Article Text

Fetal and infant markers of adult heart diseases
  1. Marjo-Riitta Järvelin
  1. Department of Public Health Science and General Practice, University of Oulu, Finland and Department of Epidemiology and Public Health, Imperial College School of Medicine, London, UK
  1. Professor Marjo-Riitta Järvelin, Department of Epidemiology and Public Health, Imperial College School of Medicine, Norfolk Place, London W2 1PG, UK email:m.jarvelin{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

There is growing evidence of an increasingly complex and multifactorial aetiology of heart diseases.1 w1 It seems likely that the large geographic variations in cardiovascular disease (CVD) morbidity and mortality,w2 even though at least partly genetic in origin, are influenced by factors acting prenatally and in early life, or by a combination of factors present throughout the life course. Changes in fetal growth pattern have been related to adult disease risk,1 and there are many theories about the underlying mechanisms affecting cell division during critical periods of tissue development. The critical periods vary according to the tissue in question, and that is why there have been attempts to explore the timing of exposure in order to predict more specifically the adult disease risk.

This article examines: firstly the historical evolution of theories on childhood factors which have an influence in adulthood; secondly what is known today about the effect of early life factors on heart disease risk; and thirdly the specific problems in longitudinal studies which explore these factors and adult disease risk.

Dawn of the “hypothesis of the 20th century”

Biological programming: a new theoretical model about the aetiology of heart disease

The dawn of modern epidemiology came after the second world war, first with ecological studies comparing CVD incidence and mortality, and subsequently multicentre cross sectional and follow up studies on CVD.w3 The studies showed that populations with high CVD mortality have high cholesterol and high blood pressure, and that smoking and obesity are common among these populations.w4This led to the lifestyle model in understanding the aetiology of chronic diseases, where the key issues are health behaviour and the interaction between genes and an adverse environment in adult life. This was consequently followed by intervention programmes, which have significantly improved heart disease risk status in many countries.w3 However, lifestyle factors only explain part of the heart disease risk, which is why other reasons have been sought. For example, in the mid 1980s Rose pointed out that the well established risk factors for coronary heart disease (CHD)—cigarette smoking, high serum cholesterol, and high blood pressure—have a limited ability to predict disease risk in adults.w5 In the large international MONICA (monitoring trends and determinants in cardiovascular disease) project,w2 w4 only 25% of the variance in CHD mortality was explained by conventional risk factors. Could childhood influences explain this gap in our understanding of the aetiology of CVD?

In Norway in the 1970s, Forsdahl2 put forward the hypothesis that the geographical differences in CVD mortality might not be related to the contemporary circumstances, but to poverty or deprivation in early life (table 1). However, the importance of fetal and early life circumstances for adult health had been suggested almost a century earlier by the chief medical officer to the Board of Education in Britain, who wrote: “recent progress has shown that the health of the adult is dependent upon the health of the child and that the health of the child is dependent upon the health of the infant and its mother”.w6

Table 1

Early fetal origin hypotheses developing studies

A new hypothesis developed following observations in the 1980s by Barker and colleagues, in accordance with Forsdahl, based upon positive relations between the areas with the highest CVD and infant mortality rates,3 and lower birth weight and increased risk of CVD mortality4 (table 1). These historical cohort studies3-5 w7 w8 and evidence from animal experiments1 w9 suggest that chronic diseases are biologically “programmed” in utero or in early infancy. Programming is the process where a stimulus or insult (for example, undernutrition, hormones, antigens, drugs or sensory stimuli) at a critical period of development induces long lasting changes in cells which in turn changes the structure or function of organs, tissues or body systems.w7 w10 In the case of heart disease, it is hypothesised that fetal undernutrition during middle gestation in particular raises the risk of later disease by the programming of blood pressure, cholesterol metabolism, blood coagulation, and hormonal settings.5 Consequently, it was suggested that the lifestyle model in the evolution of adult degenerative diseases needs to be replaced by a new model, the central feature of which is the concept of biological programming in fetal and infant life. This revolutionary model of the 20th century has received both an enthusiastic and sceptical response. Critical testing of this model is warranted owing to inevitable biases related to historical studies.

Social programming and adult diseases

During the past 10 years sociomedical research has pointed out the importance of social differences between countries and populations in explaining differences in health. This ideology has created a social programming model in parallel to the biological programming model.6 Social programming means that the effect of the early social environment on health is mediated by the social environment and school achievement during growth, and by employment opportunities, living conditions, and lifestyle factors. The social programming model is supported by various studies showing an independent effect of childhood social circumstances on adult health.7 w11

Evidence for an association between childhood factors and heart disease risk

Heart disease morbidity and mortality

The first studies reporting an association between birth weight and CHD came from Hertfordshire and Sheffield study populations.4 ,8 Both in men and women—even though the relation was weaker in women9—CHD mortality decreased progressively with increasing birth weight. Since then there have been several, mainly retrospective cohort studies which have replicated these observations and also demonstrated the association between size at birth and non-fatal CHD.w12 w13 To date, there have been over 400 papers published during the past 15 years dealing with prenatal and early life factors related to CVD mortality and disease risk.

The association between birth weight and disease outcomes is, with few exceptions,10 consistent with data based upon the older generations born in the early 1920s or 1930s from different countries. However, it is not known how these observations apply to younger generations assuming that younger generations must have had better nutritional status in early life. The historical cohorts on which these observations are mainly based are liable to bias owing to selective survival and availability of data records.

Early life factors and intermediate heart disease risk factors/conditions

The associations between markers of fetal growth and intermediate risk factors are less consistent than evidence for morbidity and mortality. These include birth measures in relation to plasma concentrations of cholesterol, apolipoprotein B,w14 and fibrinogen,11 blood pressure,12 ,13 and liability to impaired glucose tolerance and diabetes.14-16

Blood pressure has been suggested as one link between the intrauterine environment and the risk of CVD. Baker and colleagues studied the correlation between birth weight and subsequent blood pressure in three adult populations in Hertfordshire, Preston, and Sheffield in the UKw15 as well as in children of different ages.4 ,17 w16 Other studies replicating Barker's have been made on various child populations.18 ,19 w17–19 The key findings include an inverse independent relation between birth weight and subsequent systolic blood pressure, amplified by age,12 ,18 ,19 w17–19 and an association of lower birth weight and thinness at birth with an increased risk of insulin resistance,16 w20 w21 which is an important risk factor for heart diseases. Observations are not consistent; weak, non-linear or insignificant correlations between birth weight and blood pressure have been reported,20 w22 particularly among younger populations.

A correlation between possible undernutrition and serum cholesterol has been noted in men and women in some studies,w14 w21 but there are also studies which show no relation.w23 The association between body length at birth and cholesterol might reflect abnormal intrauterine growth, in which retarded trunk and visceral growth is associated with alterations in lipid metabolism. Abdominal circumference at birth, which reflects visceral growth, has been related to serum cholesterol concentration in adults.5

Theoretical models on the evolution of chronic disease

  • Lifestyle model in the 1960s-70s

  • Biological programming in fetal and infant life model in 1980s-90s

  • Social programming model in the 1990s

  • Life course model in 2000, incorporating both biological and social environments, and their interactions

Lower birth weight and weight at 1 year of age have been associated with subsequent development of type 2 diabetes mellitus in adult life. In the Hertfordshire study, the men with impaired glucose tolerance and diabetes had lower weight gain prenatally and during infancy than men without.14 The plasma 32-33 split proinsulin concentration fell with increasing weight at 1 year. All the findings were independent of current body mass index (BMI).14 In the Preston study, impaired glucose tolerance was also related to lower birth weight and smaller head circumference.21 Gestational age had no influence on the results. A follow up study of 297 women aged 60–71 years suggests, in accordance with previous studies, that those who had lower birth weight had higher plasma concentrations of glucose and insulin.w21 Obesity in adult life adds to the disadvantage of low birth weight; the women who were light at birth but are currently obese have the least favourable risk factor profile.w21 A longitudinal study of diabetes and its complications conducted among the American Indian population in Arizona, however, showed the prevalence of non-insulin dependent diabetes mellitus to be greatest not only in those with the lowest birth weights, but also in those with the highest birth weights.22 This study is supported by a study on Mexican American families.23

Patients with type 2 diabetes and hypertension often have other abnormalities, such as high plasma insulin concentrations, high serum triglyceride concentrations, low serum HDL (high density lipoprotein) concentrations, and high body mass indices and waist-to-hip ratios. This combination of abnormalities has been called syndrome X or “small baby syndrome”,w24 but may be better known as insulin resistance or metabolic syndrome. Metabolic syndrome is characterised by compensatory hyperinsulinaemiaw24 and is associated with increased mortality from CHD.w25 The association of both type 2 diabetes and hypertension with reduced fetal growth has raised the possibility that these and other components of the syndrome may have a common origin in suboptimal development at a particular stage of intrauterine life.14 ,21 In the Preston study,21 the prevalence of metabolic syndrome in both men and women decreased progressively as their birth weights increased. The association between metabolic syndrome and low birth weight was independent of gestational age and possible confounding variables, including cigarette smoking, alcohol consumption, and social class currently or at birth.

Several reports, however, have been more equivocal about the relation of birth related factors to CVD and its risks, particularly studies in adolescents and young adults, and the authors have questioned the basis and rationale for these associations and the underlying mechanisms.24 ,25 w22 w23 w26–30

The main associations between birth weight and other growth measures and heart disease are summarised in table 2.

Table 2

Summary of the main associations between birth weight and other growth measures and heart disease

Suggested biological/environmental mechanisms underlying the evolution of heart disease risk

Nutritional factors during pregnancy

There are numerous factors and mechanisms which affect both fetal growth,w31–33 and adult CVD outcomes,w4 w34which makes the analyses of the associations and their interpretation extremely complex (fig 1). Among them, in the light of early programming, are: (1) restricted maternal nutrition itself; and (2) maternal or pregnancy induced physiological, metabolic or hormone related conditions which may impair fetal nutrition or otherwise affect growth.

Figure 1

Intrauterine programming by prenatal determinants and life course factors in heart diseases (GF, growth factor).

A primary fetal origin hypothesis from the early 1990s stated that adult disease such as CVD is programmed by poor maternal nutrition during pregnancy, leading to fetal growth retardation and a permanent effect on the body's structure, physiology, and metabolism.5 w8 Based on rodent experiments and human studies it nowadays also covers other mechanisms.

Maternal nutrition

In rodents, dietary changes during gestation induce not only growth retardation but also permanent changes in metabolismw35 which can be transmitted through several generations.w36 Though well supported by animal studies,26 the evidence for similar processes in humans is patchy and complex.w23 w37 Among indicators ofmaternal nutrition in humans, low pre-pregnancy weight, height, and BMI are associated with lower birth weight.w31 w38 which in itself is associated with heart disease risk.w30 However, in men born in the 1920s and '30s, high maternal BMI together with low ponderal index was associated with their offspring's highest standardised mortality ratio for CHD. One explanation for this contradictory finding may be that, as suggested by animal studies,w39 the mothers themselves may have been smaller at birth and, as a result, accumulated more fat. Maternal height, reflecting long term nutrition, may be an even better indicator of disturbed long term nutrition than weight in relatively well nourished populations. For example, Forsen and colleagues reported that offspring of short, heavy mothers have higher rates of CHD than those of taller women.27 Small studies in humans, directly examining nutritional intake, suggested that women who have a high intake of carbohydrates in early pregnancy and a low intake of dairy protein in late pregnancy tend to have infants who are thin at birth.w40 w41

Fetal nutrition

Other indicators of possible disturbed fetal nutrition not directly related to maternal nutrition (for example, pregnancy induced hypertension, pre-eclampsia)w33 w42 have rarely been studied in relation to adult disease risk in humans. Evidence that hypertension during pregnancy in humans affects adult CVD risk is inconsistent,w43–47 although animal data are supportive.26 One difficulty, to date, has been separating pregnancy induced hypertension from essential hypertension because few studies record blood pressure measurements during pregnancy, at least not during early pregnancy, or present data on pre-pregnancy hypertension. High maternal blood pressure has, however, been associated with low birth weight of offspring,w31 w42 which in itself is associated with high blood pressure in adult life, but it is unclear to what extent this reflects maternofetal undernutrition during pregnancy or genetic factors.

Growth patterns

The growth of the fetus is a complex process which is still insufficiently understood. A key concept in the “fetal origin hypothesis” is fetal undernutrition, and its relation with adult diseases. The human evidence, as described above, is based on studies where birth measures have been related to different adult heart disease outcomes in different populations. This is strongly supported by the animal experiments, and stresses the importance of the fetomaternal environment. Barker5 has differentiated undernutrition during pregnancy by trimesters, and he suggests that the down regulation of growth during the first trimester leads to a proportionately small child who has increased risk of raised blood pressure and may possibly die of haemorrhagic stroke. Undernutrition during the second trimester leads to a disturbed fetoplacental relation, and insulin resistance or deficiency; consequently birth weight is reduced and the baby is thin, and has an increased risk of raised blood pressure, non-insulin dependent diabetes, and death from CHD. Undernourished babies during the last trimester in turn may have growth hormone resistance or deficiency, and consequently they are short but birth weight is within the normal range. These adults may have raised blood pressure, raised LDL (low density lipoprotein) cholesterol concentration, and increased risk of CHD and thrombotic stroke.

Later growth patterns, particularly catch-up growth,w48have been reported to relate to heart disease risk. For example, children who are thin at birth but become obese in later life or have high catch-up growth in infancyw48 appear to be at higher risk. However, it is not known why catch-up growth is detrimental, but one possibility is that fetal growth restriction leads to reduced cell numbers, and subsequent catch-up growth is achieved by overgrowth of a limited cell mass.

Hormonal evidence related to fetal growth and later heart disease risk

Fetal growth is also affected by several hormones, growth factors, and genetic factors (fig 1). A recently proposed underlying mechanism, based mainly on animal studies, suggests that increased blood pressure in adult life is caused by increased exposure to corticosteroids during fetal life. This might result from reduced placental 11β-hydroxysteroid dehydrogenase (11β-OHSD) activity or increased corticosteroid release secondary to disturbed nutrition.w9 w49–51 Increased exposure in turn may lead to permanent tissue damage, and programming of adult disease.1 w52There are data supporting similar mechanisms in humans—for example, studies have found that birth weight is correlated with placental 11 β-OHSD activity,w50 and cortisol concentrations in adult life correlate with birth weightw53 and adult blood pressure. w54

Insulin and insulin-like growth factors are likely to have a substantial influence on fetal growth. Insulin stimulates growth through several mechanisms: by increasing uptake and utilisation of nutrients; by direct mitogenic actions; and by increasing the release of other hormones and growth factors.w55 However, the final role of these factors in the evolution of adult disease risk is largely unknown, although it can be speculated that via the effects on fetal growth the disturbances in the regulation of these factors lead to increased risk of adult chronic diseases.

Genetic evidence

The role of genetic factors is poorly understood even though a familial aggregation of CHD and hypertension is clear. A complementary explanation for the observed associations between fetal growth and adult phenotypes could be provided by genomic variation which alters the function and/or regulation of genes influencing both phenotypes. Recently the first small genetic studies have been published which stress the importance of possible gene–environmental interaction.w56 w57 Disturbances or variations in genes which regulate either insulin or glucocorticoid action or metabolism may reduce birth weightw58 and thus possibly increase the risk of insulin resistance in adulthood. In Mexican American families, Stern and colleagues23 dissected the relation between birth weight and adult insulin resistance into two components: (1) a sporadic, environmental association betweenlow birth weight and adult insulin resistance; and (2) a genetic association betweenhigh birth weight and adult insulin resistance. This is in agreement with the studies suggesting non-linear association between birth weight and impaired glucose tolerance.22 There is a debate over whether these effects/associations are truly genetic or whether they are caused by the environment—that is, phenotypic. A future challenge is to determine the relative contributions of genesand environmental factors to the fetal and adult phenotypes.

Other possible models in the evolution of heart diseases and limitations of the studies

In Europe there are more than 20 large longitudinal studies in which the main focus has been or is to study prenatal or early life factors in relation to adult disease risk. Many of them are historical cohort studies, or data collection has started after birth retrospectively at various points of life. The most important historical cohort studies, from the point of view of the fetal origin hypothesis, are the Hertfordshire,4 ,14Preston,12 ,21 and Sheffield8 studies, as well as the Helsinki27 and Uppsala28 cohort studies.

The studies to date have had a number of important limitations that complicate interpretation. They have not been able to address the complexities of interactions between environmental and genetic factors in explaining the associations between maternal, fetal, and later life factors in the evolution of adult CVD risk. This is because they have been variously too small; retrospective and therefore subject to survival and selection biases; or prospective, but in children and adolescence and therefore have not been able to examine adult phenotypes. It has also been questioned whether a study with a completely different a priori hypothesis should be used at all for other purposes. However, the use of old data for studying early life factors is justified considering the latency between early exposure and adult outcomes. For example, Barker's studies based on early last century cohorts have been extremely valuable hypotheses developing studies, which should now be replicated in younger cohorts reaching adult age.

An important consideration and future challenge to explore from the point of view of the biological programming model is the extent to which associations between the fetal environment and adult health may be confounded by or interact with measures taken later in life.20 w1 w29 For example, adult weight and height have been reported to be stronger predictors of blood pressure than birth measures,w47 but observations from different studies are inconsistent. A further question concerns the relative influence of childhood and adult measures of socioeconomic status, and health behaviour. Several studies report a powerful association between markers of social status or wealth in childhood or adulthood, and the risk of adult chronic diseases and mortality.18 w59 The risk of premature death from CVD appears to be particularly sensitive to socioeconomic influences acting in early life,w60 but the results from different studies vary.18 ,28 w61–63 A recent review of the influence of early-life socioeconomic environment on the risk of adult disease concluded that both early-life and later circumstances are important.7

Figure 1 shows a simplified framework for the different associations between the various factors in the prenatal period and their effect on adult health. It is evident that no single model is able to explain heart disease risk. This is mainly because there is a vast amount of evidence that: (1) socioeconomic and living circumstances have an independent effect on adult health; (2) health behaviour affects disease riskw3; (3) genetic factors may have an important role in the programming process and possible gene–environment influences; and (4) the impact of chain effects and clustering of disadvantageous factors on disease risk. The clustering effect differs from programming in that it does not expect necessarily to take into account any critical period. It has been questioned if “critical periods” should be taken into account not only during fetal life but later over the life course.

It is reasonable to assume that early programming is a result of an interaction between fetomaternal environment and individual genotype. The “inborn” predisposition to later disease is in turn modified by factors along the life course. The variate of social and biological programming, the multidisciplinary life course model provides an alternative way of exploring the association between early life environment, both social and biological, and adult disease risk. This approach points out that there is a clear need to establish studies by assembling cohorts where measures of pre- and postnatal determinants have been previously recorded in different populations living under different conditions in order to explore pathways and mechanisms in the evolution of heart diseases.

Early life factors and adult heart disease risk: summary

  • A number of factors throughout the life course affect adult disease risk, starting in utero

  • A number of studies show that fetal growth is related to adult heart disease mortality, morbidity, and risk factors

  • Several factors affect fetal growth and subsequently may contribute to adult disease risk

  • There are only a few studies in humans with extensive life course data to explore the association between prenatal and infancy exposures and adult disease or risk outcomes

  • We do not know the mechanisms by which the observed associations are evoked or mediated in humans, or whether the same relations apply to older and younger cohorts


Inconsistencies between and within studies exist, and relations of varying degrees of strength have been described. With the available evidence of the relation between early life factors, intermediate CVD risk factors, disease incidence, and mortality, it remains unclear whether the associations are primarily a manifestation of intrauterine programming of CVD risk due to poor maternal nutrition itself or other influences in utero unrelated to maternal undernutrition, such as defective placentation, and hypertension or other aspects of the genetic, metabolic or circulatory milieu. Studies need to address the extent to which fetal environment and early life experiences act on adult health through independent or intermediary mechanisms, and the extent to which the associations between birth variables and disease risk are independent of later social environment and living habits.


Supplementary materials

  • Additional references for "Fetal and infant markers of adult heart diseases" by Marjo-Riitta Järvelin. Heart 2000;84:219-26.
    1. Kuh D, Ben-Shlomo Y. A life course approach to chronic disease epidemiology. Oxford University Press, 1997.
    2. WHO. World health statistics annual 1989. Geneva: World Health Organization, 1989.
    3. Dobson A, Evans A, Ferrario M, et al. Changes in estimated coronary risk in the 1980s: data from 38 populations in the WHO MONICA project. Ann Med 1998:30:199-205.
    4. The World Health Organization MONICA Project. Ecological analysis of the association between mortality and major risk factors of cardiovascular disease. Int J Epidemiol 1994;94:705-16.
    5. Rose G. Sick individuals and sick populations. Int J Epidemiol 1985;14:32-8.
    6. Anon. Annual report for 1913 of Chief Medical Officer of the Board of Education, Cd 7330. London: HMSO, 1914.
    7. Barker DJP. Fetal and infant origins of adult disease. London: BMJ Publishing Group, 1992.
    8. Barker DJP. Mothers, babies, and disease in later life. London: BMJ Publishing Group, 1994.
    9. Edwards CRW, Benediktsson R, Lindsay RS, et al. 11β-hydroxysteroid dehydrogenase: key enzymes in determining tissue-specific glucocorticoid effects. Steroids 1996;61:263-9.
    10. Lucas A. Programming by early nutrition in man. In: Bock GR, Whelan J, eds. The childhood environment and adult disease. Chichester: John Whiley and Sons; 1991:38-55.
    11. Van de Mheen HD, Stronks K, Mackenbach JP. A lifecourse perspective on socio-economic inequalities in health. The influence of childhood socio-economic conditions and selection processes. In: Bartley M, Blane D, Smith GD, eds. The sociology health inequalities. Sociology of health and illness monograph series. London, 1998.
    12. Frankel S, Elwood P, Sweetman P, et al. Birthweight, body-mass index in middle age, and incident coronary heart disease. Lancet 1996;348:1478-80.
    13. Frankel S, Elwood P, Sweetman P, et al. Birthweight, adult risk factors and incident coronary heart disease: the Caerphilly Study. Public Health 1996;110:139-43.
    14. Fall CHD, Barker DJP, Osmond C, et al. Relation of infant feeding to adult serum cholesterol concentration and death from ischaemic heart disease. BMJ 1992;304:5801-5.
    15. Barker DJP. The fetal origins of hypertension. J Hypertens 1996;14:117-20.
    16. Law CM, Barker DJP, Bull AR, et al. Maternal and fetal influences on blood pressure. Arch Dis Child 1991;66:1291-5.
    17. Whincup P, Cook D, Papacosta O, et al. Birth weight and blood pressure: cross sectional and longitudinal relations in childhood. BMJ 1995;311:773-6.
    18. Forrester TE, Wilks RJ, Bennett FI, et al. Fetal growth and cardiovascular risk factors in Jamaican schoolchildren. BMJ 1996;312:156-60.
    19. Taittonen L. Association of blood pressure with familial factors, prenatal and postnatal factors, insulin and trace elements in children and adolescents. Acta Universitatis Ouluensis Medica D 424 1997:1-69.
    20. Ravelli ACJ, van der Meulen JHP, Michels RPJ, et al. Glucose tolerance in adults after prenatal exposure to famine. Lancet 1998;351:173-7.
    21. Fall CHD, Osmond C, Barker DJP, et al. Fetal and infant growth and cardiovascular risk factors in women. BMJ 1995;310:428-32.
    22. Matthes JWA, Lewis PA, Davies DP, et al. Relation between birth weight at term and systolic blood pressure in adolescence. BMJ 1994;308:1074-7.
    23. Stanner SA, Bulmer K, Andrés C, et al. Does malnutrition in utero determine diabetes and coronary heart disease in adulthood? Results from the Leningrad siege study, a cross sectional study. BMJ 1997;315:1342-9.
    24. Reaven GM. Role of insulin resistance in human disease. Diabetes 1988;37:1595-607.
    25. Fontbonne A, Charles MA, Thibult N, et al. Hyperinsulinaemia as a predictor of coronary heart disease mortality in a healthy population: the Paris Prospective Study, 15-year follow-up. Diabetologia 1991;34:356-61.
    26. Ben-Shlomo Y, Smith GD. Deprivation in infancy or in adult life: which is more important for mortality risk? Lancet 1991;337:530-4.
    27. Churchill D, Perry U, Beevers DG. Ambulatory blood pressure in pregnancy and fetal growth. Lancet 1997;349:7.
    28. Scrimshaw NS. The relation between fetal malnutrition and chronic disease in later life. BMJ 1997;315:825.
    29. Taylor SJC, Whincup PH, Cook DG, et al. Size at birth and blood pressure: cross sectional study in 8B11 year old children. BMJ 1997;314:475-80.
    30. Clausen JO, Borch-Johnsen K, Pedersen O. Relation between birth weight and the insulin sensitivity index in a population sample of 331 young, healthy caucasians. Am J Epidemiol 1997;146:23-31.
    31. Järvelin M-R, Elliott P, Kleinschmidt I, et al. Ecological and individual predictors of birth weight among Northern Finland birth cohort for 1986. Paediatr Perinat Epidemiol 1997:11:298-312.
    32. Lumme R, Rantakallio P, Hartikainen A-L, et al. Pre-pregnancy weight and its relation to pregnancy outcome. J Obstet Gynaecol 1995;15:69-75.
    33. Innes KM, Wimsatt JH. Pregnancy-induced hypertension and insulin resistance: evidence for a connection. Acta Obstet Gynecol Scand 1999;78:263-84.
    34. Forsén T. Early growth and adult disease. Programming of coronary heart disease, Type 2 diabetes and hypertension by fetal and childhood growth. Academic dissertation, Medical Faculty of the University of Helsinki, Helsinki 2000, p 81.
    35. Langley-Evans SC, Gardner DS, Jackson AA. Association of disproportionate growth of fetal rats in late gestation with raised systolic blood pressure in later life. J Reprod Fertil 1996;106:307-12.
    36. Oh W, Gelardi NL, Cha CJ. The cross-generation effect of neonatal macrosomia in rat pups of streptozotocin-induced diabetes. Pediatr Res 1991;29:606-10.
    37. Roseboom TJ, van der Meulen JHP, Ravelli CJ, et al. Blood pressure in adults after prenatal exposure to famine. J Hypertens 1999;17:325-30.
    38. Catalano PM, Thomas AJ, Huston LP, et al. Effect of maternal metabolism on fetal growth and body composition. Diabetes Care 1998;21(suppl 2):B85-90.
    39. Anguita RM, Sigulem DM, Sawaya AL. Intrauterine food restriction is associated with obesity in young rats. J Nutr 1993;123:1421-8.
    40. Godfrey KM, Barker DJP, Robinson S, et al. Maternal birthweight and diet in pregnancy in relation to the infant's thinness at birth. Br J Obstet Gynaecol 1994;104:663-7.
    41. Godfrey KM, Forrester T, Barker DJP, et al. Maternal nutritional status in pregnancy and blood pressure in childhood. Br J Obstet Gynaecol 1997;101:398-403.
    42. Hartikainen A-L, Aliharmi RH, Rantakallio P. A cohort study of epidemiological associations and outcomes of pregnancies with hypertensive disorders. Hypertension in Pregnancy 1998;17:31-41.
    43. Walker BR, McConnachie A, Noon JP, et al. Contribution of parental blood pressures to association between low birth weight and adult high blood pressure: cross sectional study. BMJ 1998;316:834-7.
    44. Taylor SJC, Hird K, Whincup P, et al. Relation between birth weight and blood pressure is independent of maternal blood pressure. BMJ 1998;317:680.
    45. Himmelmann A, Svensson A, Hansson L. Relation of maternal blood pressure during pregnancy to birth weight and blood pressure in children. The hypertension in pregnancy offspring study. J Intern Med 1994;235:347-52.
    46. Zinner S, Rosner, B, Oh W, et al. Significance of blood pressure in infancy: familial aggregation and predictive effect on later blood pressure. Hypertension 1985;7/3:411-16.
    47. Hennessy E, Alberman E. The effects of own fetal growth on reported hypertension in parous women aged 33. Int J Epidemiol 1997;26:562-70.
    48. Eriksson JG, Forsén T, Tuomilehto J, et al. Catch-up growth in childhood and death from coronary heart disease: longitudinal study. BMJ 1999;318:127-31.
    49. Edwards C, Benediktsson R, Lindsay R, et al. Dysfunction of placental glucocorticoid barrier: link between fetal environment and adult hypertension? Lancet 1993;341:3550-7.
    50. Seckl JR, Benediktsson R, Lindsay RS, et al. Placental 11β-hydroxysteroid dehydrogenase and the programming of hypertension. J Steroid Biochem Molec Biol 1995;5/6:447-55.
    51. Kotelevtsev Y, Brown RW, Fleming S, et al. Hypertension in mice lacking 11β-hydroxysteroid dehydrogenase type 2. J Clin Invest 1999;103:683-9.
    52. Fowden AL. Endocrine regulation of fetal growth. Reprod Fertil Dev 1995;7:351-63.
    53. Phillips DIW, Barker DJP, Fall CHD, et al. Elevated plasma cortisol concentrations: a link between low birth weight and the insulin resistance syndrome? J Clin Endocrinol Metab 1998;83:757-60.
    54. Walker BR, Phillips DIW, Noon JP, et al. Increased glucocorticoid activity in men with cardiovascular risk factors. Hypertension 1998;31:891-5.
    55. Hill Dj, Petrik J, Arany E. Growth factors and the regulation of fetal growth. Diabetes Care 1998;21(suppl 2):B60-9.
    56. Hattersley AT, Beards F, Ballantyne E, et al. Mutations in the glucokinase gene of the foetus result in reduced birth weight. Nature Genetics 1998;19:268-70.
    57. Ong KKL, Phillips DI, Fall C, et al. The insulin gene VNTR, type 2 diabetes and birth weight. Nature Genetics 1999;21:263-4.
    58. Dunger DB, Ong KKL, Huxtable SJ, et al. Association of the INS VNTR with size at birth. Nature Genetics 1998;19:98-100.
    59. Kuh D, Power C, Blane D, Bartley M. Social pathways between childhood and adult health. In: Kuh D, Ben-Shlomo Y, eds. A life course approach to chronic disease epidemiology. Oxford University Press, 1997:242-73.
    60. Davey Smith G, Hart C, Ferrell C, et al. Birth weight of offspring and mortality in the Renfrew and Paisley study: prospective observational study. BMJ 1997;315:1189-93.
    61. Barker DJP, Martyn CN, Osmond C, et al. Growth in utero and serum cholesterol concentrations in adult life. BMJ 1993;307:1524-7.
    62. Koupilova I, Leon DA, Vågerö D. Can confounding by sociodemographic and behavioural factors explain the association between size at birth and blood pressure at age 50 in Sweden? J Epidemiol Community Health 1997;51:14-18.
    63. Martyn CN, Barker DJP, Jespersen S, et al. Growth in utero, adult blood pressure, and arterial compliance. Br Heart J 1995;73:116-21.