- Email: [email protected]
Sorbents in the management of uremia
Sorbents in the Management of Uremia
ELI A. FRIEDMAN, M.D. Brook!m
New
York
From the Departmentof Medicine, Downstate Medical center, 450 Clarkson Avenue, Brooklyn. New York 11203.
Requests for reprints
shouldbe addressed to Dr. Eli A. Friedman. ManuscriptacceptedSepbmtmr 19, 1976.
614
May 10, 1976
Once Pendleton and West [l] documented facile transfer of urea from plasma into the intestinal lumen, the potential for treating renal failure by extraction of nitrogen waste from the gut became self evident. Urea in the gut is degraded to ammonia by bowel bacteria to the extent that normal human feces contain no urea. In healthy volunteer subjects given an “antibiotic cocktail,” stool urea concentrations increase to blood levels while the fecal ammonia content decreases: this supports the inference that urea in the bowel is biodegraded by luminal microorganisms [2,3]. Estimates of the quantity of urea converted to ammonia in the gut have been computed by Jones and co-workers [4] (from 4 to 7 g/day in normal patients and from 17 to 50 g/day in uremic patients). Perhaps the earliest clinical attempt to treat uremic patients by lavage of nitrogen-containing intestinal wastes was reported by Kolff in his 1947 text “New Ways of Treating Uremia” [5]. Kolff created a double ended ileostomy in an isolated iteal loop with an intact blood supply in a 57 year old uremic man. As much as 0.48 g of urea/hour was removed by the patient who performed home intestinal dialysis assisted by his wife for two months until his death. Prolongation of life by intestinal dialysis has been accomplished in dog [6] and man [7,8]. Clearances during isolated jejunal loop dialysis in uremic patients were reported by Schloerb [7,8] as 5 to 10 ml/min for creatinine and 3.2 to 5.0 ml/min for uric acid, values about one third and one eighth as efficient, respectively, as obtained with peritoneal and hemodialysis. Intestinal perfusates contain smaller but significant amounts of larger molecules including aldosterone and 17-oxyhydroxycorticosteroids. By 1960, when repetitive hemodialysis was made practical by the development of an external plastic arteriovenous shunt, a slowly increasing number of patients were sustained (albeit suboptimally) by periodic intestinal dialysis. Analyzing 15 cases in the literature plus five of their own, Clark et al. [9], in 1962, concluded that intestinal dialysis “remains the best method of adjunctive management of progressive uremia.” Thereafter, the improving success rate of maintenance hemodialysis diverted interest in the quest for nitrogenous waste extraction from the gut.
The American Journal of Medicine
Volume 80
r
CHARCOAL
Attention was redirected to the value of intestinal nitrogenous waste removal by the work of Yatzidis [lo] in 1964 in which uremic patients were fed charcoal as an oral sorbent, with highly encouraging results. Sorbents are chemicals varying from single elements, such as charcoal, to complex polymers which find or adsorb gases, liquids or dissolved substances on their surface in a condensed layer. Administered in oral doses of 20 to 50 g daily, with or without sorbitol as a vehicle, Yatzidis [lo] was able to manage patients with end-stage renal failure for four to 20 months without resorting to dialytic methods. A relatively high residual renal function (creatinine clearance of 10 to 15 ml/min) in this group of patients, who were also treated with a 20 g restricted protein diet [ 111, makes it difficult to discern any specific merit of charcoal treatment of uremia. Other investigators have not been favorably impressed with the benefit of charcoal ingestion in uremia. In healthy volunteer subjects given daily oral doses of 50 g of powdered activated charcoal, nausea, vomiting and constipation were of sufficient severity to interrupt further trial [ 121. Maxwell and co-workers administered “various forms of carbon in doses up to 90 g daily” to chronically uremic dogs and found no evidence of “significant effect on the lowering of any of the metabolites associated with the uremic state.” Unless comparable protocols are employed, varying results in the assessment of ingestible sorbents by different researchers may be attributable to dissimilar experimental designs. The palatability of charcoal suspension, for example, is remarkably enhanced by chilling and flavoring with a drop of lemon juice permitting trials of at least four weeks. At present, the case for oral charcoal as even adjunctive therapy in renal failure must be judged “unproved.” OXIDIZED STARCH Promising additional data indicating that gastrointestinal sorbents can bind to and remove in feces clinically important amounts of nitrogenous wastes are provided by a series of investigations using ,oxidized starch (oxystarch) and oxidized cellulose (oxycellulose) performed by Giordano and associates [ 1416]. Corn starch or potatoe starch suspended in a solution of sodium periodate at 4’C for 24 hours slowly oxidizes to dialdehyde starch (oxystarch); cellulose treated similarly, oxidizes to oxycellulose. At body pH and temperature, each repeat unit of oxystarch (Figure 1) binds 1.5 to 1.9 moles of ammonia in vitro in an 0.3N ammonia solution; when present in excess, oxystarch will bind all the ammonia in an 0.3N solution. Oxystarch also adsorbs urea and L-
H O\
/
/c\o/ i” 0 Figure 1.
1n
Oxystarch (dialdehyde starch)
aspartic acid in vitro, but it does not bind creatinine, uric acid, L-lysine or albumin. In weanling mice fed 2 per cent oxystarch in a casein diet, growth and development are normal. In rats fed 2 per cent oxystarch in a casein diet severe diarrhea develops, whereas dogs tolerate as much as 5 per cent dietary oxystarch without apparent adverse effect. Explosive diarrhea and a cholera-like fluid and electrolyte depletion syndrome occur in dogs fed more than 10 per cent of oxystarch in their diets [ 171. In uremic patients fed 20 to 35 g of oxystarch in divided doses stool volume increases by 200 to 600 ml/day and the frequency of bowel movements is increased, but frank diarrhea does not develop [ 181. Giordano’s initial trials of oxystarch manufactured in his laboratories showed that uremic patients (creatinine clearances of 0.4 to 3.2 ml/min) tolerated divided doses of 20 g/day well for two months and that, in each case, there was a significant fall in the blood urea nitrogen level. Fecal nitrogen content increased a mean of 1,450 mg/day (range 730 to 8,050 mg/day) [ 14,151. Confirmation of increased stool nitrogen content during oxystarch treatment was provided by a double blind starch-oxystarch full balance study reported by Friedman et al. 1181. In this study seven uremic patients (creatinine clearances of 6 to 30 ml/min) were fed 29 g of oxystarch or starch daily in four equal doses added to a diet containing 40 to 50 g of protein and 2 to 4 g of salt. Blood urea nitrogen levels fell 33 per cent during oxystarch treatment from a mean of 93.1 mg/ 100 ml to a mean of 62.1 mg/lOO ml. There was no significant change in serum creatinine, plasma amino acid, uric acid and plasma glucose levels during oxystarch ingestion. Oxystarch significantly increased fecal nitrogen from a control mean of 1.4 g/24 hours to 2.5 g/24 hours. A concomitant decrease in urinary nitrogen excretion, however, from a control mean of 7.6 g/24 hours to 5.5 g/24 hours during oxystarch treatment prevented development of negative nitrogen
May 10, 1976
The
American Journal of Yediclne
Volume 60
615
SORBENTS IN THE MANAGEMENT OF UREMIA-FRIEDMAN
CONTROL
OXYSTARCH
C7
-
Meon
-
M*m
0XYSTA.W~
f
Figure 2. During minimal nitrogen ingestion, uremic patients fed oxystarch (28 g) daily have an increased fecal excretion of nitrogen and potassium, and a dounterbalancing decrease in urinary nitrogen and potassium excretion. (Reproduced with permission from Friedman et al. [ 181.)
balance. There was a significant increase in the fecal potassium content when oxystarch was ingested, ranging from 5 to 22 meq/day, which was also counterbalanced by a decrease in urinary potassium excretion. To exclude the possibility that increased fecal nitrogen content noted during oxystarch treatment was due to direct binding of unabsorbed undigested dietary nitrogen rather than bound intraintestinal nitrogen wastes (ammonia [ 191 and urea), four uremic patients were fed oxystarch while ingesting a “no protein” diet [ 181. In these patients (Figure 2) the increases in fecal nitrogen and potassium were similar to those in the previous group indicating that the origin of the extra fecal nitrogen was indeed nitrogenous wastes. Will feeding oxystarch to uremic patients have clinical import? In urine producing patients the counterbalancing decrease in urinary nitrogen excretion tends to mitigate the benefit of increased fecal nitrogen content. What will be the effect on nitrogen balance in anuric patients? Sketchy data on one uremic patient in whom the frequency of peritoneal dialyses was extended from once in seven days to once in 25 days and another who had thrice weekly hemodialysis reduced to once weekly during oxystarch therapy have been reported by Giordano et al. [ 141. Because the prescription for “adequate dialysis” is based on a combination of objective (serum creatinine and serum phosphorous levels, for example) and subjective (work tolerance, ability to sleep through the night, bone pain are examples) criteria, any therapy
616
May 10,
1976 The American Journal of Medlclne
purported to reduce the frequency or duration of dialyses must be evaluated according to a carefully controlled study. It may be anticipated that such a study, as well as a balance study of the effect of charcoal ingestion by uremic man, will be completed within the next few years. One oral sorbent, aluminum hydroxide gel, is presently in clinical use in renal failure to bind phosphate in the gut. Combined with supplemental oral calcium ingestion, in uremic patients treated with aluminum hydroxide gel, both serum phosphorous and elevated parathyroid hormone levels may be lowered to normal [ 201. Exploitation of newer technics of microencapsulation [21] opens the way to possible human use of potent sorbents previously excluded from consideration because of inherent or unknown toxicity (zirconium salts, for instance). While awaiting completion of sorbent trials in functionally anephric (undergoing dialytic maintenance) patients, several helpful animal experiments have been completed. To date, chronically uremic animals in need of dialysis have not been sustained by sorbents alone. Gavage feeding of oxystarch alone or in combination with charcoal will prolong the life of anephric rats from a control of three days to five days [ 221. Saltzman and co-workers [ 231 investigated the mechanism of sorbent-induced life extension in bilaterally nephrectomized rats fed charcoal (1 g daily), oxystarch (1 g daily) or oxystarch plus charcoal (1 g of each daily). In sorbent-treated rats the increase in blood urea concentration was less than in untreated nephrectomized controls, but they also had lower serum potassium concentrations throughout their increased life span. Both actions of oxystarch, increased fecal excretion of nitrogen and potassium, were detectable in this in vivo model of fatal acute renal failure. HEMOPERFUSION For completeness in this survey, the technic of direct exposure of blood to a sorbent, termed hemoperfusion, will be mentioned. During a single passage over granular activated charcoal, creatinine, uric acid, indican, phenols, guanidines and organic acids are nearly totally extracted from the blood. Only negligible quantities of urea, magnesium and phosphate are adsorbed on charcoal from blood. Each gram of powdered charcoal will in vitro bind simultaneously 9 mg of creatinine, 8 mg of uric acid, I.75 mg of phenols, 1 mg of guanidines and 35 mg of urea. In 1964, Yatzidis [24] devised a hemoperfusion device containing 200 g of activated charcoal in a siliconized glass cylinder 20 cm in length and 6 cm in diameter. Based on an experience of 20 human hemoperfu-
Volume 60
SORBENTS IN THE MANAGEMENT
sions at a blood flow rate of about 140 ml/min, Yatzidis estimated that 60 minutes of blood exposure to two or three 200 g charcoal columns had about the same extraction efficiency (for uremic patients) as a 4 to 6 hour hemodialysis. Hemoperfusion was advocated for the treatment of renal insufficiency, gouty arthritis, and intoxications with salicylates, barbiturates and glutethimide [ 241. Early attempts to introduce charcoal hemoperfusion into clinical management of uremic patients were unsuccessful. Of three uremic patients treated with hemoperfusion by Dunnea and Kolff [25], one died after nine perfusions and the other two were switched to maintenance hemodialysis. Noteworthy in this report is the exceptionally high creatinine clearance of 388 ml/min determined in vitro at a blood rate of 400 ml/min. Dual hazards, extraction of formed elements from blood and discharge of charcoal microemboli into efferent blood, complicate the exposure of blood to untreated charcoal. Efforts to eliminate the disadvantages of uncoated charcoal while retaining charcoal’s extraordinary adsorbing properties have concentrated on coating charcoal particles to eliminate the trapping and destruction of platelets, cells and fibrinogen. Chang et al. [26-291 have systematically studied several approaches to microencapsulation of charcoal and they developed a blood-compatible albumin-complexed polymer. In its present form, Chang’s microencapsulated “kidney” contains 300 g of double coated charcoal granules 2 to 5 mm in diameter with a surface area of 2.25 m2. This device achieves clearances in vitro, which are superior for middle molecules (MW 200 to 1,800), to hollow fiber, coil or parallel flow hemodialyzers [ 301. Periodic hemoperfusion as a sole treatment for uremia is inadequate because of the need to extract water and probably urea to maintain acceptable morbidity. Combination of hemoperfusion within series ultrafiltration of blood 1311 for water removal might prove a workable therapy for uremic patients who ingest ammonia binding sorbents such as oxidized starch. The efficiency advantage (shorter treatments) of hemoperfusion over hemodialysis in renal failure is sufficiently attractive to consider for-
mulation of regimens oral sorbent ingestion.
coupling
OF UREMIA-FRIEDMAN
hemoperfusion
with
DIALYSATE REGENERATION Most of the bulk of early hemodialysis systems was taken up by a tank to store a 100 to 180 liter dialysate reservoir. Newer compact disposable hollow fiber, coil and parallel flow hemodialyzers have permitted designs for dialysate delivery systems which “regenerate” dialysate by passage through a sorbent cartridge. A remarkable reduction in dialysate volume to 1.5 liters was devised by Gordon and co-workers 132,331 by the clever means of converting urea in dialysate to ammonium ion and carbonate by urease treatment. Ammonium ion is adsorbed by zirconium phosphate which also extracts calcium and magnesium necessitating continuous reinfusion of these ions. Zirconium oxide binds phosphate and fluoride while charcoal extracts uric acid, creatinine, guanidines, organic acid and phenols. The total weight of the sorbent cartridge is less than 2 kg. The importance of the Gordon system is not only the size reduction, which makes a suitcase travel “artificial kidney” practical, but also its clear demonstration of the value of sorbents in simplification of the therapy of uremia. In closing, it is appropriate for the nephrologist to express respect and admiration for the metabolic events occurring within the gastrointestinal tract. It has been suggested that fed the right combination of essential amino acids [34], or their a-keto precursors [35], uremic patients may actually anabolize nitrogen in urea [36]. More recent balance studies indicate that less than 6 per cent of urea might be recycled thusly [37]. Even more startling to conventional thinking is the work of Setala et al. [38], who believe that the daily ingestion of enzymes derived from soil bacteria can (and does) convert end products of protein digestion within the gut, such as creatinine and urea, into usable recycled metabolites. Uremia might thus be treated without hemodialysis, hemoperfusion or sorbent ingestion. No wonder that the nephrologist occasionally regards the gut as a displaced, hypertrophied, but very “intact nephron,”
REFERENCES 1.
2.
3.
Pendleton WR, West FE: The passage of urea between the blood and the lumen of the small intestine. Am J Physiol 101: 391.1932. Wrong 0, Metcalfe-Gibson A, Morrison RBL, Ing ST, Howard AV: In vivo dialysis of faeces as a method of stool analysis. I. Technique and rest&s in normal subjects. Clin Sci 26: 357, 1965. Wrong 0. Houghton BJ, Richards P, Wilson DR: Intestinal urea in normals and in uremic patients in urea and the kidney. (Schmidt-Nielsen B, ed), Amsterdam, Excerpta
4.
5. 6.
7.
May 10, 1976
Medica Foundation, 1970, p 46 1. Jones EA, Smallwood RA. Craigie A, Rosenoer VM: Enterohepatic circulation of urea nitrogen. Clin Sci 37: 825, 1969. Kolff WJ: New Ways of Treating Uremia. London, J. A. Churchill, 1947. Sloan H: Effect of perfusion of isolated intestinal loops on experimental uremia in dogs. Am J Physiol 166: 137, 1951. Schloerb PR: The management of uremia by perfusion of
The American Journal of Medicine
Volume 60
617
SORBENTS IN THE MANAGEMENT OF UREMIA-FRIEDMAN
8. 9.
10. 11. 12.
13.
14.
15.
16.
17.
18.
19.
20.
21. 22.
23.
618
an isolated jejunal segment. J Clin Invest 37: 1818, 1958. Schloerb PR, Lukart BL: Further observations on the isolated human jejunum. Arch Surg 86: 356, 1963. Clark JE, Templeton JY Ill, McCurdy DK: Perfusion of isolated intestinal loops in the management of chronic uremia. Trans Amer Sot Artif Int Organs 8: 246, 1962. Yatzidis H: Recherches our lepuration extra renale a I’aide do charbon actif. Nephron 1: 310, 1964. Yatzidis H: Personal communication. Denti E, Gonella M, Moriconi L. Barsott G, Giovannetti S: Attempts to remove uremic toxins through the gastrointestinal tract. Uremia (Kluthe R, Berlyne G, Burton B, eds), Stuttgart, Georg Thieme Verlag, 1972, p 239. Maxwell MH, Gordon A. Greenbaum M: Oral sorbents in medicine. Uremia (Kluthe R, Berlyne G, Burton B, eds), Stuttgart, Georg Thieme Verlag, 1972, p 220. Giordano C. Esposito R, Randazzo G: Oxystarch as a gastrointestinal sorbent in uremia. Uremia (Kluthe R, Berlyne G, Burton B, eds), Stuttgart, Georg Thieme Verlag, 1972, p 231. Giordano C, Esposito R, Pluvio M: The effects of oxidized starch on blood and fecal nitrogen in uremia. Proc Europ Dial Trans Assoc 10: 136, 1973. Giordano C, Esposito R: Studies on oxystarch and uremia. Abstracts of 8th Annual Contractors Conference, Artificial Kidney-Chronic Uremia Program, Washington D.C. NIAMDD, Jan 13-15, 1975. Sparks RE, Mason NS. Meier PM, Ltt MH, Lindan 0: Removal of uremic waste metabolites from the intestinal tract by encapsulated carbon and oxidized starch. Trans Amer Sot Artif Int Organs 17: 229, 1971. Friedman EA, Fastook J, Beyer MM, Rattazzi T, Josephson AS: Potassium and nitrogen binding in the human gut by ingested oxidized starch (OS). Trans Amer Sot Artif Int Organs 20: 161, 1974. Man NK, Drueke T, Paris J, Elizalde Monteverde C, Rondon Nucete M, Zingraff J, Jungars P: Increased nitrogen removal from the intestinal tract of uremic patients. Proc Europ Dial Trans Assoc 10: 143, 1973. Popovtzer MM, Pinggera WF, Robinette JB: Secondary hyperparathyroidism. Conservative management in patients with renal insufficiency. JAMA 231: 960, 1975. Chang TMS: Semipermeable microcapsules. Science 146: 524, 1964. Friedman EA, Laungani GB, Beyer MM: Life prolongation in nephrectomized rats fed oxidized starch and charcoal. Kidney International 7 (suppl3): 377, 1975. Saltzman M. Beyer MM, Friedman EA: Prolonged life and reduction in azotemia in anephric rats fed sorbents. Abstracts of 21st meeting of American Society for Artificial Internal Organs, Washington D.C., 1975,
May 10, 1976
The American Journel of Yedklne
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35. 36.
37. 38.
Volume 60
Yatzidis H: A convenient haemoperfusion micro-apparatus over charcoal for the treatment of endogenous and exogenous intoxication. Proc Europ Dial Transp Assoc 1: 83, 1964. Dunea G, Kolff WJ: Clinical experience with the Yatzidis charcoal artificial kidney. Trans Am Sot Artif Int Organs 11: 178, 1965. Chang TMS: Removal of endogenous and exogenous toxins by a microencapsulated adsorbent. Can J Physiol Pharm 47: 1043. 1969. Chang TMS, Malave N: The development and first clinical use of semipermeable microcapsules (artificial cells) formed from membrane-coated activated charcoal. Trans Amer Sot Artif Int Organs 16: 141, 1970. Chang TMS, Gonda A, Dirks J, Malave N: Clinical evaluation of chronic intermittent or short-term hemoperfusion in patients with chronic renal failure using semipermeable microcapsules (artificial cells) formed from membrane-coated activated charcoal. Trans Amer Sot Artif Organs 17: 246, 1971. Chang TMS, Gonda A, Coffey J, Dirks J, Burns T: ACAC microcapsule artificial kidney for long-term and shortterm management of patients with chronic renal failure. Trans Amer Sot Artif Int Organs 18: 465, 1972. Chang TMS, Migchelsen M, Coffey JF. Stark R: Serum middle molecule levels in uremia during long-term intermittent hemoperfusions with the ACAC (coated charcoal) microcapsule artificial kidney. Trans Amer Sot Artif Int Organs 20: 49, 1974. Silverstein ME, Ford CA, Lysaght MJ, Henderson LW: Treatment of severe fluid overload by ultrafiltration. N Engl J Med 291: 747. 1974. Gordon A,Greenbaum MA, Marantz LB, et al.: A sorbent based low volume recirculating dialysate system. Trans Amer Sot Artif Int Organs 15: 347, 1969. Gordon A, Better DS, Greenbaum MA, Marantz LB, Gral T. Maxwell MH: Clinical maintenance hemodtalysis with a sorbent-based low volume dialysate regeneration system. Trans Amer Sot Artif Int Organs 17: 253, 1971. Giordano C: Use of exogenous and endogenous urea for protein synthesis in normal and uraemic subjects. J Lab Clin Med 63: 231, 1963. Walser M: Ketoacids in the treatment of uremia. Clinical Nephrology 3: 180, 1975. Giordano C, Pluvio M, Esposito R: Urea index and nitrogen balance in uremic patients on minimal nitrogen intakes. Clin Nephrol 3: 168, 1975. Richards P: Nitrogen recycling in uremia: a reappraisal. Clin Nephrol 3: 166, 1975. Setala K, Heinonen H, Schreck-Purola I: Ingestion of lyophilized soil bacteria for alleviation of uremic symptoms. J lnternat Res Comm System 73: 35,’ 1973.
ELI A. FRIEDMAN, M.D. Brook!m
New
York
From the Departmentof Medicine, Downstate Medical center, 450 Clarkson Avenue, Brooklyn. New York 11203.
Requests for reprints
shouldbe addressed to Dr. Eli A. Friedman. ManuscriptacceptedSepbmtmr 19, 1976.
614
May 10, 1976
Once Pendleton and West [l] documented facile transfer of urea from plasma into the intestinal lumen, the potential for treating renal failure by extraction of nitrogen waste from the gut became self evident. Urea in the gut is degraded to ammonia by bowel bacteria to the extent that normal human feces contain no urea. In healthy volunteer subjects given an “antibiotic cocktail,” stool urea concentrations increase to blood levels while the fecal ammonia content decreases: this supports the inference that urea in the bowel is biodegraded by luminal microorganisms [2,3]. Estimates of the quantity of urea converted to ammonia in the gut have been computed by Jones and co-workers [4] (from 4 to 7 g/day in normal patients and from 17 to 50 g/day in uremic patients). Perhaps the earliest clinical attempt to treat uremic patients by lavage of nitrogen-containing intestinal wastes was reported by Kolff in his 1947 text “New Ways of Treating Uremia” [5]. Kolff created a double ended ileostomy in an isolated iteal loop with an intact blood supply in a 57 year old uremic man. As much as 0.48 g of urea/hour was removed by the patient who performed home intestinal dialysis assisted by his wife for two months until his death. Prolongation of life by intestinal dialysis has been accomplished in dog [6] and man [7,8]. Clearances during isolated jejunal loop dialysis in uremic patients were reported by Schloerb [7,8] as 5 to 10 ml/min for creatinine and 3.2 to 5.0 ml/min for uric acid, values about one third and one eighth as efficient, respectively, as obtained with peritoneal and hemodialysis. Intestinal perfusates contain smaller but significant amounts of larger molecules including aldosterone and 17-oxyhydroxycorticosteroids. By 1960, when repetitive hemodialysis was made practical by the development of an external plastic arteriovenous shunt, a slowly increasing number of patients were sustained (albeit suboptimally) by periodic intestinal dialysis. Analyzing 15 cases in the literature plus five of their own, Clark et al. [9], in 1962, concluded that intestinal dialysis “remains the best method of adjunctive management of progressive uremia.” Thereafter, the improving success rate of maintenance hemodialysis diverted interest in the quest for nitrogenous waste extraction from the gut.
The American Journal of Medicine
Volume 80
r
CHARCOAL
Attention was redirected to the value of intestinal nitrogenous waste removal by the work of Yatzidis [lo] in 1964 in which uremic patients were fed charcoal as an oral sorbent, with highly encouraging results. Sorbents are chemicals varying from single elements, such as charcoal, to complex polymers which find or adsorb gases, liquids or dissolved substances on their surface in a condensed layer. Administered in oral doses of 20 to 50 g daily, with or without sorbitol as a vehicle, Yatzidis [lo] was able to manage patients with end-stage renal failure for four to 20 months without resorting to dialytic methods. A relatively high residual renal function (creatinine clearance of 10 to 15 ml/min) in this group of patients, who were also treated with a 20 g restricted protein diet [ 111, makes it difficult to discern any specific merit of charcoal treatment of uremia. Other investigators have not been favorably impressed with the benefit of charcoal ingestion in uremia. In healthy volunteer subjects given daily oral doses of 50 g of powdered activated charcoal, nausea, vomiting and constipation were of sufficient severity to interrupt further trial [ 121. Maxwell and co-workers administered “various forms of carbon in doses up to 90 g daily” to chronically uremic dogs and found no evidence of “significant effect on the lowering of any of the metabolites associated with the uremic state.” Unless comparable protocols are employed, varying results in the assessment of ingestible sorbents by different researchers may be attributable to dissimilar experimental designs. The palatability of charcoal suspension, for example, is remarkably enhanced by chilling and flavoring with a drop of lemon juice permitting trials of at least four weeks. At present, the case for oral charcoal as even adjunctive therapy in renal failure must be judged “unproved.” OXIDIZED STARCH Promising additional data indicating that gastrointestinal sorbents can bind to and remove in feces clinically important amounts of nitrogenous wastes are provided by a series of investigations using ,oxidized starch (oxystarch) and oxidized cellulose (oxycellulose) performed by Giordano and associates [ 1416]. Corn starch or potatoe starch suspended in a solution of sodium periodate at 4’C for 24 hours slowly oxidizes to dialdehyde starch (oxystarch); cellulose treated similarly, oxidizes to oxycellulose. At body pH and temperature, each repeat unit of oxystarch (Figure 1) binds 1.5 to 1.9 moles of ammonia in vitro in an 0.3N ammonia solution; when present in excess, oxystarch will bind all the ammonia in an 0.3N solution. Oxystarch also adsorbs urea and L-
H O\
/
/c\o/ i” 0 Figure 1.
1n
Oxystarch (dialdehyde starch)
aspartic acid in vitro, but it does not bind creatinine, uric acid, L-lysine or albumin. In weanling mice fed 2 per cent oxystarch in a casein diet, growth and development are normal. In rats fed 2 per cent oxystarch in a casein diet severe diarrhea develops, whereas dogs tolerate as much as 5 per cent dietary oxystarch without apparent adverse effect. Explosive diarrhea and a cholera-like fluid and electrolyte depletion syndrome occur in dogs fed more than 10 per cent of oxystarch in their diets [ 171. In uremic patients fed 20 to 35 g of oxystarch in divided doses stool volume increases by 200 to 600 ml/day and the frequency of bowel movements is increased, but frank diarrhea does not develop [ 181. Giordano’s initial trials of oxystarch manufactured in his laboratories showed that uremic patients (creatinine clearances of 0.4 to 3.2 ml/min) tolerated divided doses of 20 g/day well for two months and that, in each case, there was a significant fall in the blood urea nitrogen level. Fecal nitrogen content increased a mean of 1,450 mg/day (range 730 to 8,050 mg/day) [ 14,151. Confirmation of increased stool nitrogen content during oxystarch treatment was provided by a double blind starch-oxystarch full balance study reported by Friedman et al. 1181. In this study seven uremic patients (creatinine clearances of 6 to 30 ml/min) were fed 29 g of oxystarch or starch daily in four equal doses added to a diet containing 40 to 50 g of protein and 2 to 4 g of salt. Blood urea nitrogen levels fell 33 per cent during oxystarch treatment from a mean of 93.1 mg/ 100 ml to a mean of 62.1 mg/lOO ml. There was no significant change in serum creatinine, plasma amino acid, uric acid and plasma glucose levels during oxystarch ingestion. Oxystarch significantly increased fecal nitrogen from a control mean of 1.4 g/24 hours to 2.5 g/24 hours. A concomitant decrease in urinary nitrogen excretion, however, from a control mean of 7.6 g/24 hours to 5.5 g/24 hours during oxystarch treatment prevented development of negative nitrogen
May 10, 1976
The
American Journal of Yediclne
Volume 60
615
SORBENTS IN THE MANAGEMENT OF UREMIA-FRIEDMAN
CONTROL
OXYSTARCH
C7
-
Meon
-
M*m
0XYSTA.W~
f
Figure 2. During minimal nitrogen ingestion, uremic patients fed oxystarch (28 g) daily have an increased fecal excretion of nitrogen and potassium, and a dounterbalancing decrease in urinary nitrogen and potassium excretion. (Reproduced with permission from Friedman et al. [ 181.)
balance. There was a significant increase in the fecal potassium content when oxystarch was ingested, ranging from 5 to 22 meq/day, which was also counterbalanced by a decrease in urinary potassium excretion. To exclude the possibility that increased fecal nitrogen content noted during oxystarch treatment was due to direct binding of unabsorbed undigested dietary nitrogen rather than bound intraintestinal nitrogen wastes (ammonia [ 191 and urea), four uremic patients were fed oxystarch while ingesting a “no protein” diet [ 181. In these patients (Figure 2) the increases in fecal nitrogen and potassium were similar to those in the previous group indicating that the origin of the extra fecal nitrogen was indeed nitrogenous wastes. Will feeding oxystarch to uremic patients have clinical import? In urine producing patients the counterbalancing decrease in urinary nitrogen excretion tends to mitigate the benefit of increased fecal nitrogen content. What will be the effect on nitrogen balance in anuric patients? Sketchy data on one uremic patient in whom the frequency of peritoneal dialyses was extended from once in seven days to once in 25 days and another who had thrice weekly hemodialysis reduced to once weekly during oxystarch therapy have been reported by Giordano et al. [ 141. Because the prescription for “adequate dialysis” is based on a combination of objective (serum creatinine and serum phosphorous levels, for example) and subjective (work tolerance, ability to sleep through the night, bone pain are examples) criteria, any therapy
616
May 10,
1976 The American Journal of Medlclne
purported to reduce the frequency or duration of dialyses must be evaluated according to a carefully controlled study. It may be anticipated that such a study, as well as a balance study of the effect of charcoal ingestion by uremic man, will be completed within the next few years. One oral sorbent, aluminum hydroxide gel, is presently in clinical use in renal failure to bind phosphate in the gut. Combined with supplemental oral calcium ingestion, in uremic patients treated with aluminum hydroxide gel, both serum phosphorous and elevated parathyroid hormone levels may be lowered to normal [ 201. Exploitation of newer technics of microencapsulation [21] opens the way to possible human use of potent sorbents previously excluded from consideration because of inherent or unknown toxicity (zirconium salts, for instance). While awaiting completion of sorbent trials in functionally anephric (undergoing dialytic maintenance) patients, several helpful animal experiments have been completed. To date, chronically uremic animals in need of dialysis have not been sustained by sorbents alone. Gavage feeding of oxystarch alone or in combination with charcoal will prolong the life of anephric rats from a control of three days to five days [ 221. Saltzman and co-workers [ 231 investigated the mechanism of sorbent-induced life extension in bilaterally nephrectomized rats fed charcoal (1 g daily), oxystarch (1 g daily) or oxystarch plus charcoal (1 g of each daily). In sorbent-treated rats the increase in blood urea concentration was less than in untreated nephrectomized controls, but they also had lower serum potassium concentrations throughout their increased life span. Both actions of oxystarch, increased fecal excretion of nitrogen and potassium, were detectable in this in vivo model of fatal acute renal failure. HEMOPERFUSION For completeness in this survey, the technic of direct exposure of blood to a sorbent, termed hemoperfusion, will be mentioned. During a single passage over granular activated charcoal, creatinine, uric acid, indican, phenols, guanidines and organic acids are nearly totally extracted from the blood. Only negligible quantities of urea, magnesium and phosphate are adsorbed on charcoal from blood. Each gram of powdered charcoal will in vitro bind simultaneously 9 mg of creatinine, 8 mg of uric acid, I.75 mg of phenols, 1 mg of guanidines and 35 mg of urea. In 1964, Yatzidis [24] devised a hemoperfusion device containing 200 g of activated charcoal in a siliconized glass cylinder 20 cm in length and 6 cm in diameter. Based on an experience of 20 human hemoperfu-
Volume 60
SORBENTS IN THE MANAGEMENT
sions at a blood flow rate of about 140 ml/min, Yatzidis estimated that 60 minutes of blood exposure to two or three 200 g charcoal columns had about the same extraction efficiency (for uremic patients) as a 4 to 6 hour hemodialysis. Hemoperfusion was advocated for the treatment of renal insufficiency, gouty arthritis, and intoxications with salicylates, barbiturates and glutethimide [ 241. Early attempts to introduce charcoal hemoperfusion into clinical management of uremic patients were unsuccessful. Of three uremic patients treated with hemoperfusion by Dunnea and Kolff [25], one died after nine perfusions and the other two were switched to maintenance hemodialysis. Noteworthy in this report is the exceptionally high creatinine clearance of 388 ml/min determined in vitro at a blood rate of 400 ml/min. Dual hazards, extraction of formed elements from blood and discharge of charcoal microemboli into efferent blood, complicate the exposure of blood to untreated charcoal. Efforts to eliminate the disadvantages of uncoated charcoal while retaining charcoal’s extraordinary adsorbing properties have concentrated on coating charcoal particles to eliminate the trapping and destruction of platelets, cells and fibrinogen. Chang et al. [26-291 have systematically studied several approaches to microencapsulation of charcoal and they developed a blood-compatible albumin-complexed polymer. In its present form, Chang’s microencapsulated “kidney” contains 300 g of double coated charcoal granules 2 to 5 mm in diameter with a surface area of 2.25 m2. This device achieves clearances in vitro, which are superior for middle molecules (MW 200 to 1,800), to hollow fiber, coil or parallel flow hemodialyzers [ 301. Periodic hemoperfusion as a sole treatment for uremia is inadequate because of the need to extract water and probably urea to maintain acceptable morbidity. Combination of hemoperfusion within series ultrafiltration of blood 1311 for water removal might prove a workable therapy for uremic patients who ingest ammonia binding sorbents such as oxidized starch. The efficiency advantage (shorter treatments) of hemoperfusion over hemodialysis in renal failure is sufficiently attractive to consider for-
mulation of regimens oral sorbent ingestion.
coupling
OF UREMIA-FRIEDMAN
hemoperfusion
with
DIALYSATE REGENERATION Most of the bulk of early hemodialysis systems was taken up by a tank to store a 100 to 180 liter dialysate reservoir. Newer compact disposable hollow fiber, coil and parallel flow hemodialyzers have permitted designs for dialysate delivery systems which “regenerate” dialysate by passage through a sorbent cartridge. A remarkable reduction in dialysate volume to 1.5 liters was devised by Gordon and co-workers 132,331 by the clever means of converting urea in dialysate to ammonium ion and carbonate by urease treatment. Ammonium ion is adsorbed by zirconium phosphate which also extracts calcium and magnesium necessitating continuous reinfusion of these ions. Zirconium oxide binds phosphate and fluoride while charcoal extracts uric acid, creatinine, guanidines, organic acid and phenols. The total weight of the sorbent cartridge is less than 2 kg. The importance of the Gordon system is not only the size reduction, which makes a suitcase travel “artificial kidney” practical, but also its clear demonstration of the value of sorbents in simplification of the therapy of uremia. In closing, it is appropriate for the nephrologist to express respect and admiration for the metabolic events occurring within the gastrointestinal tract. It has been suggested that fed the right combination of essential amino acids [34], or their a-keto precursors [35], uremic patients may actually anabolize nitrogen in urea [36]. More recent balance studies indicate that less than 6 per cent of urea might be recycled thusly [37]. Even more startling to conventional thinking is the work of Setala et al. [38], who believe that the daily ingestion of enzymes derived from soil bacteria can (and does) convert end products of protein digestion within the gut, such as creatinine and urea, into usable recycled metabolites. Uremia might thus be treated without hemodialysis, hemoperfusion or sorbent ingestion. No wonder that the nephrologist occasionally regards the gut as a displaced, hypertrophied, but very “intact nephron,”
REFERENCES 1.
2.
3.
Pendleton WR, West FE: The passage of urea between the blood and the lumen of the small intestine. Am J Physiol 101: 391.1932. Wrong 0, Metcalfe-Gibson A, Morrison RBL, Ing ST, Howard AV: In vivo dialysis of faeces as a method of stool analysis. I. Technique and rest&s in normal subjects. Clin Sci 26: 357, 1965. Wrong 0. Houghton BJ, Richards P, Wilson DR: Intestinal urea in normals and in uremic patients in urea and the kidney. (Schmidt-Nielsen B, ed), Amsterdam, Excerpta
4.
5. 6.
7.
May 10, 1976
Medica Foundation, 1970, p 46 1. Jones EA, Smallwood RA. Craigie A, Rosenoer VM: Enterohepatic circulation of urea nitrogen. Clin Sci 37: 825, 1969. Kolff WJ: New Ways of Treating Uremia. London, J. A. Churchill, 1947. Sloan H: Effect of perfusion of isolated intestinal loops on experimental uremia in dogs. Am J Physiol 166: 137, 1951. Schloerb PR: The management of uremia by perfusion of
The American Journal of Medicine
Volume 60
617
SORBENTS IN THE MANAGEMENT OF UREMIA-FRIEDMAN
8. 9.
10. 11. 12.
13.
14.
15.
16.
17.
18.
19.
20.
21. 22.
23.
618
an isolated jejunal segment. J Clin Invest 37: 1818, 1958. Schloerb PR, Lukart BL: Further observations on the isolated human jejunum. Arch Surg 86: 356, 1963. Clark JE, Templeton JY Ill, McCurdy DK: Perfusion of isolated intestinal loops in the management of chronic uremia. Trans Amer Sot Artif Int Organs 8: 246, 1962. Yatzidis H: Recherches our lepuration extra renale a I’aide do charbon actif. Nephron 1: 310, 1964. Yatzidis H: Personal communication. Denti E, Gonella M, Moriconi L. Barsott G, Giovannetti S: Attempts to remove uremic toxins through the gastrointestinal tract. Uremia (Kluthe R, Berlyne G, Burton B, eds), Stuttgart, Georg Thieme Verlag, 1972, p 239. Maxwell MH, Gordon A. Greenbaum M: Oral sorbents in medicine. Uremia (Kluthe R, Berlyne G, Burton B, eds), Stuttgart, Georg Thieme Verlag, 1972, p 220. Giordano C. Esposito R, Randazzo G: Oxystarch as a gastrointestinal sorbent in uremia. Uremia (Kluthe R, Berlyne G, Burton B, eds), Stuttgart, Georg Thieme Verlag, 1972, p 231. Giordano C, Esposito R, Pluvio M: The effects of oxidized starch on blood and fecal nitrogen in uremia. Proc Europ Dial Trans Assoc 10: 136, 1973. Giordano C, Esposito R: Studies on oxystarch and uremia. Abstracts of 8th Annual Contractors Conference, Artificial Kidney-Chronic Uremia Program, Washington D.C. NIAMDD, Jan 13-15, 1975. Sparks RE, Mason NS. Meier PM, Ltt MH, Lindan 0: Removal of uremic waste metabolites from the intestinal tract by encapsulated carbon and oxidized starch. Trans Amer Sot Artif Int Organs 17: 229, 1971. Friedman EA, Fastook J, Beyer MM, Rattazzi T, Josephson AS: Potassium and nitrogen binding in the human gut by ingested oxidized starch (OS). Trans Amer Sot Artif Int Organs 20: 161, 1974. Man NK, Drueke T, Paris J, Elizalde Monteverde C, Rondon Nucete M, Zingraff J, Jungars P: Increased nitrogen removal from the intestinal tract of uremic patients. Proc Europ Dial Trans Assoc 10: 143, 1973. Popovtzer MM, Pinggera WF, Robinette JB: Secondary hyperparathyroidism. Conservative management in patients with renal insufficiency. JAMA 231: 960, 1975. Chang TMS: Semipermeable microcapsules. Science 146: 524, 1964. Friedman EA, Laungani GB, Beyer MM: Life prolongation in nephrectomized rats fed oxidized starch and charcoal. Kidney International 7 (suppl3): 377, 1975. Saltzman M. Beyer MM, Friedman EA: Prolonged life and reduction in azotemia in anephric rats fed sorbents. Abstracts of 21st meeting of American Society for Artificial Internal Organs, Washington D.C., 1975,
May 10, 1976
The American Journel of Yedklne
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35. 36.
37. 38.
Volume 60
Yatzidis H: A convenient haemoperfusion micro-apparatus over charcoal for the treatment of endogenous and exogenous intoxication. Proc Europ Dial Transp Assoc 1: 83, 1964. Dunea G, Kolff WJ: Clinical experience with the Yatzidis charcoal artificial kidney. Trans Am Sot Artif Int Organs 11: 178, 1965. Chang TMS: Removal of endogenous and exogenous toxins by a microencapsulated adsorbent. Can J Physiol Pharm 47: 1043. 1969. Chang TMS, Malave N: The development and first clinical use of semipermeable microcapsules (artificial cells) formed from membrane-coated activated charcoal. Trans Amer Sot Artif Int Organs 16: 141, 1970. Chang TMS, Gonda A, Dirks J, Malave N: Clinical evaluation of chronic intermittent or short-term hemoperfusion in patients with chronic renal failure using semipermeable microcapsules (artificial cells) formed from membrane-coated activated charcoal. Trans Amer Sot Artif Organs 17: 246, 1971. Chang TMS, Gonda A, Coffey J, Dirks J, Burns T: ACAC microcapsule artificial kidney for long-term and shortterm management of patients with chronic renal failure. Trans Amer Sot Artif Int Organs 18: 465, 1972. Chang TMS, Migchelsen M, Coffey JF. Stark R: Serum middle molecule levels in uremia during long-term intermittent hemoperfusions with the ACAC (coated charcoal) microcapsule artificial kidney. Trans Amer Sot Artif Int Organs 20: 49, 1974. Silverstein ME, Ford CA, Lysaght MJ, Henderson LW: Treatment of severe fluid overload by ultrafiltration. N Engl J Med 291: 747. 1974. Gordon A,Greenbaum MA, Marantz LB, et al.: A sorbent based low volume recirculating dialysate system. Trans Amer Sot Artif Int Organs 15: 347, 1969. Gordon A, Better DS, Greenbaum MA, Marantz LB, Gral T. Maxwell MH: Clinical maintenance hemodtalysis with a sorbent-based low volume dialysate regeneration system. Trans Amer Sot Artif Int Organs 17: 253, 1971. Giordano C: Use of exogenous and endogenous urea for protein synthesis in normal and uraemic subjects. J Lab Clin Med 63: 231, 1963. Walser M: Ketoacids in the treatment of uremia. Clinical Nephrology 3: 180, 1975. Giordano C, Pluvio M, Esposito R: Urea index and nitrogen balance in uremic patients on minimal nitrogen intakes. Clin Nephrol 3: 168, 1975. Richards P: Nitrogen recycling in uremia: a reappraisal. Clin Nephrol 3: 166, 1975. Setala K, Heinonen H, Schreck-Purola I: Ingestion of lyophilized soil bacteria for alleviation of uremic symptoms. J lnternat Res Comm System 73: 35,’ 1973.