European Journal of Clinical Nutrition (1999) 53, 872±879
ß 1999 Stockton Press. All rights reserved 0954±3007/99 $15.00
http://www.stockton-press.co.uk/ejcn
Early nutrition, essential fatty acid status and visual acuity of
term infants at 7 months of age
EC Bakker1,2*, AC van Houwelingen1 and G Hornstra1
1
Department of Human Biology, Universiteit Maastricht, The Netherlands; and 2Present af®liation: University Hospital Maastricht,
Department of Pediatrics and Neonatology, Maastricht, The Netherlands
Objective: In term infants the relationship between visual acuity and dietary fatty acid composition is not
consistent, possibly due to confounders, which were mostly neglected in the studies concerned. In the current
study, therefore, the in¯uence of the essential fatty acid status and potential confounders on the visual acuity was
investigated.
Design: The essential fatty acid status was determined at 7 months of age in red blood cell and plasma
phospholipids of breastfed and formula-fed infants, born at term. Visual acuity was measured with Teller Acuity
Cards. Information about potential confounding factors was obtained during an interview and with a retrospective questionnaire.
Results: This study, like others, showed that the concentrations of docosahexaenoic acid (DHA, 22 : 6n-3) are
lower in plasma and red blood cell phospolipids of formula-fed infants compared to that of breastfed infants.
However, no differences in visual acuity could be found between the two groups. Moreover, no signi®cant
relationship was found between the amounts of docosahexaenoic acid in plasma and red blood cell phospholipids
and the visual acuity. Although dummy (paci®er) use showed a signi®cant positive correlation with visual acuity,
it did not in¯uence the relationship between the essential fatty acids in the infant diet and visual acuity. There
was also no confounding in¯uence of smoking habits and alcohol use during pregnancy, socioeconomic
background and other potential confounders.
Conclusions: At 7 months of age no in¯uence of fatty acid status, infant diet or potential confounders on visual
acuity was found.
Descriptors: essential fatty acids; docosahexaenoic acid; visual acuity; blood plasma; red blood cells; Teller
Acuity Cards
Introduction
The essential fatty acids (EFA) linoleic acid (18 : 2n-6) and
alpha-linolenic acid (18 : 3n-3) cannot be synthesized by
the human body (Burr & Burr, 1929). Dietary supply of
these fatty acids is important, because they are precursors
for several other fatty acids with important functions (Uauy
et al, 1989). The biologically active fatty acids, the longchain polyenes (LCP), can be synthesized from the parent
fatty acids (linoleic acid and alpha-linolenic acid) by
alternate desaturation and elongation. Some of the derived
fatty acids, particularly the LCP arachidonic acid (AA,
20 : 4n-6) and docosahexaenoic acid (DHA, 22 : 6n-3), are
found in high proportions in the central nervous system
(Anderson et al, 1992: Innis, 1991), and play an important
role in its development (Crawford, 1993). Therefore, these
LCP can also be considered essential. Low dietary intake of
*Correspondence: E C Bakker, Universiteit Maastricht, Department of
Human Biology, P O Box 616, 6200 MD Maastricht, The Netherlands.
Tel: (+31) 433 881 309; Fax: (+31) 433 670 976;
E-mail: [email protected].
Guarantor: E C Bakker.
Contributors: Adriana van Houwelingen contributed to the study design,
assisted with the ®eld work, the analyses of data and the interpretation of
the results and reviewed the paper. Gerard Hornstra supervised the project,
contributed to the study design, assisted with the analyses of data and the
interpretation of the results and reviewed the paper. Esther Bakker wrote
the protocol, carried out the ®eld work and the data analyses and wrote the
paper.
Received 22 February 1999; 25 May 1999; accepted 1 June 1999
EFA can potentially disturb the growth and development of
the central nervous system (including the retina), resulting
in functional impairment (Innis, 1991; Uauy et al, 1992).
Because the nervous system of a child mainly develops
during late pregnancy and in the ®rst postnatal year, an
adequate supply of EFA to the child during this period is
very important. The fetus is supplied with these fatty acids
by the mother via transport through the placenta. Neonates
depend on breastfeeding or arti®cial formula for their
essential fatty acid supply (Innis, 1991). Breastmilk generally contains suf®cient amounts of linoleic acid, alphalinolenic acid and LCP (including arachidonic acid and
docosahexaenoic acid) to meet the requirements of the
child (Innis, 1991). However, standard arti®cial formulas
do not contain n-3 LCP and n-6 LCP (Innis, 1991). Therefore, infants fed exclusively with standard formula are
dependent on their own synthesis of the LCP necessary
for growth and development. It is not known, however,
whether neonates possess adequate fatty acid desaturating
capacity to meet their LCP requirements.
In several studies it has been found that the functional
development of cerebral cortex and retina is slower in
formula-fed preterm infants in comparison with breastfed
premature infants (Birch et al, 1992, 1993; Carlson et al,
1993; Uauy et al, 1992; Leaf et al, 1996). In term infants,
however, this relation is not consistent. Birch et al (1992,
1993, 1998) found that term breastfed infants had signi®cantly better visual acuity scores than formula-fed infants at
6, 17 and 52 weeks after birth, measured as Visual Evoked
Essential fatty acids and visual acuity in term infants
EC Bakker et al
Potentials (VEP). In the ®rst two publications (1992, 1993)
these authors also found better preferential looking scores
in term breastfed infants as compared with term formulafed infants. Makrides et al (1993) observed a signi®cant
better VEP-acuity and higher amounts of DHA in term
breastfed infants as compared with term formula-fed
infants at 22 weeks after birth. They also observed a
positive correlation between the DHA content of red
blood cells and the VEP-acuity (Makrides et al, 1993). At
an age of 16 and 30 weeks, infants who received human
milk or LCP-enriched formula also had better VEP-acuity
scores than infants fed standard formula (Makrides et al,
1995). Furthermore, Carlson et al (1996) found that term
infants fed either human milk or DHA and AA supplemented formula had better Teller Acuity Card scores as compared with infants fed standard formula, at an age of 2
months. Beyond the age of 4 months, however, there was
no relation anymore between type of diet and visual acuity
(Carlson et al, 1996). Recently, Jùrgensen et al (1998)
found differences in swept steady-state VEP scores
between breastfed term infants and infants fed standard
formula, at 4 months of age.
Three other studies were unable to demonstrate signi®cant relations between n-3 fatty acids in the diet and visual
acuity in term infants. Auestad et al (1997) found no
differences between term infants receiving human milk,
standard formula, DHA supplemented formula and formula
supplemented with both DHA and AA at 2, 4, 6, 9 and 12
months of age. Innis et al (1994, 1996) observed no
differences in visual acuity between term infants fed
standard formula and infants fed human milk at 14 days,
3 and 9 months, in spite of lower DHA contents in plasma
and red blood cells in the formula group at 14 days and 3
months. Using the forced choice preferential looking protocol, Birch et al (1998) did not ®nd any effect of diet on
visual acuity at 6, 17, 26 and 52 weeks of age.
So, the results of studies on visual acuity in term infants
in relation to their diet are not consistent, possibly due to
the in¯uence of confounding factors like smoking habits
and alcohol use during pregnancy, and socio-economic
background. These factors may have an in¯uence on the
development of the central nervous system, as re¯ected by
the visual acuity. The studies mentioned above reported
few of these factors. Therefore, the current cross sectional
study investigates the in¯uence of the essential fatty acid
status and some potentially confounding factors on the
visual acuity at 7 months of age, in children born at term.
Materials and methods
Study population
The study population consisted of 74 healthy, singleton,
term (gestational age 37 ± 42 weeks) infants of 7 months
(range 6 ± 8 months), recruited at child health centres in the
southern part of the Netherlands. Infants with neurological
dysfunction or motor problems (according to the child
health centre physicians) were excluded. Parents of eligible
infants were provided a brief information letter. Written
informed consent was obtained from the accompanying
parent(s) of each participant. Of these infants, 48 received
human milk, whereas 26 were fed exclusively with standard
arti®cial formula-feeding, which means they never received
human milk. All infants in the human milk group were
breastfed directly, except for 2, who received the human
milk from bottles, for practical reasons. The mean duration
of breastfeeding was 16 weeks, with a range between 1 and
35 weeks. Only 12 of the infants were fully fed with human
milk until the visual acuity measurement (mean duration of
breastfeeding 31 weeks). The standard formulas used contained 11.2 ± 13.5 g linoleic acid and 1.35 ± 2.2 g alphalinolenic acid per 100 g total fatty acids, but were devoid
of LCP. Within this cross sectional study it was not
possible to take human milk samples, but from the study
of Huisman et al (1996) it is known that mature human
milk of Dutch women contains 12.8 ± 14.4 mol=100 mol
linoleic acid, 1.1 ± 1.2 mol=100 mol alpha-linolenic acid,
and 1.0 ± 1.25 mol=100 mol n-6 LCP and 0.40 mol=
100 mol n-3 LCP. From 4 months onwards, all infants
received some solid food as well, following the advice of
the child health centres. Capillary blood samples were
taken from 47 infants, of which 30 were breastfed and 17
were formula-fed. Parents from 27 infants did not give
permission to take a blood sample from their child, and 4
samples were too small to reliably determine fatty acid
pro®les in red blood cell phopholipids.
The clinical characteristics of the study population are
given in Table 1.
The study was approved by the Ethics Committee of the
University Hospital Maastricht=Universiteit Maastricht.
Sample size
Sample sizes were calculated according to standard methods (Kirkwood, 1988), based on expected group means of
0.37( 0.06) logMAR and 0.44( 0.07) logMAR (data
from Jùrgensen et al, 1998). To achieve a power of 0.90
at an overall alpha of 0.05, the sample size had to be at least
38 infants, preferably 19 children in each group. Ultimately, the human milk group consisted of 48 infants, the
formula group of 26 infants. Fatty acid pro®les of 30
breastfed and 17 formula-fed infants could be determined.
Analytical procedures and measurements
Fatty acid pro®les were determined in plasma and red blood
cell (RBC) phopholipids using the following methods.
About 1 mL capillary blood was collected in EDTA containing tubes. Within 8h after blood collection, plasma was
separated from the RBC by centrifugation (rpm ˆ 3000,
g ˆ 503, t ˆ 10min, 4 C) and collected in plastic tubes,
which were tightly closed under nitrogen and stored at
7 50 C until fatty acid analysis. The remaining RBC were
washed twice with EDTA-containing saline (Na2EDTA2H2O 28,64 g, NaC1 7 g, H2O 1000 mL), centrifuged
(rpm ˆ 1500, g ˆ 126, t ˆ 15 min, 4 C), and collected in
plastic tubes, which were tightly closed under a stream of
nitrogen and stored at 7 50 C until fatty acid analysis.
Within one week after sampling, total lipids were extracted
from the RBC as described by Bligh & Dyer (1959).
Plasma lipids were extracted as described by Folch et al
(1957).
L-a-dinonadecanoyl lecithin (PC19:0) was used as an
internal standard to calculate the quantative fatty acid
amounts. The phospholipid (PL) fraction was separated
from the total lipid extract using aminopropyl bonded
silica columns (500 mg) (Kaluzny et al, 1985). Heptadecenoic acid (17:1) was added to the samples to check carryover of free fatty acids during the phospholipid separation
procedure. The PL fraction was hydrolyzed and the
resulting fatty acids methylated with boron tri¯uoride
in methanol (Morissen & Smith, 1964). The fatty acid
composition of the PL was then determined by gas liquid
873
Essential fatty acids and visual acuity in term infants
EC Bakker et al
874
Table 1 Clinical characteristics of the study population (mean sem) and potential confounders in the relationship between essential fatty acid status and
visual acuity.
Variable
Breastfed infants (n ˆ 48)
Formula-fed infants (n ˆ 26)
t ± Test*
Duration of breastfeeding (weeks)
Age infant (weeks)
Age mother (years)
Gestational age (weeks)
Parity
Birth weight (g)
Birth length (cm)
Current weight (g)
Current length (cm)
15.9 1.45 (range 1 ± 35)
30.2 0.33
31.0 0.51
39.8 0.22
1.7 0.10
3442 71.8
50.9 0.31
7864 138.6
68.0 0.35
Ð
30.2 0.39
29.7 0.86
39.6 0.32
1.6 0.16
3267 93.4
50.3 0.50
7885 167.4
65.4 2.72
Sex (boys=girls,%)
Smoking (non ± smokers=smokers,%)
Alcohol consumption during pregnancy (non ± users=users,%)
Dummy use (non ± users=users,%)
Thumb use (non ± users=users,%)
Mothers educational level (low=middle=high,%)
Fathers job (low=middle=high,%)
Parents socio ± economic status (low=middle=high,%)
50=50
87=13
82=18
46=54
56=44
5=49=46
39=33=28
3=56=41
< 0.0001
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
w2
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
t ± Test
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
Cigarettes (a day) before pregnancy
Cigarettes (a day) during the ®rst half of pregnancy
Cigarettes (a day) during the last half of pregnancy
Cigarettes (a day) after delivery
Smoking habits of environment (total cigarettes=day) during pregnancy
Alcohol consumption (glasses=week) during ®rst half of pregnancy
Alcohol consumption (glasses=week) during last half of pregnancy
Alcohol consumption (glasses=week) during lactation
5.0 1.19
1.4 0.69
0.9 0.58
2.2 0.99
4.4 1.85
0.2 0.08
0.2 0.08
0.3 0.15
42=58
77=23
83=17
28=72
65=35
6=76=18
18=47=35
6=50=44
6.7 2.21
2.8 1.52
2.1 1.03
3.9 1.64
8.9 6.36
0.3 0.2
0.2 0.12
Ð
*N.S. : P > 0.05.
chromatography using a polar (BPX70, 50 m with 0.22 mm
ID and 0.25 mm ®lm thickness) and a non-polar column
(BP-1, 50m with 0.22 mm ID and 0.1 mm ®lm thickness,
both SGE, Bester BV, Amstelveen, The Netherlands), with
He (head pressure 370 kPa) as carrier gas. The injection
temperature was 250 C and the detection temperature
300 C. The starting temperature of the column was
160 C. After 4 min, the temperature increased up to
200 C with a rate of 6 C=min and ®nally to 270 C with
a rate of 7 C=min. The split ratio was 1 : 40.
Measurement of visual acuity
The binocular visual acuity of all 74 infants was measured
with the Teller Acuity Card method following the acuity
card procedure (Teller, 1990). This method is based on the
observer's interpretation of the infant's looking behavior.
In this method a card with a black-and-white grating of a
particular spatial frequency (stripe width) is presented to
the infant. If the infant is able to perceive this stripe pattern,
it will ®x the attention on this part of the card. If not, the
child will lose attention. The ®nest stripe pattern that can be
distinguished repeatedly is taken as an acuity estimate. All
infants were measured by the same observer (ECB), who
was blind for the feeding group the infant belonged to. A
conversion table is used to determine acuity
(cycles=degree) from the ®nest grating that can be distinguished (cycles=centimeter) as a ®xed distance of 38cm
(Teller et al, 1986, Teller, 1990). Because acuity scores are
on a logarithmic rather than a linear scale, means and
standard deviations of acuity scores cannot be determined
by simple linear addition and division (Teller, 1990).
Therefore, individual acuities were transformed to log10
cycles=degree before the analyses were conducted. After
calculating the means and SD for each group, the means
were transformed back to cycles=degree and SD expressed
in octaves (SD of log acuity scores=0.301), according to the
standard procedure (Teller, 1990).
Measurement of potentially confounding factors
The parent who accompanied the infant was interviewed
immediately after completion of the visual acuity measurement. Moreover, 62 parents completed and returned a
retrospective questionnaire, 41 of the human milk group
and 21 of the formula group. With these methods we
measured the following potential confounding factors:
duration of breastfeeding, ages of mother and infant,
gestational age, parity, birth weight and length, current
weight and length of the infant, sex, smoking and drinking
habits during pregnancy and lactation, dummy (paci®er)
and thumb use of the infant and socioeconomic status of the
family (based on maternal educational level and father's
job). Gestational age has been shown to be related to the
essential fatty acid status at birth (Houwelingen et al,
1996). Ethanol exposure causes a decrease in DHA in
brains and retinas of felines (Pawlosky & Salem, 1995).
Furthermore, smoking also decreases LCP, including DHA,
in tissues of rhesus monkeys (Brown et al, 1998). If these
factors have an in¯uence in humans as well, drinking and
smoking habits during pregnancy and lactation may disturb
the essential fatty acid metabolism in the developing
infants. Dummy use and socioeconomic status of the
family are measured because they are thought to have an
in¯uence on the functioning of the central nervous system
(Gale & Martyn, 1996). In addition, maternal age, parity,
birth weight and length, current weight and length, sex and
duration of breastfeeding are measured, because these
factors may have an in¯uence on the essential fatty acid
status as well.
Essential fatty acids and visual acuity in term infants
EC Bakker et al
Data analysis
Fatty acid data are presented as mean sem in relative
amounts (%wt=wt). Absolute fatty acid data (mg=L) are not
given, because the total amount of fatty acids did not differ
between the breastfeeding group and the formula group. In
total, 44 fatty acids were identi®ed. However, for the sake
of clarity, only 9 separate fatty acids and 11 fatty acid
combinations are reported (see TableP2a). The following
P
indexes were calculated: EFA-index ( (n-3 ‡ n-6)= (n9 ‡ n-7) (Hornstra et al, 1992), EFA-de®ciency index
(20 : 3n-9=20 : 4n-6) (Holman, 1960), DHADI (docosahexaenoic acid de®ciency index : 22 : 5n-6=22 : 4n-6) (Holman,
1986; Neuringer et al, 1986) and DHASI (docosahexaenoic
acid suf®ciency index : 22 : 6n-3=22 : 5n-6) (Hoffman &
Uauy, 1992).
To investigate the relation between visual acuity and the
EFA-status of the infants, linear regression was used with
Log10 visual acuity (cycle=deg) as dependent variable and
the fatty acids of interest (%wt=wt) as independent variables. The in¯uence of potential confounding factors on
visual acuity was studied by linear regression. The in¯uence of these factors on the relation between visual acuity
and EFA status was studied by introducing these variables
separately as continuous covariables in a multiple linear
regression analysis. The in¯uence of discrete variables on
visual acuity was analysed by Student t-tests and chi-square
tests. To study the in¯uence of diet (formula or human
milk) on visual acuity and on the status of several fatty
acids, Student's t-tests were used in which breastfed infants
were compared with formula-fed infants. In all statistical
analyses a signi®cancy level of P < 0.05 was taken, unless
mentioned otherwise.
875
Results
Clinical characteristics (mean sem) of the study population are given in Table 1. Between the breastfed and the
formula-fed infants no signi®cant differences were found
for age, gestational age, birth number, birth weight, birth
length and current weight and length. The mean duration of
breastfeeding was almost 16 weeks ( 1.5). In Table 1, the
values of some potential confounders are also given. However, no signi®cant differences were found between the
human milk group and the formula group for any of these
variables.
The fatty acid composition (mean sem, %wt=wt) of
red blood cell (RBC) phospholipids (PL) is given in Table
2a for the breastfed infants, the formula-fed infants and the
subgroup of infants who were fully breastfed until the
measurement at 31 weeks of age. Table 2b shows the
same values for plasma PL.
P
In the breastfed group the DHA and
n-3 LCP
amounts in RBC and plasma PL
P were signi®cantly higher
than in the group fed formula.
n-3 differed
P signi®cantly
only in RBC. The levels of 20 : 3n-9 and
n-9 LCP in the
RBC were signi®cantly lower in the breastfed infants as
compared with the formula-fed infants. The DHA status in
both RBC and plasma PL, as re¯ected by the values for
DHADI and DHASI, was also signi®cantly higher in the
human milk group as compared with the formula group.
The amounts of the other fatty acids were comparable
between both groups.
In both plasma and RBC the subgroup
P of fully breastfed
infants
higher levels P
of 20 : 4n-6,
n-6 LCP, 22 : 6nP had P
3,
n-3,
n-3 LCP,
pufa, EFA index and a higher
Table 2a Fatty acid composition (% of total FA; mean sem) of red blood cell phospholipids of breastfed infants (mean duration of
breastfeeding 16 and 31 weeks respectively) compared to formula-fed infants
Fatty acid*
18 : 2n-6
20 : 2n-6
22 : 4n-6
22
P: 5n-6
P n-6
n-6 LCP
18 : 3n-3
20 : 5n-3
22 : 5n-3
22
P: 6n-3
P n-3
n-3 LCP
20
P: 3n-9
P n-9 LCP
P
n-9
P n-7 ‡
P sat
P mufa
P pufa
trans
Total fa (mg=L)
EFA index
EFA def.index
DHADI
DHASI
All breastfed infants (n ˆ 29)
Fully breastfed infants (n ˆ 6)
Formula-fed infants (n ˆ 14)
11.2 0.17
10.2 0.19
2.6 0.06
0.32 0.02
26.5 0.29
15.0 0.25
0.11 0.01
0.29 0.02
1.5 0.04
2.2 0.14 (d)
4.2 0.16 (e)
4.0 0.17 (e)
0.15 0.01(e)
0.30 0.02 (e)
22.9 0.56
45.8 0.33
22.6 0.55
31.0 0.31
0.12 0.01
1000.2 43.69
1.4 0.06
0.016 0.001
0.12 0.004 (e)
7.4 0.64 (e)
10.8 0.19
11.8 0.38 (c)
2.6 0.10
0.29 0.02
27.8 0.60
16.7 0.49 (e)
0.07 0.03
0.27 0.04
1.6 0.01
3.5 0.19 (a)
5.5 0.20 (a)
5.4 0.19 (a)
0.10 0.01 (b)
0.19 0.03 (a)
17.9 0.81 (a)
48.1 0.62 (b)
17.7 0.80 (a)
33.5 0.55 (a)
0.15 0.04
1057.1 142.35
1.9 0.11 (a)
0.010 0.00 (b)
0.11 0.01
12.5 1.65 (a)
11.0 0.24
10.1 0.28
2.6 0.14
0.37 0.04
26.3 0.42
15.0 0.48
0.12 0.02
0.32 0.03
1.5 0.07
1.6 0.03
3.6 0.12
3.5 0.11
0.18 0.01
0.37 0.02
23.7 0.52
45.4 0.28
23.3 0.53
30.3 0.35
0.11 0.02
878.2 56.96
1.3 0.04
0.018 0.001
0.14 0.01
5.0 0.51
P
*18 : 2n-6 ˆ linoleic acid, 20 : 4n-6 ˆ arachidonic acid, 22 : 4n-6 ˆ adrenic acid, 22 : 5n-6 ˆ Osbond acid,
n-6 ˆ sum of all n-6 fatty acids,
P
n-6 LCP ˆ sum of the n-6 LCP : 20 : 3, 20 : 4, 22 : 4, 22P
: 5. 18 : 3n-3 ˆ alpha ± linolenic acid,P20 : 5n-3 ˆ eicosapentaenoic acid, 22 : 5n3 ˆ clupanodonic acid, 22 : 6n-3 ˆ docosahexaenoic
acid,
n-3 ˆ sum of all n-3 fattyP
acids.
n-3 LCP ˆ sum of the n-3 LCP : (20 : 5,
P
P
22 : 5, 22 : 6), 20 : 3n-9 ˆ Mead acid,
n-9P
LCP ˆ sum of the n-9 LCP (18 : 2, 20 : 3),
n-7 ‡P 9 ˆ sum of all n-7 and n-9 fatty acids.
P
sat ˆ
mufa ˆ sum of the
pufa ˆ sum of the polyunsaturated fatty
Psum of the saturated fatty acids,
Pmonounsaturated
P fatty acids,
trans ˆ sum of the trans fatty acids. EFA index ˆ
(n-3 ‡ n-6)= (n-9 ‡ n-7), EFA de®ciency index ˆ 20 : 3n-9=20 : 4n-6,
acids,
DHADI ˆ docosahexaenoic acid de®ciency index : 22 : 5n-6=22 : 4n-6. DHASI ˆ docosahexaenoic acid suf®ciency index : 22 : 6n-3=22 : 5n6. Signi®cantly different from formula-fed infants : a : P < 0.0001; b : P < 0.001; c : P < 0.005; d : P < 0.01; e : P < 0.05.
Essential fatty acids and visual acuity in term infants
EC Bakker et al
876
Table 2b Fatty acid composition (% of total FA; mean sem) of plasma phospholipids of breastfed infants (mean duration of
brestfeeding : 16 and 31 weeks respectively) compared to formula-fed infant
Fatty acid*
18 : 2n-6
20 : 4n-6
22 : 4n-6
22
P: 5n-6
P n-6
n-6 LCP
18 : 3n-3
20 : 5n-3
22 : 5n-3
22
P: 6n-3
P n-3
n-3 LCP
20
P: 3n-9
P n-9 LCP
P
n-9
P n-7 ‡
P sat
P mufa
P pufa
trans
Total fa (mg=L)
EFA index
EFA def.index
DHADI
DHASI
All breastfed infants (n ˆ 30)
Fully breastfed infants (n ˆ 6)
Formula-fed infants (n ˆ 17)
24.0 0.41
6.8 0.30
0.34 0.01
0.24 0.02
34.1 0.29
9.8 0.35
0.14 0.02
0.25 0.03
0.73 0.03
2.2 0.16 (e)
3.4 0.18
3.2 0.18 (e)
0.26 0.01
0.26 0.01
15.5 0.44
46.5 0.18
15.2 0.44
37.7 0.35
0.18 0.03
1121.0 54.28
2.5 0.11
0.040 0.003
0.70 0.03 (e)
9.9 0.87 (d)
23.0 0.81
9.0 0.77 (b)
0.37 0.02
0.22 0.01
35.3 0.47 (e)
11.9 0.76
0.10 0.04
0.21 0.05
0.82 0.07
3.6 0.28 (a)
4.8 0.24 (a)
4.7 0.26 (a)
0.20 0.02 (e)
0.22 0.02 (e)
11.9 0.88 (a)
46.7 0.49
11.7 0.86 (a)
40.4 0.61 (b)
0.33 0.10 (e)
1126.7 108.63
3.5 0.28 (a)
0.023 0.003 (e)
0.60 0.02 (e)
16.8 1.99 (a)
23.7 0.62
6.4 0.27
0.35 0.03
0.30 0.03
33.9 0.44
9.9 0.41
0.15 0.03
0.27 0.04
0.73 0.04
1.6 0.06
2.9 0.10
2.7 0.09
0.30 0.02
0.30 0.02
16.3 0.35
46.4 0.30
16.0 0.35
37.0 0.39
0.14 0.03
1150.0 113.69
2.3 0.06
0.046 0.004
0.84 0.05
6.5 0.65
*See Table 2a for legends.
DHASI than formula-fed
while lower
P infants,P
P levels
P were
found for 20 : 3n-9,
n-9 LCP,
n-7 ‡
n-9,
mufa
P
and the EFA de®ciency index. Only RBC
sat levels
were found to P
be higher in the breastfed group, whereas
differences in
trans fatty acids and DHADI were only
found in plasma.
No difference in visual acuity was observed between the
diet groups. Infants who received human milk (n ˆ 48) had
a mean visual acuity score of 6.5 cycles=degree (SD 0.41
octaves). The visual acuity score of formula-fed infants
(n ˆ 26) was 6.7 cycles=degree (SD : 0.41 octaves). The
subgroup of fully breastfed infants (n ˆ 12) had a mean
score of 6.6 cycles=degree with SD 0.48 octaves. Using
linear regression analysis, no correlation was found
between the duration of breastfeeding (weeks) and visual
acuity (r ˆ 0.035).
Multiple regression analyses were done for the total
group and for the human milk group and the formula group
separately. No signi®cant relationship was found between
visual acuity and the percentage of DHA and AA in red
blood cell (r ˆ 0.10 and 0.01 respectively for the total
group) and plasma-PL (r ˆ 0.15 and 0.17 respectively for
the total group). The other fatty acids did not show a
relationship either. No in¯uence on visual acuity was
found for the maternal smoking habits and alcohol use
during pregnancy and lactation, mother's age, gestational
age, age at visual acuity measurement, birth weight and
length, current weight and length, parity, sex, mother's
educational level, father's job, and parents' socio ± economic status. Only dummy use correlated with visual
acuity. Dummy users showed a better visual acuity
(6.9cycle=deg 0.40 octaves) than non ± dummy users
(5.9 0.42) (P < 0.05), independent of dummy use duration. In spite of this correlation, dummy use did not
interfere with the relationships between visual acuity and
either fatty acids in blood or the infant's diet. These
relationships remained unchanged when dummy use or
other factors were introduced as covariables.
Discussion
The aim of this study was to investigate in term infants the
relationship between visual acuity and dietary fatty acid
composition, and the in¯uence of potential confounders on
this relationship. At 7 months of age, visual acuity of
infants who were breastfed did not differ signi®cantly
from the visual acuity of infants who received formula
feeding, although they had signi®cantly higher amounts of
certain LCP in their plasma and red blood cell phospholipids. Moreover, the fatty acid concentrations in plasma and
red blood cells were not related to visual acuity at this age.
None of the used covariables turned out to be a confounder
of these analyses.
Visual acuity is dependent on the development of the
retina and the visual cortex, which are both rich in DHA.
The accumulation of DHA in these tissues reaches its
maximum in the last trimester of pregnancy and continues
during the ®rst months after birth (Clandinin et al,
1980a,b). In contrast to human milk, most arti®cial formulas do not contain n-6 LCP and n-3 LCP to supply the
infant with AA and DHA. The present study, like other
studies (Innis, 1992; Innis et al, 1997; Jùrgensen et al,
1996; Uauy et al, 1992), showed that the concentrations of
22 : 6n-3 and other essential fatty acids were lower in
plasma and red blood cells of formula-fed infants compared
with those of breastfed infants, whereas fatty acids of the n7 and n-9 family were higher. These differences were more
pronounced in the subgroup of infants who received human
milk until 7 months of age as compared with the total group
of breastfed infants (mean duration
of breastfeeding : 16
P
weeks). Furthermore, plasma
trans fatty acids was twice
as high in fully breastfed infants as compared with formulafed infants.
However, despite differences in essential fatty acid
values in plasma and RBC PL, no difference in visual
acuity could be found between the breastfed and formula-
Essential fatty acids and visual acuity in term infants
EC Bakker et al
fed infants. This suggests that the DHA status in the plasma
and red blood cells at the age of 7 months has no in¯uence
on the visual acuity at this age. In the current study, indeed
no signi®cant relationship was found between the amounts
of DHA in plasma or red blood cell PL and the visual
acuity.
A possible explanation for the absence of a relation
between the amount of DHA in blood and visual acuity
could be that dietary DHA in the ®rst seven months of life
is not necessary for optimal visual development. It is
possible that the DHA status at an earlier age has a greater
in¯uence on visual acuity at 7 months than has the DHA
status at 7 months itself. Possibly even the DHA intake of
mothers during pregnancy has an in¯uence, through DHA
accretion in utero. Most of the studies, including ours, do
not include EFA values at birth.
It is possible that the fatty acid status in blood (plasma
and red blood cells) is not a good indicator for the presence
of these fatty acids in the central nervous tissue. Several
studies (Anderson et al, 1992; Stinson et al, 1991; Wiegand
et al, 1991) showed that the retina can retain and recycle
DHA and other LCP. Consequently, formula-fed term
children may already have accumulated suf®cient amounts
of DHA before the age of 7 months to reach an optimal
status of this fatty acid in their retina tissue, in spite of their
lower blood DHA values at 7 months of age. This is
supported by postmortem examinations of term infants
(age at death ranged from 2 to 48 weeks) as reported by
Makrides et al (1994). They found no differences in retinal
DHA between breastfed and formula-fed infants, in spite of
differences in DHA levels in RBC PL. However, there are
animal studies in which dietary manipulation showed to
have large effects on DHA content of retinal phospholipids
of rhesus monkeys (Neuringer et al, 1986), guinea pigs
(Weisinger et al, 1995), rats (Suh et al, 1996) and felines
(Pawlosky et al, 1997). Unfortunately, in these studies, the
fatty acid values in plasma and RBC were not measured
and no breastfeeding group was included to compare with.
Probably, even the lowest PL amount of DHA found in this
study was enough for visual development. So, in spite of low
DHA levels in blood PL, formula-fed infants still have enough
DHA for their visual development. If initial DHA levels are
lower, like in premature infants (Foreman et al, 1995), an
effect of dietary DHA would be more likely. This could also
explain the consistency of results in visual acuity studies with
premature infants. All these studies showed an in¯uence of
dietary DHA on early visual acuity.
In the literature the relation between DHA and visual
function is not consistent for term infants. One explanation
for this inconsistency between studies may be the use of
different methods to measure visual functions, for example
electroretinograms, VEP, Teller Acuity Cards, etc. However, there is a close correlation between visual acuity
measured with acuity cards and both sweep and steady state
VEP in infants (Allen et al, 1992). Besides, even within
studies that use the Teller Acuity Card method in term
infants, inconsistencies are found. Thus, Innis et al (1994,
1996, 1997) used the Teller Acuity Cards in three studies
with children born at term. In the ®rst study (1994) they
compared the visual acuity of 17 breastfed infants with that
of 18 infants fed standard formula at the age of 14 d and 3
months. They did not ®nd differences in visual acuity
between both groups, despite substantial differences in
erythrocyte and plasma lipid 22 : 6n-3. In their second
study (1996) Innis and coworkers measured the visual
acuity of 433 term infants at 9 months of age. In this
study, they found no in¯uence of infant diet on visual
acuity either. In their third study (1997), they compared the
Teller acuity of breastfed children with that of children fed
different formulas at 90 days of age. Again, no differences
were found between the groups. Auestad et al (1997)
compared visual acuities of infants fed with different
formulas (with DHA, with DHA and AA and without
PUFA) and human milk. Visual acuity was measured
using both the acuity card procedure and a visual evoked
potential method and no differences were observed
between the different groups at the ages of 2, 4, 6, 9 and
12 months. Our ®ndings agree with both Innis et al (1994,
1996, 1997) and Auestad et al (1997). In their recent study,
Birch et al (1998) also found no effect of diet on visual
acuity, as measured with the forced choice looking protocol, at different time points during the ®rst 12 months of
life. In an earlier study, however, Birch et al (1993) found
that at 4 months of age, breastfed infants scored higher
Teller acuities as compared with formula-fed infants and
acuity was correlated with n-3 fatty acids in RBC membranes. Furthermore, Jùrgensen et al (1996) compared
breastfed infants with formula-fed infants at 1, 2 and 4
months. They found that the increase in visual acuity,
measured by Teller Acuity Cards, developed more rapidly
in breastfed infants compared to formula-fed infants. This
was paralleled by a decrease in the amount of DHA in RBC
of formula-fed infants, and with a signi®cantly lower level
at two and four months as compared to breast-fed infants.
Carlson et al (1996) compared term infants fed DHA and
AA enriched formula with term infants fed standard formula or human milk. They measured the visual acuity with
Teller Acuity Cards at 2, 4, 6, 9 and 12 months. At 2
months, breastfed infants and infants fed the supplemented
formula had higher acuities than term infants fed standard
formula. At later ages, these differences had disappeared
(Carlson et al, 1996).
Visual acuity, as measured with the Teller Acuity Cards,
plateaus at about 6 months of age (Teller, 1990). It is
possible that infants fed human milk reach their plateau
earlier than formula-fed infants. So, before a certain age,
differences can be found that disappear after that age. None
of the discussed studies has shown differences in visual
acuity among breastfed and formula-fed infants with Teller
Acuity Cards at this time point. Some studies did show an
effect of LCP on VEP. For instance, Birch et al (1998)
found differences in visual acuity with transient VEP at 6,
17 and 52 weeks of age, but not at 26 weeks of age. So
maybe the effect of LCP on visual acuity is only transient,
and could have disappeared at 7 months of age. This might
explain the absence of a relation between infant diet and
visual acuity at 7 months of age, and may be an explanation
for the inconsistencies between studies in infants born at
term as well.
Another explanation for these inconsistencies can be the
presence of confounding factors. For instance, all studies
comparing a human milk group with a formula group were
non-randomized. So, there may be other differences
between both groups. In the current study, we related
visual acuity at 7 months to various potentially confounding factors, like smoking and drinking habits during prregnancy, and socio-economic background. Furthermore, we
looked at the use of a dummy (paci®er) in infancy. Gale
& Martyn (1996) showed a negative in¯uence of dummy
use or the functioning of the central nervous system, as
877
Essential fatty acids and visual acuity in term infants
EC Bakker et al
878
measured by an IQ test. In the present study, of all potential
confounding factors, only dummy use showed a positive
correlation with visual acuity (P < 0.05). From the study of
Gale & Martyn, the opposite result was expected.
Although dummy use correlated with visual acuity, this
variable did not confound the relation between fatty acids
in blood and diet and the visual acuity. Other variables also
did not confound this relationship. So confounders do not
seem to play a role in this relationship, suggesting other
possible factors are responsible for the inconsistency of
results in visual acuity studies in term infants.
Both human milk and formulas in different studies have
different compositions and=or LA=ALA ratios (Jùrgensen
et al, 1996). Human milk also differs from infant formula in
contents other than DHA or PUFA. These differences may
play a role in visual development as well (Birch et al, 1992;
Jùrgensen et al, 1996).
From our study and other data available, we can conclude
that at 7 months there is no relationship between blood LCP
values and Teller Acuity in infants born at term. There are,
however, some limitations of our study that should be noted. It
is possible that the study groups were too small to detect any
in¯uence of confounding factors on the relationship between
visual acuity and LCP. Power calculations on potentially
confounding factors could not be made because data about
the in¯uence of these factors were not available. Consequently, it was not known what results could be expected.
Furthermore, not all data of all infants were complete, so
conclusions can not be extrapolated to other populations. In
the present study the acuity was measured only once at 7
months. At this age, the ®ne visual acuity development is not
yet complete; it only reaches the adult level at 3 or 4 years of
age (Catford & Oliver, 1973; Touwen, 1982). As far as we
know, adult visual acuity levels have never been measured in
relation to essential fatty acids. So, further studies on the long
term effects of n-3 fatty acids in infant nutrition on visual
acuity are also necessary.
Acknowledgements ÐThe authors wish to thank all the parents and their
infants who participated in this study and are grateful to the participating
child health centres. The statistical support of Arnold Kester, the technical
support of Marianne Simonis and Hasibe Aydeniz and the discussions with
Suzie Otto are greatly appreciated.
References
Allen D, Bennett PJ & Banks MS (1992): The effects of luminance on FPL
and VEP acuity in human infants. Vis. Res. 11, 2005 ± 2012.
Anderson RE, O Brien PJ, Wiegand RD, Koutz CA & Stinson AM (1992):
Conservation of docosahexaenoic acid in the retina. Neurobiology of
Essential Fatty Acids, eds NG Bazan et al, New York: Plenum Press
285 ± 294.
Auestad N, Montalto MB, Hall RT, Fitzgerald KM, Wheeler RE, Connor
WE, Neuringer M, Connor SL, Taylor JA & Hartmann EE (1997):
Visual acuity, erythrocyte fatty acid composition and growth in term
infants fed formulas with long chain polyunsaturated fatty acids for one
year. Pediatr. Res. 41, 1 ± 10.
Birch E, Birch D, Hoffman D, Hale L, Everett M & Uauy R (1993): Breast
feeding and optimal visual development. J. Pediatr. Opthalmol. Strabismus 30, 33 ± 38.
Birch EE, Birch DG, Hoffman DR & Uauy R (1992): Dietary essential
fatty acid supply and visual development. Invest. Opthalmol. Vis. Sci.
33, 3242 ± 3253.
Birch EE, Hoffman DR, Uauy R, Birch DG & Prestidge C (1998): Visual
acuity and the essentiality of docosahexaenoic acid and arachidonic acid
in the diet of term infants. Pediatr. Res. 44, 201 ± 209.
Bligh ED & Dyer WJ (1959): A rapid method for total lipid extraction and
puri®cation. Can. J. Biochem. Physiol. 37, 911 ± 917.
Brown KM, Morrice PC & Duthie GG (1998): Erythrocyte membrane fatty
acid composition of smokers and non ± smokers : effects of vitamin E
supplementation. Eur. J. Clin. Nutr. 52, 145 ± 150.
Burr GO & Burr MM (1929): A new de®ciency disease produced by the
rigid exclusion of fat from the diet. J. Biol. Chem. 82, 345 ± 367.
Carlson SE, Werkman SH, Rhodes PG & Tolley EA (1993): Visual acuity
development in healthy preterm infants : effect of marine ± oil supplementation Am. J. Clin. Nutr. 58, 35 ± 42.
Carlson SE, Ford AJ, Werkman SH, Peeples JM & Koo WWK (1996):
Visual acuity and fatty acid status of term infants fed human milk and
formulas with and without docosahexaenoate and arachidonate from
egg yolk lecithin Pediatr. Res. 39, 882 ± 888.
Catford GV & Oliver A (1973): Development of visual acuity Arch. Dis.
Child. 48, 47 ± 50.
Clandinin MT, Chappell JE, Leong S, Heim T, Swyer PR & Chance
GW (1980a): Intrauterine fatty acid accretion rates in human brain :
implications for fatty acid requirements Early Hum. Dev. 4,
121 ± 129.
Clandinin MT, Chappell JE, Leong S, Heim T, Swyer PR & Chance GW
(1980b): Extrauterine fatty acid accretion in infant brain : implications
for fatty acid requirements. Early Hum. Dev. 4, 131 ± 138.
Crawford MA (1993): The role of essential fatty acids in neural
development : implications for perinatal nutrition. Am. J. Clin. Nutr.
57, 703S ± 710S.
Folch J, Lees M & Sloane Stanley GH (1957): A simple-method for the
isolation and puri®cation of total lipids from animal tissues. J. Biol.
Chem. 226, 497 ± 509.
Foreman-Van Drongelen MMHP, Al MDM, Houwelingen AC v, Blanco
CE & Hornstra G (1995): Comparison between the essential fatty acid
status of preterm and full ± term infants, measured in umbilical vessel
walls. Early Hum. Dev. 42, 241 ± 251.
Gale CR & Martyn CN (1996): Breastfeeding, dummy use and adult
intelligence. Lancet 347, 1072 ± 1075.
Hoffman DR & Uauy R (1992): Essentiality of dietary omega-3 fatty acids
for premature infants; plasma and red blood cell fatty acid composition.
Lipids 27, 886 ± 895.
Holman RT (1960): The ratio of trienoic : tetraenoic acids in tissue lipids,
as a measure of essential fatty acid requirement. J. Nutr. 70, 405 ± 409.
Holman RT (1986): Control of polyunsaturated fatty acids in tissue lipids
J. Am. Coll. Nutr. 5, 183 ± 211.
Hornstra G, Al MDM, Gerrard JM & Simonis MMG (1992): Essential
fatty acid status of neonates born to Inuit mothers : comparison with
caucasian neonates and effects of diets. Prostaglandins Leukot. Essent.
Fatty Acids 45, 125 ± 130.
Houwelingen Acv, Foreman-V.Drongelen MMHP, Nicolini U, Kypros
HN, Al MDM, Kester ADM & Hornstra G (1996): Essential fatty
acid status of fetal plasma phospholipids : Similar to postnatal values
obtained at comparable gestational ages. Early Hum. Dev. 46,
141 ± 152.
Huisman M, Beusekom CM van, Lanting CI, Nijeboer HJ, Muskiet FA &
Boersma ER (1996): Triglycerides, fatty acids, sterols, mono ± and
disaccharides and sugar alcohols in human milk and current types of
infant formula milk. Eur. J. Clin. Nutr. 50, 255 ± 260.
Innis SM (1991): Essential fatty acids in growth and development. Prog.
Lipid Res. 30, 39 ± 103.
Innis SM (1992): Plasma and red blood cell fatty acid values as indexes of
essential fatty acids in the developing organs of infants fed with milk or
formulas. J. Pediatr. 120, S78 ± S86.
Innis SM, Nelson CM, Rioux MF & King DJ (1994): Development of
visual acuity in relation to plasma and erythrocyte omega ± 6 and
omega ± 3 fatty acids in healthy term gestation infants. Am. J. Clin.
Nutr. 60, 347 ± 352.
Innis SM, Nelson CM, Lwanga D, Rioux FM & Waslen P (1996): Feeding
formula without arachidonic acid and docosahexaenoic acid has no
effect on preferential looking acuity or recognition memory in health
full-term infants at 9 months of age. Am. J. Clin. Nutr. 64, 40 ± 46.
Innis SM, Akrabawi SS, Diersen-Schade DA, Dobson MV & Guy DG
(1997): Visual acuity and blook lipids in term infants fed human milk or
formulae. Lipids 32, 63 ± 72.
Jùrgensen MH, Hernell O, Lund P, Hùlmer G & Michaelsen KF (1996):
Visual acuity and erythrocyte docosahexaenoic acid status in breast-fed
and formula-fed term infants during the ®rst four months of life. Lipids
30, 99 ± 105.
Jùrgensen MH, Hùlmer G, Lund P, Hernell O & Michaelsen KF (1998):
Effect of formula supplemented with docosahexaenoic acid and g ±
linolenic acid on fatty acid status and visual acuity in term infants.
J. Pediatr. Gastroenterol. Nutr. 26, 412 ± 421.
Essential fatty acids and visual acuity in term infants
EC Bakker et al
Kaluzny MA, Duncan LA, Merritt MV & Epps DE (1985): Rapid
separation of lipid classes in high yield and purity using bonded
phase columns. J. Lipid. Res. 26, 135 ± 140.
Kirkwood BR. Essentials of Medical Statistics. Oxford : Blackwell 1988;
Leaf A, Gosbell A, McKenzie L, Sinclair A & Favilla I (1996): Long chain
polyunsaturated fatty acids and visual function in preterm infants. Early
Hum. Dev. 45, 35 ± 53.
Makrides M, Simmer K, Goggin M & Gibson RA (1993): Erythrocyte
docosahexaenoic acid correlates with the visual response of healthy,
term infants. Pediatr. Res. 33, 425 ± 427.
Makrides M, Neumann MA, Bryard RW, Simmer K & Gibson RA (1994):
Fatty acid composition of brain, retina and erythrocytes in breast and
formula-fed infants. Am. J. Clin. Nutr. 1994; 60, 189 ± 194.
Makrides M, Neumann M, Simmer K, Pater J & Gibson R (1995): Are
long ± chain polyunsaturated fatty acids essential nutrients in infancy?
Lancet 345, 1463 ± 1468.
Morissen WR & Smith LM (1964): Preparation of fatty acid methyl esters
anddimethyl acetals from lipids with boron ¯uoride ± methanol J. Lipid.
Res. 5, 600 ± 608.
Neuringer M, Connor WE, Lin DS, Barstad L & Luck S (1986):
Biochemical and functional effects of prenatal and postnatal n ± 3
fatty acid de®ciency on retina and brain in rhesus monkeys Proc.
Natl. Acad. Sci. USA 83, 4021 ± 4025.
Pawlosky RJ & Salem N (1995): Ethanol exposure causes a decrease in
docosahexaenoic acid and an increase in docosapentaenoic acid in
feline brains and retinas. Am. J. Clin. Nutr. 61, 1284 ± 1289.
Pawlosky RJ, Denkins Y, Ward G & Salem N (1997): Retinal and brain
accretion of long chain polyunsaturated fatty acids in developing
felines : the effects of corn oil-based maternal diets Am. J. Clin. Nutr.
65, 465 ± 472.
Stinson AM, Wiegand RD & Anderson RE (1991): Recycling of docosahexaenoic acid in rat retinas during n ± 3 fatty acid de®ciency J. Lipid.
Res. 32, 2009 ± 2017.
Suh M, Wierzbicki AA, Lien E & Clandinin MT (1996): Relationship
between dietary supply of long-chain fatty acids and membrane composition of long- and very long chain essential fatty acids in developing
rat photoreceptors Lipids 31, 61 ± 64.
Teller DY, McDonald MA, Preston K, Lawson Sebrits S & Dobson V
(1986): Assessment of visual acuity in infants and children : the acuity
card procedure Dev. Med. Child Neurol. 28, 779 ± 789.
Teller D (1990): Teller Acuity Card Handbook. Ohio : Vistech Consultants, Inc.
Touwen BCL The development of vision during the ®rst months of life.
Pediatric Ophthalmology. Eds J Francois & M Maione. Chichester :
John Wiley & Sons. 1982; 305 ± 309.
Uauy R, Treen M & Hoffman DR (1989): Essential fatty acid metabolism
and requirements during development Sem. Perinat. 13, 118 ± 130.
Uauy R, Birch E, Birch D & Peirano P (1992): Visual and brain function
mesurements in studies of n ± 3 fatty acid requirements of infants.
J. Pediatr. 120 (Suppl), S168 ± S180.
Weisinger HS, Vingrys AJ & Sinclair AJ (1995): Dietary manipulation of
long ± chain polyunsaturated fatty acids in the retina and brain of guinea
pigs Lipids 30, 471 ± 473.
Wiegand RD, Koutz CA, Stinson AM & Anderson RE (1991): Conservaton of docosahexaenoic acid in rod outer segments of rat retina during
n ± 3 and n ± 6 fatty acid de®ciency. J. Neurochem. 57, 1690 ± 1699.
879