Manja Fleddermann a, Hans Demmelmaira, Veit Grote a, Tatjana Nikolic b, Branka Trisic c, Berthold

Koletzko a,*
a
Dr. von Hauner Children’s Hospital, University of Munich Medical Centre, Munich, Germany b
Institute for Gynecology and Obstetrics, Clinical Centre of Serbia, Belgrade, Serbia c HiPP Study
Center, Belgrade, Serbia

articleinfo may influence infant growth. We aimed to assess the effect of a modified infant formula on growth.
Methods: In a randomized, double-blind trial, 213 healthy term infants consumed isoenergetic study formulae (intervention
formula e IF, control formula e CF) from the first month of life until the age of 120 days. IF (1.89 g protein/100 kcal) contained
Article history: a-lactalbumin (ALAB) and LC-PUFA, while CF (2.30 g protein/100 kcal) provided standard whey and no LC-PUFA.
Received 12 August 2013 Accepted Anthropometry and dietary intake were regularly assessed. A venous blood sample was obtained on day 120.
20 December 2013 Results: Both formulae were well-accepted without significant differences in health related observations. Weight gain was not
statistically different between formula groups (IF: 30.2 6.3 vs. CF: 28.3 6.5 g/day, mean SD, P ¼ 0.06). Length gain was higher
Keywords:
in IF (0.11 0.02 vs. 0.10 0.02 cm/day, P ¼ 0.02). Energy intake from formula was higher in CF at 90 and 120 days (IF: 509
Fully formula fed
Infant formula 117 and 528 123 vs. CF: 569 152 and 617 169 kcal/day, P < 0.01). Protein intake in CF was significantly higher at each
Growth
Alpha-lactalbumin
assessment. Growth per energy intake was higher in IF compared to CF for weight (6.45 2.01 vs. 5.67 2.21 g/ 100 kcal, P ¼
Low-protein 0.02) and length (0.23 0.08 vs. 0.20 0.08 mm/100 kcal, P ¼ 0.04).
summary
Conclusions: The modified infant formula with reduced protein content with added ALAB and LC-PUFA, meets infant
requirements of protein for adequate growth. The increased energetic efficiency of the new infant formula might result from
improved protein composition by added ALAB. Apparently minor differences in composition can markedly affect energetic
Background & aims: Protein
efficiency for growth.
source, macronutrient
composition and content of long The study was registered at ClinicalTrials.gov (NCT01094080).
chainpolyunsaturated fatty acids 2014 The Authors. Published by Elsevier Ltd and European Society for Clinical Nutrition and
(LC-PUFA) of infant formulae Metabolism. Open access under CC BY-NC-SA license.

higher formula protein content considering the lower nutritive value of cow’s
milk casein and whey compared to human milk proteins.10 Differences in the
Abbreviations: AA, amino acids; BeMIM, BelgradeeMunich Infant Milk Trial; BF, breastfed;
CF, control formula; FF, formula fed; IF, intervention formula; ITT, intention to treat; LC-PUFA,
amino acid (AA) composition may require higher amounts of cow’s milk
long chain-polyunsaturated fatty acids; PP, per-protocol; Trp, tryptophan. protein to ensure adequate supply of essential AA, 10 which puts additional
q burden on the renal system because of the higher nitrogen intake.11 According
Conference presentation: 46th Annual Meeting of ESPGHAN, London, May 2013. *
to the “early protein hypothesis”3 high protein intake causes rapid early
Corresponding author. Dr. von Hauner Children’s Hospital, University of Munich Medical Center, growth, which has been associated with an
Lindwurmstraße 4, D-80337 München, Germany. Tel.: þ49 (0) 89 5160 2826; fax: þ49 (0)8 9 5160 increasedriskofobesityandassociateddisordersinlaterlife.12,13 Thus, lower
7742. protein intake in infancy seems beneficial for long term health.
E-mail addresses: office.koletzko@med.uni-muenchen.de, Berthold.Koletzko@ med.uni- Davis et al. 2008 investigated the use of an infant formula with a-
muenchen.de (B. Koletzko).
lactalbumin (ALAB).14 ALAB, the predominant whey fraction in human
1. Introduction
milk, is the main source of tryptophan (Trp) of breastfed infants and enables
the reduction of protein content in infant formulae, while ensuring sufficient
Infants should be fully breastfedfor6 months whenever feasible. 1 Infants
supply of Trp and all other essential AA.15,16 The composition of milk
benefit from breastfeeding not only via immediate protection against
consumed by infants affects the energy efficiency for growth: Butte et al.
gastrointestinal and respiratory infections,2 but also via a lower risk of obesity
1990 compared the energy efficiency of infants at the ages of 1 and 4 months
and diabetes in adult life.3,4 If full breastfeeding is not possible, safe and
and found that breastfed infants had an 11% higher weight gain per 100 kcal
suitable infant formulae should be fed.5
(energetic efficiency) than formula fed infants.17
During the last decades, infant formulae have been improved for example
The inclusion of LC-PUFA into formula aimed to optimize the neuronal
by the inclusion of long chain-polyunsaturated fatty acids (LC-PUFA)6 and
development as LC-PUFA are quantitatively and qualitatively important
adaptation of content and source of protein.7,8 Protein intake of infants is lower
components of nervous tissue and provided by human milk. 18 The LC-
with breast milk than with standard infant formulae9 because of a generally
PUFAs, arachidonic acid and docosahexaenoic acid, may enhance visual

0261-5614 2014 The Authors. Published by Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. Open access under CC BY-NC-SA license.
http://dx.doi.org/10.1016/j.clnu.2013.12.007
 M. Fleddermann et al. / Clinical Nutrition 33 (2014) 588e595
acuity and mental as well as psychomotor development.19 Moreover, there
might be interactions with growth as arachidonic acid has been related to
growth and LC-PUFA derived prostaglandin may influence adipocyte
differentiation.20,21
Since the modification of protein source, macronutrient
compositionandcontentofLC-PUFAofinfantformulaemayinfluenceinfant
growthanddevelopment,modifiedformulamustbetestedtoensure they support
adequate growth.22 In present study we compared growth and blood
biochemistry of infants fed a modified infant formula with reduced protein
content and rich in ALAB as well as added LC-PUFA to a formula with
standard protein content and without LCPUFA, and a reference group of
breastfed infants. The randomized controlled trial aimed to assess the
suitability of a reduced-protein, ALAB and LC-PUFA containing formula
focusing on growth velocities, adverse events, markers of fatty acids and
protein status and energetic efficiency in infants until the age of 120 days.
2. Subjects and methods
2.1. Power calculation and randomization
Based on previous findings a weight gain between birth and age 4 months
of life of 30 g per day with a standard deviation of 6 g was assumed for sample
size estimation.23e26 The non-inferiority study assumed a power of 85% and
2.5% risk of a-error to detect a difference of 0.5 standard deviations (one-
sided test) as statistically significant required studying 70 infants per formula
group. Assuming a total loss to follow-up rate of up to 30% 100 infants were
enrolled into each formula arm.
Double blinded randomized allocation of infants to the study formulae
was stratified for gender and a block size of four was applied. A random
allocation sequence was generated by the study sponsor. The blinded
allocation was concealed for participants, support staff and investigators until
all laboratory and data analyses had been performed.
2.2. Ethical considerations
The study was approved by the Clinical Center Serbia Ethical Committee.
Written informed consent was obtained from all parents prior to study start
after the experimental protocol had been explained to them in detail.
The study was registered at ClinicalTrials.gov (NCT01094080).
2.3. Study design
The BeMIM study (BelgradeeMunich Infant Milk Trial), a randomized,
double-blind, controlled study with parallel design, was performed with two
formula groups (intervention formula e IF and control formula e CF) and a
reference group of breastfed infants (BF). Infants were recruited until the age of
28 days. Children of mothers, who could not breastfeed their healthy newborn
babies for reasons not related to this study, or who decided e in spite of all benefits
of breast milk e to start full formula-feeding within the first 28 days of life, were
randomized double blinded into one of the two formula groups. Formula infants
(FF) were fully formula fed until the age of 120 days. A reference group of
breastfed infants was recruited with intended duration of breastfeeding for at least
4 months. During the first 28 days (Baseline) and at 30, 60, 90 and 120 days of
life infants were examined and anthropometric measures were taken. During
three days before each study examination parents recorded the volume of
consumed formula and completed questionnaires on formula acceptance,
consistency and color of stool, occurrence of colic, flatulence, regurgitation, and
vomiting.
Weight was determined with a Seca 336 scale (Seca, Hamburg, Germany)
equipped with a measuring rod (Seca 232) for measuring recumbent length, and
head circumference was measured with a tape (Seca 212). All measurements were
performed in duplicate and documented with an accuracy of 10 g for weight and
589
0.1 cm for length and head circumference, respectively. The equipment was
regularly checked and calibrated to ensure accuracy of measurements. These
checks were done with a calibrated weight of 5 kg every 4 weeks. There was
never an aberration of more than 10 g. Therefore, the scales were never reset.
2.4. Study population
From Jan 2010 to May 2011, 505 infants were enrolled verbally at the
maternity ward and with flyers providing study information and contact data of
the principal investigator at the Institute for Gynecology and Obstetrics of the
Clinical Center of Serbia in Belgrade, Serbia. Eligible infants had to be born
apparently healthy from singleton pregnancies after 37e41 weeks of gestation,
with a birth weight between the 3rd and 97th weight-for-age percentile according
to the EURO-Growth charts.27 Infants with malformations, congenital heart
defects, congenital vascular diseases, severe diseases of gastrointestinal tract,
kidney, liver, central nervous system, or metabolic disease were excluded from
study participation.
2.5. Study diets
Study formulae were provided free of charge to families by HiPP GmbH &
Co. Vertrieb KG (Pfaffenhofen, Germany) in 600 g cartoons and labeled by
random numbers. The products were packed in identical white boxes and labeled
with the same product name. IF complied with the EU-directive of 200628 and CF
with the corresponding 1999 EU-directive.29 The formulas had identical whey/
Table 1
Composition of study formulae and human milk.
a
Whey:casein ratio 60:40 60:40 60:40
Energy (kcal/100 mL) 67 1.89 67 67 7.6 e
Protein (g/100 kcal) 2.2
Protein (g/100 mL) 1.3 1.5 1.2 0.2
Lipids (g/100 mL) 3.6 3.3
3.6 0.7
Carbohydrates (g/100 mL) 7.5 7.8
7.4 0.2
Nommsen et al. 1991, Food and Nutrition Board 2001.52,53
casein ratios of 60/40 and energy contents of 67 kcal/100 mL, but differed in
protein, fat and carbohydrate content (Table 1). IF had a protein content of 1.89
g/100 kcal and a fat content of 5.3 g/100 kcal, while CF contained 2.20 g/100 kcal
and 4.9 g/100 kcal, respectively. IF was rich in ALAB and supplemented with
free L-phenylalanine and L-tryptophan to meet required contents (Table 1,
supplemental data).28 Egg and fish oils providing arachidonic acid and
docosahexaenoic acid (each 10.7 mg/100 kcal) were added to IF (Table 2,
supplemental data).
2.6. Laboratory procedures
At 120 days of age venous blood samples was drawn >2 h after the last
feeding. Plasma urea, albumin, glucose and creatinine were analyzed at the
Clinical Center of Serbia. Plasma aliquots were stored at 80 C and transported on
dry ice to the Dr. von Hauner Children’s Hospital (Munich, Germany) for the
analyses of plasma AA and glycerophospholipid fatty acids.
Quantitative AA analysis was performed by LC-MS/MS.30 The samples were
deproteinized by methanolic HCl and chemically derivatized with butanolic HCl.
Amino acid butyl esters were separated on a XBridge C18 column (Waters,
Eschborn, Germany) by ionpair reversed phase chromatography (HPLC 1100 SL,
Agilent, Waldbronn, Germany) and quantified by triple quadrupole mass
spectrometry(SciexAPI2000,AppliedBiosystems,Darmstadt,Germany).
Glycerophospholipid fatty acids were determined according to Glaser et al.
(2010a) with the use of gas chromatography with flame ionization detection
(Agilent 7890A, Agilent Technologies, Mülheim an der Ruhr, Germany). 31
Derivatization of fatty acids with sodium methoxide was followed by separation
590 M. Fleddermann et al. / Clinical Nutrition 33 (2014) 588e595

of fatty acid methyl esters by gas chromatography using a BPX70 (25 m*0.22
mm, Phenomenex, Aschaffenburg, Germany) column.
2.7. Calculations
exception of mothers educational level and rate of spontaneous deliveries,
which were significantly higher for BF infants than for FF infants (P ¼ 0.002
and P ¼ 0.001, respectively). 3.1. Adverse events

The anthropometric gain (weight, length and head circumference) between 30
and 120 days of age was calculated. For calculation of the anthropometric gain
per day, the results were divided by the exact number of days between both visits
of each subject. The results were presented as weight: g/d, length: cm/d and head:
cm/d and were available for all subjects.
The daily energy and protein intake was calculated as the average intake at
30, 60, 90 and 120 days of age and was available of all four time points for 79 IF
and 80 CF infants.
The energetic efficiency was calculated as ratio of the anthropometric gain
(weight and length) from 30 to 120 days of age to the
averageenergyorproteinintakeperdayduringthestudy.Themonthly energetic
efficiency was calculated from the anthropometric gain
duringonemonthoflifeandtheenergyandproteinintakeassedatthe endof
thismonth.Theenergeticefficiencyisexpressedasweightgain/ 100 kcal or /g protein
or as length gain/100 kcal or /g protein.
The SD scores (z scores) were calculated by using the anthropometric results
relative to the growth standards of the WHO for breastfed children. 26
Both formulae were well-accepted and no differences were reported for
acceptance as well as consistency and color of stool, colic, flatulence,
regurgitation and vomiting. The total number of adverse events (adverse event
plus serious adverse event) was 21 in 88 IF infants, 41 in 92 CF infants and 45
in 185 BF infants (Table 4, supplemental data). Thus, a significantly higher
rate in adverse events was observed in CF infants compared to IF (P ¼ 0.003)
and BF infants (P ¼ 0.001), while IF and BF infants did not differ. The types
of adverse events were similarly distributed between formula groups (49%
respiratory tract infections, 24% skin infection/eczema, 10% gastrointestinal
problems, 4% urinary tract infections, and 13% others). The number of serious
adverse events was 12 in the formula groups (IF ¼ 9, CF ¼ 3) and 4 in the
reference group, with one serious adverse event in each formula group
considered a potentially association to the study formula (IF: Vomiting, blood
in stool, reflux and CF: Vomiting, blood in stool).

2.8. Data management and statistical analysis

Data were collected in a MS Access database and statistical analyses were

performed with Stata/MP 11.0 (StataCorp LP, College Station, TX, USA).
Results are presented as mean SD or medians and interquartile ranges (IQR, 25th
and 75th percentile). Pearson chi-square test was used for statistical comparison
of categorical data, t-test for normally distributed continuous variables and
KruskaleWallis test for non-normally distributed continuous variables. P-values
below 0.05 were considered as statistically significant.
The weight gain (g/d) between 30 and 120 days of life was used as primary
outcome and was calculated by t-test. For comparison of primary outcome
between both formula groups and BF group potential confounder (sex,
maternal educational level, smoking during pregnancy) were considered.
The primary (weight gain) and secondary (length and head circumference
gain) outcomes were analyzed for intention-to-treat (ITT) and per-protocol
(PP) populations. In the ITT analysis, all subjects were considered that
received study formula. In the PP analysis, only data from subjects complying
with the predefined conditions, no intake of other formula than study formula,
a maximum of 50 mL tea intake per day (mean over each 3-dayprotocol) and
less than 3 spoons complementary food per week.

3. Results

A total of 505 infants were recruited (Fig. 1) including 213 FF infants who
were randomly allocated to one of the study formulae. Of this group 207
infants finally received study formula from 19 8 days of life onward (no
significant difference between groups). During the intervention period forty
infants dropped out of the study(19.3%)becauseofparental refusal(n
¼34),medication/illness (n ¼ 4) or loss of contact (n ¼ 2). Three of the 167
infants completing the study were excluded fromthe analyses because not-
fulfilling the inclusion criteria (n ¼ 2) or because of implausible values (n ¼
1). Gender ratio, birth order, maternal age and education, smoking, gestational
age, mode of delivery, and anthropometry at birth and at study entry were not
significantly different between the randomized formula groups (Table 3,
supplemental data).
From the 185 BF infants 93 infants were lost to follow-up (50%) because
of parental refusal (n ¼ 76), medication/illness (n ¼ 3) and loss of contact (n ¼
14). Baseline characteristics were comparable to the FF infants, with the
 M. Fleddermann et al. / Clinical Nutrition 33 (2014) 588e595 591

3.2. Growth No. of infants 82 82 92

Weight gain (g/d) 30.2 6.3a a,b 28.3 6.5 a 26.7 6.4a b
There were no differences between both formula groups for weight gain Length gain (cm/d) 0.11 0.02 0.10 0.02 0.10 0.02
from 30 to 120 days of age (Table 2, ITT population). Length gain expressed Head circ. gain (cm/d) 0.05 0.01 0.05 0.01 0.05 0.02
as cm per day was significantly higher in IF (0.11 0.02 cm/day) compared to
Data are presented as mean SD. Identical letters indicating significant differences between
CF infants (0.10 0.02 cm/day, P ¼ 0.02). In BF infants weight and length gains
groups (P < 0.05, Student’s t-test).
were significantly lower than in IF infants (Table 2). No significant differences

Fig. 1. Numbers of participants for randomization, allocation, follow-up and analysis.

were observedforheadgrowthbetweenall groups(0.050.01cm/dayfor IF and
CF; 0.05 0.02 cm/day for BF infants). Determined possible confounders sex,
smoking during pregnancy, mother’s age, first pregnancy and gestational age
were not different between groups.
Additionally, we adjusted for weight or length at 30 days of age to account
for the fact that more extreme values tend to have lower or higher gains,
respectively. The adjusted estimated daily weight gain for weight from 30
days to 120 days of age was found to be 1.94 g/d (95% CI: 0.04, 39.2; P ¼
0.055) greater in IF group than in CF group. The adjusted estimated daily
length gain for length from 30 days to 120 days of age was found to be 0.05
mm/d (95% CI: 0.0006, 0.10; P ¼ 0.047) greater in IF group than in CF group.
Analyzing PP subjects, weight tended to be higher in IF compared to CF
(not significant, Table 5, supplemental data). Mean weight and length gains
of IF infants were greater than of BF infants (both P ¼ 0.01).
and protein intakes determined from the 3 day protocols are summarized in Fig.
2. Energy intake in IF and CF infants was identical at 30 and 60 days of age, but
at ages 90 and 120 days higher energy intakes were observed for CF infants (569
152 vs. 509 117 kcal/d, P ¼ 0.005 and 617 169 vs. 528 123 kcal/d,
P < 0.001, respectively). The protein intake was significantly higher in CF infants
than in IF infants (Fig. 2).
The intake of other foods than study formula was generally low. There were
no significant differences between the intake of water, tea, other formula than
study formula, breastfeeding and complementary feeding between the formula
groups during the whole study period.
In the PP population, the same differences in volume, energy and protein
intake of IF and CF infants were found as in the ITT population. Neither intake
of liquid nor breastfeeding was different between IF and CF infants (data not
shown).

Results of z-scores are depicted in Fig. 1, supplemental data.

3.4. Growth in relation to intake

3.3. Nutrient intake
The intake of study formula was similar between the groups at 30 and 60
days of age, but significantly higher in CF infants at 90 and 120 days of age
by average 90 and 140 mL, respectively. Energy
Table 2
Weight, length and head circumference gain in infants of the intention-to-treat population.
Weight and length gain from 30 to 120 days of age were related to the average
energy intake during the study as a marker for energetic efficiency of the formulae
(Fig. 3). Weight gain per 100 kcal was significantly higher in IF infants (6.45
2.01 g/100 kcal) than CF infants (5.67 2.21 g/100 kcal, P ¼ 0.02). A
corresponding difference was observed for length gain per 100 kcal (IF 0.23 0.08
mm/100 kcal and CF 0.20 0.08 mm/100 kcal,
P ¼ 0.04). The energetic efficiency during the first (birth to 30 days of age) and
second (30e60 days of age) month of life in IF and CF infants was identical, but
tended to be higher in IF during the third month (60e90 days of age). During the
fourth month (90e120 days of age), the energetic efficiency of IF was
592 M. Fleddermann et al. / Clinical Nutrition 33 (2014) 588e595
significantly higher than in CF for weight 5.16 2.96 vs. 4.12 1.64 g/100 kcal and
length 0.24 0.23 vs. 0.18 0.11 mm/100 kcal.
Weight gain by protein intake was significantly higher in IF
(3.33 1.04 g/g protein) than in CF infants (2.53 0.99 g/g protein, P < 0.001).
Fig. 2. Mean (SD) daily energy and protein intake at 30, 60, 90 and 120 day of age in intervention and control group of the intention-to-treat population. **P < 0.01, ***P < 0.001.
Significantly different between groups (Student’s t-test).
3.5. Biochemical markers
There were no significant differences between the formula groups for
albumin, urea and creatinine. Compared with the BF group, urea and creatinine
were higher and glucose was lower in both formula groups (Table 6, supplemental
data). All analyzed AA and fatty acid concentrations were within the normal
ranges for healthy infants or small children.14,32
Infants receiving IF formula showed a significantly higher plasma total fatty
acid concentration of 1225 mg/L (interquartile range: 1108, 1318) than CF infants
(1071 mg/L: 988, 1202; P < 0.001). Arachidonic acid (8.24 vs. 7.02%, P < 0.001)
and docosahexaenoic acid (4.51 vs. 1.48%, P < 0.001) percentages were higher
in IF infants (Table 7, supplemental data). The arachidonic acid level was
significantly higher in BF infants compared to both formula groups, while the
docosahexaenoic acid level was highest in IF infants, followed by breastfed
infants and CF (all group differences P < 0.01).
The concentrations of 14 AA were significantly different between groups
(Table 8, supplemental data). The total AA concentration were 2907 (2538, 3124)
mmol/L in IF and 3053 (2737, 3292) in CF group (P ¼ 0.01). The concentrations
of Ala, Asn, Asp, His, Ile, Leu, Met, Orn, Pro, Thr and Val were significantly
higher in CF infants, while Phe, Tyr and Trp were significantly higher in IF
infants. Branched chain AA concentrations of BF infants were more similar to
concentrations in the IF group, while some other AA, including Phe and Trp
tended to be more similar between BF and CF infants.
Fig. 3. Anthropometric gain per 100 kcal in intervention and control group. *P < 0.05.
Significantly different between groups (Student’s t-test).
The total AA concentration of IF was comparable to BF while essential AA
levels, but not Thr, were significantly lower in BF than in IF.
4. Discussion
4.1. Growth, acceptance and tolerance
This randomized, controlled, double-blind intervention study
demonstrated that the growth of infants fed a modified infant formula with
reduced protein with ALAB and LC-PUFA is similar and within normal
ranges for formula fed infants. The absolute growth values of the IF (30.2 6.3
g/d) and CF (28.3 6.5 g/d) were comparable (in PP and ITT population) and
similar to those found in other studies, e.g. 27.8 4.7 g/d23 and 28.1 5.4 g/d.33
In the BeMIM study, no statistical differences in weight gain were found
between the groups, whereas a tendency towards higher weight gain was
observed in IF infants. Gain in weight and length was proportional in both
study formula groups (IF: R ¼ 0.323, P ¼ 0.003; CF: 0.391, P < 0.001). Weight-
for-length z-score was higher in IF infants compared to CF infants at all time
points studied (Fig. 1, supplemental data). This is in contrast to other studies
that found lower weight-for-length z-scores in infants fed a protein reduced
formulae.7,33 To assess the long term growth and anthropometry, follow-up
visits of these children, specifically at the age of 4 and 6 years are planned.
All infants accepted IF well and for all parameters studied no negative
effects were been found. The significantly lower rate in total adverse events
(adverse event plus serious adverse event) in the IF group compared to the CF
group could be related to the ALAB addition. ALAB was proposed to have
immunostimulatory effects and antibacterial properties and thus be protective
against
infections.16
4.2. Nutrition and satiety
The nutritional intake was adequate from study formulae, did not result in
inferior growth and not in compensatory greater intake of formula,
complementary diet or liquids compared to BF infants. Our findings in energy
and protein intake differed from Fomon et al.1999. 34 In this study infants of
the lower protein group (1.7 g protein/100 kcal) consumed significantly more
formula and thus significantly more energy than the infants with the higher
protein formula. In Fomon’s study, this resulted in a higher BMI of the lower
protein group infants suggesting a higher fat deposition.34 Thus, compensation
for an inadequate protein intake was observed by Fomon et al. 1999 and it was
concluded that the formula with 1.7 g protein/100 kcal may not be safe.34 In
BeMIM study, no significantly higher intake in the IF group was found, thus
the protein content of 1.89 g/100 kcal can be considered adequate with the
protein quality provided. Plasma concentrations of urea depend on mainly
 M. Fleddermann et al. / Clinical Nutrition 33 (2014) 588e595
dietary protein intake and indicate the level of nitrogen excretion if nitrogen
intake exceeds the needs for protein synthesis and growth or dietary proteins
are imbalanced.23 Plasma urea concentrations of IF and CF infants were
similar, and both were significantly higher than in BF infants which is
consistent with a higher protein intake with both formulae. Different protein
intakes have been related to differences in growth between breastfed and
formula fed infants.35
The energy intake via study formulae was similar to previously reported
intakes of healthy term infants.7,14,36 However, IF infants had a significantly
lower energy intake than CF infants at 90 and 120 days of age. While volume
intake per meal was not different, the lower intake in the IF group was
explained by a lower meal frequency, which might indicate a higher satiety.
The adequate protein intake and a balanced AA composition of IF were
improved by substitution of b-lactoglobulin by ALAB (rich in Trp) and
addition of Trp and Phe as free AA. Thereby, a significantly higher Trp to
large neutral AA ratio (Trp/large neutral AA) in plasma was achieved in IF
(median: 0.11; IQR: 0.10; 0.13) compared to CF (0.10:0.09; 0.11, P < 0.001).
While, Trp and large neutral AA compete for transport proteins in the blood-
brain-barrier, a higher Trp/large neutral AA ratio results in higher Trp levels
in the brain.36 Trp as precursor of the neurotransmitter serotonin influences
the sleep-wake rhythm and is involved in the appetite regulation.37 According
to Halford et al. 2005 supplementation of serotonin precursors, Trp or
5hydroxytryptophan decreases food intake.38 Thus, via an increased Trp/large
neutral AA ratio, IF may have led to higher satiety and less formula intake in
the IF group.
4.3. Energetic efficiency
Both formulae appear safe with respect to infant growth. The apparently
minor differences in composition seem to affect energetic efficiency for
growth. The higher energetic efficiency of the new infant formula (IF), in
spite of its lower protein content, could be due to the improvement of protein
composition by the addition of ALAB. Davis et al. 2008 compared the growth
of infants, who were fed a formula with replacement of 2.2 g/L b-
lactoglobulin by 2.2 g/L ALAB to the growth of infants receiving an
unmodified formula (1.3 g/L ALAB).14 Growth was not statistically different
between groups, but tended to be higher in the ALAB group. Furthermore, a
slightly lower energy intake was observed in the ALAB-rich formula group.
The ALAB-rich formula group had a 0.55 g/100 kcal (calculated from energy
intake and weight gain) higher weight gain than the standard formula group.
Davis et al. 2008 observed this difference after 8 weeks of intervention. 14 In
BeMIM study a difference of 0.78 g/100 kcal was observed after 12 weeks of
intervention.
Several beneficial properties have been ascribed to ALAB in infant
formula: ALAB is highly digestible and provides a wellbalanced AA mixture,
while positively affecting mineral absorption, gut microflora and immune
function.16 Furthermore fat absorption might influence energetic efficiency of
infant formulae. The fat absorption coefficient in formula fed term infants is
about 90%.39 IF had a markedly higher phospholipid content than CF (no
phospholipids) which even exceeded phospholipid content of human milk fat
consisting of >98% triacylglycerols, 0.7% phospholipids and 0.5%
cholesterol.40 In BeMIM study, phospholipids from egg oil were used as main
source of arachidonic acid resulting in higher plasma levels of this fatty acid
in the IF group compared to CF, which did not provide LC-PUFA. The
bioavailability of LC-PUFA depends on the chemical form of fatty acid
esters. Amate et al. investigated the effect of different LC-PUFA sources,
triacylglycerols or phospholipids, on the rate of fat acid absorption in rats.41
Their results showed, that the fat absorption of some saturated fatty acids and
docosahexaenoic acid from egg-phospholipids were higher than from
triacylglycerols. Sala-Vila et al. (2004),42 argued that the incorporation of LC-
PUFA into plasma phospholipids depends more on the fatty acid composition
of the diet fed than on the source (TG or PL) of the dietary LC-PUFA.
593
Although, a positive relation between LC-PUFA, especially arachidonic acid,
and growth has been observed by Carlson et al. (1993),43 we found no
correlation between weight gain and plasma levels of arachidonic acid or
docosahexaenoic acid in BeMIM study. This is in line with several other
studies indicating that arachidonic acid and docosahexaenoic acid
supplementation do not significantly influence weight, length or head
circumference gain.6,44,45
Several meta-analyses show the protective effect of breastfeeding on
overweight.46e48 The inverse relationship between duration of breastfeeding and
the risk of overweight has been related to the protein and caloric intake. 46,47 The
multicentric European CHOP study, in 1138 healthy, formula fed infants received
a protein reduced or a higher protein infant and follow-on formula, has shown
that decreasing the protein intake from 11.7 to 7.1% of energy in the infant
formula and from 17.6 to 8.8% in the follow-on formula was associated with
normalization of early weight gain as compared to breastfed infants. 7 A higher
protein intake in formula fed infants results in higher BMI levels than in breastfed
infants and thus may increase the risk for later obesity.7,49 Thereby, the metabolic
programming of later obesity risk seems to occur within one or two years of
life.50,51 Thus, a high weight gain during this period should be avoided.
A limitation of this study is, that more than one components of the IF have
been changed compared to CF. The new modified formula (IF) according to the
European directives of 200628 was compared to a standard formula (CF) of
European directives of 1999.29 Thereby, protein source, macronutrient
composition as well as content of LC-PUFA differed between the formulas. Thus,
we can only assume that ALAB addition and improvement of dietary AA
composition was the major reason for improved energy efficiency of IF.
We conclude that an ALAB containing infant formula with a protein/energy
ratio of 1.89 g/100 kcal and improved LC-PUFA status provides an adequate
intake during the first months of life can be considered safe. This modified infant
formula is appropriate for term infants as evidenced by growth velocities,
acceptance and tolerance. We consider it most likely that the higher energetic
efficiency of the modified infant formula due to the higher content of ALAB and
Trp. The results obtained in this trial support the general conclusion that all
considerable changes in infant formula composition should be evaluated by
clinical trials inlcuding assessments of infant growth.
Source of funding
The study was financially supported by HiPP GmbH & Co Vertrieb KG
(Pfaffenhofen, Germany).
Statement of authorship
HD, VG and BK were responsible for conception and study design; TN and
NL were responsible for recruitment and study visits; BT was responsible for the
coordination of the study in Belgrade; MF performed sample analysis, and MF
and VG were responsible for statistical analyses; MF, HD and BK undertook data
interpretation; MF, HD were responsible for writing the manuscript. Finally, all
authors critically revised the manuscript.
Conflict of interest
All authors have made substantial contributions to the conception and design
of the study, or acquisition of data, or analysis and interpretation of data, drafting
the article or revising it critically for important intellectual content. Each author
has seen and approved the contents of the submitted manuscript. BT was
employed byHipp GmbH &Co. ExportKG,Austria.Otherwise,noneof the
otherauthors had any personal or financial conflict of interest.
594 M. Fleddermann et al. / Clinical Nutrition 33 (2014) 588e595

Acknowledgment 25. Fomon SJ, Ziegler EE, Nelson SE, Frantz JA. What is the safe protein-energy ratio for infant
formula. Am J Clin Nutr 1995;62:358e63.
26. WHO Multicentre Growth Reference Study Group. WHO child growth standards:
We are very grateful for the wonderful support of the study by Nina Lukic. length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass
Furthermore, it gives us a great pleasure in acknowledging the support and help index-for-age: methods and development. Geneva: World Health Organization; 2006.
of Svetlana Milenkovic and Violeta Bankovic during the recruitment. 27. Haschke F, van’t Hof MA, Euro-Growth Study Group. Euro-growth references for length,
weight, and body circumferences. J Pediatr Gastroenterol Nutr 2000;31:S14e38.
28. Commission Directive 2006/141/EC of 22 December 2006 on infant formulaeand follow-on
Appendix A. Supplementary data formulae and amending Directive 1999/21/EC p. L 401/2.
29. Commission Directive 1999/21/EC on infant formulae and follow-on formulae.
30. Harder U, Koletzko B, Peissner W. Quantification of 22 plasma amino acids combining
Supplementary data related to this article can be found at http:// derivatization and ion-pair LC-MS/MS. J Chromatogr B Analyt Technol Biomed Life Sci
dx.doi.org/10.1016/j.clnu.2013.12.007. 2011;879:495e504.
31. Glaser C, Demmelmair H, Koletzko B. High-throughput analysis of fatty acid composition of
plasma glycerophospholipids. J Lipid Res 2010;51:216e21.
References 32. Glaser C, Demmelmair H, Sausenthaler A, Herbarth O, Heinrich J, Koletzko B. Fatty acid
composition of serum glycerophospholipids in children. J Pediatr 2010;157:826e31.
1. World Health Organization (WHO). The optimal duration of exclusive breastfeeding. Report 33. Trabulsi J, Capeding R, Lebumfacil J, Ramanujam K, Feng P, McSweeney, et al. Effect of an
of an expert consultation. Genf, 2001 (document WHO/NHD/ 01.09, WHO/FCH/CAH/01.24);
a-lactalbumin-enriched infant formula with lower protein on growth. Eur J Clin Nutr
2001.
2011;65:167e74.
2. Kramer MS, Kakuma R. Optimal duration of exclusive breastfeeding. Cochrane Database Syst
34. Fomon SJ, Ziegler EE, Nelson SE, Rogers RR, Frantz JA. Infant formula with protein-energy
Rev 2012;8:CD003517.
ratio of 1.7 g/100 kcal is adequate but may not be safe. J Pediatr Gastroenterol Nutr
3. Koletzko B, von Kries R, Close-Monasterolo R, Subías JE, Scaglioni S, Giovannini M, et al.
1999;28:495e501.
Can infant feeding choices modulate later obesity risk? Am J Clin Nutr 2009;89:1502Se8S.
35. Ziegler EE. Growth of breast-fed and formula-fed infants. Nestle Nutr Workshop Ser Pediatr
4. Robinson S, Fall C. Infant nutrition and later health: a review of current evidence. Nutrients
Program 2006;58:51e63.
2012;4:859e74.
36. Hanning RM, Paes B, Atkinson SA. Protein metabolism and growth of term infants in response
5. Koletzko B, Baker S, Cleghorn G, Neto UF, Gopalan S, Hernell O, et al. Global standard for
to a reduced-protein, 40:60 whey: casein formula with added tryptophan. Am J Clin Nutr
the composition of infant formula: recommendations of an ESPGHAN coordinated
1992;56:1004e11.
international expert group. J Pediatr Gastroenterol Nutr 2005;41:584e99.
37. Lien EL. Infant formulas with increased concentrations of alpha-lactalbumin. Am J Clin Nutr
6. Fleith M, Clandinin MT. Dietary PUFA for preterm and term infants: review of clinical
2003;77:1555Se8S.
studies. Crit Rev Food Sci Nutr 2005;45:205e29.
38. Halford JC, Harrold JA, Lawton CL, Blundell JE. Serotonin (5-HT) drugs: effects on appetite
7. Koletzko B,vonKriesR,Closa-Monasterolo R,EscribanoJ,ScaglioniS,GiovanniniM, et al.
expression and use for the treatment of obesity. Curr Drug Targets 2005;6:201e13.
Lower protein in infant formula is associated with lower weightup toage two years: a
39. Lindquist S, Hernell O. Lipid digestion and absorption in early life: an update. Curr Opin Clin
randomized clinical trial. Am J Clin Nutr 2009;89:1e10.
Nutr Metab Care 2010;13:314e20.
8. Lien EL, Davis AM, Euler AR, the Multicenter Study Group. Growth and safety in term infants
40. Koletzko B, Rodriguez-Palmero M, Demmelmair H, Fidler N, Jensen R, Sauerwald T.
fed reduced-protein formula with added bovine alpha-lactalbumin. J Pediatr Gastroenterol
Physiological aspects of human milk lipids. Early Hum Dev 2001;65:S3e18.
Nutr 2004;38:170e6.
41. Amate L, Gil A, Ramirez M. Feeding infant piglets formula with long-chain polyunsaturated
9. Alexy U, Kersting M, Sichert-Hellert W, Manz F, Schoech G. Energy intake and growth of 3-
fatty acids as triacylglycerols or phospholipids influences the distribution of these fatty acids
to 36-month-old German infants and children. Ann Nutr Metab 1999;43:14e22.
in plasma lipoprotein fractions. J Nutr 2001;131:1250e5.
10. Macé K, Steenhout P, Klassen P, Donnet A. Protein quality and quantity in cow’s milk-based
42. Sala-Vila A, Castellote AI, Campoy C, Rivero M, Rodriguez-Palmero M, LopezSabater MC.
formula for healthy term infants: past, present and future. Nestle Nutr Workshop Ser Pediatr
The source of long-chain PUFA in formula supplements does not affect the fatty acid
Program 2006;58:189e205.
composition of plasma lipids in full-term infants. J Nutr 2004;134:868e73.
11. Axelsson I. Effects of high protein intakes. Nestle Nutr Workshop Ser Pediatr Program
43. Carlson SE, Werkman SH, Peeples JM, Cooke RJ, Tolley EA. Arachidonic acid status
2006;58:121e31.
correlates with first year growth in preterm infants. Proc Natl Acad Sci U S A 1993;90:1073e7.
12. Socha P, Grote V, Gruszfeld D, Janas R, Demmelmair H, Closa-Monasterolo, et al. Milk
44. Campoy C, Escolano-Margarit MV, Anjos T, Szajewska H, Uauy R. Omega 3 fatty acids on
protein intake, the metabolic-endocrine response, and growth in infancy:
child growth, visual acuity and neurodevelopment. Br J Nutr 2012;107:S85e106.
data from a randomized clinical trial. Am J Clin Nutr 2011;94:1776Se84S.
45. Auestad N, Halter R, Hall RT, Blatter M, Bogle ML, Burks W, et al. Growth and development
13. Leunissen RW, Kerkhof GF, Stijnen T, Hokken-Koelega A. Timing and tempo of first-year
in term infants fed long-chain polyunsaturated fatty acids: a double-masked, randomized,
rapid growth in relation to cardiovascular and metabolic risk profile in early adulthood. J Am
parallel, prospective, multivariate study. Pediatrics 2001;108:372e81.
Med Assoc 2009;301:2234e42.
46. Arenz S, Rückerl R, Koletzko B, von Kries R. Breastfeeding and childhood obesity e a
14. Davis AM, Harris BJ, Lien EL, Pramuk K, Trabulsi J. Alpha-lactalbumin-rich infant formula
systematic review. Int J Obes 2004;28:1247e56.
fed to healthy term infants in a multicenter study: plasma essential amino acids and
47. Harder T, Bergmann R, Kallischnigg G, Plagemann A. Duration of breastfeeding and risk of
gastrointestinal tolerance. Eur J Clin Nutr 2008;62:1294e301.
overweight: a meta-analysis. Am J Epidemiol 2005;162:
15. Wurtman RJ, Wurtman JJ. Do carbohydrates affect food intake via neurotransmitter activity?
397e403.
Appetite 1988;11(Suppl. 1):42e7.
48. Weng SF, Redsell SA, Swift JA, Yang M, Glazebrook CP. Systematic review and meta-
16. Lönnerdal B, Lien LE. Nutritional and physiologic significance of a-lactalbumin in infants. analyses of risk factors for childhood overweight identifiable during infancy. Arch Dis Child
Nutr Rev 2003;61:295e305. 2012;97:1019e26.
17. Butte NF, Smith EO, Garza C. Energy utilization of breast-fed and formula-fed infants. Am J 49. Günther ALB, Buyken AE, Kroke A. The influence of habitual protein intake in early
Clin Nutr 1990;51:350e8. childhood on BMI and age at adiposity rebound: results from the DONALD Study. Int J Obes
18. Agostoni C. Role of long-chain polyunsaturated fatty acids in the first year of life. J Pediatr 2006;30:1072e9.
Gastroenterol Nutr 2008;47:S41e4. 50. Young BE, Johnson SL, Krebs NF. Biological determinants linking infant weight gain and
19. Koletzko B, Agostini C, Bergmann R, Ritzenthaler K, Shamir R. Physiological aspects of child obesity: current knowledge and future directions. Adv Nutr 2012;3:675e86.
human milk lipids and implication for infant feeding: a workshop report. Acta Pediatr 51. Koletzko B, Brands B, Poston L, Godfrey K, Demmelmair H, Early Nutrition Project. Early
2011;100:1405e15. nutrition programming of long-term health. Proc Nutr Soc 2012b;71:371e8.
20. Massiera F, Saint-Marc P, Seydoux J, Murata T, Kobayashi T, Narumiya S, et al. Arachidonic 52. Nommsen LA, Lovelady CA, Heinig MJ, Loennerdal, Dewey KG. Determinants of energy,
acid and prostacyclin signaling promote adipose tissue development: a human health concern? protein, lipid, and lactose concentrations in human milk during the first 12 mo of lactation: the
J Lipid Res 2003;44:271e9. DARLING Study. Am J Clin Nutr 1991;53:457e65.
21. Sellmayer A, Koletzko B. Polyunsaturated fatty acids and eicosanoids in infants e 53. Food and Nutrition Board. Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty
physiological and pathophysiological aspects and open questions. Lipids 1999;34:199e205. acids, cholesterol, protein, and amino acids. Washington: The National Academies Press;
22. Koletzko B, Shamir R, Ashwell M, Early Nutrition Academy (ENA), European Society for 2001.
Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN). Quality and safety
aspects of infant nutrition. Ann Nutr Metab 2012;60:179e84.
23. Räihä NC, Fazzolari-Nesci A, Cajozzo C, Puccio G, Monestier A, Moro G, et al. Whey
predominant, whey modified infant formula with protein/energy ratio of 1.8 g/100 kcal:
adequate and safe for term infants from birth to four months. J Pediatr Gastroenterol Nutr
2002;35:275e81.
24. Turck D, Grillon C, Lachambre E, Robiliard P, Beck L, Maurin JL, et al. Adequacy and safety
of an infant formula with a protein/energy ratio of 1.8 g/100 kcal and enhanced protein
efficiency for term infants during the first 4 months of life. J Pediatr Gastroenterol Nutr
2006;43:365e71.