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

Подождите немного. Документ загружается.

and adaptability to environmental pressures that

allow pregnancies to occur under adverse circum-

stances, such as food deprivation or the need to per-

form physical labor during pregnancy. These highly

adaptive processes include limiting adipose depos-

ition, reducing diet-induced thermogenesis, and

limiting physical energy expenditure. Any reduction

in diet-induced thermogenesis is either small or of

marginal significance. It has been estimated that a

20% reduction in physical activity during pregnancy

alone would save the necessary energy to complete

pregnancy. Recent work, however, suggests that

significant energy savings from reducing physical

activity only occur in women with high physical

expenditures before pregnancy and those involved in

high physical work loads during pregnancy. Under

conditions of extreme calorie restriction, a fall in

basal metabolic rate has been observed in the first

half of pregnancy.

0007 Rigid guidelines for increasing dietary energy con-

sumption in pregnancy in well-nourished and obese

pregnant women are probably misguided. Overall,

there is a poor correlation between dietary intake

and gestational weight gains. In nonobese and under-

weight women, pregnancy outcome improves when

the minimal weight gain targets are achieved. In obese

mothers minimal or no weight gain in pregnancy does

not appear to jeopardize the pregnancy. As most

women with GDM are obese, these women should

not be expected to meet their energy costs of preg-

nancy through increased dietary intake alone. Despite

firm evidence, it is generally proposed that all obese

women should have weight gains in pregnancy to

account for the products of conception and signifi-

cant calorie restriction should be avoided.

Postabsorptive and Postprandial

Metabolism in Normal Pregnancy

0008 The metabolic changes in pregnancy insure optimal

maternal fetal fuel transfer. Glucose is the primary

fetal oxidative substrate and by late pregnancy the

fetus utilizes an estimated 17–26 g glucose per day.

This high obligatory fetal oxidative metabolism

results in an increase in maternal carbohydrate oxi-

dation and an accompanying rise in the 24-h respira-

tory quotient. Glucose is preferentially diverted from

the mother to the fetus by the fall in maternal insulin

sensitivity. This reduction in insulin sensitivity is due

to the antagonistic actions of placental hormones,

many of which are also lipolytic. The rise in circulat-

ing free fatty acids in late pregnancy also contributes

directly to the increase in peripheral insulin resist-

ance. Increased maternal insulin resistance helps

maximize the transfer of glucose postprandially to

the fetus as glucose uptake into maternal peripheral

tissues is impaired due to the insulin resistance. Post-

prandial glucose and insulin concentrations rise

during pregnancy in well-nourished women consum-

ing western-style diets and insulin resistance falls. By

contrast, little or no rise in postprandial glucose con-

centrations or insulin occurs in rural populations

consuming more traditional low glycemic index

diets and in these women insulin resistance does not

fall. This observation strongly suggests that habitual

diets and lifestyles can influence maternal glucose

tolerance and insulin sensitivity.

0009The high postprandial insulin levels associated

with decreased maternal insulin sensitivity facilitate

maternal fat deposition. The metabolic milieu of

pregnancy is very permissive for fat accumulation

and even undernourished women have a 2-kg increase

in their adipose stores. In well-nourished women

body fat increases by approximately 9% with

2–5 kg of tissue being accumulated by the end of preg-

nancy. Fatty acid oxidation falls in late pregnancy and

this too will further favor adipose accumulation.

0010Metabolic changes in maternal fat metabolism

insure a constant supply of fetal glucose in the post-

absorptive state. This is achieved by a faster switch-

over from lipogenesis to lipolysis. Human placental

lactogen and other placental hormones increase

maternal lipolysis which generates ketone bodies

and glycerol – two important gluconeogenic sub-

strates that help spare maternal glucose and amino

acids for the fetus. Ketone bodies, but not nonester-

ified acids, can cross the placenta and be used as fetal

fuels. Despite lower fasting glucose values, hepatic

glucose output increases in pregnancy due to the glu-

cogneogenic nature of many of the pregnancy-related

hormones and the increase in hepatic insulin resist-

ance. This increase in hepatic glucose output under

fasting conditions also helps maintain a steady supply

of glucose to the fetus.

Metabolic Changes Associated with GDM

0011A degree of deterioration in glucose tolerance in

pregnancy is characteristic of well-nourished women

consuming western-style diets. However, only in a

minority of women is the deterioration sufficient to

warrant the diagnosis of GDM. Most women remain

glucose-tolerant through their ability to treble their

insulin secretion in the face of increasing insulin

resistance. Women who do develop GDM have subtle

abnormalities of b-cell function and, despite being

able to maintain glucose tolerance outside pregnancy,

are unable to when faced with the high insulin

resistance levels encountered in pregnancy. A degree

of b-cell dysfunction is universal in women who

PREGNANCY/Nutrition in Diabetic Pregnancy 4755

develop GDM, both during and following pregnancy.

By the time GDM has developed insulin resistance,

both peripheral and hepatic insulin resistance are

higher than for glucose-tolerant women. There

appears to be a relatively greater b-cell defect in

the nonobese women who develop GDM than the

obese women in whom insulin resistance is a greater

contributing factor. The combination of inadequate b-

cell secretion and increased insulin resistance results in

inadequate nonesterified free fatty acid (NEFFA) sup-

pression postprandially in GDM. Postprandial in-

creases in NEFFA levels would be anticipated to

reduce insulin sensitivity further and compromise b-

cell function.

The Principle of Dietary Management

GDMs

0012 The dietary advice given in GDM needs to lessen the

metabolic abnormalities while insuring adequate

nourishment for the mother and fetus. Dietary advice

should prevent unnecessary weight gain in all preg-

nancies, especially in obese women. Dietetic advice

should in addition provide general guidelines for

healthy eating beyond the pregnancy.

General Dietary Recommendations for

GDM

0013 The diet should include the recommended vitamins

and minerals for pregnancy and potentially harmful

foods such as uncooked meats and soft cheese should

be avoided. Alcohol is potentially harmful to

pregnancy and should be actively discouraged in the

first trimester and severely limited for the rest of

pregnancy.

0014 When diabetes predates the pregnancy, 5 mg of

folate acid should be started before conception and

continued throughout the period of organogenesis

(12th gestational week/4th week from last menstrual

period). Higher doses of folic acid than 400 mg are

recommended for nondiabetic pregnancies for the

prevention of neural tube defects despite any direct

evidence that folate protects against glucose-

mediated neural tube defects. All women with GDM

diagnosed before the 12th gestational week should

start 5 mg of folic acid, as should all previous GDM

women in future pregnancies. A recent large nutri-

tional analysis of over 1000 British pregnant women

showed relatively low folate intakes – fewer than

10% of women were taking folate supplements in

early pregnancy.

0015 Dietary antioxidants protect diabetic rodents

against the teratogenic effects of hyperglycemia. Al-

though similar evidence is not available for human

pregnancies, insuring adequate antioxidants in the

diets of diabetic mothers during organogenesis seems

reasonable.

0016Calcium and vitamin D supplements during both

pregnancy and lactation should be considered for

Asian Indian women and others with poor sunlight

exposure and/or low calcium intakes. Recently a

relationship between insulin resistance and certain

vitamin D receptor polymorphism has been reported,

providing a potential association between the risk of

vitamin D deficiency and diabetes.

Recommended Maternal Weight Gains for

Normal Pregnancies

0017The published guidelines on recommended maternal

weight gains are based on large obstetric surveys in

the USA (Table 1). The maternal weight gain required

to minimize the frequency of small-for-gestational-

age (SGA) infants is higher for underweight (BMI

< 19.8 kg m

2

) than overweight or obese women.

When the prepregnancy BMI is > 35 kg m

2

with little

or no maternal weight gain, there is no added risk of

an SGA infant. Overweight (BMI 26.1–29 kg m

2

)

and obese (BMI > 29 kg m

2

) women are more sus-

ceptible to giving birth to an LGA infant and this risk

increases further as maternal weight gain increases.

The US obstetric recommendation for a 7-kg min-

imum weight gain for all obese women may not be

universally appropriate, especially for morbidly obese

women (BMI > 35 kg m

2

0018The dietary management of GDM women includes

the management of obese pregnant women. Unneces-

sary weight gain in pregnancy contributes to post-

partum obesity, which will increase the likelihood

of future GDM recurrence, diabetes, and other

obesity-related comorbidities. The American Dia-

betic Association (ADA) has endorsed dietary guide-

lines for management of diabetes in pregnancy

recommending a daily calorie allowance based on

prepregnancy ideal body weight (IBW) and current

pregnancy weight. These guidelines recommend 36–

40 kcal kg

1

when IBW < 90%, 30 kcal kg

1

for IBW

of 90–120% and 24 kcal day

1

for IBW 121–150%

and as little as 12–18 kcal kg

1

during pregnancy

tbl0001Table 1 The 1990 guidelines of the US Institute of Medicine on

maternal weight gain targets according to prepregnancy body

mass index

Underweight

(<19.8 kgm

2

)

Normal weight

(19.8^26 kgm

2

)

Overweight

(>26 kgm

2

)

Weight gain

term target

12.5–18 kg 11.5–16 kg 7.0–11.5 kg

4756 PREGNANCY/Nutrition in Diabetic Pregnancy

when the prepregnancy IBW > 150% IBW. The ADA

currently endorses the minimum recommended

weight gain of 7.0 kg for obese (BMI > 29 kg m

2

)

women, as published by the American College of

Obstetricians and Gynecologists. No equivalent

weight or daily calorie guidelines exist for the UK.

Calorie Restriction in the Obese Woman

with GDM

0019 To date there are no long-term studies on the psycho-

logical or physical development of infants whose

obese mothers are mildly calorie-restricted in late

pregnancies for GDM. Long-term follow-up studies

are available from infants born to well-nourished

Dutch women with severe calorie-restricted daily al-

lowance of 800 kcal, in late pregnancy during the 5½-

month famine of 1944–45. These infants were thinner

at birth but had normal subsequent childhood devel-

opment and were reported to be less obese than other

recruits on entry into the army at 18 years old. How-

ever, subsequent follow-up has shown that among

this cohort of children born to calorie-restricted

mothers in late pregnancy, the incidence of glucose

intolerance and diabetes was increased in middle age.

It is these long-term studies combined with genuine

concerns relating to the harmful effect of maternal

ketosis, caused by calorie restriction, on fetal neuro-

physiological and cognitive development that have

resulted in a general reluctance to advocate calorie

restriction in pregnancy.

0020 While no one recommends severe calorie restric-

tion in pregnancy, there is probably a place for

modest calorie constraint for obese women with

GDM. Women with GDM are relatively more keto-

sis-resistant to mild calorie restriction than glucose-

tolerant women; this may be attributable to their

higher hepatic glucose outputs. Theoretically, modest

calorie-restricted diets that are given as small frequent

meals containing slowly absorbed carbohydrates

will help minimize maternal ketosis as these diets

are associated with attenuated insulin responses

which delay lipolysis and therefore ketogenesis.

0021 We have previously reported that in obese GDM

women modest calorie restriction of 20–25 kcal kg

1

day

1

from the 24th gestational week reduces the

frequency of LGA infants compared with obese

glucose-tolerant women not dieted. In this study the

GDM women gained only half the weight of the

controls from 28 weeks to term (1.7 + 1.6 versus

4.1 + 3.1 kg) and this was associated with fewer

LGA infants than for obese non-GDM controls.

Other studies using hypocaloric diets have also

shown improvements in glycemic control with mild

calorie restriction in pregnancy.

0022On today’s evidence it would seem appropriate to

limit weight gain in GDM pregnancies to the bottom

rather than the top for those recommended for

average, overweight, and obese women, while setting

no minimum weight gain for the grossly obese

(> 34 kg m

2

). This would insure that following preg-

nancy there would be no overall weight gain in the

overweight women and potentially allow weight loss

in the morbidly obese woman.

The Optimal Amount and Type of Dietary

Carbohydrate and Fat for GDM

0023Glucose crosses the placenta in both a concentration-

dependent manner and by specific glucose transport-

ers. The reduction of postprandial peak glucose

concentrations has a greater impact on limiting accel-

erated fetal growth than lowering postabsorptive

levels. Therefore diets that prevent excessive post-

prandial placental uptake of glucose should be the

basis of our dietary advice to all women with GDM.

0024Controversy surrounds the optimal dietary propor-

tion that should be taken as carbohydrate. Recent

advice from the USA has advocated limiting the over-

all percentage of dietary carbohydrate to < 45%, as in

short-term studies this has been shown to improve

glycemic control. The current guidelines approved

by the ADA suggest limiting carbohydrate to 40%

of the total energy content while increasing dietary

fat to 40%. Like others, we would argue that it is the

type rather than the absolute amount of carbohydrate

that dictates postprandial glycemia. The glycemic

index of different ingested carbohydrates quantifies

their different glycemic responses. Starches, fiber, and

refined sugars are the main dietary carbohydrates and

their rates of intestinal absorption are different. The

highest postprandial glucose and insulin values occur

with refined sugars, the most rapidly absorbed, while

the lowest values are with the soluble fibers that

have the slowest absorption rates. In pregnancy, if

the carbohydrate is predominantly refined, any in-

crease above 45% of the total dietary energy content

is associated with a deterioration in glycemic control,

especially if given first thing in the morning. How-

ever, 60% of the total energy content of the diet can

be consumed as low glycemic index carbohydrates

with no change in glucose tolerance or an actual

improvement. Both during and outside pregnancy

the introduction of low-glycemic-index diets has

been shown to reduce insulin sensitivity. There is a

degree of gastric stasis in pregnancy and this itself can

further lower the glycemic index of many carbo-

hydrates.

0025The basis of the dietary management of GDM is to

reduce postprandial glycemia. This theoretically can

PREGNANCY/Nutrition in Diabetic Pregnancy 4757

be achieved either by limiting the total amount of

carbohydrate ingested or insuring that the dietary

carbohydrates consumed have a low glycemic index.

Reducing carbohydrates for breakfast has also been

advocated by some as the postprandial glucose values

tend to be highest mid-morning. However, limiting

carbohydrates will result in a higher proportion of the

diet consumed as fat, which is more energy-dense and

is likely therefore to result in weight gain. In addition,

diets that emphasize limiting the proportion of carbo-

hydrate over fat are, we believe, sending out the

wrong educational message. In practice it is often

difficult to achieve compliance when greater than

50% of the energy content of the diet is given as

carbohydrate. This is especially so with the low-

glycemic-index carbohydrates, which produce greater

satiety than refined sugars or fats.

Dietary Fat

0026 We believe the current American recommendations,

endorsed by the ADA, that 40% of the GDM diet

should be derived from dietary fat, are misguided.

The short-term dietary studies which demonstrated

a benefit of high-fat diets on postprandial blood

glucose values may, as discussed above, have been

accounted for by the high glycemic index of the

carbohydrates used in these studies. In addition, pro-

moting high-fat diets in women destined to develop

diabetes and therefore cardiovascular disease is

highly questionable. High-fat diets have also been

linked with both b-cell toxicity and increased insulin

resistance. There are currently insufficient data to

know if chain length and degree of saturation of

fatty acids are important dietary factors in determin-

ing glucose tolerance during pregnancy.

Diet and Insulin Therapy for GDM

0027 The aim of medical management of women with

GDM is to insure maternal glycemia can be kept at

a safe level. The aim of management is not to delay

insulin treatment if required. The fasting and

postprandial glucose levels at which insulin therapy

should be introduced in addition to diet have been

specified as a fasting value > 5.5 mmol l

1

and 1-h

postprandial value > 7 mmol l

1

. However, the time

that dietary management should be tried before

starting insulin is less well-specified. Even when very

tight glycemic targets are being achieved, most

women can be managed with diet alone. The

women who require insulin are the most metabolic-

ally compromised and have the highest perinatal

complications and the fastest deterioration postpar-

tum for the development of diabetes. It is important

to recognize these women early in pregnancy and not

to lay any sense of failure or noncompliance on them

or to delay the introduction of early insulin treatment.

Insulin is also occasionally introduced later in preg-

nancy for obstetric rather than pure glycemic reasons;

these would include evidence of accelerated fetal

growth and unexplained polyhydramnios.

0028The principles of dietary management for GDM

change little with the introduction of insulin. One

should continue to limit weight gain in obese

women while providing sufficient carbohydrate

snacks throughout the day to prevent hypoglycemia.

0029Whenever insulin is prescribed with the intention

of achieving euglycemia, the frequency of hypo-

glycemia increases. Although short periods of

hypoglycemia are not detrimental to the fetus, it

is both frightening and unpleasant for the woman.

In addition, after an episode of hypoglycemia the

blood glucose rises acutely, both through the

action of the counterregulatory hormones and the

usual subsequent consumption of a sugary drink.

Frequent episodes of hypoglycemia often result in

women chasing these high-rebound glucose levels

by increasing their insulin dosage and this can

result in further hypoglycemic attacks.

0030Women on insulin should have a regular intake of

carbohydrate throughout the day. This needs be taken

with each meal, as well as in a snack between meals

and at bedtime. To avoid unnecessary calorie intake,

snacks that are low in fat should be advised: fruit is

ideal for this purpose. Meanwhile, low-glycemic-

index carbohydrate should be advised for meals.

This strategy will increase the absorption period post-

prandially, thereby minimizing the risk of hypogly-

cemia. The use of the soluble fiber supplement guar

gum can be introduced if blood glucose levels cannot

be controlled adequately without increasing insulin to

levels that are causing hypoglycemia.

Long-Term Dietary Advice for the Mother

and her Child

0031Women with GDM are at greatly increased risk of

future diabetes. Many of these women are obese and

all have at least one child at increased risk of adoles-

cent obesity and diabetes. Given this, it is obvious that

dietary advice, education, and reinforcement should

continue beyond the pregnancy. Women should be

made fully aware of their added risk of both future

GDM pregnancies and type 2 diabetes and the import-

ance of not gaining weight postpartum. Low-fat diets

have been shown to minimize the recurrence of sub-

sequent GDM pregnancies. Suitable healthy living

advice should be given to all women following a

GDM pregnancy, emphasizing the benefits of such a

4758 PREGNANCY/Nutrition in Diabetic Pregnancy

diet for the whole family. Ideally, all these women

would receive annual long-term dietetic input within

the community. The case for continual follow-up of

these women will become more apparent with the

result of the ongoing diabetic prevention studies cur-

rently being carried out among high-risk groups, of

which previous GDM women are included.

The Need and Feasibility of Future Dietary

Studies in Pregnancy

0032 There remains a lack of good randomized studies on

the dietary management of GDM. Such studies are

required for both short-term pregnancy outcomes and

long-term outcomes for the mother and her child.

One of the main difficulties in conducting such stud-

ies is the control arm: even when no dietary advice is

given, once diagnosed with GDM women make life-

style changes based on family beliefs of information

gathered from a variety of sources. Also if the health

care providers are aware of the diagnosis they too

unintentionally are likely to influence lifestyle factors.

The need to blind both the women and the health care

staff of the diagnosis is difficult and often considered

unethical, as, if ignored, GDM does carry a risk to

that pregnancy.

Summary

0033 Gestational diabetes is a common complication of

pregnancy. Controlling maternal hyperglycemia with

diet alone and diet and insulin can reduce the risk of

inappropriate accelerated fetal growth. The pre-

scribed diet for GDM needs to include regular meals

and snacks. These should contain a large component

of slowly absorbed carbohydrate. As many women

with GDM are obese, the diet must not allow exces-

sive maternal weight gain as this further compromises

pregnancy outcome and adds to the mother’s risk of

future diabetes. Dietary advice in pregnancy should

extend beyond pregnancy as a woman with GDM has

a lifetime risk of future diabetes, as does her child.

There remains a need for future dietary studies in the

management of GDM.

See also: Diabetes Mellitus: Etiology; Obesity: Etiology

and Diagnosis; Pregnancy: Metabolic Adaptations and

Nutritional Requirements

Further Reading

berg A, Rydhstro

m H, Kallen B et al. (1997) Impaired

glucose tolerance during pregnancy is associated with

increased fetal mortality in preceding sibs. Acta Obste-

trica Gynecologica Scandinavica 76: 212–217.

Adams KM, Li H, Nelson RL et al. (1998) Sequelae of

unrecognised gestational diabetes. American Journal of

Obstetrics and Gynecology 178: 1321–1332.

American College of Obstetricians and Gynecologists

(1993) Nutrition during pregnancy. ACOG Technical

Bulletin 1–7.

American Diabetes Association (1995) Nutritional manage-

ment. In: Medical Management of Pregnancy Compli-

cated by Diabetes, 2nd edn. Virginia: American Diabetes

Association. pp. 47–56.

Berkowitz GS, Lapinski RH, Wein R et al. (1991) Race/

ethnicity and other risk factors for gestational diabetes.

American Journal of Epidemiology 135: 965–973.

Bianco AT, Smilen SW, Davis Y et al. (1998) Pregnancy out-

come and weight gain recommendations for morbidly

obese women. Obstetrics and Gynecology 91: 97–102.

Buchanan TA, Kjos SL, Montoro MN et al. (1994) Use of

fetal ultrasound to select metabolic therapy for pregnan-

cies complicated by mild gestational diabetes. Diabetes

Care 17: 275–283.

Butte NF, Hopkinson JM, Mehta N et al. (1999) Adjust-

ments in energy expenditure and substrate utilisation

during late pregnancy and lactation. American Journal

of Clinical Nutrition 69: 299–307.

Casey BM, Lucas MJ, McIntire DD et al. (1997) Pregnancy

outcomes in women with gestational diabetes compared

with the general obstetric population. Obstetrics and

Gynecology 90: 869–873.

Catalano PM, Huston L, Amini SB et al. (1999) Longitu-

dinal changes in glucose metabolism during pregnancy

in obese women with normal glucose tolerance and

gestational diabetes mellitus. American Journal of Ob-

stetrics and Gynecology 180: 903–916.

Dabelea D, Pettitt DJ, Hanson RL et al. (1999) Birth weight,

type 2 diabetes, and insulin resistance in Pima Indian

children and young adults. Diabetes Care 22: 944–950.

Dornhorst A and Frost G (1997) The potential for dietary

intervention postpartum in women with gestational

diabetes. Diabetes Care 20: 1635–1637.

Dornhorst A and Rossi M (1998) Risk and prevention of

type 2 diabetes in women with gestational diabetes.

Diabetes Care 21 (suppl. 1): B43–B49.

Durnin JVGA (1987) Energy requirements of pregnancy:

an integrated study in five countries. Background and

methods. Lancet ii: 895–897.

Felig DS and Naylor CD (1998) Eating for two: are guide-

lines for weight gain during pregnancy too liberal?

Lancet 351: 1054–1055.

Freinkel N (1980) Banting lecture 1980: pregnancy and

progeny. Diabetes 29: 1023–1035.

Freinkel N (1985) Effects of the conceptus on maternal

metabolism during pregnancy. In: Leibel BS and

Wrenshall GA (eds) On the Nature and Treatment of

Diabetes, pp. 679–691. Amsterdam: Exerpta Medica.

Harris MI (1988) Gestational diabetes may represent

discovery of pre-existing glucose intolerance. Diabetes

Care 11: 402–411.

Hawdon JM and Aynsley-Green A (1996) Neonatal compli-

cations, including hypoglycaemia. In: Dornhorst A and

Hadden D (eds) Diabetes and Pregnancy: An Inter-

PREGNANCY/Nutrition in Diabetic Pregnancy 4759

national Approach to Management, pp. 303–318.

Chichester, UK: John Wiley.

Hay WWJ (1991) The role of placental–fetal interaction in

fetal nutrition. Seminars in Perinatology 15: 424–433.

Hay WWJ (1994) Placental supply of energy and protein

substrate to the fetus. Acta Paediatrica Supplement 405:

13–19.

Hytten FE (1980) Nutrition. In: Hytten FE and Chamber-

lain G (eds) Clinical Physiology in Obstetrics, pp. 163–

192. Oxford, UK: Blackwell Scientific.

Hytten FE and Leitch I (1964) The Physiology of Human

Pregnancy. Oxford, UK: Blackwell Scientific.

Institute of Medicine (1990) Nutrition During Pregnancy:

Weight Gain and Nutritional Supplements. Washington,

DC: National Academy Press.

Jacobsen JD and Cousin L (1989) A population-based study

of maternal and perinatal outcome in patients with

gestational diabetes. American Journal of Obstetrics

and Gynecology 161: 981–986.

Janerich DT (1990) Alcohol and pregnancy. An epidemio-

logic perspective. Annals of Epidemiology 1: 179–185.

Jovanovic L (1998) American Diabetes Association’s fourth

international workshop-conference on gestational dia-

betes mellitus: summary and discussion: therapeutic

interventions. Diabetes Care 21 (suppl. 2): 131–137.

Kalhan S, Rossi K, Gruca L et al. (1997) Glucose turnover

and gluconeogenesis in human pregnancy. Journal of

Clinical Investigation 100: 1775–1781.

hl C (1991) Insulin secretion and insulin resistance in

pregnancy and GDM. Diabetes 40(suppl. 2): 18–24.

Major CA, Henry MJ, Veciana M et al. (1998) The effects

of carbohydrate restriction in patients with diet-con-

trolled gestational diabetes. Obstetrics and Gynecology

91: 600–604.

McFarland MB, Langer O, Conway DL et al. (1999) Diet-

ary therapy for gestational diabetes: how long is long

enough? Obstetrics Gynecology 93: 978–982.

Metzger BE, Coustan DR and the Organizing Committee

(1998) Summary and recommendations of the fourth

international workshop-conference on gestational dia-

betes mellitus. Diabetes Care 21 (suppl. 2): 161–167.

Moses RG (1996) The recurrence rate of gestational dia-

betes in subsequent pregnancies. Diabetes Care 19:

1349–1356.

Moses RG, Shand JL, Tapsell LC et al. (1997) The recur-

rence of gestational diabetes: could dietary differences in

fat intake be an explanation? Diabetes Care 20: 1647–

1650.

National Research Council (1989) Recommended Dietary

Allowance, pp. 33–34. Washington, DC: National

Academic Press.

O’Sullivan JB (1982) Body weight and subsequent diabetes

mellitus. Journal of the American Medical Association

248: 949–952.

Peterson CM and Jovanovic-Peterson L (1991) Percentage

of carbohydrate and glycaemic response to breakfast,

lunch, and dinner in women with gestational diabetes.

Diabetes 40 (suppl. 2): 172–174.

Pettitt D, Aleck K, Baird HR et al. (1988) Congenital sus-

ceptibility to NIDDM: role of intrauterine environment.

Diabetes 37: 622–628.

Silverman B, Metzger B, Cho NH et al. (1995) Impaired

glucose tolerance in adolescent offspring of diabetic

mothers. Relationship to fetal hyperinsulinism. Diabetes

Care 18: 611–617.

Spellacy WN and Goetz FC (1963) Plasma insulin in normal

late pregnancy. New England Journal of Medicine 268:

988–991.

The MRC Vitamin Research Study Group (1991) Preven-

tion of neural tube defects: the results of the Medical

Research Council Vitamin Study.

Wolever TMS, Bentum-Williams A, Jenkins DJ et al. (1995)

Physiological modulation of plasma free fatty acid con-

centrations by diet. Diabetes Care 18: 962–970.

PREMENSTRUAL SYNDROME: NUTRITIONAL

ASPECTS

F Ford, University of Sheffield, Sheffield, UK

Background

0001 Many women seek medical help for symptoms re-

lated to premenstrual syndrome (PMS). Physicians,

dietitians, and nutritionists often recommend some

form of nutrition advice, including incorporating extra

vitamin B

, vitamin B

, vitamin E, calcium, magne-

sium, essential and o-3 fatty acids, herbal remedies,

carbohydrates, low-fat and vegetarian diets, and

chocolate. Although much (but not all) of this advice

is part of a healthful diet, there is relatively little

sound scientific evidence that they relieve or prevent

symptoms.

0002PMS is a condition characterized by a number of

physical, emotional, and psychological symptoms that

appear in the second half of the menstrual cycle, after

ovulation. It can affect women in the reproductive age

4760 PREMENSTRUAL SYNDROME: NUTRITIONAL ASPECTS

from adolescence until the menopause. There is a list

of menstrual symptoms identifying over 150 different

symptoms linked to the menstrual cycle. The main

symptoms reported are irritability, depression, anx-

iety, weight gain, edema, breast pain, fatigue, and

headache. Greene and Dalton first used the term pre-

menstrual syndrome in 1953. The first author to de-

scribe the syndrome known as premenstrual tension

was Frank, in 1931. The intensity of symptoms may

vary considerably between women and for each cycle.

Severe cyclical symptoms cause suffering to many

women in their daily activity and personal relation-

ships.

0003 PMS is perceived to be a common complaint, but

data on prevalence vary according to different meas-

urement scales and cultural variations. It is estimated

that up to 95% of women suffer mild symptoms, and

5–10% of women have symptoms severe enough to

disrupt their lives in the 2 weeks before the onset

of menstruation. PMS can be defined as a regular

pattern of symptoms occurring just before the start

of menstruation and which gradually abate soon after

the start of bleeding. Severe PMS can be defined as

causing functional impairment in work, relationship,

or usual activities. The social consequences of pre-

menstrual syndrome are important and have been the

focus of much research. It has been reported that PMS

is responsible for a high incidence of crimes, alcohol-

ism, school absence, and admittance to hospital for

accidental injury.

0004 The etiology of PMS is unknown, although many

theories have been suggested. Factors such as estrogen

levels, deficiency of progesterone and progesterone

metabolites, increase of adrenal activity, subclinical

hipoglucosemy, increase of prolactin and monoami-

noxidases, imbalance of renin–angiotensin–aldoster-

one, deficiency of pyridoxine and other vitamins, and

neuroendocrine dysfunction have all been cited.

0005 The diagnosis of the PMS is essentially based on

clinical symptoms. A prospective evaluation of three

consecutive months that show these clinical symp-

toms occurring during the luteal phase of the men-

strual cycles characterize PMS. However, diagnosis of

the condition may be difficult, as there is no stand-

ardized diagnostic tool. Women’s experiences of PMS

are so varied that it is difficult to fit them into cat-

egories or even subgroups of symptoms. In the past,

some doctors have regarded PMS as a psychological

syndrome, so women report finding it difficult to

receive satisfactory treatment and attention. This

might be one of the reasons why it is a condition

that women often diagnose and treat themselves.

0006 Treatment of PMS has been largely empirical. The

following dietary components have been manipulated

for the treatment of PMS.

Vitamin B

(Pyridoxine)

0007Vitamin B

is involved in the production of prosta-

glandin E2 (which contributes to myometrial relax-

ation) and in the utilization of magnesium, so higher

levels of vitamin B

could also influence dysmenor-

rhea cramps. Vitamin B

has been used in the doses of

50–500 mg daily in the second half of the menstrual

cycle. It has been suggested that vitamin B

may act

by correcting a deficiency at the hypothalamic end of

the complex psycho-endocrine reproductive path-

ways. Pyridoxal 5

phosphate, the active form of vita-

min B

, serves as the coenzyme of a wide variety of

enzymes of amino acid metabolism. For example, it

serves as a cofactor in the metabolism of tryptophan

(the precursor of serotonin) and also in the metabol-

ism of tyrosine (leading to dopamine and noradren-

aline) and glutamate (leading to g-aminobutyric acid).

Low levels of dopamine and serotonin lead to high

levels of prolactin and aldosterone, thus explaining

the fluid retention, and the effect on the neurotrans-

mitters could explain the psychological symptoms in

PMS.

0008In recent years, a systematic review on the efficacy

of vitamin B

in the treatment of PMS concluded that:

0009Randomized placebo controlled studies of vitamin

treatment for PMS were of insufficient quality to

draw definitive conclusions.

0010Limited evidence exists to suggest that 100 mg of

vitamin B

daily (and possibly 50 mg) is likely to be

beneficial in the management of PMS.

0011Vitamin B

was significantly better than placebo in

relieving overall premenstrual symptoms and in

relieving depression associated with PMS, but the

response was not dose-dependent.

0012No conclusive evidence was found of neurological

side-effects with these doses.

0013A randomized controlled trial of sufficient power

and quality is needed to compare vitamin B

with

placebo to establish definitive recommendations

for treatment.

A systematic review published in the Cochrane Data-

base concluded that no definite results could be

reported regarding the efficacy of vitamin B

improving dysmenorrhea.

Calcium

0014Previous reports have suggested that disturbances in

calcium regulation may underlie the pathophysiologic

characteristics of PMS and that calcium supplemen-

tation may be an effective therapeutic approach. In one

study, each participant received 6 months of treatment,

involving 3 months of daily calcium supplementation

PREMENSTRUAL SYNDROME: NUTRITIONAL ASPECTS 4761

(1000 mg of calcium carbonate) and 3 months of

placebo. Three premenstrual factors (negative

affect (p ¼0.05); water retention (p ¼0.005); pain

(p ¼0.05) ) and one menstrual factor (pain

(p ¼0.05) ) were significantly alleviated by calcium.

0015 Another study evaluated the effect of 1200 mg of

elemental calcium on the luteal and menstrual phases

of the menstrual cycle in 497 women with PMS, in

a prospective, randomized, double-blind, placebo-

controlled, multicenter clinical trial. During the luteal

phase of the treatment cycle, a significantly lower

mean symptom complex score was observed in the

calcium-treated group for both the second (p ¼

0.005) and third (p < 0.001) treatment cycles. By the

third treatment cycle, calcium effectively resulted in

an overall 48% reduction in total symptom scores

from baseline compared with a 30% reduction in

placebo. All four-symptom factors were significantly

reduced by the third treatment cycle.

Magnesium

0016 Results from a recent metaanalysis have suggested

that magnesium supplements of 500 mg per day may

be a promising treatment for one of the major symp-

toms of PMS, i.e., dysmenorrhea. Magnesium treat-

ment has been shown to be more effective than

placebo, with those taking magnesium requiring

significantly less absence from work and less use of

additional medication. Minimal adverse effects have

been experienced in both magnesium and placebo

groups with no significant differences shown.

However, overall, there has been little conclusive evi-

dence for the effectiveness of magnesium for symp-

tom relief. Although all the trials have been

randomized and double blind with adequate meth-

odological quality, they were all small trials, and there

was poor measurement and reporting of pain out-

comes. Two of the included trials had withdrawal

rates of over 30%. Many of these participants may

have failed to complete the trial due to a lack of

efficacy or adverse effects from the treatments, and

so the high withdrawal rates have an impact on the

strength of the evidence. It is also unclear which

treatment regimen was more effective, as the trials

both administered treatment in quite different ways,

one daily and the other during menses only, and at

different doses. Therefore, no strong recommenda-

tion can be made about the efficacy of magnesium

until a further evaluation is carried out. A random-

ized controlled trial of magnesium with a larger

number of participants and adequate outcome meas-

urement is needed.

0017 The largest of the included trials also reported data

on the levels of prostaglandin F2 a in menstrual

blood. Overproduction of this prostaglandin has been

shown to be a substantial contributing factor to the

painful cramps associated with dysmenorrhea.

Women taking the magnesium therapy had substan-

tially lower levels of PGF2 a in their menstrual

blood than those on placebo (p < 0.05), which

mirrored the therapeutic decrease in pain experienced

by the participants. This highlights the possible bio-

logical rationale behind magnesium therapy for

dysmenorrhea; it inhibits the biosynthesis of PGF2 a

as well as having a role in muscle relaxation and

vasodilation.

Vitamin B

and Magnesium

0018Magnesium was shown to be no different in pain

outcomes from both vitamin B

and a combination

of vitamin B

and magnesium by one small trial. The

same trial also showed that a combination of magne-

sium and vitamin B

was no different from placebo

in reducing pain. However, vitamin B

alone was

more effective at reducing pain than both placebo

and a combination of magnesium and vitamin B

.In

a recent systematic review, no definite conclusions

could be made about the efficacy of the combination

of vitamin B

and magnesium supplements.

Vitamin B

(Thiamin)

0019Vitamin B

plays an important role in metabolism,

and vitamin B

deficiency can be characterized by

fatigue, muscle cramps, various pains, and a reduced

tolerance to pain, all factors that could be associated

with dysmenorrhea and PMS. Vitamin B

was

shown to be an effective treatment for dysmenorrhea,

taken at 100 mg daily, although this conclusion is

tempered slightly because it is based on only one

large randomized controlled trial (RCT). This RCT

trial demonstrated a significant effect of vitamin B

taken daily compared with placebo in reducing

pain. Results from phase one of the trial only,

before participants were crossed over to the other

treatment group, were used when calculating odds

ratios. After crossover, those swapped from vitamin

treatment to placebo maintained pain relief.

The authors of the trial interpreted this as a curative

effect of vitamin B

treatment. The length of the

trial (2 months of either treatment) makes this inter-

pretation difficult to confirm, as no longer-term

follow-up was made, and the strong placebo effect

that is typical in dysmenorrhea trials could account

for some of the maintenance of effect. Vitamin B

needs to be assessed further by a randomized trial,

preferably a large multicentered trial with different

study populations that would generate generalizable

4762 PREMENSTRUAL SYNDROME: NUTRITIONAL ASPECTS

results that could confirm the positive results of the

above trial.

Vitamin E

0020 It has been suggested that vitamin E has analgesic and

antiinflammatory properties. A randomized trial of

vitamin E for rheumatoid arthritis has shown a sig-

nificant reduction in pain parameters, which lends

further support to this theory. One trial comparing a

combination of 100 mg per day of vitamin E and

ibuprofen with ibuprofen alone on dysmenorrhea

showed no significant difference between the two

groups for the outcome of pain relief after 1 month

of treatment. Both treatment groups experienced

excellent treatment responsiveness, although this

conclusion is based on a comparison of baseline

scores with post-treatment scores, which is more

subject to bias than if a placebo control group had

been used. Overall, the addition of vitamin E supple-

ments to the diet has no further effect on pain over

that of standard treatment with ibuprofen. The treat-

ment time of 1 month of each treatment was a limita-

tion of the trial, as the efficacy of a dietary therapy

cannot really be assessed adequately over such a short

period of time.

Essential Fatty Acids

0021 Findings from uncontrolled studies have indicated

that essential fatty acid (EFA) metabolism may be

abnormal in women with PMS. EFAs are the precur-

sors of prostaglandins. Low levels of prostaglandins

are thought to lead to increased sensitivity to prolac-

tin. Women with PMS may be abnormally sensitive to

normal amounts of prolactin. This hypothesis – that

there is a deficiency in EFAs, leading to low levels of

prostaglandin E1 that may attenuate the effects of

prolactin – is one rationale for using EFAs as a treat-

ment for women with PMS.

0022 Evening primrose oil (EPO) contains two EFAs –

linoleic acid and g-linolenic acid (GLA). Linolenic

acid is needed for the synthesis of prostaglandin E,

and GLA is needed for the synthesis of prostaglandin

E1. Whilst EPO is generally accepted as a safe prod-

uct, reported side-effects include occasional nausea,

indigestion, and headache. A small number of less

common side-effects have also been documented. A

potential risk of inflammation, thrombosis, and

immunosuppression with prolonged use of GLA has

been described.

0023 EPO is available in most countries without a pre-

scription and is heavily promoted for women as being

an effective treatment for a range of conditions in-

cluding PMS. It is, however, an expensive treatment

that many women have to pay for themselves. In a

systematic literature search of clinical trials of EPO

for the treatment of PMS with a view to performing a

meta-analysis, only seven placebo-controlled trials

have been found, and randomization was clearly

indicated in only five trials. Inconsistent scoring and

response criteria made statistical pooling and hence a

rigorous metaanalysis inappropriate. The two most

well-controlled studies failed to show any beneficial

effects for EPO, although, because the trials were

relatively small, modest effects cannot be excluded.

From the current evidence, EPO is of little value in the

management of PMS.

v-3 Fatty Acids

0024One trial has compared o-3 fatty acids (fish oil) and

placebo, and data have shown that the treatment

is significantly more effective than placebo after 2

months’ administration. A meta-analysis has exam-

ined the outcome of the use of additional medication

and showed that the fish oil treatment group con-

sumed significantly fewer tablets than the placebo

group. Minimal adverse effects were significantly

more likely in the fish oil group compared with the

placebo group. None of the adverse effects were par-

ticularly serious (nausea, acne exacerbation, and dif-

ficulty swallowing capsules), but they were acute

enough to cause some women to discontinue treat-

ment. One methodological limitation of this trial is

the poor reporting of data. Pain relief was reported,

as the average of the two groups after the treatments

allocated was crossed-over, and so it is not an ad-

equate assessment of efficacy. The trial was of short

duration, only 2 months in each treatment arm,

which may not be enough time to properly assess

the effect of a dietary intervention.

0025Levels of polyunsaturated fatty acids (PUFAs) have

been correlated with menstrual pain, with higher

levels of the o-3 fatty acids associated with milder

menstrual symptoms. The PUFA o-6 is metabolized

into the specific prostaglandins associated with dys-

menorrhea, and it appears that the ratio of o-3 to o-6

is associated with menstrual symptoms, therefore a

diet higher in o-3 fatty acids is possibly associated

with less dysmenorrhea.

Herbal Remedies

0026It is advisable that women consult a qualified medical

practitioner who specializes in complementary medi-

cine before taking unproven herbal remedies. There is

a Practitioner Directory, which lists a selection of the

best doctors and practitioners who specialize in treat-

ment using complementary medicine.

PREMENSTRUAL SYNDROME: NUTRITIONAL ASPECTS 4763

Japanese Herbal Combination

(Toki-Shakuyaku-San)

0027 The combination herbal remedy Toki-shakuyaku-san

(TSS) has been compared with placebo in one trial.

Pain was assessed using a visual analog scale (VAS).

TSS was significantly more effective in reducing VAS

scores, after 2 months of treatment. This difference

was maintained after a 2-month untreated followup

period. The use of additional medication in the treat-

ment group was significantly less than in the placebo

group. One major limitation of this trial was that

participants were included in the trial using a particu-

lar traditional Chinese medicine diagnosis of a com-

plex set of symptoms as well as the typical diagnosis

of PMS. It is not clear how this would translate into

Western therapeutics, or if this remedy would help

women without this particular pattern of diagnosis.

Agnus Castus

0028 Agnus Castus has been reported to be a hormone

regulator and tonic for the nervous system, being used

for PMS to treat mood swings, depression, water

retention, and breast pain. The efficacy and tolerabil-

ity of agnus castus fruit (Vitex agnus castus L extract

Ze 440) with placebo has been investigated in women

with PMS in a prospective, randomized, placebo con-

trolled study. The main efficacy variable was a change

from baseline to end point (end of the third cycle) in

women’s self-assessment of irritability, mood alter-

ation, anger, headache, breast fullness, and other

menstrual symptoms including bloating. Results

have shown a significant improvement in the active

group compared with the placebo group (p < 0.001).

Some women have reported mild adverse events,

none of which caused discontinuation of treatment.

It has been concluded that a dry extract of agnus

castus fruit is an effective and well-tolerated treat-

ment for the relief of symptoms of PMS.

Hypericum perforatum

0029 A pilot observational study has been carried out to

investigate whether Hypericum perforatum could

relieve the symptoms of PMS in a small group of

women. Participants took hypericum tablets for two

complete menstrual cycles (1 300 mg of hypericum

extract per day standardized to 900 mg of hypericin).

The degree of improvement in overall PMS scores

between baseline and the end of the trial was 51%,

with over two-thirds of the sample demonstrating at

least a 50% decrease in symptom severity. The results

of this pilot study suggest that there is scope for

conducting a RCT to investigate the value of hyperi-

cum as a treatment for PMS.

St John’s Wort

0030St John’s Wort has been reported to raise serotonin

levels, which helps to alleviate the mild to moderate

depression associated with PMS. It should not be

taken with prescription antidepressants, progester-

one, or other hormone treatments.

Dong Quai – (

Angelica sinensis

) or Chinese

Angelica

0031Dong Quai (Angelica sinensis), or Chinese Angelica,

is sometimes known as ‘women’s ginseng‘ and is used

for PMS.

Natural Progesterone

0032Natural progesterone is sometimes prescribed where

low progesterone levels are indicated in PMS. Over-

all, there is insufficient evidence to recommend the

use of any herbal remedies apart from Agnus castus

for the treatment of PMS. The Complete German

Commission E Monographs list a number of herbal

products as being used for dysmenorrhea or men-

strual disorders, yet no evidence from clinical trials

to support the use of any of these herbal products has

been found to date.

Chocolate

0033Chocolate’s ingredients satisfy many aspects of PMS.

Its high sugar content induces the brain to make new

serotonin and make the chocolate eater feel calm and

relaxed. Moreover, chocolate is usually eaten by itself

rather than as part of a meal. Thus, there is little

chance that protein foods will interfere with the

body’s ability to make new serotonin. The high but-

terfat content of chocolate may also change premen-

strual mood. High-fat foods can have a numbing

effect on mood and on mental and physical energy,

and the combination of the fat and sugar content of

the chocolate is reported to soothe and tranquilize

women. However, chocolate has other ingredients –

like phenylethylamine – that are also claimed to have

positive mood effects. There are no reliable studies

proving this, but anecdotal evidence supports the

view that chocolate may have some ingredients that

women with PMS crave.

Carbohydrates

0034The effect of a carbohydrate-rich beverage on mood,

appetite, and cognitive function was investigated in a

small group of women with PMS. Twenty-four

women with confirmed PMS took a beverage or

placebo for three menstrual cycles after a 1-month

4764 PREMENSTRUAL SYNDROME: NUTRITIONAL ASPECTS