Secondary Prevention of Hyperkalemia With Sodium Polystyrene Sulfonate in Cardiac and Kidney Patients on Renin‐Angiotensin‐Aldosterone System Inhibition Therapy

Abstract

Background:

Hyperkalemia, induced by renin‐angiotensin‐aldosterone system inhibition (RAAS‐I) in patients with chronic kidney disease (CKD), or cardiac disease often leads to withdrawal of RAAS‐I therapy. Sodium polystyrene sulfonate (SPS) is a potassium‐binding resin used for the treatment of hyperkalemia. Recently, concerns about the safety and efficacy of SPS were raised. We report here a follow‐up of 14 patients with CKD and heart disease on RAAS‐I treatment who were treated with low‐dose daily SPS to prevent recurrence of hyperkalemia.

Hypothesis:

Daily SPS is safe and effective for secondary prevention of hyperkalemia induced by RAAS‐I therapy in CKD patients with heart disease.

Methods:

We reviewed the medical charts of the patients with CKD (nondialysis patients) and heart disease treated in our CKD clinic from 2005 to 2010 and identified all patients on RAAS‐I therapy who were treated with daily SPS (sorbitol‐free) after episodes of hyperkalemia. Data on hospitalizations, symptoms that may be attributed to SPS therapy, and electrolyte concentration levels were obtained.

Results:

Fourteen patients were treated with low‐dose SPS therapy for a total of 289 months (median length of follow‐up, 14.5 months). None of the patients developed colonic necrosis or life‐threatening events that could be attributed to SPS use. Mild hypokalemia was noted in 2 patients and responded to reducing the dose of SPS. No further episodes of hyperkalemia were recorded while patients were on the therapy. SPS was well‐tolerated during the follow‐up without need for withdrawal or reduction of the dose of RAAS‐I therapy by any patients.

Conclusions:

Low‐dose SPS was safe and effective as a secondary preventive measure for hyperkalemia induced by RAAS‐I in CKD patients with heart disease. © 2011 Wiley Periodicals, Inc.

The authors have no funding, financial relationships, or conflicts of interest to disclose.

Introduction

Renin‐angiotensin‐aldosterone system inhibition (RAAS‐I) therapy plays a pivotal cardioprotective role in certain heart diseases.1 Angiotensin converting enzyme inhibitors (ACE‐Is), angiotensin receptor blockers (ARBs), and the aldosterone antagonists spironolactone and eplerenone have been shown to improve survival for patients with severe heart failure (HF) and a reduced left ventricular (LV) ejection fraction as well as for patients with HF after myocardial infarction (MI) with LV systolic dysfunction.1, 2, 3, 4, 5, 6 The beneficial effects of RAAS‐I are not exclusive to those with heart disease. ACE‐Is, ARBs, and the direct renin inhibitor aliskiren have been shown to reduce the level of proteinuria for patients with chronic kidney disease (CKD).7, 8, 9 ACE‐Is and ARBs therapy can also slow the progression of CKD.7, 8

A major limitation in using RAAS‐I therapy is the increased incidence of hyperkalemia, especially in patients with concomitant CKD.10, 11 Patients with reduced glomerular filtration rate (GFR) were often excluded from enrollment in the large randomized studies that assessed the safety and efficacy of RAAS‐I in patients with heart diseases and, perhaps for this reason, a high incidence of hyperkalemia was not reported.5, 11 However, in clinical practice with less vigorous monitoring, RAAS‐I agents were found to be associated with a higher incidence of life‐threatening hyperkalemic events in HF patients.12, 13 The risk for hyperkalemia in these patients may also increase with the concomitant use of beta‐adrenergic blocking agents.9 The appearance of hyperkalemia may then lead the physician to withdraw RAAS‐I therapy.13

Sodium polystyrene sulfonate (SPS) (Kayexalate and Resonium‐A; Sanofi‐Aventis, Paris, France) is a potassium‐ binding resin widely used for the treatment of hyperkalemia.14 In clinical practice, SPS is often mixed with cathartics such as sorbitol to prevent constipation, which is sometimes seen with SPS.14, 15 Recently, concerns about the safety and efficacy of SPS were raised.14, 15 Importantly, colonic necrosis in patients treated with SPS given as an enema or orally has been reported.16, 17, 18 In almost all cases, colonic necrosis was associated with the use of SPS mixed with sorbitol.14, 15 The association with colonic necrosis led the US Food and Drug Administration to issue in 2009 a warning advising against the use of SPS mixed with sorbitol.19

Interestingly, we could not find in the literature reports of the prolonged routine use of low‐dose SPS therapy orally to prevent a recurrence of hyperkalemia after it had occurred in CKD and heart disease patients treated with RAAS‐I. We used this approach of secondary prevention of hyperkalemia in our CKD clinic with the rationale that successful prevention of hyperkalemic episodes may enable the patients to continue with the use of RAAS‐I therapy.

We report here on our experience in using low‐dose sorbitol‐free SPS for the secondary prevention of hyperkalemia in 14 patients with concomitant CKD and heart disease on RAAS‐I therapy who had developed hyperkalemia. Our main findings were that low‐dose SPS avoided further hyperkalemia and was well tolerated without any adverse effect attributed directly to SPS. The use of SPS enabled all patients during the follow‐up to continue with the RAAS‐I therapy without a change in dosage.

Methods

Patients and Data Recruitment

Approval of the study was granted by the Ethical Committee of the Tel‐Aviv Medical Center. We obtained medical records of all patients who were treated with RAAS‐I in our CKD clinic in the nephrology department between January 1, 2005, and December 31, 2010, and who had a combined diagnosis of CKD and heart disease. Of these patients, we identified all patients who were treated with low‐dose SPS (15 g once daily mixed in a glass of water) after at least 1 episode of hyperkalemia (potassium levels ≥6.0 mEq/L). We obtained records of all clinical data assessed during office visits (length of follow‐up, symptoms attributed to SPS therapy, weight, blood‐pressure, and heart rate), routine blood tests (including levels of potassium, sodium, calcium, phosphorus, albumin, and creatinine), and records of dietician consultations and hospital admissions.

Estimated GFR (eGFR) was calculated using Modification of Diet in Renal Disease equations.20 CKD stages were defined according to eGFR as previously described21: CKD stage 1 with GFR ≥90 mL/min/1.73 m2, CKD stage 2 with GFR between 60 and 89 mL/min/1.73 m2, CKD stage 3 with GFR between 30 and 59 mL/min/1.73 m2, CKD stage 4 with GFR between 15 and 29 mL/min/1.73 m2, and CKD stage 5 with GFR <15 mL/min/1.73 m2, or permanent renal replacement therapy. We defined LV systolic dysfunction as LV ejection fraction <50%.22

The normal concentration levels for the electrolytes presented here were defined as follows: potassium, 3.5 to 6.0 mEq/L; sodium, 132 to 153 mEq/L; calcium, 8.5 to 10.5 mg/dL; and phosphorus, 2.5 to 4.5 mg/dL.

Results

Patient Characteristics

Of a cohort of 113 patients treated in the CKD clinic who had combined CKD and heart disease, we found 14 patients on RAAS‐I therapy who were treated with sorbitol‐free SPS after at least 1 episode of hyperkalemia (serum potassium ≥6 mEq/L). They were treated for a total of 289 months (median length of follow‐up, 14.5 months; range, 7–47 months). The average interval between clinic visits for a patient was 17.3 ± 4.6 days.

The clinical characteristics of the patients are summarized in Table 1.

Table 1.

Baseline Clinical Characteristics of Patients

	75.9 ± 9.2
Age, y, mean (range)	(41–81)
Sex, male/female	9/5
CKD etiology
DN	11
IgA nephropathy	1
Unknown	2
Initial CKD stage
3	3
4	9
5‐ND	2
Heart disease
HF LV systolic	8
HF LV diastolic	2
Ischemic heart disease without HF	4
NYHA classification stage
1	5
2	4
3	1
4	0
RAAS‐I therapya
ACE‐I alone	2
ARB alone	4
ACE‐I + ARBs	3
ACE‐I + spironolactoneb	1
ARB + spironolactoneb	4

Abbreviations: ACE‐I, angiotensin converting enzyme inhibitors; ARB, angiotensin receptor blockers; CKD, chronic kidney disease; DN, diabetic nephropathy; HF, heart failure; LV, left ventricular; ND, not on dialysis; NYHA, New York Heart Association; RAAS‐I, renin‐angiotensin‐aldosterone system inhibition.

^aAll patients were treated with additional β‐adrenergic blocking agents.

^bSpironolactone daily dose of 25 mg.

CKD Characteristics

CKD characteristics are summarized in Table 1. At the initiation of SPS therapy, 3 patients were in CKD stage 3, 9 patients were in CKD stage 4, and another 2 patients were in CKD stage 5‐nondialysis (ND). During the follow‐up period, 3 patients progressed from CKD stage 4 to CKD stage 5, and 1 patient improved from CKD stage 5‐ND to CKD stage 4. One patient who progressed from CKD stage 4 to CKD stage 5 initiated hemodialysis. Eleven of the 14 patients in this cohort had diabetes mellitus type 2, and diabetic nephropathy was considered as the main cause of CKD in these patients. All 14 patients were anemic and were being treated as needed with intravenous iron and subcutaneous erythropoiesis‐stimulating agents (target hemoglobin levels between 11 to 12 g/dL).

Heart Disease Characteristics

Heart disease characteristics are summarized in Table 1. Eight patients had systolic HF. Of these 8 patients, 6 patients were diagnosed with HF and LV systolic dysfunction that was post‐MI, and 2 patients were diagnosed with HF and LV systolic dysfunction without previous known MI. Two patients had diastolic HF with a long history of chronic hypertension. Four other patients were diagnosed with ischemic heart disease without systolic or diastolic HF. The New York Heart Association Functional Classification for the 10 patients with HF is given in Table 1. All 14 patients in this cohort were treated with beta‐adrenergic blocking agents and loop diuretics.

RAAS‐I Therapy During the Follow‐up

Table 1 summarizes the RAAS‐I therapy during this study. The maximum target daily dosages for ACE‐Is were as follows: ramipril, 10 mg (5 patients); and enalapril, 40 mg (1 patient). The maximum target daily dosages for ARBs were as follows: losartan, 100 mg (6 patients); valsartan, 160 mg (4 patients); and candesartan, 32 mg (1 patient). The spironolactone daily dosage was always kept at 25 mg. Only 3 of the 14 patients did not reach target dosage of 1 RAAS‐I medication. These 3 patients were all on dual RAAS‐I therapy.

Adverse Events and Hospital Admissions

Severe adverse events and hospital admissions are summarized in Table 2. Nine of the 14 patients were admitted to the hospital for a total of 17 admissions. One in‐hospital death occurred because of sepsis complicated by pulmonary edema. Admissions due to cardiac complications included 2 acute MIs and 1 episode of chest pain. Three episodes of pulmonary edema were reported for 2 patients. MI (2 admissions) and severe sepsis (1 admission) were the precipitating factors that explained the 3 episodes of acute pulmonary edema. None of the patients in the cohort was found to have colonic necrosis. One patient was admitted for surgery for acute cholecystectomy following acute cholecystitis and another for cholangitis. Three patients were admitted a total of 9 times for recurrent skin infections. Mild transient gastrointestinal (GI) symptoms that might be attributed to the use of SPS were noted episodically in 4 patients, but none required hospitalization, and symptoms in all patients disappeared within a few days despite continuation of the SPS (Table 2).

Table 2.

Hospitalizations and Gastrointestinal Symptoms During Sodium Polystyrene Sulfonate Therapy

Cause of Hospitalizations	No. of Hospitalizations/Patientsa	Diagnosis	Outcome
Heart disease	3/2	Myocardial infarction	Dischargeb
	1/1	Chest pain	Discharge
GI disease	2/1	Acute cholecystitisc	Cholecystectomy, discharge
	1/1	Acute cholangitisc	Discharge
Non‐GI infections	1/1	Systemic sepsisd	Death
	9/3	Local infections	Discharge
GI Symptoms	No. of Patients	Mean Length of Symptoms, mo	Outcome
Nausea	0	NR	NR
Abdominal pain	1	2	Resolved, SPS therapy continued
Constipation	3	2.3	Resolved, SPS therapy continued

Abbreviations: GI, gastrointestinal; NR, not relevant; SPS, sodium polystyrene sulfonate.

^aA total 17 hospital admission for patients.

^bOne patient was readmitted for coronary angiography.

^cTransient cessation of SPS therapy during hospitalization.

^dCessation of RAAS‐I and SPS therapy during hospitalization.

Weight

During the total of 289 months of follow‐up, the average weight did not change significantly (from 75.8 ± 8.8 kg to 74.4 ± 9.7 kg).

Electrolyte Concentration Levels

Table 3 summarizes the electrolyte concentration levels at the initiation of SPS therapy and the changes in concentration levels during the 289 months of follow‐up with SPS therapy. The mean serum potassium fell from 6.4 ± 0.3 mEq/L (range, 6.0–7.1) to 4.6 ± 0.6 (range, 3.0–5.8 mEq/L), P < 0.01. The mean serum sodium remained unchanged at 139.2 ± 3.5 (range, 135–145) mEq/L to 138.4 ± 3.3 (137–145) mEq/L.

Table 3.

Electrolyte and Hemoglobin Levels During Low‐Dose Sodium Polystyrene Sulfonate Preventive Therapy

Normal Concentration Range	Concentration Levels at Initiation of SPS Therapy, Mean ± Standard Deviation (Range)	Concentration Levels on SPS Therapy, Mean ± Standard Deviation (Range)
Potassium (3.5–5.5 mmol/L)	6.4 ± 0.3 (6.0–7.1)	4.6 ± 0.6 (3.0–5.8)
Calcium (8.5–10.5 mg/dL)	9.2 ± 0.6 (7.8–10.1)	9.1 ± 0.7 (7.6–10.2)
Phosphor (2.5–4.5 mg/dL)	4.4 ± 1.3 (2.9–7.2)	4.7 ± 1.5 (2.5–7.1)
Sodium (132–153 mmol/L)	139.2 ± 3.5 (135–145)	138.4 ± 3.3 (137–145)
Hemoglobin (>12 g/dL)	11.1 ± 0.8 (9.7–12.1)	11.1 ± 0.9 (9.8–12.4)

One episode of mild asymptomatic hypokalemia was noted in 2 patients (potassium, 3.1 and 3.0 mEq/L). We stopped the resin for 1 week until the potassium returned to normal and then reduced the SPS dosage from 15 g daily to 15 g on alternate days, and we noted no recurrence of hypokalemia. No patient developed hyperkalemia after the resin was started. Hypernatremia were not found during the follow‐up period.

Two patients had mild hypocalcemia before initiation of SPS therapy (calcium concentration not lower than 7.8 mg/dL). In both cases, calcium concentration levels did not decrease during the SPS therapy period. Hypocalcemia was not observed during the follow‐up for the other 12 patients. None of the patients developed hypophosphatemia during the follow‐up.

Discussion

We report here our retrospective experience of low‐dose SPS for prevention of a recurrence of hyperkalemia after at least 1 episode of hyperkalemia (potassium levels ≥6.0 mEq/L) in 14 patients treated with RAAS‐I for combined CKD and heart disease. To the best of our knowledge, this is the first follow‐up description in the literature of use of SPS in this subset of patients. Tomino et al reported that another potassium‐binder resin, a jelly preparation of calcium polystyrene sulfonate, was safe and effective in a dose‐dependent manner for CKD patients with previous episodes of hyperkalemia. However, patients with severe heart disease were excluded from that study.23

Overall, low‐dose SPS was well tolerated without severe adverse events attributed directly to its use. Colonic necrosis, a rare complication previously associated with the use of SPS, was not observed. As the preparation used in this follow‐up was sorbitol‐free SPS, this finding is in concordance with previous reports in which colonic necrosis was mainly associated with the use of SPS mixed with sorbitol.14, 15, 16, 17, 18 Colonic necrosis due to SPS therapy has been described often in patients with concomitant severe GI disease.16, 17, 18, 19 It may well be prudent in patients with concomitant severe GI disease or severe constipation to withhold SPS therapy. Hypokalemia with SPS treatment, as well as with other potassium binders, has been well documented.24, 25 In this cohort, an episode of mild asymptomatic hypokalemia was noted in 2 different patients. We chose, in both cases, to stop the resin until the potassium returned to normal and to then reduce the SPS dosage from 15 g daily to every other day; we then noted no recurrence of hypokalemia.

SPS therapy as an ion exchanger may lead to sodium retention. Hypernatremia, edema, and worsening hypertension or acute HF have been previously described.24, 26 In the cohort presented here, hypernatremia was not found in any of the patients, and we could not detect significant weight gain during the follow‐up. Two patients were admitted to the hospital for 3 admissions because of pulmonary edema. In all 3 admissions, precipitating factors for pulmonary edema were evident, and previous attacks of pulmonary edema before the use of SPS had occurred in both patients. Previous reports of sodium retention in patients treated with SPS were mainly in infants or with large doses of the resin.24, 26 We believe that in the subset of patients with HF, it would seem prudent not to increase the low dose of SPS, as high doses might indeed contribute to the development of a positive sodium balance.

Our study has major limitations. Primarily, as a nonrandomized, retrospective study, it lacks the comparison of SPS to placebo in terms of safety and efficacy. The number of patients described here is also not large. Nevertheless, it seems that our experience justifies controlled, randomized studies to further evaluate the safety and efficacy of SPS, as the routine use of low‐dose sorbitol‐free SPS was safe and effective in our cohort.

Our study also lacks routine measurements of magnesium and bicarbonate, as it was not the common practice in our clinic to measure these routinely at the time of this study. Hypomagnesia and systemic alkalosis are rare adverse events that were previously reported with SPS therapy.24, 27 During the follow‐up, none of the patients in our study had clinical signs suggesting overt hypomagnesemia or metabolic alkalosis, and sporadic magnesium and bicarbonate levels did not reveal such impairments (data not shown). Future prospective studies for the assessment of SPS safety should carefully monitor the levels of these 2 electrolytes.

Recently, the safety and efficacy of a novel polymeric potassium binder, RLY5016, were evaluated in the double‐blind, placebo‐controlled study in patients with chronic HF (the Evaluation of RLY5016 in Heart Failure Patients [PEARL‐HF] trial).25, 28 In comparison with placebo, RLY5016 significantly reduced potassium levels, reduced the incidence of hyperkalemia, increased the proportion of patients in whom the dose of spironolactone could be increased, and was relatively well tolerated.25 As a novel agent, RLY5016 may well be substantially more costly than SPS. It is logical, in our opinion, that novel potassium binders such as RLY5016 be compared to the widely used SPS.