Hostname: page-component-7c8c6479df-xxrs7 Total loading time: 0 Render date: 2024-03-28T08:26:33.173Z Has data issue: false hasContentIssue false

The relationship between birth weight and insulin resistance in childhood

Published online by Cambridge University Press:  22 September 2009

Jesuana O. Lemos
Affiliation:
Department of Nutrition, School of Public Health, University of São Paulo, Avenida Dr Arnaldo 715, São Paulo, SP, CEP-01246-904, Brazil
Patricia H. C. Rondó*
Affiliation:
Department of Nutrition, School of Public Health, University of São Paulo, Avenida Dr Arnaldo 715, São Paulo, SP, CEP-01246-904, Brazil
Joilane A. Pereira
Affiliation:
Department of Nutrition, School of Public Health, University of São Paulo, Avenida Dr Arnaldo 715, São Paulo, SP, CEP-01246-904, Brazil
Renata G. Oliveira
Affiliation:
Department of Nutrition, School of Public Health, University of São Paulo, Avenida Dr Arnaldo 715, São Paulo, SP, CEP-01246-904, Brazil
Maria B. S. Freire
Affiliation:
Jundiaí Medical School, Rua Francisco Telles, 250, Vila Arens, Jundiaí, SP, CEP-13202-550, Brazil
Patricia B. Sonsin
Affiliation:
Department of Nutrition, School of Public Health, University of São Paulo, Avenida Dr Arnaldo 715, São Paulo, SP, CEP-01246-904, Brazil
*
*Corresponding author: Dr Patricia H. C. Rondó, fax +55 11 3061 7771, email phcrondo@usp.br
Rights & Permissions [Opens in a new window]

Abstract

Chronic diseases that are typical of adulthood may originate in intra-uterine life through inadequate fetal development. The present epidemiological cohort study of 506 healthy children aged 5–8 years evaluated the relationship between birth weight and insulin resistance in an age group that has been assessed in few similar studies. Insulin concentration was determined by chemiluminescence and insulin resistance by the homeostasis model assessment (HOMA). Blood glucose, total cholesterol and fractions (LDL cholesterol and HDL cholesterol) and TAG concentrations were determined by automated enzymatic methods. Linear regression analysis investigated the relationship between birth weight (assessed as a continuous variable and in three categories: small for gestational age, SGA; adequate for gestational age and large for gestational age) and the HOMA index, using backward stepwise selection and biological models to explain the causal pathway of the relationship. There were negative associations between birth weight (P < 0·001), SGA (P = 0·027) and the HOMA index, and a positive association between waist circumference (P < 0·001) and the HOMA index. Considering the significant associations between birth weight and waist circumference (P < 0·001) and waist circumference and insulin resistance (P < 0·001), we can probably suspect that lower birth weight is a common cause of higher waist circumference and insulin resistance. In summary, the results of the present study showed increased insulin resistance in apparently healthy, young children, who had lower weight at birth and higher measurements of waist circumference. There is a need to develop public health policies that adopt preventive measures to promote adequate maternal-fetal and child development and enable early diagnosis of metabolic abnormalities.

Type
Full Papers
Copyright
Copyright © The Authors 2009

The prevalence of chronic-degenerative diseases is related to the presence of obesity, excess energy, fat and salt intake and reduced physical activity, which are elements of the so-called nutritional transition(Reference Yajnik1). However, it has been demonstrated that diseases that are typical of adulthood may originate during intra-uterine life through inadequate fetal development(Reference De Blasio, Gatford and McMillen2Reference Mi, Cheng and Zhao9). According to the gestational phase during which intra-uterine growth restriction occurs, different systems will be affected, predisposing to diabetes mellitus type 2, arterial hypertension, obesity and dyslipidaemia(Reference Barker10Reference Yajnik13).

The theory known as the ‘Fetal origins of disease’ or the ‘Thrifty phenotype hypothesis’ suggests that individuals who are exposed to nutritional deficiency in the uterus and during the first years of life adapt themselves to this deficiency. In relation to the pancreas, the adaptation mechanism consists of decreases in the number and activity of β-cells and harm to insulin-dependent tissues, especially muscles. Therefore, to survive during periods of nutritional deprivation, fetuses change their metabolic-endocrine routes and become insulin resistant. If, over the course of life, individuals with lower weights at birth are subjected to situations of increased food intake and decreased physical activity, this adaptation may become evident, thereby predisposing towards reduced insulin resistance, obesity and diabetes mellitus type 2(Reference Eriksson, Forsén and Tuomilehto6, Reference Yudkin11, Reference Yajnik and Deshmukh14, Reference Barker, Winter and Osmmond15).

In Brazil, as in other developing countries like India, the prevalence of chronic-degenerative diseases including metabolic diseases is increasing, representing an important cause of death even in young adults(Reference Ramachandran, Snehalatha and Latha16, 17).

The present study investigated the relationship between birth weight and insulin resistance in a cohort of 506 Brazilian children aged 5–8 years, involving an age group that has been assessed in few similar studies(Reference Yajnik, Fall and Vaydia18Reference Joglekar, Fall and Deshpande24).

Materials and methods

This is an epidemiological cohort study initially involving 865 children who were followed up from birth until 5–8 years of age(Reference Rondó, Lemos and Pereira25). The children were born to women who took part in a cohort study in Jundiaí city, Brazil, between 1997 and 2000. The women attended antenatal care in twelve health units and five hospitals in Jundiaí city, and were followed before the 16th week of pregnancy to the birth of their babies. All participants were insured by the National Health Service (SUS) that assists low-income families. Women with chronic infectious diseases, metabolic diseases, cardiopathy, mental diseases, hypertension/pre-eclampsia/eclampsia, vaginal bleeding and multiple deliveries were not included in the study. The children were classified after birth as being small for gestational age (SGA), adequate for gestational age and large for gestational age(Reference Williams, Creasy and Cunningham26). Details of the study are given elsewhere(Reference Rondó, Ferreira and Nogueira27).

Out of the 865 mother–infant pairs from the previous cohort(Reference Rondó, Ferreira and Nogueira27), 745 were located and invited to participate in the present study, resulting in a final sample of 649 children, whose parents signed a free and informed consent statement and answered a general questionnaire.

The study was carried out between November 2004 and September 2006 and consisted of two phases. (1) Taking the information from the questionnaire of the first cohort study into account, the mothers who at that time were living in the city of Jundiaí and nearby municipalities were located. The mothers were invited to participate in the present study through telephone contact or by visiting their homes if they did not have a telephone. Next, a home visit was made, during which the objectives of the study were explained, and the ethical consent statement was signed by the child's parents or guardians. At the time of this home visit, a general questionnaire was applied in order to evaluate socio-economic and demographic data and any morbidity among the children. (2) The participants were contacted again by telephone to arrange to collect anthropometric data (weight, height and waist circumference) and blood samples in order to determine glucose, total cholesterol, HDL cholesterol (HDL-c), LDL cholesterol (LDL-c) and TAG concentrations.

The anthropometric data were obtained in accordance with the recommendations of Cameron(Reference Cameron28) and Jelliffe & Jelliffe(Reference Jelliffe and Jelliffe29). The children were weighed, after been fasting for 10–12 h, by a portable electronic scale (Sohnle®, model 7500, Murrhardt, Germany), with accuracy of 100 g. Their height was measured using a SECA® stadiometer (Leicester Portable Height measure model, Hamburg, Germany), with accuracy of 0·1 cm. The waist circumference measurements were obtained using a tape measure (Stanley®, model 34 103, New Britain, CT, USA) of accuracy 0·1 cm. The BMI was classified according to the Centers for Disease Control and Prevention(30) and International Obesity Task Force classifications(Reference Cole, Bellizzi and Flegal31). Mean values of waist circumference were compared with a British standard(Reference McCarthy, Jarrett and Crawley32).

Immediately after collection of the blood samples for determination of the glucose and insulin concentrations and the lipid profile, the material was centrifuged at 1500 rpm for 10 min (Tomy centrifuge, model IC-15AN, Tominaga Works Ltd, Tokyo, Japan) and then stored in a refrigerator at 0°C for a period of 24–48 h, for the biochemical tests' determinations.

Glucose was obtained by an enzymatic method using the Bayer® ADVIA 1200 clinical chemistry system (Pittsburgh, PA, USA) for glucose hexokinase. Fasting glycaemia concentration was based on the criteria currently adopted for diagnosing diabetes mellitus type 2 and was the same for adults and children(33). Insulin was assayed in serum by a chemiluminescence method using the Immulite 2000 apparatus (Immulite®, DPC, Los Angeles, CA, USA). To determine insulin resistance, the homeostasis model assessment (HOMA) method was used(Reference Matthews, Hosker and Rudenski34) by the formula: HOMA = fasting glucose (mmol/l) × fasting insulin (μU/ml)/22·5. Insulin resistance was considered to be present when HOMA was ≥ 2·5(Reference Valerio, Licenziati and Iannuzzi8, Reference D'Annunzio, Vanelli and Meschi35).

Assays for total cholesterol and HDL-c were performed by a colorimetric enzymatic method, using the Bayer® ADVIA 1200 clinical chemistry system. The concentrations of LDL-c were determined by the Friedwald(Reference Friedwald, Levy and Fredrickson36) formula. TAG concentrations were determined photometrically following an enzymatic reaction using the Bayer® ADVIA 1200 clinical chemistry system. Total cholesterol and fractions and TAG concentrations were classified in accordance with the recommendations of the First Guidelines for the Prevention of Atherosclerosis during Childhood and Adolescence(Reference Giuliano, Caramelli and Pellanda37).

The categorical variables were analysed by the χ2 test with Yates correction, applying the Fisher's exact test when the frequency was less than five. The Student's t test or the Mann–Whitney test was used to compare mean values of pairs of independent variables. The relationship between birth weight and the HOMA index, and the variables of interest were determined using Pearson's or Spearman's correlation. The impact of birth weight and other independent variables (selected according to their importance, correlation coefficient and biological collinearity) on the HOMA index was determined by multiple linear regression, using the backward stepwise selection method. The outcome – HOMA index – was analysed separately for each independent variable. The variables with descriptive level of P ≤ 0·20 (birth weight, age, sex, BMI, waist circumference, HDL-c, LDL-c, TAG and per capita income) were selected for entry into the multiple linear regression model.

In addition to the use of birth weight as a continuous variable, a second linear multiple regression model tested the association between birth weight by gestational age (in three categories: SGA, adequate for gestational age and large for gestational age) and HOMA, controlling for the variables cited above. Moreover, biological models were investigated in order to explain the causal pathway of the relationship between birth weight and insulin resistance.

In the final models, the variables with a P ≤ 0·05 were considered as being statistically significant. The software Epi Info 6.04 (Atlanta, GA, USA) and Stata version 9 (College Station, TX, USA) were used for data storage and for statistical analysis.

The present study was carried out in accordance with the Declaration of Helsinki of the World Medical Association, and it was approved by the Research Ethics Committee of the School of Public Health, University of São Paulo.

Results

The biochemical parameters were determined in the second phase of the study involving a total of 506 children. There was a loss of 32 % children, considering all the mothers who were located (n 745 mothers) in the first phase of the study, and a loss of 41·5 % children, considering the mothers who participated in the original cohort (n 865 mothers).

There was a significant difference between the mean age of the children included in the cohort (6·5 (sd 0·7) years) and the children who did not conclude the study (5·9 (sd 0·8) years; P = 0·01). However, there was no difference in relation to per capita income, birth weight or mother's schooling level. The characteristics of the children at birth are presented in Table 1. With regard to gestational age, 95·5 % were born at term ( ≥ 37 and < 42 weeks of gestation) and 53·6 % were female.

Table 1 Characteristics of the children at birth

(Mean values and standard deviations)

GA, gestational age.

Table 2 shows that most of the children (53·4 %) were between 6 and 7 years old. Approximately, 67·0 % of the participants had a per capita income of less than one Brazilian minimum monthly salary (one Brazilian minimum monthly salary = approximately US$ 77). According to the Centers for Disease Control and Prevention BMI classification(30), 45 (8·9 %) of the children were underweight, 346 (68·6 %) were normal, 72 (14·3 %) were at risk of becoming overweight and 41 (8·2 %) were overweight. Sixty-six children (twenty-one males and forty-five females) (12·9 %) were overweight, and twenty-seven children (twelve males and fifteen females; 5·3 %) were obese by the International Obesity Task Force classification(Reference Cole, Bellizzi and Flegal31). The percentages were lower to studies involving Mexican and British children with comparable ages(Reference Moraes, Beltran and Mondini38, Reference Jebb, Rennie and Cole39). Mean waist circumferences of the children were higher than the 75th percentile of a British standard(Reference McCarthy, Jarrett and Crawley32). High fasting glycaemia (5·6–6·9 mmol/l (700–1000 mg/l)) was observed in sixty-one children (12·0 %) and nineteen (3·7 %) presented an abnormal HOMA index ( ≥ 2·5 mmol/l μU/ml). Total cholesterol was high ( ≥ 4·4 mmol/l; 1700 mg/l)(Reference Giuliano, Caramelli and Pellanda37) in 168 children (33·2 %). One hundred and two children (20·2 %) presented HDL-c that was lower than recommended (1·2 mmol/l; 450 mg/l) and twenty-two (4·3 %) were in the group of children with abnormal LDL-c ( ≥ 3·4 mmol/l; 1300 mg/l).

Table 2 Socio-demographic, nutritional and biochemical characteristics of the children

(Mean values and standard deviations)

BMMS, Brazilian minimum monthly salary (R$350·00 = approximately US$ 77 per capita income); HOMA, homeostasis model assessment; HDL-c, HDL cholesterol; LDL-c, LDL cholesterol.

Table 3 shows the correlations between birth weight, gestational age, demographic, socio-economic, anthropometric and biochemical factors. Birth weight was significantly correlated with all the anthropometric parameters investigated.

Table 3 Correlation coefficients between birth weight (BW) and the variables of interest

GA, gestational age; WC, waist circumference; TC, total cholesterol; HOMA, homeostasis model assessment; Income, per capita income.

All the variables were assessed in childhood, with the exception of BW and GA.

Table 4 presents multiple linear regression models taking the HOMA index as the dependent variable and birth weight (considered as a continuous variable and categorised by gestational age as SGA, adequate for gestational age and large for gestational age) as the independent variable. In the first and second models, a backward stepwise selection method was utilised, controlling for the following variables: age; sex; BMI; waist circumference; HDL-c; LDL-c; TAG; per capita income. In the first model, the HOMA index was negatively associated with birth weight (P < 0·001) and positively associated with waist circumference (P < 0·001). In the second model, the HOMA index was positively associated with SGA (P = 0·027) and waist circumference (P < 0·001).

Table 4 Linear regression models considering insulin resistance (homeostasis model assessment, HOMA) as outcome

SGA, small for gestational age; LGA, large for gestational age.

* First model: R 2 adjusted = 0·16; second model: R 2 adjusted = 0·12; third model: R 2 adjusted = 0·11; fourth model: R 2 adjusted = 0·12; fifth model: R 2 adjusted = 0·13; sixth model: R 2 adjusted = 0·14.

There was a positive association between birth weight and waist circumference (P < 0·001), but no interaction between the two variables (P>0·05). In a univariate linear regression analysis, birth weight did not present a statistically significant association with the HOMA index (P = 0·06; Table 3).

Tu et al. (Reference Tu, West and Ellison40) reported that an inference regarding the existence and direction of an association between birth weight and an outcome occurring later in life depends on whether one adjusts or does not adjust for current weight in setting up the regression model. Therefore, based on Tu et al. (Reference Tu, West and Ellison40) study, and trying to investigate biological models, we carried out a regression analysis assessing the association between birth weight and the HOMA index adjusting the model for body weight and also for BMI. There were statistically significant associations between birth weight, weight in childhood and the HOMA index, and birth weight, BMI in childhood and the HOMA index (third and fourth models in Table 4). The first model described in Table 4 was also adjusted for weight and BMI in childhood (fifth and sixth models in Table 4); the associations between the HOMA index, birth weight and waist circumference remained significant.

Discussion

The children assessed in this cohort presented a considerable prevalence of low ( < 2500 g), deficient or insufficient (2500–2999 g) weight (Table 1); a factor of concern with regard to their potential growth and development and risk of morbidity–mortality(Reference Wallace, Levy and Matthews41).

According to the multiple linear regression analysis utilised in the present study, birth weight (as a continuous variable and categorised as SGA) and insulin resistance, assessed by the HOMA index, were associated. A reduction of 100 g in birth weight resulted in an increase of 0·026 mmol/l μU/ml in the HOMA index. The mean value of HOMA for these children was 1·0 (sd = 0·8; Table 2). Possibly, the increase in the HOMA index can have a clinical significance for the children. According to a large review on validation of the HOMA index, the parameter has an appropriate use in cohort and epidemiological studies for assessing insulin resistance(Reference Fowden, Albrecht and Pepe42).

Yajnig et al. (Reference Yajnik, Fall and Vaydia18) reported an association between insulin resistance and birth weight in 4-year-old Indian children. Four years later, his group of researchers studied the same children (n 477) and found out that the lower birth weight was associated with increased HOMA, and that the highest levels of insulin resistance syndrome variables and total and LDL cholesterol were in children of low birth weight (LBW) but high fat mass at 8 years(Reference Bavdekar, Yajnik and Fall19). Wilkin et al. (Reference Wilkin, Voss and Metcalf20) and Jeffery et al. (Reference Jeffery, Metcalf and Hosking21) reported, in the EarlyBird study, correlations between insulin resistance and weight in approximately 300 children at 5 and 8 years of age, but none with birth weight. The Pune Maternal Nutrition Study(Reference Joglekar, Fall and Deshpande24) assessed the associations of size and body proportions at birth, and growth during infancy and childhood, to body composition and CVD risk factors (insulin resistance, blood pressure, glucose tolerance and plasma lipids) in 698 children at the age of 6 years. The authors reported that smaller mid-upper arm circumference at 6 months predicted higher insulin resistance (P < 0·001), but larger mid-upper arm circumference at 1 year predicted higher systolic blood pressure (P < 0·001). Gupta et al. (Reference Gupta, Gupta and Pareek23) reported an association between insulin resistance and birth weight in 134 Indian children and adolescents from 5 to 16 years of age, controlling for waist circumference, waist:hip ratio, age, sex and BMI. Giapros et al. (Reference Giapros, Evagelidou and Challa22) observed that large for gestational age-born children aged 6·5–8 years had higher HOMA and adiponectin concentration than adequate for gestational age children.

The possible explanations for the association between birth weight and insulin resistance are the following: (a) most of the development of the pancreatic islets occurs in the uterus and, in human subjects, the mass of β-cells increases more than 130-fold between the 12th week of gestation and the 5th month after birth(Reference Yajnik44). Studies on animals have shown that nutritional deprivation within the uterus or just after birth may reduce the number of β-cells and the secretion of insulin, while noting that protein concentrations are directly related to the production of β-cells by the pancreas(Reference Hovi, Anderson and Eriksson45). In addition to lower insulin production, it can also be considered that LBW is related to glucose intolerance(Reference Krentz46); (b) under unfavourable conditions of intra-uterine nutrition, there is a reduction in the proportion of lean mass relative to body fat. This pattern favours insulin resistance as a mean of adaptation(Reference Molist-Brunet, Jimeno-Mollet and Franch-Nadal47). These abnormalities would cause structural changes that could persist into adulthood. The main strategy for fetal adaptation to deficient nutrition is selective redistribution of nutrients for the formation of fundamental organs like the brain (brain sparing). This mechanism would result in underdevelopment of the skeleton and liver, a lower proportion of muscle mass, diminished mass of β-cells from the pancreas and abnormalities in vascular tissues(Reference Yajnik, Godbole and Lubree48, Reference Levitt, Lambert and Woods49). A study carried out in South Africa(Reference McKeigue, Shah and Marmot50) involving 132 young adults indicates that the ‘fetal origins’ expression of the chronic disease phenotype is not dependent on birth weight alone, but on its interaction with subsequent fat accumulation, including measurements of waist circumference.

The present study also shows a positive association between waist circumference and the HOMA index. Mean waist circumferences of the children involved in the study were higher than mean waist circumference measurements of British children at the same age(Reference McCarthy, Jarrett and Crawley32). However, there were not many overweight or obese children, according to the Centers for Disease Control and Prevention or International Obesity Task Force classifications. The same phenomenon has been observed in Asia: central obesity even in the absence of obesity(Reference Yajnik51). Molist-Brunet et al. (Reference Molist-Brunet, Jimeno-Mollet and Franch-Nadal47) observed that as waist circumference increased the risk that insulin resistance would occur was higher. Yajnik(Reference Yajnik51), involving 8-year-old children, also found a relationship between HOMA, insulin, waist circumference and birth weight. A study carried out on the Caribbean island of Barbados showed an association between LBW girls with a family history of diabetes and higher HOMA. LBW girls without a family history of diabetes had a smaller waist circumference than normal birth weight girls. In contrast, LBW girls with a family history of diabetes had similar mean waist circumference than normal birth weight girls with a family history of diabetes(Reference Chambers, Tull and Fraser52).

As stated by Barker(Reference Barker, Winter and Osmmond15), the explanation for the accumulation of centrally located fat is that, during intra-uterine life, energy saving may occur in order to maintain brain development and functioning. These savings would generate a positive energy balance that, in adulthood, would cause an accumulation of fat, especially in the central region of the body. This centralisation of adiposity is related to insulin resistance and lower development of muscle mass, and it consists of a substrate enabling metabolic speed in providing glucose through stress-related metabolic pathways, with the aim of maintaining adequate glucose levels, particularly in the brain(Reference Bjorntorp53).

The fetal origins hypothesis, as related to insulin resistance, affirms the existence of a causal path from prenatal factors to insulin resistance. From the public health viewpoint, we want to know whether interventions undertaken in pregnancy might be of benefit to the health of the offspring later in life.

According to the present results, it is difficult to conclude that waist circumference is a potential pathway between birth weight and insulin resistance. Waist circumference in childhood may be influenced not only by birth weight(Reference Barker, Winter and Osmmond15), but probably by prenatal factors acting through effects on fetal growth (as measured by birth weight)(Reference Li, Manor and Power54) and factors occurring in childhood. Poor prenatal nutrition, for example, may act as a common cause of lower birth weight and higher waist circumference, determining insulin resistance.

Despite the non-significant result of the association between birth weight and the HOMA index (in the univariate linear regression analysis), a P-value of 0·06 has probably a biological value, considering the number of factors that could intervene in the association throughout prenatal life to childhood. However, if we consider that this association is not statistically significant, and that the associations between birth weight and waist circumference and waist circumference and insulin resistance are significant, we can probably suspect, according to Hernán et al. (Reference Hernán, Hernández-Díaz and Werler55), that lower birth weight is a common cause of higher waist circumference and insulin resistance.

Weinberg(Reference Weinberg56) asserted that aetiological inferences cannot be trusted to a regression package and that decisions about confounding and when to adjust can have a qualitative impact on our conclusions. According to the author, while we often cannot construct a complete causal pathway, we can gain important guidance from the attempt, clarify the assumptions we are making and then focus on those epidemiologic questions that are answerable.

Epidemiological studies have been conducted over the last decades in an attempt to determine the relationship between birth weight and outcomes in later life, most of them retrospective studies involving adult populations. Few prospective studies have been carried out on children(Reference Yajnik, Fall and Vaydia18Reference Joglekar, Fall and Deshpande24). To our knowledge, there is no Brazilian cohort study investigating the association between birth weight and metabolic abnormalities in this age group.

In conclusion, the results of the present study showed increased insulin resistance in apparently healthy, young children, who had lower weight at birth and higher measurements of waist circumference. There is a need to carry out large cohort epidemiological studies assessing several factors throughout pregnancy and childhood to a better understanding of the causal pathway between prenatal factors and insulin resistance. Some questions that now are quite difficult to answer in terms of causal pathways will probably be better understood, and help to develop public health policies that adopt preventive measures to promote adequate maternal-fetal and child development and enable early diagnosis of metabolic abnormalities.

Acknowledgements

The authors gratefully acknowledge Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP; Grant No. 04/04 109-8), for financing the present study, Prof. José Maria P. Souza, and the field workers Alessandra L. Santos, Rosemary F. Ferreira and Josimara F. Moura. There are no conflicts of interest to declare.

J. O. L. collected data, participated in the statistical analysis and interpretation of data, and writing of the paper. P. H. C. R. designed the study protocol, secured funding, facilitated data collection, participated in the statistical analysis and interpretation of data, and writing of the paper. J. A. P. collected data, participated in the statistical analysis and interpretation of data, and writing of the paper. R. G. O. participated in data collection, statistical analysis and writing of the paper. M. B. S. F. participated in the statistical analysis, interpretation of data and writing of the paper. P. B. S. participated in the statistical analysis, interpretation of data and writing of the paper.

References

1Yajnik, CS (2000) Interactions of perturbations in intrauterine growth during childhood on the risk of adult-onset disease. Proc Nutr Soc 59, 257265.CrossRefGoogle ScholarPubMed
2De Blasio, MJ, Gatford, KL, McMillen, IC, et al. (2007) Placental restriction of fetal growth increases insulin action, growth, and adiposity in the young lamb. Endocrinology 148, 13501358.CrossRefGoogle ScholarPubMed
3Barker, DJ (2007) The origins of the developmental origins theory. J Intern Med 261, 412417.CrossRefGoogle ScholarPubMed
4Barker, DJ (2005) The developmental origins of insulin resistance. Horm Res 64, Suppl. 3, 27.Google ScholarPubMed
5de Boo, HA & Harding, JE (2006) The developmental origins of adult disease (Barker) hypothesis. Aust N Z J Obstet Gynaecol 46, 414.CrossRefGoogle ScholarPubMed
6Eriksson, JG, Forsén, T, Tuomilehto, J, et al. (2002) Effects of size at birth and childhood growth on the insulin resistance syndrome in elderly individuals. Diabetologia 45, 342348.CrossRefGoogle ScholarPubMed
7Tanaka, Y, Kikuchi, T, Nagasaki, K, et al. (2005) Lower birth and visceral fat accumulation are related to hyperinsulinemia and insulin resistance in obese Japanese children. Hypertens Res 28, 529536.CrossRefGoogle ScholarPubMed
8Valerio, G, Licenziati, MR, Iannuzzi, A, et al. (2006) Insulin reistance and impaired glucose tolerance in obese children and adolescents from Southern Italy. Nutr Metab Cardiovasc Dis 16, 279284.CrossRefGoogle Scholar
9Mi, J, Cheng, H, Zhao, XY, et al. (2008) Developmental origin of metabolic syndrome: interaction of thinness at birth and overweight during adult life in Chinese population. Obes Rev 9 Suppl. 1, 9194.CrossRefGoogle ScholarPubMed
10Barker, DJP (2007) Obesity and early life. Obesity 8, Suppl. 1, 4549.Google ScholarPubMed
11Yudkin, JS (2007) Insulin resistance and the metabolic syndrome – or the pitfalls of epidemiology. Diabetologia 50, 15761586.CrossRefGoogle ScholarPubMed
12Law, CM (2002) Significance of birthweight for the future. Arch Dis Child Fetal Neonatal 86, 78.CrossRefGoogle ScholarPubMed
13Yajnik, CS (2004) Early life origins of insulin resistance and type 2 diabetes in India and other Asian Countries. J Nutr 134, 205210.CrossRefGoogle ScholarPubMed
14Yajnik, CS & Deshmukh, US (2008) Maternal nutrition, intrauterine programming and consequential risks in the offspring. Rev Endocr Metab Disord 9, 203211.CrossRefGoogle ScholarPubMed
15Barker, DJP, Winter, PD, Osmmond, C, et al. (1989) Weight in infancy and death from ischaemic heart disease. Lancet 334, 577580.CrossRefGoogle Scholar
16Ramachandran, A, Snehalatha, C, Latha, E, et al. (1997) Rising prevalence of NIDDM in an urban population in India. Diabetologia 40, 232237.CrossRefGoogle Scholar
17Ministério da Saúde (2004) Evolução da Mortalidade no Brasil (The Evolution of Mortality in Brazil). Vigilância em Saúde, MS: Ministério da Saúde.Google Scholar
18Yajnik, CS, Fall, CHD, Vaydia, U, et al. (1995) Fetal growth and glucose and insulin metabolism in four year old Indian children. Diabet Med 12, 330336.CrossRefGoogle ScholarPubMed
19Bavdekar, A, Yajnik, CS, Fall, CHD, et al. (1999) Insulin resistance syndrome in 8-year old Indian children. Small at birth, big at 8 years, or both? Diabetes 48, 24222429.CrossRefGoogle Scholar
20Wilkin, TJ, Voss, LD, Metcalf, BS, et al. (2004) Metabolic risk in early childhood: The EarlyBird Study. Int J Obes Relat Metab Disord 28, Suppl. 3, S64S69.CrossRefGoogle ScholarPubMed
21Jeffery, AN, Metcalf, BS, Hosking, J, et al. (2006) Little evidence for early programming of weight and insulin resistance for contemporary children: EarlyBird Diabetes Study report 19. Pediatrics 118, 11181123.CrossRefGoogle ScholarPubMed
22Giapros, V, Evagelidou, E, Challa, A, et al. (2007) Serum adiponectin and leptin levels and insulin resistance in children born large for gestational age are affected by the degree of overweight. Clin Endocrinol 66, 353359.Google ScholarPubMed
23Gupta, M, Gupta, R, Pareek, A, et al. (2007) Low birth weight and insulin resistance in mid and late childhood. Indian Pediatr 44, 177184.Google ScholarPubMed
24Joglekar, CV, Fall, CH, Deshpande, VU, et al. (2007) Newborn size, infant and childhood growth, and body composition and cardiovascular disease risk factors at the age of 6 years: The Pune Maternal Nutrition Study. Int J Obes 31, 15341544.CrossRefGoogle ScholarPubMed
25Rondó, PHC, Lemos, JO, Pereira, JA, et al. (2008) Relationship between birth weight and arterial elasticity in childhood. Clin Sci 115, 317326.CrossRefGoogle ScholarPubMed
26Williams, RL, Creasy, RK, Cunningham, MD, et al. (1982) Fetal growth and perinatal viability in California. Obstet Gynecol 59, 624632.Google ScholarPubMed
27Rondó, PHC, Ferreira, RF, Nogueira, F, et al. (2003) Maternal psychological stress and distress as predictors of low birth weight, prematurity and intra- uterine growth retardation. Eur J Clin Nutr 57, 266272.CrossRefGoogle Scholar
28Cameron, N (1984) The Measurement of Human Growth, 1st ed.London: Croom Helm.Google Scholar
29Jelliffe, DB & Jelliffe, EFP (1989) Community Nutritional Assessment, With Special Reference to Less Technically Developed Countries, 2nd ed.London: Oxford University Press.Google Scholar
30CDC – Centers for Disease Control and Prevention & National Center for Health Statistics (2000) CDC, United States of America.http://www.cdc.gov/growthcharts.Google Scholar
31Cole, TJ, Bellizzi, MC, Flegal, KM, et al. (2000) Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 320, 12401243.CrossRefGoogle ScholarPubMed
32McCarthy, HD, Jarrett, KV & Crawley, HF (2001) The development of waist circumference percentiles in British children aged 5·0–16·9 y. Eur J Clin Nutr 55, 901907.CrossRefGoogle ScholarPubMed
33American Diabetes Association Position Statement (2004) Diagnosis and classification of diabetes mellitus. Diabetes Care 27, Suppl. 1, 510.CrossRefGoogle Scholar
34Matthews, DR, Hosker, JP, Rudenski, AS, et al. (1985) Homeostasis model assessment: insulin resistance and beta cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28, 412419.CrossRefGoogle ScholarPubMed
35D'Annunzio, G, Vanelli, M, Meschi, F, et al. (2004) The SIEDP Study Group. Valori normali di- HOMA-IR in bambini e adolescents: studio multicentrico italiano. Quad Pediatr 3, 44.Google Scholar
36Friedwald, WT, Levy, RI & Fredrickson, DS (1972) Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultra centrifuge. Clin Chem 18, 499502.CrossRefGoogle Scholar
37Giuliano, ICB, Caramelli, B, Pellanda, L, et al. (2006) I Guideline for preventing atherosclerosis in childhood and adolescence. Int J Atheroscler 1, 130.Google Scholar
38Moraes, AS, Beltran, RJ, Mondini, L, et al. (2006) Prevalence of overweight and obesity, and associated factors in school children from urban area in Chilpancingo, Guerrero, Mexico, 2004. Cad Saude Publica 22, 12891301.CrossRefGoogle ScholarPubMed
39Jebb, SA, Rennie, KL & Cole, TJ (2004) Prevalence of overweight and obesity among young people in Great Britain. J Public Health Nutr 7, 461465.CrossRefGoogle ScholarPubMed
40Tu, YK, West, R, Ellison, GTH, et al. (2005) Why evidence for the fetal origins of adult disease might be a statistical artifact: the ‘reversal paradox’ for the relation between birth and blood pressure in later life. Am J Epidemiol 161, 2732.CrossRefGoogle ScholarPubMed
41Wallace, TM, Levy, JC & Matthews, DR (2004) Use and abuse of HOMA modeling. Diabetes Care 27, 14871495.CrossRefGoogle ScholarPubMed
42Fowden, AL (1985) Pancreatic endocrinology function and carbohydrate metabolism in the fetus. In Perinatal Endocrinology, vol. IV, pp. 7190 [Albrecht, E and Pepe, GJ, editors]. Ithaca, NY: Perinatology Press.Google Scholar
43Snoeck, A, Remacle, C, Reusens, B, et al. (1990) Effect of a low protein diet during pregnancy on the fetal rat endocrine pancreas. Biol Neonate 57, 107118.CrossRefGoogle ScholarPubMed
44Yajnik, CS (2002) The lifecycle effects of nutrition and body size on adult adiposity, diabetes and cardiovascular disease. Obes Res 3, 217224.CrossRefGoogle ScholarPubMed
45Hovi, P, Anderson, S, Eriksson, JG, et al. (2007) Glucose regulation in young adults with very low birth weight. N Engl J Med 358, 20532063.CrossRefGoogle Scholar
46Krentz, AJ (2002) Insulin Resistance: A Clinical Handbook, 1st ed.Oxford: Blackwell Science.CrossRefGoogle Scholar
47Molist-Brunet, N, Jimeno-Mollet, J & Franch-Nadal, J (2006) Correlation between the various measurements of obesity and the degree of resistance to insulin. Aten Primaria 37, 3036.CrossRefGoogle ScholarPubMed
48Yajnik, CS, Godbole, K & Lubree, H (2007) Fetal programming of type 2 diabetes: is sex important? Diabetes Care 30, 27542755.CrossRefGoogle ScholarPubMed
49Levitt, NS, Lambert, EV, Woods, D, et al. (2005) Adult BMI and fat distribution but not height amplify the effect of low birth weight on insulin resistance and increased blood pressure in 20-year-old South Africans. Diabetologia 48, 11181125.CrossRefGoogle Scholar
50McKeigue, PM, Shah, B & Marmot, MG (1991) Relation of central obesity and insulin resistance with diabetes prevalence and cardiovascular risk in South Asians. Lancet 337, 382386.CrossRefGoogle ScholarPubMed
51Yajnik, CS (2003) Nutrition, growth, and body size in relation to insulin resistance and type 2 diabetes. Curr Diab Rep 3, 108114.CrossRefGoogle ScholarPubMed
52Chambers, EC, Tull, ES, Fraser, H, et al. (2005) A family history of diabetes is related to abnormal insulin sensitivity in African-Caribbean girls of low birth weight: is catch-up weight important? Ethn Dis 15, 424428.Google ScholarPubMed
53Bjorntorp, P (1993) Visceral obesity: a ‘civilization syndrome’. Obes Res 1, 206222.CrossRefGoogle ScholarPubMed
54Li, L, Manor, O & Power, C (2004) Early environment and child-to-adult growth trajectories in the 1958 British birth cohort. Am J Clin Nutr 80, 185192.CrossRefGoogle ScholarPubMed
55Hernán, MA, Hernández-Díaz, S, Werler, M, et al. (2002) Causal knowledge as a prerequisite for confounding evaluation: an application to birth defects epidemiology. Am J Epidemiol 155, 176184.CrossRefGoogle ScholarPubMed
56Weinberg, RC (2005) Invited Commentary: Barker meets Simpson. Am J Epidemiol 161, 3335.CrossRefGoogle ScholarPubMed
Figure 0

Table 1 Characteristics of the children at birth(Mean values and standard deviations)

Figure 1

Table 2 Socio-demographic, nutritional and biochemical characteristics of the children(Mean values and standard deviations)

Figure 2

Table 3 Correlation coefficients between birth weight (BW) and the variables of interest

Figure 3

Table 4 Linear regression models considering insulin resistance (homeostasis model assessment, HOMA) as outcome