Kayexalate (sodium polystyrene sulphonate) in sorbitol...

Kayexalate (sodiumpolystyrene sulphonate)

in sorbitol associatedwith intestinal necrosis

in uremic patientsGeoffrey W Gardiner MD FRCPC

Can J Gastroenterol Vol 11 No 7 October 1997 573

GW Gardiner. Kayexalate (sodium polystyrene sulphonate) insorbitol associated with intestinal necrosis in uremic patients.Can J Gastroenterol 1997;11(7):573-577.

BACKGROUND: Kayexalate (sodium polystyrene sulphonate)in sorbitol is commonly used to treat hyperkalemia in patientswith renal insufficiency. Isolated case reports and one recent largeseries have documented intestinal necrosis following administra-tion of kayexalate in sorbitol.METHODS: Two patients with luminal kayexalate crystals asso-ciated with intestinal pathology were first identified in thepathology department, and clinicopathological correlation wascarried out.RESULTS: Both patients were seriously ill, had prior cardiac sur-gery and were in renal failure (uremic). Examination of autopsyand colonic resection showed luminal kayexalate crystals associ-ated with underlying mucosal necrosis, submucosal edema andtransmural inflammation.CONCLUSION: Although occurring in complex clinical set-tings, the pathological findings provide additional evidence thatkayexalate in sorbitol may be associated with intestinal necrosisand inflammation in uremic patients and that this may be a clini-cally and pathologically under-recognized iatrogenic bowel in-jury.

Key Words: Intestinal necrosis, Kayexalate, Renal failure, Sodium

polystyrene sulphonate, Sorbitol, Uremia

Kayexalate (sulfonate de polystyrène sodique)dans le sorbitol associé à la nécrose intestinalechez les patients urémiques

DONNÉES DE DÉPART : Le kayexalate (Sanofi Winthrop [sulfo-nate de polystyrène sodique]) dans le sorbitol est couramment utilisépour traiter l’hyperkaliémie chez les insuffisants rénaux. Des rapportsde cas isolés et une importante série récente ont permis de documenterdes cas de nécrose intestinale suivant l’administration de kayexalatedans le sorbitol.MÉTHODES : On a d’abord identifié deux patients présentant descristaux de kayexalate associés à une pathologie intestinale au départe-ment de pathologie et on a établi une corrélation clinico-pathologique.RÉSULTATS : Les deux patients étaient de grands malades, avaientsubi une chirurgie cardiaque et souffraient d’insuffisance rénale(urémie). L’examen d’une résection du côlon à l’autopsie a révélé laprésence de cristaux de kayexalate associés à une nécrose sous-jacentede la muqueuse, à un œdème sous-muqueux et à une inflammationtransmurale.CONCLUSION : Bien qu’ils soient survenu dans des contextes clini-ques complexes, les signes pathologiques offrent une preuve addition-nelle du fait que le kayexalate dans le sorbitol peut être associé à lanécrose et à l’inflammation intestinale chez les patients urémiques, etcela pourra se révéler être une lésion intestinale iatrogène mal identi-fiée sur le plan clinique et pathologique.

St Michael’s Hospital, Toronto, OntarioCorrespondence and reprints: Dr Geoffrey W Gardiner, St Michael’s Hospital, Department of Pathology, 30 Bond Street, Toronto, Ontario

M5B 1W8. Telephone 416-864-5858, fax 416-864-5870Received for publication April 15, 1997. Accepted July 2, 1997

BRIEF COMMUNICATION

Kayexalate is a powdered form of sodium polystyrene sul-phonate that is usually suspended in sorbitol solution

and given orally or as a retention enema for the treatment ofhyperkalemia in patients with renal insufficiency (uremia)(1). Kayexalate acts as a cation exchange resin in which so-dium ions are released and replaced with potassium ions withsubsequent loss in the stool lowering serum potassium. Sorbi-tol is a poorly absorbed sugar that is primarily degraded bycolonic bacteria and acts as an osmotic laxative to preventconstipation and fecal impaction.

Lillimoe et al (2) first reported colonic necrosis after ad-ministration of kayexalate sorbitol enemas in uremic pa-tients. This was followed by isolated case reports (3,4).Recently Rashid and Hamilton (5) reported pathologicalfindings in a series of 15 cases and concluded that this condi-tion is under-recognized by both clinicians and pathologists.This report confirms and illustrates these findings with twocases, neither of which were suspected clinically.

CASE PRESENTATIONSCase 1: A 66-year-old man underwent an aortic valve re-placement with uncomplicated surgery and immediate post-operative period. On the fourth postoperative day he devel-oped a sudden heart block with cardiogenic shock and wassuccessfully resuscitated but became acidotic and hypoglyce-mic after return to the intensive care unit. Acute liver failure(ischemic hepatitis and shock liver) developed with aspar-tate aminotransferase rising to 8129 U/L (normal less than40 U/L and alanine aminotransferase rising to 4587 U/L (nor-mal less than 35 U/L). Simultaneously, renal failure ensuedwith creatinine rising to 415 µmol/L (normal less than120 µmol/L) and urea rising to 23 mmol/L (normal 3 to7 mmol/L). A presumed metabolic neurological coma alsodeveloped. Postresuscitative blood pressure remained sta-ble, and no gastrointestinal bleeding or peritoneal signs de-veloped. Marked hyperkalemia with potassium rising to6.8 mmol/L (normal 3.5 to 5 mmol/L) necessitated the vigor-ous use of kayexalate enemas and large doses of kayexalatesorbitol solutions (240 g over two days) through a nasogastrictube. The patient died on the 10th postoperative day, six daysafter the episode of cardiogenic shock.

Necropsy confirmed shock liver (ischemic hepatitis) withcoagulative necrosis in zones II and III of the microcircula-tory acinus, and kidneys showed diffuse ischemic tubulardamage. Examination of the gastrointestinal tract foundpatchy hemorrhagic mucosal erosions of the stomach, ileumand colon (Figure 1). Light microscopy showed coagulativenecrosis of the mucosa with overlying purple rhomboidkayexalate crystals, submucosal edema and acute transmuralinflammation (Figures 2,3). The muscularis propria showedacute inflammation and myocytolysis but no frank coagula-tive necrosis. The aortic ostia of the mesenteric arteries werepatent with no evidence of atherosclerotic occlusion or ve-nous thrombosis.Case 2: A 71-year-old woman was transferred to hospital be-cause of chronic lower gastrointestinal bleeding (hemoglo-bin 76 g/L requiring transfusion of 8 U of blood over two

weeks. Two years previously she had undergone an aortic andmitral valve replacement with a difficult postoperativecourse followed by chronic renal failure with unstable serumpotassium. In addition, she had mild insulin-dependent dia-betes that was well controlled. The patient was receiving di-goxin, furosamide, warfarin sodium and kayexalate 15 g/dayorally. Blood urea nitrogen was 25.7 mmol/L and creatinine238 �mol/L. Mesenteric angiogram showed changes sugges-tive of angiodysplasia but with no well developed vascular

574 Can J Gastroenterol Vol 11 No 7 October 1997

Gardiner

Figure 1) Colon from case 1 autopsy showing segmental ulceratedplaques with adherent granular debris (kayexalate)

Figure 2) Histophathology of colon (case 1) showing sloughed colonicmucosa with impacted kayexalate crystals and acute inflammation in thesubmucosal plane (hematoxylin and eosin stain)

tuft or active bleeding. A right hemicolectomy was per-formed. Postoperatively, the patient’s course was unstable re-quiring hemodialysis, and she died on the 18th postoperativeday.

The right hemicolectomy consisted of 45 cm of colon

with a short length of ileum. Examination of the mucosarevealed focal mucosal hemorrhage with erosion and under-lying edematous bowel wall (Figure 4). Histological exami-nation showed hemorrhagic mucosal necrosis associatedwith kayexalate crystals, submucosal edema and acute trans-


Kayexalate in sorbitol

Figure 4) Right hemicolectomy specimen from case 2. Centrally there isfocal mucosal ulceration and granularity

Figure 5) Histopathology of case 2 colon showing mucosal necrosis anderosion with overlying kayexalate crystals

Figure 6) Histopathology of case 2 colon showing kayexalate crystalsover normal mucosa with submucosal edema and acute inflammation

Figure 3) Histopathology of a stomach (case 1). Necrotic and acutelyinflamed gastric mucosa with underlying thrombosed vessels and overly-ing kayexalate crystals

mural inflammation (Figures 5,6). Coagulative necrosis ofmuscle was not evident. Submucosal blood vessels werewithin normal limits for her age, and no dilated or throm-bosed vessels transversing the muscularis mucosae at sites oferosion were identified.

DISCUSSIONMany gastrointestinal complications in patients with renalfailure are reported and include peptic ulcers (6), bleeding orperforated colonic ulcers (7-9), and complications of pre-existing disease such as perforated diverticula (9). Ischemiccolitis is the most common colonic complication in renaltransplant patients, occurring in about 1% of patients (10).Electrolyte imbalance, particularly hyperkalemia, is a fre-quent complication of renal failure and its treatment withkayexalate in sorbitol may rarely result in intestinal necrosis.In an epidemiological study estimating intestinal necrosis inpostoperative patients receiving sodium polystyrene sulpho-nate, Gerstman et al (11) found an incidence of 1.8% andconcluded that sorbitol-associated complications may be anot uncommon occurrence in postoperative patients. Fur-thermore, it has been suggested that some cases of idiopathiccolonic ulcers in patients with renal failure represent kayex-alate in sorbitol-induced injury (5).

Although the precise mechanism of kayexalate in sorbitolnecrosis is uncertain, contributing factors may include ure-mia, hypovolemia and hypotension after dialysis or surgery.Indeed, elevated renin levels in uremic patients may predis-pose to angiotension-mediated mesenteric vasoconstrictionwith resulting intestinal ischemia (12). Certainly the majorclinical and morphological differential diagnoses includeischemic injury of the bowel in uremic patients.

Lillimoe et al (2) first reported five patients presentingwith extensive mucosal necrosis and transmural infarction ofthe colon following kayexalate and sorbitol enemas to treathyperkalemia in uremic patients. They further investigatedthe effects of kayexalate sorbitol enemas in normal and ure-mic rats and concluded that sorbitol is the agent responsiblefor colonic damage and that the injury was potentiated inuremic rats. In this study, when sorbitol alone or kayexalatesorbitol was given, extensive transmural necrosis developedin 80% of normal rats and all uremic rats. The pathogenesisis unclear but it was speculated that the osmotic load couldcause vascular shunting, resulting in intestinal ischemia, orthat the osmotic load from sorbitol could cause direct toxicdamage by disrupting mechanisms that regulate cell volume.While administered as a poorly absorbed sugar, sorbitol isalso an organic osmolyte (polyol) found in high concentra-tion in cell cytosol with a key role in cell volume homeostasis(13). Cellular membrane transport svstems involving potas-sium and sodium are also integral parts of cell volume regula-tory control (13).

While kayexalate crystals can be an incidental findingand are not known to cause injury, the purple irregular jag-ged crystals are a helpful histological clue to the possibilitythat sorbitol, the agent responsible for colonic necrosis, hasbeen administered. The presence of crystals should alert the

pathologist to discuss the patient’s case with the clinician. Itmust be remembered that kayexalate crystals can be foundover normal mucosa and the differential diagnosis would in-clude cholestyramine resin used in the treatment of hyper-cholesterolemia, bile-acid induced diarrhea or Clostridium

difficile colitis. Kayexalate crystals are reported to be less ba-sophilic and opaque than cholestyramine crystals and stainred with acid-fast stain (5).

Although varying in degree, the histopathological pat-tern of injury was similar in both patients, with mucosal ne-crosis and ulceration with overlying crystals and submucosaledema and transmural inflammation. The muscularis propriashowed edema and acute inflamation with myocytolysis andwas reminiscent of changes seen in toxic megacolon. Thepresence of spreading submucosal inflammation and edemaalso warrants the consideration of phlegmonous colitis (14)and other types of infectious enterocolitis including Escher-

icia coli O157:H7 that could be diagnosed by appropriatestool cultures. Kayexalate crystals are the distinguishing clueto etiology.

In case 1, because of a time interval of six days after tran-sient cardiogenic shock and absence of signs of acuteischemic bowel injury, superficial intestinal necrosis was at-tributed to kayexalate in sorbitol etiology. Large doses ofkayexalate in sorbitol were given through a nasogastric tube,and autopsy revealed patchy gastric mucosal necrosis (Figure3) with overlying kayexalate crystals. This finding supportsthe recent series from Johns Hopkins University (5) with thefirst recorded observation that gastrointestinal injury mayoccur in the stomach and small intestine, and not exclu-sively in the colon. Although it could be argued that thischange is incidental, because patients with uremia have ahigher incidence of gastritis, and gastric and duodenal ero-sions (15-17), crystals and underlying mucosal necrosiscould be attributed to a kayexalate in sorbitol etiology. Ourhospital pharmacy restricted the use of sorbitol in enemasbased on earlier adverse reports (2-4), but the findings incase 1 confirm the presence of gastric and small intestinal in-jury as a possible adverse drug effect, and the restriction oforal sorbitol administration should be considered.

Case 2 presented with intermittent lower gastrointestinalbleeding. Right-sided diverticula and angiodysplasia arecommon causes of bleeding in uremic patients (18), andthrombocytopenia and coagulation defects frequently pres-ent in uremia may also contribute to a bleeding diathesis(19). Angiodysplasia is an extremely difficult lesion for thepathologist to identify in a colectomy specimen, and intra-vascular injection techniques are often required for diagnosis(20). More practical alternative methods have also been re-ported (21). Although it is difficult to totally exclude this le-sion, no dilated venules were found traversing the muscularismucosas in the erosions, and colonic bleeding was attributedto kayexalate in sorbitol necrosis. Kayexalate in sorbitol wasgiven for chronic hyperkalemia in small doses orally.

It is unclear whether toxic effects are dose-related orwhether delayed peristalsis enhances toxicity. Chronic con-stipation is a common problem relating to medication use in

576 Can J Gastroenterol Vol 11 No 7 October 1997

Gardiner

chronic renal failure. Furthermore, the precise details ofkayexalate in sorbitol administration are often difficult to as-sess or not available.

CONCLUSIONThese two cases serve to illustrate that nephrologists, gastro-enterologists and pathologists must be alerted to the fact thatroutine therapy for hyperkalemia rarely results in significantiatrogenic bowel injury in uremic patients.

ACKNOWLEDGEMENTS: The author thanks Ms Sandra Co-hen for photographic assistance, Ms May Wong for typing themanuscript and Dr R Warren for reviewing the manuscript.

REFERENCES1. Gerstman BB, Platt R. Use of sodium polystyrene sulfonate in sorbitol

in the United States 1985-1989. Am J Kidney Dis 1991;18:619-20.2. Lillimoe KD, Romolo JL, Hamilton SR, Pennington LR, Burdeck JF,

Williams GM. Intestinal necrosis due to sodium polystyrene(Kayexalate) in sorbitol enemas: clinical and experimental support forthe hypothesis. Surgery 1987;101:266-72.

3. Wootton FT, Rhodes DF, Lee WM, Fitts CT. Colonic necrosis withKayexalate – sorbitol enemas after renal transplantation. Ann InternMed 1989;111:947-9.

4. Scott TR, Graham SM, Schweitzer EJ, Bartlett ST. Colonic necrosisfollowing sodium polystyrene sulfonate (Kayexalate) – sorbitol enemain a renal transplant patient: report of a case and review of theliterature. Dis Colon Rectum 1993;36:607-9.

5. Rashid A, Hamilton SR. Necrosis of the gastrointestinal tract inuremic patients as a result of sodium polystyrene sulfonate(Kayexalate) in sorbitol. Am J Surg Pathol 1997;21:60-9.

6. Kang JY, Wu AYT, Sutherland H, Vathsala A. Prevalance of pepticulcer in patients undergoing maintenance hemodialysis. Dig Dis Sci1988;33:774-8.

7. Huded FV, Posner GL, Tick R. Nonspecific ulcer of the colon in achronic hemodialysis patient. Am J Gastroenterol 1982;77:913-6.

8. Mills B, Zuckerman G, Sicard G. Discrete colon ulcers as a cause oflower gastrointestinal bleeding and perforation in end-stage renaldisease. Surgery 1981;87:548-52.

9. Scheff RT, Zuckerman G, Harter H, Dalnez J, Koehler R. Diverticulardiseases in patients with chronic renal failure due to polycystic kidneydisease. Ann Intern Med 1980;92:202-4.

10. Flanigan RC, Reckard CR, Lucas BA. Colonic complications of renaltransplantation. J Urol 1988;139:503-6.

11. Gerstman BB, Kirkman K, Platt R. Intestinal necrosis associated withpostoperative orally administered sodium polystyrene sulfonate insorbitol. Am J Kidney Dis 1992;20:159-61.

12. Reilly PM, Buckley GB. Vasoactive mediators and splanchnicperfusion. Crit Care Med 1993;21:555-68.

13. McManus ML, Churchwell KB, Strange K. Regulation of cell volumein health and disease. N Engl J Med 1995;333:1260-6.

14. Rosen Y, Won DM. Phlegmonous enterocolitis. Dig Dis Sci1978;23:248-56.

15. Margolis PM, Saylor JC, Geisse G, De Schryver-Keckemeti K,Harter HR, Zuckerman GR. Upper gastrointestinal disease in chronicrenal failure: a prospective evaluation. Arch Intern Med1978;138:1214-7.

16. Franzin G, Musola R, Mencarelli R. Morphological changes of thegastroduodenal mucosa in regular dialysis uremic patients.Histopathology 1982;6:429-37.

17. Mitchell CJ, Jewell DP, Lewin MR, McLaughlin JE, Moorhead JF.Gastric function and histology in chronic renal failure. J Clin Pathol1979;32:208-13.

18. Flynn CT, Chandran PKG. Renal failure and angiodysplasia of thecolon. Ann Intern Med 1985;103:154.

19. Andracry K. Uremia as a cause of bleeding. Am J Nephrol1985;5:313-9.

20. Mitsudo SM, Boley SJ, Brandt LJ, et al. Vascular ectasia of the rightcolon in the elderly: A distinct pathologic entity. Hum Pathol1979;10:585-600.

21. Thelmo WL, Vetrano JA, Wibowo A, et al. Angiodysplasia of colonrevisited: Pathologic demonstration without the use of intravascularinjection technique. Hum Pathol 1992;23:37-40.


Kayexalate in sorbitol

Submit your manuscripts athttp://www.hindawi.com

Stem CellsInternational

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Immunology ResearchHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinson’s Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttp://www.hindawi.com

Kayexalate (sodium polystyrene sulphonate) in sorbitol...

Documents

Transcript of Kayexalate (sodium polystyrene sulphonate) in sorbitol...