Introduction

Primary aldosteronism (PA) is a leading cause of secondary hypertension, with a prevalence ranging between 6% and 20%.^1,2 A higher degree of hypertension is associated with a greater prevalence of PA.^1,3 Multiple studies have shown detrimental effects of excess aldosterone on cardiac structure and function.¹

Cardiac magnetic resonance (CMR) imaging is an optimal noninvasive tool for assessing the size of heart chambers, left ventricular (LV) remodeling, and LV function.⁴ Previous studies have demonstrated, with the use of CMR, that the presence of late gadolinium enhancement (LGE) is consistent with focal cardiac fibrosis in patients with PA.^5,6 These alterations exceed the impact of hypertension alone and are therefore more pronounced than those detected in control essential hypertension patients matched for blood pressure (BP) values.⁵ However, detailed characteristics of PA patients with LGE as well as comparisons with patients without LGE on CMR are lacking. Additionally, it remains unknown whether patients with and without LGE differ in terms of heart injury, LV overload, and systemic inflammation. Therefore, we aimed to assess markers of heart injury (high‑sensitivity cardiac troponin T [hs‑cTnT]), overload (N‑terminal pro–B‑type natriuretic peptide [NT‑proBNP]), and inflammation (high‑sensitivity C‑reactive protein [hs‑CRP]) and their relationship with cardiac fibrosis in patients with PA.

Patients and methods

Results

A total of 32 patients were screened. Two of them were excluded, one due to poor image quality of the LGE images caused by breathing artifacts related to claustrophobia, and the other due to incomplete biochemical assessment. The final analysis included 30 patients, of whom 15 were men (50%). Mean (SD) age of the patients was 51 (10.4) years.

LGE was present in 21 patients (70%), who composed the LGE+ group, whereas the remaining patients were assigned to the LGE– group. All patients exhibited a nonischemic pattern from Supplementary material, Figure S1A and S1B). There were no differences in the demographic data (Table 1). Notably, neither office BP values nor the duration of hypertension differed between the 2 groups.

We subsequently compared the LV parameters between the LGE– patients and LGE+ individuals (Table 1). The groups did not differ in terms of LV dimensions or LV systolic function. However, there were differences both in the absolute LVM (P = 0.04) and the LVM indexed to body surface area (P = 0.01); the patients with myocardial fibrosis presented greater LVM. Additionally, there was a difference in global longitudinal strain (GLS), with the patients with LGE showing impaired GLS, as compared with the individuals without LGE (P = 0.02; Table 1).

Finally, we analyzed the levels of biochemical markers in both groups. In the LGE+ group, almost half of the patients had NT‑proBNP levels greater than or equal to 125 pg/ml, whereas none of the patients in the LGE– group exhibited such values (Table 1; P = 0.01). The levels of hs‑cTnT and hs‑CRP did not differ between the groups. We also checked correlations between concentrations of these 3 biomarkers and aldosterone levels. None of the analyzed biomarkers correlated with aldosterone (for troponin, ρ = 0.29; P = 0.12; for hs‑CRP, ρ = 0.21; P = 0.27; for NT‑proBNP, ρ = 0.2; P = 0.29).

There were weak positive correlations between the LVM index (LVMi) and the levels of troponin (ρ = 0.63; P <⁠0.001) and NT‑proBNP (ρ = 0.39; P = 0.03), and no correlation between the LVMi and hs‑CRP (ρ = 0.35; P = 0.06). Additionally, both absolute and indexed LVM correlated weakly with the aldosterone level (ρ = 0.42; P = 0.02 and ρ = 0.43; P = 0.01 for absolute and indexed LVM, respectively).

All analyzed markers showed a positive correlation with GLS (ρ = 0.67; P <⁠0.001; ρ = 0.45; P = 0.01; and ρ = 0.39; P = 0.04 for hs‑cTnT, NT‑proBNP, and hs‑CRP, respectively; Supplementary material, Figure S1). GLS also showed a correlation with the aldosterone level (ρ = 0.46; P = 0.01) and LVMi (ρ = 0.75; P <⁠0.001), but not with the duration of hypertension (P = 0.46).

Discussion

The main findings of our study are as follows: 1) patients with newly diagnosed PA often exhibited LGE consistent with focal cardiac fibrosis; in all cases, it was of nonischemic etiology; 2) the levels of biomarkers studied did not differ between the patients with and without focal fibrosis; 3) LV focal fibrosis in the PA patients was associated mainly with LVM; and 4) LV GLS was impaired in the patients with cardiac fibrosis and showed a correlation with all analyzed biomarkers.

In their landmark study, Freel et al⁵ used CMR to demonstrate for the first time that PA patients exhibited a high rate of cardiac fibrosis (70%, similarly to our findings), which was more pronounced than in patients with essential hypertension. They included patients both before and after causal treatment for PA. Notably, the inclusion criterion was CMR performed either before or within 1 year of adrenalectomy, or before the start of pharmacotherapy for excess aldosterone. In contrast, our analysis included only newly diagnosed cases of PA. This is particularly important since many studies and meta‑analyses have demonstrated a significant reduction in LVM after PA treatment.¹⁰ Regression of LV hypertrophy is more rapid after adrenalectomy than after specific medical therapy. This fact also explains why, in the study by Freel et al,⁵ the presence of fibrosis was independent of LVM. It can be speculated that patients after PA treatment (included in their study) demonstrated a reduction in myocardial mass, whereas the presence of fibrosis did not change, resulting in a lack of association between LVM and the presence of LGE. Additionally, although the development of cardiac fibrosis is a rather long‑term process, one cannot exclude the possibility that within 1 year between CMR and the diagnosis or specific PA treatment, new foci of LGE appeared, which could have also caused bias when the relationship between LGE status and LVM was assessed.

Data on the associations between markers of injury, overload, and inflammation and the risk of developing fibrosis in PA patients are scarce. In our study, none of the biomarkers, namely, troponin, NT‑proBNP, or CRP, were significantly associated with the presence of focal cardiac fibrosis. There are conflicting data regarding the impact of aldosterone on NT‑proBNP levels in patients with PA.^11-13 None of the studies assessed the relationship between NT‑proBNP levels and focal fibrosis in individuals with PA. Wu et al¹² demonstrated that PA patients with abnormally high NT‑proBNP levels had higher LVEDV and LVM indices and lower LVEF. We observed the same relationship; however, it was of borderline statistical significance (data not shown). Nevertheless, Wu et al¹² did not evaluate patients for the presence of LGE. In our study, all patients without focal fibrosis had normal NT‑proBNP levels. In contrast, almost 50% of the patients with LGE had elevated NT‑proBNP concentrations. This fact reflects increased atrial and ventricular wall stress in PA patients, demonstrating focal fibrosis. Whether there is a causal relationship between fibrosis development and NT‑proBNP levels in PA patients or whether this relationship is only a reflection of increased overload needs to be elucidated in further prospective studies. Notably, all patients in our study had normal LV systolic function.

We observed elevated troponin concentrations in a few patients, with a 2‑fold greater prevalence in the individuals with LGE. This difference, however, did not reach statistical significance because of the relatively small group of patients. Additionally, we detected a positive correlation between hs‑cTnT levels and LVMi, which is intuitively true. The larger the LVM is, the more troponin is released.

Aldosterone induces inflammation by various mechanisms, including but not limited to activating inflammatory pathways, generating reactive oxygen species, and promoting damage, remodeling, and fibrosis, which further exacerbates inflammation.¹⁴ While low levels of hs‑CRP reflect a low systemic inflammatory status, higher values indicate increased risk and poorer cardiovascular outcomes. In our study, none of the PA patients without focal fibrosis had elevated hs‑CRP concentrations. In the LGE+ group, almost 25% of the patients had hs‑CRP levels above the normal range. The elevated inflammatory status in the patients with LGE possibly reflects an interplay between fibrosis and inflammation and a vicious cycle in which inflammation aggravates the severity of fibrosis, and vice versa. It remains unknown whether anti‑inflammatory therapy, in addition to mineralocorticoid receptor antagonists, could limit or slow the development of fibrosis. In an animal model, finerenone, a nonsteroidal mineralocorticoid receptor antagonist, was shown to have antifibrotic and anti‑inflammatory effects on cardiomyocytes.¹⁵ More research is needed to determine whether this effect is accompanied by a reduction in fibrosis in PA patients.

Although we did not find a difference in LVEF between the LGE+ and LGE– groups, they differed in terms of LV GLS. It is not surprising that strain analysis showed impaired GLS in the patients with fibrosis, as compared with the individuals without it. However, an interesting observation is that hs‑cTnT, NT‑proBNP, and hs‑CRP levels all showed a correlation with GLS, whereby LV GLS impairment increased along with the rise in the biomarker levels. The same was true for aldosterone concentration. To the best of our knowledge, this study is the first to show a relationship between the analyzed biomarkers and LV GLS in PA patients. Although this is a cross‑sectional but not longitudinal study, one can speculate that higher aldosterone levels and elevated inflammatory status cause subclinical LV systolic dysfunction, whereas higher hs‑cTnT and NT‑proBNP levels are markers of this dysfunction.

Our study has several limitations. First, it included a relatively small number of patients, which resulted in a lack of statistical significance in some analyses. However, there is still a paucity of data on the impact of excess aldosterone on cardiac structure and function. Our study should be regarded as a preliminary, hypothesis generating investigation, and the findings should be confirmed in larger studies. Additionally, the analyses did not include an evaluation of diffuse fibrosis. Different mechanisms are responsible for the development of focal replacement fibrosis (seen as LGE foci) and diffuse fibrosis, as well as the expansion of the extracellular volume. The latter can also be measured with the use of CMR via pre- and postcontrast T1 mapping.^4,9,16-19 Although these data are available, in the present study, we focused on focal fibrosis (scar tissue formation). We did this because of space limitations and to avoid obscuring the LGE findings. Diffuse fibrosis will be assessed in a separate study. Finally, we assessed conventional biomarkers that are commonly used in clinical practice, omitting novel, potentially causal biomarkers of hypertension, or other parameters that might be related to fibrosis.²⁰ Further investigation is needed to determine whether these novel biomarkers are associated with a risk of developing cardiac fibrosis.