By Michael H. Crawford, MD, Editor
A post hoc analysis of the RALES and EMPHASIS HF trials has shown that, although treatment of heart failure with reduced left ventricular ejection fraction patients with mineralocorticoid receptor antagonists can cause a significant deterioration in renal function, the benefits outweigh the adverse effects and should not lead to automatic therapy discontinuation.
Matsumoto S, Henderson AD, Shen L, et al. Mineralocorticoid receptor antagonists in patients with heart failure and impaired renal function. J Am Coll Cardiol 2024;83:2426-2436.
Several drugs used for the management of heart failure can compromise renal function, but it has been shown recently that despite reductions in renal function, the benefits of sodium glucose cotransporter-2 inhibition or angiotensin receptor neprilysin inhibitors persist. Thus, this international group of investigators hypothesized that reductions in estimated glomerular filtration rate (GFR) to < 30 mL/min/1.73 m² in patients with heart failure (HF) would occur more frequently with mineralocorticoid receptor antagonists (MRAs), but this would not compromise the benefits of MRAs or reduce their safety to unacceptable levels.
The investigators pooled the data from the RALES (Randomized Aldactone Evaluation Study) and the EMPHASIS HF (Eplerenone in Mild Patients Hospitalization and Survival Study in HF) trials, which showed a beneficial effect in HF patients with reduced ejection fraction (HFrEF) compared to placebo. RALES enrolled patients with left ventricular EF ≤ 35% with New York Heart Association symptom levels of III or IV on an angiotensin-converting enzyme inhibitor (ACEI) and a loop diuretic. EMPHASIS HF also enrolled patients with EF ≤ 35% on a loop diuretic, but who were taking an ACEI or angiotensin receptor blocker (ARB), plus a beta-blocker if tolerated. Both trials excluded patients with serum potassium (K+) > 5.0 mmol/L and significant renal dysfunction (RALES serum creatine > 2.5 mg/dL; EMPHASIS HF GFR < 30 mL/min/1.73 m²). Both trials permitted study drug titration.
The primary endpoint for this post hoc analysis was a composite of cardiovascular (CV) mortality or hospitalization for HF. The safety endpoints were hyperkalemia and decreased renal function. Of the 4,355 patients enrolled in both trials, 295 (7%) experienced a reduction in GFR to < 30 mL/min/1.73 m². These patients tended to be older women with diabetes and with lower EF and GFR at baseline.
As expected, these patients also were more frequently those assigned to MRAs (9% vs. 5%). Compared to placebo, those assigned to MRAs experienced the primary outcome less frequently (hazard ratio [HR], 0.66; 95% confidence interval [CI], 0.59-0.73), and this benefit was observed in those who maintained a GFR > 30 mL/min/1.73 m² (HR, 0.63; 95% CI, 0.56-0.71) and those who developed a GFR < 30 mL/min/1.73 m² (HR, 0.65; 95% CI, 0.43-0.99). Also, the absolute risk of the primary outcome was reduced more frequently in those experiencing a GFR decline to < 30 mL/min/1.73 m² compared to those without such a decrease (21 vs. 9 per 100 patient-years).
An increase in K+ to > 5.5 mmol/L was more frequent in those taking MRAs (HR, 2.22; 95% CI, 1.80-2.72), especially in those who developed a GFR < 30 mL/min/1.73 m² (HR, 5.56; 95% CI, 1.68-18.37). In addition, severe hyperkalemia (> 6.0) was more frequent in those who developed a GFR < 30 mL/min/1.73 m² (2.9 vs. 1.2 per 100 patient-years). Thus, in those who developed a GFR < 30 mL/min/1.73 m², at a cost of 3/100 patient-years of severe elevations in K+, those taking an MRA had 21/100 patient-years fewer of the primary endpoint.
The authors concluded that since patients who develop a GFR < 30 mL/min/1.73 m² are at very high risk of adverse outcomes, the absolute risk reduction with an MRA is large, and such a decline in GFR should not automatically require treatment discontinuation.
COMMENTARY
The complex interplay between HF and kidney disease creates problems for the clinician. These two conditions share some of the same risk factors, such as hypertension and diabetes. Also, the onset of one in a patient with the other worsens prognosis. In addition, treatments for one may adversely affect the other. Thus, this report from an international group of investigators is of interest.
Many clinical trials of HF drugs have excluded patients with GFRs < 30 mL/min/1.73 m², as did both trials discussed in this study. However, the issue of what to do if renal function declines during HF treatment, which is not uncommon, is the subject of this post hoc analysis of two trials of MRAs in HFrEF patients.
Predictably, 7% of the patients developed a GFR < 30 mL/min/1.73 m², half within the first six months of therapy. Also, not surprisingly, these patients more often had diabetes, with lower EF and GFR at baseline. Perhaps because these were higher-risk patients, the benefit of MRAs was almost equal in those who developed a GFR < 30 and those who did not (P = 0.87 for the interaction). The trade-off was a higher incidence of hyperkalemia in those who developed a GFR < 30 mL/min/1.73 m² as expected. Surprisingly, in absolute numbers, for every three patients per 100 patient-years who developed severe hyperkalemia (> 6.0), 21 patients per 100 patient-years experienced a decrease in the risk of the primary outcome.
There are limitations to a post hoc analysis that was not prespecified, in that there may be residual unaccounted-for biases. Also, this analysis does not address the issue of initial treatment of HF patients with GFRs < 30 mL/min/1.73 m². However, for those taking MRAs who develop a GFR < 30 mL/min/1.73 m², as long as their potassium remains < 5.5, it may not be wise to just stop the MRA. Although a small range of MRA doses was permitted by physician discretion, the investigators did not address the issue of whether lowering the MRA dose was desirable or abrogated the potential benefits. They did report that more patients in the group that developed a GFR < 30 mL/min/1.73 m² did have their MRA stopped (26% vs. 19%), which may have inflated the benefits to this group in an intention-to-treat analysis. In summary, I believe this report serves to encourage us to carefully consider the potential downside to any discontinuation of guideline-directed medical therapy for HF.