Caballero B. (ed.) Encyclopaedia of Food Science, Food Technology and Nutrition. Ten-Volume Set

Подождите немного. Документ загружается.

have been used most widely and have been shown to

be safe should be preferred to anything new or

untested. Small doses of simple analgesic drugs such

as aspirin and acetaminophen (paracetamol) appear

to be safe, as do common cold remedies, which often

have these drugs as their active contents.

0028 No problems have been reported with the use of

homeopathic remedies in pregnancy. There has been

much interest in possible therapeutic benefit from the

consumption of oil of evening primrose in pregnancy.

It has been hypothesized that the linoleic acid and

g-linoleic acid in the compound may stimulate pro-

duction of vasodilatory prostaglandins and be of

benefit in lowering blood pressure in abnormal states

of pregnancy such as preeclampsia. The limited stud-

ies so far reported do not support this hypothesis.

0029 Recent reports have compared women who are on

diets rich in natural fish oil, the Faroese, with women

on the Danish mainland whose diets contain much

less fish oil. The Faroese had fewer premature babies,

fewer problems with blood pressure, and heavier

babies than the mainlanders. Unfortunately, these

studies cannot be interpreted as showing uniform

benefit from fish oil supplements as the Faroese also

lost more babies as stillbirths. A further report, com-

paring women in Aberdeen with women in the North

Sea Orkney Islands, showed the latter to have average

birth weights 250 g higher. After correction, a small

but significant proportion of the difference was found

to be related to genetic or environmental factors. The

latter included a diet in Orkney containing 30% more

fish than in Aberdeen.

0030 Ginger root may act as an antiemetic by a local

effect in the stomach. Evidence of benefit in preg-

nancy nausea exists.

0031 Garlic is widespread in the human diet. There has

always been a recognition of possible benefits of

garlic supplements on a wide range of disorders,

including hypertension, hyperlipidemia, and throm-

bosis. No studies of supplementation in pregnancy

have been reported.

Vitamin and Mineral Supplementation

0032 Some pregnant women benefit from iron and/or folic

acid supplements when dietary deficiency provokes

anemia. Although self-medication with vitamin and

mineral supplements is widespread, there is a good

case during pregnancy to restrict such supplements to

those prescribed by doctors or preparations contain-

ing less than 100% of the recommended dietary

allowance. Vitamin A in the form of retinol and its

analogs can cause congenital malformations in mega-

dosage, and vitamin D supplementation is thought to

have been the cause of idiopathic hypercalcemia. This

was a congenital syndrome consisting of abnormal

facial features, mental retardation, and abnormal

calcium metabolism. It was common in the UK in

the late 1950s and early 1960s when vitamin D sup-

plementation of a variety of foods was practiced. (The

relationship of periconceptional multivitamins to

NTDs is discussed elsewhere.)

0033It seems clear that some members of dark-skinned

races are at risk of vitamin D deficiency when resident

in temperate climates. Their newborns may be at

risk of hypocalcemia, a cause of convulsions in the

newborn period. These children may also have

delayed growth in the first year of life. Both problems

may be corrected by vitamin D supplementation in

pregnancy.

0034At various times, maternal zinc deficiency has been

proposed as a cause of both congenital malformation

and retarded fetal growth. No valid evidence has yet

emerged of any benefit derived from maternal zinc

supplementation with regard to either outcome in

humans.

Food Safety in Pregnancy

0035Pregnant women share the concerns of the general

public about food poisoning but there is a great deal

of confusion in the mind of the public and the

clinicians treating them about which foods are safe

to eat during pregnancy.

Listeria monocytogenes

0036This organism is widely distributed in the environ-

ment and the low incidence of infection suggests that

infection requires the ingestion of a large dose by a

susceptible individual. Listeriosis is a significant

problem in pregnancy because infection of the mother

can cause fetal death following transmission of the

organism across the placenta. It is not certain at the

present time whether the apparent increase in fetal

loss from this infection represents improved ascer-

tainment or a genuine increase in disease frequency.

0037Most women do not realize that the symptoms of

Listeria infection are not the same as those for some

other foodborne pathogenic organisms: Listeria symp-

toms are often described as mild ‘flu-like symptoms‘

rather than diarrhea and vomiting. The perceived risk

of Listeria infection in pregnancy for many women

appears to be much greater than the actual risk.

0038During pregnancy the following foods should be

avoided to reduce the risk of Listeria infection:

0039Mold-ripened cheese and some types of sheep and

goats’ milk cheese

0040Pa

– meat, fish, and vegetable (unless canned or

ultra heat-treated)

PREGNANCY/Safe Diet 4735

.0041 Cook–chill foods that have not been thoroughly

reheated

However, many health professionals advise women

to avoid ‘unpasteurized soft cheeses‘ which, to most

pregnant women, means avoiding all types of cream

and cottage cheese together with yogurt and fromage

frais. In fact, it is only mold-ripened cheeses with

either a skin such as that found on Brie and Camem-

bert or a mold inside such as Danish blue that should

be avoided, together with sheep and goats‘ milk

cheese that has not been made with pasteurized

milk. Unnecessary avoidance of these dairy foods

will substantially decrease calcium intake.

Toxoplasmosis

0042 Toxoplasma is a protozoan organism which can cross

the placenta when primary infection occurs in preg-

nancy. It can lead to fetal death, mental handicap,

and/or blindness in an estimated 30% of offspring

of infected mothers. To reduce the risk of infection

during pregnancy women are advised to:

0043 Avoid eating raw and undercooked meat.

0044 Always wash vegetables and salads well to remove

soil.

0045 Wear gloves when gardening.

0046 Avoid unpasteurized sheep and goats’ milk.

0047 Wash hands after contact with cats and kittens and

avoid contact with stray cats.

0048 Wear rubber gloves when cleaning out cat litter

trays or get somebody else to do this job if possible.

(See Vegetarian Diets.)

Salmonellosis

0049 Salmonella bacteria are one of the most common

causes of food poisoning in the UK, giving rise to

sickness and diarrhea. Steering clear of raw eggs and

products containing uncooked egg and exercising ap-

propriate food hygiene can reduce this risk.

Campylobacteriosis

0050 The main source of Campylobacter infection is raw

milk and poultry but a large number of foods have

been implicated in outbreaks. Again, campylobacter

may not have any direct effect on the fetus, but the

Department of Health advises women in the UK to

avoid this distressing illness.

Shellfish

0051 Pregnant women are advised not to eat oysters and

other shelled seafood, such as prawns, mussels, and

crab, unless they are part of a hot meal and have been

thoroughly cooked. When raw these foods may be

contaminated with harmful bacteria and viruses.

Bovine spongiform encephalopathy

0052In 1996 evidence of a new variant of Creutzfeldt–

Jacob disease (CJD) began to be observed, which is

most probably the result of eating beef infected with

bovine spongiform encephalopathy (BSE). A particu-

lar source of concern at the moment is vertical trans-

mission which, although common in cattle, has not

been shown to occur in humans. However, there has

recently been a possible case reported in the media of

mother to child infection.

0053Other concerns of pregnant women in the area of

food safety include the hazards of fungicide and pesti-

cide residues. and possible harmful effects of natural

food toxicants. At present it seems unlikely that there

are toxic residues on foods which can be associated

with harm in pregnancy, but this observation should

not be taken as grounds for complacency. One of the

main difficulties for scientists wishing to explore this

matter is the limited availability of reproducible

assays for the minute quantities to be studied. Simi-

larly, the study of natural food toxicants for possible

harmful effects on the fetus is in its infancy.

Conclusions

0054It is important to monitor continuously any effects

our dietary habits may be having on child-bearing

and child development, but we should be able to

achieve this without causing unnecessary scares.

Most women who consume a good general diet, and

are not addicted to tobacco, alcohol, or other drugs,

can contemplate pregnancy without concern that

their dietary habits can ‘harm the baby‘.

See also: Caffeine; Campylobacter: Campylobacteriosis;

Cholecalciferol: Properties and Determination;

Physiology; Coffee: Physiological Effects; Folic Acid:

Properties and Determination; Physiology; Food

Intolerance: Food Allergies; Food Poisoning:

Classification; Food Safety; Garlic; Listeria: Properties

and Occurrence; Detection; Listeriosis; Retinol:

Properties and Determination; Physiology

Further Reading

Department of Health (2001) Committee on Toxicity of

Chemicals in Food, Consumer Products and the Envir-

onment. Statement on the Reproductive Effects of Caf-

feine 2001/06. London: HMSO.

Department of Health and Social Security (1990) Vitamin A

and Pregnancy. PL/CMO (90) 11, PL/CNO (90).

London: Stationery Office.

Department of Health and Social Security (2000) Folic Acid

and the Prevention of Disease. London: Stationery Office.

Frisch RE and McArthur JW (1974) Menstrual cycles: fat-

ness as a determinant of minimum weight for height

4736 PREGNANCY/Safe Diet

necessary for their maintenance and onset. Science 185:

949–951.

Jensen KT, Hjollund NH and Henriksen TB (1998) Does

moderate alcohol consumption affect fertility? Follow

up study among couples planning first pregnancy. British

Medical Journal 317: 505–510.

Karen G (ed.) (1990) Maternal–Fetal Toxicology. A Clin-

ician’s Guide. New York: Marcel Dekker.

Maternal Diet, Vitamins, and

Neural Tube Defects

M L Watkins, National Center on Birth Defects and

Developmental Disabilities, Atlanta, GA, USA

S L Carmichael, March of Dimes/California Birth

Defects Monitoring Program, Oakland, CA, USA

Background

0001 The most interesting work relative to neural tube

defect (NTD) etiology and prevention in the last few

years has focused on nutritional factors. This chapter

summarizes what is known about the relation

between NTDs and various nutritional factors,

with particular emphasis on folate, and includes a

discussion of the methodologic challenges inherent

in identifying such associations.

Neural Tube Defects

0002 NTDs result from failure of normal closure of the

embryonic neural tube at one or more of five separate

closure sites. Complex forces are involved, with vary-

ing closure mechanisms at different sites within the

neural tube. Of the several NTD types (Figure 1),

spina bifida is the most common. The extent of phys-

ical disability (e.g., paralysis, bowel and bladder dys-

function) associated with spina bifida depends on the

lesion level, size, and extent of spinal nerve involve-

ment. Developmental, learning, and social problems

are common. Most children born with spina bifida in

developed countries now survive because of extensive

medical and surgical treatment. Many of these

children, however, face lifelong physical and deve-

lopmental disabilities. The death rate for affected

infants in the USA is about 10%, but the rate is

much higher in countries with less accessible medical

care. Clinical deterioration and secondary conditions

occur in many affected children in a variety of body

systems, resulting in complex medical and surgical

care. Because of a lack of information about the

natural history of untreated children, a lack of ran-

domized trials, and a lack of diagnostic criteria for

neurologic complications, there are controversies

about optimal care. The economic costs of spina

bifida are huge. The role and effectiveness of intra-

uterine myelomeningocele repair is controversial;

clinical trials have been recommended.

0003NTDs affect at least 300 000 newborns worldwide

each year. NTD rates vary considerably around the

world, with northern China reporting some of the

Anencephaly

Cranial

neuropore

Neural

fold

Neural

groove

Iniencephaly

Encephalocele

Caudal

neuropore

Somite

Closed spina bifida

Craniorachischisis Open spina bifida

fig0001Figure 1 Neural tube defects. Schematic view (from above) of a

neural tube. Main clinical types of neural tube defects. Anenceph-

aly (lateral view): the absence of the brain and skull can be total

or partial, always fatal. Craniorachischisis (posterior view):

anencephaly occurs with contiguous bony defect of the spine

and exposed neural tissue. Open spina bifida (posterior view):

bony defect of the vertebrae (in this case, of the lower thoracic

vertebrae), accompanied by exposure of neural tissue and men-

inges that are not covered by skin. Closed spina bifida (posterior

view): bony defect of the vertebrae and, if present, the herniated

meninges and neural tissue, are covered by skin. Extent of the

disability associated with spina bifida related to the level and

size of the lesion and the extent of spinal nerve involvement.

Encephalocele (lateral view): the brain and meninges herniate

through a skull defect. Iniencephaly (lateral view): the dysraphic

process in the occipital region is accompanied by severe retro-

flexion of the neck and trunk. Spina bifida and anencephaly are

more common than encephalocele, craniorachischisis, or inien-

cephaly. Adapted from Botto LD, Moore CA, Khoury MJ and

Erickson JD (1999) Neural-tube defects. New England Journal of

Medicine 341: 1509–1519, with permission.

PREGNANCY/Maternal Diet, Vitamins, and Neural Tube Defects 4737

highest rates (Figure 2). Both environmental and gen-

etic factors likely explain the variation in rates. Of

NTDs, 80% are thought to be ‘multifactorial,‘ mean-

ing that they are influenced by both genetic and en-

vironmental factors. A minority of NTDs are caused

by chromosomal abnormalities, single gene muta-

tions and teratogens. Risk factors include maternal

diabetes, use of folic acid antagonists, fever or hyper-

thermia in early pregnancy, low socioeconomic

status, and obesity. In the USA, NTD rates are higher

among Hispanic whites than non-Hispanic whites;

blacks have the lowest rates. NTD-rate monitoring

is complicated by the increasing use of prenatal diag-

nosis and selective termination, which results in lower

birth prevalence. Because the neural tube develops

very early in pregnancy (closure by 28 days postcon-

ception) before most women are aware they are preg-

nant, prevention efforts are challenging, especially in

populations with high unplanned pregnancy rates.

Approximately 95% of all NTD-affected pregnancies

occur in women with no previous history, which is

referred to as ‘occurrence.‘ However, women who

have already had an NTD-affected pregnancy have

a 10-fold greater risk for having an NTD-affected

subsequent pregnancy than women who have not.

Having more than one NTD-affected pregnancy is

referred to as ‘recurrence.‘

0004The increased prevalence of NTDs in lower socio-

economic groups, decreasing NTD rates, and sea-

sonal patterns of occurrence are consistent with a

dietary etiology for NTDs. Certain nutrients (specif-

ically folic acid) were thought to be involved because

of their role in human growth and because birth

defects occurred with maternal use of folic acid an-

tagonists (e.g., aminopterin, when used as an aborti-

facient, induced fetal NTDs). Also, animal data have

demonstrated NTD inducement resulting from vari-

ous nutritional deficiencies and excesses, suggesting

the plausibility of nutritional causes of NTDs.

Methodologic Issues in the Study of

Nutrition and NTDs

0005However plausible the link between nutrition and

NTDs, studying these associations presents inherent

challenges. Approaches to measuring nutritional

status can be broadly classified as dietary, bio-

chemical, anthropometric, or genetic. Many of the

Switzerland

Rate per 10 000 births

Denmark

Netherlands

Italy

Japan

Belgium

Czech Republic

Ireland, Northern Ireland

United states

France

Norway

Uruguay

China (South)

Spain

British Isles

Australia

Brazil

Chile

Argentina

Venezuela

Mexico

China (North)

fig0002 Figure 2 Rates of spina bifida and anencephaly by country. Cases reported in pregnancy terminations are included, where

available. Rates for Latin-American countries and Japan are estimated from hospital-based registries. Rates for other locations are

based on data from birth defects registries that usually monitor only part of a country’s birth population. The following countries and

registries are included: Switzerland (selected counties), Denmark (Odense), The Netherlands (North), Italy (Campania, Emilia-

Romagna, Toscana), Belgium (Hainaut-Namur), Czech Republic, Ireland (Dublin, Belfast), United States (Atlanta, Hawaii), France

(Bouches-du-Rhone, Paris, Strasbourg, Central East France), Norway, Spain (Basque country), UK (Glasgow, North Thames),

Australia (Victoria, South Australia), and China (selected counties). Data from the International Center for Birth Defects and the

European Registration of Congenital Anomalies (1998) World Atlas of Birth Defects, pp. 20–31. Geneva: World Health organization.

4738 PREGNANCY/Maternal Diet, Vitamins, and Neural Tube Defects

difficulties in studying nutrition and birth defects

stem from the fact that most studies are retrospective

in nature, resulting from the relatively low prevalence

of NTDs. This limitation has varied implications,

depending on the particular approach used to meas-

ure nutritional status.

0006 The food-frequency questionnaire is a tool com-

monly used to assess dietary intake retrospectively.

This method produces reasonably valid and reliable,

comprehensive, and semiquantitative data on usual

or average intake of foods and nutrients. The validity

of the data depends, however, on several factors,

including the appropriateness of the food list for the

study population; the accuracy of the nutrient data-

base; and variability in individual nutrient require-

ments and in the bioavailability and absorption of

nutrients from foods (which are affected by, for

example, food processing, fiber intake, alcohol con-

sumption, cigarette smoking, and infection). Chal-

lenges related to the analysis of dietary data include

whether to analyze intake from foods and supple-

ments separately or together; foods versus nutrients;

or single versus multiple nutrients simultaneously.

The high correlation that exists between nutrients

makes it particularly difficult to isolate the independ-

ent effects of individual nutrients.

0007 Depending on the particular nutrient, biochemical

measures of nutritional status may vary by time of

day, by the tissue from which the measurement is

made (serum and toenail zinc, for instance, reflect

more short-term versus long-term zinc status, respect-

ively), and by various host factors (e.g., genes, behav-

iors such as smoking and alcohol consumption,

health status, physical activity, stress, and medica-

tions). Furthermore, pregnancy itself is a time of

changing nutritional requirements and absorption.

These factors must be taken into consideration

when interpreting results from individual studies

and when comparing results across studies.

0008 Anthropometric measurements (e.g., height and pre-

pregnancy weight) can be recalled reasonably well.

Whether retrospectively collected data on more com-

plex measures (e.g., skinfold thickness and waist-to-hip

ratio) can serve effectively as proxies for these param-

eters around the time of conception is questionable.

0009 Genetic material is most commonly obtained from

blood samples or buccal smears. The challenge is in

discovering functionally relevant polymorphisms and

in having sample sizes large enough to detect mean-

ingful differences in risk. Recent advances in tech-

nology are enabling more rapid identification and

analysis of polymorphisms, and innovative appro-

aches to epidemiologic study design are being dev-

eloped that enable smaller numbers of subjects for

examination of certain types of genetic hypotheses.

Folic Acid and NTDs

0010In 1930, Dr. Lucy Wills discovered a factor that cured

the nutritional deficiency anemia of pregnancy

among women in India. This factor was later isolated

from spinach and named ‘folic acid‘ (folium is Latin

for ‘leaf‘). Because humans are unable to synthesize

folate, they must depend solely upon dietary sources.

Folate-rich food sources include green leafy vege-

tables, grains, legumes, certain fruits, and liver. Be-

cause heat, ultraviolet light, and air inactivate food

folate, food processing, preparation, and cooking can

reduce the amount of food folate ingested by an esti-

mated 50–95%. Bioavailability, the extent to which

folates are available for use at the cellular level, varies

widely across foods. The bioavailability of synthetic

folic acid (monoglutamate form), which is used in

cereal grain fortification and multivitamin/mineral

supplements, is estimated to be about twice that of

food folate (polyglutamate form).

0011Studies in the mid-twentieth century linked nutri-

tion and NTDs. Lower vitamin C and folate levels in

one study and poorer diet quality (in terms of macro-

nutrient and fresh fruit and vegetable intake) in an-

other were found among women with NTD-affected

children. Intervention and observational studies

followed (Figure 3). The most convincing and solid

evidence for a preventive effect of folate is provided

by two intervention trials. The international Medical

Research Council (MRC) trial, a large UK-sponsored

recurrence prevention trial, was conducted at 33 sites

in seven countries. Women who had a previous NTD-

affected pregnancy were randomized to one of four

vitamin-use groups. Among 1195 pregnancies with

known outcomes, folic acid (alone or with other vita-

mins) was associated with a 72% reduction in risk.

0012In 1992, a Hungarian randomized trial pro-

vided strong evidence for the efficacy of folic acid-

containing multivitamin supplementation to prevent

NTD occurrence. Results from this trial, combined

with results from several observational studies

(Figure 3) provided convincing evidence that occur-

rence is preventable. As a result, several countries

recommended in 1992 that women of childbearing

potential consume periconceptional folic acid daily

(most commonly 400 mg) to prevent NTDs. The rec-

ommendation was made for all women capable of

becoming pregnant (not just those planning a preg-

nancy), because these birth defects occur before many

women are aware that they are pregnant, and because

many pregnancies are unplanned. Since then, a large,

nonrandomized community intervention in China

demonstrated significant NTD reductions associated

with the use of a 400-mg supplement containing only

folic acid. Risk reductions were greater (85%) in the

PREGNANCY/Maternal Diet, Vitamins, and Neural Tube Defects 4739

high-prevalence northern countries than in the lower-

prevalence southern counties (40%). Additional

evidence for the efficacy of folic acid comes from

another study that found that periconceptional ex-

posure to folic acid antagonist medications more

than doubles the risk for NTDs and that folic acid

supplementation reduced the risk elevation associ-

ated with certain folic acid antagonists.

Efficacy Issues

0013 Based on the risk reduction observed in several stu-

dies, estimates are that at least 50% of NTDs could

be prevented by the use of folic acid. The proportion

of NTDs that are preventable may be less in low-

prevalence areas than in high-prevalence areas.

Some studies also suggest a smaller reduction in risk

for NTDs associated with folic acid use for Hispanics

than for non-Hispanic whites or blacks, and for obese

women compared with average weight women. These

observations require more study. The minimum daily

effective dose of folic acid for NTD prevention is

unknown; it may be < 400 mg for occurrence preven-

tion, but determining the effectiveness of lower

dosages is difficult because of the size and cost of

studies necessary to assess the relative efficacy of

various dosages.

0014The reasons for variability in folate efficacy are

unknown, largely because the biologic mechanism

by which periconceptional folic acid use protects

against NTDs is unknown. Folic acid plays a role in

transfer of methyl groups in the amino acid methyla-

tion cycles, a process essential for recycling homo-

cysteine back to methionine (Figure 4). It is also a

cofactor for enzymes involved in DNA and RNA

synthesis, providing single carbon units for the

synthesis of nucleotide bases, a process that is essen-

tial to the rapid cell division that occurs in early fetal

development. Extrinsic folic acid may work by in-

creasing tissue folate levels enough to override a fail-

ure in folate metabolism, or it may compensate for

deficiency or metabolic defects related to other nutri-

ents closely related to folate metabolism. Recent re-

search has focused on the role of mutations in the

genes that code for enzymes involved in folic acid

metabolism.

Approaches to Increasing Folic Acid Consumption

0015Although the mechanism is unknown, a significant

proportion of NTDs can be prevented with folic acid.

The challenge is in how to translate science into public

health policy and practice. Table 1 summarizes three

approaches for increasing folic acid consumption.

Randomized trials

U.K. '81 (recurrence)

U.K. '83 (recurrence)

Cuba '90 (recurrence)

Observational studies

U.S. '88

U.S. '89

U.S. '89b

Australia '89

U.S. '93

U.S. '95

Community interventions

China '99 (North)

China '99 (South)

0 0.5 1 1.5

Estimated relative risk and 95% confidence intervals

2 2.5

U.K.-International MRC '91 (recurrence)

Hungary '92 (occurrence)

Non-randomized trials

fig0003 Figure 3 Risk of neural tube defects and use of folic acid or multivitamin supplements: summary of studies 1980–95, by country and

year of publication. The figure shows risk estimates (boxes) and 95% confidence intervals. Risk estimates < 1 indicate a reduction in

risk among the ‘exposed‘ mothers (periconceptional use of folic acid or multivitamins). If the upper limit of confidence intervals is > 1,

the risk estimate is not statistically significant. Some studies used folic acid (varying doses); some used multivitamins. Two

randomized controlled trials (UK’ 81 and UK-international-MRC’91) and two nonrandomized trials (UK’83 and Cuba’90) examined

the efficacy of folic acid with or without other vitamins among women with a previously affected pregnancy (recurrence). One

randomized trial (Hungary’92) assessed the efficacy of a multivitamin containing folic acid among women without a previously

affected pregnancy (occurrence). Of the observational studies, five (US’88, US’89, Australia’89, US’93, US’95,) were case-control

studies, whereas one (US’89b,) was an observational cohort. The community intervention study (China’99) was conducted in two

areas, one in Northern China and one in Southern China; it assessed the effectiveness of 0.4-mg folic acid supplements (without other

vitamins) among women without a previously affected pregnancy in areas of high (North) and low (South) occurrence rates of neural

tube defects. From Botto LD, Moore CA, Khoury MJ, and Erickson JD (1999) Neural-tube defects. New England Journal of Medicine 341:

1509–1519, with permission.

4740 PREGNANCY/Maternal Diet, Vitamins, and Neural Tube Defects

Fortification of a food staple provides wide population

coverage at low cost without requiring behavior

change. The optimal fortification level is controver-

sial. Some experts had concerns that daily folic acid

intakes in excess of 1 mg could adversely affect some

persons, particularly persons with untreated cobala-

min deficiency. Folic acid can ameliorate the anemia

associated with cobalamin deficiency, possibly leading

to a failure to detect and treat the cobalamin

deficiency, which could result in the initiation or

Nucleotide

biosynthesis

5,10 methylene

tetrahydrofolate reductase

MTHFR

5,10 methylene

tetrahydrofolate

5 methyl

tetrahydrofolate

Tetrahydrofolate

Folic acid

Folic acid receptors

Methionine

Homocysteine

Methionine synthase

plus vitamin B

methionine synthase

reductase

Methyl

transferases

S-adenosyl

methionine

(SAM)

S-adenosyl

homocysteine

DNA, proteins, lipids

Methylated

DNA, proteins

lipids

Cystathionine synthase

plus vitamin B

Cystathionine

fig0004 Figure 4 Metabolic roles of folic acid and related dietary factors. Simplified schematic of some of the metabolic processes that

involve folic acid and other related factors. Folic acid (bottom left), after entering the cell (aided by folate receptors), participates in the

transfer of carbon units used for the synthesis of nucleotides (top left) or, through the conversion of homocysteine to methionine

(center), for methylation purposes. These processes are regulated by many factors including enzymes (in rectangular boxes) and

vitamins other than folic acid (e.g., vitamins B

and B

). The activity of some enzymes (e.g., methionine synthase) may be influenced

by other enzymes (e.g., methionine synthase reductase). Adapted from Botto LD, Moore CA, Khoury MJ, and Erickson JD (1999)

Neural-tube defects. New England Journal of Medicine 341: 1509–1519, with permission.

tbl0001 Table 1 Ways to increase folic acid/folate consumption to prevent neural tube defects

Approach Advantages Disadvantages Status

Fortify staple foods Proven to increase blood

folate level

Reaches almost all women

Behavior change not

necessary unless

higher intakes needed

Inexpensive

All consumers (not just target

population) increase intake;

concern that some will have

excess intake

Variable staple consumption

results in variable folic acid

intake

Sustained behavior change

required for consumption of

higher amounts

Mandatory fortification

in the USA, Canada, and several

Central and South American

and eastern European

countries; proposed in the UK

Increase use of

supplements

Efficacy proven by clinical

trials

Proven to increase blood

folate level

Relatively inexpensive

Requires sustained behavior

change

Reservations of some medical

and nutritional professionals

Varies by country and survey

question (e.g., in the USA, 30% of

reproductive-age women report

regular use; in The Netherlands,

54% of women report

‘any periconceptional use‘

and 21% ‘appropriate use‘)

Increase consumption of

foods with high levels

of natural folates (e.g.,

fruits and vegetables)

Other benefits of healthy

diet that includes

fruits and vegetables

Efficacy not proven

Much smaller increases in

blood folate levels than

supplements or fortified foods

Requires significant and

sustained behavior change

May be expensive

Various healthy diet campaigns

under way

PREGNANCY/Maternal Diet, Vitamins, and Neural Tube Defects 4741

progression of neuropathy. Although some experts in

the study of birth defects advocated a higher fortifica-

tion level, the US Food and Drug Administration ruled

that, effective January 1, 1998, all flour, corn meal,

pasta, and rice labeled as ‘enriched‘ be fortified with

folic acid at 140 mg per 100 g of cereal grain product.

Canada and several Central and South American and

eastern European countries also have mandated forti-

fication of wheat flour. The fortification level in some

countries (e.g. Chile) was established at a higher level,

with the goal of achieving a higher average daily intake

of folic acid. In 2000, the UK Committee on Medical

Aspects of Food and Nutrition Policy recommended

folic acid fortification at 240 mg per 100 g of flour. At

the time of writing, this proposal is under review but

not yet approved, although voluntary optional fortifi-

cation is allowed in the UK. Serum and red blood cell

folate levels among reproductive aged women in the

USA have more than doubled since fortification of

cereal grain products. In addition, preliminary ana-

lyses suggest an approximate 20% decline in NTD

rates postfortification.

0016 Although multivitamin supplements containing

folic acid are of proven efficacy in preventing NTDs,

their use requires a sustained behavior change by

most women. Educational campaigns in the USA

have increased awareness and knowledge about folic

acid more than they have increased actual supplement

use. In 2001, although 80% of childbearing aged

women in the USA had heard of folic acid, only

29% took a daily vitamin containing folic acid, up

only slightly from 25% in 1995. In the UK and in The

Netherlands, public-awareness campaigns signifi-

cantly increased both knowledge about, and use of,

folic acid. For example, after a 1995 campaign in

The Netherlands, use of folic acid for ‘any period

around conception‘ increased from 25% before the

campaign to 54% after, and ‘appropriate‘ use in-

creased from 5 to 21%. Whether increases in supple-

ment use can be sustained and whether levels of use

plateau at a certain level is unknown. Supplement use

is less common among younger women, less educated

women, and those with lower incomes – populations

that are often harder to reach through health educa-

tion efforts.

0017 Attempts to increase population consumption of

folate-rich foods (e.g., fruits and vegetables) through

programs like the ‘Five-a-Day‘ campaign in the USA

have generally resulted in only modest increases. One

small study found that neither dietary advice nor

folate-rich foods significantly increased women’s red

cell folate levels, whereas supplements and fortified

foods did. This could result from the lower bioavail-

ability of food folate and/or the challenges of making

dietary behavior changes.

Folate-related Nutrients

0018As noted previously, the preventive effects of folate

supplementation are more likely to result from its

compensation for metabolic errors closely related to

folate metabolism, rather than overt folate deficiency.

Folate is involved in an intricate network of reactions

that include many enzymes, nutrients, and other sub-

strates that could affect NTD risk through a variety of

mechanisms (Figure 4). The conversion of 5-methyl

tetrahydrofolate to tetrahydrofolate produces the

substrate necessary for the transmethylation of

homocysteine to methionine; the enzyme methionine

synthase (MS) catalyzes the reaction, and vitamin B

is a cofactor for this enzyme (Figure 4). Homocysteine

may also be metabolized to cystathionine via

cystathionine b synthase (CBS), of which vitamin B

is a cofactor.

0019Hyperhomocysteinemia, which has been shown to

cause NTDs in animal models, may reflect impaired

metabolism of folate, B

, or homocysteine, may

reflect a defect in the enzymes 5,10-methylenetetra-

hydrofolate reductase (MTHFR), MS, or CBS, or

may be harmful on its own. Several studies have

reported an increased NTD risk associated with

high homocysteine levels in humans, in the amniotic

fluid of mothers carrying affected fetuses, in maternal

serum during pregnancy and postpartum, and among

children with NTDs, although results have not been

entirely consistent.

0020Animal models indicate that methionine is also

important to neural tube closure. Although the mech-

anism is unknown, methionine is necessary for

protein synthesis and transmethylation reactions.

Two population-based, case-control studies have

reported a 30–50% reduction in risk for NTDs with

increasing quartiles of methionine intake. Other stud-

ies have been inconsistent: some have demonstrated

higher levels of methionine in the amniotic fluid of

mothers carrying NTD-affected fetuses, whereas an-

other study reported no difference in maternal serum

methionine post-partum.

0021Several studies have indicated lower levels of vita-

min B

in the amniotic fluid and serum of women

with affected pregnancies. Studies of methyl malonic

acid (an indicator of B

deficiency or a defect in B

metabolism) and B

carrier proteins also suggest a

defect in B

metabolism among case mothers. Other

studies have not supported these differences or have

been inconsistent in their findings depending on the

timing of and tissues used for measurement. Although

the results are mixed, they do not seem to provide

strong evidence for a relation between B

deficiency

and NTD risk. Few studies have examined the asso-

ciation between vitamin B

and risk for NTDs;

4742 PREGNANCY/Maternal Diet, Vitamins, and Neural Tube Defects

significant differences have not been observed, but

studies have involved only small sample sizes.

0022 There is some evidence that polymorphisms in the

genes for MTHFR, MS, and CBS contribute to altered

levels of these nutrients and to NTD risk, but again,

results have been mixed.

0023 The individual contributions of each of these indi-

cators of metabolic disturbances to NTD risk, and

their interdependence, are unclear at this point.

Folic acid supplementation might play a role in over-

coming the negative effects of many of these disturb-

ances. However, until the exact causes of NTDs are

determined, the understanding of the implications of

these findings and the way in which they should be

used for prevention purposes remains limited.

Other Nutrients

0024 At least 30–50% of NTDs are folate-resistant; there-

fore, other nutrients have been implicated as possible

causes of these cases. The evidence for the role of

vitamin A, zinc, and inositol in NTD etiology is

reviewed here. Other nutrients – including vitamin

C, riboflavin, iron, calcium, magnesium, selenium,

lead, and copper – also may be associated with

increased NTD risk, but evidence is very limited.

0025 The finding that isotretinoin, a synthetic derivative

of vitamin A, is teratogenic led to investigations of

whether naturally derived vitamin A is associated

with increased risk for NTD-affected pregnancy.

Most studies have found no evidence of an associ-

ation of high vitamin A levels with NTD risk, based

on intake from foods and supplements and on serum

samples taken during the first trimester of pregnancy

or postpartum. The safe level of vitamin A intake is

still debatable, in part because of its potential associ-

ation with other types of malformations and its

association with other negative health outcomes.

The effects of deficient vitamin A intake on NTD

have not been explored.

0026 Zinc is important to cellular growth and differenti-

ation and to folate absorption. Various experimental

studies in animals and case reports of women with

specific disturbances of zinc metabolism have sug-

gested an association between zinc and increased

risk for NTD-affected pregnancy. Results for analytic

studies of zinc and NTDs in humans are inconsistent.

Postpartum studies have reported NTD case mothers

to have levels of zinc in the hair that were higher,

lower, or the same as those in control mothers, higher

levels in toenails, and the same as control mothers

in serum levels. Studies of pregnant women have

reported that mothers with NTD-affected pregnan-

cies have higher levels of zinc in serum, lower levels in

white blood cells, and lower dietary intakes, whereas

other studies have reported no difference in serum

or amniotic fluid. Despite the number of studies,

inference is limited by the wide variability in zinc

measurement and by the limited understanding of

the relationships between these measures.

0027Inositol is a 6-carbon polycyclic alcohol that is

present in all cells, with especially high levels in

nerve cells. It protects against NTDs in the curly-

tailed mouse, a folate-resistant NTD model, leading

some researchers to recommend supplementation

with a combination of folate and inositol. However,

the applicability of these findings to humans has not

been studied.

Limitations

0028In general, the proposed associations reviewed here

are biologically plausible and supported by animal

models. Unfortunately, systematically testing these

associations in humans is difficult, and most existing

evidence is clinical or experimental and speculative.

Sample sizes for many of the studies have been small

(often involving < 20 case-patients). Study population

characteristics and selection methods have been

highly variable; very few have been population-

based. Few studies have adjusted for potential con-

founding or effect modification by other maternal

characteristics, including status with respect to other

nutritional factors or individual nutrients.

0029The ability to measure maternal nutritional status

is also limited. Because pregnancy is a time of tremen-

dous metabolic change, the ability of measurements

taken later in pregnancy to represent the status during

the time of neural tube closure is questionable. For

example, the fetal contribution to levels of nutrients

in amniotic fluid is uncertain and may vary by timing

of measurement and by nutrient. Also, many women

begin taking multivitamin/mineral supplements soon

after the time of neural tube closure.

0030The cause of observed variability in nutrient

levels is usually unknown, limiting the understanding

of how to use such information for preventive

measures. Nutrient metabolism is highly complex,

and many reactions are interdependent. For example,

depending on their cause, high homocysteine levels

could be remedied with supplementation of B

, me-

thionine, folate, or B

; low calcium intake may result

in increased bioavailability of lead, reduced bioavail-

ability of vitamin B

, and, accordingly, reduced me-

thionine synthase activity; and calcium, fiber, protein,

and methionine are all important to zinc bioavailabil-

ity. Therefore, it is critical that future studies examine

interrelationships, including nutrient–nutrient inter-

actions as well as gene–nutrient interactions, when-

ever possible.

PREGNANCY/Maternal Diet, Vitamins, and Neural Tube Defects 4743

Other Nutritional Factors

0031 Evidence for a link between tea and NTDs is weak. A

study in England and Wales showed an excess of

anencephaly among women drinking three or more

cups of tea per day, but no dose–response effect was

observed, and the relation was observed only in

regions with higher NTD prevalence rates. Further-

more, a study in Boston, Philadelphia, and Toronto

found no elevated risk for NTDs associated with tea

drinking. Nor is the link between coffee and NTDs

strong. In the Boston study, no significant differences

in coffee consumption were observed between the

mothers of NTD-affected infants and control

mothers. Also, no association with decaffeinated

coffee or cola drinks has been found. A Finnish

study found no association between central nervous

system defects and coffee intake. Although there are

case reports of NTDs occurring among infants of

mothers who consumed significant amounts of

alcohol during pregnancy, epidemiologic studies

have provided little evidence for a link between

NTDs and alcohol, especially for mothers consuming

minimal or moderate amounts.

0032 Some food contaminants have been proposed to

be NTD teratogens. Potato blight, caused by a fungus

(Phytophthora infestans) that rots potatoes, was

hypothesized to cause NTDs in the early 1970s,

although most evidence does not support this hypoth-

esis. Fumonisin, a potent mycotoxin prevalent in corn

and a carcinogen and cause of various animal diseases

has been suggested as a possible cause of high NTD

rates among persons living along the Texas–Mexico

border who commonly consume corn products.

Recent research suggests that fumonism may be

involved in NTD formation by inhibiting folate

uptake by cells. Also, fumonisins have been shown

to produce NTDs in mouse whole embryo cultures;

folic acid has ameliorated this teratogenic effect.

More research is needed to establish the link between

fumonisin consumption and NTD risk in humans.

0033 About one-fourth of childbearing age women in the

USA and almost as many such women in Canada, the

UK, and Continental Europe, are obese, and obesity

prevalence is increasing rapidly. Studies consistently

find a twofold elevated risk for NTDs among children

of obese women compared with mothers of average

weight, although risk estimates vary with the body

mass index cutoffs and referent groups used

(Figure 5). Most studies found a higher risk estimate

for spina bifida than anencephaly and found no risk

elevation for encephalocele. Reasons for the increased

risk are unknown. Potential explanations include

teratogenic metabolic abnormalities associated with

obesity including elevated levels of glucose, insulin,

uric acid, and endogenous estrogens. Obese women

may be more likely to be nutritionally deficient be-

cause of a self-imposed diet restriction, poor choice of

foods, or increased requirement for folate or other

nutrients. More studies are needed to help elucidate

the mechanism. The population burden of NTDs as-

sociated with obesity is substantial and increasing.

Conclusion

0034There is still much to be learned about the complex

link between nutrition and birth defects. Evidence is

emerging regarding the role of nutritional factors in

the etiology of birth defects other than NTDs (e.g.,

heart defects and orofacial clefts). Despite the chal-

lenges in studying nutritional causes of birth defects,

the prevention potential is substantial and worthy of

pursuit.

See also: Cobalamins: Properties and Determination;

Folic Acid: Properties and Determination; Physiology;

Food Fortification; Nutritional Assessment:

Importance of Measuring Nutritional Status;

Anthropometry and Clinical Examination; Biochemical

Tests for Vitamins and Minerals; Functional Tests;

Obesity: Epidemiology; Retinol: Properties and

Determination; Physiology; Vitamins: Determination;

Zinc: Properties and Determination; Physiology;

Deficiency