ORIGINAL ARTICLE: HEPATOLOGY

Effectiveness of Enteral Versus Oral Nutrition With a

Medium-Chain Triglyceride Formula to Prevent
Malnutrition and Growth Impairment in Infants
With Biliary Atresia

Rocı́o Macı́as-Rosales, yAlfredo Larrosa-Haro, zGenaro Ortı́z-Gabriel,
and §Benjamı́n Trujillo-Hernández

ABSTRACT

Objectives: The aim of this study was to compare the effectiveness of oral
What Is Known
(PO) versus enteral nutrition (EN) medium-chain triglyceride (MCT) con-
taining–formula to prevent malnutrition and growth impairment in infants  Most infants with biliary atresia develop early growth
with biliary atresia (BA) waiting for a liver transplant.
retardation and energy-protein malnutrition.
Methods: A total of 15 infants, 3 to 9 months old with BA were included.  A successful Kasai operation or liver transplant may
They were randomly assigned to either PO or EN. For 12 weeks, both groups
lead to partial catch-up growth.
received an MCT formula fortified with glucose polymers and corn oil to
reach a caloric density between 0.8 and 1 kcal/mL. The formula given to the
What Is New
PO group was administered ad libitum and that given via EN was infused
through a nasogastric tube to reach 140% of the energy intake recommended  The ad libitum ingestion of a calorically fortified
by the Dietary Recommended Intake guidelines. Protein intake was adjusted
medium-chain triglyceride formula was not enough
to 4 to 5 g/kg present weight. Outcome variables were growth and nutritional
to prevent linear growth, and head circumference
status evaluated periodically by anthropometric indicators. Biochemical and
and fat reserves impairment.
hematological variables were evaluated through the study.  Enteral nutrition at home with a medium-chain tri-
Results: Baseline clinical, nutritional, biochemical, and hematological
glyceride formula supplemented with lipid and glu-
variables showed no differences between the study groups. Baseline
cose polymers maintained linear growth, growth in
length/age was <2 SD in 10 of the 15 patients; in the PO group, it fell
head circumference, and improved fat reserves.
<3 SD, whereas in the EN group, it remained stable. Head circumference z
score dropped 0.6 SD in the PO group, whereas in the EN group it remained
stable. Triceps skinfold values improved in the infants taking EN, P < 0.001.
The frequency of adverse effects—respiratory infection and diarrhea—was
Conclusions: A 12-week EN trial with an MCT-fortified formula prevented
higher in the EN group. No biochemical or hematological differences were
malnutrition and growth impairment in infants with BA waiting for a liver
observed between the study groups throughout the study.
transplant.
Key Words: biliary atresia, enteral nutrition, growth impairment,
malnutrition, medium-chain triglyceride
Received July 23, 2014; accepted July 7, 2015.
From the Servicio Gastroenterologı́a y Nutrición, Unidad Médica de Alta
Especialidad, Hospital de Pediatrı́a, Centro Médico Nacional de (JPGN 2016;62: 101–109)
Occidente, Instituto Mexicano del Seguro Social, the yInstituto de

Nutrición Humana, Centro Universitario de Ciencias de la Salud,
Departamento de Clı́nicas de la Reproducción Humana Crecimiento y
Desarrollo Infantil, Universidad de Guadalajara, the zCentro de Inves-
tigación Biomédica de Occidente, Centro Médico Nacional de Occi-
dente, Instituto Mexicano del Seguro Social, Guadalajara, Jalisco, and
the §Unidad de Investigación y Epidemiologı́a Clı́nica, Hospital General
de Zona No. 1, Colima, México.
Address correspondence and reprint requests to Alfredo Larrosa-Haro, MD,
PhD, Instituto de Nutrición Humana, Departamento de Clı́nicas de la
Reproducción Humana, Crecimiento y Desarrollo Infantil, Centro
Universitario en Ciencias de la Salud, Universidad de Guadalajara,
Salvador Quevedo y Zubieta 750, Sector Libertad, Guadalajara, Jalisco,
México, CP 43240 (e-mail: alfredolarrosaharo@hotmail.com).
The authors report no conflicts of interest.
Copyright # 2015 by European Society for Pediatric Gastroenterology,
Hepatology, and Nutrition and North American Society for Pediatric
Gastroenterology, Hepatology, and Nutrition
DOI: 10.1097/MPG.0000000000000909
B
iliary atresia (BA) is currently the most common indication
for liver transplantation in children. It is characterized by the
progressive fibrotic obliteration of all or part of the extrahepatic
biliary tree within the first 3 months of life (1–3). The outcome of
untreated BA is progression to cirrhosis and death within 18 to
24 months (4–9). Moderate or severe acute and/or chronic malnu-
trition occur in most infants with BA during the first year of life,
even in those with a successful portoenterostomy (4,10–17). Loss
of muscle and other structural proteins, impaired bone mineraliz-
ation, and depletion of fat reserves can be devastating for growth
and development. In infants with BA, malnutrition and stunting are
likely to occur despite receiving special formulas administered
orally with medium-chain triglycerides, branched chain amino
acids, and extra glucose polymers (11–15). Ad libitum PO feeding
may not be enough to sustain growth, especially if there is delay in
liver transplantation (11). Nutritional support seems crucial for

JPGN  Volume 62, Number 1, January 2016 101

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
 Larrosa-Haro et al
most infants with BA from the time of diagnosis and through the
first 24 months of life (15–19).
The aim of this study was to compare the effectiveness of
enteral versus oral nutrition (PO) with a medium-chain triglyceride
(MCT) formula to prevent malnutrition and growth impairment in
infants with BA.
METHODS
Patients
This study was carried out at the Gastroenterology and
Nutrition Department of the Unidad Medica de Alta Especialidad
Hospital de Pediatria (Guadalajara, Mexico) from March 2009 to
March 2011. Fifteen consecutive Hispanic infants with the diag-
nosis of BA were included. Portoenterostomy was not performed on
6 of the 15 patients, 3 of each of the study groups, because they were
referred after 3 months of age. The 6 patients in whom Kasai
operation was not performed were evaluated by the liver transplant
program but none received a liver transplant during the study period.
Protocol
Design
The study design was open longitudinal intervention study.
Selection Criteria
BA was diagnosed at laparotomy by way of an operative
cholangiography (20). Infants with a birth weight <2500 g and/or
with significant associated diseases which could have a negative
impact on growth and nutrition (lung, cardiac or renal disease,
congenital malformations, etc) were not included.
Sample Size
The sample size was calculated with a formula for clinical
trials (21). Data for this purpose (d ¼ 1.19 and s ¼ 0.77) were
obtained from medium arm circumference evaluated in 2 previous
studies (22,23); the confidence level (a) was 95%, and the power
was (1b) 80%. The number of patients estimated to show differ-
ences was 8 per group.
Randomization
Consecutive patients who met the selection criteria were
randomly assigned by means of a random number table (simple
randomization) to either the oral (control) group or the enteral
(intervention) group.
Variables
The intervention variables were PO or enteral nutrition (EN).
Both groups received a medium-chain triglyceride formula for a
period of 12 weeks. The outcome variable was nutritional status; it
was evaluated every 2 weeks by anthropometry in both control and
intervention groups. Clinical adverse effects were recorded daily
while the patient was hospitalized and every week after discharge;
the caregivers had the option of 24-hour telephone contact with the
main researchers (R.M-R. and A.L-H.). For purposes of this trial,
biochemical and hematological variables were obtained at time 0
and at 12 weeks, the endpoint of the trial; however, additional
laboratory studies were performed according to the Servicio de
Gastroenterologı́a y Nutrición de la Unidad Medica de Alta Espe-
cialidad Hospital de Pediatria BA protocol.
102
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
JPGN  Volume 62, Number 1, January 2016
Nutritional Intervention
Enteral Nutrition
A hydrolyzed protein, glucose polymer, and medium-chain
triglyceride formula (Alfare, Nestlé, Vevey, Switzerland) providing
720 kcal and 24.7 g of protein per 1000 mL was administered as a
continuous infusion for a period of 18 hours. The initial caloric
input was 100% of the dietary reference intake (DRI) of energy for
age and sex according to the Food and Agriculture Organization/
World Health Organization/United Nations University reference
(24); this input was gradually increased in the following 5 to 7 days
in an attempt to reach 140% of the DRI (11–15). Protein intake was
adjusted to a maximum of 4 to 5 g/kg for present weight (16,17,25).
To reach the referred caloric input without exceeding the maximum
protein requirements, glucose polymers and corn oil were added to
the formula to reach a maximum caloric density of 0.8 to 1 kcal/mL.
PO complementary feedings were initiated at 6 months of age with
the following schedule: 6 to 7 months, fruits and vegetables; 7 to
8 months, fruits, vegetables, cereals, and chicken; 8 to 9 months, all
of the above plus red meat and legumes.
After discharge, EN was carried out at home by the parents or
guardians. Before introduction of EN, they had been given 3 to
5 days of practical training conducted by a pediatric nutritionist.
The patients were discharged from the supervision of the nutri-
tionist when their caregivers were able to adequately perform the
EN procedure. The formula was administered by means of an
enteral pump (Sentinel; OST Medical Inc, Warwick, RI) through
an 8-Fr 60-cm–long nasogastric tube (Freka Tube; Fresenius Kabi
AG, Bad Homburg, Germany); the tube was changed and placed in
the opposite nostril every other week. The infusion was continuous
for 18 hours, starting by 6 AM and ending at midnight. EN tolerance
was evaluated through clinical data (diarrhea, bloating, and vomit-
ing) and the presence of reducing substances in stools; when these
conditions occurred, the infusion volume was reduced 2.5% by the
nutritionist responsible for monitoring the procedure.
Oral Nutrition
Oral nutrition was managed ad libitum. The suggested
amount of formula was 5 to 7 ounces every 4 hours; the final
amount and schedule were determined by the patient and her or his
parents or caregivers. The type of formula and its fortification with
glucose polymers and corn oil were similar to that of the EN group.
Before discharge, the parents were also trained in the formula
preparation and feeding procedure. Complementary feedings were
managed with the same schedule as the intervention group.
Anthropometry
Standardization
Before the data were collected, the main authors performed
an anthropometric standardization trial with 15 infants with BA
(26). Consistency (intragroup individual measurements) and
validity (comparison with a criterion standard) were evaluated with
Pearson bivariate correlations. When the correlation coefficient was
<0.8, the anthropometric technique was reviewed and corrected
until intragroup and intergroup correlations >0.8 were achieved
(27,28). Criterion standard for body composition was double-x-ray
absorptiometry evaluated in a parallel study in 15 infants with
chronic liver disease (29).
Weight
Babies were weighed without a diaper on a leveled pan scale
with a beam and movable weight. The patients were placed on the
www.jpgn.org
 JPGN  Volume 62, Number 1, January 2016
scale making sure the weight was evenly distributed. Weight was
recorded to the nearest 10 g (28,30).
Recumbent Length
Length was obtained in the recumbent position using a SECA
416 Infantometer (Medical Measuring Systems and Scales, Chino,
CA). An assistant held the infant’s head, and the examiner straigh-
tened the infant’s legs, holding the feet with toes pointed upward,
and positioned the footboard against the feet. The length was
recorded to the nearest 0.1 cm (28,30).
Mid-Upper Arm Circumference
To obtain the mid upper arm circumference (MUAC),
the baby’s left arm was bent at a 908 angle at the elbow with
the upper arm held parallel to the side of the body. The distance
between the acromion and the olecranon was measured with a
fiberglass metric tape, and the midpoint between these 2 points
was marked. The baby’s left arm was then relaxed, hanging loosely
by its side. The fiberglass tape was positioned at the marked
midpoint, and the circumference was recorded to the nearest
0.1 cm (31–33).
Triceps Skinfold
Triceps skinfold (TSF) was measured with a Lange skinfold
caliper (Healthcheck Systems, Brooklyn, NY) at a previously
marked posterior left upper arm midpoint. The arm was extended
in the same relaxed position used for the MUAC. The examiner
grasped a vertical pinch of skin and subcutaneous fat between the
thumb and forefinger 1 cm above the previously marked mid-
point, gently pulling away from the underlying muscle. The skin-
fold caliper was placed at the midpoint mark while maintaining the
skinfold grasp. Readings were measured in millimeters when the
caliper came in contact with the skin and the dial reading stabilized
(31–33).
Total Upper Arm Area, Upper Arm Muscle
and Fat Areas, and Arm Fat Index
Arm areas and arm fat index were calculated according to the
formulas described by Frisancho (28); results were expressed in
square millimeters.
Reference Patterns and Indicators
of the Nutritional Status
The z scores of length-for-age and weight-for-length were
calculated with the WHO 2006 reference pattern (28). The z scores
of medium upper arm circumference, TSF, and arm areas-for-age
were calculated with the Sann reference pattern (32). The patients
were regarded as normal nutritional status when z scores were
between 2 and þ2 SD; patients with a z score 2 SD were
considered undernourished (33).
Laboratory Tests
Conjugated and nonconjugated bilirubin, aspartate amino-
transferase, alanine aminotransferase, g-glutamyltransferase,
alkaline phosphatase, serum proteins, albumin, globulin, ammonia,
total cholesterol, triglyceride, prothrombin time, partial thrombo-
plastin time, red and white cell count, platelet count, glucose, urea,
and creatinine were performed at time 0 and at 12 weeks (34).
www.jpgn.org
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
Enteral Nutrition in Infants With Biliary Atresia
Adverse Effects
Parents or guardians were instructed to communicate by
telephone with the principal investigators or attend the hospital’s
emergency department if the infant developed clinical manifes-
tations of any kind especially with regard to fever, abdominal
distension, vomiting, diarrhea, rhinorrhea, coughing, and/or diffi-
culty in breathing. In the biweekly visits, the parents were asked
about the presence of these symptoms or signs during the 2 weeks
previous to the consultation.
Statistical Analysis
Medians and interquartile range (IQR) were used as central
and dispersion trend indicators. The x2, Fisher, and Mann-Whitney
U tests were used to compare demographic, clinical, and laboratory
variables. Nutritional and anthropometric variables between the
study groups were compared with the Kruskal-Wallis and Mann-
Whitney U tests; comparison within groups was performed with the
Friedman test.
Ethical Aspects
The study was approved by the research and ethics commit-
tee of the Unidad Medica de Alta Especialidad Hospital de Pediatrı́a
and registered at the Instituto Mexicano del Seguro Social National
Registry of Intervention Studies (no. 1302-17). Parents or guardians
gave their written informed consent.
RESULTS
Patients
Seven patients were assigned to the control group and 8 to the
intervention group. The median age was 5 months (IQR 4–7) in the
latter group and 5.5 (IQR 4–6) in the former, and there was no
statistical difference between them (P ¼ 0.517). Five patients of the
7 in the control group and 5 of the 8 in the intervention group were
women, also with no statistical difference (P ¼ 0.714). Two patients
with birth weight <2500 g were excluded from the study; other
serious health problems besides BA were not identified.
The diagnosis of BA was made according to the Unidad
Medica de Alta Especialidad Hospital de Pediatria protocol
(20–22). Laparotomy was indicated when the duodenal tube test
resulted in the absence of macroscopic bile after 24 hours of
duodenal fluid collection. BA was confirmed by surgical explora-
tion of the porta hepatis, operative cholangiography, and liver
biopsy.
Baseline Clinical and Laboratory Variables
Comparison of the clinical, biochemical, and hematological
variables upon hospital admission between the study groups is
presented in Table 1; no statistical differences were found between
them with the exception of total cholesterol, which was higher in the
EN group.
Severity of Liver Damage
At admission, the severity of liver damage was assessed by
the pediatric end-stage liver disease score (35), which showed a
median score of 16 and 18 in the control and intervention groups,
respectively (P ¼ 0.485). When the patients were evaluated with the
Malatack score (36), values were 45 in the oral group and 40 in the
enteral group (P ¼ 0.399).
103
 Larrosa-Haro et al JPGN  Volume 62, Number 1, January 2016

TABLE 1. Clinical, endoscopic, and histological data in 15 infants with BA at time 0

Oral (n ¼ 7) Enteral (n ¼ 8)

Qualitative variables n % n % P

Clinical signs
Ascites 4 57.1 4 50.0 0.595
Abdominal venous markings 4 57.1 5 62.5 0.622
Hand blistering 3 42.7 3 36.5 0.622
Spiders 1 14.3 3 37.5 0.338
Endoscopy
Esophageal varices 4 57.1 6 75.0 0.426
Histology
Cirrhosis 5 71.4 5 62.5 0.573

Quantitative variables Median IQR Median IQR P

Blood
Hemoglobin, g/dL 10.7 10.3–11.6 11.1 9.3–11.7 0.536
Leucocytes, WBC/mL 16,000 8754–27,300 12,650 13,400–24,400 0.955
Platelets, platelets/mL 349,000 245,000–356,000 274,000 119,000–559,000 0.397
Prothrombin time, s 15.7 14.8–19.1 11.4 11–24 0.336
Partial thromboplastin time, sec 39.7 33.4–55 40.7 34.9–43 0.779
Biochemistry
Conjugated bilirubin, mg/dL 8.1 6.2–10 6.9 6.7–10.6 0.397
Unconjugated bilirubin, mg/dL 1.9 0.9–1.8 1.2 1–3.2 0.072
AST, IU/L 379 203–458 255 191–419 0.694
ALT, IU/L 215 138–342 166 127–285 0.694
GGT, IU/L 645 112–897 295 261–996 0.189
Alkaline phosphatase, IU/L 591 322–800 483 454–886 0.189
Albumin, g/dL 3.3 3.1–3.4 3.2 3–3.9 0.613
Ammonia, mg/dL 54 61–155 59 42–98 0.940
Total cholesterol, mg/dL 248 234–312 179 110 0.021

Data are presented as frequencies and percentages; percentages correspond to lines. Statistics, x2 or Fisher test. ALT ¼ alanine aminotransferase;
AST ¼ aspartate aminotransferase; BA ¼ biliary atresia; GGT ¼ g-glutamyltransferase; IQR ¼ interquartile range.

Nutritional Variables University recommendations (24). The energy intake in the

Intake of energy and protein were reported as the percen-
tage of the kilocalories in the DRI and the grams of protein
per kilogram of actual weight, respectively (Table 2). The
ingestion of energy and protein in the ad libitum PO-fed group
was >100% of the DRI according to the Food and Agricul-
ture Organization/World Health Organization/United Nations
EN group was significantly higher at all times of the trial, with
the exception of week 4. The protein intake was slightly higher in
the intervention group and was statistically significant at all
times.
Vitamin and inorganic nutrient intake from the MCT
formula was >100% of the DRI in both groups throughout the
trial. Besides the formula micronutrient content, all patients

TABLE 2. Energy intake expressed as percentage or the dietary recommended intake and protein intake (grams per kilograms per day) according
to the FAO/UNU/WHO recommendations in 15 infants with BA

Energy (DRI %) Protein, g/day

Oral Enteral Oral Enteral

Weeks Median IQR Median IQR P Median IQR Median IQR P

0 100 86–107 116 99–127 0.094 21.9 19.4–24 25.8 24.3–30.7 0.008
2 109 105–111 121 118–131 0.021 23.1 22.6–25 26.9 25.4–31.6 0.011
4 108 79–115 116 100–131 0.131 24.6 20.8–25.2 27.7 25.7–32.1 0.004
6 98 89–106 121 106–132 0.004 22.9 21.1–24 28.5 27.9–32.5 0.001
8 105 88–106 119 109–128 0.008 23.8 22.4–24 29.6 28.7–32.7 0.001
10 107 92–111 123 110–131 0.008 24.4 22.7–25.1 30.8 29.7–33.6 0.001
12 104 81–105 127 111–134 0.032 24.6 20.8–29.1 32.6 30.6–35.5 0.001

Energy and protein intake is reported every 2 weeks until the endpoint of the study. Statistics, Mann-Whitney U test. BA ¼ biliary atresia; DRI ¼ dietary
reference intake; FAO ¼ Food and Agriculture Organization; IQR ¼ interquartile range; UNU ¼ United Nations University; WHO ¼ World Health
Organization.

104 www.jpgn.org

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
 JPGN  Volume 62, Number 1, January 2016 Enteral Nutrition in Infants With Biliary Atresia

0 0
Enteral Length for age Oral
P = 0.697 P < 0.001
–1 –1

Z score
–2 –2

–3 –3

0 2 4 6 8 10 12 0 2 4 6 8 10 12
–4 –4
Weeks Weeks

0 0
Enteral Oral
P = 0.50 Head circumference for age P < 0.001

–1 –1
Z score

–2 –2

–3 –3

0 2 4 6 8 10 12 0 2 4 6 8 10 12
–4 –4
Weeks Weeks

FIGURE 1. Length for age and head circumference for age z scores in 15 infants with BA managed with a fortified MCT formula (ad libitum PO,
n ¼ 0.7; EN, n ¼ 8). Results are expressed as medians and IQR. P values correspond to intragroup comparisons with Friedman test. BA ¼ biliary
atresia; EN ¼ enteral nutrition; IQR ¼ interquartile range; MCT ¼ medium-chain triglyceride; PO ¼ oral nutrition.

received daily PO vitamin A palmitate (5000 IU), ergocalciferol (0.3, IQR 0.09–0.4, P ¼ 0.014), week 6 (0.4, IQR 0.01–0.5,
(400 IU), menadione (5 mg), and vitamin E (100 IU). P ¼ 0.029), week 8 (0.5, IQR 0.01–0.6, P ¼ 0.014), week 10
(0.3, IQR 0.01–0.6, P ¼ 0.004), and week 12 (0.2, IQR 0.1 to
Growth and Nutritional Status Outcome 0.8, P ¼ 0.021).

Length for Age Weight for Length

The median length for age z score at time 0 was <2 SD in
10 of the 15 patients, 5 from each study group. The outcome of this
indicator is presented in Figure 1. In the PO group, the median
length-for-age z score fell from 2.4 SD (IQR 1.1 to 2.7) to
3.2 SD (IQR 3 to 3.5), P < 0.001. In the EN group, the length-
for-age z score remained stable <2 SD (time zero 2.4, IQR 1.1
to 2.9; 12 weeks, 2.2 m IQR 1.1 to 3.7) without intragroup
difference (P ¼ 0.697). The comparison between groups at each
time of the study showed no differences.
To assess changes in the height for age z scores in the
different study times, a subtraction of the baseline from each
biweekly value was conducted and an intergroup comparison
was performed. A significant z score difference was found at weeks
8 (0.6, IQR 0.1–0.8, P ¼ 0.014), 10 (0.6, IQR 0.1–1, P ¼ 0.009),
and 12 (0.8, IQR 0.4–1.1 P ¼ 0.011).
Head Circumference-for-Age
The weight for length z scores remained stable and within
normal limits (2 to 2 SD) in both groups at all times during the
trial. In the PO group, time 0 z score was 0.7 (IQR 0.3 to 1.0)
and at 12 weeks it was 0.6 (IQR 0.1 to 1.5). In the EN group,
these values were 1.1 (IQR 0.02 to 1.9) and 0.7 (IQR 0.5
to 1). Intergroup and intragroup comparison of z score values did
not show statistical differences.
Medium Upper Arm Circumference
The median MUAC for age z score at time 0 was <3 SD in
both groups. In the PO group, MUAC fell from 3.0 (IQR 1.9 to
4) to 3.7 (IQR 3.5 to 4.8), whereas in the EN group this
indicator relocated from 3.2 (IQR 1.7 to 4) to 2.7 (IQR 1.1
to 4.4). Intergroup and intragroup comparisons, however, of
z score values were not statistically significant.

The head circumference z score at time 0 was located TSF-for-Age

> 2 SD in both groups. The outcome of this indicator is presented
in Figure 1. In the PO group, the head circumference median z
score gradually dropped from 1.5 (IQR 0.7 to 3.1) to 2.1
(IQR 1.7 to 3.8) with a loss of 0.6 SD (P < 0.001), whereas
in the EN group this indicator remained stable throughout the
intervention (1.9, IQR 1 to 3.1; 2.1, IQR 0.9 to 3.2,
P ¼ 0.650).
The significant results of the baseline head circumference z
scores subtraction from each biweekly measurement were: week 4
The TSF for age z score was <2 SD at time 0 in all of the
patients of both groups. In the PO group, this indicator remained
stable <3 SD throughout the study (time zero 3.7, IQR 2.6 to
4.2; 12 weeks 3.8, IQR 3.1 to 4.5), P ¼ 0.972. In the EN
group, the TSF moved from 4.0 (IQR 3.4 to 4.6) to 2.6 SD
(IQR 1.6 to 3.3), (P < 0.001). Intergroup comparison at 10
weeks (PO 3.9, IQR 3.2 to 4.5; EN 3, IQR 2.2
to 3.3) and 12 weeks (PO 3.8, IQR 3.1 to 4.5; EN 2.6,

www.jpgn.org 105

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
 Larrosa-Haro et al JPGN  Volume 62, Number 1, January 2016

IQR 1.6 to 3.3) showed statistical difference between the study
groups (P ¼ 0.043 and P ¼ 0.011, respectively).
The significant results of the baseline TSFs z scores sub-
traction from each biweekly measurement were: week 6 (0.6, IQR
1.1 to 0.3), week 8 (0.8, IQR 1.7 to 0.7, P ¼ 0.014), week 10
(0.9, IQR 1.4 to 0.8, P ¼ 0.040), and week 12 (0.7, IQR 1.5
to 0.5, P ¼ 0.006).
Arm Muscle Area for Age
The arm muscle area-for-age z scores were within normal
limits (2 to þ2 SD) at time 0 in both groups (Fig. 2). In the PO
group, this indicator dropped from 1.5 (IQR 1.3 to 3) to 2.6
(IQR 2.2 to 3.6), P ¼ 0.874; in the EN group, z scores remained
stable throughout the trial (time zero 1.6, IQR 1.3 to 3;
12 weeks 1.9, IQR 0.9 to 2.1), P ¼ 0.71).
Arm Fat Area for Age
The arm fat area for age z score was <2 SD at time 0 in both
groups (Fig. 2). In the PO group, the arm fat area z score remained
stable <2 SD throughout the trial (time zero 2.8, IQR 2.2 to
3.4; 12 weeks 2.2, IQR 1.9 to 2.7) (0.679). In the EN group,
this indicator moved from 3 (IQR 2.1 to 3.4) to 2.2 (IQR
1.1 to 2.9), P ¼ 0.001.
Adverse Effects
Respiratory Infection
In the PO group, 2 patients had an upper respiratory infec-
tion. In the EN group, 5 of the 8 patients had rhinitis that required
oral antibiotics and nasal infusion of 0.9% saline solution; one
of them presented bronchopneumonia requiring intravenous
antibiotics and inhalation therapy. In all of the patients, the respir-
atory infections resolved within a week. The frequency of respir-
atory infections between the study groups was not different.
Diarrhea
Three of the 8 patients having PO had loose stools and an
increase in stool output frequency. In the EN group, all of the
patients had loose stools and an increase in stool output frequency.
None of them presented with dehydration. Stool smears, stool
cultures, and reducing substances were negative in all of the
patients. The frequency of diarrhea was significantly higher in
the EN group (P ¼ 0.025).
Abdominal Distension and Vomiting
In the PO group, 3 infants presented with abdominal disten-
sion that was considered secondary to ascites. In the intervention
group, 4 patients had abdominal distension and vomiting during the
enteral infusion, and both were corrected with a lower enteral
infusion volume.
Outcome of Hematological and Biochemical
Variables
Hematologic and biochemical studies performed at time 0
and 12 weeks are presented in Table 3.
Blood Cell Count
Hemoglobin was >10 g/dL at time 0 and 12 weeks in both
study groups; at the endpoint, hemoglobin was higher in the EN

0 0

Enteral Oral
Arm fat area
P = 0.001 P = 0.915
–1 –1
Z score

–2 –2

–3 –3

0 2 4 6 8 10 12 0 2 4 6 8 10 12
–4 –4
Weeks Weeks

0 0
Enteral Oral
Arm muscle area
P = 0.117 P = 0.988
–1 –1
Z score

–2 –2

–3 –3

0 2 4 6 8 10 12 0 2 4 6 8 10 12
–4 –4
Weeks Weeks

FIGURE 2. AFA for age and arm muscle area for age z scores in 15 infants with BA managed with a fortified MCT formula (ad libitum PO, n ¼ . 7;
EN, n ¼ 8). Results are expressed as medians and IQR. P values correspond to intragroup comparisons with Friedman test. AFA ¼ arm fat area;
BA ¼ biliary atresia; EN ¼ enteral nutrition; IQR ¼ interquartile range; MCT ¼ medium-chain triglyceride; PO ¼ oral nutrition.

106 www.jpgn.org

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
 JPGN  Volume 62, Number 1, January 2016 Enteral Nutrition in Infants With Biliary Atresia

TABLE 3. Hematological and biochemical variables in 15 infants with BA at time 0 and after 12 weeks of nutritional intervention

PO (n ¼ 7) Enteral (n ¼ 8)

Variables Study times Median IQR Median IQR P

Hemoglobin, g/dL Basal 10.7 9.3–11.7 11.1 10.3–11.6 0.536

Endpoint 10.1 9.2–10.9 10.8 10.4–11.8 0.040
White blood cells, WBC/mL Basal 16,600 13,400–24,400 12,650 8754–27,300 0.955
Endpoint 12,600 10,400–16,580 9729 7800–11,083 0.021
Prothrombin time, s Basal 11.4 11–24 15.5 14.8–19.1 0.336
Endpoint 16.3 11.7–19 14.1 13.7–17.5 0.613
Partial thrombin time, s Basal 39.7 34.9–43 40.7 33.4–55 0.779
Endpoint 42.4 34.5–47 42.0 37.4–47 0.694
Platelets, platelets/mL Basal 349,000 1,19,000–5,59,000 2,74,000 2,45,000–3,56,000 0.397
Endpoint 198,000 1,10,000–4,33,000 1,80,500 1,43,000–2,79,000 0.779
Erythrocyte sedimentation velocity, mm Basal 14.2 10.5–25 10.4 10–15.1 0.189
Endpoint 14.6 10.4–25 11.3 10.1–13.7 0.121
C-reactive protein, mg/L Basal 5.6 3.3–10 5.4 3.8–11 0.955
Endpoint 4.8 3.3–16.9 4.3 3.3–10 0.779
Glucose, mg/dL Basal 80 77–83 80.1 78.5–84 0.478
Endpoint 81 80–84 81.9 78.3–84.3 0.673
Urea, mg/dL Basal 10 8–11 11 8.2–13.2 0.521
Endpoint 11 9–16 10.5 7–13 0.769
Creatinine, mg/dL Basal 0.2 0.1–0.6 0.3 0.2–0.6 0.689
Endpoint 0.2 0.1–0.8 0.2 0.1–0.7 0.496

Statistics, Mann-Whitney U test. BA ¼ biliary atresia; IQR ¼ interquartile range; PO ¼ oral nutrition.

group, P ¼ 0.040. At baseline, white blood cell count was >10,000/
mL in both groups. At 12 weeks, the median remained >10,000/mL
in the PO group, whereas it was below this cut point in the
intervention group (P ¼ 0.021).
Clotting Times and Erythrocyte
Sedimentation Rate
PT, PTT, and ESR were within normal limits at the onset of
the study, and no significant differences were observed at the
endpoint.
C-Reactive Protein
C-reactive protein values were within normal limits at the
onset of the trial in both study groups. No differences were observed
during the trial and at the end of the study.
Biochemical Variables
Glucose, urea, and creatinine serum concentration were
within normal values at time 0 and at the end of the trial; no
intragroup or intergroup differences were observed.
Liver Function Tests
Medians and IQR of liver function tests performed at the
onset and at the end of the trial are presented in Table 4. At time 0,
both groups had conjugated hyperbilirubinemia. Conjugated bilir-
ubin decreased in infants who underwent the Kasai procedure in
both groups 2 weeks after surgery (2.6, IQR 1.8–2.9; 2.1, IQR
0.4–0.9 mg/dL, EN and PO, respectively); values for groups did
not differ (P ¼ 0.307); after this transient decrease, bilirubin
increased in the EN and PO groups throughout the trial. Serum
aminotransferases were elevated in both groups at the onset and
at the endpoint of the trial with no intragroup or intergroup
differences. g-Glutamyltransferase and alkaline phosphatase were
more than the reference values in both groups at time 0; the serum
concentration of these enzymes was lower in the EN group at the
endpoint, P ¼ 0.027 and P ¼ .040, respectively. Albumin was
<3.5 g/dL at time 0, and its values decreased after 12 weeks in
both groups, with no differences between them. Cholesterol and
triglyceride concentrations were more than the reference in the PO
group upon admission, but no difference between the study groups
was observed at the end of the trial. Serum ammonia concentration
was above the reference values in both study groups at the onset and
at the end of the trial; its concentration was higher in the EN group
at the end of the study (P ¼ 0.009).
DISCUSION
Infants with BA are prone to most known mechanisms of
secondary malnutrition: dietary ingestion is not sufficient to main-
tain normal growth, intestinal absorption is impaired, the metabolic
rate is increased, and there is also impairment of some liver
macronutrient metabolic pathways (2,11–16,37). These nutritional
risk conditions lead most infants with BA to rapid growth impair-
ment and to depletion of muscle mass and adipose tissue, in other
words, to severe malnutrition. Along the first year of life, length and
head circumference of most infants with BA fall progressively
resulting in an impairment of linear and central nervous system
growth that is most likely permanent (11,17,18). Loss of muscle
mass may be an important added developmental factor to these
severe handicaps.
Although it is reasonable to assume that the probability of
malnutrition and growth impairment will be lower the earlier
portoenterostomy and/or liver transplantation is carried out, there
are a number of variables that may become involved in the nutri-
tional outcome of infants with BA. The advancement of highly
specialized liver units and nationwide liver transplantation pro-
grams in developed countries currently offer infants with BA a high
short-term probability of survival with an acceptable quality of life.
The scenario is the opposite in the developing countries, where the

www.jpgn.org 107

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
 Larrosa-Haro et al JPGN  Volume 62, Number 1, January 2016

TABLE 4. Liver function tests in 15 infants with BA at time 0 and after 12 weeks

PO (n ¼ 7) EN (n ¼ 8)

Variable Study time Median IQR Median IQR P

Conjugated bilirubin, mg/dL Basal 8.1 6.1–9.9 6.9 5–9.6 0.397

Endpoint 10.2 7.8–12.9 11.2 9.4–14.6 0.613
Unconjugated bilirubin, mg/dL Basal 1.9 1.3–3.3 1.2 0.9–1.8 0.072
Endpoint 2.3 1.3–4.2 1.4 1.5–1–9 0.462
AST, IU/L Basal 215 155–313 166 78–265 0.694
Endpoint 236 166–318 201 102–359 0.955
Alanino aminotransferase, IU/L Basal 379 281–517 255 105–404 0.694
Endpoint 268 162–452 240 134 (182–230) 0.955
GGT, IU/L Basal 645 365–1059 295 125–502 0.189
Endpoint 358 107–799 160 74–277 0.029
Alkaline phosphatase, IU/L Basal 591 413–626 483 284–808 0.189
Endpoint 622 434–1077 437 308–666 0.040
Albumin, g/dL Basal 3.3 3.1–3.7 3.2 3.1–3.3 0.613
Endpoint 3.0 2.6–3.8 2.8 2.7–3 0.779
Cholesterol, mg/dL Basal 248 218–296 179 141–251 0.021
Endpoint 202 193–216 122 54–205 0.281
Triglyceride, mg/dL Basal 190 167–229 124 105–152 0.001
Endpoint 164 134–212 104 67–158 0.397
Ammonia, mmol/L Basal 54 44–72 59 23–109 0.094
Endpoint 55 48–68 59.5 50.5–71.5 0.009

Results are expressed as medians and IQR. Statistics: Mann-Whitney U and Wilcoxon tests. Control group, intragroup (basal per 12 weeks) significant
differences: bilirubin P ¼ 0.012, GGT P ¼ 0.036, and albumin P ¼ 0.017. Intervention group, intragroup (basal per 12 weeks) significant differences:
cholesterol P ¼ 0.018 and triglyceride P ¼ 0.028. AST ¼ aspartate aminotransferase; BA ¼ biliary atresia; EN ¼ enteral nutrition; GGT ¼ g-glutamyltran-
speptidase; IQR ¼ interquartile range; PO ¼ oral nutrition.

referral of infants with cholestasis is frequently delayed, and a liver
transplant may be postponed for several months, increasing the risk
of severe malnutrition. The proportion of patients with BA in such
conditions is probably much higher worldwide.
The published experience of liver centers regarding the
mechanisms of secondary malnutrition and the nutritional inter-
vention protocols in infants and children with BA have set the
theoretical and practical basis for the nutritional approach (12–18).
The present trial, however, differs from, or adds information to,
those studies in at least 4 aspects: in our patients, the weight-for-
length indicator was not sensitive for identifying acute malnutrition
or evaluating the outcome of the 2 types of nutritional intervention
(z scores remained within normal limits [2 to þ2 SD] in both
groups during the entire trial), and this condition has already been
theoretically explained by a bias related to liver and spleen enlarge-
ment plus water retention (38); the effect of EN on linear growth
and particularly on central nervous system growth has seldom been
reported on and should be considered a key implication when
deciding on a medium- or long-term enteral nutritional intervention;
fat stores improved significantly with the nutritional intervention,
and this condition may play an important role in dealing with acute
episodes such as cholangitis or gastrointestinal bleeding, and
particularly in liver transplantation patients; ad libitum ingestion
of an MCT formula by the infants of the PO group was >100% of
the DRIs for age and sex, but it was clearly not enough to achieve
catch-up growth. These results may challenge the widely accepted
indication of an MCT PO formula as the basis of nutritional
treatment.
Our Gastroenterology and Nutrition Department has dealt
with the devastating outcome of BA over the last 2 decades. Our
efforts have always been directed toward establishing early diag-
nosis and portoenterostomy, and on a growing transplant program
(11,20,22,23). The belief that an infant with cholestatic BA will
thrive when placed on an ad libitum MCT formula with added
glucose polymers and fat-soluble vitamins because a water-soluble
compound has not been ratified in previous attempts by our group or
in the present trial: the median of energy intake in the control group
was >100% of the DRI for their age and sex, and, in spite of this, the
linear growth and nutritional status trend was one of progressive
deterioration. In contrast, enteral nutritional intervention with a
nasogastric tube by means of a continuous 18-hour-a-day infusion
achieved significant results: length and head circumference
remained at a stable z score growth speed during the 3 months
of the trial, and fat storage increased. Arm muscle area also
deteriorated in the PO group and remained stable in the EN group,
although without significance. Catch-up linear or central nervous
growth was not achieved with the EN protocol of this trial; however,
the EN group maintained growth velocity of length and head
circumference, whereas infants fed PO had impairment of growth
velocity in both variables.
The frequency of adverse effects—respiratory infection and
diarrhea—was higher in the EN group; however, from the clinical
point of view, they were not significant compared with the outcome
in growth and body composition. Biochemical and hematological
variables did not show alterations that could be associated with the
nutritional intervention.
Catch-up growth and nutritional status improvement may be
achieved with a successful liver transplant (38); however, the
amount of time necessary to reach this goal may be influenced
by a number of variables, and a long delay can result in the
deterioration of growth and nutritional status (39). EN with an
MCT formula may be an alternative that mitigates growth retar-
dation and improves the nutritional status of a patient during the
length of time before a possible liver transplantation.

108 www.jpgn.org

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
 JPGN  Volume 62, Number 1, January 2016
Acknowledgment: The authors thank Kathleen B. Schwarz,
MD, for the review, editing, and comments on this article.
REFERENCES
1. Balistreri WE, Bezerra JA, Ryckman FC. Biliary atresia and other
disorders of the extrahepatic bile ducts. In: Suchy FJ, Sokol RJ,
Balistreri WF, eds. Liver Disease in Children. New York: Cambridge
University Press; 2007:248–61.
2. Moreira RK, Cabral R, Cowles RA, et al. Biliary atresia: a multi-
disciplinary approach to diagnosis and management. Arch Pathol Lab
Med 2012;136:746–60.
3. Hartley JL, Davenport M, Kelly DA. Biliary atresia. Lancet 2009;
374:1704–13.
4. Sokol RJ, Shepherd RW, Superina R, et al. Screening and outcomes in
biliary atresia: summary of a National Institutes of Health workshop.
Hepatology 2007;46:566–81.
5. The NS, Honein MA, Caton AR, et al. National Birth Defects Prevention
Study. Risk factors for isolated biliary atresia: 1997–2002. Am J Med
Genet A 2007;143:2274–84.
6. Carvalho E, Pontes-Ivantes CA, Bezerra JA. Extrahepatic biliary
atresia: current concepts and future directions. J Pediatr (Rio J)
2007;83:105–20.
7. Sokol RJ, Mack C, Narkewicz MR, et al. Pathogenesis and outcome
of biliary atresia: current concepts. J Pediatr Gastroenterol Nutr
2003;37:4–21.
8. Davenport M. Biliary atresia: outcome and management. Indian J
Pediatr 2006;73:825–8.
9. Utterson EC, Shepherd RW, Sokol RJ, et al. The SPLIT Research
GroupBiliary atresia: clinical profiles risk factors, and outcome of 755
patients listed for a liver transplantation. J Pediatr 2005;147:180–5.
10. Schreiber RA, Kleinman RE. Biliary atresia. J Pediatr Gastroenterol
Nutr 2002;35 (suppl I):S11–6.
11. Larrosa-Haro A. Atresia de vı́as biliares. In: Larrosa-Haro A, Vásquez-
Garibay E, eds. Nestlé Nutrition Institute Workshop LATAM 2, Nutrición
y Enfermedades del Aparato Digestivo en Niños. México DF: Inter-
sistemas Editores SA de CV; 2011:247–70.
12. Chin SE, Sheperd RW, Cleghorn GJ, et al. Preoperative nutritional
support in children with end-stage liver disease accepted for liver
transplantation: an approach to management. J Gastroenterol Hepatol
1990;5:566–72.
13. Moreno LA, Gottrand F, Honden S, et al. Improvement of nutritional
status in cholestatic children with supplemental nocturnal enteral
nutrition. J Pediatr Gastroenterol Nutr 1991;12:213–6.
14. Chin SE, Shepherd RW, Cleghorn GJ, et al. Survival, growth and quality
of life in children after orthotopic liver transplantation: a 5 year
experience. J Paediatr Child Health 1991;27:380–5.
15. Charlton CP, Buchanan E, Holden CE, et al. Intensive enteral feeding in
advanced cirrhosis: reversal of malnutrition without precipitation of
hepatic encephalopathy. Arch Dis Child 1992;67:603–7.
16. Chin SE, Shepherd RW, Thomas BJ, et al. Nutritional support in
children with end-stage liver disease: a randomized crossover trial of
branched-chain amino acid supplement. Am J Clin Nutr 1992;56:158–
63.
17. Ramaccioni V, Soriano HE, Arumugam R, et al. Nutritional aspects of
chronic liver disease and liver transplantation in children. J Pediatr
Gastroenterol Nutr 2000;30:361–7.
18. Schneider BL, Brown MB, Haber B, et al., Biliary Atresia Research
Consortium. A multicenter study of the outcome of biliary atresia in the
United States, 1997 to 2000. J Pediatr 2006;148:467–74.
www.jpgn.org
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
Enteral Nutrition in Infants With Biliary Atresia
19. De Russo PA, Ye W, Shepherd R, et al. Biliary atresia research
consortium: growth failure and outcomes in infants with biliary atresia:
a report from the biliary atresia research consortium. Hepatology
2007;46:1632–8.
20. Larrosa-Haro A, Caro-López AM, Coello-Ramı́rez P, et al. Duodenal-
tube test in the diagnosis of biliary atresia. J Pediatr Gastroenterol Nutr
2001;32:311–5.
21. Jeyaseelan L, Rao PSS. Methods of determining sample sizes in clinical
trials. Indian Ped 1989;26:115–21.
22. Larrosa-Haro A, Bonal-Pérez MA, Rodrı́guez-Álvarez TH, et al. Atresia
de vı́as biliares: estudio prospectivo de manejo con nutrición enteral con
una fórmula semielemental con triglicéridos de cadena media. Rev
Gastroenterol Mex 1996;61:39.
23. Macı́as-Rosales R, Larrosa-Haro A, Bojórquez-Ramos MC. Tratamien-
to nutricio en lactantes con atresia de vı́as biliares y cirrosis avanzada en
espera de trasplante hepático: Estudio piloto. Rev Gastroenterol Mex
2003;68(suppl):S90.
24. Human Energy Requirements. FAO Food and Nutrition Technical Rep
Ser. Rome: United Nations University/World Health Organization/Food
and Agriculture Organization; 2004:11–19.
25. Kelly DA, Davenport M. Current management of biliary atresia. Arch
Dis Child 2007;92:1132–5.
26. Hurtado-López EF, Larrosa-Haro A, Vásquez-Garibay E, et al. Liver
function tests predict the nutritional status evaluated by arm anthro-
pometrical indicators. J Pediatr Gastroenterol Nutr 2007;45:451–7.
27. De Onis M, Habicht JP. Anthropometric reference data for international
use: recommendations from a World Health Organization Expert Com-
mittee. Am J Clin Nutr 1996;64:650–8.
28. Frisancho AR. Anthropometric Standards for the Assessment of Growth
and Nutritional Status. Ann Arbor: University of Michigan Press; 1993.
29. Hurtado-López EF, Vásquez-Garibay EM, Larrosa-Haro A, et al. Correla-
tion of body composition indicators evaluated by dual-energy X-ray
absorptiometry with anthropometrical indicators in infants and toddlers
with chronic liver disease. J Pediatr Gastroenterol Nutr 2009;49:E78–9.
30. Fomon SJ, Nelson SE. Size and growth. In: Fomon SJ, ed. Nutrition of
Normal Infants. St Louis, MO: Mosby-Year Book; 1993. 36–83.
31. Frisancho AR. New norms of upper limb fat and muscle areas for
assessment of nutritional status. Am J Clin Nutr 1981;34:2540–5.
32. Sann L, Durand M, Picard J, et al. Arm and muscle areas in infancy. Arch
Dis Child 1988;63:256–60.
33. WHO Expert Committee. Physical Status: The Use and Interpretation
of Anthropometry. WHO Technical Report Series 854. Geneva, Switzer-
land: World Health Organization; 1995:6–9.
34. Tietz NW. Fundamentals of Clinical Chemistry. 3rd ed. Philadelphia,
PA: WB Saunders; 1987:452–453.
35. McDiarmid SV, Anand R, Linbland AS. Principal Investigators and
Institutions of the Studies of Pediatric Liver Transplantation (SPLIT)
Research Group. Development of a pediatric end-stage liver disease
score to predict poor outcome in children awaiting liver transplantation.
Transplantation 2002;74:173–81.
36. Malatack JJ, Schaid DJ, Urbach AH, et al. Choosing a pediatric recipient
for orthotopic liver transplantation. J Pediatr 1987;111:479–89.
37. Sokol RJ, Stall C. Anthropometric evaluation of children with chronic
liver disease. Am J Clin Nutr 1990;52:203–8.
38. Pawlowska J, Socha P, Jankowska I. Factors affecting catch-up growth
after liver transplantation in children with cholestatic liver diseases. Ann
Transplant 2010;15:72–6.
39. Parker P, Stroop B, Greene H. A controlled comparison of continuous
versus intermittent feeding in the treatment of infants with intestinal
disease. J Pediatr 1981;99:360–4.
109