Long Term Effects of Infant Feeding

Much of the literature on the long-term effects of infant feeding is based on comparison of outcomes associated with human milk versus infant formula feeding. Postulated effects relate primarily to the different composition of human milk versus formula and different energy and nutrient intake by infants. The literature does not provide a clear and consistent picture of the long-term effects of feeding. When effects are found, they tend to be modest. Before discussing the results of these studies, it is important to raise several important methodological issues relevant to the interpretation of the literature.

First, breast feeding is a complex behavior chosen by mothers. Women who choose to breast-feed are likely to differ in systematic ways from those who do not. The choice to breast feed and the duration of breast feeding may be related to other short- and long-term health behaviors that affect the ultimate health outcomes of interest. To isolate the effect of infant feeding, it must be assumed that other concurrent and subsequent exposures are not systematically related to feeding history, or such exposures must be taken into account in multivari-ate analysis. Unfortunately, most studies have insufficient data to adequately control statistically for these other behaviors, particularly since they are often unmeasured or poorly measured.

Second, many studies use historical cohorts in which feeding method is recalled by the mother or based on limited records. While the decision to initiate breast feeding is likely to be accurately recalled, information about breast feeding duration and timing of introduction of other foods may be subject to recall bias.

Third, the composition of proprietary infant formulas has changed since their introduction in the 1920s. For example, sodium levels and fat sources have changed, and new ingredients such as n-3 fatty acids and nucleotides have been added recently. Therefore, results from older versus younger cohorts may differ either because true age-specific effects have emerged or because they were exposed to infant formula of different composition. Furthermore, the effects of breast and formula feeding on infant health are likely to differ depending on the environmental context.

The ideal study design for determining the long-term effects of infant feeding would require randomization to feeding regimens, and frequent follow-up of subjects up to the time when a disease risk factor or outcome is measured. Such designs are rarely ethical or feasible. An exception is a series of studies in the UK conducted by Alan Lucas and colleagues, which assessed long-term outcomes among preterm infants randomized to receive banked human milk or formula, and full-term infants whose mothers chose not to breast-feed randomized to different types of formula. While many of the studies have focused on neurodevelopment, some are now looking at other health outcomes.

Selected Outcomes Related to Infant Feeding

The following are examples of some chronic disease-related outcomes studied in relation to infant feeding. The selected outcomes are intended to be illustrative of a range of effects rather than a comprehensive treatment of all outcomes related to infant feeding.

Serum lipids Based on a systematic review of literature relating infant feeding to blood lipids in infants, adolescents, and adults, total cholesterol was found to be consistently higher in breast-fed infants compared to bottle-fed infants. No consistent differences related to feeding history were found in children and adolescents; and among adults, a majority of studies reported lower mean total cholesterol in those who had been breast-fed. The proposed but unproven mechanism for the protective effect of breast-feeding in adults is downregulation of endogenous cholesterol synthesis.

Blood pressure Differences in the sodium and fat content and composition of breast milk versus formula are thought to be the relevant determinants of long-term effects of infant feeding on blood pressure. In a recent systematic review, data were compiled to compare exclusive breast feeding to formula feeding, with adjustment for current age, sex, height, and body mass index (BMI). The analysis was based on 26 studies of systolic blood pressure and 24 studies of diastolic blood pressure. On average, subjects who were breast-fed had a modestly lower systolic blood pressure than those who had been formula fed, with an average effect of —1.10 mmHg, and no marked differences by age. However, the analysis suggested publication bias since the effect was significantly larger in small studies than large studies. The studies showed no effects of feeding on diastolic blood pressure.

Taking advantage of a 1980 randomized trial to study the effect of a low or normal sodium diet in Dutch infants, a follow-up study at age 15 years found systolic blood pressure to be 3.6 mmHg lower and diastolic to be 2.2 mmHg lower in the low-sodium group. These results suggest that sodium intake in infancy may affect blood pressure later in life.

Further evidence of the effects of diet composition comes from a long-term follow-up of the Barry Caerphilly Growth study cohort. In this study, mothers and their offspring were randomly assigned to receive a milk supplement or usual care. In young adulthood (age 23-27years), blood pressure was positively associated with dried formula milk supplement consumed in infancy. The effect was attenuated but remained significant after controlling for current BMI, suggesting an effect of diet composition independent of growth.

Reproductive function The relatively high levels of isoflavones in soy-based infant formula have raised concerns about potential effects on endocrine and reproductive function later in life. A recent retrospective cohort study of young adults who as infants had participated in controlled feeding studies during infancy found no differences associated with soy feeding across a large number of outcomes potentially susceptible to estrogenic or antiestrogenic activity of phytoestrogens, including timing of maturation, sexual development, or fertility in adolescents or adults. Another literature review reported no meaningful differences in child growth related to feeding of soy formula. However, data are limited and further randomized controlled trials are needed to provide definitive evidence.

Growth and body composition Mode of feeding may indirectly affect later disease risk through its effects on energy intake or aspects of metabolic regulation that affect growth and body composition.

Numerous studies demonstrate different growth patterns in breast- and formula-fed infants that are hypothesized to reflect differences in nutrient intakes. In fact, evidence of systematic differences in breast- and formula-fed infants has led the World Health Organization to undertake the development of growth charts for breast-fed infants. In one careful study of body composition, total energy intakes and weight velocity from 3 to 6 months of age were higher in formula-fed compared to breastfed infants. Estimates of fat and fat-free mass also indicate higher adiposity in formula-fed infants, however, none of these differences persisted into the second year of life. Similarly, in a study of nearly 5600 children who participated in the Third National Health and Nutrition Examination Survey, those who had been exclusively breast-fed for

4 months weighed less at 8-11 months than did infants who were fed in other ways, but few other meaningful differences in growth status through age

5 years were associated with early infant feeding.

Longer term effects of infant feeding have been assessed in studies that examined whether breastfeeding protects against later overweight or obesity. A recent review found inconsistent results, with some large cohort studies showing a moderate protective effect, and others showing no effect. The studies were also inconsistent in showing a dose response. An illustrative large study in 3-5-year-old children found that after adjusting for potential con-founders, risk of having a BMI > 85th percentile was reduced among exclusively breast-fed children compared with those never breast fed, but there was no reduced risk of having a BMI > 95th percentile.

The findings are typically based on retrospective studies, in which breast-feeding data derive from maternal recall. This makes it difficult, if not impossible, to control for confounding, since a mother's decisions about breast-feeding may relate to subsequent child feeding and other factors associated with overweight. Thus, it is not clear based on the available data whether the effects of infant feeding are causal or whether breast-feeding serves as a marker for other health behaviors that may affect child and adolescent growth. Recent studies among siblings, which allow control for maternal characteristics, show no protective effects of breast-feeding on obesity in adolescents and young adults.

Exposure to antigens and development of autoimmune disease The infant's diet is the main source of exposure to antigens suspected to be related to the development of autoimmune diseases. A likely protective effect of exclusive breast-feeding relates to lack of exposure to food allergens, though some other protective mechanisms related to specific substances in breast milk have been postulated. Exposure to bovine proteins by milk feeding, and to allergenic plant proteins such as those found in wheat is suspected to increase risk of developing diseases such as type 1 diabetes and celiac disease in genetically susceptible individuals.

Type 1 diabetes is one of the most prevalent chronic diseases with childhood onset. It is characterized by autoimmunity to pancreatic islet cells and is associated with a specific human leucocyte antigen (HLA) genotype. Not all individuals with the genotype develop the disease, suggesting an important role for gene-environment interactions. Hypothesized early exposures include infant feeding and entero-virus infections. Early introduction of cows' milk has received a great deal of attention as a potential risk factor. Numerous case-control studies associate increased risk with cows' milk, but a nearly equal number of studies show no effects. These retrospective studies have been criticized as suffering from recall bias and inappropriate control groups, for example, controls without the susceptible genotype. Recent prospective studies of at-risk infants in Australia and Germany found no association of type 1 diabetes with feeding of cows' milk. However, pilot study data from an international primary prevention trial suggests that eliminating cows' milk proteins in at-risk infants reduces risk of developing islet cell autoantibodies. This study also supports a role for early enterviral infections in the etiology of type 1 diabetes in genetically susceptible individuals. In fact, the research team has suggested that the effect of cows' milk may depend on viral exposures.

Recent studies suggest a role for other food antigens. A study of at-risk German children found that feeding of gluten-containing foods before 3 months of age was associated with risk of having pancreatic islet cell autoantibodies. Another study in the US also found an increased risk of islet cell autoimmu-nity among at-risk children given cereal before 3 months or after 7 months of age. Furthermore, they found that risks associated with cereal introduction were reduced by breast-feeding.

Other aspects of diet may have immunomodulatory effects. Vitamin D and the n-3 fatty acids EPA and DHA are suggested to be protective against immune-modulated diseases. For example, in a case-control study, Norwegian children given cod liver oil, a rich source of EPA and DHA, in the first year of life had significantly reduced risk of type 1 diabetes.

Type 2 diabetes Few studies have assessed the relationship of infant feeding to later development of Type 2 diabetes. Early feeding may affect patterns of insulin secretion in the newborn period, and thereby program subsequent development of metabolic control. Two studies in native American populations, one in Canada and one among Pima Indians, report a protective effect of breast-feeding on later development of Type 2 diabetes. In the Pima study, exclusive breast-feeding in the first 2 months of life was associated with a lower rate of Type 2 diabetes in children and adults. In the Canadian study, breastfeeding for more than 12 months was associated with decreased risk of Type 2 diabetes. Other studies have examined early risk factors related to subsequent development of Type 2 diabetes. For example, in a study of preterm infants randomized to human milk or formula of different composition, 32-33 split proinsulin, a marker of insulin resistance, was elevated in adolescents who had received a nutrient-enriched diet compared to those with a lower nutrient diet.

In sum, infant feeding, through nutritional adequacy, direct exposure to antigens, and protective substances provided in human milk, has the potential to alter response to subsequent exposures and to directly influence the beginning of disease processes.

Dieting Dilemma and Skinny Solutions

Dieting Dilemma and Skinny Solutions

The captivating thing about diets is that you don't get what is researched or predicted or calculated but rather, you get precisely what you expect. If the diet resonates with you then it will likely work, if it doesn't resonate, it won't.

Get My Free Ebook

Post a comment