Recurrent Hyperkalemia in Renin-Angiotensin-Aldosterone System Inhibitor (RAASi) Treatment

Renin-angiotensin-aldosterone system inhibitors (RAASis) represent a cornerstone in the treatment of proteinuric CKD, supported by multiple clinical trials showing their effectiveness in delaying its progression (1). Their use has been recommended by major consensus clinical practice guidelines (2). These medications are also a mainstay therapy for hypertension, heart failure with reduced ejection fraction, and coronary artery disease.

Given the central role of the renin-angiotensin-aldosterone system in regulating distal nephron secretion of potassium, it is not surprising that the use of RAASi markedly increases the risk of hyperkalemia, especially in those with reduced eGFR (3). This excess risk has been demonstrated in large multicenter clinical trials (4) and observational studies (5). Although the risk is greatest among those with advanced CKD (4), RAASis are also among the most common causes of hyperkalemia in patients without CKD (5). The adverse consequences of hyperkalemia are demonstrated by strong graded associations with risk of both morbidity and mortality, independent of eGFR and comorbidities (3,6).

Severe hyperkalemia (often defined as serum potassium ≥6 mEq/L) is frequently considered a medical emergency potentially requiring in-hospital care, but mild to moderate hyperkalemia is commonly treated in an outpatient setting. Historically, common ambulatory interventions for hyperkalemia include remeasurement of serum potassium, dietary potassium restriction, discontinuation or dose decrease of RAASi, initiation or dose increase of a kaliuretic diuretic, and/or initiation of potassium-binding medications (e.g., sodium polystyrene sulfonate). Although sodium polystyrene sulfonate has been widely used for more than half a century for the management of hyperkalemia, there is no high-quality evidence supporting its efficacy and safety, and patient reports and cohort studies suggested a small but significantly greater risk of severe gastrointestinal complications (3). The use of different interventions for RAASi-related hyperkalemia may vary widely, and to our knowledge, there are no randomized controlled trials on the comparative effectiveness of these common treatment options.

In recent years, two alternative potassium-reducing agents—sodium zirconium cyclosilicate and patiromer—have been shown to be effective and safe in treating hyperkalemia and have been approved by the Food and Drug Administration for nonemergent treatment. Their long-term effects on “hard” outcomes, such as major cardiovascular events and mortality, have yet to be described, however. Furthermore, the phase 3 trials demonstrating their efficacy compared active treatment with placebo, but not with other potassium-lowering strategies such as RAASi dose adjustment, potassium-restricted diets, or increasing kaliuretic diuretics.

In the context of these uncertainties regarding the longer-term effects of different ambulatory interventions for hyperkalemia in patients with CKD, the study by Hundemer et al. (7) in this issue of CJASN provides estimates of these treatment effects, although with important methodologic limitations to consider. The authors examine the 1-year risk of hyperkalemia recurrence (defined as serum K⁺ ≥5.3 mEq/L) after an episode of RAASi-related hyperkalemia in adults ≥66 years old on the basis of six mutually exclusive, but common, ambulatory hyperkalemia management strategies: (1) no intervention (continuing RAASi at the same dose with no other pharmacologic intervention); (2) RAASi discontinuation; (3) RAASi dose decrease; (4) new diuretic prescription; (5) diuretic dose increase; or (6) new sodium polystyrene sulfonate prescription. In addition, the 1-year risks of cardiovascular events and all-cause mortality were assessed as secondary outcomes. This was a population-based retrospective cohort study in Ontario, Canada, that used linked databases to identify residents treated with an RAASi and at least one outpatient hyperkalemia measurement over the span of 10 years. Models were adjusted for potential confounders, including baseline eGFR, comorbidity, and initial serum K⁺. Patients with severe hyperkalemia (>7 mEq/L) were excluded.

A total of 49,571 patients were included in the analyses. By far, most patients (77%) received no identified pharmacologic intervention, which may represent in part a “careful monitoring” approach of dietary patient education and counseling plus plans for remeasurement. Of the remaining study participants who received some kind of pharmacologic intervention, the most common intervention (74%) was RAASi discontinuation, whereas 15% were prescribed an RAASi dose decrease, 3% were newly prescribed a diuretic, 7% had an increase to their dose of a preexisting diuretic, and 1% were prescribed sodium polystyrene sulfonate. Not surprisingly, the initial potassium was higher for patients prescribed sodium polystyrene sulfonate (5.7 [5.5–6.0] mEq/L) compared with the other groups. Also, as one would expect, patients treated with a diuretic dose increase were more likely to have heart failure (59%), and those treated with sodium polystyrene sulfonate had more advanced kidney disease.

The primary outcome of recurrent hyperkalemia within 1 year occurred in 34% of participants. Compared with those with no pharmacologic intervention, the confounder-adjusted risk for recurrent hyperkalemia was lowest with RAASi discontinuation (hazard ratio [HR], 0.82; 95% confidence interval [95% CI], 0.78 to 0.85) and highest with sodium polystyrene sulfonate (HR, 1.30; 95% CI, 1.04 to 1.63), and no difference in risk was observed with the other pharmacologic interventions. With regard to morbidity/mortality at 1 year, RAASi dose decrease and diuretic dose increase were associated with a higher risk for cardiovascular events (HR, 1.35; 95% CI, 1.23 to 1.48 and HR, 1.70; 95% CI, 1.51 to 1.91, respectively), with no significant risk difference observed for the remaining interventions. Diuretic dose increase was the only pharmacologic intervention associated with a higher 1-year all-cause mortality risk (HR, 1.36; 95% CI, 1.14 to 1.62), with no significant difference in risk for the remaining interventions.

These findings suggest that when encountered with the issue of RAASi-induced hyperkalemia not requiring emergency treatment, a more conservative approach of RAASi discontinuation may result in a lower long-term risk of hyperkalemia recurrence, even when compared with sodium polystyrene sulfonate, and without any associated increased morbidity/mortality, at least within a 1-year time frame. Of note, prior single-center studies of patients with advanced CKD have suggested that stopping RAASi results in small increases in kidney function and, possibly, even delays initiation of KRT (8). That said, the results of two recent studies (8,9), also among patients with advanced CKD, suggest that discontinuing RAASi associates with greater all-cause mortality and major adverse cardiovascular events. Of note, these studies did not examine interventions after hyperkalemia specifically. Interestingly, diuretic dose increase was the only pharmacologic intervention associated with a higher 1-year all-cause morbidity and mortality risk, although residual confounding from more severe heart failure may explain these differences (see below).

Strengths of the study include the large sample size and population-based nature of the study cohort. Differences in risks of hyperkalemia and morbidity/mortality were adjusted for many potential confounders, including major cardiovascular morbidity, initial eGFR, severity of hyperkalemia, and concurrent pharmacotherapy including nonsteroidal anti-inflammatory drugs. Results were also reasonably robust to various sensitivity analyses with different definitions of recurrent hyperkalemia. However, important limitations of the study must be considered when interpreting these results. As with all observational studies, there is the potential for residual confounding when comparing various pharmacologic interventions not assigned at random. For example, although the authors adjusted for the presence or absence of diagnosed heart failure in their analyses, those treated with increasing doses of diuretics may have had more severe heart failure than those whose diuretic dose was not adjusted, thereby explaining the greater cardiovascular event rate observed in the former. The indications for the pharmacologic interventions were also not available in the database used for this study, such that it cannot be known if changes in diuretic doses or RAASi were even motivated predominately by hyperkalemia or by other indications (e.g., decompensated heart failure, worsening kidney function, or drug intolerance) that might also affect the risk of the study outcomes independently. Pharmacologic interventions provided in the emergency room were also not ascertained, such that many cases of acute sodium polystyrene sulfonate treatment were likely missed and unaccounted for. As the authors note, the association of novel potassium-lowering agents (patiromer and sodium zirconium cyclosilicate) with clinical outcomes could not be examined in this study, as these drugs had not yet been approved in Canada during the study period. Finally, the study sample included many patients with preserved eGFR and a relatively small proportion with advanced CKD, such that differential effects in these patients could not be reliably assessed.

Perhaps the most reasonable approach when faced with the issue of recurrent RAASi-induced hyperkalemia in clinical practice is to make an individualized decision that includes risk stratification, careful monitoring for adverse effects if choosing to continue therapy and/or instituting pharmacologic interventions, and consideration of the potential implications for longer-term risk of kidney failure and cardiovascular events balanced against the inherent risks associated with prolonged hyperkalemia. Although this study by Hundemer et al. (7) helps inform these risks and benefits, clearly more and better data are needed, especially among those with advanced CKD in an era where newer potassium-resin binders may change the way we manage hyperkalemia. Ultimately, data from rigorously designed observational studies and randomized clinical trials are needed to compare the effect of different pharmacologic approaches to hyperkalemia in RAASi-treated patients not only on durable reduction of serum potassium, but also on the risk of cardiovascular morbidity and mortality. In this regard, studies such as the ongoing DIAMOND trial (NCT03888066) in patients with heart failure with reduced ejection fraction and RAASi-related hyperkalemia will provide important data on the effect of novel potassium-lowering therapies on clinical outcomes. Although not specific to hyperkalemia, the STOP-ACEi trial should also provide data on the risks and benefits of RAASi discontinuation on clinical outcomes in patients with advanced CKD (10).

Disclosures

S. L. Seliger reports employment with Baltimore Veterans Affairs Medical Center, the University of Maryland Medical Center, and the University of Maryland School of Medicine; consultancy agreements with Tricida, Inc. (Endpoint Adjudication Committee); receiving research funding from Kadmon Pharmaceuticals, Palladio Biosciences, Reata Pharmaceuticals, Roche Diagnostics, Inc., and Sanofi US; patents and inventions with the University of Maryland, Baltimore and the University of Texas Southwestern (Methods for Assessing Differential Risk for Developing Heart Failure); and serving as an Associate Editor for CJASN, a member of the Medical Review Board of ESRD Network 5, the chair on the board of directors of ESRD Network 5, a member of the Endpoint Adjudication Committee of VALOR-CKD trial (Tricida Inc.), and a member of the Editorial Board of Circulation. J. Bolanos reports reports employment with Baltimore Veterans Affairs Medical Center, the University of Maryland Medical Center, and the University of Maryland School of Medicine.

Funding

None.