Search

Relationships between intakes of amino acids and total nitrogen, and blood indices of protein utilization were studied in 37 term infants fed either human milk, whey-predominant formula, or cow's milk formula as the sole nutritional source for 8 weeks. Biochemical analyses of two-hour fasting blood samples, and intakes calculated using three-day dietary records and direct analyses of milk samples were used to evaluate these relationships. Intakes of total nitrogen were positively correlated with plasma valine, leucine, isoleucine, phenylalanine, and serum urea nitrogen concentrations (r = .46 to .62, P < .01 to .001). Intakes of the four amino acids whose plasma concentrations were positively correlated with total nitrogen intakes plus four additional amino acids (threonine, tyrosine, histidine, and methionine) were correlated with their respective plasma concentrations (r = .41 to .74, P < .01 to .001). These relationships have not been previously described in term infants. Compared with values in infants fed human milk, plasma concentrations of valine, phenylalanine, methionine, and serum urea nitrogen were elevated with whey-predominant formula and cow's milk formula feeding. Values for four additional amino acids (threonine, lysine, leucine, and isoleucine) were elevated with whey-predominant formula feeding. Data indicate that altering the whey-to-casein ratio and, thus, the amino acid pattern of formulas will not achieve the desired blood indices characteristic of human milk feeding without a reduction in the total nitrogen content of formulas.

The purpose of this investigation was to provide quantitative data on amino acids ingested by human milk-fed infants. Ten term infants fed human milk ad libitum were studied at 2, 4, and 8 weeks of age. Measurements of milk volume and amino acid analyses of milk samples were performed. Both total amino acid and nitrogen intakes (mumole/kg/day) declined by approximately 34% during the study (P less than 0.001) to 0.05), primarily because of a 20% decline in respective concentrations in human milk (P less than 0.001). Consequently, markedly different amino acid intakes were observed at 2 and 8 weeks. Quantities of amino acids consumed by nearly every infant exceeded estimated Food and Agriculture Organization/World Health Organization requirements at 2 weeks. However, mean intakes for seven of nine essential amino acids studied (phenylalanine, tyrosine, methionine, threonine, valine, leucine, and lysine) were less than recommended quantities at 8 weeks. These same relationships between observed and recommended intakes were also evident at 4 weeks, except for lysine and leucine. Mean intakes of histidine and isoleucine exceeded estimated requirements at all times. These data indicate that FAO/WHO estimates of amino acid requirements may be inflated for phenylalanine, tyrosine, methionine, threonine, valine, leucine, and lysine.

To further identify and characterize the nitrogen fraction of human milk, nucleotide and total nitrogen contents were determined using high pressure liquid chromatography and Kjeldahl analyses. Five lactating women were followed longitudinally. Each provided 16 milk samples (8-10 ml each) collected before and after a single nursing, and in the morning and afternoon of a single day. This collection scheme was followed at 2, 4, 8, and 12 wk postpartum. The variance pattern of nucleotides was observed to be distinct from that of total nitrogen. As the lactation period progressed from wk 2 to 12 postpartum, levels of cytidine 5' monophosphate and adenosine 5' monophosphate declined from 594 to 321 micrograms/100 ml and from 244 to 143 micrograms/100 ml, respectively, whereas levels of inosine 5' monophosphate increased from 158 to 290 micrograms/100 ml and levels of total nucleotide nitrogen remained constant. Nucleotide nitrogen accounted for approximately 0.1-0.15% of the total nitrogen content of human milk samples analyzed. Total concentration of human milk was observed to decrease as lactation progressed and to be higher in afternoon than in morning samples. The nucleotide profile of human milk was characteristically different from that of other milks commonly used an infant feeding. It is estimated that an infant consuming human milk as a principal nutrition source would ingest 1.4-2.1 mg of nucleotide nitrogen per day.

Hyperaminoacidemia is evident in infants fed either whey-dominant or casein-dominant formula containing 2.2 g protein/100 kcal. We assessed protein metabolism in infants fed formulas with reduced protein contents and various whey/casein ratios. Term infants (n = 40) received either human milk or formula containing 1.8 g protein/100 kcal and whey/casein ratios 18:82, 34:66, or 50:50. At ages 4 and 8 weeks, growth indices and mean serum concentrations of retinol binding protein, albumin, total protein, and serum urea nitrogen were similar, as were mean plasma concentrations of total amino acids, total essential amino acids, and their ratio. Compared with infants fed human milk, those fed formula had plasma concentrations similar for valine, lysine, arginine, tyrosine, histidine, threonine, and free and total cyst(e)ine; elevated for phenylalanine, methionine, and citrulline; and depressed for taurine and tryptophan. Except for leucine, mean plasma amino acid values varied similarly among formula groups despite differences in intakes. Our data indicate that feeding formulas providing 1.8 g protein/100 kcal results in many indices of protein metabolism characteristic of human milk feeding. However, certain differences are noted, suggesting the need for further manipulation of specific amino acid patterns of infant formulas.

In Reply.— We welcome the interest in our study1 and are pleased to have the opportunity to respond to Dr Wilson's letter. At the initiation of our study, there were two equivocal reports on indices of protein metabolism in full-term infants fed human milk, whey-predominant formula, or cow's milk formula.2-4 Therefore, our objective was to assess whether specific parameters of protein metabolism responded quantitatively to dietary intakes of protein, nitrogen, and amino acids. Our results did confirm that plasma aminograms are elevated with feeding either a whey-predominant or a cow's milk formula3; however, we are the first to report that total protein (total nitrogen) intake is as important as protein quality in determining plasma amino acid response in infants.

INDICES OF PROTEIN METABOLISM IN TERM INFANTS FED HUMAN MILK, WHEY PREDOMINANT FORMULA, OR COW MILK FORMULA