Mineralocorticoid Receptor Antagonists in Dialysis

Abstract

Background

Aldosterone regulates genes controlling fluid and electrolyte balance through mineralocorticoid receptor (MR) activation. Sustained MR activation promotes inflammation, fibrosis, sodium retention, and myocardial remodeling. MR antagonists (MRAs) block aldosterone binding in the kidney, heart, and vasculature and are classified as steroidal or nonsteroidal. Although large, randomized trials have confirmed the benefits of steroidal MRAs in heart failure and of finerenone in diabetic kidney disease, these pivotal studies systematically excluded patients with kidney failure receiving dialysis. Over the past decade, several phase 2 and 3 trials have evaluated MRAs in dialysis, yielding heterogeneous and sometimes conflicting results. This review summarizes the biological rationale, evolving clinical evidence, and future directions for MR blockade in dialysis.

Summary

Pharmacokinetic studies indicate that spironolactone and eplerenone are highly protein-bound, hepatically metabolized, and not dialyzable, supporting cautious use in dialysis with potassium monitoring. Early randomized trials from East Asia suggested potential cardiovascular benefit with spironolactone 25 mg daily; however, these studies were underpowered and reported relatively few outcome events. More definitive evidence is now available from two large multicenter randomized controlled trials. In ALCHEMIST (n = 644), spironolactone 25 mg daily did not reduce major adverse cardiovascular events compared with placebo over a median follow-up of 2.7 years [hazard ratio (HR) 1.00, 95% confidence interval (CI) 0.73–1.36]. Similarly, in ACHIEVE (n = 2,538), spironolactone failed to reduce the composite of cardiovascular death or hospitalization for heart failure (HR 0.92, 95% CI: 0.78–1.09) over 1.8 years and was associated with higher rates of hyperkalemia. A contemporary meta-analysis incorporating these trials confirmed neutral efficacy but higher rates of asymptomatic hyperkalemia and gynecomastia.

Key Messages

Current evidence indicates that steroidal MRAs confer no cardiovascular or survival benefit in maintenance dialysis and modestly increase the risk of hyperkalemia and endocrine adverse effects.

Introduction

Aldosterone, secreted by the zona glomerulosa of the adrenal cortex, binds to the mineralocorticoid receptor (MR) – a cytoplasmic receptor expressed in the kidney, heart, vasculature, brain, immune cells, and fibroblasts [1]. The aldosterone-MR complex regulates genes controlling fluid and electrolyte balance [2]. However, sustained MR activation, also drives inflammation, fibrosis, hypertrophy, and sodium retention, promoting myocardial remodeling and cardiorenal injury [3].

Mineralocorticoid receptor antagonists (MRAs) block aldosterone binding to the MR in the kidney, heart, and vasculature [4]. Steroidal MRAs, such as spironolactone and eplerenone, are well established therapeutic options in heart failure, hypertension, cirrhosis, and primary aldosteronism [5, 6]. Spironolactone is a potent but nonselective MRA, associated with side effects, such as gynecomastia and menstrual irregularities. Eplerenone is more selective but less potent and has a shorter duration of action [7, 8]. Both agents increase the risk of hyperkalemia, particularly in patients with advanced kidney disease [9]. Nonsteroidal MRAs, such as finerenone and esaxerenone, offer higher selectivity, balanced distribution between the heart and the kidney, and lower risk of hyperkalemia and endocrine side effects [10–12]. In contrast to traditional MRAs that fully block receptor activation, MR modulators, such as balcinrenone and apararenone, induce distinct MR conformational changes that selectively inhibit the pro-inflammatory and pro-fibrotic gene transcription, while preserving physiological electrolyte regulation [13–15].

MRAs have now been studied in large randomized clinical trials for various indications and in different populations, including patients with end-stage renal disease. This review article will summarize the biological rationale, evolving clinical evidence, and future directions for MR blockade in patients with kidney failure requiring dialysis.

Although this review was conducted in a narrative format, a preliminary structured literature search was performed in PubMed/MEDLINE and EMBASE to identify relevant studies. Searches included combinations of the terms “mineralocorticoid receptor antagonists,” “spironolactone,” “eplerenone,” “finerenone,” “dialysis,” “hemodialysis,” “peritoneal dialysis,” “randomized controlled trial,” “cardiovascular outcomes,” “mortality,” “hyperkalemia,” and “safety.” We focused primarily on randomized controlled trials evaluating steroidal and nonsteroidal MRAs in maintenance hemodialysis and peritoneal dialysis (PD) populations. Both mechanistic endpoints (e.g., blood pressure [BP], left ventricular mass, vascular calcification, biomarkers) and hard clinical outcomes (e.g., cardiovascular events, heart-failure hospitalization, all-cause, and cardiovascular mortality) were considered. Safety outcomes, particularly hyperkalemia and endocrine adverse effects, were systematically reviewed. Relevant meta-analyses of randomized controlled trials were also examined to contextualize trial findings.

Cardiovascular Disease Burden in Patients on Dialysis

Patients with kidney failure undergoing hemodialysis or peritoneal dialysis face an alarmingly high mortality – approximately 10–15% at 1 year, 29% at 2 years, and over 50% at 5 years [16–18]. Cardiovascular mortality is 2–10 times higher than in the general population and up to 100-fold greater among adults under 45 years [19]. Cardiovascular disease remains the leading cause of death, accounting for ∼42% of cases, with sudden cardiac death contributing up to 25% [20, 21]. In a large Ontario cohort of 33,500 incident dialysis patients, the 5-year survival was less than 50% and was found to be worse than that of most solid-organ cancers, including breast, prostate, and colorectal cancer [22].

Aldosterone levels rise early in chronic kidney disease (CKD) and remain persistently elevated despite fluid overload [23]. In a post hoc analysis of the German 4D trial, patients on hemodialysis with aldosterone >200 pg/mL had a markedly higher risk of sudden cardiac death (hazard ratio [HR] 1.69, 95% confidence interval [CI]: 1.06–2.69), underscoring the potential pathogenic role of MR overactivation in this population [24]. Therefore, aldosterone blockade has been evaluated as a potential therapeutic intervention to lower cardiovascular risk in this population.

Pharmacokinetics and Pharmacodynamics of MRAs

Spironolactone is a first-generation nonselective steroidal MRA and a prodrug rapidly metabolized to active sulfur-containing metabolites – primarily canrenone, 7α-thiomethyl-spironolactone, and 6β-hydroxy-thiomethyl-spironolactone [25]. Oral bioavailability is 80–90%, with a parent half-life of ∼1.5 h and metabolite half-lives of 16–20 h, driving most of its pharmacological activity. Both parent drug and its metabolites are >90% protein-bound, undergo hepatic and biliary elimination, and are minimally excreted by the kidneys (<5%), rendering them nondialyzable [26, 27]. In a study of 14 hemodialysis patients receiving 25 mg of spironolactone thrice weekly, the mean plasma canrenone level was 13 ± 5.3 ng/mL, indicating systemic exposure without significant accumulation [28].

Eplerenone, a second-generation selective steroidal MRA, incorporates structural modifications that enhance MR selectivity, reduce endocrine side effects, but also lower potency (25 mg of spironolactone corresponds to 50 mg of eplerenone) [7, 29, 30]. Eplerenone has dose-proportional kinetics, a 4–6 h half-life, ∼50% protein binding, and is metabolized by CYP3A4 to inactive metabolites, with <5% excreted unchanged [31, 32].

Finerenone, a third-generation nonsteroidal MRA, exhibits high MR selectivity, balanced cardiac-renal distribution, and no off-target hormonal effects [11]. Oral bioavailability is ∼43%, with hepatic CYP3A4/5 metabolism to inactive metabolites, a 2–3 h half-life, and <1% renal excretion [33, 34]. No formal pharmacokinetic or dialytic clearance studies exist in maintenance dialysis populations, but high protein binding and nonrenal elimination suggest negligible dialyzability.

Established Benefits of MRAs in Non-Dialysis Populations

MRAs in Heart Failure with Reduced and Preserved Ejection Fraction

Steroidal MRAs have robust evidence in heart failure with reduced ejection fraction (HFrEF), earning a class I, Level A recommendation, whereas their role in heart failure with preserved ejection fraction (HFpEF) remains less well established (class IIb) [35]. Spironolactone first demonstrated a survival benefit in severe HFrEF in the RALES trial [36]. Eplerenone also showed consistent benefit across the heart failure spectrum in the subsequent EPHESUS and EMPHASIS-HF clinical trials in patients with post myocardial infarction heart failure or compensated HFrEF (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000551881) [37, 38].

The TOPCAT trial evaluated spironolactone in HFpEF but did not show a significant reduction in the composite outcome of cardiovascular death, aborted cardiac arrest, or heart failure hospitalization (HR 0.89, 95% CI: 0.77–1.04) [39]. Notably, marked regional discrepancies emerged: patients enrolled in Russia and Georgia had unusually low event rates and minimal changes in biochemical parameters, suggesting poor adherence to the study drug and inconsistent application of the clinical trial inclusion criteria. Participants from the Americas experienced a significant benefit (HR 0.82, 95% CI: 0.69–0.98), whereas no benefit was seen in the Eastern European cohort (HR 1.10, 95% CI: 0.79–1.51) [40]. More recently, the FINEARTS-HF trial enrolled 6,001 patients with heart failure with mildly reduced or preserved ejection fraction and demonstrated that finerenone significantly reduced the composite of cardiovascular death and total heart failure events (RR 0.84, 95% CI: 0.74–0.95) [41]. These data thus support a class IIa recommendation for finerenone in HFpEF, compared with class IIb for steroidal MRAs (online suppl. Table 1) [42].

MRAs in Non-Dialysis Dependent CKD

While earlier subgroup analyses or surrogate-endpoint studies with steroidal MRAs hinted at a possible cardiovascular and renal benefit in CKD, the larger pragmatic trial BARACK-D failed to demonstrate any benefit in stage 3 CKD [27, 43, 44]. The BARACK-D trial assessed low-dose spironolactone (25 mg daily) in 1,434 patients with mean estimated glomerular filtration rate (eGFR) of 43 mL/min/1.73 m². Over 3 years, the primary composite cardiovascular outcome was similar between the spironolactone and usual-care groups (16.7% vs. 16.0%; HR 1.05, 95% CI: 0.81–1.37). Tolerability was limited, with nearly two-thirds discontinuing therapy within 6 months, predominantly due to a drop in eGFR or hyperkalemia. No significant changes were observed in CKD progression, albuminuria, or BP, suggesting that spironolactone should be reserved for CKD patients with concurrent heart failure indications [45].

In contrast, the nonsteroidal MRA finerenone has demonstrated consistent cardiorenal protection in diabetic CKD. In the FIGARO-DKD trial (n = 7,437), finerenone reduced the composite endpoint of cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, or heart failure hospitalization by 13% (HR 0.87, 95% CI: 0.76–0.98) [46]. The FIDELIO-DKD trial (n = 5,734) showed that finerenone reduced the risk of CKD progression (HR 0.82, 95% CI: 0.73–0.93) and cardiovascular events (HR 0.86, 95% CI: 0.75–0.99) [47]. In the pooled FIDELITY analysis (n = 13,026), finerenone demonstrated robust and consistent efficacy across eGFR and albuminuria subgroups, lowering the risk of kidney (HR 0.77, 95% CI: 0.67–0.88) and cardiovascular (HR 0.86, 95% CI: 0.78–0.95) events [48]. Finerenone is now incorporated into the guideline-directed therapy for diabetic kidney disease, in addition to renin-angiotensin system (RAAS) inhibitors, sodium glucose co-transporter-2 (SGLT-2) inhibitors, and glucagon-like peptide-1 receptor agonists, based on consistent cardiorenal benefit demonstrated in large, randomized trials [49]. Ongoing studies, including FIND-CKD (NCT05047263), are evaluating whether nonsteroidal MRAs provide similar benefits in patients with nondiabetic CKD.

Importantly, the population enrolled in BARACK-D differed substantially from those included in contemporary nonsteroidal MRA trials (online suppl. Table 2). Participants in BARACK-D were older, predominantly non-diabetic, largely normoalbuminuric, and had stage 3b CKD with comparatively lower baseline cardiovascular risk. In contrast, the FIDELIO-DKD and FIGARO-DKD trials enrolled patients with diabetic CKD and significant albuminuria – populations at higher cardio-renal risk and therefore more likely to derive benefit from MR blockade. These differences limit direct comparisons between steroidal and nonsteroidal MRA trials across heterogeneous CKD populations [50]. Importantly, all pivotal MRA trials excluded participants with eGFR <25–30 mL/min/1.73 m² and those receiving maintenance dialysis, leaving the efficacy and safety of MRAs in this population uncertain.

Phase 2 Studies Evaluating Safety and Tolerability of MRAs in Dialysis

Early Physiologic Studies of MRAs in Dialysis

Sugarman and Brown (1988) conducted the earliest physiological crossover study evaluating MR blockade in hemodialysis. Seven anephric patients underwent three 72-h interdialytic interventions: control (no therapy), deoxycorticosterone acetate (DOCA; 10 mg/day intramuscularly), and spironolactone (300 mg/day orally). Acute and dietary potassium loads were administered, with serial measurements of serum, salivary, and fecal electrolytes. Deoxycorticosterone acetate attenuated the post-load rise in serum potassium by expanding the apparent volume of potassium distribution (55% vs. 34% in control), whereas spironolactone amplified the potassium increase and modestly reduced the distribution volume (35%). Neither intervention affected fecal or salivary potassium excretion. This study showed that aldosterone promotes extrarenal potassium buffering via cellular uptake, while MR blockade blunts this response – providing the first in vivo evidence of aldosterone’s role in potassium regulation in dialysis patients [51].

Non-Randomized Interventional Studies of MRAs in Dialysis

In early pilot studies, low-dose spironolactone appeared generally safe in maintenance hemodialysis. Hussain et al. [52] (2003) treated 15 patients with 25 mg daily for 4 weeks; mean potassium remained stable, though two withdrew for severe hyperkalemia (>7.5 mmol/L), one after a missed dialysis session (Table 1). A contemporaneous Swiss study using 12.5–25 mg thrice weekly in 14 patients reported no hyperkalemia [28]. Similarly, Matsumoto et al. [53] (2008) followed 61 Japanese oligoanuric patients treated with spironolactone 25 mg daily for 6 months; mean potassium levels slightly increased (from 5.0 ± 0.7 to 5.2 ± 0.7 mmol/L), with no value exceeding 6.8 mmol/L, supporting the relative safety of low-dose spironolactone under close monitoring.

Table 1.

Phase 2 studies assessing safety and tolerability of MRAs in dialysis

Author	Hussain et al. [52]	Saudan et al. [28]	Matsumoto et al. [53]	Yongsiri et al. [54]	PHASE [55]	SPin-D [56]
Year of publication	2003	2003	2008	2013	2015	2019
Study design	Single-arm, interventional study	Non-randomized, open-label interventional study	Single-arm, interventional study	Double-blind, placebo-controlled, randomized cross-over trial	Double-blind, placebo-controlled, parallel-group, randomized controlled trial	Double-blind, placebo-controlled, parallel-group, dose-finding randomized controlled trial
Patient population	HD >4 months	HD	Oligoanuric HD patients on dialysis >2 years	PD >3 months with hypokalemia or requiring potassium supplements	HD ≥3 months	HD ≥6 months or 3–6 months with stable dry weight without recent hospitalizations
MRA	Spironolactone 25 mg once daily	Spironolactone 25 mg thrice weekly	Spironolactone 25 mg once daily	Spironolactone 25 mg once daily	Eplerenone 50 mg daily	Spironolactone 12.5 mg, 25 mg, and 50 mg once daily
Total Patients	15	35	61	20	154	129
Duration of follow-up	4 weeks	4 weeks	6 months	4 weeks	13 weeks	40 weeks
Primary outcome	Change in predialysis serum potassium	Change in predialysis serum potassium	Change in predialysis serum potassium	Change in serum potassium	Permanent discontinuation due to hyperkalemia or hypotension	Hyperkalemia (>6.5 mmol/L)
						Hypotension requiring ED visit or hospitalization
Age (years)	54 (33–87)a	54±14b	67±14	52±12	62±15b	56±12
		59±17c			63±14c
Males	10 (67)	–	18 (36)	8 (40)	96 (62)	85 (66)
Diabetes	–	7 (20)	13 (26)	13 (65)	68 (44)	66 (51)
CAD	–	–	–	–	29 (19)	28 (22)
Heart failure	–	–	–	–	14 (9)	21 (16)
Dialysis vintage (years)	–	5.6 (0.7–15.3)b	8.4±6.9	2.0±0.9	3 (2–7)	3.4 (1.9–6.1)
		2 (0.9–8.3)c
RAAS inhibitor use	6 (40)	19 (54)	33 (66)	0	65 (42)	39 (30)
Safety outcomes (%)
Hyperkalemiad
MRA arm	2 (13)	0	7 (14)	0	9 (12)	16 (21)
Placebo arm	0	0	–	0	2 (3)	9 (18)
Gynecomastia	​	–	​	​	–	​
MRA arm	1 (7)	​	3 (5)	0	​	3 (4)
Placebo arm	–	​	​	0	​	2 (4)
Hypotension
MRA arm	–	–	–	0	16 (21)	14 (18)
Placebo arm	–	–	–	0	14 (18)	8 (16)
Discontinuation due to AE
MRA arm	2 (13)	0	6 (10)	0	3 (4)	8 (10)
Placebo arm	–	0	–	0	2 (3)	11 (22)
Discontinuation due to any cause
MRA arm	2 (13)	0	11 (18)	0	14 (19)	13 (17)
Placebo arm	–	0	–	0	9 (13)	14 (28)
Trial Limitations	Open-label trial	Non-randomized open-label trial	Open-label trial	​Potassium supplements were adjusted to keep serum potassium in target range	Short follow-up	Modest sample size
	No control arm		No control arm		Residual renal function not measured	Relatively healthier population

Continuous variables are reported as mean ± SD or median (IQR) as appropriate. Categorical variables are presented as proportions (%).

AE, adverse event; CAD, coronary artery disease; ED, emergency department; HD, Hemodialysis: IQR, interquartile range; MRA, mineralocorticoid receptor antagonist; PD, peritoneal dialysis; PHASE, Pilot trial of hemodialysis patients undergoing aldosterone antagonism with eplerenone; RAAS, renin-angiotensin-aldosterone system inhibitors; SD, standard deviation; SPin-D, Spironolactone in Dialysis.

^aMean (range).

^bMRA arm.

^cPlacebo arm.

^dHyperkalemia defined as serum potassium >6.5 mmol/L.

Phase 2 Randomized Controlled Trials Evaluating Safety and Tolerability of MRAs in Dialysis

The Pilot Trial of Hemodialysis Patients Undergoing Aldosterone Antagonism with Eplerenone (PHASE, 2015) was a multicenter, double-blind, placebo-controlled trial conducted across five Canadian centers (Table 1; Fig. 1). Adults on maintenance hemodialysis for >3 months were randomized 1:1 to eplerenone 50 mg daily (n = 77) or placebo (n = 77) for 13 weeks. The primary endpoint-permanent study drug discontinuation due to hyperkalemia or hypotension–occurred in 3 (4%) patients receiving eplerenone and 2 (2.8%) receiving placebo (absolute difference 1.2%; 95% CI: -4.7 to 7.1), meeting the non-inferiority criterion. Adherence (≥80% of doses taken) was comparable between the two groups (79.9% vs. 76.6%). Eplerenone increased the incidence of hyperkalemia >6.5 mmol/L (11.7% vs. 2.6%; relative risk (RR) 4.5, 95% CI: 1.0–20.2), with four severe cases (>7.0 mmol/L) versus none in the placebo group, whereas hypotension rates were similar (20.8% vs. 18.2%). The authors concluded that, although eplerenone modestly increased hyperkalemia risk, it did not lead to permanent discontinuation, supporting the option of MRA therapy under close monitoring. Key limitations, however, included a short follow-up period and lack of residual renal function assessment, which may have influenced potassium balance [55].

Fig. 1.

Overview of randomized controlled trials evaluating mineralocorticoid receptor antagonists in dialysis. CIMT, carotid intima-media thickness; CV, cardiovascular; LVMi, left ventricular mass index; MRA, mineralocorticoid receptor antagonists.

The Spironolactone in Dialysis (SPin-D, 2019) trial, a double-blind, placebo-controlled, dose-finding study by the Hemodialysis Novel Therapies Consortium, randomized 129 maintenance hemodialysis patients in a 2:1:1:1 ratio to placebo (n = 51) or spironolactone 12.5 mg (n = 27), 25 mg (n = 26), or 50 mg (n = 25) daily. Participants were evaluated weekly during a 6-week dose-escalation phase and monthly thereafter for a total of 40 weeks. Dose reductions were allowed before discontinuation, and the study drug was stopped if potassium exceeded 7.0 mmol/L. The primary safety outcomes were hyperkalemia (>6.5 mmol/L) and hypotension requiring emergency department visit or hospitalization. The cohort was predominantly African American (71%), male (66%), and relatively young (mean age 56 ± 12 years). Hyperkalemia occurred at similar rates between placebo and all spironolactone groups combined (0.50 vs. 0.49 events per patient-year) but increased significantly at the 50 mg dose (0.89 events per patient-year). Hypotension showed a similar dose-dependent pattern. Although exploratory analyses suggested minor improvements in diastolic function (mitral annular E′ velocity) and coronary flow reserve with the 50 mg dose, these were not statistically significant. The trial was methodologically robust, with standardized monitoring and centralized event adjudication, but limited by its modest sample size and relatively healthy population with near-normal baseline cardiac function [56].

Hypokalemia is a more frequent concern in PD [57, 58]. A meta-analysis of 24 studies (>60,000 PD patients) reported a 17.7% prevalence of hypokalemia (<3.5 mmol/L), associated with increased risk of all-cause mortality (HR 1.49, 95% CI: 1.18–1.89), cardiovascular mortality (HR, 1.50, 95% CI, 1.19–1.88) and peritonitis (HR 1.42, 95% CI: 1.17–1.73) [59]. Thus, spironolactone may be repurposed to correct or prevent hypokalemia in PD. In a randomized, double-blind, cross-over trial by Yongsiri et al., 24 PD patients with a history of hypokalemia or requiring chronic potassium supplementation received spironolactone 25 mg daily or placebo for 4 weeks, separated by a 2-week washout period. Serum potassium did not differ significantly between the treatment periods and only one moderate hyperkalemia episode (5.6 mmol/L) occurred. Although adjustment of potassium supplements during the study may have confounded the results, the trial nevertheless supports the short-term safety of spironolactone in PD [54].

Phase 2 Studies Evaluating the Efficacy of MRAs in Dialysis

Effect on BP Regulation

MRAs lower BP by blocking aldosterone-mediated sodium retention and vascular remodeling, thereby reducing volume overload and arterial stiffness [60, 61]. They are particularly effective in resistant hypertension; however, their antihypertensive efficacy and safety in dialysis remain underexplored [62, 63].

In a crossover randomized controlled trial of eight oligoanuric hemodialysis patients (Gross et al. [64] 2005), spironolactone 50 mg twice daily reduced predialysis systolic BP without affecting potassium levels or dry weight, suggesting a hemodynamic effect independent of changes in volume status (Table 2). In a parallel-group randomized controlled trial of 76 hemodialysis and peritoneal patients with refractory hypertension (Ni et al. [65] 2014), spironolactone (25–50 mg daily) reduced 24-h ambulatory systolic and diastolic BP over 12 weeks compared with placebo (Table 2; Fig. 1). A small crossover trial in patients with recurrent intradialytic hypertension also demonstrated fewer hypertensive episodes with spironolactone 50 mg [66]. Collectively, these mechanistic studies indicate that spironolactone may lower both ambulatory and intradialytic BP in dialysis without increasing hyperkalemia risk, supporting its potential as a second-line antihypertensive agent under careful monitoring.

Table 2.

Phase 2 randomized controlled trials assessing efficacy of MRA in dialysis

Author	Gross et al. [64]	Taheri et al. [67]	Vukusich et al. [68]	Ni et al. [65]	Nagoya Spiro Study [69]	Feniman-De-Stefano et al. [70]	MiREnDa [71]	Gueiros et al. [72]	SV-CAPD trial [73]
Year of publication	2005	2009	2010	2014	2014	2015	2018	2019	2022
Study design	Double-blind, placebo-controlled, cross-over, randomized controlled trial	Double-blind, placebo-controlled, parallel-group randomized controlled trial	Double-blind, placebo-controlled, parallel-group randomized controlled trial	Double-blind, placebo-controlled, parallel-group randomized controlled trial	Open-label, parallel-group, randomized controlled trial	Double-blind, placebo-controlled, parallel-group randomized controlled trial	Double-blind, placebo-controlled, parallel-group, randomized controlled trial	Open-label, parallel-group, randomized controlled trial	Double-blind, placebo-controlled, parallel-group randomized controlled trial
Patient population and setting	Oligoanuric HD patients on dialysis >3 months	HD patients with NYHA-FC class III or IV symptoms receiving RAASi	HD >18 months	HD or PD >3 months with refractory hypertension	PD patients with NYHA-FC class I or II symptoms receiving RAASi >3 months	HD patients with LVMi >51 g/m2	Maintenance HD	PD >6 months with coronary calcification score >30	PD
MRA	Spironolactone 50 mg twice daily	Spironolactone 25 mg thrice weekly after dialysis	Spironolactone 50 mg thrice weekly after dialysis	Spironolactone 25–50 mg daily	Spironolactone 25 mg daily/eplerenone 50 mg daily	Spironolactone 12.5–25 mg daily	Spironolactone 50 mg once daily	Spironolactone 25 mg once daily	Spironolactone 25 mg once daily
Total patients	8	16	66	76	158	17	97	33	34
Follow-up duration	2 weeks	6 months	24 months	12 weeks	2 years	6 months	40 weeks	12 months	6 months
Primary outcome	Change in predialysis systolic BP	Change in LVEF and LVMi	Change in carotid intima-media thickness	Change in 24-h ambulatory BP at 12 weeks from baseline	Change in LVMi and LVEF	Change in LVMi	Change in LVMi by CMR	Change in coronary calcification score by CT	Change in coronary calcification score by CT
Age (years)	53±10	60±7a	50±5a	56±12a	57±12a	52±19a	60±13	70±9a	46±16a
		57±9b	56±4b	55±14b	56±14b	56±11b		61±9b	55±12b
Males	3 (38)	11 (69)	34 (64)	46 (61)	104 (66)	9 (53)	75 (77)	8 (50)	22 (65)
Diabetes	3 (38)	10 (63)	0	–	62 (39)	9 (53)	32 (33)	8 (50)	23 (68)
CAD	–	9 (56)	–	–	–	0	36 (37)	2 (13)	9 (27)
Heart failure	0	16 (100)	–	–	0	0	4 (4)	–	–
Dialysis vintage (years)	1.7±1.2	–	7.9±1.3a	4.7±0.8a	0.5 (0.2–1.8)a	2.2±1.8a	3.5 (1.3–6.2)	0.8 (0.5–2)a	–
			8.8±1.2b	4.6±1.2b	0.6 (0.2–1.9)b	4.0±6.6b		4.4 (2.8–6)b
RAASi use	0	16 (100)	0	41 (54)	158 (100)	3 (18)	55 (57)	8 (50)	17 (50)
Efficacy outcomes	Spironolactone 50 mg twice daily given for 2 weeks reduced predialysis systolic BP	Spironolactone 25 mg thrice weekly reduced LVMi and improved LVEF over 6 months	Spironolactone 50 mg thrice weekly slowed or even reversed CIMT progression over 24 months	Spironolactone 25–50 mg given for 12 weeks reduced 24-h ambulatory systolic and diastolic BP	Spironolactone 25 mg daily reduced LVMi and improved LVEF over 24 months	Spironolactone 12.5–25 mg daily reduced LVMi over 6 months	Spironolactone 50 mg daily did not improve LVMi, LVEF, 6-min walk-test distance or 24-h ambulatory BP after 40 weeks of treatment	Spironolactone 25 mg daily did not attenuate the progression of coronary artery calcium score after 12 months of treatment	Spironolactone 25 mg once daily reduced the progression of coronary artery calcium score after 6 months of treatment
Safety outcomes	No risk of hyperkalemia reported	No risk of hyperkalemia reported	No significant BP change or hyperkalemia observed	No difference in serum potassium vs. placebo	No significant hyperkalemia reported	No significant change in BP or serum potassium	Moderate hyperkalemia with spironolactone treatment	Well-tolerated, with no serious adverse events or hyperkalemia reported	No severe hyperkalemia or treatment discontinuation due to adverse events
					Gynecomastia occurred in 14% of patients
Limitations	Pilot study with small number of patients	Very small sample size, unclear randomization and allocation concealment, operator-dependent echocardiographic assessment	Patients with diabetes were excluded from the study	Small sample size–not powered for clinical outcomes, inadequate description of randomization and allocation concealment, unclear adjustments to concomitant antihypertensive therapy and ultrafiltration	Open-label design, unclear randomization and allocation concealment. no standardized potassium-monitoring protocol, operator-dependent echocardiographic assessment, limited generalizability	Very small sample size, short follow-up of 6 months, placebo group older with longer dialysis vintage, randomization and allocation concealment not described, operator-dependent echocardiographic assessment	Short follow-up	Very small sample size, high dropout rate, open-label design, imbalance in baseline dialysis vintage	Small sample size, short 6-month follow-up, baseline imbalances, better phosphate control in the spironolactone arm may have contributed to improved calcification scores

Continuous variables are reported as mean ± SD or median (IQR) as appropriate. Categorical variables are presented as proportions (%).

BP, blood pressure; CAD, coronary artery disease; CI, confidence interval; CIMT, carotid intima-media thickness; CMR, cardiac magnetic resonance imaging; CT, computed tomography; HD, hemodialysis; LVEF, left ventricular ejection fraction; LVMi, left ventricular mass index (indexed to height); MiREnDa, Mineralocorticoid Receptor Antagonists in End-Stage Renal Disease; MRA, mineralocorticoid receptor antagonist; NYHA-FC, New York Heart Association- Functional Class; PD, peritoneal dialysis; RAASi, renin-angiotensin-aldosterone system inhibitors; SD, standard deviation; SV-CAPD, Spironolactone to prevent the progression of Vascular Calcification Among Peritoneal Dialysis patients; 24-h, 24-h.

^aMRA arm.

^bPlacebo arm.

Effect on Vascular Calcification and Remodeling

Vascular calcification is a major complication in dialysis, manifesting as medial (arteriosclerotic) and intimal (atherosclerotic) calcium-phosphate deposition that contributes to arterial stiffness, systolic hypertension, left ventricular hypertrophy, and cardiovascular mortality [74, 75]. It is now recognized as an active, cell-mediated process rather than a passive physicochemical phenomenon [76]. Under uremic conditions, vascular smooth muscle cells undergo osteogenic trans-differentiation driven by hyperphosphatemia, oxidative stress, uremic toxins, and inflammation [77]. This transformation involves upregulation of osteogenic transcription factors (Runx2, BMP-2, osteocalcin) and depletion of endogenous calcification inhibitors such as fetuin-A, matrix Gla protein, and pyrophosphate [78].

Recent evidence implicates aldosterone and MR signaling in “vascular osteoinduction” [79–81]. Aldosterone promotes vascular calcification by upregulating the type III sodium-dependent phosphate transporter Pit-1 (SLC20A1), enhancing phosphate uptake and activating the MR-Pit-1-Runx2 signaling pathway [82, 83]. Experimental uremic models show that MRAs attenuate calcification by downregulating Runx2, Pit-1, and osteopontin, and by reducing oxidative stress and apoptosis without altering BP [84–86]. In adenine-induced CKD rats, Tatsumoto et al. [87] (2015) demonstrated that spironolactone (50–100 mg/kg/day for 8 weeks) dose-dependently reduced aortic calcium and phosphate deposition and suppressed MR-mediated signaling (Runx2, Pit-1, Sgk-1). Human data, however, remain limited [88, 89]. In a pilot study by Nitta et al. [90], five hemodialysis patients with aortic calcification received spironolactone 50 mg daily for a mean of 3.1 years. The aortic calcification index declined significantly (27.0 ± 12.8% to 18.6 ± 11.8%; p = 0.003), accompanied by reductions in plasma osteopontin, without changes in parathyroid hormone or calcium–phosphate product, suggesting suppression of osteogenic activity and vascular mineralization.

Two randomized controlled trials in PD evaluated coronary artery calcification using computed tomography (Table 2). In the Thailand pilot trial (n = 34), spironolactone 25 mg daily was compared with placebo for 6 months; although calcification scores numerically decreased in the treatment arm, between-group differences were not statistically significant [73]. In a Brazilian double-blind randomized controlled trial (n = 33), spironolactone 25 mg daily for 12 months similarly did not significantly alter coronary calcification progression, though treatment was well-tolerated [72].

Aldosterone promotes vascular remodeling through inflammation, oxidative stress, and fibrosis, and carotid intima-media thickness serves as a surrogate marker of such structural vascular changes [91–93]. In a double-blind randomized controlled trial from Chile, 66 nondiabetic hemodialysis patients were randomized to spironolactone 50 mg thrice weekly or placebo for 24 months. Carotid intima-media thickness, assessed by high-resolution ultrasonography, progressed in the placebo group but was attenuated or regressed in several segments with spironolactone [68] (Fig. 1; Table 2).

Effect on Cardiac Hypertrophy and Remodeling

Left ventricular hypertrophy is highly prevalent among dialysis patients due to chronic pressure and volume overload, anemia, and arterial stiffness [94, 95]. It is a powerful predictor of cardiovascular morbidity and mortality, predisposing to heart failure, arrhythmias, and sudden cardiac death [96–98].

In a double-blind placebo-controlled trial, Taheri et al. [67] randomized 16 hemodialysis patients with HFrEF to spironolactone 25 mg thrice weekly or placebo for 6 months and observed a modest reduction in left ventricular mass index (LVMi) with treatment (Table 2). Another small trial allocated 17 hemodialysis patients with LVMi >51 g/m² to spironolactone 12.5–25 mg daily or placebo for 6 months. Spironolactone produced a modest but significant reduction in LVMi without affecting BP or potassium levels [70].

The Nagoya Spiro Study (2014), a multicenter open-label randomized controlled trial across 12 Japanese hospitals, enrolled 158 PD patients with mild heart failure on background RAAS inhibition (Table 2; Fig. 1). Participants received spironolactone 25 mg daily (later replaced by eplerenone 50 mg daily) or standard of care for 2 years. MRA therapy improved LVMi at 6, 18, and 24 months. In men with baseline LVMi >50 g/m², LVMi declined progressively with treatment, while remaining stable in controls. Whereas, in men with baseline LVMi <50 g/m², LVMi did not change during the follow-up with treatment, but increased in the control group. These effects were absent in women. Limitations included its open-label design, uncertain allocation concealment, operator-dependent echocardiography, and the ethnically homogeneous cohort [69].

The Mineralocorticoid Receptor Antagonists in End-Stage Renal Disease (MiREnDa) trial (2018) was designed to address these methodological limitations. After a 2–4-week placebo run-in and baseline cardiovascular phenotyping, 97 hemodialysis patients were randomized 1:1 to spironolactone (n = 50) or placebo (n = 47) for 40 weeks. Randomization was stratified by left ventricular hypertrophy status and RAAS inhibitor use. The primary endpoint was the change in LVMi from baseline to 40 weeks measured by cardiac magnetic resonance imaging. LVMi change did not differ between the two groups (−2.86 ± 11.9 vs. +0.41 ± 10.8 g/m²; p = 0.34). Similarly, there was no difference in secondary outcomes, including 24-h ambulatory systolic BP, diastolic BP, and 6-min walk distance [71].

MiREnDa was a well conducted trial with rigorous and blinded endpoint adjudication [99]. However, its 40-week duration may have been insufficient to detect structural cardiac regression. Compared with MiREnDa, participants in the Nagoya Spiro Study were younger, had shorter dialysis vintage, preserved residual renal function, milder heart failure, and less baseline hypertrophy – factors that may explain their greater responsiveness to MR blockade [69]. Importantly, these studies were small, largely mechanistic trials evaluating surrogate endpoints and were not powered to assess hard clinical outcomes.

Phase 3 Randomized Controlled Trials of MRAs in Dialysis

Early Phase 3 Randomized Controlled Trials Conducted in East Asia

The Dialysis Outcomes Heart Failure Aldactone Study (DOHAS trial) was a single-center (5 dialysis units), open-label, randomized controlled trial conducted in oligoanuric, prevalent hemodialysis patients in Japan between 2008 and 2011 (Table 3; Fig. 1). Participants were randomized to receive spironolactone 25 mg once daily (n = 157) or standard of care (n = 152) for 3 years. The primary endpoint was a composite of death or hospitalization due to cardiovascular or cerebrovascular events, adjudicated by an independent blinded committee. A total of 309 patients were enrolled, though recruitment fell short of the planned 600 participants. Treatment discontinuation occurred in 29% of the spironolactone group – mainly due to gynecomastia or breast tenderness – and in 16% of the control group, largely from transfer to other centers. Baseline characteristics were imbalanced: the control group had a longer dialysis vintage, higher cardiothoracic ratio, and lower serum albumin. Over 3 years, the composite primary outcome occurred in 9 patients in the spironolactone group versus 23 in the control group (adjusted HR 0.38, 95% CI: 0.17–0.83). All-cause mortality was also lower in the spironolactone arm (10 vs. 30 deaths; adjusted HR 0.33, 95% CI: 0.16–0.69). Serum potassium remained stable, with only three discontinuations due to hyperkalemia in the spironolactone arm [100].

Table 3.

Phase 3 randomized controlled trials of mineralocorticoid antagonists in dialysis

Author	DOHAS [100]	Lin et al. [101]	ALCHEMIST [102]	ACHIEVE [103]
Year of publication	2014	2016	2025	2025
Study design	Single-center, parallel-group, open-label, randomized controlled trial	Multicenter, double-blind, parallel-group, placebo-controlled, randomized controlled trial	Multicenter, double-blind, placebo-controlled, parallel-group, randomized controlled trial	Multicenter, double-blind, placebo-controlled, parallel-group, randomized controlled trial
Patient population and setting	Oligoanuric HD patients on dialysis >2 years	HD or PD >3 months	HD >45 days with additional CV disease or risk factors	HD or PD >3 months
MRA	Spironolactone 25 mg daily	Spironolactone 25 mg daily	Spironolactone 25 mg daily	Spironolactone 25 mg daily
Total patients	309	253	644	2538
Duration of follow-up	3 years	1 year	2.7 (1.4–4.0)	1.8 (0.9–3.4)
Primary outcome	Death or hospitalization for CV or cerebrovascular eventsa	Composite of death from CV or cerebrovascular events, sudden cardiac death or aborted cardiac arresta	Major adverse cardiovascular eventsb	Composite of cardiovascular mortality and hospitalization for heart failure
Age (years)	67±12c	70±11c	71 (63–78)	61±11c
	68±11d	71±8d	​	62±11d
Males	203 (66)	153 (61)	444 (69)	1607 (63)
Diabetes	–	–	446 (69)	1407 (55)
CAD	–	–	274 (43)	304 (12)
Heart failure	–	–	74 (12)	290 (11)
Atrial fibrillation	33 (11)	–	44 (22)	228 (9)
Dialysis vintage (years)	8.3±6.9c	3.5±1.5c	1.7 (0.7–4.5)	2 (1–4.4)c
	10.6±7.9d	3.6±1.5d	​	2 (0.9–4.5)d
RAAS inhibitor use	129 (42)	214 (85)	164 (26)	903 (36)
Efficacy outcomes, HR (95% CI)
Composite CV outcome	0.38 (0.17–0.83)	0.42 (0.26–0.78)	1.00 (0.73–1.36)	0.92 (078–1.09)
Cardiovascular mortality	0.43 (0.17–1.11)	0.33 (0.13–0.85)	1.26 (0.86–1.85)	0.89 (0.74–1.08)
All-cause mortality	0.33 (0.16–0.69)	0.52 (0.29–0.94)	1.01 (0.79–1.30)	0.95 (0.83–1.09)
Hospitalization for heart failure	–	–	0.41 (0.17–1.00)	0.97 (0.72–1.30)
Safety outcomes, HR (95% CI) or number (%)
Hyperkalemia	3 (2%) in MRA arme	3 (2%) in MRA armf	1.12 (0.88–1.43)f	1.54 (1.07–2.22)e
Gynecomastia
MRA arm	16 (10)	4 (3)	14 (4)	19 (2)
Control arm	0	0	4 (1)	6 (1)
Hypotension	–	–	–	​
MRA arm	​	​	​	7 (1)
Control arm	​	​	​	10 (1)
Discontinuation due to AEs
MRA arm	23 (14.6)	5 (4)	75 (23)	294 (23.3)
Placebo arm	0	0	74 (23)	241 (18.9)
Limitations	Open-label design, randomization details not reported, baseline characteristics imbalanced, low event rate, adverse events reported only in spironolactone arm, limited generalizability	Low event rate, short follow-up duration, outcomes not adjudicated by blinded committee, incomplete potassium monitoring and adverse event reporting, limited generalizability	Stopped prematurely due to lack of funding, high dropout rate (23%)	Stopped early due to futility

Continuous variables are reported as mean ± SD or median (IQR) as appropriate. Categorical variables are presented as proportions (%).

ACHIEVE, aldosterone blockade for health improvement evaluation in end-stage renal disease; AE, adverse event; ALCHEMIST, aldosterone antagonist chronic hemodialysis interventional survival trial; BP, blood pressure; CAD, coronary artery disease; CI, confidence interval; CV, cardiovascular; DOHAS, dialysis outcomes heart failure aldactone study; HD, hemodialysis; HR, hazard ratio; IQR, interquartile range; MRA, mineralocorticoid receptor antagonist; PD, peritoneal dialysis; RAAS, renin-angiotensin-aldosterone system; SD, standard deviation.

^aCardiovascular or cerebrovascular events were defined as new occurrence or exacerbation of heart failure unresponsive to ultrafiltration; malignant ventricular arrhythmias (ventricular fibrillation or sustained ventricular tachycardia); new or recurrent acute myocardial infarction; new occurrence or exacerbation of angina pectoris; dissecting aneurysm of the aorta; stroke; new or recurrent transient ischemic attack; and sudden cardiac death.

^bMajor adverse cardiovascular events was defined as a composite of cardiovascular death, nonfatal myocardial infarction, acute coronary syndrome, stroke and hospitalization for heart failure.

^cMRA arm.

^dControl arm.

^eHyperklemia was defined as serum potassium >6.5 mmol/L.

^fHyperklemia was defined as serum potassium >6 mmol/L.

Despite these encouraging findings, the trial had several notable limitations. The open-label design without placebo control may have introduced performance and detection bias, as adverse events such as hyperkalemia were reported only in the spironolactone arm. Baseline imbalances favored the treatment group and the study lacked a predefined hyperkalemia management protocol. Key methodological details – randomization, allocation concealment, adherence, and co-interventions – were insufficiently described. Moreover, the trial was underpowered for its primary endpoint due to premature termination and fewer-than-expected events (32 vs. 126 required) [104]. The exclusively Japanese cohort of long-term, stable hemodialysis patients also limits generalizability to incident or more diverse populations. Nonetheless, DOHAS remains the first long-term randomized controlled trial to suggest mortality and cardiovascular benefits with an MRA in dialysis patients.

Following this, Lin et al. [101] conducted a double-blind, placebo-controlled randomized controlled trial in 253 hemodialysis and PD patients across three centers in Southeast China (Table 3; Fig. 1). Eligible participants had been on dialysis for at least 3 months and had no symptoms of heart failure. Patients were randomized 1:1 to receive spironolactone 25 mg daily or placebo for 1 year. The primary outcome was a composite of death due to cardiovascular or cerebrovascular events, aborted cardiac arrest, and sudden cardiac death, with secondary endpoints including all-cause mortality and echocardiographic measures. Participants had a mean age of 70 years, 61% were male, and mean dialysis vintage was 3.5 years. Over 1 year, the composite primary outcome occurred in 9 (7.2%) patients in the spironolactone group and 23 (18%) in the placebo group (adjusted HR 0.42, 95% CI: 0.26–0.78). All-cause mortality was similarly reduced (12 vs. 25 deaths; adjusted HR 0.52, 95% CI: 0.29–0.94). Adverse events were infrequent, with hyperkalemia (>6 mmol/L) occurring in only three patients receiving spironolactone.

Although these results suggested a substantial cardioprotective effect, several limitations warrant caution. The trial was underpowered, with only 37 composite events and had a short follow-up of 1 year. Outcomes were not adjudicated by an independent blinded committee, raising the potential for detection bias. Furthermore, details regarding potassium monitoring, adherence, and adverse event reporting – particularly in the placebo arm – were limited. Despite these weaknesses, the study by Lin et al. reinforced the cardioprotective signals observed in DOHAS, renewing interest in the potential role of MRAs in dialysis [100, 105–108].

Large Multicenter Phase 3 Randomized Controlled Trials in Dialysis Population

Building on earlier mechanistic and small-scale studies, two pivotal multicenter phase-3 randomized controlled trials – ALCHEMIST and ACHIEVE – were undertaken to definitively assess the cardiovascular efficacy and safety of spironolactone in patients receiving maintenance dialysis.

The ALCHEMIST Trial

The Aldosterone Antagonist Chronic Hemodialysis Interventional Survival Trial (ALCHEMIST) was a multicenter, double-blind, placebo-controlled randomized trial conducted across 64 dialysis centers in France, Belgium, and Monaco (Table 3; Fig. 1). Adults on thrice-weekly hemodialysis for ≥3 months with at least one cardiovascular comorbidity or risk factor – such as diabetes, cardiovascular disease, left ventricular hypertrophy, reduced ejection fraction, wide QRS, left bundle branch block, or elevated C-reactive protein – were eligible. Following informed consent, participants entered a 4-week open-label run-in phase receiving spironolactone 25 mg every other day. Those with serum potassium <5.5 mmol/L were then randomized 1:1 to spironolactone (titrated up to 25 mg daily as per a pre-specified potassium-monitoring algorithm) or matching placebo. The primary endpoint was time to first major adverse cardiovascular event (MACE) – a composite of cardiovascular death, nonfatal myocardial infarction, acute coronary syndrome, stroke, or hospitalization for heart failure – adjudicated by an independent blinded committee. Secondary outcomes included the individual MACE components, all-cause mortality, hyperkalemia (>6 mmol/L), BP changes, and new-onset atrial fibrillation.

Between June 2013 and November 2020, 1,442 patients were screened, 794 entered the run-in, and 644 were randomized (320 to spironolactone and 324 to placebo). Median age was 70.8 years, 69% were male, and median dialysis vintage was 1.7 years. The trial was terminated early due to funding limitations. Over a median follow-up of 2.7 years, the primary composite outcome occurred in 78 spironolactone-treated and 79 placebo-treated patients (10.66 vs. 10.70 events per 100 patient-years; HR 1.00, 95% CI: 0.73–1.36). Among secondary outcomes, hospitalization for heart failure (HR 0.41, 95% CI: 0.17–1.00), nonfatal cardiovascular events (including resuscitated cardiac arrest, HR 0.66, 95% CI: 0.43–0.99), and incident atrial fibrillation (odds ratio [OR] 0.52, 95% CI: 0.27–0.99) were less frequent with spironolactone. There were no differences in other secondary outcomes.

Hyperkalemia (>6 mmol/L) occurred in 42% of participants receiving spironolactone and 41% on placebo (HR 1.12, 95% CI: 0.88–1.43). Gynecomastia occurred in 14 (4%) patients in the spironolactone arm versus 4 (1%) with placebo. Predialysis BP remained unchanged throughout follow-up. Subgroup analyses showed no significant effect modification except by age: patients <75 years tended toward benefit, whereas those ≥75 years had a neutral or adverse trend [102].

The ALCHEMIST trial was rigorously designed with frequent potassium monitoring, dose titration, and blinded endpoint adjudication. However, early termination and a 23% dropout rate limited statistical power. Spironolactone 25 mg daily did not reduce MACE in high-risk hemodialysis patients but suggested possible reductions in heart-failure hospitalizations and atrial fibrillation.

The ACHIEVE Trial

The Aldosterone blockade for health improvement evaluation in end-stage renal disease (ACHIEVE) trial was an international, multicenter, double-blind, placebo-controlled randomized trial conducted across 143 dialysis centers in 12 countries (Table 3; Fig. 1). Adults aged ≥45 years – or ≥18 years with diabetes – on hemodialysis or PD for ≥3 months were eligible. All participants entered a 7-week open-label run-in phase receiving spironolactone 25 mg daily, with serum potassium measured at weeks 1, 2, 3, and 7. Only those maintaining potassium <6.0 mmol/L and ≥80% adherence were randomized 1:1 to spironolactone 25 mg daily or placebo. The primary endpoint was a composite of cardiovascular death or hospitalization for heart failure. Secondary outcomes included all-cause and cause-specific mortality, all-cause hospitalization, and severe hyperkalemia (>6.5 mmol/L or any hyperkalemia leading to hospitalization or treatment discontinuation) [109].

Between September 2017 and October 2024, 3,689 patients were screened, 3,565 entered the run-in, and 2,538 were randomized (1,260 spironolactone; 1,278 placebo). Mean age was 62 years, 63% were male, 84% were on hemodialysis, and median dialysis vintage was 2 years; 36% were receiving RAAS inhibitors. At the planned 75% interim analysis, the Data Safety Monitoring Board recommended early termination for futility. After a median follow-up of 1.8 years, the primary composite outcome occurred in 258 spironolactone vs. 276 placebo participants (10.5 vs. 11.3 events per 100 patient-years; HR 0.92, 95% CI: 0.78–1.09). No secondary endpoints significantly favored spironolactone, though a trend toward lower cardiac mortality was observed (HR 0.81, 95% CI: 0.64–1.03) [103].

Subgroup analyses indicated a potential sex interaction, with benefit in men (HR 0.81, 95% CI: 0.66–0.99) but not in women (HR 1.23, 95% CI: 0.91–1.67; p-interaction = 0.02). This persisted after multivariable adjustment, consistent with the smaller mechanistic study by Ito et al. mentioned above which also showed greater left ventricular mass regression in male dialysis patients, although such differences have not been observed in non-dialysis MRA trials [110]. Severe hyperkalemia was more frequent with spironolactone (5.0 vs. 3.2 events per 100 patient-years; HR 1.54, 95% CI: 1.07–2.22), though events requiring hospitalization or treatment discontinuation were comparable and largely driven by asymptomatic potassium elevations >6.5 mmol/L. Gynecomastia occurred in 1.5% vs. 0.5% and treatment discontinuation rates were 23% and 19%, respectively. Overall, spironolactone 25 mg daily did not reduce major cardiovascular outcomes in dialysis patients but modestly increased asymptomatic hyperkalemia.

Despite its neutral results, the ACHIEVE trial was a large-scale, international study with rigorous design and had similar findings with the aforementioned ALCHEMIST study. The early futility-stop and high baseline event rates (cardiovascular mortality 8.3, all-cause mortality 17.0, and hospitalization 88.1 events per 100 patient-years) underscore both the heavy cardiovascular burden and the ongoing unmet need for effective therapies in this high-risk group.

Evidence Synthesis and Meta-Analyses of MRAs in Dialysis Populations

Early quantitative syntheses of MRAs in dialysis were based on small, single-center randomized trials, most evaluating spironolactone at the dose of 25–50 mg daily [111–114]. These meta-analyses suggested striking reductions in cardiovascular and all-cause mortality, with pooled RR of 0.34 (95% CI: 0.15–0.75) and 0.40 (95% CI: 0.23–0.69), respectively [115]. A more rigorous appraisal was provided by the Cochrane review, which included 16 randomized trials enrolling 1,446 patients on maintenance hemodialysis or PD. MRA therapy appeared to radically reduce all-cause mortality (RR 0.45, 95% CI: 0.30–0.67) and cardiovascular mortality (RR 0.37, 95% CI: 0.22–0.64) with a number needed to treat for an additional beneficial outcome of 14, though the certainty of evidence was rated moderate. This benefit was not associated with an increase in serious hyperkalemia (RR 1.41, 95% CI: 0.72–2.78), with low certainty of evidence [116]. However, these apparent benefits were mainly driven by small, heterogeneous studies with short follow-up, incomplete allocation concealment, absence of blinded adjudication, and variable potassium monitoring.

The publication of the large, methodologically robust ALCHEMIST and ACHIEVE trials substantially reshaped the evidence base [102, 103]. An updated meta-analysis (2025) incorporating 19 randomized trials (n = 4,675) reevaluated MRA efficacy and safety using a Bayesian random-effects model (online suppl. Table 3). When restricted to low-risk-of-bias studies (PHASE, SPin-D, MiREnDa, ALCHEMIST, and ACHIEVE; n = 3,562), MRA therapy showed no significant effect on cardiovascular mortality (OR 0.98, 95% CI: 0.80–1.20; I² = 2.9%), all-cause mortality (OR 0.97, 95% CI: 0.84–1.12; I² = 0.0%), or heart-failure hospitalizations (OR 0.70, 95% CI: 0.30–1.65; I² = 71.1%). The previously observed survival advantage disappeared once large phase 3 trials were incorporated. This updated synthesis also confirmed a modest but statistically significant increase in hyperkalemia, particularly for potassium ≥6.5 mmol/L (OR 1.50, 95% CI: 1.11–2.03; I² = 0.0%) and in adverse events, such as gynecomastia (OR 3.02, 95% CI: 1.57–5.81; I² = 0.0%) [117].

Evidence Gaps, Challenges, and Future Directions

Despite the compelling cardiovascular benefits of MRAs in heart failure, these effects have not translated to the dialysis population, despite their high cardiovascular burden. This discrepancy likely reflects the distinct cardiovascular pathophysiology in kidney failure, driven by chronic pressure-volume overload, intradialytic myocardial stunning, vascular calcification, and non-atherosclerotic processes fueled by uremia, inflammation, and oxidative stress. Competing non-cardiovascular mortality, diagnostic overlap between heart failure and volume overload, and heterogeneity in cardiac remodeling further obscure treatment effects [118–120].

This is not the first instance where a proven cardioprotective therapy has failed in dialysis. The 4D, AURORA, and SHARP trials found no cardiovascular benefit of statins in dialysis, while FOSIDIAL demonstrated no effect of fosinopril on cardiovascular death, myocardial infarction, stroke, or heart failure hospitalization, contrasting sharply with benefits seen in earlier CKD [121–124]. Currently, SGLT-2 inhibitors are being evaluated in advanced CKD and in dialysis populations [125, 126].

Whether this marks the end of the road or a new direction for MRAs in dialysis remains uncertain. The dialysis population is heterogeneous and potential benefit may exist in currently unrecognized subgroups [127–129]. Most dialysis patients with heart failure have the HFpEF phenotype, which may respond more favorably to nonsteroidal MRAs-like finerenone, yet to be tested in this setting [130–132]. Aldosterone exerts both genomic (fibrotic and inflammatory) and non-genomic (oxidative and vascular) effects via MR activation. While MRAs primarily block genomic pathways, aldosterone synthase inhibitors act upstream at CYP11B2, suppressing aldosterone synthesis and thereby blocking both signaling arms [133–135]. Ongoing trials with these agents will assess their therapeutic potential in CKD, HFrEF, and HFpEF populations [136, 137]. Whether targeting aldosterone synthesis directly through aldosterone synthase inhibition could complement or even surpass MRA therapy-offering a novel pathway for cardiovascular protection in advanced CKD and dialysis – remains to be determined [134, 138]. It is also likely that any single intervention might not be sufficient to reverse the profound dysregulation associated with end-stage renal disease and that multitarget treatment approaches be required to significantly alter prognosis in the dialysis population [128]. Until then, effective cardiovascular prevention in dialysis remains an aspiration yet to be realized.

Conclusions

Evidence has evolved from early optimism based on small, heterogeneous studies to more neutral findings from large phase 3 trials (ALCHEMIST, ACHIEVE). Current data indicate that steroidal MRAs offer no cardiovascular or survival benefit in maintenance dialysis and modestly increase the risk of hyperkalemia and other adverse effects. Future studies should focus on identifying patient subgroups most likely to derive benefit and on determining whether earlier or more precisely targeted interventions could translate into meaningful cardiovascular improvements.

Acknowledgments

The figures were created using the BioRender software.

Conflict of Interest Statement

Dr. Elenjickal is supported by an educational grant from Pfizer. Dr. Mavrakanas has received speaker honoraria and has served on advisory boards for Bayer, Novo Nordisk, and Böhringer-Ingelheim. He has also received research grants from AstraZeneca, Bayer, and Pfizer. He is supported by a Fonds de Recherche Santé Quebec (FRSQ) Clinician Scholar award and a Kidney Research Scientist Core Education and National Training (KRESCENT) program New Investigator Award.

Funding Sources

No funding was received for this project.

Author Contributions

E.J.E. wrote the first draft and created the figures. T.A.M. devised the proof outline, critically revised the manuscript, and provided supervision.

Funding Statement

No funding was received for this project.