The role of estrogen in the sex difference for the risk factors of heart failure with preserved ejection fraction

Abstract

Heart failure with preserved ejection fraction (HFpEF) is a major subtype of heart failure, primarily characterized by a normal or mildly reduced left ventricular ejection fraction along with left ventricular diastolic dysfunction. Recent studies have shown that the prevalence of HFpEF is higher in women than that in men, particularly in postmenopausal women. Concurrently, it has been observed that the incidence of risk factors contributing to HFpEF (such as obesity, hypertension, diabetes, and atrial fibrillation) also notably increases post-menopause, affecting the incidence of HFpEF. This review aimed to examine the relationship between estrogen and risk factors associated with HFpEF, clarifying the underlying mechanisms through which estrogen affects these risk factors from epidemiological and pathophysiological perspectives. This review also provides a comprehensive understanding of the association between estrogen and the risk factors for HFpEF, thus helping explore potential targets for HFpEF treatment.

Heart failure (HF) is a global health concern with high incidence and mortality rates. Approximately 2% of the developed Western population suffers from HF, and the prevalence of the disease rapidly increases with age [1,2,3]. Based on the left ventricular ejection fraction (EF) at the time of diagnosis, HF is classified into three types: heart failure with reduced ejection fraction (HFrEF), which refers to an EF less than 40%; heart failure with mid-range ejection fraction (HFmrEF), defined as an EF between 41 and 49%; and heart failure with preserved ejection fraction (HFpEF), ranging from EF greater than 50% [4]. Recent epidemiological studies have found that HFpEF accounts for at least 50% of all HF cases, with an increasing incidence and female preference [5]. Among patients with HFpEF, the number of female patients is nearly twice that of male patients, whereas male patients with HF primarily exhibit a reduced ejection fraction [6]. Data from a European population revealed that the incidence of both types of HF increases with age, and the incidence of HFpEF increases more rapidly with age than that of HFrEF. However, at any age, the incidence of HFpEF is higher in women than that in men, particularly in postmenopausal women [5]. Among individuals aged 25–49 years, the incidence rates, when stratified by age and sex, increased from 0% (males) and 1% (females) to 4–6% in males and 8–10% in females aged 80 years and above [7].

Biological sex is a negligible factor that significantly affects the structure and function of the cardiovascular system. The XY sex chromosome, sex-specific gene expression, sex hormones, and sex reproductive organs show the most significant differences, which may contribute to the pathology of heart diseases [8]. Estrogen is the major sex hormone responsible for this sex-specific difference. Notably, the increased risk of cardiovascular diseases risk in postmenopausal women suggests that changes in estrogen levels may contribute to the occurrence and development of HFpEF in females, further indicating a potential protective role of estrogen in patients with HFpEF [9].

The prevalence of HFpEF is closely associated with risk factors, such as hypertension, obesity, diabetes, and atrial fibrillation (AF), as shown in Fig. 1. The increase in the risk associated with these factors shows an upward trend with age, especially in the female population, significantly increasing with a decline in hormone levels during menopause. Therefore, this review aimed to clarify the relationship between estrogen and these risk factors for HFpEF, providing novel therapeutic targets to combat this devastating disease worldwide from a sex perspective.

Fig. 1

Estrogen and HFpEF-related risk factors. Common risk factors of HFpEF include obesity, hypertension, diabetes, and atrial fibrillation. As estrogen levels in women decline, the prevalence and severity of these risk factors continue to increase, directly or indirectly influencing the onset and progression of HFpEF

Interplay among estrogen, obesity, and HFpEF

Obesity is a major risk factor for HFpEF [10]. Globally, nearly 880 million adults are obese, and this number continues to rise annually [11]. The global prevalence of obesity is higher in women than that in men [12], and there is a sex bias in body fat distribution and energy homeostasis [13]. The number of women with severe obesity is approximately as many as twice the number of men (i.e., class 2 obesity with a body mass index (BMI) ≥ 35 kg/m² and class 3 obesity with a BMI ≥ 40 kg/m²) [14]. Statistics show that the obesity rate among American women aged 40–59 years is 65%, and for women aged 60 years and above, it is as high as 73.8%, demonstrating an increasing trend of being overweight and obese in postmenopausal women [15,16,17]. It is well known that obesity greatly elevates the risks of cardiovascular disease (CVD) and even HF [18], which is particularly evident in HFpEF throughout an individual’s lifetime [19, 20]. Epidemiological studies indicate that nearly 80% of patients with HFpEF are overweight or obese (i.e., BMI ≥ 25 kg/m²) [21, 22]. Among patients with HF, a higher BMI is closely associated with HFpEF [23]. A large study comprising 22,618 individuals from four community cohorts (the Cardiovascular Health Study, PREVENT, Framingham Heart Study, and MESA) revealed that there is a 34% increase in the incidence of HFpEF with every 1 kg/m² increase in BMI [6, 23, 24]. Obesity and the resulting cardiac metabolic changes significantly elevate the risk of HFpEF in women, implying a stronger link between obesity and HFpEF in women than that in men [23].

Obesity leads to hemodynamic changes, resulting in compensatory alterations in cardiac morphology and ventricular function. These symptoms include increased cardiac output, left ventricular hypertrophy, and impaired diastolic and systolic function in both ventricles [25]. The diastolic function may also be impaired by limiting energy utilization efficiency or increasing myocardial lipid content [26, 27], the latter of which is associated with increased left ventricular diastolic tension [28, 29]. Excess adipose tissue (AT) often leads to plasma volume expansion and impaired left ventricular relaxation, which is likely mediated by systemic inflammation and microvascular rarefaction [30]. These changes can increase left ventricular filling pressure, manifesting as signs and symptoms of HF, and ultimately resulting in restricted ventricular dilation [31, 32].

Moreover, obesity influences the neurohormonal system, promoting the development of HFpEF by increasing aldosterone and leptin levels and suppressing natriuretic peptide activity [33,34,35]. This modulation occurs by activating the renin–angiotensin–aldosterone system (RAAS), which elevates aldosterone synthesis. Such activation triggers a cascade of downstream effects, including sodium retention, plasma volume expansion, elevated systemic inflammation, intensified kidney and heart fibrosis, increased arterial stiffness, and deterioration of left ventricular diastolic function [19, 33, 36, 37]. Patients with obesity and HFpEF are highly likely to experience diastolic dysfunction, as well as cardiac enlargement due to increased epicardial fat, leading to pericardial restraint and enhanced ventricular interaction [19, 38]. Compared with non-obese patients with HFpEF and controls, those with obesity and HFpEF have worse exercise capacity, elevated biventricular filling pressures during exercise, reduced pulmonary arterial vasodilator reserves, inadequate cardiac output, and decreased cardiopulmonary function during exercise [19].

Although it was once believed that overall fat accumulation increases cardiovascular risk, the detrimental effects of obesity on the cardiovascular system may also be linked to AT distribution within the body [39]. Obesity can be classified based on fat distribution into peripheral obesity, with fat predominantly distributed in the hips and limbs as subcutaneous AT, and central obesity, with fat primarily distributed around the abdominal viscera as visceral AT. Central obesity is a major risk factor for CVD, whereas peripheral fat appears to offer protective effects [40,41,42]. AT distribution is influenced by multiple factors, and sex hormone levels play a significant role [43]. Typically, men have a lower total fat mass but a higher proportion of abdominal fat than women. In contrast, women tend to have more total fat distributed predominantly as subcutaneous fat in the hips, thighs, and other peripheral areas than men [44].

In premenopausal women, estrogen (E2) promotes subcutaneous fat accumulation, inhibits visceral fat deposition, and reduces CVD risk [45]. Estrogen acts in synergy with AT genes, increasing the subcutaneous AT mass in reproductive-aged women and reducing the visceral AT mass, thereby producing a protective cardiac metabolic effect. However, the loss of estrogen after menopause increases the total AT mass, partially reversing the protective distribution of AT in women [46]. When estrogen levels become sufficiently low, women experience visceral fat accumulation, possibly due to the direct effects of estrogen on AT distribution [47]. After menopause, subcutaneous AT and visceral AT levels both increase, with the relative rise in visceral AT being twice that of total AT [48]. Evidence has shown that postmenopausal women are three times more likely to develop obesity-related metabolic syndrome than premenopausal women [49]. Fat distribution in women also shifts significantly with changes in estrogen levels, with central obesity becoming more ronounced in postmenopausal women. Metabolic abnormalities are also more closely related to visceral fat accumulation and plasma estrogen levels than to BMI alone [50]. During aging, while subcutaneous fat deposition occurs universally in women, visceral adiposity progressively dominates, posing a heightened risk for metabolic and cardiovascular complications [51].

The epicardial adipose tissue (EAT), a specialized form of visceral fat, is located within the pericardium and closely surrounds the myocardium. They are both metabolically active complex tissues [52]. In addition to storing triglycerides to supply energy to the myocardium, the EAT also secretes pro-inflammatory adipokines and pro-oxidative substances, contributing to adverse cardiac effects [53, 54]. The EAT exerts primarily detrimental effects on the heart by (i) promoting inflammation and driving related complications and (ii) facilitating myocardial interactions that result in diastolic dysfunction [53, 54].

The decline in estrogen levels after menopause exacerbates obesity and inflammation via various mechanisms. Reduced estrogen levels not only directly alter fat distribution and increase visceral fat accumulation but also promote systemic low-grade inflammation and a local pro-inflammatory environment, worsening endothelial dysfunction [55, 56]. Adipose tissues, including EAT, secrete higher levels of pro-inflammatory cytokines, such as TNF-α, IL-6, and CRP [57,58,59], which reduce the bioavailability of nitric oxide (NO). As a result, microvascular dysfunction, impaired diastolic function, and myocardial fibrosis are more pronounced compared to individuals with lower levels of these pro-inflammatory cytokines. [60, 61]. Collectively, these processes drive the development of HFpEF. Estrogen/progesterone hormone replacement therapy has been shown to reduce visceral AT, fasting blood glucose, and insulin levels in postmenopausal women [43, 44]. Therefore, estrogen-related therapy may mitigate the increased cardiovascular risk after menopause by reducing the risk of obesity and its associated detrimental effects, offering a potential strategy for HFpEF prevention and management.

Interplay among estrogen, hypertension, and HFpEF

Patients with long-standing hypertension show a significantly increased risk of HF, which is a significant and highly associated clinical consequence of hypertension-mediated organ damage [62,63,64]. The incidence of hypertension in premenopausal women is usually lower than that in men of the same age; however, after menopause, the incidence of hypertension in women sharply increases, leading to the gradual disappearance of this sex difference [63]. As age increases, women become more susceptible to hypertension than men, especially in the 65–74-year-old age group, where 45% of women and 41% of men are reported to have hypertension, with the incidence in women surpassing that in men [65, 66]. Hypertension is common in females over 60 years of age [67], with sex-associated steroid hormones identified as the key players behind this sex-related difference [68]. It is known that E2 has potent acute and chronic vasodilatory activity, ultimately leading to blood pressure (BP) reduction [69,70,71]. Most animal studies indicate that E2 is involved in various mechanisms preventing hypertension, such as stimulating NO-mediated vasodilator pathway [72, 73]. Previous research suggests that all three types of ERs have protective effects on the cardiovascular system. Specifically, the activation of ERα has been shown to reduce endothelial dysfunction [74], while the activation of ERβ can lower BP, vascular constriction, and vascular resistance, and alleviate cardiac hypertrophy [75, 76]. Additionally, GPR30 lowers BP, stimulates vasodilation, and reduces vascular smooth muscle cell proliferation and migration [77]. The action of estrogen may be beneficial for premenopausal women, exerting a protective effect on the cardiovascular system, as shown in Fig. 2 [71]. In contrast, its loss may render the hearts of postmenopausal women more vulnerable to cardiovascular risks [78].

Fig. 2

Mechanisms by which estrogen influences hypertension. E2 binds to membrane GPR30, activating the MAPK/ERK or PI3K pathways, enhancing eNOS activity, and rapidly increasing NO production, leading to vasodilation and blood pressure reduction. Simultaneously, E2 inhibits the NF-κB pathway, reducing the expression of inflammatory mediators and mitigating damage to the vascular endothelium. E2 also interacts with ERα and ERβ, which then translocate to the nucleus and bind directly to estrogen response elements, regulating the transcription of target genes, such as VEGF, to promote angiogenesis and eNOS for vasodilation and blood pressure regulation

This notable sex difference also affects HF progression. In the Framingham Heart Study, after adjusting for age, the risk of HF in hypertensive patients increased twofold in men and 3.2-fold in women [79, 80]. As the incidence of hypertension continues to increase, it primarily leads to a higher occurrence of HFpEF in women and a higher occurrence of HFrEF in men [81, 82]. Proper BP control is widely recognized as the cornerstone in the prevention and clinical management of HFpEF [83,84,85]. Randomized clinical trials have shown that following the strategies outlined in the American College of Cardiology (ACC)/American Heart Association (AHA) hypertension treatment guidelines can reduce the incidence of HFpEF by 40% in the next 2–8 years of life [86, 87]. Further epidemiological investigations have found that the incidence of hypertension in patients with HFpEF ranges from 55 to 90% [88,89,90]. Compared to patients with HFrEF, those with HFpEF are more likely to have a history of hypertension [81, 91]. In patients diagnosed with HFpEF, hypertension may be the sole cause or one of several causes or comorbidities.

Hypertension causes left ventricular hypertrophy, diastolic dysfunction, left ventricular fibrosis, left atrial dilation, and macrovascular and microvascular sclerosis [92,93,94]. In females, hypertension induces concentric hypertrophy of the left ventricle, leading to diastolic dysfunction, and the clinical course of HF is generally benign and often characterized by HFpEF [80, 82]. In males, eccentric hypertrophy with chamber dilatation is more common and ultimately leads to systolic dysfunction [95,96,97]. At any given age, the aortic root diameter in females is smaller than that in males. This may indicate a relatively delayed external remodeling response of the aortic arch during hemodynamic load, potentially leading to increased pulse pressure [98]. Increased load-induced impairment of left ventricular diastolic function makes females more susceptible to the adverse effects of arteriosclerosis and early wave reflection on left ventricular diastolic function [99]. Arterial-ventricular coupling in males remains stable with age, whereas it decreases with age in women [99]. This difference may exacerbate the disadvantages of hypertension and HFpEF in women. Females are also more prone to idiopathic pulmonary arterial hypertension with enhanced pulmonary arterial vasoconstriction and intrinsic pulmonary arterial remodeling, affecting them four times more than males [100]. Compared to HFpEF patients without pulmonary hypertension (PH), patients with PH-HFpEF are more often female, have more severe symptoms, and more frequently exhibit right ventricular hypertrophy and right atrial dilation with higher right atrial pressures [101]. In summary, the interplay between hypertension and HFpEF is influenced by notable sex differences, with females exhibiting unique pathophysiological responses, such as greater susceptibility to diastolic dysfunction and PH, underscoring the need for sex-specific approaches in the prevention and management of HFpEF.

Interplay among estrogen, diabetes, and HFpEF

Approximately 30.3 million individuals in the United States currently have diabetes, with an additional 84.1 million categorized as prediabetic, and the prevalence of diabetes is projected to increase by 54% by 2030 [102, 103]. As of 2023, China, the country with the largest number of patients with diabetes worldwide, has over 118 million people affected by this disease [104]. Individuals with diabetes exhibit a significantly elevated incidence of HF compared with their non-diabetic counterparts, with an approximately 2–fivefold increased risk [105, 106]. It is estimated that diabetes coexists in approximately 30–40% of patients with HF and a deteriorated clinical condition and higher all-cause and cardiovascular mortality than that in patients with HF without type 2 diabetes (T2D) [107].

Significant sex differences have also been observed in the development of diabetes, particularly T2D. Overall, the incidence of T2D is higher in men than in women, especially in the middle-aged and young populations. However, as age increases, the incidence of diabetes in women gradually increases, surpassing that in men after 70 years of age [108,109,110]. Furthermore, the early onset of diabetes in women may be more insidious; the age-specific prevalence of previously undiagnosed diabetes and impaired glucose tolerance (IGT) defined by isolated post-load hyperglycemia is higher in women than that in men [110]. Additionally, women may be more prone to developing diabetes-related complications, such as CVD, than men [111, 112]. Additionally, gestational diabetes is a risk factor specific to women, as those with gestational diabetes have a higher risk of developing T2D in the future [113]. Moreover, women with a history of GDM appear to have a nearly tenfold higher risk of developing T2D than those with normoglycemic pregnancies [114].

Similarly, significant sex-based differences were observed in the transition from diabetes to HF. Based on a study that followed 5209 men and women aged 30–62 years for 18 years and explored the relationship between the incidence of congestive HF and previous diabetes status, the results showed that diabetes increased the risk of HF more significantly in women than that in men, with the risk increasing fivefold in women compared to a 2.4-fold increase in men [115]. Furthermore, women with diabetes have a significantly higher risk of developing congestive HF than men. Even after adjusting for other factors, such as age, BP, and cholesterol, sex differences remained evident [115]. These findings suggest that the risk of death associated with T2D is significantly higher in women than that in men [116]. Two population-based studies conducted in Scotland and Sweden confirmed this trend, revealing an elevated mortality risk due to T2D in both sexes, with a markedly greater effect in women [117, 118]. Among patients diagnosed with HF and T2DM, nearly half are identified with HFpEF, particularly among older participants and especially in the cohort of females with hypertension and T2D [119].

The molecular mechanisms through which DM contributes to the development of HFpEF are complex and diverse. From a pathophysiological perspective, the occurrence of HFpEF is often highly correlated with pathological features induced by diabetes, such as volume overload, peripheral injury, metabolic disorders, release of pro-inflammatory factors, and skeletal muscle damage. Patients with diabetes often exhibit excessive activation of the neurohormonal system and altered sodium handling, potentially leading to cardiac congestion, cardiorenal syndrome, and reduced diuretic response [120]. In addition, hyperglycemia can lead to the upregulation of sodium-glucose co-transporter-2 (SGLT2), thereby increasing proximal renal sodium reabsorption, resulting in volume expansion and reduced diuretic response [121]. Patients with HFpEF and concomitant diabetes demonstrate manifestations of volume overload, often presenting with higher mechanical ventilation and requirements for dialysis/ultrafiltration, worse renal function, and prolonged hospital stays owing to increased in-hospital volume and poor diuretic response at discharge, leading to an increased rate of HF readmissions [102, 122]. In the EMPA-REG OUTCOME (BI 10773 [Empagliflozin] Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients) trial, empagliflozin was associated with a reduction in major adverse cardiovascular events and a significantly reduced rate of HF hospitalizations [123]. SGLT2 inhibitors (SGLT2is) promote diuresis and natriuresis by restoring sodium delivery to the distal renal tubule without activating the sympathetic nervous system. This achieves natriuretic therapy and alleviates volume overload, thereby reducing the incidence of HFpEF and DM in patients [124]. SGLT2is also reduce myocardial Na/H exchanger activity, reverse electrolyte imbalances, and increase the sodium ion concentration in the mitochondria of patients with HF [125, 126]. In addition, they improve myocardial metabolism by optimizing energy metabolism, enhancing oxygen delivery, promoting ATP storage, increasing mitochondrial oxygen uptake and conversion efficiency, and facilitating energy supply by utilizing glucose-derived ketone bodies [127, 128]. Furthermore, SGLT2is can reduce cardiomyocyte apoptosis and improve myocardial fibrosis by alleviating oxidative stress, decreasing TGF-β production, and regulating macrophage polarization [129, 130]. Insulin resistance in patients with diabetes may lead to increased utilization of free fatty acids by myocardial cells, potentially resulting in mitochondrial dysfunction, production of toxic lipid intermediates, and increased reactive oxygen species [131, 132]. High glucose-induced advanced glycation end products can impair microvascular function and reduce nitric oxide availability [131]. These processes may contribute to structural changes in myocardial cells, endothelial dysfunction, and HF.

In the female population, diabetes and the resulting HFpEF exhibit particularly distinct manifestations. Studies have demonstrated that women with diabetes often show higher levels of insulin resistance [133, 134] and are more susceptible to glucose intolerance and a persistent hyperglycemic state, whereas men with diabetes predominantly exhibit elevated fasting blood glucose levels [135]. Insulin resistance and persistent hyperglycemia can induce a systemic inflammatory response [136], aggravating cardiac fibrosis and diastolic dysfunction [137]. In women, the risk of diabetes is closely linked to a decline in estrogen levels, as lower E2 levels during menopausal transition are associated with an increased risk of diabetes, reflecting the effects of ovarian aging [138]. Research further suggests that postmenopausal women, despite enhanced insulin sensitivity, exhibit reduced non-insulin-dependent glucose-handling capabilities and diminished pancreatic insulin secretion, as indicated by decreased plasma C-peptide levels [139]. Longitudinal studies have highlighted that women undergoing oophorectomy (leading to abrupt and severe E2 deficiency) face a markedly higher risk of developing diabetes than those undergoing natural menopause, with a 57% increased risk of diabetes observed over 9 years [140, 141]. Diabetes research in female rodent models, including oophorectomized Sprague–Dawley rats, revealed worse glucose tolerance than in animals whose ovaries were not removed [142,143,144,145]. Large randomized controlled trials have indicated that menopausal hormone therapy (MHT) can reduce the incidence of T2D in women [142, 143]. However, given the complex balance between risks and benefits and because the impact of MHT on diabetes prevention is not a primary outcome in RCTs, MHT is neither appropriate nor FDA-approved for preventing T2D in women [144]. The role of estrogen under hyperglycemic conditions in female diabetic patients can also lead to sex-specific endothelial dysfunction, which is potentially linked to the estrogen-induced increase in PKCβ expression and O²⁻ production [113], improving vasodilation capacity by elevating NO production, reducing vascular resistance, and aiding in the maintenance of the EF. Moreover, hyperglycemia can alter the balance of ERs, thereby increasing oxidative stress and vasoconstrictor levels [146,147,148].

Previous studies have confirmed that estrogen compounds effectively protect the survival and function of pancreatic β-cells in response to various diabetic injuries [149,150,151,152]. GLP-1 is an agonist that targets pancreatic β-cells and stimulates insulin secretion, thereby lowering blood glucose levels [153]. As both GLP-1 and E2 have beneficial effects in diabetes treatment, some studies have proposed combining these two therapeutic strategies to exploit their synergistic effects, offering a new treatment approach for diabetes and related cardiovascular diseases such as HFpEF.

Compared to single GLP-1 or E2 agonists, GLP1-E2 combination therapy showed greater efficacy in lowering hyperglycemia, protecting β-cells, and alleviating diabetes-related complications [154,155,156]. Research has indicated that this combination therapy significantly enhances glucose-stimulated insulin secretion in human islets, outperforming single agonists [157]. The underlying mechanism of action involves GLP-1 entering β-cells through the GLP-1 receptor (GLP-1R), releasing E2 under lysosomal acidification, and activating ERα, inducing multiple protective signaling pathways. GLP1-E2 significantly enhances anti-apoptotic pathways (e.g., PI3K/Akt signaling), suppresses inflammation and oxidative stress, and improves insulin sensitivity and glucose metabolism, providing substantial protection for pancreatic β-cells [154]. Additionally, by precisely targeting the release of estrogen, GLP1-E2 avoids the systemic side effects associated with conventional estrogen therapy, such as adverse effects on the breast and uterus [158]. GLP1-E2 is particularly suitable for high-risk populations, such as postmenopausal women, and has great potential for preventing and managing diabetes and HFpEF. Future research should validate its safety and efficacy, optimize drug design to enhance therapeutic outcomes, and promote broader clinical applications.

Cardiac hypertrophy, myocardial wall thickening, and increased left ventricular mass are markedly pronounced in women with T2D [159]. A hallmark of diabetic cardiomyopathy is left ventricular hypertrophy, cardiac remodeling, progression of cardiac diastolic dysfunction, and subsequent clinical signs of HF with a normal EF [160]. By promoting the expression of the estrogen receptor ERα, activating the Akt signaling pathway, and inhibiting the TGF-β signaling pathway, estrogen regulates collagen expression and suppresses cardiac hypertrophy and fibrosis, ultimately protecting H9c2 cardiomyocytes from damage caused by advanced glycation end products (AGEs), preventing diabetes-induced cardiac fibrosis in db/db mice, and improving cardiac function [161]. Thus, estrogen and its receptors greatly affect the development of diabetes and associated HF, with changes in estrogen levels potentially increasing the risk of cardiac disease, leading to a increased incidence of HF.

Interplay among estrogen, AF, and HFpEF

Epidemiological studies have shown that the incidence of AF and HFpEF is increasing annually, with an increasing rate of comorbidities. For instance, there is a higher prevalence of AF in patients with HFpEF, ranging from 15 to 40%, depending on the research cohorts and diagnostic methods for AF [81]. AF has also been identified as a major risk factor of new-onset HFpEF, and its presence often predicts the occurrence of HFpEF more effectively than HFrEF [24]. Additionally, the development of AF may have a greater impact on patients with HFpEF than those with HFrEF and is highly associated with adverse outcomes in patients with HFpEF [162,163,164,165,166]. AF is an independent risk factor for new-onset HFpEF in women but not in men [167]. Furthermore, among the participants with AF, the incidence of HFpEF is higher in women than that in men [168].

AF is a common complication in patients with HFpEF, and a strong interplay exists between the two conditions. AF can serve as an early marker of HFpEF, potentially indicating changes in cardiac structure and function [169]. AF can lead to left ventricular dilation, impaired atrial function, and aggravated atrial fibrosis, which may directly trigger HFpEF [170]. Additionally, AF may result in irregular ventricular rates, particularly irregularities owing to variability in atrioventricular conduction, which may lead to decreased ventricular function and cardiac output [171]. More critically, AF may also cause serious complications, such as thrombus formation and embolism, further exacerbating the condition of patients with HFpEF [172].

In contrast, HFpEF promotes the development of AF. A recent study revealed that in a group of 450 patients newly diagnosed with HFpEF, after a median follow-up of 3.7 years, 32% of patients progressed to AF. This translates to approximately 69 new cases of AF per 1000 patients annually [165]. In patients with HFpEF, the compliance and mechanics of the left atrium (LA) gradually deteriorate, leading to LA enlargement, potentially progressing to permanent AF [169]. The most common mechanism by which HFpEF leads to AF is structural and functional LA remodeling [173, 174]. Compared to age-matched control groups, the LA volume in patients with HFpEF significantly increases, which is more pronounced than that in patients with hypertensive heart disease without HF [173]. LA remodeling manifests differently in different HF subtypes. Although eccentric remodeling is more common in HFrEF, it is characterized by increased LA stiffness, which is more likely to trigger AF [175]. Abnormal ventricular diastolic function in patients with HFpEF may also lead to increased atrial pressure, triggering AF [169].

Although the incidence of AF is higher in men than that in women [176, 177], women exhibit more severe symptoms, including atypical symptoms, and have a worse quality of life. Furthermore, camapered to men, women face a higher risk of adverse events related to AF, such as stroke and death [178, 179]. Notably, women generally develop AF later than men [180]. Data from the Framingham Heart Study show that 74% of female patients with AF are aged ≥ 70 years, while the corresponding proportion in men is only 58% [181]. The incidence of AF is lower in premenopausal women; however, it significantly increases after menopause, especially in women aged > 50 years, suggesting a close association between changes in estrogen levels and the occurrence and development of AF [182]. The effect of estrogen replacement therapy (HRT) on AF has shown contradictory results across studies, indicating that the effect of this treatment on AF risk may be complex. According to long-term follow-up data from the Italian Tamoxifen Study Group, women treated with tamoxifen had a significantly higher risk of developing AF than those in the placebo group, with a relative risk of 1.74 [183], suggesting that certain hormone-related drugs may be associated with AF occurrence. In addition, the Women’s Health Study found that the use of estrogen-only replacement therapy (ERT) was associated with an increased risk of AF, whereas combined estrogen and progesterone therapy did not show a similar association [184]. This indicates that estrogen alone may contribute to an increased risk of AF. A retrospective cohort study further explored the impact of different types of HRT on the risk of AF and revealed that the incidence of AF, stroke, and major adverse cardiac events (MACE) was lower in the estradiol treatment group than that in the conjugated equine estrogen (CEE) treatment group. The incidence of AF was significantly high in women receiving CEE therapy, with an adjusted risk ratio of 1.96, suggesting that CEE may increase the risk of AF [185]. Meanwhile, in women surviving myocardial infarction, the incidence of AF was lower among those receiving HRT, especially in the group aged ≥ 80 years, where the risk of AF was significantly reduced in women receiving overall HRT or vaginal estrogen. This suggests that HRT may help reduce the risk of AF [186]. However, except for estradiol-only HRT, other types of HRT (such as combined estrogen therapy) have been observed to increase the risk of AF in some cases, indicating that the effect of HRT on AF is not only influenced by ERs but may also involve the effects of other female steroids [187].

AF and HFpEF share similar risk factors, including hypertension, obesity, diabetes, and chronic inflammatory states [188, 189]. This is particularly evident in postmenopausal women, wherein a decline in estrogen levels significantly exacerbates these risk factors. The lack of estrogen may contribute to accelerated atrial fibrosis, increased cardiac load, and the promotion of metabolic dysregulation, thereby intensifying the interaction between AF and HFpEF [165, 190,191,192]. Although HRT has not shown survival benefits in large studies, it can improve diastolic function in postmenopausal women [193, 194]. This finding may help explain why, in some cases, HRT has been observed to reduce the incidence of AF and may indirectly or directly impact the occurrence of HFpEF. However, further research is needed to clarify the exact role of HRT and to potentially minimize the side effects of estrogen in the treatment of AF and HFpEF.

Molecular mechanisms of sex differences in animal HFpEF models

The prevalence of HFpEF is characterized by a significant sex imbalance, with a maximum of two-thirds of patients being female in some epidemiological studies; female patients often exhibit more severe symptoms than males [195, 196]. However, some perspectives have questioned the elevated presence of females in HFpEF, suggesting that this might be due to the longer average lifespan of females and their tendency to enter old age, as the risk of HF generally increases with age [5, 99]. However, the exact role of sex in the development of HFpEF remains unclear. To gain a deeper understanding of the association of sex and HFpEF, one research approach involves exploring sex differences through animal models, which allow the control of genetic and environmental factors and the direct study of sex-specific factors in HF, such as through procedures like gonadectomy.

Aging model (age-related HFpEF phenotype)

Age is a key factor in the development of HFpEF, and several models have been used to study how aging affects sex differences in HFpEF. The senescence-accelerated mouse model SAMP8 is one such example, in which aging leads to left ventricular stiffness, diastolic dysfunction, and myocardial fibrosis [197, 198]. Interestingly, variations in left ventricular structure and function in FVB/N inbred mice and Fischer 344 rats are sex-related [197, 199]. Male FVB/N mice exhibit diastolic dysfunction as early as 12 months of age, whereas female mice do not exhibit similar symptoms. In contrast, aged female Fischer 344 rats are more prone to diastolic dysfunction and left ventricular remodeling than their male counterparts [200, 201].

Spontaneously hypertensive rat (SHR) model (hypertension-induced model)

The SHR model is another commonly used model for studying HFpEF. In this model, hypertension occurs naturally, and the rats exhibit features similar to those of HFpEF, such as left ventricular hypertrophy and diastolic dysfunction. Studies comparing male and female SHR have shown sex-related differences in the progression of HF. For example, male SHRs tend to exhibit a more rapid onset of noticeable aortic dysfunction and left ventricular hypertrophy than females [202]. Additionally, female SHRs showed delayed matrix metalloproteinase-2 activity in the aorta and heart, along with higher levels of AT2R and MasR mRNA, demonstrating relatively less fibrosis and better elastin preservation [202]. These findings suggest that estrogen plays a protective role in the progression of hypertension-related HFpEF.

Diabetic HFpEF model

Diabetes is an important risk factor for HFpEF. Studies using animal models of diabetes, particularly streptozotocin (STZ)-induced T1DM models, have provided crucial experimental evidence for understanding the role of hormonal and metabolic dysregulation in HFpEF development. STZ-induced T1DM has become a widely accepted animal model for studying diabetes-related CVD and HF [203, 204]. Both male and female diabetic mice exhibit characteristic features of HFpEF, including decreased cardiac output and increased cardiac ANP mRNA expression, indicating the onset of HF [205]. Additionally, despite preservation of the EF, STZ-treated rats showed a reduction in left ventricular end-diastolic volume, suggesting left ventricular diastolic dysfunction [205].

Notably, in a diabetic mouse model, diastolic dysfunction was more pronounced in female mice than that in male mice. Specifically, the E/E’ ratio in female mice significantly increased, while no significant change was observed in male mice [204]. This indicates that STZ-induced diabetes makes female mice more sensitive to diastolic dysfunction, even though their glucose levels are lower than those in male mice [204]. These findings highlight the significant role played by sex in diabetes-induced cardiovascular pathological changes.

High-fat diet (HFD) combined with L-NAME (obesity and hypertension model)

One of the most commonly used HFpEF animal models is the HFD combined with L-NAME (an NO synthase inhibitor). This model mimics the common risk factors for obesity and hypertension, both of which are prevalent in patients with HFpEF. In studies using this model, male and female mice showed distinct differences in their responses to the induced HFpEF phenotype. One study showed significant differences in cardiac function and structure between male and female mice in an HFpEF model by feeding C57BL/6N mice an HFD combined with L-NAME for 7 weeks. Male mice exhibited obvious concentric left ventricular hypertrophy and diastolic dysfunction when on this diet, characterized by increased heart mass, left ventricular remodeling index, and E/E′ ratio. In contrast, the changes in female mice were milder, with a smaller increase in the E/E′ ratio and insignificant variations in heart mass and ventricular remodeling index [206]. This finding suggests that in HFpEF, female mice may have a certain degree of protective effect compared to male mice. Ovariectomy surgery was performed and these animals were subsequently subjected to HFpEF treatment for 15 weeks. The reults showed that the heart weight and diastolic function of female mice were similar. This suggests that the protective effect of the female sex on HFpEF may not be entirely mediated by female hormones. However, this experiment was only conducted in young animals, and research on HFpEF in older animals and humans has not yet been conducted. Additionally, using ovariectomy to simulate the physiological menopausal state may not be entirely accurate, and the effects of male hormones have not been thoroughly studied. In another study using the same HFD combined with L-NAME-fed HFpEF C57BL/6 J mouse model, contradictory differences were observed compared to the previous study [207]. Although both male and female mice showed increased body weight, fat mass, left ventricular mass, and fat pad weight, as well as decreased exercise tolerance, the E/A ratio, E/E′ ratio, and lung weight in female mice were higher than those in male mice [206, 207]. The authors explained that this difference might be due to the different genetic backgrounds of the mice and pointed out a significant gene-sex interaction [208]. Mechanistically, in mice and humans, males often show higher mitochondrial DNA content and mitochondrial activity than females, which is partially regulated by sex hormones. RNA-seq analysis further revealed that acyl-CoA synthetase long-chain family member 6 (Acsl6) is a predominant downstream player of sex hormones that regulate mitochondrial activity and metabolic alterations underlying sex bias [207].

Despite employing nearly identical methods, these two studies yielded diametrically opposite results, as shown in Fig. 3. Researchers have suggested that this discrepancy may stem from the use of different mouse strains. However, the specific mouse strain that has the highest correlation with and reflects the pathology of HFpEF in humans remains unclear. Future studies should consider utilizing a more acute animal model as well as age, sex, and genetic background to achieve a more comprehensive understanding of the role of sex in HFpEF in humans.

Fig. 3

Molecular mechanisms of sex difference in animal HFpEF models. In the HFpEF model induced by a high-fat diet combined with L-NAME, significant differences in HFpEF phenotypes were observed between male and female mice. In C57BL/6N mice, male mice exhibited concentric left ventricular hypertrophy, diastolic dysfunction, and significant cardiac remodeling, whereas female mice showed relatively milder changes, with a lower increase in the E/E′ ratio and minimal cardiac remodeling. However, in C57BL/6 J mice, although both male and female mice exhibited obesity and reduced exercise tolerance, female mice had higher E/A and E/E′ ratios and increased lung weight compared to males. Estrogen may also influence mitochondrial gene expression in cardiomyocytes, leading to a further reduction in mtDNA content and more pronounced mitochondrial dysfunction in female mice. These two studies yielded contradictory results, and further research is needed to determine which model better reflects the pathological features of human HFpEF

Conclusion

The role of sex in HFpEF is complex and involves multiple factors, necessitating further research to elucidate the specific mechanisms underlying sex differences and how these differences influence the development and progression of HFpEF. Estrogen is closely associated with the occurrence and development of HFpEF. Through various mechanisms, such as regulating fat distribution, lowering BP, improving blood glucose levels, and reducing the occurrence of AF, estrogen reduces the incidence of risk factors, thus affecting HFpEF (Table 1). Therefore, a decline in estrogen levels may interact with these risk factors, leading to the weakening effect of estrogen on the cardiovascular system and collectively increasing the risk of HFpEF in women. Hence, future research should further explore the specific mechanisms of estrogen in various phenotypes of risk factors in HFpEF, as well as the potential value of HRT in the treatment of HFpEF. Overall, this review provides a comprehensive understanding for the role of estrogen in HFpEF in a sex-specific manner.

Table 1 Summary of current research on the impact of estrogen on HFpEF risk factors