Chronic heart failure (CHF) is one of the most acute problems in modern cardiology. According to the national registers, in the Russian Federation, there are around 8 mln people with this pathology, and its rate is constantly increasing [1]. Economic costs of the treatment of patients with CHF in the developed countries occupy 2–3% of the whole healthcare budget, and a five-year survival rate does not exceed 50%.

For a long time, CHF was associated with a reduction of the left ventricular (LV) ejection function. However, with time, it became evident that a significant place in the structure of the disease was occupied by patients with preserved ejection fraction (pEF). Recent research data showed that the severity and prognosis of CHFpEF were comparable with CHF with reduced EF (CHFrEF) at the populational level. Besides, according to some analyses, the rate of hospitalizations for CHFpEF exceeds the rate of hospitalizations for CHFrEF [2][3]. The situation is aggravated by an increasing morbidity rate with CHFpEF and the lack of reliable methods of treatment to decrease the lethality rate. This provides the need for a search for effective tools of stratifications of the risks of CHFpEF that would promote the optimization of therapy and provide a scientific basis for further studies of pharmaceutical drugs.

A significant contribution to the development and progressing of CHF is made by the activation of the renin-angiotensin-aldosterone system (RAAS) and the hyperproduction of its final effector aldosterone. Numerous studies showed that an increase in the plasma level of aldosterone led to a significant increase in the development of myocardial infarction, atrial fibrillation, cerebrovascular events, chronic kidney disease, cardiovascular and all-cause lethality [4][5][6][7]. In patients with primary hyperaldosteronism, cardiovascular pathology and lethality several-fold exceeds these rates in the general population [5][8][9].

Among patients with CHF, the largest evidence-based database of the prognostic significance of aldosterone and the effectiveness of its blockers was obtained in patients with CHFrEF. At the same time, the data on the role of hyperaldosteronemia in patients with CHFpEF are sparse, and the application of mineralocorticoid receptor antagonists has not been successful so far. Earlier, the author’s studies showed that hyperaldosteronemia in patients with CHFpEF often developed after a long-term application of RAAS blockers [10] because of closely associated worsening of carbohydrate metabolism [11], development of prognostically unfavorable ventricular arrhythmias [12], and progressing of LV remodeling [13].

The study aimed to evaluate the prognostic value of secondary hyperaldosteronism in patients with heart failure with preserved ejection fraction (HFpEF).

A total of 158 patients with CHFpEF took part in the prospective study. The main criteria of study entry included primary hyperaldosteronism, secondary hyperaldosteronism of renal or hepatic genesis, and application of mineralocorticoid receptor antagonists for 6 weeks before the study. A detailed description of the criteria of entry and exclusion is described in a previously published article [11]. All patients signed informed consent for participation in the study. The protocol of the study and the form of informed consent for patients were approved by the local ethical committee on the issues of bioethics of the Donetsk National Medical University named after M. Gorky (protocol No. 2 dated April 22, 2016).

At the stage of patients’ inclusion in the study, the following data were collected: patients' sex, age, smoking status (during the study), NYHA functional class, associated diseases, pharmaceutical therapy, blood pressure (BP), body mass index, the main laboratory or echocardiographic parameters.

Laboratory analyses included the evaluation of the level of aldosterone, hemoglobin, sodium, creatinine, NT-proBNP, low-density lipoprotein cholesterol (CH-LDL), fasting glucose, glycated hemoglobin (HbA1c), blood potassium.

The levels of aldosterone were measured by the immunoenzyme method with a photometer Multiskan (Thermo Electron, Germany) and test-systems DRG (Germany). Blood samples were collected after a 30-minute rest in a lying position in the morning after fasting, 3-4 hours after waking up. Hyperaldosteronemia was diagnosed in plasma levels of aldosterone higher than the upper threshold of the referent interval (> 160 pg/ml).

The level of serum creatinine was measured by the method of Jaffe. Glomerular filtration rate was calculated by the formula CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration). The level of NT-proBNP was measured with a qualitative immunologic assay with Cardiac Reader (Roche, Germany) and standard kits (Roche Diagnostics). The threshold value of NT-proBNP or the verification of the CHF was 125 pg/ml. The evaluation of the glucose level in venous blood was made after fasting with colorimetric glucose oxidation, HbA1c – with a turbidimetric method, CH-LPL – with a method of a direct enzymatic assay using an analyzer Olympus 480 (Beckman Coulter, USA). The level of potassium in the blood serum was measured with ion-selective electrodes and an automated biochemical analyzer Cobas C 311 (Roche Diagnostics, Germany).

Transthoracic echocardiography was performed in M-modal, two dimensions, and Doppler modes with a system of ultrasound heart examination Aplio MX SSA-780 A (Toshiba Medical Systems Corporation, Japan) in the position of the patient lying on the left side or back. The study was performed in the left parasternal position by the long and short axes, as well as in the apical four- and two-chamber positions. The volume of the left atrium (LA) relative to the body surface area was identified. The end-diastolic and end-systolic volumes, left ventricular ejection fraction (LVEF), the thickness of the posterior wall of LV in the diastole, interventricular septum thickness, relative thickness of LV walls, and the myocardial mass index of LV were evaluated. The hypertrophy of LV was diagnosed in men with the myocardial mass index of LV ≥ 115 g/m2 and women ≥ 95 g/m2.

The condition of the diastolic function of LV was evaluated with an impulse Doppler study of the transmitral flow and tissue Doppler imaging of the diastolic contraction of LV. Such parameters as the maximal rate of the early diastolic filling (E) and systolic filling of atriums (A), septal and lateral velocity of the mitral valve orifice (e’), the indexed volume of the LA, and maximal rate of tricuspid regurgitation were evaluated. Diastolic dysfunction of LV was diagnosed in patients with a minimum of three of the following signs:

After the initial examination, patients were followed up. The primary endpoint was any-cause death. Patients were followed up for 36 months after the inclusions in the study or death depending on what occurred earlier. The secondary endpoints of the study were cardiovascular death, non-cardiovascular death, hospitalization for decompensation of CHF, changes in the NYHA functional class, and a significant decrease in LVEF.

The identification of the fact of hospitalization for CHF and classification of the death cause were made by two independent clinical specialists based on medical records and/or death certificates. In the case of disagreement, the final decision was made by the third clinical specialist. Cardiovascular lethality was additionally classified as death from fatal myocardial infarction, CHF, or other causes (sudden cardiac death, death from a stroke, death after cardiovascular intervention). Sudden cardiac death was defined as sudden unexpected death or death that occurred within 1 hour after the onset of the symptoms. Non-cardiovascular death was classified as death from fatal infection, malignant neoplasms, and other causes.

The evaluation of the NYHA functional class and LVEF was performed in survived patients 36 months after the beginning of the study. A significant decrease in LVEF was a criterion of a patient’s transfer from the category of pEF to the category of intermediate EF (iEF) or reduced EF (rEF).

The study results were analyzed on a PC using the software package Jamovi 1.2.2. Categorical data were expressed in percent, continuous data were expressed as the mean and standard error of the mean (M ± SD) or a median and interquartile interval Me (Q1-Q3). A comparison between the groups was performed using Student’s t-test for normal data distribution and Wilcoxon’s test for not normal distribution. To compare categorical parameters, the method of analysis of the contingency tables with the χ2 test was used. To reveal the factors associated with hyperaldosteronemia, one-factor and multi-factor step linear regression analyses were performed. The association between the features was expressed via the odds ratio (OR) with 95% CI and Cramer’s V-test. The differences were statistically significant at p < 0.05.

The initial laboratory study revealed hyperaldosteronemia in 59 out of 158 patients (37.3%, 95% CI: 30.0–45.0%). Significant differences were revealed between patients with high and normal aldosterone levels by age, NYHA functional class, rate of associated conditions, structural parameters of the myocardium, and levels of laboratory parameters (Table 1).

Table 1

The median follow-up was 32 months (28-38 months). During this time, 50 patients (37.6%) died. Cardiovascular death was registered in 32 patients (20.3%). Other causes of death were registered in 18 cases (11.4%). Sixty-five patients (41.1%) were hospitalized for CHF worsening. A detailed characteristic of outcomes is presented in Table 2.

A comparative analysis showed that patients with hyperaldosteronemia got hospitalized more frequently for decompensation of CHF and had a lower survival rate in comparison with patients with initially normal aldosterone levels. A detailed analysis of the causes of lethal outcomes showed that the differences in the levels of general lethality were observed due to cardiovascular death, while other causes of lethality did not differ between the groups. The most frequent cause of cardiovascular death was CHF. The differences in the rate of lethality from fatal myocardial infarction and other cardiovascular events were not statistically significant (Table 2). The analysis of certain causes of non-vascular death did not reveal any statistically significant associations with the level of aldosterone either.

Table 2

By the end of the follow-up period, in 40 (37%) out of 108 patients that survived, the NYHA functional class increased in comparison with the baseline data, and 35 (32.4%) of them moved from the category of pEF or rEF. The share of patients with NYHA functional class worsening was significantly higher than in the group of hyperaldosteronemia (Table 3).

Table 3

One-factor regression analysis showed that the presence of hyperaldosteronemia was associated with the general and cardiovascular lethality, hospitalization and death from CHF, worsening of the NYHA functional class, and decrease in LVEF (Table 4). However, after the correction for age, comorbid conditions, received therapy, echocardiographic parameters, and the level of biomarkers, the association between a high level of aldosterone and cardiovascular death was lost. In comparison with patients with a normal level of aldosterone, the presence of hyperaldosteronemia was associated with a significant increase in the risk of death from all the causes (corrected OR 1.64; 95% CI 1.23–7.65, P = 0.033), cardiovascular events (OR 1.86; 95% CI 1.02–6.97, P = 0.046), and death from CHF (OR 1.56; 95% CI 1.14–11.3, P = 0.021). In the multi-actor model, the influence of hyperaldosteronemia on CHF remained significant. In particular, it affected the rate of hospitalizations for decompensation of CHF, progression of the functional class of CHF, and decrease in EF.

 Table 4

Presently, a high prognostic significance of the level of aldosterone in plasma was demonstrated in patients with different forms of ischemic heart disease (IHD) [14][15][16][17]. The OPERA [18] study showed that higher levels of aldosterone in patients with myocardial infarction, even if they remained within the physiological norm, were associated with an increased risk of ventricular and supraventricular rhythm disturbances, repeated infarction, stroke, CHF, and death. A major study LURIC that included more than 3,000 patients with IHD showed that the plasma level of aldosterone was an independent predictor of cardiovascular events and all-cause death even when the level of aldosterone was lower than the upper threshold. The analysis of certain causes of death showed aldosterone was associated with a higher risk of sudden cardiac death and fatal stroke. This association was revealed both in patients with chronic IHD and acute coronary syndrome. In patients with the coronary syndrome, the association between the level of aldosterone and cardiovascular lethality remained even in patients with normal or minimally reduced systolic fraction of the LV, which confirmed the reports on a positive effect of mineralocorticoid receptor antagonists in patients with myocardial infarction regardless of the presence of systolic dysfunction [19][20]. A detailed analysis showed that the level of aldosterone higher than 48 pg/ml (lower threshold of the norm) could lead to damage to target organs.

The results of the study ILLUMINATE [21] on the evaluation of the effectiveness of cholesterol ester transfer protein inhibitor (CETP) torcetrapib, which was early terminated, also confirmed an important prognostic role of hyperaldosteronemia. This is the first drug from this pharmacological group with a mechanism of action to increase the level of HDL cholesterol. The application of torcetrapib in patients with cardiovascular diseases led to a significant increase in the level of HDL cholesterol by 70% and a decrease in the level of LDL cholesterol by 25%. Still, it was associated with an increase in the rate of cardiovascular complications and lethality associated with an increase in the concentration of aldosterone and BP. In 2012, the results of the 3rd phase of dal-OUTCOMES [22] were published that evaluated the effectiveness of the second experimental CETP drug dalcetrapib. The study was early terminated when the results of the intermediate analysis did not show a decrease in the risk of cardiovascular complications in patients that received dalcetrapib in comparison with placebo despite an increase in the concentration of HDL cholesterol by 30%. According to some experts, one of the reasons for negative study results could be a minor but clinically significant increase in the level of systolic BP that could be caused by hyperaldosteronemia. The effect of CEPT inhibitors on the level of aldosterone was confirmed in the study ACCELERATE [23] dedicated to the evaluation of the effectiveness of evacetrapib in patients with high cardiovascular risk. The drug contributed to an increase in the level of HDL cholesterol but did not change the plasma concentration of aldosterone. Still, an improvement of lipid profile was not associated with a decrease in cardiovascular events and lethality.

An unfavorable effect of hyperaldosteronemia on the prognosis was also shown in patients with CHF. In 1990, the results of a small study CONSENSUS showed that the level of aldosterone was an independent predictor of death in patients with severe decompensated CHF [24]. Further clinical studies confirmed that higher concentrations of aldosterone were associated with an increased lethality rate in patients with CHF in a long-term follow-up. This risk increases in patients with a combined increase in the levels of aldosterone and cortisol [25].

However, the conducted studies on the evaluation of the influence of aldosterone on the prognosis of CHF included either patients with rEF or patients with different systolic functions of the LV. The author was the first to demonstrate the influence of aldosterone on the prognosis of the outcome in the population of patients with CHFpEF. The results of the present study showed that an increase in the plasma level of aldosterone was widespread among patients with CHFrEF even outside the phase of decompensation and registered in 37% of cases. It could be the event of secondary aldosteronism that is associated with arterial hypertension, different extra-cardiac conditions (obesity, chronic obstructive pulmonary disease, renal dysfunction, diabetes mellitus), and aldosterone escape phenomena during the treatment with RAAS inhibitors. Possible mechanisms of association between hyperaldosteronemia and associated conditions are described in detail in previous publications [26].

It should be noted that the influence of hyperaldosteronemia on the development and outcome of CHF also remained after such generally accepted predictors of cardiovascular prognosis as age, structural parameters of the myocardium, and the level of some biomarkers. This suggests that the inclusion of aldosterone in the existing models of CHF prognosis could improve their predictive value and optimize medical therapy in patients with a high risk.

Presently, there is no significant evidence database for the treatment of patients with CHFrEF. The results of the TOPCAT study and data of major meta-analyses show that mineralocorticoid receptor antagonists decrease lethality and the rate of hospitalizations for cardiovascular events only in patients with rEF but not pEF [27, 28]. At the same time, it is known that CHFpEF is a quite diverse event by the cause and mechanisms of the development. Probably, aldosterone antagonists will not prove the effectiveness in the improvement of the prognosis in a wide population of patients with CHFpEF but could contribute to a decrease in the risk of lethality in a certain category of patients. This study suggests that the indication of mineralocorticoid receptor blockers can be feasible in patients with hyperaldosteronemia. However, further studies are needed for the verification of this hypothesis.

Apart from the main aim of the study to evaluate the prognostic significance of hyperaldosteronemia in patients with CHFpEF, several additional tasks were fulfilled. Three-year survival of patients with CHFpEF was observed in 68.4% of patients even with optimal pharmaceutical therapy. Besides, it was revealed that during a 3-year follow-up, every third patient with CHFpEF moved to the category of iEF or rEF, and hyperaldosteronemia remained an independent predictor of worsening of the systolic function of the LV.

A certain limiting factor of the study is a relatively small sampling of patients, which limited the possibility of the study of associations between the level of aldosterone and certain causes of cardiovascular lethality such as sudden cardiac death, stroke, and cardiosurgical interventions. Further major studies will provide answers to these questions.

In patients with CHFpEF, a high aldosterone level in the blood is associated with a significant increase in the risk of hospitalizations for disease decompensation, CHF progression, and disorders in the systolic function of the LV. The presence of hyperaldosteronemia in this cohort of patients is an independent predictor of general and cardiovascular death and death from CHF.