Hyperkalemia as a Constraint to Therapy With Combination Renin‐Angiotensin System Blockade: The Elephant in the Room

The renin‐angiotensin‐aldosterone system (RAAS) is an established mediator of progressive renal and cardiovascular disease. Angiotensin II plays a key role in the progression of cardiovascular and renal disease, and drugs that interfere with angiotensin II actions such as angiotensin‐converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) can offer cardiovascular and renal protection beyond their effects on blood pressure. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ Recently, a number of experimental and clinical investigations have implicated aldosterone, independent of angiotensin II, in the pathogenesis of progressive cardiovascular and renal disease. ⁷ , ⁸ , ⁹ These studies have generated intense interest in aldosterone receptor antagonism, either alone or in combination with an ACEI or an ARB, in the treatment of a wide range of cardiovascular diseases as well as progressive renal injury.

Blockade of the renin‐angiotensin system (RAS) with either an ACEI or an ARB is important in patients with hypertension, heart failure, myocardial infarction, stroke, and various forms of nephropathy, the most common of which is due to diabetes mellitus. Support for RAS blockade derives from large randomized, controlled, clinical trials demonstrating that these agents attenuate progression of renal disease. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ Lowering blood pressure with other antihypertensive agents has also been shown to slow renal disease progression.

Clinical evidence suggests that blockers of the RAS are necessary adjuncts to antihypertensive regimens in delaying progression of chronic kidney disease. Both ACEIs and ARBs are recommended by numerous hypertension and diabetes guidelines around the world for patients with kidney disease due to nondiabetic and diabetic etiologies. ¹⁰ , ¹¹ , ¹² The consistency of their benefits is well described in the literature. The earliest reports demonstrated the advantage of ACEIs in patients with type 1 diabetes. ¹ These early studies were extended to many ACEIs in both diabetic nephropathy and nondiabetic kidney disease. Jafar and associates ¹³ published a meta‐analysis indicating that the renal protective effects of ACEIs are also seen in patients with nondiabetic kidney disease. More recently, ARBs as part of a treatment regimen have been demonstrated to benefit patients with type 2 diabetes and chronic kidney disease. Clearly, these effects are important because diabetic nephropathy constitutes the most common cause of end‐stage renal disease (ESRD) in the western world. ⁴ , ⁵ Both the Irbesartan Diabetic Nephropathy Trial (IDNT) ⁵ and Reduction of Endpoints in NIDDM With the Angiotensin II Antagonist Losartan (RENAAL) ⁴ studies conclusively demonstrated that using an ARB as part of a successful multidrug blood pressure–lowering regimen reduces the risk of the composite end point of time to doubling of serum creatinine level, ESRD, and all‐cause death. Subsequent clinical experience has established that RAS‐blocking drugs are critically important when used as part of a successful blood pressure–lowering regimen in patients with all forms of chronic kidney disease.

Carefully conducted clinical trials have demonstrated that dual blockade of the RAAS with the combination of an ACEI added to an ARB or an aldosterone receptor antagonist added to existing therapy with either an ACEI or an ARB can reduce cardiovascular end points in high‐risk patients with evidence of cardiovascular disease. ⁷ , ⁸ Spironolactone 25 mg added to standard care including an ACEI improved outcomes in patients with stage III and IV congestive heart failure. ⁷ Eplerenone 25 to 50 mg added to either an ACEI or an ARB plus a β‐blocker reduced end points in patients with a reduced ejection fraction following myocardial infarction. ⁸ The use of aldosterone antagonists in adults in addition to an ACEI as therapy with an ARB in patients whose BP is difficult to control is also becoming common.

The proverbial “elephant in the room” is the potential for hyperkalemia developing. An important consideration in assessing the feasibility of RAAS blockade using a combination of renin‐angiotensin or aldosterone blockers is the risk of provoking hyperkalemia. Unfortunately, the growing practice of prescribing dual RAAS blockade in patients with congestive heart failure and following myocardial infarction has been associated with higher rates of hyperkalemia and its complications, especially in patients with underlying renal impairment. Consequently, hyperkalemia among those receiving dual RAAS blockade has become a growing public health concern. ¹⁴ A recent population‐based time series analysis suggested that there has indeed been a significant increase in the use of spironolactone among patients with congestive heart failure treated with an ACEI, and this growing practice has been associated with higher rates of hyperkalemia and its complications. ¹⁵ Clearly, this constitutes the most common limitation of either combining 2 RAAS blockers or up‐titration of an ACEI or an ARB.

The pathophysiologic basis for the development of hyperkalemia is incomplete or impaired adaptability by the kidney. Patients with chronic kidney disease have a diminished glomerular filtration rate and impaired distal potassium secretion, which together result in the reduced capacity to excrete potassium and blunted kaliuresis in response to potassium challenge. Nevertheless, the kidney is able to adapt its potassium excretion to maintain potassium balance despite very low levels of glomerular filtration rate, provided that aldosterone secretion and tubular responsiveness to aldosterone are intact and oliguria is not present. Patients with chronic kidney disease are therefore at particular risk for hyperkalemia should they receive a medication that further impedes potassium excretion by interfering with the RAAS such as an ACEI, an ARB, or an aldosterone receptor antagonist. Because dual blockade interferes at 2 separate points of a primary mechanism by which the failing kidney maintains critical potassium balance, an aldosterone receptor antagonist added to either an ACEI or an ARB could increase the risk of hyperkalemia even further.

What Constitutes Hyperkalemia?

The spectrum of possible untoward events attributable to hyperkalemia is extensive, ranging from single episodes of hyperkalemia, sustained hyperkalemia, or—as in the Ongoing Telmisartan Alone and in Combination With Ramipril Global Endpoint Trial (ONTARGET)—cases of hyperkalemia severe enough to necessitate dialytic intervention. Although severe hyperkalemia is generally considered a life‐threatening event, there is little agreement on what constitutes mild, moderate, or severe hyperkalemia. In a comprehensive review, Levinsky ¹⁶ defined “minimal” hyperkalemia as a serum potassium concentration <6.5 mmol/L accompanied by only electrocardiographic changes, whereas others ¹⁷ have defined a potassium concentration >6.0 mmol/L as severe hyperkalemia. Moreover, few definitions take into account that the toxic effects of a given potassium concentration are dependent on the baseline value and the rate of increase in potassium concentration as well as the acid‐base status and serum calcium concentration. ¹⁸ , ¹⁹

Regretfully, reporting of hyperkalemia has been haphazard, inconsistent, and often confounded by concomitant interventions. Using regimens utilizing the addition of an aldosterone blocker to an ACEI exemplifies these drawbacks. Focusing only on extreme and unsustained instances underestimates the incidence of treatment‐induced hyperkalemia and consequently precludes a critical assessment of both the magnitude and severity of this complication.

Vacillations of Plasma Potassium: A Caution

Importantly, defining hyperkalemia based on 1 or 2 plasma determinations can be misleading, as highlighted by the case of the patient depicted in the Figure. The patient participated in a clinical study consisting of forced titration of eplerenone to a dosage of 200 mg/d in addition to baseline angiotensin‐converting enzyme inhibition. The patient’s serum potassium value increased to 6.3 mEq/L. The patient continued therapy with an ACEI and eplerenone 200 mg/d despite the study protocol mandating that eplerenone should be discontinued with the onset of severe hyperkalemia. As can be seen, the patient’s serum potassium value subsequently decreased spontaneously to 4.3 mEq/L despite the persistent provocation of treatment with an ACEI and full doses of eplerenone. This illustrative case demonstrates the vicissitudes and vagaries of spontaneous changes in potassium and emphasizes the requirement for multiple potassium determinations to delineate the potential risk of hyperkalemia.

Figure.

 Illustrative case demonstrating a dissociation between serum potassium and the drug treatment regimen. As can be seen, the patient's serum potassium value decreased spontaneously from 6.3 to 4.3 mEq/L despite the persistent provocation of treatment with an angiotensin‐converting enzyme inhibitor and full doses of eplerenone (EPL) 200 mg/d.

How often hyperkalemia supervenes in patients treated with aldosterone blockers when added to an ACEI or ARB is unsettled. Whereas a few reports have suggested that it constitutes a problem, ¹⁵ , ²⁰ a careful review of these reports discloses several confounding factors. Svensson and colleagues ²⁰ reported that 36% of 125 patients with congestive heart failure treated with spironolactone added to an ACEI had a serum potassium value >5 mmol/L, but the patients were elderly (mean age, 72.9 years). Juurlink and associates ¹⁵ recently examined trends in the rate of prescriptions for spironolactone and in the rate of hospitalization for hyperkalemia before and after the publication of Randomized Aldactone Evaluation Study (RALES) among older patients who received ACEIs. Among patients treated with ACEIs who had recently been hospitalized for heart failure, the spironolactone prescription rate was 34 per 1000 patients in 1994; it increased immediately after the publication of RALES to 149 per 1000 patients by late 2001 (P<.001). The rate of hospitalization for hyperkalemia rose from 2.4 per 1000 patients in 1994 to 11.0 per 1000 patients in 2001 (P<.001), and the associated mortality rose from 0.3 per 1000 in 1994 to 2.0 per 1000 patients in 2001 (P<.001). Regretfully, many confounders qualify the applicability of this study. There were no direct measures of potassium or creatinine concentration, adherence to medications, use of nonprescription drugs, or clinical details surrounding death. Indeed, many of the patients hospitalized for hyperkalemia may have died of another illness. The diagnostic coding for hyperkalemia has not been validated; moreover, many patients hospitalized for hyperkalemia may have also had volume contraction or renal insufficiency related to spironolactone therapy. In addition, the authors were unable to identify adverse outcomes that occurred before admission. These considerations underscore the multiple confounders that can obscure rigorous reporting of the incidence of hyperkalemia attributable to the RAAS‐blocking intervention. The risks of hyperkalemia are greater in patients with heart failure than in hypertensive individuals with preserved cardiac function.

Do the Risks of Hyperkalemia Vary With the Drugs in the Regimen?

The available studies to date suggest that the risk of hyperkalemia may be less in regimens in which an aldosterone blocker is used as add‐on therapy. ²¹ , ²² As an example, Bianchi and associates ²¹ reported that hyperkalemia was relatively uncommon when an aldosterone blocker was added to an ACEI or an ARB in patients with diabetic or nondiabetic nephropathy and a baseline glomerular filtration rate of >55 mL/min/1.73 m². In 42 patients with chronic kidney disease treated with spironolactone 25 mg/d added to an ACEI or an ARB for 8 weeks, 5 patients had a serum potassium level of 5.5 mEq/L (all had an estimated glomerular filtration rate >60 mL/min/1.73 m², and the potassium value was <6.0 mEq/L in all cases). Of greater clinical relevance, data from recent clinical trials indicate that hyperkalemia is not more likely to develop in diabetic hypertensive patients with microalbuminuria or macroalbuminuria when treated with low‐dose eplerenone added to an ACEI. ²² Again, hyperkalemia is not common in hypertensive patients without evidence of renal disease who are treated with an ARB plus an ACEI or an aldosterone blocker.

The Importance of Drug Doses in Determining Hyperkalemia

The incidence of hyperkalemia associated with low‐dose eplerenone or spironolactone treatment underscores the importance of reviewing the dosing of each component of the regimen in assessing the relative risk of hyperkalemia. A forced titration study with eplerenone in hypertensive patients with type 2 diabetes reported a reduction in proteinuria at dosages of 200 mg/d, both as monotherapy and when coadministered with enalapril 10 mg. ²³ However, this high dosage of eplerenone was associated with an increased risk of hyperkalemia in this population. In a subsequent study, lower doses of eplerenone, when coadministered with enalapril, would produce a similar antialbuminuric effect in these patients without producing the hyperkalemia observed previously. ²⁴ In that study, the substantial reduction in the urinary albumin‐to‐creatinine ratio in the eplerenone 50 mg/enalapril treatment arm was not accompanied by significant increases in the incidences of either sustained or severe hyperkalemia compared with placebo/enalapril treatment. Whereas the incidence of sustained and severe potassium elevations with eplerenone 100 mg treatment was not significantly higher than with eplerenone 50 mg treatment, in some cases they were numerically higher and, further, there did not appear to be an incremental benefit for reduction of albuminuria with this higher dose. A dosing regimen of eplerenone 50 mg with an ACEI therefore may confer the desired antialbuminuric benefit while reducing the risk of hyperkalemia. It should, however, be noted that the majority (90%) of the study patients had a baseline estimated glomerular filtration rate >50 mL/min/1.73 m² at entry (estimated glomerular filtration rate varied widely from 34 to 153 mL/min/1.73 m²). Consequently, our results cannot yet be extrapolated to patients with type 2 diabetes and an estimated glomerular filtration rate <50 mL/min/1.73 m². Similar benefits may be obtained with moderate doses of spironolactone (50 mg) plus an ACEI.

A Proposed Approach to Define Severe and Sustained Hyperkalemia

As detailed in a recent study, we advocate using quantitative methodology to characterize hyperkalemia. Sustained hyperkalemia can be defined as a serum potassium level >5.5 mmol/L on 2 consecutive occasions 1 to 3 days apart, and severe hyperkalemia defined as a serum potassium level ≥6.0 mmol/L on any occasion. The role of glomerular filtration rate as a determinant of hyperkalemia has also been defined. A Cochran‐Mantel‐Haenszel test was performed to analyze the association between the incidence of hyperkalemia and treatment within each estimated glomerular filtration rate quartile and the association of the incidence of hyperkalemia between treatment and estimated glomerular filtration rate quartile (Table). Using this quantitative approach, we demonstrated in one study that the incidence of eplerenone‐induced hyperkalemia in patients with an estimated glomerular filtration rate >60 mL/min/1.73 m² did not differ significantly from placebo.

Table.

 Baseline eGFR as a Determinant of Eplerenone‐Induced Hyperkalemia

eGFR Quartile (mL/min/1.73 m 2)	PBO	EPL50	EPL100	P Valuea
	No. (%)	No. (%)	No. (%)	Within Quartile b	Treatment*Quartile Association c
Sustained hyperkalemia (serum potassium >5.5 mmol/L on 2 consecutive occasions)
<61	1/25 (4.0)	1/20 (5.0)	1/19 (5.3)	0.84	0.27
61–73	0/19 (0.0)	1/26 (3.8)	1/20 (5.0)	0.37
74–84	0/25 (0.0)	0/22 (0.0)	0/19 (0.0)	–
≥85	0/19 (0.0)	0/21 (0.0)	1/24 (4.2)	0.26
Severe hyperkalemia (serum potassium ≥6.0 mmol/L)
<61	2/25 (8.0)	1/20 (5.0)	0/19 (0.0)	0.22	0.12
61–73	0/19 (0.0)	1/26 (3.8)	1/20 (5.0)	0.37
74–84	0/25 (0.0)	0/22 (0.0)	1/19 (5.3)	0.18
≥85	1/19 (5.3)	0/21 (0.0)	3/24 (12.5)	0.29
Total	88	89	82	259

Reproduced from Epstein et al, 2006.²⁴ Abbreviations: eGFR, estimated glomerular filtration rate; PBO, placebo + enalapril 20 mg; EPL50, eplerenone 50 mg + enalapril 20 mg; EPL100, eplerenone 100 mg + enalapril 20 mg. ^aBased on the Cochran‐Mantel‐Haenszel test. ^b P Value testing for the association between the incidence of hyperkalemia and treatment within each quartile of eGFR. ^c P Vlue testing for association in the incidence of hyperkalemia between treatment and quartile of eGFR.

Proposal for Novel Source Test to Estimate Tissue Potassium Levels to Detect Impending Hyperkalemia

Approximately 3500 mEq of potassium are stored in the body of a 70‐kg human being, predominantly as intracellular ion. When we sample blood to determine the serum level of potassium, we are determining potassium stores in the extracellular fluid, which contains 60 to 70 mEq, or merely 2% of body potassium. Consequently, estimation of the magnitude of a deficit or excess of total body potassium as extrapolated from the serum potassium level constitutes a remarkably imprecise determination. Delgado and Delgado‐Almeida ²⁵ have marshaled evidence suggesting that the use of red blood cell potassium level monitoring constitutes a proxy for measuring intracellular potassium levels in other cells including cardiac tissue. Consequently, the development of a test to monitor red blood cell potassium may provide a practical point of service test to anticipate and detect impending hyperkalemia and guide rational dosing of RAAS blockers in patients with cardiovascular disease and progressive kidney disease.

Conclusions

Accruing evidence confirms that hyperkalemia constitutes the most prominent constraint to prescribing combination therapy with 2 RAAS blockers. Unfortunately, the data for hyperkalemia are often anecdotal, utilizing isolated plasma determinations without uniform sampling intervals. Furthermore, the data are usually poorly defined and not quantitative, merely noting the occurrence of hyperkalemia. I propose that future clinical investigations incorporate systemic determinations of serum potassium in the protocol to better assess the frequency and severity of hyperkalemia. Because the toxic effects of a given elevated serum potassium concentration is dependent on the baseline value as well as the rate of increase, clinical trials should mandate concomitant electrocardiographic assessments to ascertain the functional significance of the increases in serum potassium. Concomitant determinations of estimated glomerular filtration rate will facilitate an examination of the role of underlying renal function as a determinant of the propensity for the development of hyperkalemia.