7/25/2019 Differences in Fat Content and Fatty Acid Proportions Among Colostrum

1/8

Original Article

Differences in fat content and fatty acid proportions among colostrum,transitional, and mature milk from women delivering very preterm,preterm, and term infants

Carolina Molt-Puigmart a,b, Ana Isabel Castellote a,b, Xavier Carbonell-Estrany c,M. Carmen Lpez-Sabater a,b,*

a Department of Nutrition and Food Science, Faculty of Pharmacy, University of Barcelona. Avda. Joan XXIII s/n CE-08028, Barcelona, Spainb CIBER of Epidemiology and Public Health, Spainc Hospital Clnic, Institut de Ginecologia, Obstetrcia i Neonatologia, Agrupaci Sanitria Hospital Clnice Hospital Sant Joan de Du. IDIBAPS (Institut dInvestigacions Biomdiques

August Pi i Sunyer), University of Barcelona, Barcelona, Spain

a r t i c l e i n f o

Article history:

Received 22 December 2009

Accepted 18 July 2010

Keywords:

Conjugated linoleic acid

DHA

Human milk

Milk banking

Preterm infants

Creamatocrit

s u m m a r y

Background & aims:Human milk composition changes according to gestational age and stage of lactation,

but infants fed banked human milk often receive pooled milk. We studied the changes in fat content and

fatty acid proportions throughout lactation in very preterm, preterm, and full term milk, and the

differences among gestational age groups.

Methods:Samples from women delivering before 30 (n 10), between 30 and 37 (n 10), and between

38 and 42 (n 23) weeks of gestation were analyzed.

Results: Fat content was higher in very preterm than in preterm and full term samples (p < 0.05).

Medium-chain saturated fatty acids, alpha-linolenic acid, and rumenic acid proportions increased

(p < 0.05) during lactation, while those of most long-chain saturated fatty acids and most long-chain

polyunsaturated fatty acids from the n-3 and n-6 families decreased (p < 0.05). In colostrum and

transitional milk, medium-chain saturated fatty acid proportions were highest in the very preterm group,and decreased with gestational age (p < 0.05).

Conclusions: The differences in fat and fatty acids of human milk obtained at different gestational ages

and stages of lactation may impact preterm infants health. Therefore they could be taken into account

when feeding newborns banked human milk and when designing infant formulas or human milk

fortiers.

2010 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

1. Introduction

Mothers own milk (MOM), with appropriate fortication for

some premature or very-low birth weight (VLBW) infants, is the

best food for newborns.1e3 However, breastfeeding premature or

VLBW infants is still a challenge because of their physiological and

neurological immaturity, because they are alert for very short

periods of time, and because they may present inappropriate suck/

swallow/breathe coordination. In spite of this, many neonatal units

help mothers to start and maintain an appropriate milk produc-tion,4 not only because some premature newborns may still be able

to breastfeed, but also because those that cannot suck directly at

the breast can still be fed milk extracted from their own mothers.5

When, despite the efforts, a woman is unable to feed her child with

her own milk or cannot supply enough milk, or when breastfeeding

is contraindicated, preterm infants can be fed preterm formula or

pasteurized donor human milk (either as a sole diet or as

a complement of MOM). Two points should be emphasized in

regard to these different feeding strategies for prematures:

a) unfortied MOM and specially donor milk may fail to full the

nutritional requirements of preterm newborns2; b) a higher rate of

Abbreviations: VLBW, very-low birth weight; FT, full term; PT, preterm; VPT,

very preterm; SFAs, saturated fatty acids; MCSFAs, medium-chain saturated fatty

acids; LCSFAs, long-chain saturated fatty acids; MUFAs, monounsaturated fatty

acids; PUFAs, polyunsaturated fatty acids; LCPUFAs, long-chain polyunsaturated

fatty acids; LA, linoleic acid; CLA, conjugated linoleic acid; AA, arachidonic acid;

ALA, alpha-linolenic acid; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid.

* Corresponding author. Department of Nutrition and Food Science, Faculty of

Pharmacy, University of Barcelona. Avda. Joan XXIII s/n CE-08028, Barcelona, Spain.

Tel.: 34 93 402 45 12; fax: 34 93 403 59 31.

E-mail addresses: carolinamolto@ub.edu (C. Molt-Puigmart), aicastellote@ub.

edu (A.I. Castellote), XCARBO@clinic.ub.es (X. Carbonell-Estrany), mclopez@ub.

edu,mclopez@ub.edu(M.C. Lpez-Sabater).

Contents lists available at ScienceDirect

Clinical Nutrition

j o u r n a l h o m e p a g e : h t t p : / / w w w . e l se v i e r . c o m / l o c a t e / cl n u

0261-5614/$ e see front matter 2010 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

doi:10.1016/j.clnu.2010.07.013

Clinical Nutrition 30 (2011) 116e123

Downloaded from ClinicalKey.com at Universitas Andalas June 30, 2016.For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

mailto:carolinamolto@ub.edumailto:aicastellote@ub.edumailto:aicastellote@ub.edumailto:XCARBO@clinic.ub.esmailto:mclopez@ub.edumailto:mclopez@ub.edumailto:mclopez@ub.edumailto:mclopez@ub.eduhttp://www.sciencedirect.com/science/journal/02615614http://www.elsevier.com/locate/clnuhttp://dx.doi.org/10.1016/j.clnu.2010.07.013http://dx.doi.org/10.1016/j.clnu.2010.07.013http://dx.doi.org/10.1016/j.clnu.2010.07.013http://dx.doi.org/10.1016/j.clnu.2010.07.013http://dx.doi.org/10.1016/j.clnu.2010.07.013http://dx.doi.org/10.1016/j.clnu.2010.07.013http://www.elsevier.com/locate/clnuhttp://www.sciencedirect.com/science/journal/02615614mailto:mclopez@ub.edumailto:mclopez@ub.edumailto:mclopez@ub.edumailto:XCARBO@clinic.ub.esmailto:aicastellote@ub.edumailto:aicastellote@ub.edumailto:carolinamolto@ub.edu

7/25/2019 Differences in Fat Content and Fatty Acid Proportions Among Colostrum

2/8

growth gain has been observed when feeding prematures preterm

formulas instead of unfortied human milk but, in the former case,

they become more prone to suffer from necrotizing enterocolitis

(NEC).6 Therefore, the use of MOM or donor milk with human milk

fortiers could be seen as the currently preferable option. However,

partly because of the high variability in milk composition, there is

no consensus in regard to what the optimal composition and

quantity of these fortiers is. Some neonatal units now opt to apply

individual fortication, in which fortication is done in base of the

specic composition (of macronutrients) of each milk sample or on

the metabolic responses of the infant during feeding. Individual

fortication seems to be a good strategy to follow, but it entails

obvious logistic difculties and it may not be an achievable goal in

milk banks with scarce resources; in addition, some studies warn

about the fact that fortication is not risk-free. In this context, we

consider that a better knowledge on the intrinsic variability of

human milk composition will help neonatologists, human milk

banks, and manufacturers of infant formulas and human milk

fortiers to improve existing feeding strategies, design alternative

ones, and optimize the use of fortiers.

Fat is known to be the most variable macronutrient in milk.

Apart from its role as energy supplier, it is also source of essential

fatty acids and vehicle for fat-soluble vitamins. Many studies havedescribed the changes in fatty acid composition from colostrum to

mature milk.7e26 However, information on whether these changes

are also found in milk from mothers delivering before the 30th

week of gestation, and on whether very preterm (VPT) milk fatty

acid composition differs from that of preterm (PT) and full term (FT)

milk is scarce. Our aim was, therefore, to study the differences in

human milk fatty acid composition of colostrum, transitional, and

mature milk of Spanish women with VPT, PT, and FT deliveries.

Differences in fat content were also assessed.

2. Participants and methods

2.1. Participants and timing for milk collection

Between June 2006 and May 2008, 50 women living in the

Barcelona metropolitan area who delivered at the Hospital Clnic

from Barcelona participated in the study. Women had to provide

one sample of colostrum (2e4 days post-partum), one of transi-

tional milk (8e12 days post-partum), and one of mature milk

(28e32 days post-partum). Exclusion criteria were: illness of the

mother (including metabolic disorders such as diabetes or gesta-

tional diabetes), eclampsia, HIV infection, drug abuse, treatment

with psychoactive drugs or narcotics, and women delivering small

for gestational age babies. Infants weight at birth was recorded.

Milk samples were classied according to the gestational age as

VPT (less than 30 weeks of gestation), PT (30e37 weeks of gesta-

tion), and FT (38e42 weeks of gestation). Of these 50 women, 7

abandoned the study for different reasons. A total of 43 womensuccessfully provided milk samples at the three time points. Of this

total, 23 women delivered FT infants,10 delivered PT infants, and10

delivered VPT infants. All procedures were approved by the

hospitals ethical committee, and written informed consent was

obtained from all women.

2.2. Breast milk collection

Colostrum was collected at the hospital by an experienced nurse

and one of the authors (CMP). For the collection of transitional and

mature milk, women were visited at home by the same trained

researcher. Milk was collected between 8:00 and 12:00 a.m., at the

end of the feeding, in sterile polypropylene tubes by mechanical

expression of either or both breasts with a breast pump (Ameda,

Zug, Switzerland). Milk was transported to the laboratory in

iceboxes in less than 2 h after collection. Several aliquots were

made from each milk sample, and they were stored at 80 C until

analyzed.

2.3. Creamatocrit analysis and fat estimation

Creamatocrit was determined as described by Lucas et al.

27

withslight modications. In short, milk was drawn by capillarity into

glass capillary tubes, which were then sealed at one end with clay

and centrifuged for 15 min at 12000g(10,921 rpm) and 25 C. The

cream layer and length of total milk column were read under

a magnifying glass with a vernier calliper within 30 min of

centrifugation to prevent the cream column to unpack. The

creamatocrit was calculated as the ratio between the cream layer

and the length of the total milk column and expressed as

percentage. Each samplewas analyzed in triplicate. Milk fat content

was calculated from creamatocrit values through the formula

proposed in the same study by Lucas et al. 27

2.4. Fatty acid analysis

Fatty acid methyl esters were prepared with sodium methylate

and methanolic boron triuoride and extracted into hexane

following the method developed by Molt-Puigmart et al.28

Subsequently, they were separated and quantied by fast-gas

chromatography with ame ionization detection according to the

same method. Each sample was analyzed in duplicate.

2.5. Statistical analysis

First, a descriptive analysis of clinical and biological data was

performed. According to the KolmogoroveSmirnov test, maternal

and infants clinical characteristics were normally distributed

(except for parity and number of babies in the delivery), as well as

creamatocrit values, fat concentration, and most of the fatty acids.Normal distribution was assumed for the statistical analysesof fatty

acids, while validity of the obtained results was checked by

repeating the statistical analyses after performing log10 trans-

formation of those following skewed distributions. Creamatocrit

values, fat concentration, fatty acid proportions and maternal

characteristics were expressed as mean standard deviation.

Differences in maternal and infantsclinical data were assessed by

a one-way ANOVA and by the KruskaleWallis non-parametrical

test for normally and not normally distributed variables, respec-

tively. For each of the gestational age groups, a repeated-measures

analysis of variance test (GLM) was used to assess differences in

fatty acid proportions, creamatocrit values, and fat concentrations

with time (stage of lactation). To compare gestational age groups at

each stage of lactation, an ANOVA (one-way) analysis was used.Differences between gestational age groups were subsequently

identied by using the Bonferroni post-hoc test. Differences asso-

ciated withp values lower than 0.05 were considered to be statis-

tically signicant. Pearson correlation coefcients were used to

study correlations between fatty acids. The SPSS statistical software

(Version 15, SPSS Inc, Chicago, Illinois, USA) was used for all data

analyses.

3. Results

Mean gestational ages were 26.74 (1.12), 33.49 (1.86) and

40.50 (1.11) weeks for women delivering VPT, PT, and FT infants,

respectively. Maternal age ranged from 22 to 42 years old. Other

clinical data of the participants are shown in Table 1.

C. Molt-Puigmart et al. / Clinical Nutrition 30 (2011) 116e123 117

Downloaded from ClinicalKey.com at Universitas Andalas June 30, 2016.For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

7/25/2019 Differences in Fat Content and Fatty Acid Proportions Among Colostrum

3/8

3.1. Creamatocrit and fat content

Differences in creamatocrit values among stages of lactation and

gestational age groups are shown inFig. 1.Values obtained for VPT,PT, and FT milk tended to increase from colostrum to transitional

milk, reaching statistical signicance in the VPT and PT groups

(p 0.031 and 0.007, respectively). Values for mature milk were

found to be between those of colostrum and transitional milk. VPT

milk creamatocrit values at each stage of lactation were higher than

those of FT milk at the same stage of lactation (p < 0.05). The same

was true when comparing VPT with PTsamples, but it did not reach

statistical signicance in the case of transitional milk (p 0.107).

There was no statistically signicant difference between creama-

tocrit values of PT and FT milk. Mean fat content (SD) in colos-

trum, transitional, and mature milk was 4.05 1.62, 4.76 1.62,

and 4.67 1.19 g/100 mL, respectively, in VPT samples; 2.58 1.88,

3.75 1.24, and 2.98 1.75, respectively, in PT samples; and

2.60 1.48, 3.11 1.53, and 3.06 1.29, respectively, in FT samples.Statistically signicant differences in the fat content among groups

were the same as those observed for the creamatocrit.

3.2. Differences in fatty acid proportions according to stage of

lactation

Several differences were found between colostrum, transitional,and mature milk saturated fatty acids (SFAs), monounsaturated

fatty acids (MUFAs), and polyunsaturated fatty acids (PUFAs)

proportions (Tables 2e4, respectively). The observed differences

were maintained even after log-transforming fatty acids with

skewed distributions.

As shown inTable 2, the sum of SFAs increased in a signicant

way from colostrum to transitional PT and FT milk, and remained

stable in mature milk. Instead, VPT SFAs proportions did not

signicantly vary with the progression of lactation. When looking

at the indivual SFAs, we observed that C10:0 and C12:0 increased

during lactation in the three gestational age groups with

a maximum in transitional milk, SFAs with 20 or more carbon

atoms decreasedor tended to decrease during the same period, and

SFAs with 14e

18 carbon atoms had a variable behaviour. Differ-ences seen in the sum of medium-chain SFAs (MCSFAs,

C10:0 C12:0) reected the ones just described for the individual

fatty acids (Fig. 2a). When summing up long-chain SFAs propor-

tions (LCSFAs, C14:0), we observed a decrease during lactation in

VPT and PT, but not in FT samples (Fig. 2b).

Despite some oscillations, the sum of MUFAs proportions

remained quite constant during lactation (Table 3). However, more

differences were found when looking at the MUFAs individually. Of

note, percentages of MUFAs with more than 18 carbons (C20:1 n-9,

C22:1 n-9, and C24:1 n-9) considerably decreased, in a signicant

way, from colostrum to mature milk. Instead, oleic acid (C18:1 n-9)

proportions slightly increased in the VPT and PT groups during the

same period.

Regarding PUFAs (Table 4) from the n-6 family,transisomers oflinoleic acid (t9t12-andc9t12,t9c12-C18:2) and linoleic acid (C18:2

n-6, LA) proportions remained quite stable from colostrum to

mature milk, C18:3 n-6 and c9t11-CLA proportions increased, and

those of PUFAs with more than 18 carbons decreased. In the n-3

family, alpha-linolenic acid (C18:3 n-3, ALA) proportions increased,

those of eicosapentaenoic acid (C20:5 n-3, EPA) did not vary, and

those of C22:5 n-3 and docosahexaenoic acid (C22:6 n-3, DHA)

decreased with progression of lactation. Fig. 3a and b shows the

evolution of arachidonic acid (C20:4 n-6, AA) and DHA throughout

lactation. When n-3 and n-6 fatty acids were summed up, the

percentages of PUFAs from both families (PUFAs n-6 and PUFAs n-3)

decreasedduring lactationin PTand FT milk, as also didthose of the

n-6 and n-3 long-chain PUFAs (LCPUFAs n-6 and LCPUFAs n-3) in all

groups.

Table 1

Characteristics of women and infants included in the study.

Characteristicsc Very preterm

(n 10)

Preterm

(n 10)

Full term (n 23) p valued

Maternal age (years) 33.10 3.56 32.80 5.33 30.83 4.88 0.327a

M atern al wei ght pr evious p regnan cy (kg) 6 5.9 0 6.85 64.15 11.76 55.91 6.48 0.004a

Maternal height (m) 1.67 0.05 1.60 0.06 1.60 0.07 0.025a

BMI previous pregnancy (kg/m2) 23.75 1.97 25.09 5.25 21.97 2.49 0.050a

Weight gain during pregnancy 8.4 3.75 13.55 3.48 11.92 5.04 0.050

a

Parity 1.40 0.67 1.60 0.49 1.61 0.86 0.607b

Number of babies in the delivery 1.10 0.30 1.20 0.40 1.00 0.00 0.103b

Number of previous pregnancies 1.33 0.96 1.25 0.44 1.75 1.73 0.795a

Time from last pregnancy (months) 25.25 21.64 68.50 53.79 11.05 8.30 0.354a

Weight of the newborns at birth (kg) 0.97 0.25 2.21 0.58 3.67 0.71 0.000a

a Statistical differences were assessed by one-way ANOVA.b Statistical differences were assessed by the KruskaleWallis non-parametrical test.c Mean SD.d Signicant differences (p < 0.05) are indicated in bold.

Fig.1. Creamatocrit values (%) in colostrum (black bars), transitional (white bars), and

mature milk (striped bars), from women delivering very preterm (VPT), preterm (PT),

and full term (FT) infants. Lower case letters indicate differences among colostrum,

transitional, and mature milk in each gestational age group (same letters indicate

signicant differences, according to GLM and LSD post-hoc test). Symbols indicate

differences among gestational age groups at each stage of lactation (same symbols

indicate signicant differences, according to ANOVA and Bonferroni post-hoc test).

Signi

cance:p