Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
294_N01-162 8/4/2002 3:35 pm Page 294 (Black plate) © International Epidemiological Association 2002 Printed in Great Britain International Journal of Epidemiology 2002;31:294–299 REVIEWS Epidemiological challenges in studying the fetal origins of adult chronic disease Matthew W Gillman Keywords Fetal origins of adult disease, lifecourse, socioeconomic status, fetal growth, geneenvironment interaction, nutrition, epidemiological study design, fetal programming Accepted 23 October 2001 The relatively new research field of the fetal origins of adult disease has matured greatly in the past decade. Recent systematic overviews provide strong evidence that robust inverse epidemiological associations exist between birthweight and later blood pressure/hypertension1 and glucose intolerance/type 2 diabetes.2 Consistent associations are also evident for the inverse associations of birthweight with coronary heart disease incidence and mortality.3,4 Studies with high follow-up rates and adjustment for social and economic variables have to a large extent mitigated earlier criticisms that observed associations were highly subject to confounding and selection bias.5 Many questions remain, however, about the contributing roles of postnatal growth and development, underlying biological mechanisms, and the importance of the epidemiological findings to public health. This review outlines current challenges—stated in the form of propositions—that epidemiologists in this field currently face. 1. The Lifecourse Approach Helps Put Fetal Influences into Perspective The lifecourse approach to chronic disease posits that fetal life is one important period of development, but not the only one, in determining susceptibility to adult chronic disease.6 Many studies have now shown that associations of birth size with later cardiovascular outcomes are modified by body mass index in adulthood or, indeed, in childhood.7,8 The highest risk is typically among adults who were born small and become overweight during childhood or adulthood. Examples of outcomes that fit this profile include insulin resistance in Indian 8-year-old boys and girls,9 blood pressure levels among 50-year-old Swedish men,10 and incidence of coronary heart disease among middle to older age Welsh men.4 In another example, body mass index at age 11 appeared to modify the association of ponderal index at birth with adult coronary disease risk in Finnish males.7 From this observed phenomenon some have inferred that that postnatal ‘catch up growth’ modifies intrauterine influences. There are two problems with this inference. First, most studies Department of Ambulatory Care and Prevention, Harvard Medical School and Harvard Pilgrim Health Care, and the Department of Nutrition, Harvard School of Public Health, Boston, MA, USA. to date include only two weight points—one at birth and one later in life. Without a third, intermediate, point, it is difficult to distinguish whether the findings represent an interaction between fetal and postnatal exposures or merely some facets of postnatal growth itself.11 Some recent studies, however, have had multiple postnatal measures of height and weight. Overall they have suggested that both intrauterine and postnatal growth are important, but it is not clear exactly which periods are important or even whether increased or decreased postnatal growth is harmful. For example, in a Finnish cohort, adult hypertension was associated with both lower birthweight and accelerated growth in the first 7 years of life.12 In contrast, in a small group of Chinese adults, both lower ponderal index at birth and a decrease in ponderal index from 6 to 18 months of life were related to increased systolic blood pressure level at age 30 years.13 Moreover, in two recently reported studies from the UK, catch up growth in the first 6 months of life was not clearly related to young adult blood pressure at all, although birthweight was.14,15 In another study, growth patterns exemplified by lower birthweight followed by reduced growth until age 1 seemed to confer the highest risk for ischaemic heart disease later in life, but the analysis was confined to men.16 These disparate results call for additional serial growth data to determine which periods of growth contribute most to the observed interactions between fetal and postnatal size in determining health outcomes. A second issue is that catch up growth traditionally refers to the first 6–12 months after birth, not later growth. The standard interpretation is that catch up growth represents realignment of one’s genetic growth potential after intrauterine growth restriction. One advantage of this interpretation is that it allows for fetal growth restriction at any birthweight—even large fetuses can be growth restricted relative to their genetic potential—thus obviating the need for arbitrary definitions such as ‘small for gestational age’. This distinction is important in the fetal origins of adult disease, since observed associations of birthweight with later outcomes generally span the entire birthweight spectrum, not just at the ‘low birthweight’ end. The major problems, however, are in the measurement of catch up growth. Some investigators resort to a post hoc, empirical definition, using the observed infant growth patterns relative to birthweight.17 A second possibility is that mid-parental height gives a good 294 294_N01-162 8/4/2002 3:35 pm Page 295 (Black plate) CHALLENGES RELATING TO FETAL PROGRAMMING estimate of the child’s growth potential, but usually for height, not relative weight, and the estimation is better after 2 years of age than in infancy. Third, while needing confirmation, recent data raise the possibility that levels of maternal insulin-like growth factor I during pregnancy may reflect fetal growth restriction.18 A final issue with measuring catch up growth is that even in the first 6 months of life, postnatal environmental factors can also influence growth. For example, while having been breastfed seems to protect against obesity later in childhood,19 breastfed infants in developed countries generally exhibit higher body mass during the first year of life that formula-fed infants.20 In sum, there are three possible general inferences from observations that the highest risk for cardiovascular outcomes lies among individuals both born small and achieving high body size later in life. The first is that fast postnatal growth itself increases risk. The second is that factors related to fetal growth restriction increase risk and the third involves an interaction between the two. Thus studies of the fetal origins of adult disease will benefit from having multiple measures of height and weight through the lifecourse, to distinguish prenatal influences from influences occurring at various points in infancy, childhood, and later. They will also benefit from a clear definition of catch up growth and a non-circular method of measuring it. In the meantime, authors should clarify what periods of life their measurements cover, preferably reserving the term catch up growth for its traditional meaning of realignment to genetic potential during infancy. 2. Adjustment for Achieved Body Mass Index, Despite Being Controversial, Is Often Helpful While the points above concern effect modification by achieved body size, this section addresses adjustment for it. Some epidemiological associations, for example the inverse relationship of birthweight with blood pressure, appear stronger—or in some studies, appear only—after statistical adjustment for adult (or childhood) body mass index.21,22 How, or even whether, to make such adjustments is controversial. Some of the controversy seems to emanate from the fact that epidemiologists from a variety of fields—perinatal, cardiovascular, paediatric—with different lexicons, are converging on the same research topics. Also, perinatal epidemiologists have traditionally been concerned with avoiding ‘low birthweight’. In fact, increasing birthweight at this end of the spectrum, primarily by avoiding preterm births,23 would reduce neonatal and infant adverse outcomes. From this perspective, birthweight is an entity in and of itself, with aetiological significance. In contrast, in the field of fetal origins of adult disease, birthweight is not a monolith. Not all determinants of birthweight are salient, so increasing birthweight is not necessarily helpful. Further, this new paradigm is concerned with the entire spectrum of birthweight, rather than just the lower parts of the distribution. Adjusting for later body size is therefore appropriate; it can elucidate underlying aetiological pathways. Figure 1 demonstrates this argument, using the example of blood pressure as an outcome. If it is true that (1) birthweight is positively related to later body mass index,24 (2) birthweight is inversely related to later blood pressure—the association of 295 Figure 1 Why adjustment for attained body mass index is warranted. See text (‘Challenge No. 2‘) for explanation. BMI = body mass index; BP = blood pressure; + indicates direct association; – indicates inverse association primary interest, and (3) body mass index is positively related to contemporaneous blood pressure (a well-known relationship), then the prenatal factors explaining pathways #1 and #2 (labelled Factors A and Factors B, respectively) are almost surely different. Adjusting for body mass index could elucidate some of the factors underlying the inverse birthweight-blood pressure association, and is thus potentially useful. One empirical example that serves to illustrate this point is the relationship between birthweight and development of type 2 diabetes. Unadjusted for attained body mass index, the shape of the relationship is a ‘U’ or reverse ‘J’, with increased risk at both ends of the birthweight spectrum.25 But this observation obscures the fact that there are probably two underlying mechanisms. At high birthweights, maternal gestational diabetes is causing obesity and excess diabetes risk in the offspring. Adjustment for attained body mass, however, reveals an inverse and linear association across the birthweight spectrum, reflecting a second pathway with reduced risk at higher birthweights, a pattern sometimes attributed to the influence of a ‘thrifty phenotype’.26 3. Social and Economic Factors Are Not Just Confounders Many decades span the time from birth until cardiovascular diseases declare themselves. Most of these diseases are related to both lower birthweight and poorer socioeconomic circumstances in adulthood. Thus controlling for these circumstances seems necessary to isolate prenatal influences. On the other hand, for some social class is stable across generations. Under these conditions, the same socioeconomic factors that affect postnatal life may affect prenatal life as well, and could thus determine the putative exposures under study. In this view, the social and economic factors are part of the causal pathway, both before and after birth. For others, social circumstances change over time. Depending on the disorder, the place, and the era, societal factors can have impact in different points in the lifespan. In the UK in the 294_N01-162 8/4/2002 3:35 pm Page 296 296 (Black plate) INTERNATIONAL JOURNAL OF EPIDEMIOLOGY late 20th century, for example, low childhood social class is more strongly associated with risk of stroke than is adult social class. For coronary heart disease, however, childhood and adulthood social factors appear to play equal roles.27 The challenges for future analyses are to consider the explanatory as well as confounding effects of social and economic factors; and to take into account changes in these factors across the lifespan. As these effects are being clarified, it will be important to assess how such societal factors affect physiological mechanisms underlying the epidemiological phenomena. Some factors may operate through their associations with behavioural factors. For example, maternal diet may differ by social class. In contrast, other societal factors may directly affect biological responses; for example, accumulated stress may influence the hypothalamic-pituitary-adrenal axis independent of lifestyle choices. Alternatively, the two kinds of pathways may actually work in concert with each other. In rodent models, stress during gestation causes reduced levels of maternal licking/grooming of her offspring, which are in turn related to levels of offspring stress reactivity through effects on systems that regulate activity of corticotropin-releasing hormone.28 4. Fetal Growth and Size at Birth Are Not Synonymous In studies that can separate length of gestation from overall birthweight, most observed associations between birthweight and later cardiovascular outcomes are not due to length of gestation.29 Epidemiologists have called the remaining factor in birthweight ‘fetal growth’. However, fetal size at birth, adjusted for length of gestation, is only a snapshot of the trajectory of fetal growth throughout gestation. Some, but clearly not all, epidemiological studies have found stronger associations with later outcomes for thinness or shortness at birth, or even abdominal or other body circumferences, rather than birthweight itself. Authors have hypothesized trimester-specific perturbations of growth of the fetus or particular organs to explain these phenomena.30 Whether such hypotheses have merit cannot be fully addressed with crude measures of fetal size at birth, considering not only the measurement error inherent in length and circumference measures, but also the non-specificity of even accurate measures at birth for determining fetal growth trajectories.31,32 Ongoing animal studies of fetal growth will continue to inform the epidemiology. In addition, one way studies of humans can be refined is to take advantage of serial ultrasound measures to measure fetal growth parameters throughout pregnancy. For example, preliminary data from one study suggested that femur length at 24 weeks was more strongly inversely associated with blood pressure at age 6 years than were either head or abdominal circumferences.33 In that study, however, changes in ultrasound measurements from 18 to 38 weeks were not clearly related to the outcome. In addition, the dearth of data linking reduced gestational age at birth with adult health outcomes is chiefly a function of the historical data available to date, not necessarily because the link is absent. When most of the participants in the currently published long-term studies were born, relatively few preterm infants survived. To what extent gestational age predicts adult outcomes is an unresolved question, but an important one. In general, determinants of preterm birth and fetal growth are not the same.34 5. Some Determinants of Fetal Growth Will Be Red Herrings While the observed epidemiological associations appear to implicate fetal growth as a major factor, overall fetal size at birth is an amalgam of many determinants. Some of these determinants may be related to susceptibility to adult disease, and others may not. Conversely, some prenatal determinants of adult outcomes may not be related to fetal growth. While investigations of fetal growth are likely to reveal some underlying processes, the need exists to investigate more directly prenatal determinants of postnatal outcomes, irrespective of effects on birthweight. One example of size at birth potentially being a good proxy for an underlying causal pathway derives from the hypothesis that a congenital nephron deficit underlies predisposition to hypertension in adult life.35 A recent study shows that kidney volume is smaller in adults who were thinner at birth, even after adjustment for current body size.36 In contrast, a potential example of a prenatal exposure causing postnatal disease without effects on birthweight is the newly described association of paracetamol use by mothers in late pregnancy and risk of asthma in their preschool age offspring.37 Maternal smoking during pregnancy is a good example of a prenatal exposure whose deleterious effect on fetal growth is well-known but whose influence on long-term offspring health is uncertain. Perhaps investigation of the seemingly paradoxical protective effect of smoking against the development of pre-eclampsia38 will offer insights. 6. Both Genes and Environment Are Important. Even Distinction Between the Two Is Difficult Many critics of the ‘fetal origins hypothesis’ have taken to task the epidemiological studies for not accounting for possible genetic explanations. Although rare genetic mutations do not explain the findings of population-based studies, they may provide a paradigm for understanding. For example, a rare mutation in the glucokinase gene causes drastically reduced birthweight and offspring insulin resistance.39 No analogous common allelic variation has yet been discovered, but it is possible that some exist to help explain the observed phenomena. In addition, maternal, paternal, and fetal genes may all play roles. While inherited fetal gene expression can underlie susceptibility to disease, it is also likely that the maternal genome codes for substances that affect the fetal environment. As such, what is genetic for the mother is environmental for the fetus. One example of this principle is that it appears that the maternal thermolabile variant of methylenetetrahydrofolate reductase may be associated with the risk of neural tube defects in the offspring, even in the face of a normal fetal genotype.40 This consideration also highlights the close relationship between genes and environment and the need to examine interactions between them. Not only can maternal gene expression alter the fetal environment, but it is also possible that the maternal gestational environment can affect fetal gene expression. 294_N01-162 8/4/2002 3:35 pm Page 297 (Black plate) CHALLENGES RELATING TO FETAL PROGRAMMING For example, in a mouse model, feeding the mother foods with different quantities of methyl donors can alter offspring coat colour through modification of gene expression.41,42 More substantial progress in humans will occur when investigators can assess interactions among environmental factors and candidate genes as their identities become available. Ultimately, there will be probably be less distinction between genes and environment than most researchers currently consider. 7. Maternal Nutrition Does Not Equal Fetal Nutrition Scientists and the public have been confused by statements about ‘undernutrition’ explaining the observed epidemiological phenomena. Except at the extremes of intake, maternal energy and macronutrient intake have relatively little impact on birthweight. In addition, little or no firm evidence suggests that maternal intake of total energy or macronutrients during pregnancy, or maternal body composition, have substantial impact on offspring risk of chronic disease. In fact, some empirical results seem diametrically opposed to each other. For example, results of one abstract presented at the 2001 World Congress on the fetal origins of adult disease suggested that a high dietary protein/carbohydrate ratio in the third trimester was associated with lower offspring blood pressure, whereas the authors of the next presentation observed higher blood pressures among offspring of mothers consuming a relatively high protein, low carbohydrate diet in late pregnancy.43,44 In contrast, what is often termed ‘fetal nutrition’ is likely to be vital. The oxygen and nutrients that support fetal growth and development rely on the entire nutrient supply line, beginning with maternal consumption and body size, but extending to uterine perfusion, placental function, and fetal metabolism. Interruptions of the supply line at any point could result in programming the fetus for future risk of cardiovascular disease.31 Some of the confusion arises because a common animal model for interrupting the supply line is to alter maternal gestational intake.45 Interrupting the supply line at other points, say by underperfusing the placenta, can have analogous effects. One should use the animal experimental data only as models, as they are intended. The placenta is likely to be a key to understanding associations of fetal growth with adult health, owing to its inherent endocrine function as well as its metabolic and transfer capacities.31 The lack of good animal models for placentation has hampered understanding of the pathways by which the fetus receives nutrients and oxygen from the mother, and what processes alter these pathways. Possible ways to study placental function in population-based studies include placental morphological pathology, as well as using specimens of placenta, maternal prenatal blood or other biological specimens, and umbilical cord blood to measure markers of altered blood flow, endocrine and transport characteristics, or activity of specific enzymes. These biological specimens can also be used to investigate other mechanisms underlying the observed epidemiological findings, including maternal and fetal endocrine pathways,46 inflammatory and immune function,47 and changes in the central and peripheral nervous system,48,49 among others. Still, researchers should not dismiss the possibility that maternal dietary intake could be an important supply line 297 component in humans. In investigating maternal dietary impact, investigators should concentrate on pathways that could involve specific foods and micronutrients in addition to energy and macronutrient intake, which appear to be relatively unimportant on their own. For example, recent data from 27 pregnant women suggest that while overall gestational protein intake is unimportant, relative partitioning from protein oxidation to deposition, perhaps related to micronutrient factors, is associated with birthweight.50 Relevant examples from perinatal epidemiology include that fatty acids of the n-3 family prolong gestation and reduce the incidence of preterm birth, and inadequate folic acid intake causes neural tube defects.51,52 Future research may reveal that fatty acids, folate, or other nutrients may also be involved in programming of chronic disease risk. 8. New Epidemiological Study Designs Will Help Many of the extant epidemiological findings have emanated from historical cohort studies, taking great advantage of serendipitous clinically recorded perinatal variables. The next generation of studies, however, should overcome the limitations of the earlier studies. These limitations include lack of research quality measures, limited information on covarying factors, and absence of biological specimens. Although no one study can address all methodological issues, three types of studies will be especially valuable in the coming years: 1. New prospective studies of pregnant women (including ones that even start pre-conceptionally) that follow their offspring over time, collecting biological samples to explore mechanisms, as well as research quality epidemiological data.53 These have the advantage of collecting detailed prospective data and examining childhood cardiovascular risk factors and diseases such as asthma, but must wait decades for adult disease outcomes to occur. 2. Follow-up of adult cohorts in which biological samples and prospectively collected epidemiological data are available from the past.54 These have the advantage of research quality data and emerging disease outcomes, but may lack important exposure or covariate information, such as maternal diet during pregnancy. 3. Follow-up into childhood or adulthood of randomized controlled trials conducted during pregnancy.55 These have the advantage of isolating an environmental ‘insult’ during pregnancy and so can come close to proving that programming occurs. However, this type of study is limited to examining a single or limited number of exposures. 9. To Make Policy Recommendations Now Regarding Fetal Programming Is Dangerous Strong evidence exists for programming in humans. First is the robust epidemiological findings of associations between size at birth and later health outcomes. Second is the experimental animal evidence that in utero environmental insults produce lifelong alterations in metabolism, physiology, and pathology, 294_N01-162 8/4/2002 3:35 pm Page 298 298 (Black plate) INTERNATIONAL JOURNAL OF EPIDEMIOLOGY whether or not the exact perturbations have analogy in human systems. One area of great importance is implications for developing countries undergoing the epidemiological transition from infectious disease to chronic disease, chiefly cardiovascular disease, as well as the ‘nutrition’ transition from an active, low-calorie lifestyle to a sedentary, high-calorie lifestyle.56 Available data indicate that a lower birthweight combined with higher attained body mass confers the highest risk for cardiovascular disease. Successive generations of individuals in developing countries are likely to have ever larger proportions with that high-risk profile. Efforts to prevent development of obesity in areas undergoing the epidemiological and nutritional transitions are paramount. It is premature to make policy recommendations regarding fetal programming. Since birthweight is only a marker for underlying aetiological pathways, the true aetiological factors are unknown. Until they are identified, it is impossible to calculate attributable risks or design interventions to modify them. Interventions to increase birthweight per se would be at best inefficient, probably ineffective, and potentially harmful. Ultimately, the promise of the fetal origins paradigm is that attending to the health of women of reproductive age will have profound impact on the well-being of their progeny. 4 Frankel S, Elwood P, Sweetnam P, Yarnell J, Davey Smith G. Birthweight, body-mass index in middle age, and incident coronary heart disease. Lancet 1996;348:1478–80. 5 Gillman MW, Rich-Edwards J. The fetal origins of adult disease: from sceptic to convert. Pediatr Perinat Epidemiol 2000;14:192–93. 6 Kuh D, Ben-Shlomo Y. A Life Course Approach to Chronic Disease Epidemiology. New York: Oxford University Press, 1997. 7 Eriksson JG, Forsen T, Tuomilehto J, Winter PD, Osmond C, Barker DJP. Catch-up growth in childhood and death from coronary heart disease: a longitudinal study. BMJ 1999;318:427–31. 8 Whincup PH, Bredow M, Payne F, Sadler S, Golding J. Size at birth and blood pressure at 3 years of age. The Avon Longitudinal Study of Pregnancy and Childhood (ALSPAC). Am J Epidemiol 1999;149: 730–39. 9 Bavdekar A, Yajnik CS, Fall CHD et al. The insulin resistance syndrome in eight-year-old Indian children; small at birth, big at eight years or both? Diabetes 1999;48:2422–29. 10 Leon DA, Koupilova I, Lithell HO et al. Failure to realise growth potential in utero and adult obesity in relation to blood pressure in 50 year old Swedish men. BMJ 1996;312:401–06. 11 Lucas A, Fewtrell MS, Cole TJ. Fetal origins of adult disease—the hypothesis revisited. BMJ 1999;319:245–49. 12 Eriksson J, Forsen T, Tuomilehto J, Osmond C, Barker D. Fetal and childhood growth and hypertension in adult life. Hypertension 2000; 36:790–94. 13 Cheung YB, Low L, Osmond C, Barker D, Karlberg J. Fetal growth Acknowledgements Many of the ideas expressed here grew out of a Harvard Medical School conference, Fetal Origins of Adult Cardiovascular Disease: New Directions, which I chaired in November 1999. The contributions of the 48 attendees of that conference, which included epidemiologists, biomedical scientists, and health policy researchers, were instrumental in the genesis of this paper. The list of attendees and the conference agenda are available upon request. I also acknowledge the helpful comments of Janet Rich-Edwards, conference co-chair, on a draft of this manuscript. Conference funding was supplied by Maternal and Child Health Bureau of Health Resources and Services Administration (6-MCJ-000102–45–1), Department of Ambulatory Care and Prevention at Harvard Medical School/Harvard Pilgrim Health Care, SmithKline Beecham Consumer Healthcare, Harvard Center for Children’s Health, Department of Nutrition at Harvard School of Public Health, Council on Epidemiology and Prevention of the American Heart Association, Mead Johnson Nutritionals, The Procter and Gamble Co. This work was also supported by grants from the US National Institutes of Health (HD 34568, HL 64925, HL 68041) and by Harvard Medical School and the Harvard Pilgrim Health Care Foundation. References and early postnatal growth are related to blood pressure in adults. Hypertension 2000;36:795–800. 14 McCarthy A, Ben-Schlomo Y, Elwood P, Davies D, Davey Smith G. The relationship between birth weight, catch up growth in infancy and blood pressure in young adulthood. Pediatr Res 2001;50:45A. 15 Law CM, Shiell AW, Syddall HE. Growth in utero, infancy and childhood: which influences adult blood pressure? Pediatr Res 2001;50:11A. 16 Eriksson JG, Forsen T, Tuomilehto J, Osmond C, Barker DJP. Early growth and coronary heart disease in later life: longitudinal study. BMJ 2001;322:949–53. 17 Dunger DB, Ong KKL, Huxtable SJ et al. Association of the INS VNTR with size at birth. Circulation 1998;96:2233–38. 18 Holmes RP, Holly JM, Soothill PW. A prospective study of maternal serum insulin-like growth factor-I in pregnancies with appropriately grown or growth restricted fetuses. Br J Obstet Gynaecol 1998;105: 1273–78. 19 Gillman MW, Rifas-Shiman SL, Camargo CA Jr et al. Risk of overweight among adolescents who had been breast fed as infants. JAMA 2001;285:2461–67. 20 Dewey KG, Peerson JM, Brown KB et al. Growth of breast-fed infants deviates from current reference data: a pooled analysis of US, Canadian, and European data sets. Pediatrics 1995;96:495–503. 21 Leon DA. Fetal growth and later disease: epidemiological evidence from Swedish cohorts. In: O’Brien PMS, Wheeler T, Barker DJP (eds). Fetal Programming. Influences on Development and Disease in Later Life. London: RCOG Press, 1999, pp.12–29. 1 Huxley RR, Shiell AW, Law CM. The role of size at birth and postnatal 22 Whincup PH. Fetal origins of cardiovascular risk: evidence from catch-up growth in determining systolic blood pressure: a systematic review of the literature. J Hypertens 2000;18:815–31. studies in children. In: O’Brien PMS, Wheeler T, Barker DJP (eds). Fetal Programming. Influences on Development and Disease in Later Life. London: RCOG Press, 1999, pp.30–40. 2 Law CM, Fall CHD, Newsome CA, Shiell AW, Phillips DIW, Shier R. Is birthweight related to glucose and insulin metabolism?—a systematic review. Pediatr Res 2001;50:60A. 23 Wilcox AJ. On the importance—and the unimportance—of birth 3 Rich-Edwards JW, Stampfer MJ, Manson JE et al. Birthweight and 24 Parsons TJ, Powers C, Logan S, Summerbell CD. Childhood predictors the risk of cardiovascular disease in adult women. BMJ 1997; 315:396–400. of adult obesity: a systematic review. Int J Obes Relat Metab Disord 1999;23:S1–107. weight. Int J Epidemiol 2001;30:1233–41. 294_N01-162 8/4/2002 3:35 pm Page 299 (Black plate) CHALLENGES RELATING TO FETAL PROGRAMMING 25 Rich-Edwards JW, Colditz GA, Stampfer MJ et al. Birthweight and the risk for type 2 diabetes mellitus in adult women. Ann Intern Med 1999;130:278–84. 299 42 Wolff GL, Kodell RL, Moore SR, Cooney CA. Maternal epigenetics and methyl supplements affect agouti gene expression in Avy/a mice. FASEB J 1998;12:949–57. 26 Hales CN, Barker DJP. Type 2 (non-insulin-depedent) diabetes 43 Haselden S, Shiell AW, Campbell-Brown M, Robinson S, Godfrey KM, mellitus: the thrifty phenotype hypothesis. Diabetologia 1992;35: 595–601. Barker DJP. A high meat, low carbohydrate diet in pregnancy; relation to the offspring’s blood pressure and glucose-insulin metabolism at age 30. Pediatr Res 2001;50:24A. 27 Davey Smith G, Hart C, Blane D, Hole D. Adverse socioeconomic conditions in childhood and cause specific adult mortality: prospective observational study. BMJ 1998;316:1631–35. 28 Francis D, Diorio J, Liu D, Meaney MJ. Nongenomic transmission across generations in maternal behavior and stress responses in the rat. Science 1999;286:1155–58. 29 Leon DA, Lithell HO, Vagero D et al. Reduced fetal growth rate and increased risk of death from ischemic heart disease: cohort study of 15 000 Swedish men and women born 1915–29. BMJ 1998;317: 241–45. 30 Barker DJP. Mothers, Babies, and Disease in Later Life. 2nd Edn. London: Harcourt Brace & Co. Ltd, 1998. 31 Harding JE. The nutritional basis of the fetal origins of adult disease. Int J Epidemiol 2001;30:15–23. 32 Kramer MS, McLean FH, Olivier M, Willis DM, Usher RH. Body proportionality and head and length ‘sparing’ in growth-retarded neonates: a critical appraisal. Pediatrics 1989;84:717–23. 33 Blake KV, Gurrin LC, Stanley FJ, Landau LI, Beilin LJ, Newnham JP. Intrauterine growth and subsequent blood pressure in childhood. Pediatr Res 2001;50:42A. 34 Kramer MS. Determinants of low birth weight: methodological assess- ment and meta-analysis. Bull World Health Organ 1987;65: 663–737. 44 Roseboom TJ, van der Meulen JH, van Montfrans GA et al. Maternal nutrition during gestation and blood pressure in later life. J Hypertens 2001;19:29–34. 45 Jackson AA. Maternal and fetal demands for nutrients: significance of protein restriction. In: O’Brien PMS, Wheeler T, Barker DJP (eds). Fetal Programming. Influences on Development and Disease in Later Life. London: RCOG Press, 1999, pp.246–70. 46 Welberg LA, Seckl JR. Prenatal stress, glucocorticoids and the programming of the brain. J Neuroendocrinol 2001;13:113–28. 47 Prescott SL, Macaubas C, Smallacombe T, Holt BJ, Sly PD, Holt PG. Development of allergen-specific T-cell memory in atopic and normal children. Lancet 1999;353:1187–88. 48 Plagemann A, Harder T, Melchior K, Rake A, Rohde W, Dorner G. Elevation of hypothalamic neuropeptide Y-neurons in adult offspring of diabetic mother rats. NeuroReport 1999;10:3211–16. 49 Young JB, Morrison SF. Effects of fetal and neonatal environment on sympathetic nervous system development. Diabetes Care 1998;21: B156–60. 50 Duggleby SL, Jackson AA. Maternal amino acid oxidation during pregnancy and birth weight. Pediatr Res 2001;50:3A. 51 Olsen SF, Secher NJ, Tabor A, Weber T, Walker JJ, Gluud C. cause of adult hypertension and progressive renal injury? Curr Opin Nephrol Hypertens 1993;2:691–95. Randomised clinical trials of fish oil supplementation in high risk pregnancies. Fish Oil Trials in Pregnancy (FOTIP) Team. Br J Obstet Gynaecol 2000;107:382–95. 36 Singh GR, White AV, Spencer J, Hoy WE. Intrauterine growth 52 MRC Vitamin Study Research Group. Prevention of neural tube influences kidney size and function: the renal dimension of the fetal origins hypothesis. Pediatr Res 2001;50:30A. defects: results of the Medical Research Council Vitamin Study. Lancet 1991;338:131–37. 37 Shaheen SO, Newson RB, Sherriff AM et al. Analgesic use in 53 Rich-Edwards J, Krieger N, Majzoub J, Zierler S, Lieberman E, pregnancy and atopic disease in early childhood. Pediatr Res 2001; 50:5A. Gillman M. Maternal experiences of racism and violence as predictors of preterm birth: rationale and study design. Paediatr Perinat Epidemiol 2001;15:124–35. 35 Brenner BM, Chertow GM. Congenital oligonephropathy: an inborn 38 Xiong X, Wang FL, Davidge ST et al. Maternal smoking and preeclampsia. J Reprod Med 2000;45:727–32. 39 Hattersley AT, Tooke JE. The fetal insulin hypothesis: an alternative explanation of the association of low birthweight with diabetes and vascular disease. Lancet 1999;353:1789–92. 40 Wenstrom KD, Johanning GL, Owen J, Johnston KE, Acton S, Tamura T. Role of amniotic fluid homocysteine level and of fetal 5, 10-methylenetetrahydrafolate reductase genotype in the etiology of neural tube defects. Am J Med Genet 2000;90:12–16. 54 Seidman LJ, Buka SL, Goldstein JM, Horton NJ, Rieder RO, Tsuang MT. The relationship of prenatal and perinatal complications to cognitive functioning at age 7 in the New England Cohorts of the National Collaborative Perinatal Project. Schizophr Bull 2000;26: 309–21. 55 Belizan JM, Villar J, Bergel E et al. Long term effect of calcium supplementation during pregnancy on the blood pressure of offspring: follow up of a randomised controlled trial. BMJ 1997;315:281–85. 41 Morgan HD, Sutherland HGE, Martin DIK, Whitelaw E. Epigenetic 56 Murray CJL, Lopez AD. Alternative projections of mortality and inheritance at the agouti locus in the mouse. Nat Genet 1999;23: 314–18. disability by cause 1990–2020: Global Burden of Disease Study. Lancet 1997;349:1498–504.