Abstract
Background: Birth weight and high weight gain the first years of life is associated with later body composition and increased risk of obesity. A detailed description of body composition at 3 years of age and factors related to the distribution of fat and fat-free mass is interesting, since body composition build up in childhood to a certain degree will be retained later in life. At the same time, research have shown that increased tempo of growth including length gain is related to an early nadir in the BMI-curve with a subsequent increase in BMI (adiposity rebound) which is also associated with later increased adiposity risk. Growth in early childhood is highly related to the levels of insulin-like growth factor-I.
Objectives: Data used in this thesis is from the prospective cohort study SKOT which includes 330 children examined at 9, 18 and 36 months of age. The main objective with this PhD study was to look at which factors were related to body composition at 3 years of age with a special focus on factors related to later obesity risk. A methodological objective was to be able to predict body composition by using bioelectrical impedance and anthropometry. Secondly, an objective was to identify growth periods with relation to body composition at 3 years and whether the impact of growth was modified by nutrition. A third objective was to examine the relation between IGF-I levels at 9 and 36 months and body composition at 3 years. Since the IGF-I level was expected to be affected by nutrition this was included in the analyses.
Results: In paper I, we showed that a simple model including bioelectrical impedance, height and weight was as good to predict fat-free mass as a more comprehensive model including the sum of subscapularis and triceps skinfolds as well. Prediction error for FFM was 3.0 % for both equations. Since fat mass constitutes a smaller part of the total weight, the relative prediction error was larger for fat mass (10.5 % in the comprehensive model and 12.0 % in the simple model). Results from paper II showed that birth weight and weight gain the first 0 - 5 months were strongly related to body composition at 3 years. No measures of adiposity (BMI, fat mass, fat mass index and skinfold thickness) were significantly associated to weight gain after 5 months of age. Two interesting effects modifications by full breastfeeding on the effect of birth weight and early growth on later fat mass index was found. With children that were fully breastfed for less than one months as reference group, we found that full breastfeeding for 4 - 5 months reduced the effect of early weight gain on fat mass index by 47 % (p = 0.05) while full breastfeeding for 6 months eliminated the effect of early growth on fat mass index (p = 0.002). Likewise, full breastfeeding for 6 months eliminated the effect of birth weight on fat mass index (p = 0.002). No effect modification was seen between breastfeeding and early growth on fat-free mass index at 3 years. In paper III, IGF-I concentrations at 9 and 36 months were positively related to height, weight, BMI, fat mass and fat free mass at 3 years, but not associated with percentage of body fat and fat mass index. A change in IGF-I from 9 to 36 months was positively related to fat-free mass and fat-free mass index but not with BMI, fat mass and fat mass index. Children breastfed at 9 months had lower IGF-I levels at 9 months but reached the same IGF-I levels at 36 months as infants not breastfed at 9 months. No associations were seen between IGF-I variables at 36 months and current protein, milk or meat intake at 36
months but IGF-I variables were negatively associated with intake of fat and saturated fat expressed as energy percentages.
Discussion: The predictive equation that was generated in paper I enabled good predictions of fat-free mass and fat mass for 233 children compared with the 101 children with high quality DXA scans. Due to the age-dependent variation in the hydration level of fat-free mass we expect that the predictive equation can be used in other studies with a study population at 2 - 4 years of age. Result from paper II confirmed that birth weight and early weight gain are important for later body composition which has been shown in a large number of studies. A lack of studies has focused on whether the adverse effects of rapid weight gain are modified by diet. We found that full breastfeeding for 4 - 6 months had a considerable modifying effect on the effect of weight gain on fat mass index, and full breastfeeding for 6 months eliminated the effect of birth weight on fat mass index. Of course parental choices of duration of full breastfeeding and introduction to solids can easily be influenced by early growth patterns as well as related to their health behavior in general. However, the interactions are physiologically plausible and should be investigated in other study settings. Results from paper III showed that total and free IGF-I at both 9 and 36 months were positively related to most measures of anthropometry and body composition at 3 years. We found no clear associations between IGF-I levels and early development of obesity. It is possible that IGF-I levels are related to later risk of obesity through increased tempo of weight gain and linear growth leading to early adiposity rebound but this was not visible at 3 years. We had expected that intake of protein and cow’s milk was positively associated with IGF-I values but this was not the case in this study. The negative associations between IGF-I levels and current intake of fat and saturated fat expressed as energy percentages could be related to the modulating effect of human milk on IGF-I, as human milk also contains high amounts of fat and saturated fat. However, the relation between diet, IGF-I and risk of early development of obesity is very complex and needs to be investigated further.
Conclusion and perspectives: This thesis represent a throughout examination of body composition at 3 years in a large group of healthy well-nourished Danish children. Birth weight and early weight gain were positively related with fat and fat-free mass at 3 years and the result indicates that the first five months of life is a sensitive period where it is possible to reduce the impact of high birth weight and rapid weight gain on later fat mass by continuing full breastfeeding for 4 - 6 months. The results therefore support the current recommendations for nutrition in infancy and are relevant in the discussion of early prevention of obesity. IGF-I was related to linear growth, fat mass and fat-free mass at 3 years but a relation between IGF-I and later obesity risk is speculative. A continuation of the SKOT cohort with a follow-up visit when the children have reached 7 - 8 years of age would be highly relevant and enable further elaboration on the findings presented in paper II and III. At this age most children have passed the adiposity rebound and a role for IGF-I on accelerated growth and age of the adiposity rebound could be further elucidated. Also a follow-up study of the same children would provide more knowledge of the relation between early growth and infant nutrition on later risk of overweight and obesity.
Objectives: Data used in this thesis is from the prospective cohort study SKOT which includes 330 children examined at 9, 18 and 36 months of age. The main objective with this PhD study was to look at which factors were related to body composition at 3 years of age with a special focus on factors related to later obesity risk. A methodological objective was to be able to predict body composition by using bioelectrical impedance and anthropometry. Secondly, an objective was to identify growth periods with relation to body composition at 3 years and whether the impact of growth was modified by nutrition. A third objective was to examine the relation between IGF-I levels at 9 and 36 months and body composition at 3 years. Since the IGF-I level was expected to be affected by nutrition this was included in the analyses.
Results: In paper I, we showed that a simple model including bioelectrical impedance, height and weight was as good to predict fat-free mass as a more comprehensive model including the sum of subscapularis and triceps skinfolds as well. Prediction error for FFM was 3.0 % for both equations. Since fat mass constitutes a smaller part of the total weight, the relative prediction error was larger for fat mass (10.5 % in the comprehensive model and 12.0 % in the simple model). Results from paper II showed that birth weight and weight gain the first 0 - 5 months were strongly related to body composition at 3 years. No measures of adiposity (BMI, fat mass, fat mass index and skinfold thickness) were significantly associated to weight gain after 5 months of age. Two interesting effects modifications by full breastfeeding on the effect of birth weight and early growth on later fat mass index was found. With children that were fully breastfed for less than one months as reference group, we found that full breastfeeding for 4 - 5 months reduced the effect of early weight gain on fat mass index by 47 % (p = 0.05) while full breastfeeding for 6 months eliminated the effect of early growth on fat mass index (p = 0.002). Likewise, full breastfeeding for 6 months eliminated the effect of birth weight on fat mass index (p = 0.002). No effect modification was seen between breastfeeding and early growth on fat-free mass index at 3 years. In paper III, IGF-I concentrations at 9 and 36 months were positively related to height, weight, BMI, fat mass and fat free mass at 3 years, but not associated with percentage of body fat and fat mass index. A change in IGF-I from 9 to 36 months was positively related to fat-free mass and fat-free mass index but not with BMI, fat mass and fat mass index. Children breastfed at 9 months had lower IGF-I levels at 9 months but reached the same IGF-I levels at 36 months as infants not breastfed at 9 months. No associations were seen between IGF-I variables at 36 months and current protein, milk or meat intake at 36
months but IGF-I variables were negatively associated with intake of fat and saturated fat expressed as energy percentages.
Discussion: The predictive equation that was generated in paper I enabled good predictions of fat-free mass and fat mass for 233 children compared with the 101 children with high quality DXA scans. Due to the age-dependent variation in the hydration level of fat-free mass we expect that the predictive equation can be used in other studies with a study population at 2 - 4 years of age. Result from paper II confirmed that birth weight and early weight gain are important for later body composition which has been shown in a large number of studies. A lack of studies has focused on whether the adverse effects of rapid weight gain are modified by diet. We found that full breastfeeding for 4 - 6 months had a considerable modifying effect on the effect of weight gain on fat mass index, and full breastfeeding for 6 months eliminated the effect of birth weight on fat mass index. Of course parental choices of duration of full breastfeeding and introduction to solids can easily be influenced by early growth patterns as well as related to their health behavior in general. However, the interactions are physiologically plausible and should be investigated in other study settings. Results from paper III showed that total and free IGF-I at both 9 and 36 months were positively related to most measures of anthropometry and body composition at 3 years. We found no clear associations between IGF-I levels and early development of obesity. It is possible that IGF-I levels are related to later risk of obesity through increased tempo of weight gain and linear growth leading to early adiposity rebound but this was not visible at 3 years. We had expected that intake of protein and cow’s milk was positively associated with IGF-I values but this was not the case in this study. The negative associations between IGF-I levels and current intake of fat and saturated fat expressed as energy percentages could be related to the modulating effect of human milk on IGF-I, as human milk also contains high amounts of fat and saturated fat. However, the relation between diet, IGF-I and risk of early development of obesity is very complex and needs to be investigated further.
Conclusion and perspectives: This thesis represent a throughout examination of body composition at 3 years in a large group of healthy well-nourished Danish children. Birth weight and early weight gain were positively related with fat and fat-free mass at 3 years and the result indicates that the first five months of life is a sensitive period where it is possible to reduce the impact of high birth weight and rapid weight gain on later fat mass by continuing full breastfeeding for 4 - 6 months. The results therefore support the current recommendations for nutrition in infancy and are relevant in the discussion of early prevention of obesity. IGF-I was related to linear growth, fat mass and fat-free mass at 3 years but a relation between IGF-I and later obesity risk is speculative. A continuation of the SKOT cohort with a follow-up visit when the children have reached 7 - 8 years of age would be highly relevant and enable further elaboration on the findings presented in paper II and III. At this age most children have passed the adiposity rebound and a role for IGF-I on accelerated growth and age of the adiposity rebound could be further elucidated. Also a follow-up study of the same children would provide more knowledge of the relation between early growth and infant nutrition on later risk of overweight and obesity.
| Original language | English |
|---|
| Publisher | Department of Nutrition, Exercise and Sports, Faculty of Sciences, University of Copenhagen |
|---|---|
| ISBN (Print) | 978 87 7611 667 5 |
| Publication status | Published - 2013 |