ArticlePDF Available

Growth of infants consuming whey-predominant term infant formulas with a protein content of 1.8 g/100 kcal: A multicenter pooled analysis of individual participant data

Authors:

Abstract and Figures

Background: High protein intake during infancy may contribute to obesity later in life in infants who are not exclusively breastfed. Lowering the protein content of infant formula so it is closer to that of mature breast milk may reduce long-term risk of overweight or obesity in formula-fed infants. Objective: We assessed the effects of whey-predominant formulas with a protein content of 1.8 g/100 kcal (lower than that in most current formulas and closer to breast milk) on infant growth by comparing against WHO growth standards and breastfed infants. Design: A multicenter pooled analysis was conducted with the use of individual participant data (n = 1882) from 11 randomized controlled trials of healthy term infants. Mixed-effects models that used ANCOVA were generated to estimate weight-for-age z score (WAZ), as well as length-for-age, BMI-for-age, and head circumference-for-age z scores at age 4 mo in infants fed a lower-protein infant formula (LPF) or a lower-protein infant formula with additional active ingredients (probiotics, prebiotics, or both) (LPFA) and breastfed infants. Estimates, including 95% CIs, were compared with a ±0.5 SD of WHO growth standards, a benchmark for clinically significant differences. Results: The 95% CIs for pooled estimates of WAZ were within ±0.5 SD of WHO growth standards for the LPF [0.07 (-0.16, 0.29)] and LPFA [0.22 (0.01, 0.43)] groups. WAZ was higher in the LPF (P < 0.001) and LPFA (P = 0.003) groups than in the breastfed infants, likely because breastfed infants had a relatively low WAZ [-0.23 (-0.51, 0.05)] compared with WHO growth standards. The 95% CIs for all other z scores in the LPF and LPFA groups were within ±0.5 SD of WHO growth standards, except for head circumference, for which the upper limit of the 95% CI slightly exceeded 0.5 SD. No difference was observed in any z scores between the LPF and LPFA groups. Conclusion: Whey-predominant infant formula with a lower protein content that more closely resembles that of breast milk supports healthy growth comparable to the WHO growth standards and close to breastfed infants.
No caption available
… 
Content may be subject to copyright.
Growth of infants consuming whey-predominant term infant formulas
with a protein content of 1.8 g/100 kcal: a multicenter pooled analysis
of individual participant data
1,2
Dominik D Alexander,
3,4
Jian Yan,
5
* Lauren C Bylsma,
3
Robert S Northington,
5
Dominik Grathwohl,
6
Philippe Steenhout,
7
Peter Erdmann,
8
Evelyn Spivey-Krobath,
8
and Ferdinand Haschke
9
3
EpidStat Institute, Ann Arbor, MI;
4
EpidStat Institute, Seattle, WA;
5
Research and Development, Nestle
´Nutrition, King of Prussia, PA;
6
Nestle
´Research
Center, Lausanne, Switzerland;
7
Nestle
´Health Science and
8
Nestle
´Nutrition, Vevey, Switzerland; and
9
Paracelsus Medical University, Salzburg, Austria
ABSTRACT
Background: High protein intake during infancy may contribute to
obesity later in life in infants who are not exclusively breastfed.
Lowering the protein content of infant formula so it is closer to that
of mature breast milk may reduce long-term risk of overweight or
obesity in formula-fed infants.
Objective: We assessed the effects of whey-predominant formu-
las with a protein content of 1.8 g/100 kcal (lower than that in
most current formulas and closer to breast milk) on infant growth
by comparing against WHO growth standards and breastfed
infants.
Design: A multicenter pooled analysis was conducted with the use
of individual participant data (n= 1882) from 11 randomized con-
trolled trials of healthy term infants. Mixed-effects models that used
ANCOVA were generated to estimate weight-for-age zscore (WAZ),
as well as length-for-age, BMI-for-age, and head circumference–for-
age zscores at age 4 mo in infants fed a lower-protein infant
formula (LPF) or a lower-protein infant formula with additional
active ingredients (probiotics, prebiotics, or both) (LPFA) and
breastfed infants. Estimates, including 95% CIs, were compared
with a 60.5 SD of WHO growth standards, a benchmark for clin-
ically significant differences.
Results: The 95% CIs for pooled estimates of WAZ were within
60.5 SD of WHO growth standards for the LPF [0.07 (20.16,
0.29)] and LPFA [0.22 (0.01, 0.43)] groups. WAZ was higher in
the LPF (P,0.001) and LPFA (P= 0.003) groups than in the
breastfed infants, likely because breastfed infants had a relatively
low WAZ [20.23 (20.51, 0.05)] compared with WHO growth stan-
dards. The 95% CIs for all other zscores in the LPF and LPFA
groups were within 60.5 SD of WHO growth standards, except for
head circumference, for which the upper limit of the 95% CI
slightly exceeded 0.5 SD. No difference was observed in any
zscores between the LPF and LPFA groups.
Conclusion: Whey-predominant infant formula with a lower protein
content that more closely resembles that of breast milk supports healthy
growth comparable to the WHO growth standards and close to
breastfed infants. Am J Clin Nutr doi: 10.3945/ajcn.116.130633.
Keywords: multicenter, individual participant data, pooled anal-
ysis, low-protein infant formula, breast milk, infant growth, pro-
biotics, prebiotics
INTRODUCTION
Accumulating scientific evidence has shown that the kinetics
of early growth, such as rapid weight gain after birth, may be
associated with later risk of obesity and possible chronic disease
outcomes (1–4). Breastfeeding, compared with feeding tradi-
tional (high-protein) formulas, has been identified as a protective
factor against obesity later in life (5–8). Although the underlying
biological mechanisms are not entirely clear, one theory, known
as the “early protein hypothesis,” attributes this possible protective
relation to the protein content of feedings (5, 6). Specifically,
formula-fed infants may be exposed to a high amount of protein
that may increase their risk of later undesirable health outcomes.
Therefore, a better understanding of the potential relation be-
tween lower protein intake in early infancy and growth may
have important implications for obesity prevention.
Because of the difference in protein quality between breast milk
and infant formula, a higher protein content in many infant for-
mulas traditionally has been required to ensure that infants receive
adequate amounts of amino acids for growth and development (9).
However, advances in protein technology have led to the de-
velopment of a higher-quality whey-predominant protein (10) that
is used to manufacture lower-protein infant formula (LPF)
10
and
1
FundedbyNestle
´Nutrition, Vevey, Switzerland. Nestec provided study prod-
ucts and funding support for all 11 studies included in the analysis. This is a free
access article, distributed under terms (http://www.nutrition.org/publications/
guidelinesand-policies/license/) that permit unrestricted noncommercial use, dis-
tribution, and reproduction in any medium, provided the original work is properly
cited.
2
Supplemental Figures 1 and 2 and Supplemental Table 1 are available from
the “Online Supporting Material” link in the online posting of the article and
from the same link in the online table of contents at http://ajcn.nutrition.org.
*To whom correspondence should be addressed. E-mail: jian.yan@rd.
nestle.com.
Received January 19, 2016. Accepted for publication July 28, 2016.
doi: 10.3945/ajcn.116.130633.
10
Abbreviations used: BMIAZ, BMI-for-age zscore; HCAZ, head
circumference–for-age zscore; IPD, individual participant data; LAZ, length-
for-age zscore; LPF, lower-protein infant formula; LPFA, lower-protein
infant formula with additional active ingredients (probiotics, prebiotics, or
both); RCT, randomized controlled trial; WAZ, weight-for-age zscore.
Am J Clin Nutr doi: 10.3945/ajcn.116.130633. Printed in USA. Ó2016 American Society for Nutrition 1of10
AJCN. First published ahead of print September 7, 2016 as doi: 10.3945/ajcn.116.130633.
Copyright (C) 2016 by the American Society for Nutrition
lower-protein infant formula with additional active ingredients
(probiotics, prebiotics, or both) (LPFA). At 1.8 g protein/100 kcal,
the protein-to-energy ratio of LPF and LPFA is closer to that of
breast milk and represents the lowest regulatory permissible limit
for protein in infant formula in the United States and the Euro-
pean Union (11–13).
The protein concentration of LPF and LPFA has been dem-
onstrated to be safe while supporting early growth patterns and
metabolic outcomes closer to those of breastfed infants (14).
Although individual randomized controlled trials (RCTs) have
demonstrated that LPFs and LPFAs support adequate infant
growth, to our knowledge, anthropometric outcomes from these
randomized trials have not been synthesized with the use of
systematic methodology. Furthermore, as noted in the comments
on infant formula supplemented with probiotics and/or prebiotics
in 2010 by the European Society for Pediatric Gastroenterology,
Hepatology, and Nutrition (15), better understanding of the health
effects of such formulas compared with formula without probiotics
or prebiotics is warranted. Although preliminary pooled analyses of
weight-for-age zscore (WAZ) from some studies have been re-
ported in 2 recent reviews (16, 17), to our knowledge, no study to
date has systematically pooled all of the individual-level growth
data from different trials for infants receiving LPF, including those
containing active ingredients, and compared them against the
WHO growth standards and with breastfed infants. Pooled anal-
ysis of individual data across multiple trials can enhance the sta-
tistical precision to estimate growth parameters in infants fed LPF
or LPFA while accounting for geographic and cultural variability.
Therefore, we conducted a pooled analysis with the use of
individual participant data (IPD) across 11 randomized trials of
LPF and LPFA. In a traditional meta-analysis, summary statistics
are combined across studies, which may be subject to design
variation, use of differing analytic metrics, and variable defini-
tions of exposures and outcomes. Our approach allowed us to
perform a pooled analysis of IPD across 11 primary RCTs, thus
making it possible to create unified variable definitions and adjust
for influencing covariates such as birth characteristics (18).
Specifically, we 1) evaluated the growth of infants fed LPF or
LPFA and breastfed infants by comparing the anthropometric
zscores against the 2006 WHO growth standards (19), with
WAZ as the primary endpoint, and 2) compared anthropometric
zscores in LPF-fed, LPFA-fed, and breastfed infants.
METHODS
Study design and inclusion of individual studies
Currently, the most commonly used infant formula globally
with 1.8 g protein/100 kcal is a whey-predominant infant formula
(Supplemental Table 1) from Nestle
´Nutrition. This formula
has been tested in multiple clinical trials that have included
infant growth as an outcome, thereby providing a rich
source of data on infant formula with a specific and homoge-
neous protein content, in terms of both protein quantity and
quality. We had access to the participant-level data for all in-
cluded trials, which allowed the use of a more rigorous IPD
pooled analysis design than would a traditional meta-analysis
design based on summary statistics reported by the individual
studies. Eligibility criteria for inclusion of trials in the pooled
analysis included the following: 1) double-blind, randomized
controlled design; 2) evaluation of healthy term infants; 3) infants in
$1 study arm fed whey-predominant infant formula with a protein
content of 1.8 g/100 kcal; 4) results that included WAZ, length-for-
age zscore (LAZ), BMI-for-age zscore (BMIAZ), and head
circumference–for-age zscore (HCAZ); and 5) infants either ex-
clusively formula-fed from #4wkofageto$4moofage
(formula-fed arms) or exclusively breastfed from birth to $4mo
of age (breastfed arms). All included trials had parallel infant
formula group designs, with some having a nonrandomized
breastfed reference group as well. Research staff and infants’
parents were blinded to the type of formula during study follow-
up. All trials were conducted in accordance with the Declaration
of Helsinki and Good Clinical Practices and were approved by
respective institutional ethics committees.
Comparison with 2006 WHO Child Growth Standards
In the current analysis, zscores were calculated to compare
infants’ growth with the 2006 WHO standards. Zscores represent
the number of SD units above or below the median of the WHO
standard curves for healthy child growth, which were developed
based on the WHO Multicenter Growth Reference Study (19).
The WHO growth standards have been widely accepted as the
standard on how infants and young children should grow (20).
Data synthesis and statistical analyses
We pooled individual-level data across trials while following
a modified version of the Preferred Reporting Items for a Sys-
tematic Review and Meta-Analysis of Individual Participant Data
guidelines to ensure the validity and accurate reporting of data(21).
IPD from all trials were merged into a single database. To be
included in the pooled analysis, a subject needed to have data at
birth and month 4 and also have data for the 2 additional covariates
included in the statistical model, infant sex and delivery type. This
allowed us to take advantage of the poolingstructure of the analysis
by harmonizing covariates across different centers. All data were
carefully reviewed for missing values or data entry errors according
to standard quality control procedures. This was done for both
baseline information and outcome measurements. Among the few
database errors identified were negative ages, no birth weight
data, and inconsistent head circumference data. These accounted
for ,10 infants throughout all study centers, and data for these
participants were omitted from the analyses. The data manage-
ment flow diagram is presented in Supplemental Figure 1. Baseline
categorical data were analyzed with the use of a chi-square test, and
baseline continuous data were analyzed with the use of ANOVA.
Overall comparison of the 3 groups was conducted first, and if this
was significant at P#0.05, pairwise comparisons were done.
The primary objective was to assess the growth of LPF-fed,
LPFA-fed, and breastfed infants by comparing anthropometric
zscores with WHO growth standards. The primary endpoint was
WAZ at 4 mo of age. Four months was selected as a clinically
relevant time point that would avoid significant confounding
from the introduction of complementary feeding between 4 and
6 mo of age. To achieve the primary objective, WAZ, LAZ,
BMIAZ, and HCAZ were calculated with the use of publicly
available macros from the WHO website. The zscore estimates
at 4 mo from the individual studies were derived by ANCOVA
while adjusting for corresponding birth zscores, infant sex, and
2of10 ALEXANDER ET AL.
type of delivery. A 2-step procedure that used mixed-effects
(fixed and random) models was used for all primary analyses
(18). I
2
was calculated for the primary endpoint (WAZ) as an
indicator of the proportion of heterogeneity in the meta-analysis
model. In the first step, fixed-effects models with the use of IPD
were generated with ANCOVA that corrected for corresponding
birth zscores, infant sex, delivery type, and study. In the second
step, random-effects models were created to estimate the overall
weighted group mean effects while accounting for between-study
variance, with the use of the study-specific group mean estimates
and variance data. This allowed us to use and control for the in-
dividual data in all infants, estimate the within-group variance for
each study, and produce overall summary effects while accounting
for both within- and between-study variability. Finally, the zscore
estimates and 95% CIs were compared with 60.5 SD of the WHO
growth standards. Specifically, when the lower bound of the 95%
CI was .20.5 SD and the higher bound of the 95% CI was ,0.5
SD, anthropometric parameters were considered to be not statis-
tically or clinically different from the WHO growth standards. The
selection of 60.5 SD as the clinically significant benchmark for
WAZ is consistent with previous studies (20, 22, 23) and with the
recommendation from the American Academy of Pediatrics to use
3 g/d as a clinically relevant difference in weight gain in infant
feeding clinical trials (24). A growth difference of 3 g/d will result
in a 366 g difference in weight gain after 4 mo. When compared
against the WHO growth standard at 4 mo, 366 g translates to
a 0.52 SD for girls and a 0.50 SD for boys. Thus, as a measure of
clinical relevance, a 3-g/d difference in weight gain translates into
a60.5 SD for WAZ at 4 mo. The same benchmark of 60.5 SD
was also used as an indication of clinical significance for other
anthropometric parameters in the study, and this is consistent with
previous studies that applied 0.5 SD as a benchmark for assessing
infant LAZ and HCAZ (20, 22, 23).
The secondary objective of our analysis was to compare the
growth of infants fed LPF or LPFA with that of breastfed infants.
To achieve this objective, zscores were compared between the
3 feeding groups with the use of ANCOVA with feeding group
as the factor in the model while adjusting for corresponding birth
zscores, infant sex, delivery type, and study. As an exploratory
analysis, we evaluated the rate of weight gain based on change in
WAZ from birth to 4 mo of age (WAZ at 4 mo minus WAZ at
birth) between the 3 feeding groups. The change in WAZ
was classified as “slow” (,20.67), “gradual” (20.67 to 0.67), or
“rapid” (.0.67). Such classification based on increments of 0.67
SD has been used in previous studies (25, 26), and the 3 cate-
gories for change in WAZ are equivalent to downward, aver-
age, and upward crossings of major weight percentiles, with
a 0.67 SD change corresponding to a change in one major
percentile band (e.g., 50th to 75th) on a growth chart (25).
Rapid early weight gain (change of .0.67) has been linked to
an increased risk of overweight in later life (25). The distri-
bution of infants in these 3 categories was compared between
the LPF, LPFA, and breastfed groups with the use of a chi-
square test. A Fisher’s exact test subsequently was used for
pairwise comparisons. In previous studies (25, 26), the clas-
sification was applied to the change in WAZ from birth to
age 6 mo. However, to avoid potential confounding effects of
complementary feeding on the rate of weight gain, we focused
on the change from birth to 4 mo, when infants still were fed
exclusively with either formula or breast milk.
All statistical analyses were conducted with the use of SAS
Statistical Software, version 9.1. Data reported are estimated
means (95% CIs) unless otherwise noted.
RESULTS
Characteristics of infants included in the pooled analysis
Individual-level data from 11 RCTs were included in the
multicenter pooled analysis (Table 1). Studies were conducted
between 1998 and 2008 and were performed in 6 different
countries on 4 different continents. Ten studies used LPF as an
infant feeding arm; 9 studies used LPFA that contained prebi-
otics, probiotics, or both; and 5 studies used breastfed infants as
a reference arm. The overall dropout rate among all study par-
ticipants was w24%, with slightly higher rates of dropout in the
breastfed groups than in the formula-fed groups (30% compared
with 22%). We pooled data from a total of 1882 healthy term
infants; 737 in the LPF group, 965 in the LPFA group, and 180
in the breastfed group met all predefined requirements for the
WAZ endpoint analysis. The number of infants included in the
other growth endpoint analyses was somewhat lower because of
missing baseline values (Table 2). The proportion of male and
female infants was balanced (w1:1) in each of the 3 feeding
groups. There was a significantly higher rate of cesarean section
delivery in the LPF and LPFA groups than in the breastfed group,
and the 2 groups of formula-fed infants had significantly lower
WAZ and BMIAZ at birth than did the breastfed group (Table 2).
WAZ at 4 mo of age
The pooled WAZ estimate for the LPF group was 0.07 (20.16,
0.29) based on analysis of 737 infants from 10 studies (Figure 1).
The 95% CI range for the pooled estimate was well within
60.5 SD of the WHO growth standards; this was also true for
7 of the 10 studies with LPF arms. Results from individual studies
were relatively homogeneous, with the exception of the study
conducted in China (31), in which the mean estimate for WAZ
was 0.92 and the lower bound of the 95% CI was 0.79 (which
is .0.5 SD). A sensitivity analysis conducted without the data
from this study resulted in a pooled WAZ estimate of 20.03
(20.12, 0.05) based on analysis of 554 infants. Furthermore, the
I
2
heterogeneity test was 94% with the China study included, but
the model became homogeneous after removal of this study (I
2
=8%).
The pooled WAZ estimate for the LPFA group was 0.22 (0.01,
0.43) based on analysis of 965 infants from 9 studies (Figure 1).
Similar to the LPF analysis results, the 95% CI for the pooled
estimate was within 60.5 SD; this was also true for 6 of 9 studies
with LPFA arms. The China study (31) again appeared to be an
outlier, with a mean WAZ estimate of 0.81 and a lower bound of
the 95% CI of 0.68. When this study was removed in a sensitivity
analysis (n= 776), the pooled WAZ estimate was 0.14 (0.02, 0.26)
and the proportion of variance due to heterogeneity (I
2
)changed
from 92% to 42%.
The pooled WAZ estimate for the breastfed group was 20.23
(20.51, 0.05) based on analysis of 180 infants from 5 studies
(Figure 1), with an I
2
of 64%. The pooled mean WAZ estimate
was skewed slightly lower than the WHO growth standards by
data from 2 French studies (32, 33), both of which had mean
WAZ estimates and corresponding lower bounds of the 95% CIs
that were ,20.5 SD.
FORMULA PROTEIN CONTENT AND INFANT GROWTH 3of10
When compared between the 3 feeding groups (Table 3), the
WAZ estimates were significantly higher in the LPF (P,0.001)
and LPFA (P= 0.003) groups than they were in the breastfed
group. The mean WAZ differences were 0.30 (0.14, 0.46) for
the LPF group compared with breastfed infants and 0.24
(0.08, 0.39) for the LPFA group compared with breastfed
infants; both differences were ,0.5 SD. No significant dif-
ference was detected between the LPF and LPFA groups on
WAZ estimates (P=0.17).
LAZ at 4 mo of age
The pooled LAZ estimates for the LPF, LPFA, and breastfed
groups (Figure 2) were 20.02 (20.20, 0.17), 0.14 (20.06, 0.34)
and 0.19 (0.04, 0.34) based on analysis of 699, 921, and 180
infants from 10, 9, and 5 studies, respectively. The 95% CIs of
the pooled estimates were all within 60.5 SD. No statistically
significant differences in LAZ estimates were detected between
the LPF, LPFA, and breastfed groups (P.0.16, Table 3).
BMIAZ at 4 mo of age
The pooled BMIAZ estimates for the LPF and LPFA groups
(Figure 3) were 0.07 (20.19, 0.33) and 0.11 (20.10, 0.33)
based on analysis of 699 and 921 infants from 10 and 9 studies,
respectively. The 95% CIs for both groups were within 60.5 SD.
Similar to with the WAZ results, the China study (31) appeared
to be an outlier, with mean BMIAZ estimates of 0.89 and 0.69
with lower bounds of the 95% CIs of 0.70 and 0.52 (both of
which were .0.5 SD).
The pooled BMIAZ estimate for the breastfed group (Figure 3)
was 20.53 (20.79, 20.28) based on analysis of 180 infants
from 5 studies. Results from individual studies were homoge-
neous, with mean BMIAZ estimates for all individual studies
close to or below 20.5 SD. When compared with the LPF and
LPFA groups (Table 3), the breastfed group had significantly
lower BMIAZ (P,0.001 for breastfed infants compared with
the LPF group and P= 0.001 for BF infants compared with the
LPFA group). The difference was driven by the low BMIAZ
for the breastfed group, with BMIAZ estimates for the LPF
TABLE 1
Studies included in the multicenter IPD pooled analysis
1
Study, year Study arms included in the pooled analysis Reference
Italy, 1998 Formula-fed: 1.8 g protein/100 kcal Ra
¨iha
¨et al., 2002 (14)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs
Breastfed
Italy, 1999 Formula-fed: 1.8 g protein/100 kcal Barclay et al., 2003 (27)
Formula-fed: 1.8 g protein/100 kcal + prebiotics (Raftilose)
Formula-fed: 1.8 g protein/100 kcal + probiotics (BB12)
Formula-fed: 1.8 g protein/100 kcal + prebiotics (Raftilose) + probiotics (BB12)
Australia, 2002 Formula-fed: 1.8 g protein/100 kcal Gibson et al., 2009 (28)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + probiotics (B. lactis CNCM
I-3446)
Italy, 2003 Formula-fed: 1.8 g protein/100 kcal + LCPUFAs Puccio et al., 2007 (29)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + prebiotics (BMOSs) + probiotics
(BL999)
France, 2003 Formula-fed: 1.8 g protein/100 kcal + LCPUFAs Chouraqui et al., 2008 (30)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + probiotics (BL999 + LPR)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + prebiotics (BMOSs) + probiotics
(BL999 + LPR)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + prebiotics (BMOSs) + probiotics
(BL999 + ST11)
China, 2003 Formula-fed: 1.8 g protein/100 kcal + LCPUFAs Wu et al., 2016 (31)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + probiotics (BL999)
France, 2005a Formula-fed: 1.8 g protein/100 kcal Putet et al., 2016 (32)
Breastfed
France, 2005b Formula-fed: 1.8 g protein/100 kcal Hascoe
¨t et al., 2011 (33)
Formula-fed: 1.8 g protein/100 kcal + probiotics (BL999)
Breastfed
Italy, 2005 Formula-fed: 1.8 g protein/100 kcal + LCPUFAs Meli et al., 2014 (34)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + prebiotics (BMOSs)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + prebiotics + probiotics
(BMOSs + BL999 + LPR)
Breastfed
South Africa, 2007 Formula-fed: 1.8 g protein/100 kcal + LCPUFAs Cooper et al., 2015 (35)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + prebiotics (BMOSs) + probiotics
(B. lactis CNCM I-3446)
Greece, 2008 Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + probiotics (B. lactis CNCM I-3446) Baglatzi et al., 2016 (36)
Formula-fed: 1.8 g protein/100 kcal + LCPUFAs + probiotics (B. lactis CNCM
I-3446, higher level)
Breastfed
1
BB12, Bifidobacterium lactis BB12; BL999, Bifidobacterium longum (ATCC BAA999); BMOS, bovine milk–derived oligosaccharides; IPD, individual
participant data; LCPUFA, long-chain PUFA; LPR, Lactobacillus rhamnosus CGMCC 1.3724; ST11, Lactobacillus paracasei CNCM I-2116.
4of10 ALEXANDER ET AL.
and LPFA groups being similar to the WHO growth standards
(Figure 3).
HCAZ at 4 mo of age
The pooled HCAZ estimates for the LPF and LPFA groups
(Figure 4) were 0.37 (0.19, 0.55) and 0.45 (0.28, 0.62) based on
analysis of 736 and 959 infants from 10 and 9 studies, respectively.
Results from individual studies of these 2 groups were homoge-
neous, with mean HCAZ estimates close to 0.5 SD, except for
those for the South Africa study (35), in which mothers of infants
were HIV-positive. Interestingly, infants fed LPF and LPFA in the
South Africa study exhibited mean HCAZ estimates close to 1 SD.
The pooled HCAZ estimate for the breastfed group (Figure 4) was
0.27 (0.13, 0.4) based on analysis of 178 infants from 5 studies. No
significant differences between the LPF, LPFA, and breastfed
groups were detected on HCAZ estimates (P.0.4, Table 3).
Rate of weight gain based on change in WAZ from birth to
age 4 mo
The proportion of infants in weight gain categories based on
change in WAZ differed (P,0.001) between the LPF, LPFA,
and breastfed groups (Figure 5A). Specifically, compared with
TABLE 2
Baseline characteristics of infants included in the IPD pooled analysis according to feeding groups
1
LPF LPFA BF
Studies included, n10 9 5
Female 50 49 49
Cesarean section delivery 43 52 27*
Birth WAZ (n)20.10 60.84 (737) 20.12 60.90 (965) 0.14 60.77 (180)
#
Birth LAZ (n) 0.03 61.07 (700) 0.09 61.08 (921) 0.21 60.91 (180)
Birth BMIAZ (n)20.17 61.08 (700) 20.27 61.13 (921) 0.05 60.99 (180)
#
Birth HCAZ
2
(n) 0.14 61.07 (554) 0.22 61.06 (775) 0.30 60.93 (180)
1
Values are means 6SDs or percentages, unless otherwise indicated. Categorical data were analyzed with the use of
a chi-square test, and continuous data were analyzed with the use of ANOVA. An overall comparison between 3 groups was
conducted first, and if significant at P#0.05, pairwise comparisons between 2 groups were obtained. *LPF and LPFA are
significantly different from BF and are also different from each other at P#0.05.
#
LPF and LPFA are significantly different
from BF but not different from each other. BF, breastfed; BMIAZ, BMI-for-age zscore; HCAZ, head circumference–for-
age zscore; IPD, individual participant data; LAZ, length-for-age zscore; LPF, lower-protein infant formula; LPFA, lower-
protein infant formula with additional active ingredients (probiotics, prebiotics, or both); WAZ, weight-for-age zscore.
2
Birth HCAZ of infants from the study conducted in China (31) (which included LPF and LPFA arms) was not available.
FIGURE 1 IPD pooled analysis of WAZ in infants at 4 mo of age for the LPF, LPFA, and BF groups. Values are estimated means calculated from
ANCOVA with the use of random-effects models adjusted for birth WAZ, infant sex, delivery type, and study. The solid circles represent the estimated mean
from individual studies, and the horizontal lines represent the 95% CIs for the mean. The diamonds represent the pooled mean estimate, with the horizontal
tips of the diamond representing the lower and upper limits of the 95% CIs. BF, breastfed; IPD, individual participant data; LPF, lower-protein infant formula;
LPFA, lower-protein infant formula with additional active ingredients (probiotics, prebiotics, or both); WAZ, weight-for-age zscore.
FORMULA PROTEIN CONTENT AND INFANT GROWTH 5of10
the breastfed group, the LPF (P,0.001) and LPFA (P,0.001)
groups had a lower proportion of infants in the slow category
(22% and 18%, respectively, compared with 42%), a similar
proportion of infants in the gradual category (48% and 50%,
respectively, compared with 49%) and a higher proportion of
infants in the rapid category (30% and 32%, respectively,
compared with 9%). No difference was detected between the
LPF and LPFA groups (P= 0.13).
The China study (31) again appeared to be an outlier (Sup-
plemental Figure 2), with 57% and 54% of infants in the rapid
category for the LPF and LPFA groups, respectively. In addition,
the China study did not include a breastfed reference group;
therefore, no Chinese infants were included in the breastfed group
in the current analysis. Further analysis that excluded data from
the China study showed that the proportions of infants in weight
gain categories in the LPF (P,0.001) and LPFA (P,0.001)
groups still differed from those of the breastfed group, but were
numerically closer (Figure 5B), i.e., 21% and 26% instead of 30%
and 32% in the rapid category for the LPF and LPFA groups, re-
spectively. After the China study was excluded, the LPF group
differed from the LPFA group (P= 0.015), with a higher proportion
of infants in the slow category (27% compared with 22%), no
difference in the gradual category (52% compared with 52%), and
a lower proportion in the rapid category (21% compared with 26%).
DISCUSSION
We conducted a comprehensive pooled analysis on individual
data from 1882 healthy term infants in 11 RCTs to evaluate the
effects of a whey-predominant infant formula with lower protein
content with and without added active ingredients (prebiotics,
probiotics, or both) on growth parameters at 4 mo of age. Using
TABLE 3
Anthropometric zscore differences between feeding groups at age 4 mo
1
LPF vs. BF LPFA vs. BF LPFA vs. LPF
WAZ difference 0.30 (0.14, 0.46)* 0.24 (0.08, 0.39)* 20.06 (20.15, 0.03)
LAZ difference 0.04 (20.15, 0.23) 20.03 (20.21, 0.15) 20.07 (20.17, 0.03)
BMIAZ difference 0.37 (20.19, 0.33)* 0.11 (20.10, 0.33)* 20.53 (20.79, 20.28)
HCAZ difference 0.02 (20.16, 0.19) 20.01 (20.18, 0.15) 20.03 (20.12, 0.06)
1
Values are estimated zscore mean differences (95% CIs) between feeding groups (LPF 2BF, LPFA 2BF, and LPFA 2
LPF) calculated from ANCOVA while adjusting for corresponding birth zscores, infant sex, delivery type, and study.
*P#0.003. BF, breastfed; BMIAZ, BMI-for-age zscore; HCAZ, head circumference–for-age zscore; LAZ, length-for-age
zscore; LPF, lower-protein infant formula; LPFA, lower-protein infant formula with additional active ingredients (pro-
biotics, prebiotics, or both); WAZ, weight-for-age zscore.
FIGURE 2 IPD pooled analysis of LAZ in infants at 4 mo of age for the LPF, LPFA, and BF groups. Values are estimated means calculated from
ANCOVA with the use of random-effects models adjusted for birth LAZ, infant sex, delivery type, and study. The solid circles represent the estimated mean
from individual studies, and the horizontal lines represent the 95% CIs for the mean. The diamonds represent the pooled mean estimate, with the horizontal
tips of the diamond representing the lower and upper limits of the 95% CIs. BF, breastfed; IPD, individual participant data; LAZ, length-for-age zscore; LPF,
lower-protein infant formula; LPFA, lower-protein infant formula with additional active ingredients (probiotics, prebiotics, or both).
6of10 ALEXANDER ET AL.
pooled analysis methodology of IPD, we were able to further
analyze study data by harmonizing the covariates under study
with uniform analytic metrics (e.g., correcting for baseline
characteristics). Our pooled analyses generated summary associ-
ations with greater precision (i.e., enhanced statistical power) than
any of the individual studies. By virtue of combining individual
data, we created a single, larger analysis with greater analytic
control. This methodology is widely used (e.g., in the Harvard
Pooling Project of Prospective Studies of Diet and Cancer; https://
www.hsph.harvard.edu/pooling-project/about-the-study/).
The results of our analyses showed that a whey-predominant
infant formula with a protein content of 1.8 g/100 kcal supports
healthy growth that is comparable to the WHO growth standards.
Specifically, the WAZ, LAZ, and BMIAZ pooled estimates and
95% CIs at 4 mo in LPF- or LPFA-fed infants were well within
60.5 SD of the WHO growth standards. However, there was
some degree of data inflection within some of the models. This
mainly was due to outlier results from one study conducted in
Shanghai, a major urbanized city in China. The estimates for
WAZ and BMIAZ at 4 mo of age for LPF- and LPFA-fed infants
from the Chinese study (31) were considerably higher than those
from the other studies included in the analysis. This observation
is consistent with a recent publication (37) based on the Chinese
fourth National Survey on the Physical Growth and Development
of Children, which reported that urban Chinese infants were
heavier than those included in the WHO Multicenter Growth
Reference Study. Regardless, the exclusion of data from this
Chinese study did not modify the pooled results significantly. The
results of our analyses also showed that including specific active
ingredients (i.e., prebiotics, probiotics, or both) in LPF did not
significantly affect growth parameters at 4 mo of age. This is con-
sistent with a recent systematic review by Szajewska et al. (38),
which showed that infants fed Bifidobacterium lactis–supplemented
formula grew at a rate that was similar to that of infants fed
unsupplemented formula. One major difference between our
analysis of formula supplemented with active ingredients and
that of Szajewska et al. (38) is that our LPFA group included
several different types of active ingredients: different strains of
probiotics (e.g., B. lactis CNCM I-3446, BL999) or prebiotics
(bovine milk–derived oligosaccharides), or both. An interesting
finding is that the change in WAZ from birth to 4 mo of age
differed between the LPF and LPFA groups (Figure 5) after the
China study was excluded; however, interpretation of this finding
should be made in the context of the exploratory and descriptive
nature of this particular analysis.
It is noteworthy that HCAZ estimates of LPF- and LPFA-fed
infants were consistently higher than the WHO growth standards,
with infants in the South Africa study of HIV-positive mothers
exhibiting the greatest deviation. HCAZ estimates in breastfed
infants included in the analysis were also similarly higher than
the WHO standard. These HCAZ results mirrored the findings
from a recent systematic review (20), which showed that the
WHO head circumference data are at the lower end of head
circumference measurements from large studies of economically
advantaged children in 30 countries. The observed difference
between the WHO head circumference standard and recent studies
(including ours) may be the result of differences in head cir-
cumference measurement techniques; however, as noted in Natale
et al. (20), there is still a sizable difference between the WHO head
circumference data and a large European study that used a strict
FIGURE 3 IPD pooled analysis of BMIAZ in infants at 4 mo of age for the LPF, LPFA, and BF groups. Values are estimated means calculated from
ANCOVA with the use of random-effects models adjusted for birth BMIAZ, infant sex, delivery type, and study. The solid circles represent the estimated mean
from individual studies, and the horizontal lines represent the 95% CIs for the mean. The diamonds represent the pooled mean estimate, with the horizontal
tips of the diamond representing the lower and upper limits of the 95% CIs. BF, breastfed; BMIAZ, BMI-for-age zscore; IPD, individual participant data; LPF,
lower-protein infant formula; LPFA, lower-protein infant formula with additional active ingredients (probiotics, prebiotics, or both).
FORMULA PROTEIN CONTENT AND INFANT GROWTH 7of10
standardized measurement technique that mirrored the WHO study
methodology (39). Collectively, our IPD pooled analysis findings
support the conclusion of Natale et al. (20), and suggest that
additional research may be needed to justify using a single in-
ternational head circumference standard.
An interesting finding from our IPD pooled analysis was that
breastfed infants appeared to deviate from the WHO growth
standards for weight, length, and BMI to a larger extent than did
LPF- or LPFA-fed infants. Specifically, breastfed infants in our
study tended to be lighter and longer, and therefore manifested
FIGURE 4 IPD pooled analysis of HCAZ in infants at 4 mo of age for the LPF, LPFA, and BF groups. Values are estimated means calculated from
ANCOVA with the use of random-effects models adjusted for birth HCAZ (birth HCAZ for the China study (31) was not available, and birth WAZ was used
instead for this study), infant sex, delivery type, and study. The solid circles represent the estimated mean from individual studies, and the horizontal lines
represent the 95% CIs for the mean. The diamonds represent the pooled mean estimate, with the horizontal tips of the diamond representing the lower and
upper limits of the 95% CIs. BF, breastfed; HCAZ, head circumference–for-age zscore; IPD, individual participant data; LPF, lower-protein infant formula;
LPFA, lower-protein infant formula with additional active ingredients (probiotics, prebiotics, or both); WAZ, weight-for-age zscore.
FIGURE 5 Rate of weight gain between birth and 4 mo of age for the LPF, LPFA, and BF groups. IPD from all studies (A). The LPF (P,0.001) and
LPFA (P,0.001) groups were significantly different from the BF group, whereas no difference between the LPF and LPFA groups was detected (P= 0.13).
IPD from all studies excluding the China study (31) (B). The LPF (P,0.001) and LPFA (P,0.001) groups were significantly different from the BF group,
and the LPF group also was significantly different (P= 0.015) from the LPFA group. The WAZ change was calculated as WAZ at 4 mo of age minus WAZ at
birth. Values are percentage of infants in weight-gain categories based on WAZ change as slow (,20.67), gradual (20.67 to 0.67), or rapid (.0.67). A chi-
square test was used to compare the LPF, LPFA, and BF groups, and a Fisher’s exact test subsequently was used for pairwise comparisons. BF, breastfed; IPD,
individual participant data; LPF, lower-protein infant formula; LPFA, lower-protein infant formula with additional active ingredients (probiotics, prebiotics, or
both); WAZ, weight-for-age zscore.
8of10 ALEXANDER ET AL.
a significantly lower BMI than the WHO growth standards.
Multiple reasons may underlie this finding. For example, the
breastfed infants in our analysis were from France, Italy, and
Greece, all of which are countries not included in the WHO study.
In addition, the sample size (n= 180) of the breastfed group was
relatively small. Nonetheless, the relatively lower weight and
BMI of the breastfed infants included in our analysis was ac-
ceptable, as it was within 22 SD of the WHO standards. The
considerably homogeneous results of low BMIAZ in breastfed
infants in the 5 European studies included in our analysis also
are in agreement with the BMI zscores of the breastfed group
in the European Childhood Obesity Trial (6), which included
w500 breastfed infants from 5 European countries (Germany,
Belgium, Italy, Poland, and Spain). Furthermore, the breastfed
group from the Euro-Growth study also manifested relatively
low WAZ (compared to WHO Growth Standards) at 4 mo of age
(40). Additional research is warranted to explore any potential
long-term implications of the relatively lower weight and BMI
in early infancy in European infants.
Because early rapid weight gain has been shown to be asso-
ciated with obesity risk in later life (25, 26, 41, 42), the rate of
weight gain based on the change in WAZ was explored. In the
LPF, LPFA, and breastfed groups, w50% of the infants were in
the gradual weight-gain category. However, a relatively greater
proportion of breastfed infants (42% compared with #27% of
formula-fed infants) were in the slow category, and a relatively
greater proportion of formula-fed infants ($21% compared with
9% of breastfed infants) were in the rapid category, despite
consumption of formulas with a lower protein content closer to
that of breast milk. The observed difference in the rate of weight
gain may arise in part from a higher milk intake in formula-fed
infants than in breastfed infants (43). It also may be a statistical
caveat, because there were fewer infants in the breastfed group
(5 studies; n= 180). In addition, our ability to interpret these
findings is somewhat restricted because it is uncertain whether
the rate of weight gain in the breastfed group was skewed or
truly reflective of how infants should grow. For example, it is
theoretically desirable to have all infants in the gradual weight
gain category; however, only w50% of the breastfed infants in
the current analysis were in the gradual category, whereas 42%
were in the slow category.
One of the major strengths of our analysis was the ability to
pool individual-level data across 11 RCTs, with greater analytic
control being facilitated by the ability to harmonize covariates,
definitions, and analytic metrics. Another key strength was the
randomized controlled design of the included studies. This type
of design allows for greater control of influential confounding
factors, and is less susceptible to residual confounding than
observational studies. Moreover, there is control over the allo-
cation and, in theory, the compliance with the type of infant
feeding, given the experimental components of RCTs. Finally,
the analysis of data over a decade of research, at different study
centers, and across numerous geographic regions may have
enhanced the generalizability of our research findings.
Some limitations of our analysis also warrant mention. We
focused on infant formula with a lower-protein content (1.8 g
protein/100 kcal) and a specific protein quality (whey-predominant);
thus, findings may not be applicable to LPFs with a different protein
quality (e.g., casein-predominant). Maternal height and BMI (im-
portant factors influencing child growth, which were not controlled
in the analysis) may differ between the infant population evaluated
in the WHO growth reference study and the studies included in this
analysis. In addition, the RCTs used in the current analysis differed
somewhat in terms of operational methodology; not all trials in-
cluded both LPF and LPFA arms, and fewer trials included arms
with breastfed infants, thus resulting in a variable proportion of
infants representing the 3 study groups. However, given the fact that
we were able to conduct the analysis on individual-level data, we
had analytic latitude such that we could synchronize similar ex-
posure groups and outcome classifications.
In conclusion, it is well established that obesity is a major
public health burden in most developed countries, with a fore-
telling of risk in developing countries worldwide. Accumulating
scientific evidence suggests that excessive protein intake during
infancy may increase the subsequent risk of obesity. The results
of our pooled analysis of IPD from 11 RCTs have indicated that
feeding with a whey-predominant formula with a lower protein
content (1.8 g/100 kcal, lower than most currently available infant
formulas) that is closer to that of human milk, with or without pre-
or probiotics, supports healthy early growth comparable to the
WHO growth standards and close to that of breastfed infants. The
different rate of weight gain based on the change in WAZ between
breastfed infants and those fed LPF or LPFA warrants further
investigation.
We thank the lead investigators of the 11 studies: Niels Ra
¨iha
¨(Italy, 1998;
Italy, 1999), Maria Makrides (Australia, 2002), Giuseppe Puccio (Italy,
2003; Italy, 2005), Jean-Pierre Chouraqui (France, 2003), Weiping Wang
(China, 2003), Guy Putet and Jean-Charles Picaud (France, 2005a), Jean-
Michel Hascoe
¨t (France, 2005b), Peter A Cooper (South Africa, 2007), and
Christos Costalos (Greece, 2008).
The authors’ responsibilities were as follows—JY, RSN, DG, PS, PE, ES-K,
and FH: conceptualized and designed the study; RSN and DG: conducted the
analysis; DDA, JY, and LCB: wrote the paper with input from all authors; and
all authors: read and approved the final manuscript. DDA and LCB are em-
ployees of EpidStat Institute; JY, RSN, DG, PS, PE, and ES-K are employees
of Nestle
´; and FH was the chairman of the Nestle
´Nutrition Institute at the time
of initiation of the study.
REFERENCES
1. Singhal A, Lanigan J. Breastfeeding, early growth and later obesity.
Obes Rev 2007;8(Suppl 1):51–4.
2. Toschke AM, Grote V, Koletzko B, von Kries R. Identifying children at
high risk for overweight at school entry by weight gain during the first
2 years. Arch Pediatr Adolesc Med 2004;158:449–52.
3. Stettler N. Nature and strength of epidemiological evidence for origins
of childhood and adulthood obesity in the first year of life. Int J Obes
(Lond) 2007;31:1035–43.
4. Wells JC. The programming effects of early growth. Early Hum Dev
2007;83:743–8.
5. Weber M, Grote V, Closa-Monasterolo R, Escribano J, Langhendries
JP, Dain E, Giovannini M, Verduci E, Gruszfeld D, Socha P, et al.
Lower protein content in infant formula reduces BMI and obesity risk
at school age: follow-up of a randomized trial. Am J Clin Nutr 2014;
99:1041–51.
6. Koletzko B, von Kreis R, Closa R, Escribano J, Scaglioni S, Giovannini
M, Beyer J, Demmelmair H, Anton B, Gruszfeld D, et al. Can infant
feeding choices modulate later obesity risk? Am J Clin Nutr 2009;89:
1502S–8S. Corrected and republished from: Am J Clin Nutr 2009;
90(1):248.
7. Monasta L, Batty GD, Cattaneo A, Lutje V, Ronfani L, Van Lenthe FJ,
Brug J. Early-life determinants of overweight and obesity: a review of
systematic reviews. Obes Rev 2010;11:695–708.
8. Inostroza J, Haschke F, Steenhout P, Grathwohl D, Nelson SE, Ziegler
EE. Low-protein formula slows weight gain in infants of overweight
mothers. J Pediatr Gastroenterol Nutr 2014;59:70–7.
FORMULA PROTEIN CONTENT AND INFANT GROWTH 9of10
9. Lo
¨nnerdal B. Infant formula and infant nutrition: bioactive proteins of
human milk and implications for composition of infant formulas. Am J
Clin Nutr 2014;99:712S–7S.
10. Erdmann P, Neumann F. Process for treatment of a lactic raw materials.
Treatment of lactic materials. US patent 6787158 B1. 2014 Sep 7.
11. FDA. Title 21: food and drugs. Washington (DC): Office of the Federal
Government; 2015.
12. World Health Organization. CODEX alimentarius: standard for infant
formula and formulas for special medical purposes intended for infants.
Revised standard for infant formula. Rome (Italy): Food and Agriculture
Organization of the United Nations; 1981.
13. European Union. Commission Delegated Regulation (EU) 2016/127.
Official Journal of the European Union 2016;59.
14. Ra
¨iha
¨NC, Fazzolari-Nesci A, Cajozzo C, Puccio G, Monestier A,
Moro G, Minoli I, Haschke-Becher E, Bachmann C, Van’t Hof M, 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;35:275–81.
15. Braegger C, Chmielewska AF, Decsi TF, Kolacek S, Mihatsch W,
Moreno LF, Piescik MF, Puntis J, Shamir R, Szajewska H, et al.
Supplementation of infant formula with probiotics and/or prebiotics:
a systematic review and comment by the ESPGHAN committee on
nutrition. J Pediatr Gastroenterol Nutr 2011;52:238–50.
16. Haschke F, Steenhout P, Grathwohl D, Haschke-Becher E. Evaluation
of growth and early infant feeding: a challenge for scientists, industry
and regulatory bodies. World Rev Nutr Diet 2013;106:33–8.
17. Haschke F, Grathwohl D, Detzel P, Steenhout P, Wagemans N. Post-
natal high protein intake can contribute to accelerated weight gain of
infants and increased obesity risk. Edition edited. In: Fewtrell MS,
Haschke F, Prescott SL, editors. 85th Nestle Nutrition Institute work-
shop, London, November 2014: Preventive aspects of early nutrition.
London, 2016.
18. Blettner M, Sauerbrei W, Schlehofer B, Scheuchenpflug T, Friedenreich
C. Traditional reviews, meta-analyses and pooled analyses in epidemi-
ology. Int J Epidemiol 1999;28:1–9.
19. WHO. WHO child growth standards: length/height-for-age, weight-
for-age, weight-for-length, weight-for-height and body mass index-for-
age, methods and development. Geneva (Switzerland): World Health
Organization; 2006.
20. Natale V, Rajagopalan A. Worldwide variation in human growth and
the World Health Organization growth standards: a systematic review.
BMJ Open 2014;4(1):e003735.
21. Stewart LA, Clarke M, Rovers M, Riley RD, Simmonds M, Stewart G,
Tierney JF. Preferred reporting items for systematic review and meta-
analyses of individual participant data: the PRISMA-IPD statement.
JAMA 2015;313:1657–65.
22. WHO Multicentre Growth Reference Study Group. Assessment of
differences in linear growth among populations in the WHO Multi-
centre Growth Reference Study. Acta Paediatr Suppl 2006;450(Suppl):
56–65.
23. WHO. An evaluation of infant growth: the use and interpretation of
anthropometry in infants. Bull World Health Organ 1995;73:165–74.
24. American Academy of Pediatrics. Clinical testing of infant for-
mulas with respect to nutritonal suitability for term infants. Elk
Grove (IL): American Academy of Pediatrics Committee on Nu-
trition; 1988.
25. Ekelund U, Ong K, Linne Y, Neovius M, Brage S, Dunger DB,
Wareham NJ, Rossner S. Upward weight percentile crossing in infancy
and early childhood independently predicts fat mass in young adults:
the Stockholm Weight Development Study (SWEDES). Am J Clin
Nutr 2006;83:324–30.
26. Ong KK, Loos RJ. Rapid infancy weight gain and subsequent obe-
sity: systematic reviews and hopeful suggestions. Acta Paediatr
2006;95:904–8.
27. Barclay D, Puccio G, Fazzolari-Nesci A, Giammanco A, Ra
¨iha
¨N,
Carrie-Fa
¨ssier A, Brown C, Chauffard F, Grathwohl D, Hager C, et al.
Growth and tolerance of a whey-based starter infant formula with
enhanced protein efficiency and containing Pro-, Pre-, or synbiotics. A
randomized controlled trial in term infants. J Pediatr Gastroenterol
Nutr 2003;37:388.
28. Gibson RA, Barclay D, Marshall H, Moulin J, Maire JC, Makrides M.
Safety of supplementing infant formula with long-chain polyunsaturated
fatty acids and Bifidobacterium lactis in term infants: a randomised
controlled trial. Br J Nutr 2009;101:1706–13.
29. Puccio G, Cajozzo C, Meli F, Rochat F, Grathwohl D, Steenhout P.
Clinical evaluation of a new starter formula for infants containing live
Bifidobacterium longum BL999 and prebiotics. Nutrition 2007;23:1–8.
30. Chouraqui JP, Grathwohl D, Labaune JM, Hascoet JM, de Montgolfier
I, Leclaire M, Giarre M, Steenhout P. Assessment of the safety, tol-
erance, and protective effect against diarrhea of infant formulas
containing mixtures of probiotics or probiotics and prebiotics in
a randomized controlled trial. Am J Clin Nutr 2008;87:1365–73.
31. Wu BB, Yang Y, Xu X, Wang WP. Effects of Bifidobacterium sup-
plementation on intestinal microbiota composition and the immune
response in healthy infants. World J Pediatr 2016;12:177–82.
32. Putet G, Labaune JM, Mace K, Steenhout P, Grathwohl D, Raverot V,
Morel Y, Picaud JC. Effect of dietary protein on plasma insulin-like
growth factor-1, growth, and body composition in healthy term infants:
a randomised, double-blind, controlled trial (Early Protein and Obesity
in Childhood (EPOCH) study). Br J Nutr 2016;115:271–84.
33. Hascoe
¨t JM, Hubert C, Rochat F, Legagneur H, Gaga S, Emady-Azar
S, Steenhout PG. Effect of formula composition on the development of
infant gut microbiota. J Pediatr Gastroenterol Nutr 2011;52:756–62.
34. Meli F, Puccio G, Cajozzo C, Ricottone GL, Pecquet S, Sprenger N,
Steenhout P. Growth and safety evaluation of infant formulae con-
taining oligosaccharides derived from bovine milk: a randomized,
double-blind, noninferiority trial. BMC Pediatr 2014;14:306.
35. Cooper PA, Bolton KD, Velaphi SC, Pecquet SS, Steenhout PG.
Normal growth of infants born from HIV + mothers, after normal or
caesarean delivery, fed a pre and probiotics supplemented starter for-
mula. The European Society for Paediatric Gastroenterology Hepatology
and Nutrition Amsterdam, 2015.
36. Baglatzi L, Gavrili S, Stamouli K, Zachaki S, Favre L, Pecquet S,
Benyacoub J, Costalos C. Effect of infant formula containing a low
dose of the probiotic Bifidobacterium lactis CNCM I-3446 on immune
and gut functions in C-section delivered babies: a pilot study. Clin Med
Insights Pediatr 2016;10:11–9.
37. Zong XN, Li H. Construction of a new growth references for China
based on urban Chinese children: comparison with the WHO growth
standards. PLoS One 2013;8:e59569.
38. Szajewska H, Chmielewska A. Growth of infants fed formula supple-
mented with Bifidobacterium lactis Bb12 or Lactobacillus GG: a system-
atic review of randomized controlled trials. BMC Pediatr 2013;13:185.
39. van’t Hof MA, Haschke F. The Euro-Growth Study: why, who, and
how. J Pediatr Gastroenterol Nutr 2000;31(Suppl 1):S3–13.
40. Ziegler EE. Obesity prevention: is the protein intake in infants too
high? The Nestle Nutrition Institute Newsletter. South East Asia and
Pacific Rim. 2013;(7):3.
41. Nettleton JA, Jebb S, Riserus U, Koletzko B, Fleming J. Role of dietary
fats in the prevention and treatment of the metabolic syndrome. Ann
Nutr Metab 2014;64:167–78.
42. Druet C, Stettler N, Sharp S, Simmons RK, Cooper C, Smith GD,
Ekelund U, Levy-Marchal C, Jarvelin MR, Kuh D, et al. Prediction of
childhood obesity by infancy weight gain: an individual-level meta-
analysis. Paediatr Perinat Epidemiol 2012;26:19–26.
43. Kramer MS, Guo T, Platt RW, Vanilovich I, Sevkovskaya Z, Dzikovich
I, Michaelsen KF, Dewey K. Feeding effects on growth during infancy.
J Pediatr 2004;145:600–5.
10 of 10 ALEXANDER ET AL.
... Yet, since vitamin D and iron have been identified to have critical effects on body growth, we had subsequently investigated the trajectory of growth pattern in our subjects. Our findings regarding body weight was consistent with several studies that also demonstrated lighter weight in breastfed infants when compared to formula-fed infants [15][16][17] . Rapid and excess weight gain during the first year of life was reported to be a predictor of later obesity. ...
... In contrast to body weight that showed universally smaller weight in breastfed infants, reports on body height were more conflicting. Some reports showed no difference in mean z-scores for height between breastfed and formula-fed infants 17,21 . However, similar to our observation, several studies had shown exclusive breastfeeding to be associated with shorter stature 15,16,22,23 . ...
Article
Full-text available
This study aimed to compare the trajectory of serum 25(OH)D, micronutrient levels, and anthropometric measurements between exclusively breastfed and mixed-fed children. This is a prospective cohort study. Anthropometric measurements of the children were obtained during scheduled clinical visits. Tests for 25(OHD), ferritin, zinc and complete blood count were performed yearly until 3 years of age. Clinical records and questionnaires on dietary habits were obtained. The results showed that despite official recommendations on vitamin D/iron supplements for breastfed children, less than 10% of our exclusively breastfed children received regular supplements. Thus, after 1 year, the odds for having iron deficiency anemia and vitamin D insufficiency were 9 [95% CI, 4–19] and 6 [95% CI, 2–16], respectively. Longitudinal follow-up showed the prevalence of iron deficiency to decrease from 34% at 1 year to 2% at age 3 years. However, the prevalence of vitamin D insufficiency remained persistently high throughout the first three years of life (60% at 1 to 44% at 3 years). Very few children had zinc deficiency. Anthropometric measurements showed exclusively breastfed children to have lower mean z-scores for body weight and height when compared to mixed-fed children after 12 months. In conclusion, children who were exclusively breastfed for longer than 4 months without proper supplement were more likely to have transient iron deficiency anemia and persistent vitamin D insufficiency. Their growth became relatively slower after infancy. Whether this was associated with underlying inadequate serum vitamin D and iron level remains an important issue to be explored.
... In the current study, despite the modest sample sizes of the PP population in the intervention groups, the Concept IF group demonstrated an equivalent daily weight gain compared to the Control IF group after adjustment for birthweight and ethnicity as confounding factors. Moreover, the mean infant growth outcomes in all randomized formula groups of the current study were close to the median of the WHO growth standards (within ±0.5 z-score bandwidth), indicative of adequate infant growth [35,36]. Together, these findings support the conclusion of the previous (fully powered) equivalence study [34] demonstrating that an infant formula with large, milk phospholipid-coated lipid droplets (containing dairy lipids) supports adequate infant growth. ...
... Hence, caution should be used in the interpretation of the outcomes of the current study in relation to its nutritional interventions. The inclusion of Malay, Indian as well as Chinese infants in this study may have introduced additional confounding, given the previously observed differences in growth trajectories [33][34][35]. Moreover, the milk feeding characteristics of these ethnic populations might be different as well, e.g., evident from the higher number of Chinese infants in the breastfed reference group versus the randomized IF groups. ...
Article
Full-text available
Lipids are essential for healthy infant growth and development. The structural complexity of lipids in human milk is not present in infant milk formula (IF). A concept IF was developed mimicking more closely the structure and composition of human milk fat globules. The current study evaluates whether a concept IF with large, milk phospholipid-coated lipid droplets (mode diameter 3 to 5 μm) is equivalent to standard IF with regard to growth adequacy and safety in healthy, term Asian infants. In this randomized, double-blind, controlled trial, infants were randomized after parents decided to introduce formula. Infants received a standard IF with (Control) or without the specific prebiotic mixture scGOS/lcFOS (9:1 ratio; Control w/o prebiotics), or a Concept IF with large, milk phospholipid-coated lipid droplets and the prebiotic mixture. A group of 67 breastfed infants served as a reference. As a priori defined, only those infants who were fully intervention formula-fed ≤28 days of age were included in the equivalence analysis (Control n = 29; Control w/o prebiotics n = 28; Concept n = 35, per-protocol population). Primary outcome was daily weight gain during the first four months of life, with the difference between the Concept and Control as the key comparison of interest. Additionally, adverse events, growth and tolerance parameters were evaluated. Equivalence of daily weight gain was demonstrated between the Concept and Control group after additional correction for ethnicity and birthweight (difference in estimated means of 0.1 g/d, 90%CI [−2.30, 2.47]; equivalence margin +/− 3 g/d). No clinically relevant group differences were observed in secondary growth outcomes, tolerance outcomes or number, severity or relatedness of adverse events. This study corroborates that an infant formula with large, milk phospholipid-coated lipid droplets supports adequate growth and is well tolerated and safe for use in healthy infants.
... Compared to human milk, total protein in infant formula has traditionally been higher to ensure sufficient protein quality to support normal growth. However, increasing evidence suggests infant formulas that have lower protein concentrations, closer to human milk, support adequate growth and potentially benefit later growth outcomes [11,[19][20][21][22][23][24][25]. For iron, an important functional component of many proteins and an essential mineral for brain development, human milk content is relatively low (~0.35 mg/L) and reflected by the dietary iron requirement (0.27 mg/day) through to 6 months of age [26]. ...
Article
Full-text available
Inclusion of bovine-derived milk fat globule membrane (bMFGM) or bMFGM components in infant formulas (IFs) may support healthy brain development. This double-blind, prospective trial evaluated growth, tolerance, and iron status in infants receiving added bMFGM and modified protein, iron, and arachidonic acid (ARA) concentrations in IF. Healthy term infants were randomized to: control (marketed, routine cow’s milk-based IF/100 kcal: 2.1 g protein, 1.8 mg iron, 34 mg ARA) or INV-MFGM (investigational cow’s milk-based IF/100 kcal: 1.9 g protein, 1.2 mg iron, 25 mg ARA and whey protein-lipid concentrate, 5 g/L (source of bMFGM)). Anthropometrics, stool characteristics, fussiness, and gassiness through day 365 and blood markers of iron status at day 365 were evaluated. The primary outcome was rate of weight gain from 14–120 days of age. Of 373 infants enrolled (control: 191, INV-MFGM: 182), 275 completed the study (control: 141; INV-MFGM: 134). No group differences in growth rate (g/day) from day 14–120 or study discontinuation were detected. Few group differences in growth or parent-reported fussiness, gassiness, or stool characteristics were detected. No group differences were detected in hemoglobin, hematocrit, or incidence of anemia. In healthy term infants, bMFGM and modified protein, iron, and ARA concentrations in a cow’s milk-based IF were well-tolerated, associated with adequate growth throughout the first year of life, and supported normal iron status at one year of age.
... В пользу указанной теории свидетельствуют результаты 11 клинических исследований (1882 участника), согласно которым дети, получавшие питание с содержанием белка (1,8 г на 100 ккал) и аминокислотным профилем, сопоставимым с профилем грудного молока (ГМ), имели массо-ростовые показатели, сравнимые с таковыми у детей на естественном вскармливании [6]. ...
Article
Full-text available
E.A. Pyr’eva, A.I. Safronova, E.A. Netunaeva, M.I. Timoshina Federal Research Center for Nutrition & Biotechnology, Moscow, Russian Federation This paper discusses the introduction of complementary feeding considering the novel data on its role in ensuring the health and development of a child. One of the most controversial issues is protein intake that significantly changes both quantitatively and qualitatively after introducing supplementary food. Excessive protein intake is associated with metabolic load and growth acceleration that is considered the predictor of overweight and obesity, metabolic syndrome and diabetes. The importance of meat puree in the first year of life (also from the perspective of optimizing nutrition to achieve adequate zinc and iron levels) is highlighted. Domestic and international studies on the effects of proteins of various origins (meat, milk, etc.) on growth and the risk of obesity are addressed. The importance of qualitative rather than quantitative composition of consumed protein is emphasized. Keywords: young children, supplementary food, protein, meat, growth, insulin-like growth factor. For citation: Pyr’eva E.A., Safronova A.I., Netunaeva E.A., Timoshina M.I. Role and sources of protein in the nutrition of young children. Russian Journal of Woman and Child Health. 2021;4(1):65–69. DOI: 10.32364/2618-8430-2021-4-1-65-69.
... 52 Some exclusively breastfed infants experience excessive weight gain during the first 6 months and may later have a catch-down effect after starting solid foods. [53][54][55] Thus, in an attempt to understand what variable components of human milk might result in weight gain, recent studies investigated the link between HMO composition and the effects on infant growth and body composition. [18][19][20][21] Small studies link HMOs to growth in early infancy affecting both weight-for-age z-scores (WAZ) and length-for-age z-scores (LAZ) depending upon the specific HMO. ...
Article
Full-text available
The relationship between human milk oligosaccharides (HMOs) and infant growth and adiposity is not fully understood and comprehensive studies are missing from the current literature. We screened and recruited 370 healthy, pregnant women and their infants from seven European countries. Breastmilk samples were collected using standardized procedures at six time points over 4 months, as were infant parameters. Correlations and associations between HMO area under the curve, anthropometric data, and fat mass at 4 months were tested. Lacto-N-neotetraose had a negative correlation with the change in length (rs = −0.18, P = 0.02). Sialyllacto-N-tetraose c (LSTc) had a positive correlation with weight for length (rs = 0.19, P = 0.015). Infants at the 25th upper percentile were fed milk higher in 3′-sialyllactose and LSTc (P = 0.017 and P = 0.006, respectively) compared to the lower 25th percentile of the weight-for-length z-score gain over 4 months of lactation. No significant associations between growth and body composition and Lewis or secretor-dependent HMOs like 2′-fucosyllactose were identified. Changes in the HMO composition of breastmilk during the first 4 months appear to have little influence on infant growth and body composition in this cohort of healthy mothers and infants. Modest associations exist between individual HMO and infant growth outcomes at least in healthy growing populations. Our study provides a comprehensive investigation of associations between all major HMO and infant growth and adiposity including several time points. Certain groups of HMOs, like the sialylated, may be associated with adiposity during the first months of lactation. HMO may modulate the risk of future metabolic disease. Future population studies need to address the role of specific groups of HMOs in the context of health and disease to understand the long-term impact.
... Increasing research suggests that high protein intake may be detrimental to metabolic risk factors such as for overweight in infants and toddlers [46][47][48] by increasing body weight in the absence of improvements in fat-free mass [49]. It is believed that protein consumption stimulates the secretion of insulin-like growth factor I, which leads to cellular proliferation, accelerated growth and increased body fat [50,51]. ...
Article
Full-text available
The present study aimed to assess energy intake, nutrient profile and food sources in Spanish children participating in the EsNuPI (“Estudio Nutricional en Población Infantil Española”) study. Plausibility of energy intake and adequacy of nutrient intakes to international recommendations were analyzed in a final sample of 1448 subjects (728 boys and 720 girls) and one group representative of the 1 to <10 years old urban Spanish children (reference sample (n = 707)) who consumed milk and one of the same age who consumed adapted milk over the last year (adapted milk consumers sample (n = 741)) were compared. Both groups completed data of a face-to-face and a telephone 24-h dietary recalls. Both the reference and the adapted milk consumers samples reported an adequate daily energy intake (1503 kcal/day and 1404 kcal/day); and a high contribution to total energy from protein (16.5% and 15.6%) and fat (36.5% and 35.9%). Also, a high percentage of children from both samples were below the lower limit of the recommendations for carbohydrates (47.8% and 39.3%). As the percentage of plausible energy reporters was high for both groups (84.7% and 83.5%, respectively), data for the whole sample were analyzed. Milk and dairy, cereals, meat and derived products, fats and oils, bakery and pastry, fruits and vegetables contributed to about 80% of the total energy intake in both groups. However, the reference sample reported significantly more contribution to energy from cereals, meat and meat products, bakery and pastry and ready to cook/eat foods; meanwhile, the adapted milk consumers sample reported significantly more energy from milk and dairy products, fruits and eggs. Those results suggest that adapted milk consumers have better adherence to the food-based dietary guidelines. Further analyses are warranted to characterize food patterns and the quality of the diet in the EsNuPI study population.
... Manufacturers who need to rely on alternate methods of study design to assess growth need a sound scientific basis to do so. While there have been a few reviews examining primary studies on selected nutritional attributes of infant formulas (4,5) there is no review assessing the potential impact of deviations from the GMS criteria. How likely is it that an alternate study design will reach a conclusion different than that which would have been found under the GMS criteria? ...
Article
Full-text available
Growth of infants fed isocaloric infant formulas differing in nutritional content was studied. Twenty-three of 109 randomized clinical trials reported some difference in weight, length or head circumference between formula groups. Logistic regression demonstrated no relationship between finding a significant difference in a growth outcome with enrollment prior to 15 d or observation of 15 wk, parameters specified in regulation. Sample size and year of publication also were not correlated with report of a significant growth difference, though reporting separate data by sex was (p = 0.012). The difference in mean weight gain between control and test formula groups was comparable to that between formula-fed and breast-fed infants (1 g/d) and smaller than that between male and female infants (4 g/d). Encouraging alternate study designs with flexible enrollment ages and infants who transition from breastfeeding to formula would gain information on physiologic outcomes and common feeding behaviors, as well as growth.
Preprint
Background. Increasing evidence demonstrate that concentration of protein in infant formula >1.9g/100Kcal with high levels of insulinogenic aminoacids is associated with accelerated weight gain, increased fat mass accumulation and risk of adiposity. Purpose of this study was to conduct a systematic review to determine the metabolic effects in infants feed with infant formula optimized in protein. Methods. Systematic review was conducted according PRISMA Statement. RCTs with one intervention group (infant formula with 1.6-1.9gr of protein/100Kcal) and at least one comparative control group (infant formula with >1.9gr of protein/100Kcal) were included. Standardized mean differences (SMD), through random model were calculated. Results. 15 RCT were included. Optimized protein in infant formula was associated with less gain of BMI at 24 months of follow-up (SMD -0.25, IC95% -0.36 to -0.13, p 0.01) and less fat mass accumulation (SMD -0.68, IC95% -0.98 to -0.37, p 0.01). Optimized protein was also associated to less gain of weight, weight/age Z-score, weight/length Z-score, BUN (mmol/dL) and IGF1 (ng/ml). No effect on length/age Z-score was observed. Conclusions. Robust evidence showed optimized protein (1.6gr/100Kcal to 1.9gr/100Kcal) in infant formula is associated with metabolic benefits in infants with less weight gain, BMI and fat mass accumulation.
Article
Full-text available
Introduction: human milk oligosaccharides (HMOs) are an important component of human milk supporting the development of a balanced intestinal microbiota and immune protection in breastfed infants. Randomized controlled trials (RCTs) have demonstrated that infant formulas supplemented with the HMOs 2'-fucosyllactose (2'FL) and lacto-N-neotetraose (LNnT) are safe, well-tolerated, and support normal growth. This Real-World Evidence (RWE) study aimed to evaluate growth and tolerance in infants consuming a formula supplemented with 1 g/L of 2'FL and 0.5 g/L of LNnT, and included a mixed-feeding group never studied before in RCTs. Participants and methods: this open-label, prospective study was conducted at six centers in Spain, and included healthy, exclusively breastfed infants (BF group), an exclusively formula-fed group (FF) who received a milk-based formula with 2' FL and LNnT, and a group mixed fed with both formula and human milk (MF), for 8 weeks. Co-primary outcomes were growth (anthropometry) and gastrointestinal tolerance (Infant Gastrointestinal Symptom Questionnaire, IGSQ). Secondary outcomes included formula satisfaction and adverse events (AEs). Results: 159 infants completed the study (66 FF, 48 MF, and 45 BF). Mean z-scores for growth were similar between all groups and within ± 0.5 of WHO medians at week 8. Composite IGSQ scores demonstrated low GI distress in all groups, with no significant group differences at baseline, week 4, or week 8. Incidence of AEs was low overall, and comparable across groups. Conclusions: in this RWE study examining a HMO-supplemented infant formula, growth and tolerance outcomes were similar to RCT findings, supporting the effectiveness of this early feeding option.
Article
Background: A high protein intake in early life is associated with a risk of obesity later in life. The essential amino acid requirements of formula-fed infants have been reassessed recently, enabling a reduction in total protein content and thus in protein intake. Objectives: We aimed to assess the safety of an infant formula with a modified amino acid profile and a modified low-protein (mLP) content in healthy term-born infants. Outcomes were compared with a specifically designed control (CTRL) infant formula. Methods: In this double-blind, randomized controlled equivalence trial, infants received either mLP (1.7 g protein/100 kcal; n = 90) or CTRL formula (2.1 g protein/100 kcal; n = 88) from enrollment (age ≤ 45 d) to 6 mo of age. A breastfed group served as a reference (n = 67). Anthropometry and body composition were determined at baseline, 17 wk (including safety blood parameters), and 6 mo of age. The primary outcome was daily weight gain from enrollment up until the age of 17 wk (at an equivalence margin of ±3.0 g/d). Results: Weight gain from baseline (mean ± SD age: 31 ± 9 d) up to the age of 17 wk was equivalent between the mLP and CTRL formula groups (27.9 and 28.8 g/d, respectively; difference: -0.86 g/d; 90% CI: -2.36, 0.63 g/d). No differences in other growth parameters, body composition, or in adverse events were observed. Urea was significantly lower in the mLP formula group than in the CTRL formula group (-0.74 mmol/L; 95% CI: -0.97, -0.51 mmol/L; P < 0.001). Growth rates, fat mass, fat-free mass, and several essential amino acids were significantly higher in both formula groups than in the breastfed reference group. Conclusions: Feeding an infant formula with a modified amino acid profile and a lower protein content from an average age of 1 mo until the age of 6 mo is safe and supports an adequate growth, similar to that of infants consuming CTRL formula. This trial was registered at www.trialregister.nl as Trial NL4677.
Article
Full-text available
Background: In the absence of breast-feeding and its immunomodulatory factors, supplementation of starter infant formula (IF) with probiotics is currently used to support immune functions and gut development. Aim: To assess whether immune-related beneficial effects of regular dose (10(7) CFU/g of powder) of the probiotic Bifidobacterium lactis CNCM I-3446 (hereafter named B. lactis) in starter IF supplementation can be maintained with starter IF containing a low dose (10(4) CFU/g of powder) of B. lactis. Method: This trial was designed as a pilot, prospective, double-blind, randomized, single-center clinical trial of two parallel groups (n = 77 infants/group) of C-section delivered infants receiving a starter IF containing either low dose or regular dose of the probiotic B. lactis from birth to six months of age. In addition, a reference group of infants breast-fed for a minimum of four months (n = 44 infants), also born by C-section, were included. All groups were then provided follow-up formula without B. lactis up to 12 months of age. Occurrence of diarrhea, immune and gut maturation, responses to vaccinations, and growth were assessed from birth to 12 months. The effect of low-dose B. lactis formula was compared to regular-dose B. lactis formula, considered as reference for IF with probiotics, and both were further compared to breast-feeding as a physiological reference. Results: Data showed that feeding low-dose B. lactis IF provides similar effects as feeding regular-dose B. lactis IF or breast milk. No consistent statistical differences regarding early life protection against gastrointestinal infections, immune and gut maturation, microbiota establishment, and growth were observed between randomized formula-fed groups as well as with the breast-fed reference group. Conclusion: This pilot study suggests that supplementing C-section born neonates with low-dose B. lactis-containing starter formula may impact immune as well as gut maturation similarly to regular-dose B. lactis, close to the breast-feeding reference.
Article
Full-text available
Aim: To assess differences in length/height among populations in the WHO Multicentre Growth Reference Study (MGRS) and to evaluate the appropriateness of pooling data for the purpose of constructing a single international growth standard. Methods: The MGRS collected growth data and related information from 8440 affluent children from widely differing ethnic backgrounds and cultural settings (Brazil, Ghana, India, Norway, Oman and the USA). Eligibility criteria included breastfeeding, no maternal smoking and environments supportive of unconstrained growth. The study combined longitudinal (birth to 24 mo) and cross-sectional (18-71 mo) components. For the longitudinal component, mother-infant pairs were enrolled at delivery and visited 21 times over the next 2 y. Rigorous methods of data collection and standardized procedures were applied across study sites. We evaluate the total variability of length attributable to sites and individuals, differences in length/height among sites, and the impact of excluding single sites on the percentiles of the remaining pooled sample. Results: Proportions of total variability attributable to sites and individuals within sites were 3% and 70%, respectively. Differences in length and height ranged from -0.33 to +0.49 and -0.41 to +0.46 standard deviation units (SDs), respectively, most values being below 0.2 SDs. Differences in length on exclusion of single sites ranged from -0.10 to +0.07, -0.07 to +0.13, and -0.25 to +0.09 SDs, for the 50th, 3rd and 97th percentiles, respectively. Corresponding values forheightranged from -0.09 to +0.08, -0.12 to +0.13, and -0.15 to +0.07 SDs. Conclusion: The striking similarity in linear growth among children in the six sites justifies pooling the data and constructing a single international standard from birth to 5 y of age.
Article
Full-text available
The effect of protein intake on growth velocity in infancy may be mediated by insulin-like growth factor-1 (IGF-1). This study aimed to determine the effects of formulae containing 1·8 (F1·8) or 2·7 g (F2·7) protein/418·4 kJ (100 kcal) on IGF-1 concentrations and growth. Healthy term infants were randomly assigned to receive F1·8 (n 74) or F2·7 (n 80) exclusively for the first 4 months of life. A group of breast-fed infants (n 84) was followed-up simultaneously (reference). Growth and body composition were measured at 0·5, 4, 6, 12, 36, 48 and 60 months of life. The IGF-1 concentrations at 4 months (primary outcome) were similar in the F1·8 (67·1 (sd 20·8) ng/l; n 70) and F2·7 (71·2 (sd 27·5) ng/l; n 73) groups (P=0·52). Both formula groups had higher IGF-1 concentrations than the breast-fed group at 4 and 9 months of age (P≤0·0001). During the first 60 months of life, anthropometric parameters in the F1·8 group were lower compared with the F2·7 group, and the differences were significant for head circumference from 2 to 60 months, body weight at 4 and 6 months and length at 9, 12 and 36 months of age. There were no significant differences in body composition between these two groups at any age. We conclude that, in formula-fed infants, although increased protein intake did not affect the IGF-1 concentration during the first 12 months of life, it did affect length and head circumference growth, suggesting that factors other than IGF-1 could play roles in determining growth velocity.
Article
Full-text available
Systematic reviews and meta-analyses of individual participant data (IPD) aim to collect, check, and reanalyze individual-level data from all studies addressing a particular research question and are therefore considered a gold standard approach to evidence synthesis. They are likely to be used with increasing frequency as current initiatives to share clinical trial data gain momentum and may be particularly important in reviewing controversial therapeutic areas. To develop PRISMA-IPD as a stand-alone extension to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) Statement, tailored to the specific requirements of reporting systematic reviews and meta-analyses of IPD. Although developed primarily for reviews of randomized trials, many items will apply in other contexts, including reviews of diagnosis and prognosis. Development of PRISMA-IPD followed the EQUATOR Network framework guidance and used the existing standard PRISMA Statement as a starting point to draft additional relevant material. A web-based survey informed discussion at an international workshop that included researchers, clinicians, methodologists experienced in conducting systematic reviews and meta-analyses of IPD, and journal editors. The statement was drafted and iterative refinements were made by the project, advisory, and development groups. The PRISMA-IPD Development Group reached agreement on the PRISMA-IPD checklist and flow diagram by consensus. Compared with standard PRISMA, the PRISMA-IPD checklist includes 3 new items that address (1) methods of checking the integrity of the IPD (such as pattern of randomization, data consistency, baseline imbalance, and missing data), (2) reporting any important issues that emerge, and (3) exploring variation (such as whether certain types of individual benefit more from the intervention than others). A further additional item was created by reorganization of standard PRISMA items relating to interpreting results. Wording was modified in 23 items to reflect the IPD approach. PRISMA-IPD provides guidelines for reporting systematic reviews and meta-analyses of IPD.
Article
Full-text available
Background: A limited number of nondigestible oligosaccharides are available for use in infant formula. This study evaluated growth and safety in infants fed formula supplemented with a mixture of bovine milk-derived oligosaccharides (BMOS). This mixture, which was generated from whey permeate, contains galactooligosaccharides and other oligosaccharides from bovine milk, such as 3'- and 6'-sialyllactose. We hypothesized that growth in infants fed BMOS-supplemented formula would be noninferior to that in infants fed standard formula. Methods: Healthy term infants ≤14 days old were randomly assigned to standard formula (control; n = 84); standard formula with BMOS (IF-BMOS; n = 99); or standard formula with BMOS and probiotics (Bifidobacterium longum, Lactobacillus rhamnosus) (IF-BMOS + Pro; n = 98). A breastfed reference group was also enrolled (n = 30). The primary outcome was mean weight gain/day from enrollment to age 4 months (noninferiority margin: -3.0 g/day). Results: 189 (67.3%) formula-fed infants were included in the primary analysis. Mean differences in weight gain between the control and IF-BMOS and IF-BMOS + Pro groups were <1 g/day, with 97.5% confidence intervals above -3.0 g/day, indicating noninferior weight gain in the BMOS formula groups. Compared with control, infants in the BMOS groups had more frequent (p < 0.0001) and less hard (p = 0.0003) stools. No significant differences were observed between the control and BMOS groups in caregivers' reports of flatulence, vomiting, spitting up, crying, fussing, and colic. When based on clinical evaluation by the investigator, the incidence of colic was higher (p = 0.01) in IF-BMOS than in control; the incidence of investigator-diagnosed colic was not significantly different in control and IF-BMOS + Pro (p = 0.15). Stool bifidobacteria and lactobacilli counts were higher with IF-BMOS + Pro compared with control (p < 0.05), whereas Clostridia counts were lower (p < 0.05) in both BMOS groups compared with control. Conclusions: Infant formula containing BMOS either with or without probiotics provides adequate nutrition for normal growth in healthy term infants. Further studies are needed to fully explore the digestive tolerance of BMOS formula. Trial registration: ClinicalTrials.gov NCT01886898 . Registered 24 June 2013.
Article
Full-text available
A symposium on the health significance of dietary fat in the prevention and treatment of the metabolic syndrome (MetS) was held at the 20th International Congress of Nutrition in Granada, Spain, on September 19, 2013. Four nutrition experts addressed the topics of dietary fat and obesity, effects of dietary fat quality in obesity and insulin resistance, influence of early nutrition on the later risk of MetS and the relative merits of high- or low-fat diets in counteracting MetS. Participants agreed that preventing weight gain and achieving weight loss in overweight and obese patients were key strategies for reducing MetS. Both low-fat and low-carbohydrate diets are associated with weight loss, but adherence to the diet is the most important factor in achieving success. Avoidance of high saturated fats contributes to lower health risks among obese, MetS and diabetic patients. Further, healthy maternal weight at conception and in pregnancy is more important that weight gain during pregnancy for reducing the risk of obesity in the offspring. The effects of different polyunsaturated fatty acids on MetS and weight loss require clarification. © 2014 S. Karger AG, Basel.
Article
Worldwide, 38% of women are now overweight (BMI 25-30) or obese (BMI ≥30). There is increasing evidence that maternal obesity can result in unfavorable (epigenetic) pre- and postnatal programming of important genes of the offspring. Infants of overweight mothers show faster weight gain during infancy, which is associated with higher risk of obesity during childhood and adult life. This can have lifelong consequences such as increased risk of noncommunicable diseases. Many studies indicate that infants of obese and nonobese mothers who were fed traditional (high-protein) formulas gain more rapidly weight than breastfed infants. An updated meta-analysis (n = 1,150) indicates that infants from four continents who were fed a whey-based, low-protein (1.8 g/100 kcal) formula with an essential amino-acid profile closer to breast milk grow in accordance with the World Health Organization (WHO) growth standard (0-4 months). A new experimental low-protein (1.61-1.65 g protein/100 kcal) formula for infants between 3 and 12 months of age was recently tested in two randomized clinical trials. One trial in the general US population indicates lower weight between 4 and 12 months of age in infants fed the low-protein formula when compared to infants on the high-protein formula (p = 0.031). Weight gain was not inferior to the WHO growth standards. Longitudinal analysis of odds ratios from 4 to 12 months of age showed a lower incidence of infants with weight >85th percentile in the low-protein group compared with the high-protein group (p = 0.015). In the second trial, which was conducted in Chile and included infants of mothers with BMI >25, infants fed the low-protein formula gained less weight between 4 and 12 months (p = 0.022) and until 24 months (p = 0.031) than the high-protein group. Weight gain was similar to the breastfed reference group. In both trials, biomarkers of protein metabolism (insulin-like growth factor-1 and C-peptide) of the low-protein groups were closer to breastfed infants than the respective biomarkers of the high-protein groups. Health economic analyses indicate that feeding low-protein formulas to nonbreastfed infants would result in cost savings for both the individual and the society. Preventive measures against childhood and adult obesity should include promotion of breastfeeding for 6 months or longer, and use of low-protein formulas in nonbreastfed infants.
Article
In reviewing the growth of infants who live under favourable conditions and are fed according to WHO feeding recommendations, the Working Group found significant differences between the growth patterns of these infants and the patterns reflected in the NCHS-WHO international reference. Given the short- and long-term consequences of growth failure, and the dangers of both the premature introduction of complementary foods and their undue delay described as the 'weanling's dilemma', the Working Group concluded that use of the current NCHS-WHO reference appears to accentuate the difficulty of avoiding these extremes rather than to help ensure optimal infant nutritional management. The Working Group identified the following requirements: (a) a new reference which will enhance the nutritional management of infants; (b) the reference population should reflect current health recommendations because of the frequent use of such reference data as standards; (c) evaluation, in a broad range of settings, of the practical utility of using reference data based on infants for whom the WHO feeding recommendations are being followed; (d) close investigation of the effects of different complementary foods on the growth of infants who are being fed according to the WHO recommendations; (e) criteria for evaluating abnormal growth patterns; (f) research for identifying proxy measures for length; and (g) evaluation of reference data based on other anthropometric measurements, such as skinfold thickness and arm and head circumferences.
Article
Intestinal microbiotas are thought to be the most important source of maturational stimuli to the development of the immune system. However, few studies have focused on the development of T helper (Th) 1 immune response and antibody response to vaccinations in healthy infants, especially in a large cohort. Through this randomized, double-blind control trial, we investigated the effects of Bifidobacterium longum BB536 (BB536) supplementation on intestinal microbiota composition and the immune response in term infants. In total, 300 healthy newborns were recruited, randomized and fed formula either supplemented with BB536 or with no supplementation. Stool samples were analyzed at months 2, 4 and 11. The representative cytokine for Th1 [interferon-γ (IFN-γ)] and Th2 [interleukin-4 (IL-4)] secretion cells were measured using enzyme-linked immunospot assay at 4 and 7 months of age. The antibody response to vaccines was measured at months 7 and 11. A total of 264 infants completed the study. The amount of bifidobacteria and the bifidobacteria/Enterobacteriaceae ratio (B/E) were significantly higher in the BB536 supplementation group at months 2 and 4. The number of IFN-γ secretion cells and the ratio of IFN-γ/IL-4 secretion cells were increased in the BB536 supplementation group at 7 months. Moreover, the higher value of B/E in the early stages seems to be related to the increased Th1 response. No difference was observed between groups in the antibody response after vaccination. BB536 has positive effects on establishing a healthy intestinal microbiota early in life, and it also plays an important role in improving the Th1 immune response.
Article
Background The use of review articles and meta-analysis has become an important part of epidemiological research, mainly for reconciling previously conducted studies that have inconsistent results. Numerous methodologic issues particularly with respect to biases and the use of meta-analysis are still controversial. Methods Four methods summarizing data from epidemiological studies are described. The rationale for meta-analysis and the statistical methods used are outlined. The strengths and limitations of these methods are compared particularly with respect to their ability to investigate heterogeneity between studies and to provide quantitative risk estimation. Results Meta-analyses from published data are in general insufficient to calculate a pooled estimate since published estimates are based on heterogeneous populations, different study designs and mainly different statistical models. More reliable results can be expected if individual data are available for a pooled analysis, although some heterogeneity still remains. Large prospective planned meta-analysis of multicentre studies would be preferable to investigate small risk factors, however this type of meta-analysis is expensive and time-consuming. Conclusion For a full assessment of risk factors with a high prevalence in the general population, pooling of data will become increasingly important. Future research needs to focus on the deficiencies of review methods, in particular, the errors and biases that can be produced when studies are combined that have used different designs, methods and analytic models.