University of Groningen

Brain-selective nutrients in pregnancy and lactation

Stoutjesdijk, Eline

DOI:
10.33612/diss.146373942

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.

Document Version
Publisher's PDF, also known as Version of record

Publication date:
2020

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Stoutjesdijk, E. (2020). Brain-selective nutrients in pregnancy and lactation. University of Groningen.
https://doi.org/10.33612/diss.146373942

Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the
author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the
number of authors shown on this cover page is limited to 10 maximum.

Download date: 15-09-2021

 6
Summary & epilogue
Chapter 6
Summary
This thesis reports on studies in the Nutritional Science, that have been conducted from
an evolutionary prospective. In Chapter 1 we illuminate the origin and importance of
‘brain-selective nutrients’. ‘Brain-selective nutrients’ is a collective term encompassing
those nutrients needed especially for normal human brain development. They are not
specific for brain, but are used elsewhere in the body as well1. If the requirement for any
of the brain-selective nutrients is not met at the correct stage of development, permanent
retardation results. There is good evidence that homo sapiens originated from the (East)
African land-water ecosystem where brain-selective nutrients, including the long-chain
polyunsaturated fatty acids (LCP) eicosapentaenoic acid (EPA) and docosahexaenoic acid
(DHA), vitamin D, vitamin B12, the combination of folate, choline and betaine, iron, zinc,
copper, selenium and iodine, are abundantly available. The land-water ecosystem is where
our Palaeolithic ancestors are likely to have hunted and gathered from about 2.5 million
to 10,000 years ago; the so called Palaeolithic Era. It is in this Era that our genes adapted to
the environment to ultimately provide us with the current relatively large brain according
to the principles of Darwin’s ‘adaptation to the conditions of existence’. This view on our
evolutionary background implies that knowledge on the land-water ecosystem to which
our ancestors were exposed is key to the understanding of who we are and what our
relatively large brain needs. Human milk is homo sapiens’ only food that, to a large extent,
becomes determined by our own genetic machinery and may thus still provide us with
strong clues of our current physiology and the conditions to which we adapted in the past.
The first 1,000 days, i.e. from conception to about 2 years of age, are considered to constitute
a critical window of growth and development. Maternal nutrition prior to conception and
during pregnancy and lactation provides the embryo, fetus and breastfed infant with the
essential nutrients for brain development, healthy growth and an optimal immune system.
The World Health Organization (WHO) recommends exclusive breast milk for the first 6
months of life. To meet the infant’s needs, the composition of breast milk changes during
the lactation period. Human milk’s physiological background encouraged the notion that
breast milk is the ‘perfect food for infants’: ‘breast is best’. However when comparing the
breast milk compositions of mothers from all over the world, some similarities, but also
many differences, can be noted.
Despite the many recommendations, the optimal micronutrient status of pregnant- and
lactating women and their breastfed infants are currently unclear. The requirements for
0-6 month-old infants are almost exclusively derived from the breast milk composition
of apparently healthy mothers. Since lifestyle has changed drastically during the past
about 10.000 years, it is nowadays difficult to define such requirements. The close relation
278
Summary & epilogue
between maternal and infant nutritional status implies that the requirements for infants
could at best derive from (breast milk and blood) samples of life-long adequately fed
mothers. Evidence of maternal ‘adequacy’ should in that case be provided.
In this thesis we focus on the brain-selective nutrients: vitamin B12, vitamin D, LCPω3 and
various (trace) elements during pregnancy and lactation. To find clues to optimise maternal
and infant status, and the milk contents of these micronutrients, we studied populations
inhabiting various geographical regions with different cultural backgrounds, and living at
different latitudes. We selected populations in the Netherlands, Curaçao, Vietnam, Malaysia
and Tanzania. The populations exhibited several differences in life-long lifestyles. There
is a huge difference in exposure to sunlight, ranging from mostly low in the Netherlands
to lifetime-high in Tanzania. The selected populations also have different diets: typically
Western diets in the Netherlands and Curaçao, Asian diets in Malaysia and Vietnam, and
diets determined by culture and the availability of food in Tanzania, such as low- (Maasai),
high- (Sengerema) and very high- (Ukerewe) freshwater fish intakes. The conducted studies
may provide us with insights in similarities and differences in maternal nutrient status and
breast milk compositions with the ultimate aim to find clues for their respective optimality
from an evolutionary perspective. Based on earlier derived optimal status of vitamin D
(25(OH)D) and EPA+DHA in traditionally living women in Tanzania, we also investigated,
in Dutch women, what supplemental vitamin D and EPA+DHA dosages were needed to 6
reach these target levels of adequacy at the pregnancy’s end and in breast milk at 4 weeks
postpartum (PP). We also examined the iodine status of pregnant Dutch women and the
effect of iodine supplementation during pregnancy and lactation.
Vitamin B12
Maternal vitamin B12 insufficiency during pregnancy is a folate-independent risk factor
for neural tube defects. Low vitamin B12 status is prevalent in breastfed infants and has
been associated with neurological symptoms, and in severe cases causes brain atrophy
and physical symptoms like abnormal pigmentation, hypotonia, enlarged liver and
spleen, anorexia and food refusal, failure to thrive, diarrhoea, delayed motor function and
regurgitations. The current vitamin B12 adequate intake (AI) established by the institute of
medicine (IOM) for 0-6-month-old infants is 0.4 μg/day. It is based on the average vitamin
B12 concentration in breast milk of 9 apparently healthy mothers. With an estimated daily
intake of 0.78 L mature milk, the 0.4 μg/day AI corresponds with a breast milk vitamin B12
concentration of 378 pmol/L.
The current IOM-AI is based on a small number of observations, while milk vitamin B12
has been assayed with methods that may have co-measured vitamin B12 analogues. As
a consequence, there is no well-established AI. Breast milk of women with a functionally
279
Chapter 6
adequate vitamin B12 status, as derived from the assay of both plasma vitamin B12 and
methylmalonic acid (MMA) concentrations, may be used to estimate a ‘Functional
Adequate Concentration’ (FAC) of milk vitamin B12. Lactating women with an adequate
vitamin B12 status are likely to have enabled their offspring to build adequate vitamin B12
stores during pregnancy and to produce milk with adequate amounts of vitamin B12.
In Chapter 2 we evaluated milk vitamin B12 concentrations of various populations, using
the IOM-AI for 0-6-month-old infants and the FAC for milk B12, the latter as calculated from
‘vitamin B12-adequate’ Vietnamese mothers. We collected breast milk samples from women
living in the Netherlands, Curaçao, Vietnam: Halong Bay, Phu Tho, Tien Giang, Ho Chi Minh
City, Vietnam, Hanoi, Malaysia: Kuala Lumpur, Tanzania-Ukerewe and Ruvu-Maasai. We
also collected blood from the Vietnamese women and analysed plasma vitamin B12 with
an Electro-ChemiLuminescence immunoassay and MMA with Liquid chromatography-
tandem mass spectrometry (LC-MS/MS). Milk vitamin B12 was analysed by Immulite 1000.
We established a FAC of 248 pmol/L, based on 86 Vietnamese mothers exhibiting plasma
vitamin B12 concentrations >221 pmol/L and plasma MMA <210 nmol/L, which are generally
considered to reflect an adequate vitamin B12 status. We found a mean milk vitamin B12 in
the entire population of 221 pmol/L (range: <111-667 pmol/L).
Other data (geometric mean in pmol/L; range; % below AI; % below FAC) were: The
Netherlands (n=43; 170; n.d.-667; 88%; 74%), Curaçao (n=10; 169; n.d-538; 70%; 60%),
Vietnam: Halong-Bay (n=20; 411; 154-646; 35%; 10%), Phu Tho (n=22; 201; n.d.-593; 86%;
64%), Tien Giang (n=20; 185; n.d.-463; 80%; 65%), Ho Chi Minh City (n=18; 293; 155-579;
61%; 50%), Hanoi (n=21; 219; n.d.-475; 90%; 67%), Malaysia: Kuala Lumpur (n=20; 274; n.d.-
611; 65%; 40%), Tanzania-Ukerewe (n=19; 239; 130-604; 79%; 63%) and Maasai (n=18; 178;
n.d.-401; 94%; 83%). Most importantly we found that 77% of the milk samples contained
vitamin B12 levels below the IOM-AI of 378 pmol/L and a total of 59% contained vitamin
B12 below the FAC of 248 pmol/L. Our data revealed a disturbingly high prevalence of low
milk vitamin B12, especially in the Maasai and the Dutch. The latter observation contrasts
with the adequate vitamin B12 intake of Dutch women of reproductive age, as established
by the Dutch National Food Consumption Survey 2007-2010. Our results are in agreement
with other studies, showing that mothers with adequate vitamin B12 status apparently
produce milk with low vitamin B12, while their infants have adequate vitamin B12 status.
These findings may point at the importance of adequately built fetal vitamin B12 stores
during pregnancy. More studies are needed to obtain insight into the determinants of
vitamin B12 transplacental transport, the size of infant stores, and postnatal vitamin B12
mobilization from these stores. The current 248 pmol/L milk vitamin B12 cut-off, as derived
from ‘vitamin B12 functionally-adequate mothers, needs confirmation from other countries
or intervention studies.
280
Summary & epilogue
Vitamin D
Pregnant and lactating women and their breastfed infants are at risk of vitamin D deficiency.
It is estimated that, dependent on definition, the worldwide occurrence of vitamin D
insufficiency in pregnant women ranges from 8 to 100%. Low vitamin D status in pregnant
women is associated with several pregnancy complications, including pre-eclampsia and
gestational diabetes, and with offspring growth and developmental delays. Low vitamin
D status in infancy is associated with rickets, autoimmune diseases like diabetes mellitus
(DM) type 1, asthma and multiple sclerosis. From an evolutionary point of view, cutaneous
synthesis of parent vitamin D by exposure to UVB may be our principal source of vitamin
D. Since it is advised to keep the infant out of direct sunlight, the most important postnatal
vitamin D sources are breast milk and formula. Breast milk from Western mothers is low
in vitamin D and it is consequently advised to supplement the infant with 10 μg vitamin
D/day.
In Chapter 3.1, we compared the milk anti-rachitic activity (ARA) of mothers from different
populations with the current AI. The current vitamin D AI for 0-6-month-old breastfed
infants is 10μg/day, corresponding with an milk ARA of 513 IU/L. We were particularly
interested to see whether milk from women with lifetime abundant sunlight exposure
reached the AI. We measured milk ARA of mothers with different cultural backgrounds
and fish intakes, living at different latitudes. Mature milk was derived from 181 lactating 6
women in the Netherlands, Curaçao, Vietnam, Malaysia and Tanzania. Milk ARA and plasma
25-hydroxyvitamin D [25(OH)D] were analysed by LC-MS/MS and milk fatty acids were
analysed by gas chromatography (GC)-flame ionisation detector (FID). None of the mothers
reached the milk vitamin D AI. We found the following milk ARA (n; median; range) in the
various populations: the Netherlands (n=9; 46 IU/l; 3–51), Curaçao (n=10; 31 IU/l; 5–113),
Vietnam: Halong Bay (n=20; 58 IU/l; 23–110), Phu Tho (n=22; 28 IU/l; 1–62), Tien Giang
(n=20; 63 IU/l; 26–247), Ho Chi Minh City (n=18; 49 IU/l; 24–116), Hanoi (n=21; 37 IU/l; 11–118),
Malaysia: Kuala Lumpur (n=20; 14 IU/l;1–46) and Tanzania-Ukerewe (n=21; 77 IU/l; 12–232)
and Maasai (n=20; 88 IU/l; 43–189). We also collected blood samples of the lactating women
in Curaçao, Vietnam and from Tanzania-Ukerewe, and found that 33.3% had plasma 25(OH)
D levels between 80 and 249.9 nmol/l, 47.3% between 50 and 79.9 nmol/l and 19.4%
between 25 and 49.9 nmol/l. Milk ARA correlated positively with maternal plasma 25(OH)D
(range 27–132 nmol/l, r=0.40) and milk EPA+DHA (0.1–3.1 g%, r=0.20), and negatively with
latitude (2°S-53°N, r= −0.21). Milk ARA of mothers with lifetime abundant sunlight exposure
was not even close to the vitamin D AI for 0–6 months old infants. We also collected blood
and milk samples of six lactating mother-infant pairs from Ukerewe, Tanzania. The lactating
mothers and their infants had adequate vitamin D status, but their milk ARA was low.
Together with the finding that pregnant women in Sengerema have higher plasma 25(OH)
D compared with non-pregnant and lactating counterparts, our findings may suggest that,
281
Chapter 6
during pregnancy, parent vitamin D from adipose tissue (AT) and 25(OH)D from muscle
tissue become mobilized from these maternal stores for subsequent transplacental transfer
to build fetal stores. Our data may point at the importance of adequate maternal vitamin
D status during pregnancy to support the building of adequate fetal vitamin D stores.
In Chapter 3.2 we investigated what supplemental vitamin D dosages are needed to
reach vitamin D adequacy (25(OH)D 80–249 nmol/L) in pregnant Dutch women. Healthy
pregnant women in the Netherlands were randomized to 10 μg (n=10), 35 μg (n=11), 60 μg
(n=11) and 85 μg (n=11) vitamin D3/day from 20 gestational weeks (GW) to 4 weeks PP. The
participants also received increasing dosages of fish oil supplements and a multivitamin.
Treatment allocation was not blinded. Parent vitamin D and 25(OH)D were measured in
maternal plasma at 20 GW, 36 GW and 4 weeks PP, and in milk at 4 weeks PP. Median 25(OH)
D and parent vitamin D at 20 GW were 85 (range 25-131) nmol/l and ‘not detectable’ (nd)
(range nd-40) nmol/l. Both 25(OH)D and parent vitamin D increased, seemingly dose-
dependently, from 20 to 36 GW and decreased from 36 GW to 4 weeks PP. We found that
35 μg vitamin D3/day was needed to augment 25(OH)D to adequacy (80–249 nmol/l) in
>97.5% of the participants at 36 GW, while >85 μg/day was needed to reach this criterion
at 4 weeks PP. The lower dose needed in pregnancy may relate to mobilisation of maternal
vitamin D stores during pregnancy. The magnitudes of the 25(OH)D increments from 20
to 36 GW and from 20 GW to 4 weeks PP diminished with supplemental dose and related
inversely to 25(OH)D at 20 GW. Milk ARA related to vitamin D3 dose, but the infant adequate
intake of 513 IU/l was not reached. Using linear extrapolation, we estimated that a 213 μg/
day supplement may be needed to reach the infant AI. This dosage seems in line with
another study that found that 160 μg/day during lactating was able to increase milk ARA
above the IOM-AI. However, although perfectly safe with no adverse effects observed, this
vitamin D dose is well above the current upper limit of 100 μg/day for adults. We concluded
that vitamin D3 dosages of 35 and >85 μg/day were needed to reach adequate maternal
vitamin D status at 36 GW and 4 weeks PP, respectively.
The classical function of vitamin D and its metabolites is in bone and calcium homeostasis.
Among these, 1,25-dihydroxyvitamin D [1,25(OH)2D] is considered to be an active hormone,
with the parent vitamin D and 25(OH)D acting predominantly as pre-hormones. It has
been suggested that many cell types may take up the parent vitamin D and convert it
themselves into the active 1,25(OH)2D hormone for autocrine and paracrine functions.
In addition, it has recently been shown that vitamin D is a more potent stabiliser of the
endothelium than 25(OH)D and 1,25 (OH)2D indicating an active functional role of parent
vitamin D.
282
Summary & epilogue
In Chapter 3.3 we investigated the parent vitamin D contents of plasma and AT of
Tanzanian women with lifetime abundant sunlight exposure in relation to earlier published
25(OH)D data. Blood samples derived from non-pregnant women (n=29), pregnant
women (n=78) and 3 days (n=27) and 3 months (n=29) lactating women, and their
infants in Sengerema. We also studied women at delivery (n=16) and 1-3 months (n=8)
PP and their infants in Ukerewe. AT derived from Sengerema women (n=21) taken during
Caesarean section. Parent vitamin D and plasma 25(OH)D were analysed by LC-MS/MS. In
Sengerema, median (range) plasma parent vitamin D concentrations were 125-42 nmol/L
in non-pregnant women, 21 (<4-82) nmol/L in all pregnant women taken together, 9 (<4-
35) nmol/L at 3 days PP lactating women and 18 (<4-47) nmol/L at 3 months PP lactating
women. The median parent vitamin D in cord blood was below 4 (<4-5) nmol/L. Higher
plasma parent vitamin D and serum 25(OH)D were found in pregnant, compared to non-
pregnant women and lactating counterparts. In Ukerewe, maternal median plasma parent
vitamin D concentrations were 225-67 nmol/L at delivery and 19 (<4-46) nmol/L at 1-3
months lactation. The infant plasma parent vitamin D was below 4 (<4-7) nmol/L at delivery
and 11 (<4-23 nmol/L) at 1-3 months PP. Maternal and infant 25(OH)D and parent vitamin D
concentrations were correlated. Median (range) parent vitamin D contents of subcutaneous
and abdominal AT were 15433-557 nmol/kg. With this figure we estimate that the whole
body AT vitamin D content of Sengerema women amounts to 7.0-8.7 mg. Theoretically,
this store could support the release of 20 μg vitamin D/day for one year. Like serum 25(OH) 6
D, plasma parent vitamin D increased during pregnancy. The feto-placental unit may be
exposed to higher circulating parent vitamin D and 25(OH)D, compared with circulating
concentrations in the non-pregnant and lactating states. However, this observation is likely
to be confined to women with high vitamin D stores. This association, together with the
increase of vitamin D binding protein (DBP) during pregnancy, the intra-uterine growth
of a sizeable vitamin-D-naive fetal AT compartment synthesized from polar precursors,
and the possible storage of 25(OH)D in fetal muscle, indicates that both maternal parent
vitamin D and 25(OH)D may contribute to the building of fetal stores.
The low milk ARA of women with lifetime abundant sunlight exposure and their high
vitamin D status and stores, point at the importance of building adequate infant vitamin
D stores during pregnancy. We therefore conclude that adequate postnatal infant vitamin
D status may depend on adequate maternal vitamin D status during pregnancy. Our data
shift the attention from an adequate maternal status during pregnancy and lactation, to
adequacy prior to conception.
283
Chapter 6
Long chain polyunsaturated ω3 fatty acids
The LCPω3 status in adults around the globe has been estimated to be low to very-low,
with sizeable variation. The cause is the low global consumption of seafood. Almost 80%
of the global population does not reach fish intake recommendations. LCP, notably EPA
and DHA, are important during pregnancy and lactation, since they support fetal and
infant (neuro)developments. Low LCPω3 status during pregnancy is related to preterm
delivery, preeclampsia, gestational diabetes, suboptimal infant brain development, and
the development of postpartum depression. Recommendations for EPA+DHA intakes
from governments and health organizations amount to 200–667 mg/day. The Global
Organization for EPA and DHA Omega 3S (GOED) recommends a daily EPA+DHA intake
of 700 mg for pregnant and lactating women, which seems close to the daily 1,000 mg
EPA+DHA from seafood during pregnancy that is associated with the lowest risk of low
verbal IQ at 8-year-old children. The same research group also found that a milk DHA of
about 1 g% was related to the lowest risk of postpartum depression. An erythrocyte (RBC)
EPA+DHA content of 8 g% confers optimal cardiovascular and mental health in adults.
Data from our group showed that mothers with lifetime high fish intakes and an RBC
DHA status of 8 g% at delivery give birth to infants with an RBC DHA of 7–8 g%. At lower
maternal DHA status, biomagnification occurred, whereas at higher status bio-attenuation
was observed. After 3 months exclusive breastfeeding, maternal RBC DHA was 7–8 g%,
while infant RBC DHA had increased to 8 g%. The corresponding breast milk DHA content
at 3 months PP was 1 g%.
In Chapter 4 we established what supplemental dosages of EPA+DHA were needed to
augment RBC EPA+DHA to 8 g% at the pregnancy end and milk EPA+DHA to 1 g% at 4
weeks PP. Healthy pregnant women in the Netherlands were randomised to 225+90 (n=9,
group A), 450+180 (n=9, group B), 675+270 (n=11, group C) and 900+360 (n=7, group D)
mg DHA+EPA/day from 20 GW to 4 weeks PP. The participants also received increasing
dosages of vitamin D3 and a multivitamin. Samples were collected at 20 and 36 GW and 4
weeks PP. Treatment allocation was not blinded. RBC- and milk fatty acids were analysed
by GC-FID. Median RBC EPA+DHA at 20 GW were 5.5 (range 3.3-8.5) g%. Groups A-D did not
exhibit differences between RBC EPA+DHA contents at the start (p=0.267). At 36 GW the
RBC EPA+DHA medians (ranges) had increased to 6.5 (5.5–8.6) g% for group A, 7.4 (6.2–9.3)
g% for group B, 8.7 (8.1–10.4) g% for group C and 9.5 (6.0–11.3) g% for group D (p<0.01 for
between-group differences). The medians (ranges) of milk EPA+DHA at 4 weeks PP were
0.36 (0.27–0.75) g% for group A, 0.81 (0.56–1.06) g% for group B, 1.01 (0.71–1.31) g% for group
C, and 1.08 (0.68–1.68) g% for group D. Linear regression revealed that needed dosages
rounded at 750 mg of EPA+DHA/day were necessary to reach RBC EPA+DHA of 8 g% at
the pregnancy end, while about 1,000 mg of EPA+DHA/day was needed to reach milk
EPA+DHA of 1 g% at 4 weeks PP. RBC EPA+DHA increments depended on baseline values.
284
Summary & epilogue
In other words, there was an RBC EPA+DHA ceiling effect: an RBC EPA+DHA maximum
seems to be reached at about 10 g%. The supplement had no potentially adverse effects
on milk AA. The milk EPA/AA ratio increased but remained within the physiological range
of 0.17–1.08 g/g.
We conclude that a daily 1,000 mg EPA+DHA supplement in pregnancy may optimize
LCPω3 status of both mother and child. This dosage is in excellent agreement with the
dosage recommended for the secondary prevention of cardiovascular disease and the
primary prevention of affective disorders.
Elements
Most micronutrients are important for infant growth and development. Heavy metals,
however, can be toxic and become transferred to breast milk upon maternal exposure
to high dietary levels, or by other environmental means. Recently, inductively coupled
plasma-mass spectrometry (ICP-MS) has become more widely available, making it possible
to simultaneously analyse breast milk elements with high sensitivity and selectivity. Few
studies have investigated into a wide range of breast milk essential and toxic elements in
different populations. One of the brain-selective nutrients that depends clearly on maternal
status and intake is iodine. Iodine as a building block of thyroid hormone, is e.g. important
for fetal and infant brain development. Pregnant and lactating women are vulnerable to 6
iodine deficiency. From the time that the maximally permitted iodine content of bakery
salt in the Netherlands was reduced, the iodine intake has decreased by 20-25%. Iodine
deficiency, especially during pregnancy, occurs in many West-European countries. The
current iodine status of pregnant and lactating Dutch women, is unknown.
In Chapter 5.1, by using ICP-MS, we determined six essential macro-elements (potassium,
sodium, phosphorus, calcium, sulphur, magnesium), seven essential micro-elements (zinc,
iron, copper, iodine, selenium, manganese, molybdenum) and three potentially toxic
elements (arsenic, cadmium, bromine) in breast milk samples of 206 healthy mothers
living in five countries and ten locations in The Netherlands (n=43), Curaçao (n=10), Vietnam
(n=101; Halong Bay, Phu Tho, Tien Giang, Ho Chi Minh City, Hanoi), Malaysia (n=20; Kuala
Lumpur) and Tanzania (n=32; Ukerewe, Maasai). Concentrations were similar to those of
previous reports using the same technique, but iodine was similar or lower. The outcomes
were compared with the IOM-AI for 0-6-month-old infants, as converted to mature milk
concentrations, assuming a milk intake of 780 mL/day. For the total population we found
that current medians (ranges) milk concentrations in mmol/L for potassium 14.2 (9.4-19.8),
calcium 7.0 (3.5-9.7) and phosphorus 4.7 (1.3-6.7) were somewhat higher, sodium 6.4 (2.6-
39.4) was about equal and magnesium 1.3 (0.7-1.9) was somewhat lower. Sulphur was
4.1 (2.8-10.2) mmol/L. Expressed in µmol/L, copper 5.2 (1.6-26.1) was somewhat higher,
285
Chapter 6 Summary & epilogue

manganese 0.07 (0.02-1.52) about equal, while zinc 32.0 (4.5-89.5), iron 5.5 (1.1-21.7), iodine 0.7
(nd-6.3), selenium 0.19 (0.09-0.55) and molybdenum 0.02 (nd-2.89) were somewhat lower.
Bromine 11.9 (3.8-44.1) µmol/L and arsenic 3.8 (nd-168) nmol/L were above established cut-
offs in 38/206 and 3/206 samples, respectively, while cadmium 0.55 (nd-7.27) nmol/L was
below established cut-offs. When the sum of potassium, calcium, sodium and magnesium
was set at 100%, we found significant negative associations between sodium vs potassium,
sodium vs. calcium and sodium vs. magnesium, and positive relations between potassium
vs. magnesium and calcium vs. magnesium. Ratios in mol/mol were: potassium/sodium
2.2 (0.4-5.8), calcium/magnesium 5.6 (2.9-10.4), zinc/copper 6.0 (1.0-27.9), iodine/selenium
3.3 (0.2-27.1), iodine/bromine 0.05 (nd-0.41) and selenium/sulphur 0.05 (0.03-0.10). Median
ratios compared well with those of an infant formula, except for the higher 8.5 zinc/copper
ratio in the formula.
We conclude that the milk concentrations of the currently investigated elements were
comparable with those reported by others using ICP-MS, and that for most elements
there were no major differences between the median concentrations of the investigated
essential elements and the concentrations needed to reach the IOM-AIs for 0-6-month-
old infants. Exceptions were copper, iodine, selenium and molybdenum, which had
over 20% deviation from the IOM-AI. The essential macro-elements potassium, calcium,
sodium and magnesium exhibited interrelations, which is conceivable from both functional
(aiming at infant homeostasis) and mechanistic (interacting transporters) points of
view. The concentrations of various essential microelements in milk and some of their
investigated ratios were subject to high inter-individual variation. Of special concern
are the current AIs of essential elements that are not tightly regulated in milk, notably
iodine and selenium. These are known to exhibit interaction (in e.g. thyroid hormone
synthesis), are clearly dependent on maternal intake, and display concentrations that in
apparently healthy mothers are unlikely to provide information on adequacy, because of
the current widespread poor iodine and selenium status, numerous confounding factors
(e.g. goitrogens), and the current neglect of iodine’s many extra-thyroidal functions. An AI
based on functional outcomes instead of median breast milk concentrations of apparently
healthy mothers seems to provide stronger evidence for micro-element requirements of
0-6-month-old infants.
women are 150 and 100 μg/l, respectively. AIs of iodine for infants aged 0-6 months
are 1.1 μmol/l (IOM) or 0.5 μmol/l (Nordic Council recommendations). The median UICs
(percentages below cut-off) were 102 μg/l (83%) at 20 GW, 144 μg/l (56%) at 36 GW and
112 μg/l (40%) at 4 weeks PP. The UIC of women, who took iodine-containing multivitamin
prior to 20 GW (61% of the women), did not differ from the UIC of counterparts who did
not. However there was a trend in the dose-response: the women ingesting the highest
doses seemed to have the highest UIC. We found correlations between iodine status at
20 GW with 36 GW and between 36 GW and 4 weeks PP. The median breast milk iodine
concentration was 1.2 μmol/l (range 0.5-3.0); 33% and 0% of the infants had estimated
iodine intakes below the IOM-AI and Nordic-AI, respectively. We did not find a correlation
between UIC at 4 weeks PP and breast milk iodine concentration.
This pilot study suggests a high prevalence (83%) of iodine deficiency during pregnancy.
The insufficiency was not entirely corrected by the use of a daily 150 μg iodine supplement.
The median breast milk iodine concentration seems adequate. Due to the potentially
severe and easily preventable consequences of iodine insufficiency for both mother and
her offspring, further studies, using a representative sample of the Dutch population, are
urgently needed to establish the current Dutch iodine status of pregnant and lactating
women. Iodine supplementation, preferably prior to conception, could be advised to
optimise thyroid hormone stores with positive effect on both mother and child. Iodine 6
supplements are to be administered to people with adequate selenium status, which
makes the simultaneous study of both elements of utmost importance.

In Chapter 5.2, we describe the outcomes of a pilot study, in which we examined the
iodine status of 36 pregnant Dutch women. From 20 GW until 4 weeks PP, they ingested
150 μg iodine/day in the form of a multivitamin supplement for pregnant and lactating
women. Twenty-four hour urine samples were collected at 20 and 36 GW and at 4 weeks PP.
A breast milk sample was collected at 4 weeks PP. Iodine concentrations were analysed by
ICP-MS. Cut-off values for the urinary iodine concentration (UIC) for pregnant and lactating

286 287
Chapter 6
Epilogue
The genetic constitution (‘nature’; ‘hardware’) plays a crucial role in the infant’s phenotype.
However, phenotype and thereby future health, does not only depend on genetics,
but also on many other factors, of which early nutrition and lifestyle (‘nurture’) are
among the most important. Also the diet and lifestyle of previous generations play an
(evolutionary conceivable) role, which refers to the so called ‘Barker Hypothesis’2 also
named ‘Developmental Origins of Health and Disease’ (DOHaD), or ‘Predictive Adaptive
Response’3. The ‘Barker Hypothesis’ is nowadays widely accepted to find its basis in
epigenetics (‘software’), with the goal to function as an erasable short-to-intermediate
(days-few generations) term adaptation to the predicted future environment.
Evolution proceeds by positive or negative selection of mutations, mainly driven by a
changing environment. The epigenetic machinery is also under genetic control. In other
words, the ‘adaptations to the conditions of existence’ of Darwin are ultimately of genetic
nature and need to be rapid to prevent extinction when selection pressure is high, but
may on the other hand be slow when selection pressure is low. The current trend in
many Western countries is that we become older, but that the number of years without
chronic illness declines. In other words: on the average, people with typically Western
lifestyles do not age in optimal health. It is increasingly acknowledged that we deal with
a conflict4 between our ancient genome that does not adapt at rapid speed in view of our
relatively long generation time and our current environment. Taken together, our currently
suboptimal environment simply does not force us to adapt with rapidity, e.g. within a
couple of generations, because of the relatively mild nature of the exerted selection force.
A really strong selection force, such as exerted by a novel infectious agent, would strongly
reduce the population with survival of only the resistant subjects, causing what is named
a ‘bottle neck’, with subsequent outgrowth of the resistant type.
The evolution theory predicts that our (epi)genetically determined physiology has,
amongst others, become fine-tuned on the nutrient intakes during the preceding millions
of years. Considering the above-mentioned slow adaptation at low selection pressure
that notably affects us after reproductive age, it seems conceivable that for optimal
health, and healthy aging, we have to stay close to the intakes and status on which our
evolutionary established physiology is based. Our current Western lifestyle has, however,
drastically changed during the past 10,000 years, notably due to changes resulting from
the agricultural and industrial revolutions, that is: from the beginning of the era that started
with significant human impact on the Earth’s geology and ecosystems, also named the
Anthropocene. These changes had great impact on our dietary composition. The current
Western diet consists e.g. of abundant amounts of sugar, fat, meat and notably refined
288
Summary & epilogue
food products, and low amounts of vegetables, fruit and (shell)fish. More specifically, it is
characterized by, amongst others, high intakes of (high glycaemic load) carbohydrates, low
intakes of fibre and micronutrients, and an abnormal fatty acid composition that is notably
high in linoleic acid and low in LCPω3. The deviation from a land-water ecosystem in which
we most probably evolved, has importantly affected the intake of the brain-selective
nutrients that explain the unprecedented unique growth of our relatively large brain.
A shortage of brain-selective nutrients, like vitamin B12, vitamin D, LCPω3 and elements
like iron, iodine and selenium, is widespread, while these are still abundantly available in
shellfish, seaweed and fish livers.
The Nutritional Science has unfortunately experienced little influence of what biology
considers being of exclusive importance to understand who we currently are. Theodosius
Dobzhansky, the founder of evolutionary medicine, provided this insight to us and
coined the legendary expression: ‘Nothing in biology makes sense except in the light of
evolution’. In contrast, for its recommendations, the Nutritional Science has relied heavily
on observations of what Western populations consume and its epidemiologically-founded
associations with disease, as corrected by what were considered to be ‘confounders’. In
other words: it was a predominantly statistical affair. This trend was followed by the firm
believe in ‘Evidence Based’ data, that refer to David Sackett’s ‘Evidence Based Medicine’,
but was not meant for the Nutritional Science in the first place, and, moreover, almost 6
exclusively misinterpreted as the outcome of randomized controlled trials (RCTs) and
the meta-analyses thereof. Regarding what ‘Evidence Based Medicine’ is not, Sackett
explained that it is: neither old-hat nor impossible to practice, not ‘cook-book’ medicine,
not cost-cutting medicine and not restricted to randomized trials and meta-analyses.
Another problem with the current Nutritional Science is that the studies are almost
exclusively directed at single nutrients, and that consequently most, if not all, current
recommendations for nutrients are derived thereof. Regarding this issue Marion Nestle
recommended ‘not to take the nutrient out of the context of food, food out of the context
of diet, and diet out of the context of lifestyle’. To this list we may add the increasingly
importance ‘to not take lifestyle out of the context of sustainable health for us all’.
This thesis, although in its analytical assays also directed at single nutrients, tried to
bring more insight into the current status and needs of brain-selective nutrients during
pregnancy and lactation. The common denominator of the investigated single nutrients
is their importance to brain development and their meeting at high concentrations in the
food that derives from in the land-water ecosystem. We showed large variations in breast
milk vitamin B12, vitamin D, and elements, both between- and within- the investigated
populations. Because of the encountered huge variation, it seems inappropriate to
establish an AI for 0-6 month-old infants from the average nutrient contents in breast
289
Chapter 6
milk of mothers who are ‘apparently healthy’ and ‘well-nourished’. We feel that there is
a need to establish RDAs or AIs for exclusively breastfed infants that are based on breast
milk from mothers with proven adequate status or based on infant functional outcomes.
Studies that investigate the diurnal variation of breast milk nutrients in both Western
populations and traditionally living populations may provide us with more insight into
the regulation of nutrient secretion into human milk and may additionally provide us
with practical guidelines for breast milk collection. Maybe breast milk is not meant to
provide a monotonous nutrient concentration. We need to develop sensitive and specific,
preferably non-invasive, biomarkers to find clues for optimal infant nutrient status. This
thesis also showed the importance of adequate fetal stores, which underlines the need to
develop new analytic methods and other innovations for their quantification. An example
of a potentially useful innovation is the Helios Smart Ring® that measures the amount of
sunlight to which a subject is exposed and from this information estimates the amount
of vitamin D that has been synthesized.
The nutrient content of breast milk is only part of a larger picture. As said, the maternal-
fetal axis during pregnancy allows the building of fetal stores that may become utilized
in postnatal life. We showed that human milk with nutrient concentrations below the AI
may not necessarily cause postnatal inadequacy; an adequate maternal status during
pregnancy may lead to an adequate neonatal store. The examples were vitamins D and B12.
We therefore suggest to shift the attention from studies starting from merely pregnancy
and lactation to studies starting from pre-conception, that is: in women of childbearing
age. This may not only be important for vitamins D and B12, but also for iodine and
EPA+DHA, and others. Pregnant women in the Netherlands are currently advised to use a
supplement providing 10 μg vitamin D/day. We concluded that in Dutch pregnant women
vitamin D3 dosages of 35 and >85 μg/day are needed to maintain adequate vitamin D
status in >97.5% of the mothers during pregnancy and lactation, respectively. The lower
dose needed in pregnancy may indicate the mobilisation of vitamin D (metabolites) from
maternal stores and therefore points at maternal depletion. It underlines the need of
adequate vitamin D status and stores in women of childbearing age to guarantee sufficient
pre-conceptional stores and thereby somewhat counteract postconceptional losses. In
line with this notion, we also observed that the EPA+DHA status of the investigated Dutch
women was insufficient, despite their relatively high fish intakes. A dosage rounded at
750 mg of EPA+DHA/day was needed to reach RBC-EPA+DHA of 8 g% at the pregnancy
end and about 1,000 mg of EPA+DHA/day was needed to reach a milk EPA+DHA of 1 g%.
This would correspond to the consumption of about 60-80 g of salmon per day. Like the
women in Tanzania, it seems preferable to have a pre-conceptional LCPw3 status of 8 g%
prior to conception, which will benefit both mother and offspring.
290
Summary & epilogue
This thesis also showed a high prevalence of iodine insufficiency in pregnant Dutch
women. The insufficiency was not entirely corrected by the use of a daily 150 μg iodine
supplement. Although other studies established the importance of adequate iodine status
during notably early pregnancy and lactation, the functional benefits of supplemental
iodine are currently unclear. Since selenium insufficiency is likely to be prevalent in the
Netherlands as well, iodine supplementation studies should also include a selenium
supplement to avoid imbalances; imbalances that are increasingly envisioned to be at
the basis of thyroid autoimmune disease and glandular cancer, like those of the breasts,
prostate and stomach. Obstetricians and midwifes perform regular checks during
pregnancy. We suggest inquiring about the consumption of food from the land-water
ecosystem and supplements containing brain-selective nutrients, including vitamins D and
B12, iodine, selenium and LCPω3. This may lead to a personalized dietary advice in support
of the developing fetus and to prevent the infant’s possible parasite behavior causing
maternal depletion. This will certainly occur at low maternal iodine status.
We are not even close to the understanding of nature on a molecular basis, let alone
the understanding of any positive health effect of nutrients in the light of an even
larger picture, that is: healthy diet, exercise, lifestyle, chronic stress, microbiome and
pollution (e.g. smoking, fine dust). What certainly does not seem to be preferred is the
continuation of studies on the associations between single nutrients and health outcomes 6
in epidemiological settings, performing short-term experiments using food, but especially
doing single nutrient supplement studies to improve some health outcome. Alarming the
world with studies on the calculated risks and number of deaths due to the consumption
of e.g. saturated fat and sodium, as extrapolated from their epidemiological associations
with disease, should no longer take place. They undermine the authority of the Nutritional
Science. There is still little knowledge on the relationship of dietary patterns and health
outcomes, and combinations of nutrients with health outcomes. A holistic approach, by
using high quality diets with abundant amounts of e.g. (shell)fish, vegetables and fruits,
might show more beneficial effects than corresponding studies with single nutrients.
High quality diets are to be preferred over supplements due to the richness of micro- and
macronutrients in such foods, allowing nutrient interactions that we currently do not even
understand. Taken together, we feel that studying whole food and nutrient interactions
could lead to a better understanding of what a healthy diet looks really like.
From the studies in this thesis, it is clear that multiple subclinical insufficiencies/
suboptimalities of brain-selective nutrients do exist. This state of affairs is disturbing,
since deficiencies of brain-selective nutrients are not only related to loss of IQ and mental
retardation initiated during pregnancy and infancy, but also to neuropsychiatric diseases
like depression in later life. Even more disturbing is the knowledge that they are easy to
291
Chapter 6
correct, such as the joint iodine and selenium suboptimalities. Currently, depression is a
major global public health concern, but also a common comorbidity among patients
with other chronic diseases5. It would, on the other hand, nowadays be impossible to
live the lives that our ancestors lived, while the land-water ecosystem of our ancestors
has become largely destroyed. Even worse, we find the current ecosystem polluted with
persistent chemicals, (micro)plastic, nanoparticles and drugs, that all accumulate in our
food system. Not the least: there are pressing concerns on the sustainability of our current
way of life. How we can change our lifestyles to halt and even reverse the rapidly changing
environment is one of the most important questions of this time. Economic growth and
prosperity seem mostly in the first place.
Whatever the solution, it seems important to remind the discussion that we cannot change
our physiology. Massive adoption of a sustainable vegan diet is for instance not a solution.
While there is intense debate on the dietary pattern that keeps us healthy and how much
we should eat without intoxicating ourselves and destroy the various ecosystems, it is
meanwhile common knowledge that we currently do not eat sufficient amounts of fruit,
vegetables and (shell)fish and consume too much rapid carbohydrates and ultra-refined
food. Solutions for this persisting problem may be found in educating the public starting
in school, joining force of consumers to convince food industry to provide us with healthy
choices and last but not least by including lifestyle courses into the medical curriculum (see
below). As long as we insist on RCTs with lifestyle to provide us with proof for prevention
it seems unjustified to complain about the costs of drugs and other medical interventions
that are endorsed by influential organizations, following their judgment on the more easily
performed RCTs with single drugs or devices.
A holistic approach not only entails nutrition, but improvements of lifestyle in general and,
unlike the subject of this thesis, not only during a specific part of life, but throughout the
entire life span. For example lifestyle factors like unhealthy diets, poor physical activity,
insufficient sleep, smoking and other adverse factors constitute risks for developing and
progression of typically Western diseases such as type 2 Diabetes Mellitus (DM). Therefore
Dutch guidelines for the treatment of type 2 DM suggest improving health by quitting
smoking, improving the quality and time of physical activity, weight loss and healthy
nutrition, although difficult to guide for most/some doctors6. Fortunately, Dutch healthcare
seems to experience a paradigm shift towards a more personalised and preventive
healthcare: ‘the right treatment for the right patient at the right time’, concomitant with a
shift from ‘disease thinking’ to a focus on prevention and health promotion.
By promoting a healthy lifestyle, the individual subject is placed in the centre of the (primary
and secondary) prevention of disease. However, in order to be effective, also adjustments
292
Summary & epilogue
within the society are to be made. Temptations and distractions are everywhere. It is also
important to educate (future) doctors and other healthcare providers. Until now, little
attention has been paid to prevention and to lifestyle and nutrition in their educational
trajectories7, but some improvements have certainly been made. These improvements
do not only concern future doctors, but also those practising to date. For instance, there
are initiatives to improve patient recovery after surgery through preoperative lifestyle
intervention, comprising improvement of physical condition, diet, treating possible
anaemia, encourage cessation of smoking and/or alcohol consumption and improvement
of psychological resilience and vulnerability. Patients do recover better after surgery if they
are in good physical and mental health.
Also health insurance companies may encourage individuals to adopt a healthy lifestyle.
Together with the government they may improve health by providing correct information
on healthy food and lifestyle and by choosing the healthier option that is better visible and
available. The food industry provides logos to inform costumers what products are healthy
choices. Unfortunately the current logos cause confusion and do not always indicate the
healthiest products. The healthy choice is often less visible and people are seduced with
an unhealthier choice. The same holds for the food in restaurants of companies where
one may find ample options for a diet with (high glycaemic load) carbohydrates, highly
refined food products and a shortage of fruits, vegetables and a healthy protein sources. 6
As a resident in clinical chemistry, I believe there is also a growing role for my future
profession. With both general and in-depth knowledge of (patho)biochemistry and
(patho)physiology, the clinical chemist is a generalist with a lateral vision that is needed
in laboratory health assessment. There are several ways in which a clinical chemist can
contribute to health assessment and the needed adjustments to arrive at a healthy lifestyle.
By applying laboratory measurements, subclinical deficiencies can be demonstrated.
That is: prior to the appearance of clinical symptoms. For instance: current markers like
high sensitivity C-reactive protein (measured on several different occasions), triglycerides/
HDL-cholesterol ratio (marker of ‘small sense LDL’), arachidonic acid/eicosapentaenoic
acid ratio (marker of pending exaggerated inflammatory response with poor resolution)
and trimethylamine N-oxide (a non-causal mortality risk factor) can provide information
on ‘chronic low grade inflammation’, insulin resistance, atherogenic dyslipidaemia,
and cardiovascular disease and mortality risks. Perhaps not in line with the need to be
economical in ordering laboratory tests, profiling of biomarkers could provide us with
more information. Assessment of the predictive values of such multiple tests is of course
necessary. By combining biomarkers, it might be possible to show (sub)clinical deficiencies
and provide a personal advice, similar to the already existing Nutriprofiel® that measures
hemoglobin, iron, folate, and vitamins B6, B12 and D. If used correctly, point-of-care testing
293
Chapter 6
may be another way to gain easier access to nutrient status or risk. Workers in clinical
chemistry, universities and analytical industries, may engage in innovations leading to
easy, sensitive and specific measurements of health, for example measuring breath-, saliva-
or urine-compositions to gain an overview of nutrient status. A different approach, by
combining common sense, innovations, point-of-care testing and big data derived from
many individuals, would probably bring us new insights into a healthy lifestyle and the
prevention of disease. I believe that we can make a difference by working with other
healthcare providers and health insurance companies.
294
Summary & epilogue
References
(1) Cunnane SC. Survival of the Fattest: The Key to Human Brain Evolution. 1st ed. Singapore: World
Scientific Publishing Company; 2005.
(2) Godfrey KM, Barker DJ. Fetal nutrition and adult disease. Am J Clin Nutr 2000 May;71(5 Sup-
pl):1344S-52S.
(3) Gluckman PD, Buklijas T, Hanson MA. Chapter 1 - The Developmental Origins of Health and Dis-
ease (DOHaD) Concept: Past, Present, and Future. In: Rosenfeld CS, editor. The Epigenome and
Developmental Origins of Health and Disease. 1st ed.: Elsevier; 2016. p. 1-15.
(4) Cordain L, Eaton SB, Sebastian A, Mann N, Lindeberg S, Watkins BA, et al. Origins and evolution of
the Western diet: health implications for the 21st century. Am J Clin Nutr 2005 Feb;81(2):341-354.
(5) Li H, Ge S, Greene B, Dunbar-Jacob J. Depression in the context of chronic diseases in the United
States and China. Int J Nurs Sci 2018 Nov 29;6(1):117-122.
(6) Nederlands Huisartsen Genootschap. NHG-standaard. Diabetes mellitus type 2. 2018; Available at:
https://www.nhg.org/standaarden/volledig/nhg-standaard-diabetes-mellitus-type-2. Accessed
12/28, 2019.
(7) Crowley J, Ball L, Hiddink GJ. Nutrition in medical education: a systematic review. Lancet Planet
Health 2019 Sep;3(9):e379-e389.
295