Introduction

While the first indication of a significant protective effect of breast-feeding against childhood obesity was published in 1981 by Kramer,¹ these findings were only confirmed recently when several large-scale epidemiological studies with an adequate study design and sufficient power were published.^{2, 3, 4, 5, 6, 7, 8, 9, 10, 11} Most of these studies were narratively reviewed in 2003,¹² however, without systematic identification of relevant studies. This is the first review on this issue applying a systematic identification of relevant studies and a prospective study protocol with a focus on careful adjustment of potential confounding factors. Publication bias and potential heterogeneity including its possible causes are assessed and a weighted odds ratio (OR) is calculated applying the established methods for meta-analyses.¹³

Methods

Inclusion criteria

The systematic review and the subsequent meta-analysis were conducted using a prospective study protocol, defining the inclusion criteria to be met on a priori by a person initially not familiar with the study results. For the assessment of the association between breast-feeding and obesity in childhood we included population-based cohort, cross-sectional or case–control studies. In cross-sectional studies outcome and exposure are assessed at the same time. However, in cross-sectional studies on breast-feeding the exposure is ascertained retrospectively at the time of assessment of obesity, whereas in cohort studies the exposure is assessed in the first year of life and obesity is measured later on. In retrospective cohort studies, the information on the exposure is retrospectively abstracted from files documented at recruitment. The first step was a systematic review of all eligible studies on children older than 1 y at the last follow-up stage. Studies had to be published in English, French, Italian, Spanish or German. In the second step a meta-analysis was conducted. Only studies with adjustment for at least three of the following relevant confounding or interacting factors birth weight, parental overweight, parental smoking, dietary factors, physical activity and socioeconomic status were included in this meta-analysis. As suggested by Liberatos et al,¹⁴ we accepted parental education as an appropriate indicator of socioeconomic status. Other inclusion criteria were: comparable risk estimates as OR or relative risk had to be reported and age at the last follow-up had to be between 5 and 18 y; feeding-mode had to be assessed and reported and obesity as outcome had to be defined by body mass index (BMI) percentiles 90, 95 or 97 kg/m², to allow for comparison of the studies. We did not require all studies to use identical reference values. If risk estimates were calculated for different percentile values in a particular study, the estimate for the highest percentile-value was included in the meta-analysis.

Identification of studies

The systematic computerized literature search of published studies was carried out in January 2004. The search was conducted in Silver Platter Medline from 1966 to December 2003, in Embase from 1988 to December 2003, in the Cochrane Library 2/2003 and in Google (internet search engine) using the following search terms: 'breast-feeding or breast feeding or breastfeeding' and 'overweight or adiposity^* or fat^* or obesity' and 'child^* or adolescent^*'. No further limitations were made in order to be as sensitive as possible. Identification of further studies was made through handsearch in original articles and reviews on the subject found by electronic search.

Data extraction and statistical analysis

Data of each study were extracted using a standardized protocol to assess mode of feeding, outcome and potential confounders. The pooled crude and adjusted odds ratios (AOR) of the studies meeting the inclusion criteria were calculated in a fixed-effects model. The assumption of the fixed-effects model is that all studies measure the same effect. Heterogeneity was tested using a Q-test.¹⁵ Additionally, random-effects pooled estimates using the DerSimonian and Laird method¹⁶ were calculated, in which the study effects are treated as normally distributed around a population mean. Sensitivity analyses were carried out to detect potential sources of heterogeneity by testing the stability of the findings across different approaches in study design, exposure ascertainment and selection of study participants.¹³ The studies were stratified by the definition of breast-feeding, definition of obesity, number of potential confounders considered for in the adjustment, type of the study or age of the children at follow-up. Additionally, we assessed the potential impact of inclusion of other studies—not matching the inclusion criteria for the meta-analysis—on the pooled estimates.

Publication bias was assessed by a funnel plot. The log of the ORs, representing estimates of the effects of breast-feeding on childhood obesity, were plotted against the inverse of the standard error of the log OR as a measure of precision reflecting the study size. The aim is to assess symmetry as an indicator of the absence of publication bias. A funnel plot regression was additionally computed.¹⁷ In this approach, the degree of funnel plot asymmetry can be measured by the intercept from a linear regression of the standardized effect sizes against precision. In the absence of publication bias this intercept will be zero.

All analyses were performed with SAS software package, version 8.

Results

Identification of studies

The applied search strategy yielded 953 potentially relevant publications in Medline and Embase. One additional article was found in the citations of the relevant studies by manual search.¹⁸ No additional studies were found in the Cochrane Library and Google search. The evaluation of the 954 publications is shown in Figure 1.

Figure 1.

Breast-feeding and obesity. Identification and evaluation of the studies for the meta-analysis.

Full figure and legend (84K)

Results of the systematic review

In total, 28 studies met the inclusion criteria for the systematic review, 19 of them were not eligible for the meta-analysis^{1, 2, 11, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34} because they did not meet the inclusion criteria of the meta-analysis (Table 1). Most of the studies excluded from the meta-analysis were small, only nine of them had a sample size of more than 100 per feeding group.^{2, 11, 22, 23, 25, 26, 30, 32, 33} Six of these nine studies did not report adequate assessment and adjustment for potential confounding factors so that inclusion criteria were not met.^{22, 23, 25, 30, 32, 33} The study of Wadsworth was published as a letter to the editor and adjusted for potential confounding factors, but did not report an AOR for breast-feeding.³³ In the case–control study of Kramer,¹ obesity was defined as weight/height >120% or skinfold thickness >95th percentile or both >90th percentile, which did not match our study criteria. The cohort study of Armstrong et al² included children at an age of 39–42 months at follow-up. The recently published study of Parsons et al²⁶ did not report AORs for the age group of 7–16 y, whereas in the study of Victora et al¹¹ the inclusion criteria of definition of overweight or obesity were not met. The studies of Kramer¹ and of Armstrong et al² showed a significant protective effect of breast-feeding, whereas the studies of Wadsworth et al,³³ Parsons et al²⁶ and Victora et al¹¹ failed to show such an association. Reasons for exclusion from the meta-analysis are shown in Table 1.

Table 1 - Characteristics of the studies that examined the association between breast-feeding and obesity in children, but did not meet the inclusion criteria for the meta-analysis.

Full table

Results of the meta-analysis

Nine studies were eligible for the meta-analysis comprising more than 69 000 children included in the final analysis. Table 2 shows characteristics of these studies and potential confounders, for which adjustment was made.

Table 2 - Characteristics of the studies that met the inclusion criteria for the meta-analysis.

Full table

Six studies used cutoff values for the respective BMI percentiles based on internal^{8, 9} or national^{4, 5, 6, 10} reference populations, whereas two studies used the reference populations according to the recommendations of the international obesity task force (IOTF)³⁵ or Rolland-Cachera et al.³⁶

One of the studies included was a cohort study with longitudinal follow-up of the participants at several ages.⁷ For this study, conducted by Poulton et al, we calculated a weighted estimate for the age-groups between 5 and 18 y. Children who had never been breastfed were compared to children breastfed for more than 6 months. This weighted AOR was used in the meta-analysis.

The pooled crude OR for breast-feeding and obesity defined as BMI 90th, 95th or 97th percentile could be calculated for six studies^{3, 4, 5, 8, 9, 18} (Figure 2). The three other studies lacked information to calculate crude OR with confidence intervals.^{6, 7, 10} Under the fixed-effects assumption, the OR was 0.67, 95% CI (0.62, 0.73). In the random-effects model an almost identical OR was found (data not shown). There was no indication of significant heterogeneity of the studies: the null hypothesis of homogeneity could not be rejected (Q=7.8, df=5, P>0.1).

Figure 2.

Effect of breast-feeding vs formula feeding on childhood obesity: crude odds ratios of six studies and pooled odds ratio.

Full figure and legend (11K)

The AOR calculated for nine studies^{3, 4, 5, 6, 7, 8, 9, 10, 18} (Figure 3) was 0.78, 95% CI (0.71, 0.85) for both fixed and random-effects model, suggesting that there was no heterogeneity between the studies (Figure 3). The homogeneity hypothesis could not be rejected by the Q-test (Q=9.4, df=8, P>0.3). Although the studies appeared to be homogeneous, sensitivity analyses were performed, comparing the studies according to the following criteria: cohort study or cross-sectional study, different definitions of breast-feeding, different definitions of obesity, different age-groups and number of potential confounders considered for adjustment. The protective effect of breast-feeding was more pronounced in studies with adjustment for less than seven potential confounding factors compared to adjustment for seven or more potential confounding factors (Table 3).

Figure 3.

Effect of breast-feeding vs formula feeding on childhood obesity: covariate-adjusted odds ratios of nine studies and pooled odds ratio.

Full figure and legend (14K)

Table 3 - Sensitivity analyses of studies that met the inclusion criteria for the meta-analysis.

Full table

Eight of the studies provided data about a relationship between breast-feeding duration and risk of overweight or obesity in later childhood (Table 4). Four studies demonstrated an inverse association of breast-feeding duration and prevalence of obesity both in the crude and the adjusted analysis,^{4, 6, 8, 9} whereas one study found a dose–response contingency in the crude analysis that lost statistical significance after adjustment.⁷ Three studies found no significant effect of duration of breast-feeding.^{5, 10, 18}

Table 4 - Characteristics of the studies that met the inclusion criteria for the meta-analysis and examined the association of breast-feeding duration with childhood obesity.

Full table

The funnel plot showed an asymmetric pattern, which was due to a particular study (Figure 4). The funnel plot regression analysis did not reject the null hypothesis of symmetry (df=8, P=0.71), suggesting that there was no publication bias.

Figure 4.

Funnel plot. Log (AOR) of the individual studies (horizontal axis) plotted against 1/se log (AOR) (vertical axis).

Full figure and legend (7K)

Discussion

A classical meta-analysis requires randomized controlled trials. Randomization of breast-feeding on an individual level is not ethical, however. Trials with randomized intervention on hospital level as recently published by Kramer et al³⁷ might provide data that will allow for better assessment of the effects of breast-feeding on obesity. Unfortunately, there are no cluster-randomized controlled trials on breast-feeding and obesity published yet. So far only cohort, cross-sectional and case–control studies have been performed on this issue.

Definition of obesity

It has been reported that breast-feeding does not shift the whole distribution of BMI to the left, but only the upper tail as clearly shown in the publication of Koletzko et al.³⁸ We are not aware of other studies on the interdependencies of breast-feeding and childhood obesity taking into account both an effect on the median and the upper tail of the distribution. Therefore, we can only hypothesize that the impact on the upper tail of the distribution is the genuine effect of breast-feeding. Interestingly, the obesity epidemic in children as opposed to adults is predominantly caused by an increase of the BMI distribution in the upper percentiles.^{39, 40} Therefore, it appears reasonable to consider exposures, which affect the upper tail mainly. We confined our analyses to studies with overweight and obesity as end points. Although the studies used different percentiles for the definition of obesity, the results were comparable and the sensitivity analysis showed no difference between a cutoff at the 90th, 95th and 97th percentile.

Definition of breast-feeding

The assessment of exposure to breast-feeding differed from study to study: Gillman et al⁴ compared children who were mostly or only breastfed in the first 6 months with children mostly or only formula fed, while most of the studies compared children who were never breastfed with children ever breastfed (Table 2). However, in a sensitivity analysis, homogeneity between the studies stratified by different definitions of breast-feeding could not be rejected.

Residual confounding

Although only studies with adjustment for at least three confounding variables were included in this meta-analysis, residual confounding remains a potential limitation. Residual confounding is present if a confounding factor cannot be assessed with sufficient precision and randomization is not feasible. This represents a widespread problem in epidemiological research.^{13, 41} Socioeconomic status in particular is likely to confound the relationship between breast-feeding and childhood obesity. The number of indicators to assess socioeconomic status used in the studies considered varied: in those using one indicator only there might be residual confounding. Even in the study of Bergmann et al,³ however, which used all three indicators parental education, profession and income, socioeconomic status could not fully explain the protective effect of breastfeeding on childhood obesity, supporting the findings of Liberatos et al,¹⁴ that parental education is a highly relevant indicator of socioeconomic status.

In our meta-analysis potential confounding factors as birth weight, maternal overweight, maternal smoking and socioeconomic status contributed to a difference between adjusted and crude OR of up to 0.14 in the individual studies. Even in the three studies, in which the investigators adjusted for more than six potential confounding factors, the protective effect of breast-feeding remained significant with a pooled OR of 0.78, however (Table 3). This indicates that further adjustment for more relevant variables might reduce the calculated effect, but the protective effect of the breast-feeding is unlikely to be reduced to zero.

Publication bias

The funnel plot of the included studies showed an asymmetric pattern, which was due to the particular study of Bergmann et al.³ This study is diverging to the left-hand side. The funnel plot regression, however, indicated no significant publication bias. Both approaches have limitations. The shape of a funnel plot can show great variation depending on the way of construction.⁴² We used the method suggested by Sterne et al.⁴³ The validity of the regression method is limited because of low statistical power in meta-analyses of less than 20 studies.⁴³

To assess potential selection bias, a pooled estimate of all eligible studies that reported AORs with confidence intervals—including also studies excluded from the original meta-analysis^{2, 3, 4, 5, 6, 7, 8, 9, 10, 18, 26}—was calculated. The AOR of 0.77 (95% CI: 0.72, 0.82) was similar to the base case including only studies matching the inclusion criteria^{3, 4, 5, 6, 7, 8, 9, 10, 18} (AOR 0.78). Nevertheless publication bias cannot be excluded definitely, because some studies, which found no significant effect in a crude analysis, did not report adjusted estimates and therefore had to be excluded from the meta-analysis. Including these studies might reduce the protective effect of breast-feeding; however, most of the recently published studies with weak or absent effects in the crude analysis presented estimates with adjustment for confounding.^{5, 7, 10, 11, 18, 26}

Possible mechanisms

The inverse association between breast-feeding and obesity suggests a role for breast-feeding in the reduction of obesity prevalence in later life. Possible explanations were extensively discussed in earlier studies, reviews and editorials^{12, 44} and include behavioural and hormonal mechanisms and differences in macronutrient intake. Higher plasma-insulin concentrations in bottle-fed compared to breast-fed infants could stimulate fat deposition and lead to early development of adipocytes.⁴⁵ Bioactive factors in breast-milk might modulate growth factors, which inhibit adipocyte differentiation in vitro.^{46, 47} Furthermore, protein intake and amount of energy metabolisation is lower in breastfed than in formula-fed infants.⁴⁸ A longitudinal study showed a significant association between early protein intake and later BMI,⁴⁹ suggesting that a higher protein intake early in life might increase the risk of later obesity. In animal studies the availability of protein during fetal and early postnatal development was found to have a long-term effect on the metabolic programming of glucose metabolism and body composition in later life.^{50, 51} These pathways—alone or in combination—provide plausible explanations for a protective effect of breast-feeding against obesity.

Conclusion

This meta-analysis indicates that breast-feeding is associated with a small but consistent protective effect against obesity risk in later childhood.

References

1. Kramer MS. Do breast-feeding and delayed introduction of solid foods protect against subsequent obesity? J Pediatr 1981; 98: 883–887. | PubMed | ChemPort |
2. Armstrong J, Reilly JJ. Breastfeeding and lowering the risk of childhood obesity. Lancet 2002; 359: 2003–2004. | Article | PubMed | ISI |
3. Bergmann KE, Bergmann RL, Von Kries R, Bohm O, Richter R, Dudenhausen JW, Wahn U. Early determinants of childhood overweight and adiposity in a birth cohort study: role of breast-feeding. Int J Obes Relat Metab Disord 2003; 27: 162–172. | Article | PubMed | ChemPort |
4. Gillman MW, Rifas-Shiman SL, Camargo Jr CA, Berkey CS, Frazier AL, Rockett HR, Field AE, Colditz GA. Risk of overweight among adolescents who were breastfed as infants. JAMA 2001; 285: 2461–2467. | Article | PubMed | ISI | ChemPort |
5. Hediger ML, Overpeck MD, Kuczmarski RJ, Ruan WJ. Association between infant breastfeeding and overweight in young children. JAMA 2001; 285: 2453–2460. | Article | PubMed | ISI | ChemPort |
6. Liese AD, Hirsch T, von Mutius E, Keil U, Leupold W, Weiland SK. Inverse association of overweight and breast feeding in 9 to 10-y-old children in Germany. Int J Obes Relat Metab Disord 2001; 25: 1644–1650. | Article | PubMed | ChemPort |
7. Poulton R, Williams S. Breastfeeding and risk of overweight. JAMA 2001; 286: 1449–1450. | PubMed | ISI | ChemPort |
8. Toschke AM, Vignerova J, Lhotska L, Osancova K, Koletzko B, Von Kries R. Overweight and obesity in 6- to 14-year-old Czech children in 1991: protective effect of breast-feeding. J Pediatr 2002; 141: 764–769. | Article | PubMed | ISI |
9. von Kries R, Koletzko B, Sauerwald T, von Mutius E, Barnert D, Grunert V, von Voss H. Breast feeding and obesity: cross sectional study. BMJ 1999; 319: 147–150. | PubMed | ChemPort |
10. Li L, Parsons TJ, Power C. Breast feeding and obesity in childhood: cross sectional study. BMJ 2003; 327: 904–905. | Article | PubMed | ChemPort |
11. Victora CG, Barros F, Lima RC, Horta BL, Wells J. Anthropometry and body composition of 18 year old men according to duration of breast feeding: birth cohort study from Brazil. BMJ 2003; 327: 901. | Article | PubMed |
12. Dewey KG. Is breastfeeding protective against child obesity? J Hum Lact 2003; 19: 9–18. | Article | PubMed |
13. Egger M, Schneider M, Davey Smith G. Spurious precision? Meta-analysis of observational studies. BMJ 1998; 316: 140–144. | PubMed | ChemPort |
14. Liberatos P, Link BG, Kelsey JL. The measurement of social class in epidemiology. Epidemiol Rev 1988; 10: 87–121. | PubMed | ISI | ChemPort |
15. Normand SL. Meta-analysis: formulating, evaluating, combining, and reporting. Stat Med 1999; 18: 321–359. | Article | PubMed | ISI | ChemPort |
16. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: 177–188. | Article | PubMed | ISI | ChemPort |
17. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997; 315: 629–634. | PubMed | ChemPort |
18. O'Callaghan MJ, Williams GM, Andersen MJ, Bor W, Najman JM. Prediction of obesity in children at 5 years: a cohort study. J Paediatr Child Health 1997; 33: 311–316. | PubMed |
19. Agras WS, Kraemer HC, Berkowitz RI, Hammer LD. Influence of early feeding style on adiposity at 6 years of age. J Pediatr 1990; 116: 805–809. | PubMed |
20. Agras WS, Kraemer HC, Berkowitz RI, Korner AF, Hammer LD. Does a vigorous feeding style influence early development of adiposity? J Pediatr 1987; 110: 799–804. | PubMed |
21. Elliott KG, Kjolhede CL, Gournis E, Rasmussen KM. Duration of breastfeeding associated with obesity during adolescence. Obes Res 1997; 5: 538–541. | PubMed | ISI | ChemPort |
22. Frye C, Heinrich J. Trends and predictors of overweight and obesity in East German children. Int J Obes Relat Metab Disord 2003; 27: 963–969. | Article | PubMed | ChemPort |
23. Haschke F, van't Hof M. The influence of nutritional and genetic factors on growth and BMI until 5 years of age. Monatsschr Kinderheilk 2003; 151 (Suppl 1): S54–S57. | Article |
24. Ivanovic D, Ivanovic R, Buitron C. Nutritional status, birth weight and breast feeding of elementary first grade Chilean students. Nutr Rep Int 1987; 36: 1347–1361. | PubMed |
25. Langnase K, Mast M, Danielzik S, Spethmann C, Muller MJ. Socioeconomic gradients in body weight of German children reverse direction between the ages of 2 and 6 years. J Nutr 2003; 133: 789–796. | PubMed | ChemPort |
26. Parsons TJ, Power C, Manor O. Infant feeding and obesity through the lifecourse. Arch Dis Child 2003; 88: 793–794. | Article | PubMed | ChemPort |
27. Ravelli AC, van der Meulen JH, Osmond C, Barker DJ, Bleker OP. Infant feeding and adult glucose tolerance, lipid profile, blood pressure, and obesity. Arch Dis Child 2000; 82: 248–252. | Article | PubMed | ChemPort |
28. Richter J. Influence of duration of breast-feeding on body-weight-development. Arztl Jugendkd 1981; 72: 166–169. | PubMed |
29. Scaglioni S, Agostoni C, Notaris RD, Radaelli G, Radice N, Valenti M, Giovannini M, Riva E. Early macronutrient intake and overweight at five years of age. Int J Obes Relat Metab Disord 2000; 24: 777–781. | Article | PubMed | ChemPort |
30. Strbak V, Skultetyova M, Hromadova M, Randuskova A, Macho L. Late effects of breast-feeding and early weaning: seven-year prospective study in children. Endocr Regul 1991; 25: 53–57. | PubMed |
31. Tanasescu M, Ferris AM, Himmelgreen DA, Rodriguez N, Perez-Escamilla R. Biobehavioral factors are associated with obesity in Puerto Rican children. J Nutr 2000; 130: 1734–1742. | PubMed | ChemPort |
32. Tulldahl J, Pettersson K, Andersson SW, Hulthen L. Mode of infant feeding and achieved growth in adolescence: early feeding patterns in relation to growth and body composition in adolescence. Obes Res 1999; 7: 431–437. | PubMed | ISI | ChemPort |
33. Wadsworth M, Marshall S, Hardy R, Paul A. Breast feeding and obesity. Relation may be accounted for by social factors. BMJ 1999; 319: 1576. | PubMed | ChemPort |
34. Wolman PG. Feeding practices in infancy and prevalence of obesity in preschool children. J Am Diet Assoc 1984; 84: 436–438. | PubMed |
35. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320: 1240–1243. | Article | PubMed | ChemPort |
36. Rolland-Cachera MF, Cole TJ, Sempe M, Tichet J, Rossignol C, Charraud A. Body mass index variations: centiles from birth to 87 years. Eur J Clin Nutr 1991; 45: 13–21. | PubMed | ChemPort |
37. Kramer MS, Chalmers B, Hodnett ED, Sevkovskaya Z, Dzikovich I, Shapiro S, Collet JP, Vanilovich I, Mezen I, Ducruef T, Shishko G, Zubovich V, Mknuik D, Gluchanina E, Dombrovskiy V, Ustinovitch A, Kot T, Bogdanovich N, Ovchinikova L, Helsing E. Promotion of Breastfeeding Intervention Trial (PROBIT): a randomized trial in the Republic of Belarus. JAMA 2001; 285: 413–420. | Article | PubMed | ISI | ChemPort |
38. Koletzko B, von Kries R. Are there long term protective effects of breast feeding against later obesity? Nutr Health 2001; 15: 225–236. | PubMed | ChemPort |
39. Flegal KM, Troiano RP. Changes in the distribution of body mass index of adults and children in the US population. Int J Obes Relat Metab Disord 2000; 24: 807–818. | Article | PubMed | ChemPort |
40. Kalies H, Lenz J, von Kries R. Prevalence of overweight and obesity and trends in body mass index in German pre-school children, 1982–1997. Int J Obes Relat Metab Disord 2002; 26: 1211–1217. | Article | PubMed | ChemPort |
41. Blettner M, Sauerbrei W, Schlehofer B, Scheuchenpflug T, Friedenreich C. Traditional reviews, meta-analyses and pooled analyses in epidemiology. Int J Epidemiol 1999; 28: 1–9. | Article | PubMed | ISI | ChemPort |
42. Tang JL, Liu JL. Misleading funnel plot for detection of bias in meta-analysis. J Clin Epidemiol 2000; 53: 477–484. | Article | PubMed | ISI | ChemPort |
43. Sterne JA, Egger M, Smith GD. Systematic reviews in health care: investigating and dealing with publication and other biases in meta-analysis. BMJ 2001; 323: 101–105. | Article | PubMed | ChemPort |
44. Dietz WH. Breastfeeding may help prevent childhood overweight. JAMA 2001; 285: 2506–2507. | Article | PubMed |
45. Lucas A, Sarson DL, Blackburn AM, Adrian TE, Aynsley-Green A, Bloom SR. Breast vs bottle: endocrine responses are different with formula feeding. Lancet 1980; 1: 1267–1269. | Article | PubMed | ChemPort |
46. Hauner H, Röhrig K, Petruschke T. Effects of epidermal growth factor (EGF), platelet-derived growth factor (PDGF) and fibroblast growth factor (FGF) on human adipocyte development and function. Eur J Clin Invest 1995; 25: 90–96. | PubMed |
47. Petruschke T, Röhrig K, Hauner H. Transforming growth factor beta (TGF-beta) inhibits the differentiation of human adipocyte precursor cells in primary culture. Int J Obes Relat Metab Disord 1994; 18: 532–536. | PubMed | ChemPort |
48. Whitehead RG. For how long is exclusive breast-feeding adequate to satisfy the dietary energy needs of the average young baby? Pediatr Res 1995; 37: 239–243. | PubMed | ChemPort |
49. Rolland-Cachera MF, Deheeger M, Akrout M, Bellisle F. Influence of macronutrients on adiposity development: a follow up study of nutrition and growth from 10 months to 8 years of age. Int J Obes Relat Metab Disord 1995; 19: 573–578. | PubMed | ChemPort |
50. Desai M, Hales CN. Role of fetal and infant growth in programming metabolism in later life. Biol Rev Camb Philos Soc 1997; 72: 329–348. | Article | PubMed |
51. Burns SP, Desai M, Cohen RD, Hales CN, Iles RA, Germain JP, Going TC, Bailey RA. Gluconeogenesis, glucose handling, and structural changes in livers of the adult offspring of rats partially deprived of protein during pregnancy and lactation. J Clin Invest 1997; 100: 1768–1774. | PubMed |

Acknowledgements

This study has been carried out with support from the Bavarian Ministry for Environment, Health and Consumer Protection, Germany and partial financial support from the Commission of the European Communities, specific RTD programme 'Quality of Life and Management of Living Resources', QLK1-2001-00389 'Childhood Obesity: Early Programming by Infant Nutrition?' (CHOPIN). It does not necessarily reflect its views and in no way anticipates the Commission's future policy in this area.