies be implemented in this group? How can multidisciplinary and coordinated care be developed? A better understanding of these issues could provide a valuable opportunity to address this major public health problem [4]. This article briefly reviews the bidirectional epidemiological burden and pathophysiological interplay shared by HFpEF and CKD, discusses some of the controversial aspects regarding HFpEF diagnosis and screening in CKD patients, and focuses on the most effective therapeutic approaches to improve symptoms and prognosis in this high-risk population.
A Bidirectional Relationship
The Epidemiological Burden of CKD among HFpEF Patients
It is well known that CKD is a common condition in HFpEF. In recent landmark trials, such as Dapagliflozin Evaluation to Improve the Lives of Patients with Preserved Ejection Fraction Heart Failure (DELIVER) and Finerenone Trial to Investigate Efficacy and Safety Superior to Placebo in Patients with Heart Failure (FINEARTS-HF), nearly half of the patients included had an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 [5, 6]. This finding is consistent with data from real-world registries, where the percentage may rise to 70% when considering the presence of albuminuria in patients with an eGFR ≥60 mL/min/1.73 m2 [7]. Furthermore, patients with heart failure (HF) show a pronounced decline in eGFR over time, even after adjusting for other well-known progression risk factors. This decline is even more pronounced in patients with HFpEF [8]. Additionally, CKD is not a bystander in HFpEF; it significantly impacts prognosis. In a meta-analysis including 57 studies with 1,076,104 patients that investigated the effect of CKD on mortality risk in HF, moderate renal impairment (hazard ratio [HR] 1.59; 95% confidence interval [CI]: 1.49–1.69; p < 0.001) and severe renal impairment (HR, 2.17; 95% CI: 1.95–2.40; p < 0.001) were independent predictors of mortality [9]. Interestingly, the presence of CKD had a greater prognostic impact in patients with more preserved left ventricular ejection fraction (LVEF) [9].
The Epidemiological Burden of HFpEF among CKD Patients
HF prevalence increases in patients with CKD. According to the US Renal Data System annual report, in patients aged ≥66 years, the prevalence of HF ranges from 6.4% in non-CKD patients to 26.5% in stage 3 CKD and 40.2% in stage 4 or 5 CKD [10]. Interestingly, HFpEF appears to be the most prevalent form of HF, particularly among patients with end-stage renal disease (ESRD) [11, 12]. CKD is also associated with an increased risk of incident HF. In a contemporary pooled analysis from 3 community-based cohort studies, including 14,462 participants with and without CKD and no prevalent cardiovascular (CV) disease, patients with CKD stage 3 or higher (i.e., eGFR <60 mL/min/1.73 m2) had an increased risk of incident HF (adjusted risk difference 2.3; 95% CI: 1.2–3.3) [13]. This risk is higher at more advanced stages of CKD [14]. Finally, although HFpEF is associated with poor prognosis in both CKD and non-CKD patients, its detrimental prognostic significance is greater in patients with CKD [10].
The Pathophysiological Interplay
Several factors may explain the epidemiological relationship between CKD and HFpEF. First, both conditions share common risk factors such as age, obesity, and type 2 diabetes mellitus (T2DM). However, beyond these overlapping predisposing factors, there are true mechanistic pathways in which dysfunction in one organ ultimately affects the other [15]. One distinctive feature of HFpEF pathogenesis in CKD patients is a chronic inflammatory state that leads to endothelial dysfunction, which ultimately contributes to myocardial remodeling, hypertrophy, and fibrosis [1]. Indeed, a decline in eGFR is associated with an increase in inflammatory cytokines. This systemic proinflammatory state affects the coronary microvascular endothelium, leading to increased production of reactive oxygen species, decreased nitric oxide availability, microvascular dysfunction, and the induction of hypertrophy and fibrosis [1, 16]. Mitochondrial dysfunction, specific profibrotic factors, and overactivation of the renin-angiotensin-aldosterone system (RAAS) also contribute to this process [17]. An accelerated CV aging process in CKD, triggered by arterial stiffness and calcification, further promotes the development of left ventricular hypertrophy by increasing afterload [18]. Additionally, an often underestimated factor in HF pathogenesis in CKD is congestion. This relationship has been increasingly recognized as a major contributor to renal dysfunction in HFpEF, owing to the so-called “renal tamponade hypothesis” [19]. This role is likely bidirectional, as impaired salt and water handling in advanced CKD may lead to increased preload, further promoting cardiac remodeling.
Diagnostic Approach to HFpEF in CKD Patients
The Symptomatic CKD Patient
The diagnosis of HFpEF relies on a combination of clinical signs and symptoms resulting from structural and/or functional cardiac abnormalities, supported by elevated NP levels or objective evidence of congestion, in the presence of an LVEF >50% [20‒22]. Two diagnostic scoring systems for HFpEF, the H2FPEF and HFA-PEFF scores, have been developed and may aid in the diagnostic evaluation of patients with CKD and exertional dyspnea [23, 24]. Both approaches are summarized in Figure 1. The H2FPEF score focuses on 4 comorbidities commonly associated with HFpEF (obesity, hypertension, atrial fibrillation, and advanced age) and includes 2 echocardiographic criteria, with a total score ranging from 0 to 9 [24]. In contrast, the HFA-PEFF score, proposed by the Heart Failure Association of the European Society of Cardiology (ESC), emphasizes echocardiographic measures of functional and morphological abnormalities, along with NP levels. Each of the scores overrepresents particular HFpEF populations, which may explain that only 72% of patients have concordant classification according to both systems [25]. In the CKD population, the H2FPEF score may select older patients with obesity and atrial fibrillation, conditions frequently encountered in this context. Conversely, the HFA-PEFF score may place greater weight on NP levels. While both scoring systems demonstrate similar diagnostic and prognostic utility, neither has been specifically validated in CKD. Currently, there is no definitive evidence favoring one system over the other in the context of cardiorenal syndrome [25, 26]. However, in our opinion, if HFA-FPEF is used, specific recommended adjustment of NT-proBNP levels according to recent consensus published by the HF Association of the ESC should be used. The latter suggests increasing cut point for NT-proBNP by 35% when eGFR is below 30 mL/min/1.73 m2, by 25% for eGFR between 30 and 45 mL/min/1.73 m2, and by 15% for eGFR between 45 and 60 mL/min/1.73 m2 [27].
HFpEF diagnosis in symptomatic patients with CKD. Based on diagnostic scores by Reddy et al. and Pieske et al. [23, 24]. AF, atrial fibrillation; BMI, body mass index; CKD, chronic kidney disease; LAVi, indexed left atrial volume; LV, left ventricle; LVMi, left ventricular mass index; PASP, pulmonary artery systolic pressure; RWT, relative wall thickness; SR, sinus rhythm; TRV, tricuspid regurgitation velocity.
HFpEF diagnosis in symptomatic patients with CKD. Based on diagnostic scores by Reddy et al. and Pieske et al. [23, 24]. AF, atrial fibrillation; BMI, body mass index; CKD, chronic kidney disease; LAVi, indexed left atrial volume; LV, left ventricle; LVMi, left ventricular mass index; PASP, pulmonary artery systolic pressure; RWT, relative wall thickness; SR, sinus rhythm; TRV, tricuspid regurgitation velocity.
Both scores may categorize many patients into the intermediate probability range. In such cases, invasive hemodynamic testing (right heart catheterization) can reveal elevated filling pressures and assist in diagnosing HFpEF. However, some patients with exertional dyspnea may exhibit normal filling pressures at rest, normal NP levels, and no signs of congestion, yet still show abnormal hemodynamic responses during exercise. In this subset of patients, right heart catheterization with exercise can uncover these abnormal responses and aid in HFpEF diagnosis [28]. Combining cardiopulmonary exercise testing with invasive hemodynamic measurements or echocardiography may help exclude noncardiac causes of functional limitation and identify cardiac contributors to dyspnea, such as reduced heart rate reserve, exercise-induced pulmonary hypertension, or impaired diastolic reserve [29]. These insights can inform treatment strategies [30].
Why to Approach Presymptomatic HFpEF in CKD
In CKD, the development of HFpEF should be understood as a continuum, where structural or functional cardiac abnormalities and biochemical alterations may exist before clinical signs and symptoms become apparent. According to the universal definition of HF, presymptomatic HF includes stages A and B. Stage A represents patients at risk of HF who have no current or prior symptoms or signs of HF, nor structural cardiac changes or elevated biomarkers of heart disease [20]. Although not explicitly addressed in the universal definition, patients with moderate to very high CKD risk should be considered at stage A HFpEF, emphasizing the importance of proactive risk factor management in this population. Stage B, referred to as pre-HF, encompasses patients with cardiac structural or functional abnormalities (pre-HF type 1), molecular alterations such as elevated NPs (pre-HF type 2 or heart stress), or the combination of both (pre-HF type 3 or subclinical HF) [31]. Pre-HF is particularly prevalent in CKD, with its frequency increasing as kidney function deteriorates and albuminuria levels rise [32]. For example, in the Copenhagen Chronic Kidney Disease Echocardiographic Study – a prospective cohort of outpatients aged 30–75 years across CKD stages G1 to G5 (predialysis) – stage B HF was identified in 35.3% of participants, even though NP levels were not assessed. Patients with stage B HF had a significantly higher incidence of all-cause mortality compared to stage A HF patients, with rates of 2.4 per 100 patient-years (95% CI: 1.6–3.5) versus 1.1 per 100 patient-years (95% CI: 0.7–1.7), respectively (p < 0.001) [32].
These findings underscore the importance of actively screening pre-HF in CKD patients to identify those at risk, facilitate timely interventions, and optimize follow-up strategies. Early detection can help manage risk factors effectively and determine whether further examinations or referral to a cardiologist is warranted [27, 31].
How to Approach HFpEF Screening in CKD
Diagnosing pre-HF or stage B HF relies on identifying structural and functional cardiac abnormalities and elevated NP levels. However, CKD complicates the interpretation of these components. A clearer understanding of these abnormalities in CKD enables a stepwise approach to asymptomatic CKD patients, progressing from detecting “heart stress” to identifying signs and symptoms of HFpEF (Fig. 2).
HFpEF screening in asymptomatic patients with CKD. AF, atrial fibrillation; CA125, carbohydrate antigen 125; CKD, chronic kidney disease; CPET, cardiopulmonary exercise testing; eGFR, estimated glomerular filtration rate; HF, heart failure; HFpEF, heart failure with preserved left ventricular ejection fraction; LA, left atrium; LV, left ventricle; NT-proBNP, amino-terminal pro-brain natriuretic peptide; PA, pulmonary artery; SR, sinus rhythm; TR, tricuspid regurgitation.
HFpEF screening in asymptomatic patients with CKD. AF, atrial fibrillation; CA125, carbohydrate antigen 125; CKD, chronic kidney disease; CPET, cardiopulmonary exercise testing; eGFR, estimated glomerular filtration rate; HF, heart failure; HFpEF, heart failure with preserved left ventricular ejection fraction; LA, left atrium; LV, left ventricle; NT-proBNP, amino-terminal pro-brain natriuretic peptide; PA, pulmonary artery; SR, sinus rhythm; TR, tricuspid regurgitation.
The initial step in this approach is the detection of a “stressed heart,” or pre-HF type 2, which requires elevated NP levels. Although NPs are typically elevated in advanced kidney dysfunction, these high concentrations should not be dismissed as normal in CKD, as they carry even greater prognostic significance than in patients with normal kidney function [33]. Due to their strong prognostic value and widespread availability, NP testing could be performed annually in asymptomatic CKD patients to identify heart stress as a gateway to further evaluations when rule-in concentrations are observed. For results in the gray zone, earlier retesting of NP levels may be warranted (Fig. 2). The <125 pg/mL NT-proBNP threshold for ruling out HF in nonacute settings remains applicable. However, in patients with significant kidney dysfunction, higher rule-in thresholds are needed, and recent guidelines propose stepwise adjustments to NT-proBNP cutoffs [27]. Importantly, due to limited evidence, NP testing for HF screening is discouraged in dialysis patients [34].
The second step involves identifying cardiac structural abnormalities, which are highly prevalent in CKD. Left ventricular hypertrophy, for instance, is found in 40% of CKD patients and up to 80% of those with ESRD [35]. Other abnormalities, such as those described by the Acute Dialysis Quality Initiative (ADQI) XI Workgroup, are nearly ubiquitous in CKD patients [36]. However, whether these abnormalities represent pre-HF type 1 remains debated, as some criteria have failed to predict HF hospitalization risk [36]. In our view, echocardiography should be considered a secondary diagnostic step for asymptomatic patients with rule-in NP concentrations to identify structural and functional cardiac abnormalities (Fig. 2) [21].
The final step involves identifying the signs and symptoms of HFpEF, such as exertional dyspnea or fatigue. These symptoms are nonspecific and can often be attributed to poor physical fitness or sarcopenia in CKD patients, which may contribute to the underdiagnosis of HFpEF and subsequent worse outcomes [37]. This has recently been shown in a prospective cohort study including 825 patients with CKD G1 to G5 stages in which 15.3% of patients were found to have stage C or D HF, in contrast to the initial 8.8% of patients that had a HF diagnosis coded at baseline [32]. This highlights the need for proactive HF screening in CKD patients, particularly when heart stress and cardiac structural abnormalities are present. In this context, cardiopulmonary exercise testing can help establish maximal aerobic capacity and differentiate between cardiac and noncardiac causes of functional limitation (Fig. 2).
Additionally, assessing congestion is critical. Emerging tools like point-of-care ultrasound, bioimpedance analysis, or biomarkers unaffected by renal dysfunction, such as carbohydrate antigen 125 (CA125), may aid in the early detection and characterization of congestion in CKD patients (Fig. 2) [38].
Treatment of Patients with Cardiorenal Disease and HFpEF
The management of concomitant HFpEF and CKD poses significant challenges in daily clinical practice due to fragmented specialized care and the underrepresentation of these patients in randomized clinical trials for HF. The kidney-related exclusion criteria from major HFpEF trials are summarized in Table 1. Furthermore, HF pharmacological therapies can influence renal physiology and potentially worsen kidney function, leading to the historical practice of using HF drugs at very low doses or avoiding them entirely in patients with CKD. Consequently, patients with CKD are less likely to initiate and adhere to guideline-directed medical therapy for HF [39]. However, recent advancements have improved the management of patients with HFpEF and CKD. Multidisciplinary teams and the establishment of dedicated cardiorenal units are being implemented to provide holistic and specialized care for patients with cardiorenal syndrome [40]. Additionally, emerging therapies have demonstrated dual benefits in slowing CKD progression while also improving symptoms, quality of life, and outcomes in patients with HFpEF [5, 6, 41‒43].
Main trials conducted in patients with HF with LVEF >40%
Trial . | Year . | Intervention . | Primary endpoint . | Primary results . | Renal exclusion criteria . | Mean eGFR . | Patients with CKD . | Kidney endpoint . | Subgroup analysis in patients with CKD . |
---|---|---|---|---|---|---|---|---|---|
Irbesartan in Heart Failure with Preserved Ejection Fraction Study (I-PRESERVE) | 2008 | Irbesartan target dose 300 mg daily vs. placebo | Death from any cause or hospitalization for a CV cause | No significant differences (hazard ratio [HR] = 0.95; 95% confidence interval [CI]: 0.86–1.05; p = 0.35) | Creatinine ≥2.5 mg/dL | 72 (57, 87) mL/min/1.73 m2 | 30% with eGFR <60 mL/min/1.73 m2 | No | No |
Candesartan in Heart Failure-Assessment of Reduction in Mortality and Morbidity (CHARM-Preserved) | 2003 | Candesartan target dose 320 mg daily vs. placebo | CV death or unplanned admission for worsening HF | No significant differences (HR = 0.89; 95% CI: 0.77–1.03; p = 0.118) | Creatinine >3 mg/dL | NA | 34.7% with eGFR <60 mL/min/1.73 m2 | No | There was no evidence that the beneficial effect of candesartan was modified by baseline eGFR |
Prospective Comparison of ARNI with ARB Global Outcomes in HF with Preserved Ejection Fraction (PARAGON-HF) | 2019 | Sacubitril/valsartan target dose 97/103 mg bid vs. placebo | Total hospitalizations for HF and death from CV causes | No significant differences (rate ratio, 0.87; 95% CI: 0.75–1.01; p = 0.06) | eGFR <30 mL/min/1.73 m2 | 63±19 mL/min/1.73 m2 | 47% with eGFR <60 mL/min/1.73 m2 | Secondary endpoint Composite renal endpoint, defined as follows:
| Reduction in kidney endpoint with sacubitril/valsartan in patients with and without diabetes (HR 0.54, 95% CI: 0.33–0.89, and HR 0.42, 95% CI: 0.19–0.91, respectively) |
Empagliflozin Outcome Trial in Patients with Chronic Heart Failure with Preserved Ejection Fraction (EMPEROR-Preserved) | 2021 | Empagliflozin 10 mg daily vs. placebo | CV death or hospitalization for HF | Reduction with empagliflozin (HR, 0.79; 95% CI: 0.69–0.90; p < 0.001) | eGFR < 20 mL/min/1.73 m2 or requiring dialysis | 60.6±19.8 mL/min/1.73 m2 | 53.5% had eGFR <60 mL/min/1.73 m2 or urine albumin-to-creatine ratio >300 mg/g | Secondary endpoint: GFR slope Prespecified analysis: composite renal endpoint | Empagliflozin slowed the slope of eGFR decline by 1.43 (1.01–1.85) mL/min/1.73 m2/yr in patients with CKD and 1.31 (0.88–1.74) mL/min/1.73 m2/yr in patients without CKD (interaction p = 0.70) Empagliflozin did not reduce the prespecified kidney outcome in patients with or without CKD (with CKD: HR 0.97, 95% CI: 0.71–1.34; without CKD: HR 0.92, 95% CI: 0.58–1.48; interaction p = 0.86) |
Dapagliflozin Evaluation to Improve the Lives of Patients with Preserved Ejection Fraction Heart Failure (DELIVER) | 2022 | Dapagliflozin 10 mg daily vs. placebo | Worsening HF or CV death | Reduction with dapagliflozin (HR, 0.82; 95% CI: 0.73–0.92; p < 0.001) | eGFR <25 mL/min/1.73 m2 | 61±19 mL/min/1.73 m2 | 49% had an eGFR <60 mL/min/1.73 m2 | Secondary endpoint: slope of eGFR Post hoc composite kidney endpoint | Dapagliflozin attenuated the decline in eGFR from baseline (difference, 0.5; 95% CI: 0.1–0.9 mL/min/1.73 m2 per yr; p = 0.01) No significant differences in kidney endpoint (HR, 1.08; 95% CI: 0.79–1.49) |
Semaglutide Treatment Effect in People with Obesity (STEP-HFpEF and STEP-HFpEF DM) | 2024 | Semaglutide target dose 2.4 mg sc weekly vs. placebo | Change from baseline to week 52 in KCCQ-CSS and bodyweight | Reduction with semaglutide KCCQ-CSS 7·5 points (95% CI: 5·3–9·8); p < 0·0001; difference in bodyweight at week 52 −8·4% (−9·2 to −7·5); p < 0·0001 | ESRD or dialysis | STEP-HFpEF DM 69.0 (50.0–88.0) mL/min/1.73 m2 | NA | No | NA |
Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist (TOPCAT) | 2014 | Spironolactone up to 45 mg daily vs. placebo | CV death, aborted cardiac arrest, or hospitalization for HF | No significant differences (HR, 0.89; 95% CI: 0.77–1.04; p = 0.14) | eGFR of <30 mL per minute per 1.73 m2 of body surface area or creatinine ≥2.5 mg/dL | 58.4±19.4 mL/min/1.73 m2 | 39% in both groups had eGFR < 60 mL/min/1.73 m2 | Post hoc analysis differences in eGFR slopes | Spironolactone led to a higher decline of rGFR over time −2.5 [−3.4 to −1.6] mL/min/1.73 m2/yr; p < 0.001 |
FINerenone trial to investigate Efficacy and sAfety superioR to placebo in paTientS with Heart Failure (FINEARTS-HF) | 2024 | Finerenone up to 20 mg daily if eGFR ≤60 mL/min/1.73 m2 or 40 mg daily if eGFR >60 mL/min/1.73m2 vs. placebo | Total worsening HF events and death from CV causes | Reduction with finerenone (rate ratio, 0.84; 95% CI: 0.74–0.95; p = 0.0079) | eGFR <25 mL/min/1.73 m2 | 62.1±19.7 mL/min/1.73 m2 | 48.1% with an eGFR <60 mL/min/1.73 m2 | Composite renal endpoint (defined as sustained decrease in eGFR ≥50% relative to baseline over at least 4 wk, sustained eGFR decline <15 mL/min/1.73 m2, or initiation of dialysis or renal transplantation) | No significant differences among groups (HR, 1.33; 95% CI: 0.94–1.89) |
Trial . | Year . | Intervention . | Primary endpoint . | Primary results . | Renal exclusion criteria . | Mean eGFR . | Patients with CKD . | Kidney endpoint . | Subgroup analysis in patients with CKD . |
---|---|---|---|---|---|---|---|---|---|
Irbesartan in Heart Failure with Preserved Ejection Fraction Study (I-PRESERVE) | 2008 | Irbesartan target dose 300 mg daily vs. placebo | Death from any cause or hospitalization for a CV cause | No significant differences (hazard ratio [HR] = 0.95; 95% confidence interval [CI]: 0.86–1.05; p = 0.35) | Creatinine ≥2.5 mg/dL | 72 (57, 87) mL/min/1.73 m2 | 30% with eGFR <60 mL/min/1.73 m2 | No | No |
Candesartan in Heart Failure-Assessment of Reduction in Mortality and Morbidity (CHARM-Preserved) | 2003 | Candesartan target dose 320 mg daily vs. placebo | CV death or unplanned admission for worsening HF | No significant differences (HR = 0.89; 95% CI: 0.77–1.03; p = 0.118) | Creatinine >3 mg/dL | NA | 34.7% with eGFR <60 mL/min/1.73 m2 | No | There was no evidence that the beneficial effect of candesartan was modified by baseline eGFR |
Prospective Comparison of ARNI with ARB Global Outcomes in HF with Preserved Ejection Fraction (PARAGON-HF) | 2019 | Sacubitril/valsartan target dose 97/103 mg bid vs. placebo | Total hospitalizations for HF and death from CV causes | No significant differences (rate ratio, 0.87; 95% CI: 0.75–1.01; p = 0.06) | eGFR <30 mL/min/1.73 m2 | 63±19 mL/min/1.73 m2 | 47% with eGFR <60 mL/min/1.73 m2 | Secondary endpoint Composite renal endpoint, defined as follows:
| Reduction in kidney endpoint with sacubitril/valsartan in patients with and without diabetes (HR 0.54, 95% CI: 0.33–0.89, and HR 0.42, 95% CI: 0.19–0.91, respectively) |
Empagliflozin Outcome Trial in Patients with Chronic Heart Failure with Preserved Ejection Fraction (EMPEROR-Preserved) | 2021 | Empagliflozin 10 mg daily vs. placebo | CV death or hospitalization for HF | Reduction with empagliflozin (HR, 0.79; 95% CI: 0.69–0.90; p < 0.001) | eGFR < 20 mL/min/1.73 m2 or requiring dialysis | 60.6±19.8 mL/min/1.73 m2 | 53.5% had eGFR <60 mL/min/1.73 m2 or urine albumin-to-creatine ratio >300 mg/g | Secondary endpoint: GFR slope Prespecified analysis: composite renal endpoint | Empagliflozin slowed the slope of eGFR decline by 1.43 (1.01–1.85) mL/min/1.73 m2/yr in patients with CKD and 1.31 (0.88–1.74) mL/min/1.73 m2/yr in patients without CKD (interaction p = 0.70) Empagliflozin did not reduce the prespecified kidney outcome in patients with or without CKD (with CKD: HR 0.97, 95% CI: 0.71–1.34; without CKD: HR 0.92, 95% CI: 0.58–1.48; interaction p = 0.86) |
Dapagliflozin Evaluation to Improve the Lives of Patients with Preserved Ejection Fraction Heart Failure (DELIVER) | 2022 | Dapagliflozin 10 mg daily vs. placebo | Worsening HF or CV death | Reduction with dapagliflozin (HR, 0.82; 95% CI: 0.73–0.92; p < 0.001) | eGFR <25 mL/min/1.73 m2 | 61±19 mL/min/1.73 m2 | 49% had an eGFR <60 mL/min/1.73 m2 | Secondary endpoint: slope of eGFR Post hoc composite kidney endpoint | Dapagliflozin attenuated the decline in eGFR from baseline (difference, 0.5; 95% CI: 0.1–0.9 mL/min/1.73 m2 per yr; p = 0.01) No significant differences in kidney endpoint (HR, 1.08; 95% CI: 0.79–1.49) |
Semaglutide Treatment Effect in People with Obesity (STEP-HFpEF and STEP-HFpEF DM) | 2024 | Semaglutide target dose 2.4 mg sc weekly vs. placebo | Change from baseline to week 52 in KCCQ-CSS and bodyweight | Reduction with semaglutide KCCQ-CSS 7·5 points (95% CI: 5·3–9·8); p < 0·0001; difference in bodyweight at week 52 −8·4% (−9·2 to −7·5); p < 0·0001 | ESRD or dialysis | STEP-HFpEF DM 69.0 (50.0–88.0) mL/min/1.73 m2 | NA | No | NA |
Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist (TOPCAT) | 2014 | Spironolactone up to 45 mg daily vs. placebo | CV death, aborted cardiac arrest, or hospitalization for HF | No significant differences (HR, 0.89; 95% CI: 0.77–1.04; p = 0.14) | eGFR of <30 mL per minute per 1.73 m2 of body surface area or creatinine ≥2.5 mg/dL | 58.4±19.4 mL/min/1.73 m2 | 39% in both groups had eGFR < 60 mL/min/1.73 m2 | Post hoc analysis differences in eGFR slopes | Spironolactone led to a higher decline of rGFR over time −2.5 [−3.4 to −1.6] mL/min/1.73 m2/yr; p < 0.001 |
FINerenone trial to investigate Efficacy and sAfety superioR to placebo in paTientS with Heart Failure (FINEARTS-HF) | 2024 | Finerenone up to 20 mg daily if eGFR ≤60 mL/min/1.73 m2 or 40 mg daily if eGFR >60 mL/min/1.73m2 vs. placebo | Total worsening HF events and death from CV causes | Reduction with finerenone (rate ratio, 0.84; 95% CI: 0.74–0.95; p = 0.0079) | eGFR <25 mL/min/1.73 m2 | 62.1±19.7 mL/min/1.73 m2 | 48.1% with an eGFR <60 mL/min/1.73 m2 | Composite renal endpoint (defined as sustained decrease in eGFR ≥50% relative to baseline over at least 4 wk, sustained eGFR decline <15 mL/min/1.73 m2, or initiation of dialysis or renal transplantation) | No significant differences among groups (HR, 1.33; 95% CI: 0.94–1.89) |
CKD, chronic kidney disease; CV, cardiovascular; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; HF, heart failure; KCCQ, Kansas City Cardiomyopathy Questionnaire.
Diuretics
Diuretics are recommended for HFpEF to alleviate symptoms and relieve congestion [21]. However, their use in patients with cardiorenal disease presents significant challenges. Loop diuretics effectively increase renal sodium and water excretion but may have acute and long-term effects on kidney function. Acutely, they can reduce eGFR due to mechanisms such as sympathetic nervous system and RAAS activation, changes in tubular and interstitial pressures, and alterations in volume status. Despite this, transient and mild serum creatinine elevations following effective diuresis and decongestion are considered normal and associated with better outcomes [44]. In the long term, observational data suggest that higher doses of loop diuretics are linked to a more rapid decline in GFR. However, this association may be influenced by selection bias, as sicker patients with more significant congestion and progressive renal decline are often prescribed higher diuretic doses [45]. Patients with CKD typically require higher doses to maintain an adequate diuretic response, particularly as eGFR declines and diuretic efficiency diminishes [45, 46]. For HFpEF and CKD patients, a thorough multiparametric assessment of volume status, congestion severity, and diuretic response is crucial. In this sense, merging strategies have demonstrated potential benefits in guiding diuretic therapy. A CA125-guided diuretic strategy has been shown to improve decongestion, eGFR, and other renal parameters at 72 h in patients with acute HF and renal dysfunction [47]. Similarly, assessing natriuresis provides a valuable measure of diuretic response. In the Pragmatic Urinary Sodium-based treatment algoritHm in Acute Heart Failure (PUSH-AHF) trial, which included 20% HFpEF patients, natriuretic-guided therapy improved short-term natriuresis and diuresis but did not impact 180-day outcomes [48]. No specific subgroup analysis was made in patients with preserved ejection fraction, but a subgroup analysis revealed that patients with low eGFR may derive specific benefit, with significantly fewer adverse events at 180 days in the natriuretic-guided arm (p for interaction = 0.017) [49]. These findings suggest that a personalized approach targeting adequate natriuresis could be particularly beneficial for patients with acute HF and CKD.
Loop diuretics remain the cornerstone of diuretic therapy [21]. Although torsemide has been proposed to offer pharmacokinetic advantages over furosemide in CKD patients, clinical trials have not shown superior long-term outcomes with torsemide compared to furosemide following acute HF hospitalization. These findings were consistent across patients with reduced, mildly reduced, or preserved ejection fraction [50]. In cases of suboptimal loop diuretic response or diuretic resistance, combining diuretics targeting different nephron sites is recommended [21, 22]. Recent trials have evaluated the addition of acetazolamide or hydrochlorothiazide to loop diuretics, demonstrating improved natriuresis, diuresis, and congestion resolution with combination therapy [51, 52]. A prespecified analysis of the Acetazolamide in Decompensated Heart Failure with Volume Overload (ADVOR) trial, with a median eGFR of 40 (30–52) mL/min/1.73 m2, showed that acetazolamide was safe and increased the likelihood of successful decongestion regardless of renal function. Benefits were particularly pronounced in patients with lower eGFR [53]. Additionally, the benefits of acetazolamide were of baseline LVEF [54]. The Combining loop with thiazide diuretics for decompensated heart failure (CLOROTIC) trial, with a median eGFR of 43 (14–109) mL/min/1.73 m2, found no safety differences across eGFR ranges, though the diuretic effect of hydrochlorothiazide was attenuated in patients with eGFR <45 mL/min/1.73 m2 [51]. No treatment effect modification was found according to baseline LVEF [55]. Finally, tolvaptan may be a safe and effective alternative for patients with HF and advanced CKD, although there are no available data regarding treatment effect modification according to LVEF [56].
SGLT2 Inhibitors
SGLT2 inhibitors (SGLT2is) have transformed the management of cardio-renal-metabolic diseases, demonstrating significant benefits across the cardiorenal spectrum. In a meta-analysis of 78,607 patients across 9 trials, GLT2i consistently improved CV outcomes across a wide range of kidney functions, including patients with an eGFR <30 mL/min/1.73 m2 [41]. Among patients with CKD of various etiologies, these drugs also reduce HF-related events. For instance, a prespecified analysis of the DAPA-CKD trial showed that dapagliflozin reduced the risk of CV death and HF hospitalization by 29%, regardless of the presence of HF at baseline [57]. This underscores the utility of SGLT2is in preventing clinical HF in CKD patients in the “pre-HF” stage. SGLT2is have also marked a milestone in HF management, being the first therapy to robustly improve outcomes in patients with stage C HFpEF [5, 58]. In the Empagliflozin, Health Status, and Quality of Life in Patients With Heart Failure and Preserved Ejection Fraction (EMPEROR-Preserved) and DELIVER trials, SGLTi reduced the composite of CV death and HF hospitalization by 20% (HR, 0.80; 95% CI: 0.78–0.87; p < 0.001). While CV death reduction was not statistically significant (HR, 0.88; 95% CI: 0.77–1.00; p = 0.052), HF hospitalizations were reduced by 26% (HR, 0.74; 95% CI: 0.67–0.83; p < 0.001). Consequently, the European Society of Cardiology (ESC) guidelines recommend both empagliflozin and dapagliflozin as class IA treatments for patients with HF and LVEF >40% [21, 22]. In both trials, patients with CKD were well represented (Table 1). In the prespecified subgroup of patients with CKD, SGLT2is demonstrated safety and efficacy irrespective of baseline eGFR or KDIGO categories [59, 60]. Regarding kidney function, they have consistently shown nephroprotective effects. A meta-analysis of 15 randomized controlled trials, encompassing over 90,000 patients, revealed a 37% reduction in the risk of kidney disease progression compared to placebo, alongside a comparable reduction in acute kidney injury risk [61].
The primary mechanism behind these benefits is their ability to modulate inter- and intraglomerular hemodynamics without affecting the number of functional nephrons. An initial relatively small eGFR decrease may be expected after SGLT2i initiation in HFpEF. In the DELIVER trial, 40% of patients on dapagliflozin experienced an initial eGFR decline >10% compared to 25% on placebo (OR, 1.9; 95% CI: 1.7–2.1; p < 0.001). Notably, this initial eGFR drop did not correlate with an increased long-term risk of CV or renal events [62]. In the long term, SGLT2is attenuated eGFR decline by 1.0 mL/min/1.73 m2/year in EMPEROR-Preserved and 0.5 mL/min/1.73 m2/year in DELIVER [60, 63]. Interestingly, among patients with advanced CKD (eGFR <30 mL/min/1.73 m2), the placebo-corrected absolute eGFR change was minimal (0.2; 95% CI: −1.6 to +2.0), reaffirming the safety of these drugs in HFpEF and CKD. This smaller response may be explained by reduced activation of the tubuloglomerular feedback in advanced CKD [64]. In HFpEF, larger eGFR declines may occasionally occur. A post hoc analysis of the combined DAPA-HF and DELIVER datasets showed that 347 participants (3.2%) experienced eGFR deterioration below 25 mL/min/1.73 m2, mostly beyond the first month [65]. Despite this, dapagliflozin-treated patients had a lower risk of primary composite outcomes compared to placebo, supporting the continuation of SGLT2is in these cases after thorough clinical evaluation. When should SGLT2is be discontinued? Current KDIGO guidelines endorse continuation at eGFR levels below 20 mL/min/1.73 m2, emphasizing the importance of individualized care in specialized cardiorenal programs [42]. Ongoing trials, such as RENAL LIFECYCLE (A RCT to Assess the Effect of Dapagliflozin on Renal and Cardiovascular Outcomes in Patients With Severe CKD) trial (NCT05374291), may further inform the use of SGLT2is in severe CKD.
Glucagon-Like Peptide-1 Agonists
Glucagon-like peptide-1 (GLP-1) receptor agonists have changed the treatment of patients with T2DM and/or obesity. These drugs effectively reduce body weight, visceral fat, glycemia, blood pressure, and inflammation, while also improving outcomes in these conditions. In patients with HFpEF and obesity, once-weekly semaglutide 2.4 mg significantly outperformed placebo in achieving weight loss and improving health status. This therapy also led to a 15% greater reduction in NT-proBNP levels and a 43% greater reduction in CRP levels compared to placebo [66]. Furthermore, a pooled analysis of 3,743 patients with a history of HFpEF from 4 trials showed that semaglutide reduced the risk of a composite endpoint of CV death and worsening HF events by 31% (HR, 0.69; 95% CI: 0.53–0.89; p = 0.0045) [67]. Recently, tirzepatide, a dual glucose-dependent insulinotropic polypeptide and GLP-1 receptor agonist administered weekly, led to a lower risk of a composite of death from CV causes or worsening HF (HR, 0.62; 95% CI: 0.41–0.95; p = 0.0026) in patients with HFpEF and obesity [68]. These findings are expected to shape future recommendations for managing the obesity-related HFpEF phenotype.
GLP-1 receptor agonists have also demonstrated benefits across the cardio-renal-metabolic spectrum. In a large trial involving patients with T2DM and CKD, semaglutide 2.4 mg reduced the risk of kidney and CV events compared to placebo. The trial reported a 21% reduction in kidney-specific outcomes, an improvement in the eGFR slope, and a 20% reduction in all-cause mortality over a median follow-up of 3.4 years [69].
Angiotensin Receptor-Neprilysin Inhibitor
RAAS inhibitors (RAASis) have been foundational in the treatment of CKD [42]. However, historically, no RAASi has convincingly improved outcomes in patients with HFpEF. In the Prospective Comparison of ARNI with ARB (angiotensin-receptor blockers) Global Outcomes in HF with Preserved Ejection Fraction (PARAGON-HF) trial, sacubitril/valsartan did not significantly reduce the composite risk of HF hospitalizations and CV death compared to valsartan in patients with HF and LVEF >45% [70]. Although it narrowly missed statistical significance, exploratory analyses revealed potentially meaningful benefits in specific subgroups, particularly women and patients with LVEF at the lower end of the spectrum. Consequently, both the ESC and American College of Cardiology/American Heart Association (ACC/AHA) guidelines provide a class 2b recommendation for sacubitril/valsartan in the treatment of HFpEF [21, 22].
Could sacubitril/valsartan benefit patients with both HFpEF and CKD? Sacubitril inhibits the clearance of NPs, thereby enhancing their biological effects. NPs promote diuresis and natriuresis, decrease sympathetic activity, increase GFR through afferent arteriolar vasodilatation, and increase the glomerular capillary ultrafiltration coefficient by inducing relaxation of the contractile intraglomerular mesangial cells [45]. In a combined analysis of the Prospective comparison of ARNI with ARB Given following stabiLization In DEcompensated HFpEF (PARAGLIDE) and PARAGON-HF trials (n = 5,262), sacubitril/valsartan was associated with a lower rate of a renal composite endpoint (defined as >50% decline in eGFR from baseline, progression to ESRD, or renal death) compared to control (HR, 0.60; 95% CI: 0.44–0.83; p = 0.002). These findings support a nephroprotective effect of angiotensin receptor-neprilysin inhibitor (ARNI) in patients with HFpEF [71].
Mineralocorticoid Receptor Antagonists
In the heart, kidneys, and vasculature, overactivation of the mineralocorticoid receptor contributes to endothelial dysfunction, vascular remodeling, inflammation, fibrosis, proteinuria, and glomerular and tubular injury. Mineralocorticoid receptor antagonists (MRAs) reduce sodium and fluid retention, while exerting anti-inflammatory, anti-remodeling, and anti-fibrotic effects [43, 44].
Steroidal MRAs have a class IA recommendation for the treatment of HFrEF [21, 22]. However, their efficacy in HFpEF is less robust. In the Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist (TOPCAT) trial on patients with HF and LVEF >45%, spironolactone did not significantly reduce the primary outcome of CV death, aborted cardiac arrest, or HF hospitalizations compared to placebo [72]. Nevertheless, patient selection and protocol implementation issues arose outside the Americas. In the American cohort, where elevated NP levels were required for enrollment, a potential benefit was observed [73]. Evidence for steroidal MRAs in CKD is similarly limited. Recently, spironolactone failed to improve CV outcomes in patients with stage 3B CKD [74].
Finerenone, a selective nonsteroidal MRA, offers distinct physicochemical properties compared to steroidal MRAs, including differences in tissue penetration, distribution, and a shorter half-life. Unlike spironolactone, finerenone does not cross the blood-brain barrier and is equally distributed in the heart and kidneys. Finerenone has been evaluated in two large landmark randomized clinical trials in patients with T2DM and a broad range of CKD, reducing kidney and CV outcomes when added to RAASi therapy [75, 76]. In the Finerenone in Reducing Kidney Failure and Disease Progression in Diabetic Kidney Disease (FIDELIO-DKD) trial, finerenone reduced the risk of a composite of kidney failure by 18%, as well as the risk of a secondary endpoint of major CV events [75]. In the Finerenone in Reducing Cardiovascular Mortality and Morbidity in Diabetic Kidney Disease (FIGARO-DKD) trial, finerenone reduced the rate of the primary outcome, a composite of CV death, nonfatal myocardial infarction, nonfatal stroke, or HF hospitalization driven by a statistically significant lower incidence of HF hospitalization [76]. In both trials the occurrence of hyperkalemia was higher in the finerenone group compared with the placebo group. However, the rate of adverse events was similar between the 2 arms [75, 76]. A prespecified pooled analysis of 13,026 patients from both trials reported a 14% reduction in the composite CV outcome and a 22% reduction in HF hospitalizations with finerenone, highlighting its role in preventing clinical HF in T2DM-CKD patients in the “pre-HF” stage [77].
Recently, the results of the landmark FINEARTS-HF trial have been published. The trial enrolled 6,001 patients with LVEF ≥40%, making it one of the largest trials in this population [6]. Baseline median eGFR was 62 mL/min/1.73 m2, and 40% showed any degree of albuminuria. Finerenone significantly reduced the risk of CV death and worsening HF events, driven by a significant 18% reduction in the risk of worsening HF (Table 1). The risk of hyperkalemia was higher than with placebo, although it rarely caused hospitalizations and caused no deaths [6]. As observed with SGLT2is, finerenone did not result in a lower risk of secondary kidney composite outcomes in this trial (Table 1), reflecting the complexity of HFPEF, but overall kidney event rates were low in such trials.
In summary, treatment landscape for patients with HFpEF has seen significant advancements in recent years. For the first time, the concept of optimal medical therapy has been extended to include patients with HF and LVEF >40%. SGLT2 inhibitors and finerenone have demonstrated reductions in the risk of CV death and worsening HF in patients with HFpEF. In addition, GLP-1 receptor agonists improve health status and potentially HF-related events in patients with obesity-related HFpEF, and ARNI may be beneficial in selected patients. All these therapies are not only safe for patients with CKD but also offer potential renal benefits. Diuretics remain essential for relieving congestion, while coordinated care through cardiorenal units and comprehensive management of comorbidities are crucial for optimizing patient outcomes (Fig. 3).
Treatment recommendations for patients with concomitant HFpEF and CKD. CKD, chronic kidney disease; LVEF, left ventricular ejection fraction; SGLT2is, SGLT2 inhibitors; WHF, worsening heart failure. *up to stage 5 renal disease. ⱡup to eGFR down to 25 mL/min/1.73 m2.
Treatment recommendations for patients with concomitant HFpEF and CKD. CKD, chronic kidney disease; LVEF, left ventricular ejection fraction; SGLT2is, SGLT2 inhibitors; WHF, worsening heart failure. *up to stage 5 renal disease. ⱡup to eGFR down to 25 mL/min/1.73 m2.
Conflict of Interest Statement
The authors have no conflicts of interest to declare.
Funding Sources
This study was not supported by any sponsor or funder.
Author Contributions
G.N.-M. and E.S. contributed equally to conceptualization, writing, review, and editing.