Cholestasis

Parenteral nutrition (PN) is an essential component in the care of premature and ill infants. The incidence of parenteral nutrition-associated cholestasis (PNAC) ranges from 7.4 to 84%. One substance in PN solutions that has been implicated in PNAC is aluminum. Aluminum loading in animals and humans causes hepatic accumulation and damage. The degree of aluminum contamination of PN solutions has decreased over time, but contamination still significantly exceeds levels that are safe for human neonates. Further study into the relationship between aluminum contamination in neonatal PN solutions and the development of PNAC is necessary.

Key words: parenteral nutrition, cholestasis, aluminum, contamination

Introduction

Parenteral nutrition (PN) is an essential component of the care of premature and ill infants. Cholestasis is a form of liver injury that occurs because of reduced bile flow from the liver into the duodenum. Cholestasis causes liver damage, significant illness, and in some cases, death. Infants and neonates are at particular risk for this complication. Parenteral nutrition-associated cholestasis (PNAC) was first described by Peden et al. in 1971.1 The incidence of PNAC in neonates is between 7.4 and 84% depending on several factors such as low birth weight, young gestational age, hypoxia at birth, presence or absence of sepsis, and whether the neonate had gastrointestinal surgery.2-4 Research performed by Arnold et al. discovered that 25% of neonates with intestinal pathology developed PNAC after a mean of 18 days of PN therapy.5 The authors also found that the number of days on PN, birth weight, and having undergone gastrointestinal surgery were significant predictors of PNAC.5 Forty to sixty percent of children on long-term PN develop some degree of hepatic dysfunction.6 Cannalicular cholestasis has been shown to present in 85% of infants after only 10 days on PN. After 3 weeks, most infants on PN develop bile duct proliferation and 80% exhibit periportal fibrosis.

Since PNAC was first described, many components of PN have changed. Originally, PN consisted mainly of dextrose and casein hydrolysate amino acids. It was not until the early 1980s that lipid became an integral component. During this time, the source of amino acids also changed from casein hydrolysates to synthetic crystalline amino acids. Along with the changes in solution composition, there was a shift from hyperalimentation to balancing provision with requirement. Through all of these changes cholestatic liver disease has endured.

Many different theories exist as to the etiology of PNAC. One aspect of PN that was implicated was the absence of enteral nutrition while on PN. This mechanism was felt to be secondary to reduced cholecystokinin (CCK) levels. CCK is responsible for stimulating emptying of the gallbladder, thus preventing biliary sludge formation.7 Although researchers found that CCK did reduce biliary sludge formation, it did not significantly reduce the incidence of PNAC.8-11 Lack of enteral nutrition is suggested to lead to intestinal stasis, which alters enterohepatic circulation and leads to the development of toxic bile acids that cause cholestasis.12-15 This suggestion conflicts with those of other researchers who have found progressive intrahepatic cholestasis despite the use of partial enteral nutrition as an adjunct to PN.16-18 It therefore appears that the role of enteral nutrition in the development of PNAC is controversial.

Another hypothesis states that the PN solution itself or some constituent therein is toxic to the liver. The fatty acid composition of parenteral lipid emulsions has been investigated. Researchers have found that people fed fish oil-based emulsions developed less cholestasis than people fed traditional soybean products.19 This venue continues to be explored. Other lipid components that have been investigated include plant sterols. Plant sterols have been shown to cause a reduction in bile secretion in experimental models.20,21 The mechanism was related to a reduction in cannalicular secretory activity.20 However, the presence of PNAC was not verified.17

It is important to note that cholestasis occurred prior to the routine use of parenteral lipid emulsions in nutritional support. This suggests that whereas lipid emulsions may contribute to PNAC, they are unlikely to be the root cause. Parenteral amino acid solutions have also been investigated in relation to PNAC. Some researchers found that the incidence of cholestasis was directly related to the volume of amino acids infused because they affected the cannalicular membrane of the hepatocyte;22,23 others studying the effect of specific amino acid solutions on cholestasis found that the addition of a methyl donor amino acid, such as serine or methionine, may protect against the development of PNAC.24 Later research suggested that methionine may be toxic to the liver, thereby causing cholestasis (Figure 1).25-29

Aluminum

One constituent of PN solutions that has been implicated in PNAC is the contaminant aluminum. Aluminum is a nonessential toxic metal that is the third most abundant component of the earth's crust.30 Oral exposure to aluminum occurs through the ingestion of aluminum-containing medications, food, and water. The human gastrointestinal tract is an effective barrier against aluminum accumulation as typical absorption is

Once aluminum is absorbed, it can either remain free in the blood or it can become bound to the plasma protein transferrin.29 Transferrin is thought to be the mechanism by which aluminum is delivered to the liver.33 Once it is in the liver, aluminum has been found (via electron microscopy) to be deposited in lysosomes of macrophages,34 in lipofuscin granules of hepatocytes,35 and in the phagolysosomes of hepatocytes, where they create severe ultrastructure lesions.36 The major route of aluminum elimination is via renal excretion with minor excretion in bile.29 In cases of low glomerular filtration rates, as seen in premature neonates with underdeveloped renal systems, the free aluminum is not removed. The bound aluminum in complex with transferrin is also nonfilterable.29,32 These two factors therefore lead to aluminum accumulation in the premature neonate.

Accumulated aluminum is also toxic in other organ systems implicated in the pathogenesis of osteopenic bone disease,7,37,38 microcytic anemia, and encephalopathy.32 In relation to metabolic bone disease, aluminum has been shown to impair mineralization by competing with calcium for deposition in bone matrix as well as by impairing parathyroid secretion and calcitriol formation.32,34,39

Aluminum and PNAC

In 1984, Klein et al.40 investigated hepatic aluminum accumulation in children on total parenteral nutrition. Klein evaluated five children, 18 to 34 months of age, with PNAC who had been on PN for 18 to 33 months. In addition to elevated liver enzymes (serum glutamic-oxaloacetic transferase, serum glutamic-pyruvic transferase, alkaline phosphatase, total/direct bilirubin), increased serum, urine, and hepatic aluminum concentrations were also found. Hepatic aluminum concentrations were between 5 and 27 times normal concentrations. Histopathologic findings of liver biopsies included periportal fibrosis, bile duct proliferation, inflammatory cells, cellular necrosis, and nonuniform nodular regeneration. Klein et al. speculated that the abnormal accumulation of aluminum in the liver of children might contribute to the development of PNAC.

Klein et al.41 furthered this work in 1986 using a piglet model to investigate the association between aluminum and PNAC. Eight six-week-old piglets were divided into two groups. The experimental animals received 1.5 mg. kg^sup -1^. day^sup -1^ of parenteral aluminum via central catheter for 50 days. The control animals received heparinized saline via a central catheter for the same 50 days. At the conclusion of the study period, the animals were euthanized and serum and liver samples were taken. The authors found significant increases in serum bile acids and alkaline phosphatase levels in the experimental animals compared with the controls, which suggested that cholestasis had developed. The researchers also found the presence of aluminum in the lysosomes of hepatocytes of experimental animals when they were examined by electron microscopy.

In 1988, Klein et al.42 studied hepatobiliary dysfunction in relation to aluminum loading in rats. The animals were placed into experimental and control groups. The experimental groups received aluminum via central catheters while the control groups received saline. The experimental groups received aluminum in doses of 5 mg. kg^sup -1^. day^sup -1^ for 14 days, 1 mg. kg^sup -1^. day^sup -1^ for 14 days, or 5 mg. kg^sup -1^. day^sup -1^ for 7 days. After the study period, the animals were anaesthetized and had bile duct and bladder cannulae inserted. Bile and urine were collected for 3 hours. After the 3-hour period, the animals were euthanized. The researchers found that serum bile acids were greater in the experimental groups than in the control groups; there were higher concentrations in the 14-day group than in the 7-day group. Biliary flow was reduced by 33% in the 5 mg. kg^sup -1^. day^sup -1^ group but not in the 1 mg. kg^sup -1^. day^sup -1^ group. Hepatic aluminum concentration correlated with bile flow but not with serum bile acid concentrations. The authors concluded that aluminum loading might be associated with the pathogenesis of parenteral nutrition-induced hepatocellular dysfunction.

Moreno et al.43 reported on a prospective observational study of 35 newborn infants who required PN support. The infants were on PN for 15.6 days + or - 8.7 days. Indications for PN support were low birth weight, necrotizing enterocolitis, and abdominal surgery. Aluminum contents of all substances provided were determined. Serum and urinary aluminum concentrations were tested for all neonates receiving PN as were aluminum concentrations from brain, bone, liver, and kidney at the time of autopsy in deceased infants. A control group of 13 neonates of similar gestational age were also followed. Moreno found that the PN solutions accounted for 88.7% of the total aluminum intake. Serum and urine aluminum levels were greater in the group receiving PN than the control group. Tissue deposition of aluminum at autopsy was greater in the neonates receiving PN than in the controls. The authors also found that aluminum contamination of PN solutions ranged from 4.16 to 7.26 [mu]mol/L, representing an intake of 0.62 [mu]mol. kg^sup -1^. day^sup -1^ (16.7 [mu]g. kg^sup -1^. day^sup -1^). The researchers concluded that neonates are susceptible to accumulation of aluminum in tissues while on PN.

In 1999, Popinska et al.44 evaluated the aluminum content of commonly used PN solutions. They quantified the amount of aluminum in the major components of the PN solutions and found that aluminum was present in concentrations of approximately 10.6 [mu]g/kg of body weight in infants. This is significantly greater than the American Society of Clinical Nutrition/American Society of Enteral and Parenteral Nutrition Joint Commission (ASCN/ASPEN) recommendation of 2 [mu]g. kg^sup -1^. day^sup -1^.45

Heyman et al.46 showed that adults and teenagers receiving aluminum infusions at or below 2 [mu]g. kg^sup -1^. day^sup -1^ had no signs of accumulation. It is important to note that both Moreno and Popinska demonstrated that neonates receive significantly more than 2 [mu]g. kg^sup -1^. day^sup -1^ of aluminum infusion from present-day PN solutions. In addition, no evidence exists to indicate a safe level of aluminum contamination for neonates. Because premature neonates have underdeveloped renal systems that are incapable of metabolizing aluminum, levels far below 2 [mu]g. kg^sup -1^. day^sup -1^ may pose significant health risk to neonates.

Conclusions

PNAC has been a complication of intravenous nutrition for nearly 30 years. Many theories have been proposed and continue to be explored. Research has shown that the individual components of the PN solution are associated with the development of PNAC. An important commonality to this relationship is that all of the major components of PN are contaminated with aluminum to varying degrees; this includes vitamin and mineral preparations, amino acids, and to a lesser degree dextrose solutions and lipid emulsions. Research to date on aluminum contamination and hepatic dysfunction suggests a strong relationship although much of the experimental research used non-neonatal models with aluminum doses far exceeding the level of contamination seen today. Research by Moreno43 and Popinska44 demonstrated that neonates and infants are being exposed to greater aluminum loads than are safe for teens and adults. In light of the underdeveloped organ systems of the premature neonate, potential morbidity and mortality caused by aluminum contamination of parenteral nutrition formulae are unacceptable. Further research into the relationship between aluminum-contaminated PN solutions and the development of PNAC in neonates and infants is therefore necessary.

1. Peden V, Witzleben C, Skelton M. Total parenteral nutrition. J Pediatr. 1971;78:180-181.

2. Cohen C, Olsen M. Pediatric total parenteral nutrition. Liver histopathology. Arch Pathol Med. 1981; 105:152-156.

3. Bell R, Ferry G, Smith E, et al. Parenteral nutrition-related cholestasis in infants. JPEN. 1986;10:356-359.

4. Merrit R. Cholestasis associated with total parenteral nutrition. J Pediatr Gastroenterol Nutr. 1986;5: 9-22.

5. Arnold CJ, Miller GG, Zello GA. Parenteral nutrition associated cholestasis in infants with intestinal failure. FASEB J. 2002;16:A281.

6. Kelly D. Liver complications of pediatric parenteral nutrition-epidemiology. Nutrition. 1998;14: 153-157.

7. Sitzman JV, Pitt HA, Steinborn PA, Pasha Z, Sanders R. Cholecystokinin prevents parenteral nutrition induced biliary sludge in humans. Surg Gynecol Obstet. 1990;170:25-31.

8. Moss RL, Amii LA. New approaches to understanding the etiology and treatment of total parenteral nutrition-associated cholestasis. Semin Pediatr Surg. 1999;8:140-147.

9. Touloukian R, Seashore J. Hepatic secretory obstruction with total parenteral nutrition in the infant. J Pediatr Gastroenterol Surg. 1975;10:353-360.

10. Aynsley-Green G. Plasma hormone concentrations during enteral and parenteral nutrition in the human newborn. J Pediatr Gastroenterol Nutr. 1983;2: S108-S112.

11. Greenberg G, Walman S, Christofides N, Bloom S, Jeejeebhoy K. Effect of total parenteral nutrition on gut hormone release in humans. Gastroenterology. 1981;80:988-993.

12. Rooney J, Hill D, Danks D. Jaundice associated with bacterial infection in the newborn. Am J Dis Child. 1971;122:39-41.

13. Nolan J. The role of endotoxin in liver injury. Gastroenterology. 1975;69:1346-1356.

14. Utili R, Abernathy C, Zimmerman H. Cholestatic effects of Escherichia coli endotoxin on the isolated perfused rat liver. Gastroenterology. 1976;70:248-253.

15. Palmer R, Ruban Z. Production of bile duct acid hyperplasia and gallstones by lithocholic acid. J Clin Invest. 1966;45:1255-1267.

16. Hodes J, Grosfeld J, Weber T, Schreiner R, Fitzgerald J, Mirkin L. Hepatic failure in infants on total parenteral nutrition: clinical and histopathological observations. J Pediatr Surg. 1982;17:463-468.

17. Moss R, Das J, Raffensberger J. Total parenteral nutrition-associated cholestasis: clinical and histopathologic correlation. J Pediatr Surg. 1993;28: 1270-1275.

18. Dermican M, Ergun O, Avanoglu S, Yilmaz F, Ozok G. Determination of serum bile acids routinely may prevent delay in diagnosis of total parenteral nutrition-induced cholestasis. J Pediatr Surg. 1999;34: 565-567.

19. Van Aerde J, Duerksen D, Gramlich L, et al. Intravenous fish oil emulsion attenuates total parenteral nutrition-induced cholestasis in newborn piglets. Pediatr Res. 1999;45:202-208.

20. Clayton P, Bowron A, Mills K, Massoud A, Casteels M, Milla P. Phytosterolemia in children with parenteral nutrition-associated cholestatic liver disease. Gastroenterology. 1993;105:1806-1813.

21. Iyer K, Spitz L, Clayton P. New insight into mechanisms of parenteral nutrition-associated cholestasis: role of plant sterols. J Pediatr Surg. 1998;33: 1-6.

22. Black D, Suttle E, Whitington P, Whitington G, Kornes S. The effect of short-term total parenteral nutrition on hepatic function in the human neonate: a prospective randomized study demonstrating alteration of hepatic cannalicular function. J Pediatr. 1981;99:445-449.

23. Vileisis R, Inwood R, Hunt C. Prospective controlled study of parenteral nutrition-associated cholestatic jaundice: effect of protein intake. J Pediatr. 1980; 96:893-897.

24. Belli D, Fournire L-A, Lepage G, et al. Total parenteral nutrition-associated cholestasis in rats: comparison of different amino acid mixtures. JPEN. 1987;11:67-73.

25. Moss R, Haynes A, Pastuszyn A, Glew R. Methionine infusion reproduces liver injury of parenteral nutrition cholestasis. Pediatr Res. 1999;45:664-668.

26. Preisig R, Rennert O. Biliary transport and cholestatic effects of amino acids. Gastroenterology. 1989;73:1240.

27. Moss R, Das J, Ansari G. Hepatobiliary dysfunction during total parenteral nutrition is caused by infusate, not route of administration. J Pediatr Surg. 1993;28:391-397.

28. Coran A, Drongowski R. Studies on the toxicity and efficacy of a new amino acid solution in pediatric parenteral nutrition. JPEN. 1987;11:368-377.

29. Cooper A, Betts J, Periera G, Ziegler M. Taurine deficiency in severe hepatic dysfunction complicating total parenteral nutrition. J Pediatr Surg. 1984; 19:462-466.

30. Greger J, Sutherland J. Aluminum exposure and metabolism. Crit Rev Clin Lab Sci. 1997;34:439-474.

31. Davis A, Spillane R, Zublena L. Aluminum: a problem trace metal in nutrition support. Nutr Clin Pract. 1999;14:227-231.

32. Recknagel S, Bratter P, Chrisafidou A, Gramm H, Kotwas J, Rosick U. Parenteral aluminum loading in critical care medicine part I: aluminum content of infusion solutions and solutions for parenteral nutrition, Infusionsther Transfusionsmed. 1994;21:266-273.

33. Klein G, Goldblum R, Moslen M, et al. Increased biliary transferrin excretion following parenteral aluminum administration to rats. Pharmacol Toxicol. 1993;72:373-376.

34. Fiejka M, Fiejka E, Dougaszek M. Effect of aluminum hydroxide administration on normal mice: tissue distribution and ultrastructure localization of aluminum in liver. Pharma Toxicol. 1996;78:123-128.

35. Galle P, Giudicelli C. Electron microprobe ultrastructure localization of aluminum in hepatocytes. Nouv Presse Med. 1982;11:1123-1125.

36. Galle P, Giudicelli C, Nebout T, Baglin A, Fries D. Ultrastructural localization of aluminum in hepatocytes of hemodialyzed patients. Ann Pathol. 1987; 7:163-170.

37. Klein G. Aluminum in parenteral solutions revisited-again. Am J Clin Nutr. 1995;61:449-456.

38. Koo W. Parenteral nutrition-related bone disease. JPEN. 1992;16:386-394.

39. Shike M, Shils M, Heller A, et al. Bone disease in prolonged parenteral nutrition: osteopenia without mineralization defect. Am J Clin Nutr. 1986;44: 89-98.

40. Klein G, Berquist W, Ament M, Coburn J, Miller N, Alfrey A. Hepatic aluminum accumulation in children on total parenteral nutrition. J Pediatr Gastroenterol Nutr. 1984;3:740-743.

41. Klein G, Sedman A, Heyman M, et al. Hepatic abnormalities associated with aluminum loading in piglets. JPEN. 1987;11:293-297.

42. Klein G, Heyman M, Lee T, et al. Aluminum-associated hepatobiliary dysfunction in rats: relationships to dosage and duration of exposure. Pediatr Res. 1988;23:275-278.

43. Moreno A, Dominguez C, Ballabriga A. Aluminum in the neonate related to parenteral nutrition. Acta Pediatr. 1994;83:25-29.

44. Popinska K, Kierkus J, Lyszkowska M, et al. Aluminum contamination of parenteral nutrition additives, amino acids solutions, and lipid emulsions. Nutrition. 1999;15:683-686.

45. ASCN/ASPEN Working Group on Standards for Aluminum Content of Parenteral Nutrition Solutions. Parenteral drug products containing aluminum as an ingredient or a contaminant: response to Food and Drug Administration notice of intent and request for information. JPEN. 1991;15:194-198.

46. Heyman M, Klein G, Wong A, et al. Aluminum does not accumulate in teenagers and adults on prolonged parenteral nutrition containing free form amino acids. JPEN. 1986;10:86-87.

Chris J. Arnold, R.D., Grant G. Miller, M.D., and Gordon A. Zello, Ph.D.

Mr. Arnold and Dr. Zello are with the College of Pharmacy and Nutrition, University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7M 0Z9. Dr. Miller is with the College of Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, Canada.

Copyright International Life Sciences Institute and Nutrition Foundation Sep 2003
Provided by ProQuest Information and Learning Company. All rights Reserved