SPECIAL ARTICLE |
Despite the general acknowledgment that breastfeeding is the most desirable means of feeding normal infants, it is evident that many infants are currently fed infant formulas and that evaluation of the adequacy of such formulas is an important topic in the field of pediatrics. During early infancy, growth is the most sensitive index of health, and growth is therefore a critical component in evaluation of the adequacy of a new infant formula. To use growth as a component of formula evaluation, it is necessary to select a reference population, and there is an ongoing debate about which of several reference populations is most suitable.
On first consideration, the breastfed infant may seem a logical choice as a reference. It can be pointed out that the lactation process has been modified by evolutionary forces over millions of years, and human milk is generally agreed to be the model for development of infant formulas. Nowhere is the superiority of human milk over infant formula more clearly evident than in protection against infection. The major antiinfective components of human milk have not been incorporated into currently available infant formulas, and there is little likelihood that this will occur in the future. Nevertheless, based on an argument that will be presented, the overwhelming evidence that breastfeeding is the preferred way to nourish infants does not logically lead to the conclusion that data on growth of breastfed infants are suitable for assessment of growth of formula-fed infants.
| GROWTH OF BREASTFED AND FORMULA-FED INFANTS |
|---|
|
|
|---|
It is generally accepted that the growth of breastfed infants does not conform with the most commonly used reference data,3,4 and it has been proposed that a new set of reference data be assembled based on growth of breastfed infants living in advantageous circumstances. Because suitable reference data on growth of breastfed infants beyond the first few months of life are not available, this effort can only be applauded. Nevertheless, reference data on growth of breastfed infants, however valuable and necessary for assessment of growth of breastfed infants, may not be suitable for assessment of growth of formula-fed infants.
On a gender-specific basis, from 8 to 42 days of age, gains in weight and length of breastfed and formula-fed infants are similar, whereas from 42 to 112 days of age, formula-fed infants grow more rapidly than do breastfed infants (Table 1). Greater (but not significantly greater) gains by formula-fed than by breastfed infants were also reported by Dewey et al5 from 1 to 3 and 2 to 4 months of age (lack of significance is possibly the consequence of small numbers of subjects and large standard deviations).
|
It may seem heretical to suggest that at least some breastfed infants fail to meet their potential for growth of fat-free tissue. However, as stimuli to evolutionary change, other factors are considerably more important than rate of growth. An understanding of the effect of evolutionary factors on the composition of human milk requires acceptance of 2 major considerations: 1) the forces driving evolution are focused on the mother-offspring unit rather than on the infant; and 2) changes in living conditions can occur in 100 years,whereas evolutionary adaptation requires thousands of years.
| MOTHER-OFFSPRING UNIT |
|---|
|
|
|---|
As stated by Dugdale11: "Simple arithmetic shows that where there is a conflict of interests, the welfare of the mother outweighs that of the infant. The dyad hypothesis suggests that maximum evolutionary gain is obtained when protein and energy levels in breast-milk are just high enough to prevent prohibitive infant mortality rates, but low enough to spare the mother. The anti-infective constituents of breast-milk are very small in bulk, so they place a minimum metabolic load on the mother but have a large benefit for the child."
Animals with short gestation periods are likely to resolve a conflict between the welfare of the mother and that of the offspring in favor of the mother. Thus, when food supply is short, a rodent dam may kill all or part of the litter, presumably choosing to sacrifice the present offspring in favor of maintaining her health to be able to invest in the next litter. In animals with long gestation periods, the investment in the current offspring is quite high, and the mother may continue to invest her nutritional resources to support her infant despite some personal short-term disadvantage. In this context, the milk produced by undernourished Gambian women was reported to be of the same quantity and quality as that produced by well-nourished women in the United Kingdom.14 Nevertheless, although a short-term disadvantage to the mother may be acceptable in promoting the welfare of the species, it is clear that a long-term disadvantage to the mother is unacceptable.
| PROTEIN |
|---|
|
|
|---|
However, several writers20,21 have examined evidence about hunter-gatherers and primitive agricultural societies and point out that, in such societies, protein intakes are likely to be marginal and energy intakes may be erratic because of variable climatic conditions and variable success in food preservation and storage. Moreover, congregations of persons living in fixed locations without adequate sanitation facilities contribute to the spread of infectious diseases. Perhaps for these reasons, early European Homo sapiens, who enjoyed an abundance of animal protein (and, presumably, experienced fewer infectious diseases), were an average of 6 inches taller than their descendents who lived after the development of farming.21 Although we have no data on the protein concentration of human milk during the long interval of time when diets were high in protein, it seems reasonable to speculate that the protein content of human milk was greater than at present.
One might ask whether subsistence on diets much lower in protein over a period of 15 000 to 30 000 years could be sufficient to result in a decrease in the protein concentration of human milk, a change that would benefit the mother. Based on the discussion by Smith22 concerning spread of a favorable gene, it may be calculated that in 250 generations (ie, in
50006000 years), a gene present in 1% of a population might spread to 67% of the population. The best example of rapid spread of a favorable gene is the development of lactose tolerance of the adult, which presumably developed as a genetic adaptation to milk drinking.23,24 In times of food shortage, the ability to tolerate milk consumption probably had major survival value. A similar genetic change that permitted a lactating woman to limit the protein content of her milk would also be expected to enhance her survival. From the evolutionary point of view, it is irrelevant that, in a number of industrialized countries, abundant supplies of protein have been present in the diet for
100 years. It is obvious that our genetic pool, modified during at least 15 000 to 30 000 years of subsistence on marginal protein intakes, could not yet have responded to such a recent change.
| OTHER NUTRIENTS OF INTEREST |
|---|
|
|
|---|
Vitamin K
It is well known that the microorganisms of the gastrointestinal tract produce menaquinones, which, if ingested, exert vitamin K activity. A synthetic menaquinone (menaquinone-4) is used in Japan rather than phylloquinone (vitamin K1) for administration to the newborn to prevent vitamin K-deficiency bleeding.25 However, the extent of absorption of menaquinones from the site of their formation in the intestinal tract (ie, if not ingested) seems to be low. Thus, menaquinones cannot be counted on for meeting the needs for vitamin K for individuals living under current conditions of sanitation in industrialized countries. With what we would now consider poor environmental sanitation and poor personal hygiene, one can imagine that through all but the last hundred years or so of our evolutionary history, menaquinones produced by fecal microorganisms were widely available in the infants environment, and in fact the infant was likely to receive, inadvertently, a substantial oral dose of menaquinones during the birth process and was likely to receive supplements of menaquinones from contamination of the immediate environment, including the surface of his mothers breasts.
Vitamin D
Over all but the last 100 or 200 years of human history, the great majority of human infants met their needs for vitamin D by converting (with the aid of sunlight) 7-dehydrocholesterol in the skin to previtamin D3 and then to vitamin D3. Currently, vitamin D deficiency in breastfed infants, especially dark-skinned breastfed infants, is not uncommon26 and reflects the inadequate exposure of some infants to sunlight because of seasonally low availability of sunlight, atmospheric pollution, or purposeful avoidance of exposure.
Iron
Iron ranks among the most abundant elements in our environment. Although it occurs primarily in highly insoluble forms, in some soils it occurs in forms soluble enough to be toxic to plants.27 Hallberg and Björn-Rasmussen28 found that
30% of contamination iron in some diets was exchangeable with the remainder of iron in the diet (ie, was bioavailable). As stated by Hallberg29: "Considering the high iron content of many soils... the intake of soil iron must be considered in human nutrition, especially in the diets in developing countries where the intake of soil iron may be quite considerable." The concentrations of iron in milks of many precocial mammals (those whose young are relatively mature at birth) are low. The young of these animals generally have ready access to iron in the environment. It is well known that piglets suckled in clean environments without access to soil develop iron deficiency. Such deficiency can be prevented by providing supplements of iron or putting dirt in the pen.30
On the other hand, concentrations of iron in milk of altricial mammals (those whose young are quite immature at birth) are greater. A young marsupial remains attached to its mothers nipple for weeks or months and would be unlikely to ingest iron from sources other than milk. Iron concentration of milk of the tammar wallaby ranges from
10 to >20 mg/L during the period before the young marsupial emerges from the pouch.31 When the young marsupial emerges from the pouch (and presumably has access to iron in the environment), the iron concentration of the milk rapidly decreases to
5 mg/L. Concurrently, the energy concentration of the milk increases
2.7-fold, indicating that the iron content per unit of metabolizable energy has decreased remarkably.
During most of our human history, the infant, even during the early months of life, was exposed to an environment in which dirt, and therefore iron, was ingested from the environment: from the skin and clothing of the mother and other caretakers, from the bedding on which the infant slept, and, by 3 to 5 months of age, directly from the floors of living quarters. Thus, it is reasonable to speculate that infants were able to achieve adequate iron intake despite the quite limited intake of iron from human milk. However, in industrialized countries during the last 100 years or so, improvements in sanitation have been successful in excluding contamination iron from the diet.
The observation that iron deficiency is common in infants in developing countries should not be assumed to indicate lack of bioavailability of iron inadvertently consumed from environmental sources. Studies of iron nutritional status in developing countries have been conducted primarily in areas in which the population density exaggerates the problems of sanitation with high rates of infection and parasitic infestation. Even areas now classified as rural have greater populations than was the case during most of human history. Living conditions were quite different when the human population was sparse.
| FINAL COMMENTS AND SPECULATION |
|---|
|
|
|---|
Although intakes of protein were luxurious during most of human history, the increase in the human population and concurrent decrease in the biomass greatly decreased protein intake beginning 30 000 to 15 000 years ago. I speculate that evolutionary change since 30 000 years ago has been responsible for a nutritional compromise that favored survival of the mother through limiting the drain on her body reserves during lactation. The lactating woman was therefore able to conserve body protein by limiting its concentration in her milk.
According to my speculation, some breastfed infants fail to meet their potential for growth of fat-free tissue because of marginal intakes of energy, protein or other nutrients. The more-rapid growth of formula-fed than of breastfed infants may merely reflect the more-generous intakes of energy and essential nutrients by formula-fed infants. In the absence of evidence that the lesser gain in fat-free tissue by breastfed infants is advantageous, it seems unsound to select the gains of breastfed infants as our reference for growth of formula-fed infants. If one were to accept the conclusion that the growth of breastfed infants is the most suitable reference for growth of formula-fed infants, it would seem logical to require the infant-formula industry to devise formulas that would result in lesser gains in fat-free tissue than are now observed with formula-fed infants. In my view, this course of action is unacceptable. Adequate reference data on growth of formula-fed infants are available,3 and I believe they should be used for evaluation of growth of infants fed a new formula.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Address correspondence to Samuel J. Fomon, MD, Department of Pediatrics, University of Iowa, University of Iowa Hospitals and Clinics, 200 Hawkins Dr, Iowa City, Iowa 52242-1083. E-mail: samfomon{at}aol.com
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. H. Lende Review of Fomon's "Assessment of Growth of Formula-Fed Infants: Evolutionary Considerations" Pediatrics, November 1, 2004; 114(5): 1365 - 1365. [Full Text] [PDF] |
||||
![]() |
J. Brines Adult Lactose Tolerance Is Not an Advantageous Evolutionary Trait Pediatrics, November 1, 2004; 114(5): 1372 - 1372. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||