Regardless of the cause, neonates with intestinal failure (IF) require parenteral nutrition (PN) to survive, grow, and develop (1). PN is often associated with severe complications; including intestinal failure–associated liver disease (IFALD) (2). The incidence rates of IFALD, mainly reported in retrospective studies, vary from 30% to 60% in neonates receiving PN (2–7).
Various causative factors have been evaluated for significance in contributing to IFALD, including duration of PN, septic episodes, prematurity, low birth weight, massive intestinal resection, lack of enteral nutrients (8,9), and different components of PN including plant sterols (PS) (10–12), and the etiology is proposed to be multifactorial (13). Normally, only 5% to 10% of dietary PS is absorbed from the small intestine (14), and their serum levels reflect intestinal cholesterol absorption (15–17). Cholestanol (as ratio to cholesterol in serum), partly of dietary, partly of cholesterol synthesis origin, reflects serum levels of PS and intestinal absorption of cholesterol (16,17) and is a sensitive indicator of cholestasis in several cholestatic diseases including primary biliary cirrhosis and biliary atresia (18–21). The serum ratios of PS markedly increase in adults and children on PN with IFALD (10–12,22,23). Transition from soy-based to olive oil–based PN containing less PS decreases elevated liver enzymes, whereas fish oil–based emulsions devoid of PS have been suggested to reverse cholestasis in neonates (24,25). In vitro, PS, especially stigmasterol, antagonizes nuclear farnesoid X receptor (FXR) regulating bile acid homeostasis in hepatocytes (26,27). Moreover, in our previous study in children with IF on long-term PN, serum PS levels associated with glutamyl transferase (GT) and appeared to mirror liver fibrosis (28). The contribution of PS to IFALD in neonates with immature liver function and biliary secretion has not been evaluated before.
To this end, we performed a prospective controlled nationwide follow-up study among neonates with IF during and after discontinuation of PN with known amount of parenteral PS and nutritional content of PN. Serum PS and other noncholesterol sterols, including cholesterol precursors, were repeatedly measured to evaluate the significance of PS in neonatal IFALD by linking them to the biochemical markers of cholestasis and liver function. In addition, the results of neonates were compared with those of children on PN with more mature liver function and healthy controls to determine putative differences in their PS metabolism, and, furthermore, to understand the influence of parenterally administered PS on IFALD.
All patients younger than 17 years receiving prolonged PN (>28 days), including daily infusions of carbohydrates and protein with or without lipids, in Finland's 5 university hospitals (Helsinki, Tampere, Turku, Kuopio, and Oulu) between November 2009 and November 2010 were prospectively identified and recruited. Patients with a malignant primary disease were excluded.
Blood samples were collected for analyses of serum cholesterol, squalene and noncholesterol sterols, including PS, cholesterol precursor sterols and cholestanol, fat-soluble vitamins, and biochemical markers of liver function at the time of inclusion, 2 and 4 weeks after newly started PN, and thereafter every 12 weeks during PN for 1 year. If a patient weaned off PN during the follow-up, blood samples were collected 1 month after discontinuation of PN. The exact composition and amount of PN, medications including antimicrobial therapy, growth, and septic episodes were recorded until the end of follow-up or death. Antimicrobial therapy was recorded as intravenous antibiotic days, including intravenous antibiotic use 1 month before inclusion in children and between birth and inclusion in neonates. Collected clinical data included gestational age, birth weight and height, underlying diagnoses, surgical procedures and complications, anatomy and length of the remaining bowel, central venous catheter complications, and patient outcome. Anatomical details of the remaining bowel were recorded from the operative records. Percentage of age-adjusted small intestinal and colon length was assessed according to Struijs et al (29).
A total of 39 patients on prolonged PN, 28 neonates and 11 children, were identified (Table 1). At inclusion, the mean age of neonates was 50 (28–126) days and that of older children was 6.9 (2.1–16.6) years. The underlying reason for prolonged PN was IF in 71% of neonates and in all of the children. Anatomy of the remaining bowel is displayed in Table 1. Access to health care, including antenatal care, is available for all and the costs are covered by the national insurance system in Finland. The OECD ranking for infant mortality in 2007 was 2.7/1000 live births in Finland, whereas the average for OECD countries was 4.9 (30).
All of the patients were included in the epidemiological part of the study, including diagnosis, sex, birth weight, gestational age, associated anomalies, and outcome. Informed consent for blood samples and collection of clinical data was received from all of the children and 21 neonates (21/28), including liver enzymes in 19 (19/28) and noncholesterol sterols and squalene in 18 (18/28) neonates. One month after discontinuation of PN, 15 neonates (15/23) participated in blood tests.
Healthy day-case surgery (ie, inguinal hernia, umbilical hernia, undescended testes) patients (10 infants and 22 older children) matched for age and sex without evidence of intestinal disease, diabetes, or dyslipidemia served as controls. Neonates receiving prolonged PN were compared with children receiving PN and with healthy controls.
Serum cholesterol and noncholesterol sterols, including PS (stigmasterol, sitosterol, avenasterol, campesterol), cholestanol and cholesterol precursor sterols (cholestanol, desmosterol, lathosterol), and squalene, were measured from nonsaponifiable material by gas-liquid chromatography–gas-liquid chromatography, on a 50-m-long SE-30 nonpolar capillary column (Ultra 1 Column, Agilent Technologies, Palo Alto, CA), with 5α-cholestane as the internal standard (31,32). Noncholesterol sterols and squalene in the serum were expressed as ratios to the cholesterol concentration of the same gas-liquid chromatography run, that is, 100× μg/mg of cholesterol (called ratios in the text), to exclude the effects of varying serum lipoprotein concentration. Parenteral PS and cholesterol intake were calculated based on parenteral emulsions analyzed for PS and cholesterol as described above and published previously (28).
Serum alanine aminotransferase (ALT), GT, aspartate aminotransferase (AST), bilirubin, bile acids, thromboplastin time, and vitamins A, E, and 25-hydroxyvitamin D were determined by routine hospital methods.
Definition of IFALD
IFALD was defined as >1.5 times increase above the upper limit of normal of at least 2 of the 3 following parameters during follow-up: direct bilirubin >35 μmol/L, GT >75 U/L, ALT >65 U/L (2). According to this definition, the patients were divided into IFALD and non-IFALD groups.
Incidence and mortality rates were calculated according to the official statistics of the Finnish population (age 0–16 years) in 2009 (33,34) and patient statistics of the Finnish neonatal intensive care units. Unless otherwise stated, the data are expressed as medians (range). Nonparametric methods were used to compare continuous variables between groups (Mann-Whitney U test). Correlations were analyzed with the Spearman rank correlation test. P < 0.05 was considered significant.
An informed consent was obtained from each patient, control, and/or their parents included in the study. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki, as reflected in a priori approval by the ethical committees of the respective university hospitals.
Epidemiology of PN in Neonates and Children
A total of 39 patients, 28 neonates and 11 children, were identified (Table 1). The pediatric units of the 5 university hospitals treat all prematurely born newborns and pediatric patients on prolonged PN in Finland allowing all-inclusive nationwide patient identification. In the study population yearly nationwide incidence of PN continuing for 28 days was 38.2 (patients/million inhabitants younger than 17 years). Of all of the neonates born during follow-up in Finland, 0.05% (28/60,430, 46:100,000 live births) and 2.0% (24/1223) of those born before 37 weeks of gestation treated in Finnish neonatal intensive care units required prolonged PN.
Septic Episodes and Use of Intravenous Antibiotics
Blood culture positive septic episodes were more frequent among neonates (61%, 1.6/patient) than in children (18%, 1.0/patient) during PN (P < 0.05). The most prevalent pathogen was Staphylococcus epidermidis (56% of blood cultures in neonates, 33% in children). Neonates received more often intravenous antibiotics before inclusion (30 antibiotic days [7–97] and 0 [0–15], P < 0.001), but not during follow-up (5 antibiotic days [0–64] and 0 [0–120], P = 0.367), compared with children.
The total duration of PN was 65 days (range 30–388) in neonates and 4.6 years (range 148 days–17.5 years) in children. A total of 23 (82%) neonates achieved intestinal autonomy after a median of 49 days (range 30–300) on PN and 1 child after successful intestinal transplantation. The percentage of total daily energy provided in PN was 70% (range 30%–100%) in neonates and 25% (range 5%–100%) in children (P = 0.142). All of the neonates received PN fat as olive oil–based regimen (Clinoleic, Baxter, Helsinki, Finland), 4 of them combined with a median dose of 0.82 g · kg−1 · day−1 (0.0–1.0) of fish oil–based PN emulsion (Omegaven, Fresenius Kabi, Germany). In children, PN fat was olive oil–based emulsion in 10 patients and soybean oil–based (Intralipid, Fresenius Kabi, Sweden) in 1 patient. Per kilogram of body weight neonates received higher amounts of parenteral energy, fat, carbohydrates, stigmasterol, and avenasterol than children (P < 0.05–0.001) (Table 2).
Liver Function Tests
Neonates had significantly higher median serum GT, direct bilirubin, and bile acids during PN compared with children. Serum prealbumin and thromboplastin time were under the normal range in 33% and 89% of neonates, respectively, and significantly lower when compared with children (Table 3).
Serum Fat-soluble Vitamins During PN
In neonates and children, the serum levels of vitamin E (29 [15–47] and 18 [9–38] μmol/L) and 25-hydroxyvitamin D (55 [10–13] and 48 [18–98] nmol/L) were within the normal range. The median vitamin A level was close to the normal range in neonates (0.9 [0.3–2.0] μmol/L) and in the normal range in children (1.2 [0.8–2.2] μmol/L].
Serum Noncholesterol Sterols and Squalene During PN
The median serum ratios of cholestanol and PS, that is, stigmasterol, sitosterol, avenasterol, campesterol, and total serum PS, were increased among neonates in relation to healthy controls and children by 2- to 22- and 2- to 5-fold (P < 0.005), respectively. Of the cholesterol precursor sterols, only lathosterol was doubled in neonates when compared with healthy controls (P < 0.005, Table 4). The median serum ratio of squalene was increased compared with controls (P < 0.005, Table 4).
Comparison of Neonates With And Without IFALD
IFALD occurred more frequently in neonates (63%, 12/19) than in children (27%, 3/11, P < 0.05). The neonates with IFALD had significantly higher serum ALT, AST, bile acids, and direct bilirubin compared with neonates without IFALD (Table 3). Of the various risk factors of IFALD and variables assessed, the serum ratios of cholestanol, stigmasterol, sitosterol, avenasterol, and total PS differed significantly between neonates with and without IFALD (Table 5).
Relation of PN, Serum PS, and Liver Function During PN in Neonates
The total duration of PN positively reflected serum ratios of cholestanol, stigmasterol, avenasterol (Table 6, Figs. 1 and 2), ALT (r = 0.557, P < 0.05), and AST (r = 0.483, P < 0.05), the highest serum sitosterol (r = 0.502, P < 0.05), and the highest total PS (r = 0.509, P < 0.05). The PN percentage of total energy was positively related to the highest serum stigmasterol (r = 0.483, P < 0.05) and campesterol ratios (Table 6). The amount of PN fat, carbohydrates, protein, and total or individual PS were not associated with serum markers of liver function or total or individual PS ratios (data not shown).
The serum ratios of cholestanol, stigmasterol, sitosterol, and avenasterol were positively associated with serum AST, ALT, and direct bilirubin. Serum cholestanol positively reflected serum bile acids (r = 0.615, P < 0.05) and stigmasterol ratio (r = 0.713, P < 0.005). Serum total PS ratio correlated with serum ALT and direct bilirubin (Table 6).
The remaining age-adjusted percentage of small bowel was inversely related to serum cholestanol (r = −0.555, P < 0.05) and bile acids (r = −0.604, P < 0.05). The neonates without ileocecal valve (ICV; n = 4) had higher serum cholestanol ratios (607 [510–743] 100× μg/mg of cholesterol) than those (n = 14) with ICV (406 [235–5503] 100× μg/mg of cholesterol) during PN (P < 0.01). The remaining age-adjusted colon length was negatively associated with the highest serum ALT, stigmasterol, avenasterol, sitosterol, and total PS (r = −0.539 to −0.476, P < 0.05).
Liver Function and Serum PS 1 Month After Discontinuation of PN in Neonates
One month after discontinuation of PN, 25% of neonates still fulfilled the criteria for IFALD with 4.2, 2.2, and 47 times higher serum stigmasterol and cholestanol ratios and direct bilirubin level in relation to those of neonates with normalized liver function (Table 7). The stigmasterol ratio correlated positively with the cholestanol ratio (r = 0.632, P < 0.05) after discontinuation of PN.
In this prospective nationwide study we assessed the significance of parenteral PS on IFALD among neonates. IFALD was more frequent among neonates with significantly higher serum PS, in relative terms especially stigmasterol, during PN with an association with elevated serum ALT, AST, and direct bilirubin. Moreover, serum stigmasterol and cholestanol levels remained high after weaning off PN in neonates with persisting IFALD. Although the number of patients is limited, this is the first prospective nationwide study on the interplay of prolonged PN, PS, and IFALD in neonates.
In our previous study on IF children receiving PN long term with mean age of 6.3 years, the distribution of individual PS in the serum closely paralleled that of infused lipid regimen (28). In the present study, serum PS during PN was higher in neonates than in children receiving PN or healthy controls and the distribution of individual PS was weighted with high stigmasterol levels in neonates with IFALD during and after weaning off PN. In addition, the duration of PN, but not the amount of PN lipids or PS, correlated with serum stigmasterol, avenasterol, cholestanol, and aminotransferases. Serum PS (stigmasterol, sitosterol, and avenasterol), aminotransferases, and direct bilirubin were interrelated during PN. In general, serum PS reflects the balance between input and biliary excretion and is expected to accumulate as a result of liver damage (35). In the present study, the high levels of serum PS suggest that PS, especially stigmasterol, accumulates in neonates on PN presumably due to insufficient hepatobiliary excretion capacity.
Previously, PN-associated liver disease has been attributed to defects in the bile canalicular ABC-transporter family of proteins, including multidrug resistance 1 and 2 (mdr1, mdr2) genes encoding P-glycoproteins, responsible for transport of bile components out of hepatocytes (36). The mdr2-deficient mice are described as developing a cholestatic liver disease and an inability of the liver to secrete phospholipids into bile (37,38). In a piglet model, the administration of PN reduced the basal bile flow and bile acid secretion (39). Tazuke et al (38) observed a significant decline in the expression of mdr2 mRNA and protein expression during PN administration in hepatic samples of mice. In a piglet model, Iyer et al (40) found a progressive accumulation of PS in serum, liver, and bile, together with raised serum bile acid levels and decreased bile acid secretion. PS are eliminated by biliary secretion via 2 canalicular ATP-binding cassette transporters, ABCG5 and ABCG8, by hepatocytes (35). In healthy humans, the hepatic clearance rate of sitosterol is higher than that of campesterol but lower in both compared with that of cholesterol (41). Interestingly, PS, especially stigmasterol, antagonizes the nuclear FXR regulating mdr2 gene and further bile acid homeostasis in hepatocytes in vitro (26,27,42). Gene knockout studies demonstrate that mice lacking the FXR hepatoprotective mechanisms are ultrasensitive to the bile acid–induced liver injury, and, furthermore, treatment of rats with FXR-agonist protects against cholestasis (43,44). Our results of the significantly increased serum plant sterol levels, especially that of stigmasterol, in neonates with IFALD may be associated with pathogenesis of PN-associated liver disease, as suggested by the previous findings in the animal and in vitro studies.
In the present study, due to the small numbers and low incidence of sepsis we did not find an association between IFALD and its known risk factors, including septic episodes, birth weight, and gestation age (8,9,13). Despite this, an association between elevated serum PS levels, especially stigmasterol, was identified, suggesting that it may be an important independent risk factor for IFALD in neonates. A major bowel resection is an established risk factor for IFALD (13,22), and we found some evidence that interruption of the enterohepatic circulation caused by resection of the terminal ileum and ICV exacerbates cholestasis by removing the negative feedback normally exerted on 7α-hydroxylation of cholesterol, which is the rate-limiting step in bile acid production (45). Serum cholestanol had a positive correlation with serum AST, ALT, direct bilirubin, and bile acids, and a negative correlation with the remaining age-adjusted bowel length supporting this concept. The close positive relation between the serum ratio of cholestanol and bilirubin further supports the previous findings, suggesting that the serum cholestanol/cholesterol ratio is a useful marker of cholestasis in IFALD (18–21).
One month after discontinuation of PN, serum PS levels remained high in neonates, especially those with persistent IFALD putatively mirroring the high bilirubin levels and cholestasis. There are no previous reports of serum PS levels after discontinuation of PN in neonates, but the serum bilirubin level is reported to normalize by 4 months after a 1-month plateau phase after discontinuation of PN (46). Further studies, with longer follow-up after discontinuation of PN, are needed to detect the clearance rate of serum PS and significance of high remaining serum PS on persisting IFALD in neonates.
In conclusion, IFALD is frequent among neonates on PN, with an association to markedly increased serum PS compared with healthy neonatal controls and children on PN with more mature liver function. The high levels of serum PS suggest that PS, especially stigmasterol, accumulate in neonates on PN, presumably due to insufficient hepatobiliary excretion capacity. The serum levels of PS remain high after the discontinuation of PN in neonates with persistent IFALD. Elevated serum PS level, especially stigmasterol, may be an independent risk factor for IFALD.
1. Mirtallo J, Canada T, Johnson D, et al. Safe practices for parenteral nutrition. JPEN J Parenter Enteral Nutr
2. Kumpf VJ. Parenteral nutrition-associated liver disease in adult and pediatric patients. Nutr Clin Pract
3. Kelly DA. Liver complications of pediatric parenteral nutrition: epidemiology. Nutrition
4. Sondheimer JM, Asturias E, Cadnapaphornchai M. Infection and cholestasis in neonates with intestinal resection and long-term parenteral nutrition. J Pediatr Gastroenterol Nutr
5. Suita S, Masumoto K, Yamanouchi T, et al. Complications in neonates with short bowel syndrome and long-term parenteral nutrition. JPEN J Parenter Enteral Nutr
6. Misra S, Ament ME, Vargas JH, et al. Chronic liver disease in children on long-term parenteral nutrition. J Gastroenterol Hepatol
7. Kubota A, Yonekura T, Hoki M, et al. Total parenteral nutrition-associated intrahepatic cholestasis in infants: 25 years’ experience. J Pediatr Surg
8. Beath SV, Davies P, Papadopoulou A, et al. Parenteral nutrition-related cholestasis in postsurgical neonates: multivariate analysis of risk factors. J Pediatr Surg
9. Spencer AU, Neaga A, West B, et al. Pediatric short bowel syndrome—redefining predictors of success. Ann Surg
10. Clayton PT, Bowron A, Mills KA, et al. Phytosterolemia in children with parenteral nutrition-associated cholestatic liver disease. Gastroenterology
11. Clayton PT, Whitfield P, Iyer K. The role of phytosterols in the pathogenesis of liver complications of pediatric parenteral nutrition. Nutrition
12. Llop M, Virgili N, Moreno-Villares JM, et al. Phytosterolemia in parenteral nutrition patients: implications for liver disease development. Nutrition
13. Kelly DA. Intestinal failure-associated liver disease: what do we know today? Gastroenterology
14. Salen G, Ahrens EK Jr, Grundy SM. Metabolism of beta-sitosterol in man. J Clin Invest
15. Tilvis RS, Miettinen TA. Serum plant sterols and their relation to cholesterol absorption. Am J Clin Nutr
16. Nissinen MJ, Gylling H, Miettinen TA. Responses of surrogate markers of cholesterol absorption and synthesis to changes in cholesterol metabolism during various amounts of fat and cholesterol feeding among healthy men. Br J Nutr
17. Miettinen TA, Tilvis RS, Kesäniemi YA. Serum cholestanol and plant sterol levels in relation to cholesterol metabolism in middle-aged men. Metabolism
18. Nikkilä K, Höckerstedt K, Miettinen TA. High cholestanol and low campesterol-to-sitosterol ratio in serum of patients with primary biliary cirrhosis before liver transplantation. Hepatology
19. Gylling H, Vuoristo M, Färkkilä M, et al. The metabolism of cholestanol in primary biliary cirrhosis. J Hepatol
20. Nikkilä K, Nissinen MJ, Gylling H, et al. Serum sterols in patients with primary biliary cirrhosis and acute liver failure before and after liver transplantation. J Hepatol
21. Pakarinen MP, Lampela H, Gylling H, et al. Surrogate markers of cholesterol metabolism in children with native liver after successful portoenterostomy for biliary atresia. J Pediatr Surg
22. Cavicchi M, Beau P, Crenn P, et al. Prevalence of liver disease and contributing factors in patients receiving home parenteral nutrition for permanent intestinal failure. Ann Intern Med
23. Hallikainen M, Huikko L, Kontra K, et al. Effect of parenteral serum plant sterols on liver enzymes and cholesterol metabolism in a patient with short bowel syndrome. Nutr Clin Pract
24. Puder M, Valim C, Meisel JA, et al. Parenteral fish oil improves outcomes in patients with parenteral nutrition-associated liver injury. Ann Surg
25. Pálová S, Charvat J, Kvapil M. Comparison of soybean oil- and olive oil-based lipid emulsions on hepatobiliary function and serum triacylglycerols level during realimentation. J Int Med Res
26. Wang H, Chen J, Hollister K, et al. Endogenous bile acids are ligands for the nuclear receptor FXR/BAR. Molecular Cell
27. Carter BA, Taylor OA, Prendergast DR, et al. Stigmasterol, a soy lipid-derived phytosterol, is an antagonist of the bile acid nuclear receptor FXR. Pediatr Res
28. Kurvinen A, Nissinen MJ, Gylling H, et al. Effects of long-term parenteral nutrition on serum lipids, plant sterols, cholesterol metabolism, and liver histology in pediatric intestinal failure. J Pediatr Gastroenterol Nutr
29. Struijs MC, Diamond IR, de Silva N, et al. Establishing norms for intestinal length in children. J Pediart Surg
31. Miettinen TA, Koivisto P. Paumgartner G, Stiehl A, Gerok W. Non-cholesterol sterols and bile acid production in hypercholesterolemic patients with ileal by-pass. Bile Acids and Cholesterol in Health and Disease
. Lancaster, PA:MTP Press; 1983. 183–187.
32. Miettinen TA. Cholesterol metabolism during ketoconatzole treatment in man. J Lipid Res
35. Hazard SE, Patel SB. Sterolins ABCG5 and ABCG8: regulators of whole body dietary sterols. Pflugers Arch
36. Trauner M, Meier PJ, Boyer JL. Molecular pathogenesis of cholestasis. N Engl J Med
37. Smit JJ, Schinkel AH, Oude Elferink RP, et al. Homozygous disruption of the murine mdr2 P-glycoprotein gene leads to a complete absence of phospholipid from bile and to liver disease. Cell
38. Tazuke Y, Kiristioglu I, Heidelberger KP, et al. Hepatic P-glycoprotein chances with total parenteral nutrition administration. JPEN J Parenter Enteral Nutr
39. Duerksen DR, Van Aerde JE, Chan G, et al. Total parenteral nutrition impairs bile flow and alter bile composition in newborn piglet. Dig Dis Sci
40. Iyer KR, Spitz L, Clayton P. New insights into mechanisms of parenteral nutrition-associated cholestasis: role of plant sterols. J Pediatr Surg
41. Sudhop T, Sahin Y, Lindenthal B, et al. Comparison of the hepatic clearances of campesterol, sitosterol, and cholesterol in healthy subjects suggests that efflux transporters controlling intestinal sterol absorption also regulate biliary secretion. Gut
42. Kast HR, Goodwin B, Tarr PT, et al. Regulation of multidrug resistance-associated protein 2 (ABCC2) by the nuclear pregame X receptor, farnesoid X-activated receptor, and constitutive androstane receptor. J Biol Chem
43. Sinal CJ, Tohkin M, Miyata M, et al. Targeted disruption of the nuclear receptor FXR/BAR impairs bile acid and lipid homeostasis. Cell
44. Liu Y, Binz J, Numerick MJ, et al. Hepatoprotection by farnesoid X receptor agonist GW4064 in rat models of intra- and extrahepatic cholestasis. J Clin Invest
45. Russell DW. The enzymes, regulation, and genetics of bile acid synthesis. Annu Rev Biochem
46. Pichler J, Horn V, MacDonald S, et al. Sepsis and its etiology among hospitalized children less than 1 year of age with intestinal failure on parenteral nutrition. Transplant Proc