Asia Pacific J Clin Nutr (l992) 1, 67-72

Serum bile acid fractions in neonates on total parenteral nutrition is lithocholic acid responsible for the occurrence of cholestasis?

Akio Kubota, Kenji Imura, Akira Okada*, Shinkichi Kamata*, Riichiro Nezu* and Hisayoshi Kawahara*.

Department of Pediatric Surgery, Osaka Medical Center and Research Institute for Maternal and Child Health, Osaka, Japan and *Department of Pediatric Surgery, Osaka University Medical School, Osaka, Japan.

In order to determine whether lithocholic acid (LCA) contributes to the occurrence of total parenteral nutrition (TPN) associated intrahepatic cholestasis (IHC) in neonates, we investigated the serum bile acid fractions of neonates on TPN. Twenty-five surgical neonates, receiving TPN for more than 2 weeks were studied. TPN associated IHC was defined as serum defect bilirubin greater than 2.0 mg/dL. Serum bile acid fractions were examined by HPLC using 3a -hydroxy steroid dehydrogenase. Eight patients (32%; IHC group) developed TPN associated IHC. Serum direct bilirubin concentrations in the non IHC and IHC groups were 0.99 and 3.31 mg/dL respectively. Serum total bile acid levels in both groups were 14.4 and 71.6 mmol/ml respectively. Glycine and taurine conjugated cholic and chenodeoxycholic acids could be detected, and unconjugated and secondary (deoxycholic and lithocholic) bile acid were detected in trace levels in both the IHC and non-IHC groups. In conclusion, LCA is unlikely to be a causative factor in TPN associated IHC in neonates.

Introduction

Intrahepatic cholestasis especially in neonates frequently develops during the course of total parenteral nutrition. Despite the vast number of investigations dealing with its aetiology, the cause of this TPN associated liver dysfunction remains unclear. This is now considered to be related to various factors, including immaturity, early fasting, surgical operations, underlying diseases, overloading or imbalance of macronutrients, deficiency of trace elements, and infection^1,2,3. Our previous study revealed that in addition to energy overloading, coexistence of infection and intestinal stasis play major roles in IHC in neonates⁴.

Capron et al assumed that intestinal anaerobic bacterial overgrowth could be a significant contributing factor to the occurrence of IHC associated with TPN, and they showed that metronidazole, a drug which suppresses anaerobic intestinal organisms, prevented the occurrence of liver dysfunction during TPN in patients with chronic inflammatory bowel disease (CIBD)^5,6. We also demonstrated a beneficial effect of metronidazole on TPN-associated liver dysfunction in surgical neonates⁷. The results of these studies suggest that intestinal overgrowth of anaerobic bacteria implicated in the occurrence of hepatic dysfunction associated with TPN via certain mediators. Fouin-Fonunet et al., in 1982, suggested a role of lithocholic acid (LCA) in the IHC associated with TPN in patients with CIBD⁸. In this study, in order to determine whether LCA contributes to the occurrence of TPN-associated IHC in neonates, we investigated the serum bile acid fractions.

Subjects and methods

Twenty-five surgical neonates receiving TPN for more than 2 weeks at our institutions between 1984 and 1987 were studied. Their age ranged from 14 to 24 days, with a mean age of 17.4± 2.6 days. Their underlying diseases are shown in Table 1. The nutritional regimen contained 21% glucose, 2.5% amino acid, and electrolytes, vitamins and trace elements. The solution was delivered through a catheter placed in a central vein, and provided 7-100 kcal/kg/day continuously. The amino acid mixture consisted of a formula devised for paediatric use⁹. Blood samples for bile acid analysis were collected in the morning under complete fasting with continuous TPN infusion. Serum was collected by centrifugation immediately after sampling of venous blood, and was stored at -80° C. The serum samples were deproteinised and desalted with SEP-PAK C18¹⁰, and fractionated into three groups using lipophilic gel chromatography¹¹. Each fraction was then applied to a high-performance liquid chromatography system using an ODS column, and the fractionated cholic (CA), chenodeoxycholic (CDCA), deoxycholic (DCA) and lithocholic acids (LCA) were applied to the 3a -hydroxy steroid dehydrogenase-immobilised column¹². NADH that was generated in proportion to 3a -hydroxy bile acids was measured by fluorometer (Figure 1). The external standard of unconjugated, glycine conjugated and taurine conjugated CA, CDCA, DCA and LCA were purchased from P-L Biochemicals, Inc, Milwaukee, Wis. In this system using 3a -HSD, sulfated and glucuronidated bile acids were not detected, since sulphate and glucuronide are conjugated with bile acid at the position of 3a -OH.

All results are expressed as mean± SD. Statistical analysis was performed by Student's t-test, and a probability of 5% or less was considered significant.

Table 1. Underlying diseases.

Oesophageal atresia	6
Duodenal atresia	5
Diaphragmatic hernia	5
Omphal ocele/gastroschisis	3
Midgut volvulus	3
Jejunal/Ileal atresia	2
Hirschsprung's disease	2
Necrotizing enterocolitis	1
CIIPS*1	1
Total	25

*¹CIIPS: chronic idiopathic intestinal pseudo-obstruction syndrome.

Figure 1. Procedure of serum bile acid analysis.-IHPLC: high-performance liquid chromatography *²3a -HSD: 3a -hydroxy steroid dehydrogenase *³NAD: nicotinamide adeninedinucleotide. *⁴NADH: dihydro-NAD.

Table 2. Liver function tests.

	DB*1 (mg/dl)	TBA*2 (nmol/ml)	GOT*3 (iu/l)	GPT*4 (iu/l)	g -GTP*5 (iu/l)	AL-P*6 (iu/l)
non-IHC*7	0.99± 0.59	14.4± 4.5	36.0± 55.0	14.6± 19.4	95.0± 66.1	302.4± 105.7
IHC	3.60± 1.69	71.7± 34.9	107.6± 166.6	23.8± 12.1	108.3± 77.4	239.1± 105.7
P values	P<0.01	P<0.01	NS	NS	NS	NS

*¹DB: direct bilirubin. *²TBA: total bile acid. *³GOT: glutamic ozaloacetic tranaminase. *⁴GPT: glutamic-pyruvic transaminase. *^5g -GTP: g -glutamyl transpeptidase. *⁶AL-P: alkaline phoshatase. *⁷IHC: intrahepatic cholestasis

Results

Liver function tests

Table 2 shows the results of routine liver function tests. Out of 25 patients, 8 (32%) developed cholestasis with a serum concentration of direct bilirubin greater than 2.0 mg/dl during the first month of life. These 8 patients were included in the IHC group, and the other 17 in the non-IHC group. The mean serum concentrations of direct bilirubin in the non-IHC and IHC groups were 0.99 and 2.60 mg/dl, respectively, and those of total bile acid were 14.4 and 71.7 nmol/ml, respectively. Thus, total bile acid concentration was significantly higher in the IHC group. There was no significant difference in glutamic oxaloacetic transaminase, glutamicpyruvic transaminase, g -glutamyl transpeptidase and alkaline phosphatase between the non-IHC and IHC groups.

Serum bile acid fractions

In the non-IHC group, the mean glycine- and taurine conjugated cholic acid were 4.9 and 3.1 nmol/ml, respectively (Table 3). Glycine- and taurine-conjugated chenodeoxycholic acids were 3.1 and 2.3 nmol/ml, respectively. However unconjugated bile acids or secondary bile acids, such as deoxycholic or lithocholic acid were not detected or were detected only in trace levels. In the IHC group, these fractions were detected in high amounts; glycine-conjugated bile acids were approximately four times higher than in the non-IHC group, and taurine-conjugated bile acids were approximately seven times higher than in the non-IHC group. However, in the IHC group, unconjugated and secondary bile acids were not detected or were detected only in trace levels. There was no marked increase in any single fraction in the IHC group.

Discussion

Certain bile acids, in particular the monohydroxylated secondary bile acid, LCA, are known to be cholestatic in a variety of animal species^13-19. The histologic changes produced in LCA-treated animals^13,15,16 are similar to the changes observed in neonates and infants with TPN-associated IHC^4,20-24. This suggests that the histologic changes seen in patients receiving TPN could be attributed to LCA. Intrinsic LCA is normally produced in the small intestine and colon by anaerobic bacterial dehydroxylation of chenodeoxycholic acid²⁵. Recently, several investigators have reported the beneficial effects of metronidazole, a drug which suppresses intestinal anaerobic organisms²⁶, on TPN-associated liver dysfunction in adult patients with CIBD^5,6, in surgical neonates⁷, and in animals²⁷. These reports suggest that hepatotoxic substances produced by anaerobic intestinal bacteria contribute to the occurrence of liver dysfunction during TPN. Fouin-Fortunet et al. showed that in TPN patients with CIBD, the biliary concentration of LCA was significantly higher in patients with hepatic dysfunction than in patients with normal liver function⁸. Balistreri et al. found increased serum levels of sulfated LCA in infants receiving TPN who developed IHC^28,29. Above studies suggest that intrinsic LCA may also cause liver dysfunction. However, no reports have yet shown that intrinsic LCA causes TPN-associated liver dysfunction in infants. We therefore investigated the serum bile acid fraction in neonates on TPN in this study, and attempted to determine whether LCA contributes to the occurrence of TPN-associated liver dysfunction in neonates. Sulfate and glucuronide conjugated bile acid were not analysed; however, Stiehl demonstrated that approximately 20% of serum LCA in infants with cholestasis was non-sulphated and non-glucuronated LCA³⁰, indicating that if sulphated or glucuronated LCA is increased, nonsulphated or glucuronated LCA is also increased. The present study showed that no single fraction, including LCA, was increased, even in patients with IHC, suggesting that LCA is unlikely to be a causative factor in TPN-associated IHC in neonates. In the previous studies reported by Stiehl or Farrell et al., serum LCA concentrations were as high as 2-4 m g/ml (approximately 4-8 mmol/ml). As reported by Stiehl et al., these concentrations were much lower than the amount given to animals to induce cholestasis, 120-240 mol/kg intravenous infusion^14,17,18,19and 0.1-1% in oral feeding^l3,15. And he concluded that it unlikely that such concentrations of monohydroxy bile acids measured in their patients were responsible for the cholestasis³⁰. Furthermore, Cano et al. reported that in 8 non-CIBD patients with TPN-induced cholestasis, biliary LCA was normal; and in 6 cases with normal enterohepatic cycle where bile acid composition was normal, LCA represented less than 1% of total bile acids³¹. They concluded that LCA could not account for the occurrence of cholestasis in their patients.

Table 3. Serum bile acid fractions.

Fraction	non-IHC group	IHC-group	P values
G*1 -CA*7	4.9± 2.5	22.8± 16.6	<0.05
T*2 -CA	3.1± 2.0	22.6± 12.2	<0.05
U*3 -CA	trace	trace	-
G-CDCA*8	3.1± 1.1	12.7± 7.7	<0.05
T-CDCA	2.3± 1.2	16.6± 10.2	<0.05
U-CDCA	trace	trace	-
G- DCA*9	trace	trace	-
T- DCA	trace	trace	-
U- DCA	trace	trace	-
G- LCA*10	trace	trace	-
T- LCA	trace	trace	-
U- LCA	trace	trace	-
Total CA	8.5± 4.1	46.3± 23.2	<0.05
Total CDCA	5.4± 1.6	24.7± 16.6	<0.05
Total DCA	trace	trace	-
Total LCA	trace	trace	-
Total Cly*4	8.0± 3.1	35.5± 23.4	<0.05
Total Tau*5	5.9± 3.0	39.5± 20.4	<0.05
Total Unc*6	trace	trace	-
TBA	14.4± 4.5	76.1± 34.4	<0.05

*¹G: Glycine conjugated. *²: taurine conjugated. *³: Unconjugated *⁴Gly: Glycine conjugated bile acids. *⁵:Taurine conjugated bile acid. *⁶Unc: Unconjugated bile acids. *⁷CA: cholic acid. *⁸CDCA: Chenodeoxycholic acid. *⁹DCA: deoxycholic acid. *¹⁰DCA: lithocholic acid.

Figure 2. Mechanisms of TPN-associated intrahepatic cholestasis.

The results of our previous study showing that coexistence of infection and intestinal stasis were two major contributing factors in the occurrence of IHC during TPN in neonates4, and the beneficial effects of metronidazole on this TPN-associated liver dysfunction567'7 indicate the role of another sepsis-mediated mechanism such as endotoxin release or portal bacteremia (Figure 2). Cholestatic effects of endotoxin have been demonstrated in isolated perfused rat liver by Utili et al.³². Because young infants are highly susceptible to endotoxin, it is reasonable to presume that endotoxin may induce liver dysfunction in neonates on TPN. The observation of pericholangitis in cases of TPN-associated liver dysfunction^4,21,22, supports the possibility that ascending cholangitis is caused by portal bacteremia secondary to intestinal bacterial overgrowth, and results in intrahepatic cholestasis. Thus, portal bacteremia is another possible route for sepsis-mediated mechanism. Further investigations are necessary to clarify the relationship between intestinal or systemic bacterial proliferation and TPN-associated liver dysfunction in infants.

Correspondence: Akio Kubota, 840 Murodo-cho, Izumi, Osaka 590 02, Japan.

References

1. Merritt RJ. Cholestasis associated with total parenteral nutrition. I Pediatr Gastroenter Nutr 1986; 5:9-22.
2. Bell RL, Ferry GD, Smith EO, et al. Total parenteral nutrition-related cholestasis in infants. JPEN 1986; 10:356-359.
3. Drongowski RA, Coran AG. An analysis of factors contributing to the development of total parenteral nutrition-induced cholestasis. JPEN 1989; 13: 586-589.
4. Kubota A, Olcada A, Nezu R, et al. Hyperbilirubinemia in neonates associated with total parenteral nutrition. JPEN 1988; 12: 602-606.
5. Capron JP, Gineston JL, Herve MA, et al. Metronidazole in prevention of cholestasis associated with total parenteral nutrition. Lancet 1983; 1: 446-447.
6. Lambert JR, Thomas SM. Metronidazole prevention of serum liver enzyme abnormalities during total parenteral nutrition. 1985; JPEN 9: 501-503.
7. Kubota A, Okada A, Imura K, et al. The effect to metronidazole on TPN-associated liver dysfunction in neonates. J Pediatr Surg 1990; 25: 618-621.
8. Fouin-Foutunet H, LeOuernec L, Erlinger S, et al. Hepatic alterations during total parenteral nutrition in patients with inflammatory bowel disease: possible consequence of lithocholic toxicity. Gastroenterol 1982; 82: 932-937.
9. Imura K, Fukui Y, Kawahara H, et al. Clinical studies on a newly devised amino acid solution for neonates. JPEN 12: 1988; 496-504.
10. Goto J, Kato H, Saruta Y, et al. Separation and determination of bile acids in human bile by high-performance liquid chromatography. J Liquid Chromatogr 1980; 3: 991-1003.
11. Goto J, Hasaegawa M Kato H, et al. A new method for simultaneous determination of bile acids in human bile without hydrolysis. Clinical Chimica Acta 1978; 87: 141-147.
12. Kamada S, Maeda M, Tsuji A. Separation and determination of bile acid by high-performance liquid chromatography using immobilized 3a -hydroxy steriod dehydrogenase and an electrochemical detector. J Chromatogram 1982; 239: 773-783.
13. Hunt RD, Leveille GA, Sauberlich HE. Dietary bile acids and lipid metabolis. III. Effects of lithocholic acid in mammalian species. Proc Soc Exp Biol Med 1964; 115: 277-280.
14. Javitt NB. Cholestasis in rats induced by taurolithocholate. Nature 1966; 210: 1262-1263.
15. Palmer RH, Hruban Z. Production of bile duct hyperplasia and gallstone by lithocholic acid. J Clin Invest 1966; 45: 1255-1267.
16. Fischer CD, Cooper NS, Rothschild MA, et al. Effect of dietary chenodeoxycholic acid and lithocholic acid in the rabbit. Am J Dig Dis 1974; 19: 877-886.
17. Layden TJ, Boyer JL. Taurolithocholate-induced cholestasis: taurocholate, but not dehydrocholate, reverses cholestasis and bile canalicular membrane injury. 1977; 73: 120-128.
18. Kakis G, Yousef IM. Pathogenesis of lithocholate- and taurocholate-induced intrahepatic cholestasis in rats. Gastroenterol 1978; 75: 595-607.
19. Yousef IM, Tuchweber B, Vonk RJ, et al. Lithocholate cholestasis-sulfated glycolithocholate-induced intrahepatic cholestasis in rats. Gastroenterol 1981, 80: 233-241.
20. Cohen C, Olsen MM. Pediatric total parenteral nutrition: liver histopathology. Arch Patrol Lab Med 1980; 105: 152-156.
21. Dahms BB, Halpin TC Jr. Serial liver biopsies in parenteral nutrition-associated cholestasis of early infancy. Gastroenterol 1981; 81: 136-144.
22. Benjamin DR. Hepatobiliary dysfunction in infants and children associated with long-term total parenteral nutrition. A clinical-pathologic study. Am J Clin Pathol 1981; 76: 276-283.
23. Hodes JE, Grosfeld JL, Weber TR, et al. Hepatic failure in infants on total parenteral nutrition (TPN): Clinical and histopathologic observations. J Pediatr Surg 1982; 17: 463-468.
24. Hughes CA, Talbot IC, Ducker DA, et al. Total parenteral nutrition in infancy: effect on the liver and suggested pathogenesis. Gut 1983; 24: 241-248.
25. Macdonald IA, Bokkenheuser VD, Winter J, et al. Degradation of steroids in the human gut. J Lipid Res 1983; 24: 675-700.
26. Goldman P. Metronidazole. N Engl J Med 1980; 20: 1212-1218.
27. Freund HR, Muggia-Sullam M, LaFrance R, et al. A possible beneficial effect of metronidazole in reducing TPN-associated liver function derangements. J Surg Res 1985; 38: 356-363.
28. Balistreri WF, Suchy FJ, Farell MK, et al. Pathologic versus physiologic cholestasis: elevated serum concentration of a secondary bile acid in the presence of hepatobiliary disease. J Pediatr 1981; 98: 399-402.
29. Farell MK, Balistreri WF, Suchy FJ: Serum-sulfated lithocholate as an indicator of cholestasis during parenteral nutrition in infants and children. JPEN 1982; 3: 30-33.
30. Stiehl A, Becker M, Czygan P, et al. Bile acids and their sulphated and glucuronidated derivatives in bile, plasma, and urine of children with intrahepatic cholestasis: effects of phenobarbital treatment. Eur J Clin Invest 1980; 10: 307-316.
31. Cano N, Gerolami A. Intrahepatic cholestasis during total parenteral nutrition. Lancet 1983; 1 (8331): 985.
32. Utili R, Abernathu CO, Zimmermann HJ. Cholestatic effects of escherichia coli endotoxin on the isolated perfused rat liver. Gastroenterol 1976; 70: 248-253.

Copyright © 1996 [Asia Pacific Journal of Clinical Nutrition]. All rights reserved.
Revised: January 14, 1999 .

to the top