No Need for Urgent Revisiting of Kalaemia Levels in Guidelines Despite Use of Mineralocorticoid Receptor Antagonists: Bring in More Evidence

The Times They Are A-Changin'. Bob Dylan (1964). When Bob Dylan released The Times They Are A-Changin' in the summer of 1964, aldosterone had already been isolated and characterized (done in 1953), and the mineralocorticoid receptor antagonist (MRA) spironolactone was gaining recognition (in the early 1960s). For aldosterone and MRAs, the next 30 years represented the Dark Ages, with only a few converts, largely from endocrinology, keeping the flame alive. In the 1990s, things started to change. The first sparks of renewed life for aldosterone came from the studies of Karl Weber, who showed that inappropriate aldosterone status in rats produced cardiac hypertrophy and global cardiac fibrosis.1 These and subsequent findings from other laboratories prompted the Randomized Aldactone Evaluation Study (RALES), headed by Bertram Pitt, which was prematurely terminated in August 1998 and published in September 1999.2 Briefly, RALES showed that the addition of a modest dose of spironolactone to patients with heart failure (HF) with reduced ejection fraction (HFrEF) and New York Heart Association (NYHA) class III–IV provided a 30% improvement in survival and a 35% improvement in morbidity/hospitalization. The follow-up 2003 Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS) trial, also under the precise baton of Pitt, demonstrated similar results with eplerenone (a selective aldosterone antagonist) among patients with myocardial infarction complicated by left ventricular dysfunction.3 The third pivotal MRA trial was Eplerenone in Mild Patients Hospitalization and Survival Study in Heart Failure (EMPHASIS-HF), rigorously conducted by Faiez Zannad and published in 2011.4 EMPHASIS-HF randomized 2737 patients with HFrEF and NYHA class II symptoms to either eplerenone or placebo, in addition to optimal medical therapy with angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and beta-blockers. At a mean follow-up of 1.8 years, eplerenone significantly reduced the rate of the composite primary outcome of cardiovascular death or HF hospitalization (18.3% vs. 25.9%; hazard ratio 0.63). The number needed to treat (NNT) to prevent one primary outcome at one year was 19, and the NNT to postpone one death per year was 51. No further pivotal MRA trials have been conducted in HFrEF, but these outstanding studies had a profound early impact on guidelines. Indeed, the European Society of Cardiology (ESC) HF guidelines gave MRA a class IB indication in the updates for 2001, 2005, and 2008, and MRAs were upgraded to class IA in the 2012 and 2016 versions.5 We have known for decades that, as expected, MRAs raise potassium levels.6 Although spironolactone and eplerenone protect against hypokalaemia, there are more episodes of hyperkalaemia, occasionally life-threatening. Since 2001, the ESC HF guidelines have recommended monitoring potassium levels at 1 and 4 weeks after initiation/uptitration and at 8 and 12 weeks and 6, 9, and 12 months. Furthermore, the guidelines indicate that the MRA dose be halved if K > 5.5 mmol/L and discontinued if K > 6.0 mmol/L.5 In the current opinion paper by Butler et al.7 published in this issue of the Journal, the authors propose a re-evaluation of current guidelines on potassium thresholds, in particular in HF patients on MRA therapy. Given the importance of the topic and the expertise of the authors, we respectfully provide our opinion to enrich and stimulate the debate. For the following reasons, we feel that there is a need for additional and robust evidence before lowering potassium thresholds. First, a revisit of HF guidelines for hyperkalaemia management to lower potassium values may affect MRA initiation, which is currently well-established with a class IA indication. At present, only slightly over 50% of patients even in the best-managed cohorts take MRAs, which are life-saving medications in use after decades of evidence-based trials. Is this because of medical inertia or even malpractice? We believe not. In real-life clinical practice, well into the 21st century, patients with HFrEF tend to be senior citizens with multiple co-morbidities, including renal dysfunction [49% with estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 in our series8] and frailty. Furthermore, lowering the potassium level threshold in guidelines may provide a basis for a more conservative approach, leading to MRA dose reduction or discontinuation—actions that can deprive patients of a therapy proven to improve clinical outcomes. Second, the evidence of a U-shaped association for prognosis relative to potassium levels is compelling.9 Nevertheless, the prognostic value of hyperkalaemia is not fully incontrovertible, in particular in patients with acute HF. Indeed, in the setting of acutely decompensated HF, the association between hyperkalaemia and the risk of mortality is conflicting or neutral.10, 11 On the other hand, more research is needed to fully elucidate whether the upper end of the U-shaped relationship reflects only harm from higher potassium levels or that higher levels of potassium mirror other important co-morbidities such as renal impairment. In addition, at the lower end of the U-shaped curve, in the presence of hypokalaemia, the risk for adverse events should not be underestimated. A recent report showing the continuum of serum potassium values found that the hazard ratios for all-cause mortality are higher in hypokalaemia compared to hyperkalaemia.9 The advent of new agents that reduce or bind potassium should keep us alert to the risks of hypokalaemia, which are often neglected. Third, the progressive incorporation of sacubitril/valsartan in patients with HFrEF may affect renal function and potassium levels. A sub-analysis of PARADIGM-HF data suggests that use of an MRA with sacubitril/valsartan rather than enalapril leads to less severe hyperkalaemia in patients with chronic HF.12 Indeed, potassium increases were more common with enalapril and MRA than with sacubitril/valsartan and MRA. These data suggest that neprilysin inhibition, which increases levels of natriuretic peptides, attenuates the risk for hyperkalaemia when MRAs are combined with other inhibitors of the renin–angiotensin–aldosterone system in HF patients. Recent reports adding to this line of evidence have shown that compared with enalapril, sacubitril/valsartan leads to a slower rate of decrease in the eGFR and improved cardiovascular outcomes, even in patients with chronic kidney disease.13 This effect was even more relevant in patients with diabetes (39.1% of patients in our HF cohort8). The authors concluded that neprilysin inhibition attenuates the tendency of diabetes to accelerate the deterioration of renal function in patients with chronic HF.14 Fourth, the high expectations put on potassium-binding agents are optimistic, possibly overly so.15 More work is needed to determine whether deploying potassium-binding agents to expand the use of MRAs in HF patients with hyperkalaemia would translate into improved clinical outcomes.16 The authors' proposal to revisit potassium levels is legitimate, but the clinical trial thresholds for MRA treatment initiation and dose change/discontinuation, as well as the frequency of monitoring, remain the only evidence-based advice we can give. A premature revision in guideline potassium levels without enough new evidence may alarm clinicians to avoid MRAs and ultimately be counterproductive for the patient with HF. A.B.G. was supported by grants from the Ministerio de Educación y Ciencia (SAF2014-59892), Fundació La MARATÓ de TV3 (201502, 201516), CIBER Cardiovascular (CB16/11/00403), and AdvanceCat 2014-2020. Conflict of interest: none declared.