566853

research-article2014
PENXXX10.1177/0148607114566853Journal of Parenteral and Enteral NutritionWang et al

Original Communication
Journal of Parenteral and Enteral
Nutrition
Effect of an Olive Oil–Based Lipid Emulsion Compared Volume XX Number X
Month 201X 1­–9
With a Soybean Oil–Based Lipid Emulsion on Liver © 2015 American Society

Chemistry and Bile Acid Composition in Preterm for Parenteral and Enteral Nutrition
DOI: 10.1177/0148607114566853
Infants Receiving Parenteral Nutrition: A Double-Blind, jpen.sagepub.com
hosted at
Randomized Trial online.sagepub.com

Ying Wang, MD, PhD1,2*; Ke-Jun Zhou, PhD2,3*; Qing-Ya Tang, MD1,2;
Li Hong, MD, PhD4; Yi Feng, MD, PhD4; Li-Na Lu, MD1; Wei-Ping Wang, MD1;
and Wei Cai, MD, PhD1,2,3

Abstract
Background: The pathogenesis of parenteral nutrition (PN)–associated liver dysfunction is multifactorial. Lipid emulsions may be one of
the putative mechanisms. Our aim was to comparatively assess the effect of parenteral olive oil– and soybean oil–based lipid emulsions on
liver chemistry and bile acid composition in preterm infants. Methods: We performed a double-blind, randomized clinical study in which
103 preterm infants were randomly assigned to PN using either soybean oil–based lipid emulsion (SO; n = 51) or olive oil (OO)–based
lipid emulsion (OO; n = 52). The primary end point was liver chemistry. The secondary end point was the plasma bile acid composition.
Results: One hundred infants completed this study. In the SO group, the serum direct bilirubin was significantly higher after PN for 7 days
compared with the OO group. Bile acids increased over time in both treatment groups. However, specific differences in the change in
bile acid composition over time were noted between groups. Conclusions: Differences in direct bilirubin and bile acid composition were
observed over time between the 2 groups. Considering the long-term use of lipid emulsions in higher risk babies, these findings might be
useful for understanding the pathogenesis of PN-associated liver dysfunction. (JPEN J Parenter Enteral Nutr. XXXX;xx:xx-xx)

Keywords
olive oil; parenteral nutrition; liver chemistry; bile acid; lipid emulsion

Clinical Relevancy Statement hepatobiliary dysfunction remains a significant life-threatening

Intravenous lipid emulsions (ILEs) serve as a major source of
nonprotein energy. Their benefits include reduction of the poten-
tial side effects of high glucose intake, provision of required
essential fatty acids (EFAs), and improved nitrogen balance.
Preterm neonates are more vulnerable to parenteral nutrition–
associated cholestasis (PNAC) due to physiologic immaturity
and exposure to other risk factors during their stay in the inten-
sive care unit. Previous studies have shown that olive oil–based
lipid emulsion is safe and well tolerated by preterm neonates. In
this study, we compared the effects of olive oil– and soybean
oil–based lipid emulsions used in parenteral nutrition on the
liver chemistry and bile acid composition of preterm infants and
found that the direct bilirubin and bile acid composition were
different between the 2 groups. Given a paucity of such data, this
clinical trial adds information for evolving our understanding of
the mechanisms of PNAC in neonates. However, longer treat-
ment trials of olive oil–based lipid emulsions are required.
Introduction
The development of parenteral nutrition (PN) has improved
the prognosis for extremely premature infants. However,
complication associated with PN. Cholestasis may develop in
40%–60% of infants1 and up to 85% of neonates2 who require
long-term PN. Evidence of liver dysfunction may occur as
early as 14 days after PN is initiated in neonates,3 and up to 25%
From the 1Department of Clinical Nutrition, Xin Hua Hospital, School of
Medicine, Shanghai Jiao Tong University, Shanghai, China; 2Shanghai
Key Laboratory of Pediatric Gastroenterology and Nutrition, Shanghai,
China; 3Shanghai Institute of Pediatric Research, Shanghai, China; and
4
Department of Clinical Nutrition, Shanghai Children’s Medical Center,
School of Medicine, Shanghai Jiao Tong University, Shanghai, China.
*These authors contributed equally to this work and share first
authorship.
Financial disclosure: This study was supported in part by grants from
the National Natural Science Foundation of China (No. 81100631) and
Shanghai Key Laboratory of Pediatric Gastroenterology and Nutrition
(No. 11DZ2260500).
Received for publication August 12, 2014; accepted for publication
December 5, 2014.
Corresponding Author:
Wei Cai, MD, PhD, Department of Clinical Nutrition, Xin Hua Hospital,
Kongjiang Rd 1665, Shanghai 200092, China.
Email: caiw1978@163.com

Downloaded from pen.sagepub.com at Universidad Nacional Aut Mexic on January 18, 2015
2 Journal of Parenteral and Enteral Nutrition XX(X)
of children with parenteral nutrition–associated cholestasis
(PNAC) progress to biliary cirrhosis and liver failure and
require transplantation.4
The development of cholestasis is related to a number of fac-
tors. In neonates, it is associated with prematurity, recurrent
sepsis (particularly catheter-related sepsis), use of soybean lipid
emulsions, and lack of enteral feeding. The association of cho-
lestasis with parenteral lipid infusions has been well studied.5,6
In adults, cholestasis is related to the volume of parenteral lipid
infusion (>1 g·kg−1·d−1), while in children it is related to the
type of lipid and the method of administration.5,7,8 The contribu-
tion of certain components of PN, particularly lipids, to the
development of PNAC is a topic of much debate. Soybean for-
mulations are composed mainly of ω-6 polyunsaturated fatty
acids (PUFAs), which may contribute to the development of
PNAC due to their proinflammatory properties9 via the accu-
mulation of hepatic phospholipids and phytosterols, which
impair biliary secretion.5,10,11 Some centers have attempted to
modify the lipid administration method by reducing the amount
of lipids,12 cycling PN,13,14 or temporarily removing lipids15 in
an effort to prevent or treat PNAC. The use of lipid emulsions
containing fish oil has recently gained support, as fish oil con-
tains ω-3 PUFAs, which are anti-inflammatory in nature.9 The
treatment of PNAC with fish oil–based emulsions seems to be
safe and effective.15 However, fish oil–based emulsion is
administered at a lower dose (1 g·kg−1·d−1) than soybean oil–
based lipid emulsion.16 Another alternative is an olive oil–based
lipid emulsion, which is prepared from a mixture of soybean oil
(20%) and olive oil (80%); it contains a lower proportion (20%)
of PUFAs and a high proportion of monounsaturated fatty acids
(MUFAs, 60%). Moreover, a study by Xu et al17 showed that
the total phytosterol intake from the olive oil and soybean oil
lipid emulsion was 137 mg/d and 220 mg/d, respectively, in
patients receiving 100 g/d of lipid. In addition, several studies
have reported that olive oil–based emulsion was safe and well
tolerated by preterm neonates.18,19
This prospective randomized double-blind study compared
the effects of olive oil– and soybean oil–based lipid emulsions
used in PN on the liver chemistry and bile acid composition of
preterm infants.
Patients and Methods
Patients
We screened 118 preterm infants admitted to the neonatal inten-
sive care unit (NICU) of Xin Hua Hospital and Shanghai
Children’s Medical Center (School of Medicine, Shanghai Jiao
Tong University) for inclusion in this study from February 2012
to July 2013. The inclusion criteria were birth weight <2000 g,
admission to the hospital within 72 hours after birth, and admin-
istration of PN for 14 days or more. The exclusion criteria were
administration of PN before screening, calorie intake from
enteral nutrition (EN) >10% when enrolled, obstructive
Downloaded from pen.sagepub.com at Universidad Nacional Aut Mexic on January 18, 2015
jaundice, suspected or identified biliary tract atresia, neonatal
hepatitis, liver or renal markers present at >2 times the normal
level, congenital abnormalities associated with metabolism,
major chromosomal diseases, cytomegalovirus infection, viral
hepatitis, and congenital or acquired immune deficiency.
Finally, 103 neonates who met the criteria were randomly
assigned to either the soybean oil–based lipid emulsion (SO;
Intralipid, Fresenius Kabi, Wuxi, Jiangsu, China) group (n = 52)
or the olive oil–based lipid emulsion (OO; ClinOleic, Baxter,
Lessines, Belgium) group (n = 51). Two infants in the OO group
and 1 infant in the SO group were on full enteral feeding earlier
than 14 days; hence, 100 infants completed this study.
Ethical approval from the Research Ethical Committee, Xin
Hua Hospital, School of Medicine, Shanghai Jiao Tong
University, was obtained, and written informed consent was
obtained from the parents of all the infants before enrollment.
The trial registration number is NCT01786759.
Randomization
In this double-blind, randomized clinical trial, treatment assign-
ment cards were created with a unique randomization code and
placed in sequentially numbered opaque envelops. The cards
were pulled in sequential order, and the randomization number
was used to assign the patient to either one of the treatment
groups. Investigators, parents, and all the physicians and nurses
involved in patient care were blinded to the group assignment,
and the randomization schedule was made available only to the
pharmacist who supervised the administration of PN. The
infants were followed up until they were discharged from the
hospital; in the case of death, severe adverse effects, or parent’s
refusal to continue with PN, the treatment was discontinued.
Nutrition Support
PN support was provided according to the “Guidelines for the
Use of Nutritional Support in Critically Ill Neonates.”20 The
starting dose of the lipid emulsion was 1.0 g·kg−1·d−1 and was
increased by 0.5–1.0 g·kg−1·d−1 up to 3 g·kg−1·d−1; pediatric
amino acid solution was started at a dose of 1.5–2.0 g·kg−1·d−1
and was increased up to 3.5–4.0 g·kg−1·d−1. PN was decreased
when enteral intake increased, and PN was withheld once
patients received more than 80% of the recommended calorie
intake via the enteral route. The 2 solutions looked identical to
the clinicians. The “all-in-one” solutions contained lipids
(Intralipid or ClinOleic) (see Table 1), amino acids (18AA-11;
Treeful, Shanghai, China), glucose solution, minerals, trace ele-
ments, water-soluble vitamins (Soluvit; Fresenius Kabi, Wuxi,
Jiangsu, China), and fat-soluble vitamins (Vitalipid; Fresenius
Kabi, Wuxi, Jiangsu, China). The “all-in-one” nutrition solution
was infused continuously for 24 hours. Because no breast milk
bank is available in NICUs throughout China, all the infants
were fed with preterm formula, and the amount of EN given
was recorded by nurses every day.
Wang et al 3

Table 1.  Contents of the Fat Emulsions Used in the Study. (Agilent Technologies, Santa Clara, CA), using the method
described by Janzen et al21 with small modifications. In brief,
Ingredient OO SO
50 µL plasma samples was mixed with 150 µL methanol (with
Contents, g/100 mL 0.25 µmol/L d4-CA and 0.05 µmol/L d4-CDCA as the internal
  Soybean oil 4 20 standards) and then incubated for 10 minutes at room tempera-
  Olive oil 16 0 ture. After centrifugation for 10 minutes at 20,000 g, 10 µL of
  Egg yolk phospholipids 1.2 1.2 the supernatant was injected and separation was achieved
 Glycerol 2.25 2.25 using a Luna C18 column (50 × 2.0 mm, 3 µmol/L) (Pheno-
  Sodium oleate 0.03 - mex, Aschaffenburg, Germany), which was protected by a C18
  α-Tocopherol, mg/L 30 27 4 × 2.0–mm precolumn from Phenomex (Aschaffenburg, Ger-
Fatty acid, mol% many). Data acquisition was performed using the Masshunter
  Palmitic acid (C16:0) 13.5 11.0 software (version B.05; Agilent Technologies).
  Stearic acid (C18:0) 2.9 4.3
  Oleic acid (C18:1ω-9) 59.5 21.0
  Linoleic acid (LA-C18:2ω-6) 18.5 52.0 Statistical Analysis
  Linolenic acid (ALA-C18:3ω-3) 2.0 7.0
Randomization was carried out using computer-generated sim-
  Arachidonic acid (AA-C20:4ω-6) 0.3 0.3
  Eicosapentaenoic acid <0.02 <0.02
ple random tables in a 1:1 ratio. Sample size was determined
(EPA-C20:5ω-3) after consideration of 4 factors: (1) an expected difference in
  Docosahexaenoic acid 0.12 0.34 Dbi of about 16.5% between SO and OO groups, based on pre-
(DHA-C22:6ω-3) vious studies; (2) type I error of 0.05; (3) type II error of 20%;
PUFA, % 20.5 59.0 and (4) an allowed dropout rate of 20%. Sample size was cal-
Phytosterols, mg, per 100 g of lipid17 137 ± 1.30 220 ± 2.86 culated at 118 patients, with 59 patients in each group. The
results are expressed as mean ± SD. Data were analyzed using
Data provided by the manufacturer. AA, arachidonic acid; ALA, α-
linolenic acid; LA, linolenic acid; OO, olive oil–based emulsion; PUFA,
SPSS 21.0 for Windows (SPSS, Inc, an IBM Company,
polyunsaturated fatty acid; SO, soybean oil. Chicago, IL). Data for the 2 groups were compared by using a
t test or Mann-Whitney nonparametric test. A 2-factor repeated-
measures analysis of variance was applied to the difference
Measurement Parameters between values of the 2 groups at different times. For the liver
Liver chemistry and bile acid composition were conducted in chemistry and bile acid composition, a t test or Mann-Whitney
the plasma samples collected just before PN intervention, after nonparametric test was used to analyze the differences between
7 days and 14 days of PN, respectively. the following values: before the commencement of PN (PN-0)
and 7 days after PN (PN-7), PN-0 and 14 days after PN (PN-
Liver chemistry. The primary end point was liver chemistry 14), and PN-7 and PN-14. A linear mixed-model regression
tests assessed by the automated colorimetric method (Beck- analysis was used for repeated-measures fixed and random
man SYNCHRON LX20 system, Beckman Coulter, CA, effects within and between groups. Bonferroni correction was
USA). Liver tests included total bile acid (TBA), alanine ami- used to conserve the overall type I error at α = 0.05. A P value
notransferase (ALT), aspartate aminotransferase (AST), alka- of <.05 was considered statistically significant. Comparative
line phosphatase (AKP), γ-glutamyltransferase (GGT), total morbidity and mortality were reported as relative risk (RR) and
bilirubin (Tbi), and direct bilirubin (Dbi). The reference ranges 95% confidence intervals (CIs).
were as follows: TBA, 0–10 µmol/L; ALT, 0–75 U/L; AST,
8–38 U/L; AKP, 42–121 U/L; GGT, 16–73 U/L; Tbi, 0.2–1.2
Results
mg/dL; and Dbi, 0–0.4 mg/dL.
One hundred preterm infants were included in this study (see
Bile acids.  The secondary end point was the plasma bile acid Figure 1). There were no significant differences between the 2
composition. Stock solutions of 500 µmol/L of primary bile groups with regard to the demographic data and weight gain
acids (cholic acid [CA], glycocholic acid [GCA], taurocholic (g/d), head circumference (cm/w), days to regain birth weight,
acid [TCA], chenodeoxycholic acid [CDCA], glycochenode- the number of days on the ventilator, and the length of hospital
oxycholic acid [GCDCA], and taurochenodeoxycholic acid stay (see Table 2). There were no significant differences
[TCDCA]) were used for calibration. Standard curves (range, between the groups in nutrition characteristics either (see Table 3).
1.0–20,000 nmol/L) were obtained by diluting the mixed stock PNAC was diagnosed when the conjugated bilirubin (DBi)
solution with activated charcoal desalted plasma. The concen- was >2 mg/dL. Two infants in the OO group and 2 infants in
trations of the 6 primary bile acids were measured with an Agi- the SO group were confirmed as having PNAC, and thus, there
lent 1290 high-performance liquid chromatography coupled was no difference between the 2 groups with regard to the inci-
with an Agilent 6490 triple quadrupole mass spectrometer dence of PNAC. Three infants in the OO group and 1 infant in

Downloaded from pen.sagepub.com at Universidad Nacional Aut Mexic on January 18, 2015
4 Journal of Parenteral and Enteral Nutrition XX(X)

118 infants with gestaonal age <37wk

and maybe parenteral nutrion for 14
days or more screened

Refused to
parcipate (n=12)
Died <72 hours of age
(n=3)
103 Infants randomized

52 received OO parenteral nutrion 51 received SO parenteral nutrion

2 PN <14d 1PN <14d

50 completed study 50 completed study

Figure 1.  Flowchart depicting the process for screening and assigning the preterm infants into the OO and SO groups. The numbers of
infants who were screened and randomized are shown. A total of 103 infants were randomly assigned into the OO or SO group. Two
infants in the OO group and 1 infant in the SO group were on full enteral feeding earlier than 14 days, so they were excluded. Finally,
100 infants completed this study. OO, olive oil–based emulsion; PN, parenteral nutrition; SO, soybean oil.

Table 2.  Demographics and Outcomes.a

Variable OO (n = 50) SO (n = 50) P Value

Male/female, No. 26:24 31:19 .317
Age, median (IQR), d 2.00 (1.00–2.00) 1.00 (1.00–2.00) 1.279
Gestational age, wk 32.20 ± 1.72 30.85 ± 4.99 .074
Birth weight, g 1486.66 ± 253.89 1469.83 ± 250.52 .742
Weight gain, g/d 12.77 ± 6.93 11.78 ± 7.13 .485
Head circumference, cm/w 0.41 ± 0.23 0.49 ± 0.24 .246
Days to regain birth weight, g 11.13 ± 5.66 12.30 ± 5.23 .301
Days on ventilator, median (IQR), d 0.00 (0.00–4.00) 0.00 (0.00–6.00) .820
LOS, d 41.66 ± 22.65 40.81 ± 17.55 .837
a
Values are presented as mean ± SD unless otherwise indicated. IQR, interquartile range; LOS, length of hospitalization; OO, olive oil–based emulsion;
SO, soybean oil.

the SO group were diagnosed with necrotizing enterocolitis
(NEC), 8 infants in the OO group and 7 infants in the SO group
were confirmed to have sepsis (positive blood or cerebrospinal
fluid culture for bacteria or fungi obtained after 72 hours of age
in the presence of compatible clinical signs of septicemia), and
1 infant in the SO group was diagnosed with bronchopulmo-
nary dysplasia (BPD). There was no significant difference in
the morbidity of NEC, sepsis, or BPD.
Three deaths occurred in this study, 2 in the OO group and 1
in the SO group. Furthermore, 2 infants in the OO group and 2
infants in the SO group were withdrawn from the study. Of
these, 1 infant in the OO group was transferred to another hospi-
tal, and 3 families declined to participate in the study. There was
no significant change in mortality according to the intention to
treat. The RR was 1.532, and the 95% CI was 0.245–9.587.
The results of the liver chemistry analysis of PN-0, PN-7,
and PN-14 are shown in Table 4. No statistically significant

Downloaded from pen.sagepub.com at Universidad Nacional Aut Mexic on January 18, 2015
Wang et al 5
Table 3.  Nutrient Intake.a
Nutrient Intake OO (n = 50) SO (n = 50)
−1 −1
Lipid, g·kg ·d 1.45 ± 0.28 1.41 ± 0.19
Amino acid, g·kg−1·d−1 2.30 ± 0.40 2.17 ± 0.40
Glucose, g·kg−1·d−1 8.50 ± 1.57 7.75 ± 0.17
PN calories, 56.20 ± 9.15 54.62 ± 8.00
kcal·kg−1·d−1
EN calories, 42.28 ± 22.46 41.86 ± 28.27
kcal·kg−1·d−1
PN duration, d 27.86 ± 21.46 23.32 ± 10.48
a
Values are presented as mean ± SD. EN, enteral nutrition; OO, olive
oil–based emulsion; PN, parenteral nutrition; SO, soybean oil.
difference was observed between the groups with regard to
TBA, ALT, AST, AKP, GGT, and Tbi. In the SO group, the
serum level of Dbi was significantly higher at PN-7 (P < .05),
but this was not observed in the OO group. Moreover, the level
of Dbi was significantly different between the 2 groups, accord-
ing to linear mixed-model regression analysis (P = .039).
The bile acid levels at PN-0, PN-7, and PN-14 are shown in
Table 5. In the SO group, the TCA, GCA, C all, TCA/CA,
GCA/CA, TCA + GCA/CA, TCDCA/CDCA, GCDCA/CDCA,
and TCDCA + GCDCA/CDCA values increased significantly
with time (P < 0.05). In the OO group, the level of TCA, GCA,
C all, C all/CDC all, CA/CDCA, and GCDCA/CDCA increased
over time (P < 0.05). However, there was no significant differ-
ence between the 2 groups.
Discussion
Long-term PN in premature infants is associated with hepato-
biliary disorders such as cholestasis. Soybean-based lipid
emulsions are generally used for PN, but due to hepatobiliary
complications associated with their use, there is a need to
explore alternative lipid emulsions. This study was conducted
to compare an olive oil–based emulsion with a soybean oil–
based emulsion with regard to the prevention of PN-associated
liver disease in neonates.
Olive oil lipid emulsion is low in PUFAs but high in
MUFAs. The study by Goulet et al22 showed that despite the
low content of PUFAs in the olive oil lipid emulsion compared
with the soybean oil lipid emulsion, it did not significantly
alter the plasma triene/tetraene ratio. In another study, children
who received short-term treatment with an olive oil–based
emulsion showed either no EFA deficiency or the resolution of
an EFA deficiency.23 Administration of lipids 0.2–0.4 g·kg−1·d−1
from the SO lipid emulsion used would be required for a child
with a total energy intake of 418.4 kJ (100 kcal)·kg−1·d−1 to
meet this recommended intake. In our study, the infants
received an average intake of lipids 1.4 g·kg−1·d−1 intrave-
nously from the lipid emulsion, so their EFAs requirements
appear to have been met.
Downloaded from pen.sagepub.com at Universidad Nacional Aut Mexic on January 18, 2015
An elevated Dbi is considered the prime indicator of cholesta-
sis, which is typically characterized by Dbi as a concentration of >2
P Value
mg/dL.24 We designed a double-blind, randomized clinical study to
.082 eliminate some factors, such as sepsis. In our study, the serum level
.151 of Dbi increased significantly in the SO group after 7 days of PN,
.124 but this was not the case after 14 days. This is probably because
.073 within the 7 days after birth, the infants were mainly on PN sup-
port, which was decreased when enteral intake was increased with
.166 age. Inability to establish enteral feeding is common in infants, par-
ticularly premature infants. Furthermore, liver dysfunction appears
.233
to develop more frequently in infants who are unable to tolerate
any enteral feeding compared with those who are partially fed
enterally. Lack of EN results in reduced hepatocellular bile acid
and bile secretion and reduced gallbladder contractility. One mech-
anism may be the reduction of food-stimulated release of gastroin-
testinal hormones such as cholecystokinin (CCK).24
Soybean oil–based emulsions, rich in PUFAs, are the most
common form of lipid emulsion administered to preterm
infants in the world. In vitro and in vivo studies have shown
that a reduction in the PUFA supply from 60% to 20% of the
total fatty acid intake results in reduced lipid peroxidation.25,26
Therefore, olive oil emulsion may have the potential to reduce
oxidative stress, since it contains a much lower percentage of
PUFAs (20%). Our previous study in rabbits indicated that oxi-
dative damage may be one of the essential mechanisms of
PN-associated liver dysfunction.27 In agreement with our
results, the study by Goulet et al7 in pediatric patients found
that an olive oil lipid emulsion was more suitable for the pre-
vention of lipid peroxidation.
Earlier reports have shown that phytosterols are associated
with phytosterolemia and the severity of cholestasis.28–30 In the
study by Clayton et al,31 children receiving long-term paren-
teral lipid infusion were reported to have high levels of plasma
phytosterols, which were associated with cholestatic liver dis-
ease. Llop et al28 reported the development of liver dysfunction
in patients infused with parenteral lipid emulsion; their blood
bilirubin and AST levels were significantly correlated with
blood phytosterol concentration, which was significantly
higher in the patients receiving PN than in the healthy controls
on normal diets. However, a recent study by Vlaardingerbroek
et al32 disproved the prevailing hypothesis that phytosterolemia
contributes to hepatic cholestasis and injury. They showed that
PN-fed pigs given soybean oil developed cholestasis and ste-
atosis that were both prevented with both fish oil and SMOF
emulsions (SMOF is a blend of lipids including soybean oil,
medium-chain triglycerides, olive oil, and fish oil and thus
contains phytosterols). They found that the relative patterns in
plasma and liver phytosterol concentrations were weakly cor-
related with markers of cholestasis and liver injury among lipid
emulsion groups. Furthermore, the expression of hepatic farne-
soid X receptor (FXR) and its target genes (CYP7A1, CYP27A1,
CYP8B1, BSEP, MRP3, NTCP) among the lipid emulsion
groups was also not closely correlated with the liver phytos-
terol content, which did not support the idea that phytosterols
 6
Table 4.  Liver Chemistry.

OO SO
a
Liver Chemistry PN-0 PN-7 PN-14 P PN-0 PN-7 PN-14 Pa P
TBA, µmol/Lb 9.40 (7.60–12.55)c 14.11 ± 10.40d 20.26 ± 12.16e .001 8.40 (6.15–13.15) 13.10 (7.85–18.60) 25.84 ± 14.60e,f .009 .159
ALT, IU/Lg 7.00 (6.00–9.00) 7.81 ± 2.40 8.39 ± 4.09 .625 6.00 (5.00–8.00) 7.57 ± 2.82 7.00 (5.00–12.00) .424 .502
AST, IU/Lh 52.44 ± 33.48 24.50 (19.00–33.00)i 29.15 ± 14.27e .000 43.00 (37.75–57.50) 29.58 ± 15.05i 22.00 (17.00–28.00)e .000 .100
AKP, IU/Lj 202.36 ± 70.03 268.90 ± 87.79i 292.59 ± 114.27e .000 211.84 ± 67.30 252.48 ± 94.02i 295.48 ± 113.85e .000 .780
GGT, IU/Lk 98.00 (164.00–236.50) 100.00 (76.00–151.00)i 138.78 ± 104.80e .018 214.60 ± 105.35 111.80 ± 59.97i 89.00 (61.00–155.00)e .000 .356
Tbi, mg/dLl 2.75 (2.21–5.05) 8.35 ± 2.88i 4.13 ± 2.64f .000 38.80 (31.75–44.53) 9.00 ± 3.65i 3.83 ± 3.03f .000 .614
Dbi, mg/dLm 0.55 ± 0.23 0.75 (0.55–0.93) 0.73 ± 0.36 .076 0.59 ± 0.18 1.01 ± 1.14i 0.67 ± 0.30f .008 .039

AKP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; Dbi, direct bilirubin; GGT, γ-glutamyltransferase; OO, olive oil–based emulsion; PN, parenteral nutrition;
PN-0, before commencement of PN; PN-7, after 7 days of PN; PN-14, after 14 days of PN; SO, soybean oil; TBA, total bile acid; Tbi, total bilirubin.
a
Intragroup comparison between different time points.
b
OO group: n = 47; SO group: n = 49.
c
Median (interquartile range) (all such values).
d
Mean ± standard deviation (all such values).
e
P < .05 compared with PN-0 and PN-14 groups (analysis of variance with Bonferroni correction).
f
P < .05 compared with PN-7 and PN-14 groups (analysis of variance with Bonferroni correction).
g
OO group: n = 50; SO group: n = 50.
h
OO group: n = 50; SO group: n = 50.
i
P < .05 compared with PN-0 and PN-7 groups (analysis of variance with Bonferroni correction).
j

Downloaded from pen.sagepub.com at Universidad Nacional Aut Mexic on January 18, 2015
OO group: n = 50; SO group: n = 50.
k
OO group: n = 50; SO group: n = 50.
l
OO group: n = 50; SO group: n = 50.
m
OO group: n = 50; SO group: n = 50.
 Table 5.  Plasma Levels (µmol/L) of Conjugated Bile Acids.

OO SO
a
Bile Acids PN-0 PN-7 PN-14 P PN-0 PN-7 PN-14 Pa P
CAb 0.44 ± 0.37c 0.03 (0.02–0.04)d 0.03 (0.02–0.21) .206 0.05 (0.03–0.09) 0.04 (0.02–0.09) 0.04 (0.02–0.12) .059 .839
e
CDCA 0.33 ± 0.25 0.02 (0.01–0.03) 0.38 ± 0.34 .805 0.04 (0.02–0.06) 0.02 (0.01–0.03) 0.02 (0.01–0.06) .133 .333
TCAf 1.65 ± 0.91 2.53 ± 2.28 3.50 ± 2.02g .007 1.49 (0.61–2.12) 2.04 (0.88–3.33) 3.93 ± 2.90g .006 .624
GCAh 1.00 ± 0.61 1.70 ± 1.49 2.93 ± 2.18g,i .001 1.15 (0.58–1.72) 1.17 (0.70–3.50) 2.27 (1.41–4.22)g .017 .443
TCDCAj 7.44 ± 4.96 7.99 ± 5.91 8.33 ± 7.85 .903 4.93 (1.87–8.10) 6.16 (2.96–9.91) 9.84 ± 8.03 .319 .753
GCDCAk 1.10 ± 0.88 1.42 (0.84–2.71) 1.62 (0.82–3.39) .052 0.77 (0.54–1.25) 1.57 (0.73–3.16) 2.17 (1.16–4.17) .070 .995
C all 1.14 (0.00–2.72) 4.32 ± 3.54 6.57 ± 3.91g,i .001 2.41 (0.80–3.78) 3.12 (1.07–6.22) 7.75 ± 6.32g .004 .490
CDC all 8.58 ± 5.69 10.31 ± 7.72 3.80 (0.00–10.49) .476 5.24 (2.21–8.36) 5.99 (2.60–11.68) 13.43 ± 10.76 .177 .819
C all/CDC all 0.40 ± 0.23 0.42 ± 0.23 0.80 ± 0.56g,i .001 0.56 ± 0.38 0.60 ± 0.40 0.52 (0.36–0.84) .330 .456
CA/CDCA 1.45 ± 0.92 1.76 ± 1.10 2.85 ± 2.15g,i .008 1.79 ± 1.24 2.16 ± 1.29 2.56 ± 2.09 .107 .863
TCA/CA 63.13 ± 62.55 76.42 ± 65.63 87.78 ± 84.07 .536 35.67 ± 33.35 50.83 (19.50–86.65) 74.18 (15.70–161.68)g .049 .601
GCA/CA 35.01 ± 31.64 55.16 ± 46.84 61.63 ± 56.50 .165 16.49 (12.38–43.94) 34.38 (11.90–100.50) 54.14 (15.05–115.77)g .014 .325
TCA + GCA/CA 98.15 ± 92.78 131.57 ± 103.43 149.41 ± 137.78 .339 42.50 (30.32–94.68) 92.98 (30.04–192.80) 128.36 (29.85–299.83)g .021 .448
TCDCA/CDCA 312.99 ± 340.34 340.34 ± 230.81 371.84 ± 354.89 .804 189.86 ± 182.85 203.55 (141.37–496.62)l 273.34 (74.96–548.56)g .037 .755
GCDCA/CDCA 40.24 ± 33.53 104.13 ± 89.49l 113.62 ± 85.20g .005 36.68 ± 51.83 68.43 (29.73–168.43)l 72.56 (34.08–186.49)g .005 .716
TCDCA + 353.23 ± 300.34 444.47 ± 304.92 485.46 ± 421.85 .458 140.07 (72.59–343.94) 302.94 (176.82-702.69)l 365.67 (109.49-747.64)g .017 .732
GCDCA/CDCA

CA, cholic acid; C all, CA + GCA + TCA; CDC all, CDCA + GCDCA + TCDCA; CDCA, chenodeoxycholic acid; GCA, glycocholic acid; GCDCA, glycochenodeoxycholic acid; OO, olive oil–based
emulsion; PN, parenteral nutrition; PN-0, before commencement of PN; PN-7, after 7 days of PN; PN-14, after 14 days of PN; SO, soybean oil; TCA, taurocholic acid; TCDCA, taurochenodeoxycholic

Downloaded from pen.sagepub.com at Universidad Nacional Aut Mexic on January 18, 2015
acid.
a
Intragroup comparison between different time points.
b
OO group: n = 45; SO group: n = 46.
c
Mean ± standard deviation (all such values).
d
Median (interquartile range) (all such values).
e
OO group: n = 45; SO group: n = 46.
f
OO group: n = 45; SO group: n = 46.
g
P < .05 compared with PN-0 and PN-14 groups (analysis of variance with Bonferroni correction).
h
OO group: n = 45; SO group: n = 46.
i
P < .05 compared with PN-7 and PN-14 groups (analysis of variance with Bonferroni correction).
j
OO group: n = 45; SO group: n = 46.
k
OO group: n = 45; SO group: n = 46.
l
P < .05 compared with PN-0 and PN-7 groups (analysis of variance with Bonferroni correction).

7
8 Journal of Parenteral and Enteral Nutrition XX(X)
are strongly associated with PNAC and antagonism of hepatic
FXR function. In our study, the level of Dbi was found to sig-
nificantly differ between the 2 groups, and the higher number
of bile acids showed changes in the SO group compared with
the OO group; however, OO also contains phytosterol, which
is in agreement with the report by Vlaardingerbroek et al.
In preterm babies, canalicular cholestasis is the first histologic
sign of PN-induced hepatobiliary dysfunction and can be detected
as early as 5 days after initiation of PN. After 2 weeks of PN,
intracellular cholestasis is detected.33 There is evidence that
reduced enterohepatic circulation of bile acids and increased lev-
els of phytosterols, which are highly concentrated in many lipid
emulsions used for PN, might increase the risk of cholestasis.34
Bile acids, as other important makers of liver injury and cholesta-
sis, are widely investigated to screen for cholestatic diseases. The
study by D’Apolito et al35 showed that in newborns receiving PN,
short-term PN use is associated with an early increase in some
conjugated bile acids. This was supported by our research as the
significant increase of TCA, GCA, and GCDCA in both OO and
SO groups (Table 5). Although bile acids increased over time in
both treatment groups, some specific differences in the change in
bile acid composition over time were noted between groups. We
found that the ratios of total conjugated cholic acid/cholic acid
(GCA + TCA/CA) and total conjugated chenodeoxycholic acid/
chenodeoxycholic acid (GCDCA + TCDCA/CDCA) were sig-
nificantly elevated in the SO group over time, while no signifi-
cant increase was noted in the OO group. This may indicate that
the enterohepatic circulation in the SO group was more severely
reduced. Primary bile acids (CA and CDCA) were synthesized in
liver and conjugated with glycine and taurine. Then the conju-
gated bile acids were secreted into intestine though bile ducts.
Most of these bile acids were deconjugated and then reabsorbed
in the liver.36 Thus, the reduced enterohepatic circulation will
increase the ratio of conjugated bile acids/unconjugated bile acid.
Conclusion
Our study found that there were differences in the direct biliru-
bin and bile acid composition between the olive oil–based lipid
emulsion and soybean oil lipid emulsion in preterm infants.
Larger and long-term randomized trials should be conducted in
the future to test the effects of different lipid emulsions to deter-
mine which emulsion is safe and optimal for neonatal infants.
Acknowledgments
The authors thank the staffs of the neonatal intensive care unit in
Xin Hua Hospital and Shanghai Children’s Medical Center
(School of Medicine, Shanghai Jiao Tong University) for carrying
out the study. We also thank all the families that participated in
this study.
References
1. Kelly DA. Intestinal failure–associated liver disease: what do we know
today. Gastroenterology. 2006;130:S70-S77.
Downloaded from pen.sagepub.com at Universidad Nacional Aut Mexic on January 18, 2015
2. Nehra D, Fallon EM, Puder M. The prevention and treatment of intestinal
failure–associated liver disease in neonates and children. Surg Clin North
Am. 2011;91:543-563.
3. Christensen RD, Henry E, Wiedmeier SE, Burnett J, Lambert DK.
Identifying patients, on the first day of life, at high-risk of develop-
ing parenteral nutrition–associated liver disease. J Perinatol. 2007;27:
284-290.
4. Diamond IR, de Silva N, Pencharz PB, Kim JH, Wales PW. Neonatal
short bowel syndrome outcomes after the establishment of the first
Canadian multidisciplinary intestinal rehabilitation program: preliminary
experience. J Pediatr Surg. 2007;42:806-811.
5. Cavicchi M, Beau P, Crenn P, Degott C, Messing B. Prevalence of liver
disease and contributing factors in patients receiving home parenteral
nutrition for permanent intestinal failure. Ann Intern Med. 2000;132:
525-532.
6. Colomb V, Goulet O, De Potter S, Ricour C. Liver disease associated with
long-term parenteral nutrition in children. Transplant Proc. 1994;26:1467.
7. Goulet O, Joly F, Corriol O, Colomb-Jung V. Some new insights in
intestinal failure–associated liver disease. Curr Opin Organ Transplant.
2009;14:256-261.
8. Colomb V, Jobert-Giraud A, Lacaille F, Goulet O, Fournet JC, Ricour C.
Role of lipid emulsions in cholestasis associated with long-term parenteral
nutrition in children. JPEN J Parenter Enteral Nutr. 2000;24:345-350.
9. Marik PE. Maximizing efficacy from parenteral nutrition in critical
care: appropriate patient populations, supplemental parenteral nutrition,
glucose control, parenteral glutamine, and alternative fat sources. Curr
Gastroenterol Rep. 2007;9:345-353.
10. Diamond IR, Sterescu A, Pencharz PB, Wales PW. The rationale for the
use of parenteral omega-3 lipids in children with short bowel syndrome
and liver disease. Pediatr Surg Int. 2008;24:773-778.
11. Guglielmi FW, Regano N, Mazzuoli S, et al. Cholestasis induced by total
parenteral nutrition. Clin Liver Dis. 2008;12:97-110.
12. Cober MP, Killu G, Brattain A, Welch KB, Kunisaki SM, Teitelbaum DH.
Intravenous fat emulsions reduction for patients with parenteral nutrition–
associated liver disease. J Pediatr. 2012;160:421-427.
13. Cowles RA, Ventura KA, Martinez M, et al. Reversal of intestinal failure–
associated liver disease in infants and children on parenteral nutrition:
experience with 93 patients at a referral center for intestinal rehabilitation.
J Pediatr Surg. 2010;45:84-87.
14. Jensen AR, Goldin AB, Koopmeiners JS, Stevens J, Waldhausen JH, Kim
SS. The association of cyclic parenteral nutrition and decreased incidence
of cholestatic liver disease in patients with gastroschisis. J Pediatr Surg.
2009;44:183-189.
15. Rollins MD, Scaife ER, Jackson WD, Meyers RL, Mulroy CW, Book LS.
Elimination of soybean lipid emulsion in parenteral nutrition and supple-
mentation with enteral fish oil improve cholestasis in infants with short
bowel syndrome. Nutr Clin Pract. 2010;25:199-204.
16. Puder M, Valim C, Meisel JA, et al. Parenteral fish oil improves outcomes
in patients with parenteral nutrition–associated liver injury. Ann Surg.
2009;250:395-402.
17. Xu Z, Harvey KA, Pavlina T, et al. Steroidal compounds in commercial
parenteral lipid emulsions. Nutrients. 2012;4:904-921.
18. Deshpande GC, Simmer K, Mori T, Croft K. Parenteral lipid emulsions
based on olive oil compared with soybean oil in preterm (<28 weeks’ ges-
tation) neonates: a randomised controlled trial. J Pediatr Gastroenterol
Nutr. 2009;49:619-625.
19. Webb AN, Hardy P, Peterkin M, et al. Tolerability and safety of olive oil–
based lipid emulsion in critically ill neonates: a blinded randomized trial.
Nutrition. 2008;24:1057-1064.
20. Working Group of Pediatrics, Chinese Society of Parenteral and Enteral
Nutrition, Working Group of Neonatology, Chinese Society of Pediatrics,
Working Group of Neonatal Surgery, Chinese Society of Pediatric
Surgery. CSPEN guidelines for nutrition support in neonates. Asia Pac J
Clin Nutr. 2013;22:655-663.
Wang et al 9
21. Janzen N, Sander S, Terhardt M, et al. Rapid quantification of conjugated
and unconjugated bile acids and C27 precursors in dried blood spots and
small volumes of serum. J Lipid Res. 2010;51:1591-1598.
22. Goulet O, de Potter S, Antébi H, et al. Long-term efficacy and safety of
a new olive oil–based intravenous fat emulsion in pediatric patients: a
double-blind randomized study. Am J Clin Nutr. 1999;70:338-345.
23. Munck A. Comparison of fatty acid status in TPN-dependent children
receiving two lipid emulsions with different essential fatty acid (EFA)
content [abstract]. Clin Nutr. 1993;12(suppl 2):17.
24. Kumpf VJ. Parenteral nutrition–associated liver disease in adult and pedi-
atric patients. Nutr Clin Pract. 2006;21:279-290.
25. Wispe JR, Bell EF, Roberts RJ. Assessment of lipid peroxidation in new-
born infants and rabbits by measurements of expired ethane and pentane:
influence of parenteral lipid infusion. Pediatr Res. 1985;19:374-379.
26. Tomsits E, Rischak K, Szollar L. Long-term effects of unsaturated fatty
acid dominance on the release of free radicals in the rat. Pediatr Res.
1994;36:278-282.
27. Hong L, Wang X, Wu J, Cai W. Mitochondria-initiated apoptosis triggered
by oxidative injury play a role in total parenteral nutrition–associated liver
dysfunction in infant rabbit model. J Pediatr Surg. 2009;44:1712-1718.
28. Llop JM, Virgili N, Moreno-Villares JM, et al. Phytosterolemia in par-
enteral nutrition patients: implications for liver disease development.
Nutrition. 2008;24:1145-1152.
Downloaded from pen.sagepub.com at Universidad Nacional Aut Mexic on January 18, 2015
29. Iyer KR, Spitz L, Clayton P. BAPS prize lecture: new insight into mecha-
nisms of parenteral nutrition–associated cholestasis: role of plant sterols.
British Association of Paediatric Surgeons. J Pediatr Surg. 1998;33:1-6.
30. Bindl L, Lutjohann D, Buderus S, Lentze MJ, Bergmann KV. High plasma
levels of phytosterols in patients on parenteral nutrition: a marker of liver
dysfunction. J Pediatr Gastroenterol Nutr. 2000;31:313-316.
31. Clayton PT, Bowron A, Mills KA, Massoud A, Casteels M, Milla PJ.
Phytosterolemia in children with parenteral nutrition–associated choles-
tatic liver disease. Gastroenterology. 1993;105:1806-1813.
32. Vlaardingerbroek H, Ng K, Stoll B, et al. New generation lipid emulsions
prevent PNALD in chronic parenterally fed preterm pigs. J Lipid Res.
2014;55:466-477.
33. Moss RL, Das JB, Raffensperger JG. Total parenteral nutrition–associ-
ated cholestasis: clinical and histopathologic correlation. J Pediatr Surg.
1993;28:1270-1274.
34. Child P, Kuksis A. Investigation of the role of micellar phospholipid in the
preferential uptake of cholesterol over sitosterol by dispersed rat jejunal
villus cells. Biochem Cell Biol. 1986;64:847-853.
35. D’Apolito O, Pianese P, Salvia G, et al. Plasma levels of conjugated bile
acids in newborns after a short period of parenteral nutrition. JPEN J
Parenter Enteral Nutr. 2010;34:538-541.
36. Chiang JY. Bile acids: regulation of synthesis. J Lipid Res. 2009;50;
1955-1966.