CLINICAL COMMENTARY www.jasn.org

Ion-Exchange Resins for the Treatment of Hyperkalemia: Are They Safe and Effective?

Richard H. Sterns, Maria Rojas, Paul Bernstein, and Sreedevi Chennupati

Rochester General Hospital and University of Rochester School of Medicine and Dentistry, Rochester, New York

ABSTRACT Sodium polystyrene sulfonate (SPS), an ion-exchange resin designed to bind po- the FDA were to consider SPS in sorbitol tassium in the colon, was approved in 1958 as a treatment for hyperkalemia by the and Kayexalate powder as new therapeu- US Food and Drug Administration, 4 years before drug manufacturers were re- tic agents today, would they be approved quired to prove the effectiveness and safety of their drugs. In September 2009, as safe and effective? citing reports of colonic necrosis, the Food and Drug Administration issued a Synthetic cation-exchange resins, warning advising against concomitant administration of sorbitol, an osmotic ca- which became available to industry in thartic used to prevent SPS-induced fecal impaction and to speed delivery of resin 1935, are insoluble polymers with a mo- to the colon, with the powdered resin; however, a premixed suspension of SPS in lecular structure resembling a crystal lat- sorbitol, the only preparation stocked by many hospital pharmacies, is prescribed tice to which reactive carboxylic or sul- routinely for treatment of hyperkalemia. We can find no convincing evidence that fonic groups are attached. When a resin SPS increases fecal potassium losses in experimental animals or humans and no is placed in a solvent, the resin’s reactive evidence that adding sorbitol to the resin increases its effectiveness as a treatment groups, which may be preloaded to form for hyperkalemia. There is growing concern, however, that suspensions of SPS in sodium, potassium, calcium, ammo- sorbitol can be harmful. It would be wise to exhaust other alternatives for manag- nium salts, or combinations of these ing hyperkalemia before turning to these largely unproven and potentially harmful ions, exchange their bound cations with therapies. cations dissolved in the solvent.3 Medical applications of synthetic res- J Am Soc Nephrol 21: 733–735, 2010. doi: 10.1681/ASN.2010010079 ins were first explored in 1946.3 At ap- proximately the same time, clinical mea- surements of serum potassium became Kayexalate (sodium polystyrene sulfo- widely. A 6-month single-center survey available and the clinical consequences nate [SPS]), an ion-exchange resin de- from an emergency department reported of “potassium intoxication” were first signed to exchange sodium for potas- 1001 patients who received SPS in sorbi- recognized.4 In an era when hemodialy- sium in the colon, is approved for use in tol for hyperkalemia and only 188 pa- sis was not easily available, the newly ac- the treatment of hyperkalemia. In Sep- tients who received other potassium- cessible resins were embraced as a possi- tember 2009, the US Food and Drug Ad- lowering therapies.2 An unpublished ble treatment for hyperkalemia. ministration (FDA) posted safety label- hospital-wide survey of inpatient prac- Studies of dogs showed that a carbox- ing changes for Kayexalate powder on its tice patterns at our own hospital yielded ylic acid exchange resin in the hydrogen web site. The warning reads, “Cases of similar results with 1800 doses of SPS in cycle causes metabolic acidosis with a colonic necrosis and other serious gas- sorbitol given in the past year, a 30% in- large increase in fecal sodium and potas- trointestinal adverse events (bleeding, crease from a survey 2 years earlier. sium and a reciprocal decrease in urinary ischemic colitis, perforation) have been Does SPS in sorbitol deserve its cur- reported in association with Kayexalate rent popularity? Do benefits from the use. The majority of these cases reported drug outweigh its risks? The FDA first Published online ahead of print. Publication date the concomitant use of sorbitol.. . . Con- approved Kayexalate for the treatment of available at www.jasn.org. comitant administration of sorbitol is hyperkalemia on June 5, 1958, 4 years be- Correspondence: Dr. Richard H. Sterns, Rochester 1 General Hospital, 1425 Portland Avenue, Rochester, not recommended.” fore passage of the Kefauver-Harris Drug NY 14621. Phone: 585-922-4242; Fax: 585-922- Despite this, prepackaged suspen- Amendments, which require drug man- 4440; E-mail: [email protected] sions of SPS in sorbitol, an osmotic laxa- ufacturers to prove the effectiveness of Copyright ᮊ 2010 by the American Society of tive, are still on the market and used their products before marketing them. If Nephrology

J Am Soc Nephrol 21: 733–735, 2010 ISSN : 1046-6673/2105-733 733 CLINICAL COMMENTARY www.jasn.org cation excretion5; however, even at a drugs that were already on the market, concentration rose slightly (0.4 mEq/L) dosage of 30 to 60 g/d (equivalent to 200 ruled Kayexalate powder “effective.” on placebo and did not change during to 400 g/d in humans), a sodium-cycled Even without placebo controls, devel- the course of 12 hours in response to a resin, which exchanges sodium rather opment of hypokalemia in a patient with single dose of 30 g of resin in water, 30 g than hydrogen for potassium, caused no oliguria and hyperkalemia makes it diffi- of resin in 60 g of sorbitol, or 60 g of change in acid-base balance, little in- cult to deny that the resin is effective sorbitol alone.10 crease in fecal potassium losses, and no when taken for several days. Evidence of At the same time, evidence has grown change in urinary potassium excretion or effectiveness in the crucial first 24 hours that mixtures of resin in sorbitol may be plasma potassium concentration.5 of therapy is more tenuous. Hypokalemic harmful.11 By 2005, the FDA had re- In 1953, Evans et al.6 introduced a sul- effects of the resin cannot be distinguished ceived 35 adverse event reports of serious fonate resin charged with sodium, a pre- from the effects of the extremely-low-po- bowel injuries associated with both oral cursor of the modern resin we know as tassium diets and the concurrent adminis- and rectal administration of the mixture, Kayexalate. The investigators noted the- tration of large quantities of dextrose used many of them fatal. Extensive transmural oretical advantages of this preparation, in these studies. infarction of the colon and ileum was ob- because it avoids absorption of ammo- Soon after the introduction of Kayex- served with SPS crystals adherent to the nium, which is metabolized to urea and alate, it was recognized that the agent could mucosa and in luminal debris.11–13 The causes acidosis; however, we were unable cause severe constipation and life-threat- authors of one of the published reports to find data then or now showing that ening intestinal impactions. These obser- identified sorbitol as the culprit, because Kayexalate or its precursors increase fe- vations and the desire to speed delivery of similar bowel lesions are induced in rats cal potassium losses in experimental an- the drug to the colon led to another study with and without uremia by administer- imals. Evans et al. reported uncontrolled in 1961 of administration of resin in sorbi- ing enemas of 70% sorbitol, with or with- data showing potassium binding in the tol, a widely used over-the-counter os- out SPS, but not by giving SPS alone.11 A stool and a hypokalemic effect in four pa- motic laxative; seven patients with oliguria variety of other serious complications tients with renal failure and a single nor- were treated with three daily doses of the from SPS in sorbitol have been reported, mal volunteer.6 On the basis of necropsy resin in sorbitol, and three were given sor- including mucosal lesions in the esopha- studies of one patient who died of renal bitol alone.8 A gradual steady decrease in gus, stomach, and duodenum; fatal failure after several days of resin therapy the serum potassium over 5 days was seen chemical pneumonitis after aspiration; and another patient who died of hyper- in all cases. On the basis of limited data and rectal stenosis as a result of a foreign kalemia 18 hours after the oral adminis- without statistical analysis, the authors body reaction to SPS crystals.14–16 tration of the resin, the authors con- concluded that “sorbitol alone is as effec- Responding to concerns about ad- cluded that most potassium binding tive as a combination.” verse events, recommendations for con- occurs in the colon. This explains why For two decades, FDA-approved la- comitant or postdosing use of sorbitol the hypokalemic effect of the resin is de- beling for Kayexalate powder included were removed from FDA-approved la- layed after the first dose, reflecting the recommendations encouraging its ad- beling for the powdered resin in 2005. time required for resin to reach the co- ministration with sorbitol. In 1982, a According to a verbal communication lon, and prolonged after the last, reflect- convenient premade suspension of SPS (January 4, 2010) with the company’s ing continued delivery of orally ingested in sorbitol was approved for commercial President and its Medical Consultant, resin to the colon. distribution in the United States. A sur- Carolina Medical, the largest manufac- In 1961, Scherr et al.7 reported the vey of Drugstore and Hospital Pharma- turer of premixed suspensions, met with largest clinical experience with Kayex- ceutical purchase data in the United the FDA in 2006 and was allowed to con- alate suspended in water in an uncon- States between 1985 and 1989 showed a tinue marketing its product on the basis trolled study of patients with acute and 52% increase in the overall use of SPS of the following evidence: The Carolina chronic renal failure, using the newly ap- during this period, all of it accounted for formulation contains 33% sorbitol and proved medication provided by the by growth in the use of the prepackaged all reports of adverse gastrointestinal manufacturer, Winthrop Laboratories. suspension.9 By 1989, 62% of SPS was events followed administration of resin In 23 of 30 cases, the plasma potassium purchased as the prepackaged suspen- in 70% sorbitol; at the time, the company fell by at least 0.4 mEq/L in the first 24 sion. Today, in many hospital pharma- had received no adverse gastrointestinal hours. Two patients with pretreatment cies, SPS in sorbitol is the only stocked event reports since the formulations’s hyperkalemia (6.1 and 7.4 mEq/L) devel- preparation. approval in 1982 (and since then only oped hypokalemia (3.3 and 2.3 mEq/L) There is no evidence that adding sor- one adverse event in a critically ill pa- while receiving 40 g/d oral resin for 2 and bitol to potassium-binding resin makes tient), despite extensive use of the prod- 6 days. On the strength of this study and it more effective in correcting hyperkale- uct (approximately 5 million doses year- several smaller case series, the FDA’s mia. Recent studies of patients with nor- ly); and a study of rats conducted by the Drug Efficacy Study Implementation Pro- mokalemia and mild hyperkalemia and company showed that enemas of 70% gram, charged with reviewing pre-1962 with ESRD found the serum potassium sorbitol caused bowel necrosis whereas

734 Journal of the American Society of Nephrology J Am Soc Nephrol 21: 733–735, 2010 www.jasn.org CLINICAL COMMENTARY the 33% did not. The FDA asked all other fit against uncontrolled studies showing exchange resin and sorbitol: A preliminary re- manufacturers of premixed resin to re- harm. It would be wise to exhaust other port. N Engl J Med 264: 111–115, 1961 9. Gerstman BB, Platt R: Use of sodium poly- formulate their products, and this rec- alternatives for managing hyperkalemia styrene sulfonate in sorbitol in the United ommendation has been in force since before turning to these largely unproven States, 1985–1989. Am J Kidney Dis 5: 619– September 2007 (CDER DRUG INFO, and potentially harmful therapies. 621, 1991 [email protected], written com- 10. Gruy-Kapral C, Emmett M, Santa Ana CA, munication, December 3, 2009). Curi- Porter JL, Fordtran JS, Fine KD: Effect of single dose resin-cathartic therapy on serum ously, 70% sorbitol continues to be mar- DISCLOSURES potassium concentration in patients with keted as an over-the-counter osmotic None. end-stage renal disease. J Am Soc Nephrol laxative. 9: 1924–1930, 1998 More recent data call the safety of the 11. Lillemoe KD, Romolo JL, Hamilton SR, Pen- 33% sorbitol formulation into question. nington LR, Burdick JF, Williams GM: Intes- REFERENCES tinal necrosis due to sodium polystyrene A study published in 2009 described 11 (Kayexalate) in sorbitol enemas: Clinical and new cases of colonic necrosis associated 1. US Food and Drug Administration: Kayex- experimental support for the hypothesis. with SPS in sorbitol, four of them fatal, alate (sodium polystyrene sulfonate) pow- Surgery 101: 267–272, 1987 12. Gerstman BB, Kirkman R, Platt R: Intestinal identified over 9 years in a single center.13 der. Available at: http://www.fda.gov/Safety/ MedWatch/SafetyInformation/ucm186845. necrosis associated with postoperative In contrast to previous reports, only two htm. Accessed January 27, 2010 orally administered sodium polystyrene sul- of these cases were postoperative and 2. Joshi P, Beaulieu J, Shemin D: The effect of fonate in sorbitol. Am J Kidney Dis 20: 159– only four of the patients had ESRD. Of a single dose of polystyrene sulfonate (SPS) 161, 1992 13. McGowan CE, Saha S, Chu G, Resnick MB, the fatal cases, three patients were admit- and sorbitol in hyperkalemic patients with kidney disease [Abstract]. J Am Soc Nephrol Moss SF: Intestinal necrosis due to sodium ted with noncritical illnesses and devel- 19: 355A, 2008 polystyrene sulfonate (Kayexalate) in sorbi- oped symptoms of intestinal injury be- 3. Elkington JR, Squires RD, Klingensmith WC: tol. South Med J 102: 493–497, 2009 tween 3 hours and 11 days after receiving Cation exchange resin in the treatment of 14. Idowu MO, Mudge M, Ghatak NR: Kayexalate congestive heart failure. Circulation 5: 747– (sodium polystyrene sulfonate) aspiration. oral SPS in sorbitol for serum potassium 753, 1952 Arch Pathol Lab Med 129: 125, 2005 concentrations ranging from 5.7 to 6.8 4. Finch CA, Sawyer CG, Flynn JM: Clinical 15. Chatelain D, Brevet M, Manaouil D, Yzet T, mEq/L.13 Personal communications from syndrome of potassium intoxication. Am J Regimbeau JM, Sevestre H: Rectal stenosis the author of this report (C. McGowan, Med 1: 337–352, 1946 caused by foreign body reaction to sodium 5. Danowski TS, Greenman L, Mateer FM, Par- polystyrene sulfonate crystals (Kayexalate). December 10, 2009) and from the author sons WB, Weigand H, Mermestein H, Peters Ann Diagn Pathol 11: 217–219, 2007 of another case report of a fatality17 (A. JH: Carboxylic cation exchange resin effects 16. Abraham SC, Bhagavan BS, Lee LA, Rashid Thomas, December 17, 2009), indicate in dogs. J Clin Invest 30: 984–994, 1951 A, Wu TT: Upper gastrointestinal tract in- that at least some of these patients had 6. Evans BM, Hughes Jones NC, Milne MD, Yel- jury in patients receiving Kayexalate (so- lowlees H: Ion-exchange resins in the treat- dium polystyrene sulfonate) in sorbitol: been treated with the 33% formulation. ment of anuria. Lancet 265: 791–795, 1953 Clinical, endoscopic, and histopathologic If Kayexalate or SPS in sorbitol were 7. Scherr L, Ogden DA, Mead AW, Spritz N, findings. Am J Surg Pathol 25: 637–644, presented to the FDA as new drugs with the Rubin AL: Management of hyperkalemia 2001 data available today, it is doubtful that ei- with a cation-exchange resin. N Engl J Med 17. Thomas A, James BR, Landsberg D: Co- 264: 115–119, 1961 lonic necrosis due to oral Kayexalate in a ther would pass muster. Clinicians must 8. Flinn RB, Merrill JP, Welzan WR: Treatment of critically-ill patient. Am J Med Sci 337: weigh uncontrolled studies showing bene- the oliguric patient with a new sodium ion- 305–306, 2009

J Am Soc Nephrol 21: 733–735, 2010 Ion-Exchange Resins for Treatment of Hyperkalemia 735