Search

Randomized Trial of Feeding Nutrient-Enriched vs Standard Formula to Premature Infants during the First Year of Life

Comparison of Two Dexamethasone Treatment Regimens on Outcomes of Growth and Bone Metabolism in Piglets † 1523

Nutrition, Growth, Bone Metabolism and the Ontogeny of the GH-IGF-I Axis to Term Corrected Age in Very Low Birth Weight Infants Treated with Dexamethasone in Early Life • 1585

The Influence of Breast Compared to Formula Feeding of Preterm Infants to Three Months Corrected Age on Growth and Body Composition to One Year † 1557

REDUCED PROTEIN SYNTHESIS AND GROWTH IN DEXAMETHASONE-TREATED PIGLETS IS ATTENUATED WITH ADJUVANT GROWTH HORMONE±INSULIN-LIKE GROWTH FACTOR-I THERAPY • 1442

PREMATURE INFANTS (PI) FED BREAST MILK OR FORMULA HAVE SIMILAR GROWTH BUT DIFFERENT BODY COMPOSITION IN THE FIRST YEAR. † 1917

Premature infants (PI) with bronchopulmonary dysplasia (BPD) commonly experience growth failure beyond early infancy. Elevated basal energy expenditure (EE) combined with inadequate energy intake (EI) may be responsible. Objective: To describe patterns of energy intake and expenditure during recovery from BPD. Methods and Subjects: A longitudinal descriptive trial in which BPD infants (n=38, birth weight=867± 194 g, gestational age=26 ± 1.4 wk) were fed a standard premature formula (810 kcal/L) prior to 36.5 ± 1.6 wk post-mentural age(PMA), and then were fed an energy enriched formula (900 kcal/L) until 3 mo corrected age (CA). Measurement of EE by oral dosing of doubly labelled water(0.3 g/kg 18O and 0.1 g/kg deuterium with urine analysis by mass spectrometry at 1 and 7 days) was done at 34.5 ± 1.3 wk PMA(in-hospital), at 1 (0.75 ± 0.4) and 3 (3.3 ± 0.4) mo CA at home. Energy intake at home was determined by weighed intake.Results: The Table shows EE as a% of EI(EE/EI) and total EE of BPD infants compared to reference infants (REF): 34.5 wk-healthy PI (Jensen et al, 1992); 1 and 3 mo-infants born at term (Davies et al, 1990). EI at 34.5 wk, 1 and 3 mo (119±8, 118±20 and 114±24 kcal/kg/d, respectively) met or exceeded recommended EI for infants. Although EE of BPD infants was more variable than term REF infants, they appeared to have greater basal EE, depleting the proportion of energy available for storage. Weight gain patterns supported this trend, since no catch-up growth was observed from 1 to 3 mo CA. Conclusions: Prolonged excessive EE during recovery from BPD reflects the need for aggressive nutritional therapy beyond the neonatal period, and possibly beyond 3 mo CA. (Funded by MOH and Wyeth-Ayerst, USA)

The effects of resistance exercise upon leucine oxidation and whole body protein synthesis were studied using stable isotope methodology. L-[1-13C]leucine was used as a tracer to calculate leucine oxidation and whole body protein synthesis in six healthy, fed, male athletes in response to a 1 h bout of circuit-set resistance exercise. The measurements were performed prior to, during, and for 2 h after exercise, and corrections were made for background 13CO2/12CO2 breath enrichment and bicarbonate retention factor changes. Results demonstrated significant (P less than 0.01) increases in the background 13CO2/12CO2 breath enrichment at 1 and 2h after exercise and in the bicarbonate retention factor (P less than 0.01) during exercise. At 15 min after exercise, the bicarbonate retention factor was significantly (P less than 0.05) lower than at rest. There were no significant effects of exercise on leucine oxidation or flux, whole body protein synthesis, or the rate of appearance of endogenous leucine. We concluded that circuit-set resistance exercise did not affect the measured variables of leucine metabolism. In addition, large errors in calculating leucine oxidation and whole body protein synthesis during resistance exercise can occur if background 13CO2/12CO2 breath enrichment and bicarbonate retention factor changes are not accounted for.

The fecal monitoring technique for measuring the absorption of Zn, Mn, Se, and Fe was studied in eight male piglets (mean +/- SEM birthweight (bw) = 1695 +/- 50 g) using high resolution gamma spectrometry. Four d old piglets were fed a complete liquid milk diet for five d prior to the orogastric administration of an isotope dose (75Se, 54Mn, 59Fe, and 65Zn) equilibrated with the liquid milk diet. 51CrCl3 was used as a fecal marker but was found to be partially absorbed. Stool samples were collected daily for 15 d, counted, and then the daily fecal excretion was calculated. Results indicate that endogenous excretion for each of the isotopes was not constant but decreased exponentially with time. The pattern of endogenous excretion varied between elements. An improved method for calculating the endogenous excretion was therefore developed. This method is based on the pattern of endogenous excretion in three-four d old male piglets (mean +/- SEM bw = 2060 +/- 75 g) injected intravenously with the same isotopes and on the level of endogenous excretion in orally fed animals in the postabsorptive phase of excretion. These findings have important implications for the estimation of endogenous excretion in future fecal monitoring absorption studies in order to minimize underestimation of true absorption.

Milk from mothers giving birth prematurely was analyzed for Na, Cl, K, Mg, Ca and P concentrations. The data presented are from analyses of milk samples representative of complete 24-hour expressions and collected serially over the first 29 days of lactation from mothers giving birth at term (FT) and mothers giving birth prematurely (PT). Mineral composition of FT and PT milks was similar during the first month lactation. With the exceptions of Mg and P, the concentrations of the minerals studied were higher initially than at the end of the fourth week of lactation. From these data, intakes of premature infants fed their own mothers' milk can be estimated and compared to predicted mineral requirements for the premature infant. On the basis of this comparison, we suggest that the quantities provided of Na, Cl, K and Mg, but not Ca and P, would be adequate to meet requirements of premature infants during the early weeks of life.

The lactose, lipid, protein, and total energy content of milk produced during the first 4 wk of lactation was determined in milk from mothers giving birth prematurely and at term. Milk samples analyzed were representative of complete 24-h expressions. Over the first 4 wk of lactation lactose concentration increased, but nitrogen concentration decreased by an average of 25%. Lipid and total energy concentration increased by 25% from the 1st to 2nd wk and then remained stable in both groups. Although these changes with duration of lactation were similar in both groups, premature compared to full term milk was consistently 20 to 30% higher in total energy and lipid concentration, 15 to 20% higher in total protein and, after the 1st wk, approximately 10% lower in lactose. By comparing predicted nutrient intakes to estimated requirements of premature infants fed 150 to 200 ml/kg/day of their own mothers' milk, we predict that the quantities of protein and energy provided would be adequate to meet the needs of the infant during the early weeks of life.

The association of various diuretic therapies with the renal handling of minerals, important factors in the development of nephrocalcinosis and osteopenia, was studied in low birth weight infants. Twenty-four-hour urine specimens (n = 65) were collected from 30 patients who were treated with (1) furosemide with or without spironolactone and hydrochlorothiazide (2) spironolactone with hydrochlorothiazide, (3) spironolactone alone, or (4) no diuretic (control; i.e., after diuretic). Hypercalciuria (urinary calcium greater than or equal to 0.15 mmol/kg/day) was observed in all but the control group. Covariate analysis demonstrated a significant effect of sodium, calcium, and vitamin D intakes (p less than 0.01) and sodium excretion (p less than 0.05) on urinary calcium excretion. Treatment with any of these diuretics in neonates may be associated with abnormal renal losses of calcium, sodium, chloride, and potassium. From a nutritional perspective, neonates requiring long-term diuretic therapy thereby require special consideration, including monitoring of mineral excretion and renal ultrasonography.

Although the biochemistry of milk from mothers giving birth prematurely has been described in some detail, it was not known whether premature birth would affect vitamin D status of mothers, the vitamin D activity of milk or the concentration of vitamin D binding protein in milk. In the present study, plasma and milk samples were collected from eight mothers giving birth prematurely and compared to samples obtained from nine mothers giving birth at term. Vitamin D intake from milk and vitamin D supplements was at least 400 IU/day in all mothers. Complete 24-hour expressions of milk were collected between 14 and 21 days postpartum. No differences were observed in plasma D 2 (5.1+0.7 and 3.7+1.0 ng/ml, mean±SEM) or milk D 2 (0.19±0.06 and 0.12±0.02 ng/ml) in term versus preterm groups, respectively. In contrast, plasma vitamin D 3 (0.7±0.1 and 2.7±0.5, ng/ml, p 3 (0.14±0.02 and 0.23±0.03, ng/ml, p 2 and 25-hydroxyvitamin D 3 , no differences between mother groups could be discerned in plasma or milk. Differences in the synthesis and/or transport of vitamin D and DBP which are reflected in PT milk and plasma levels in the early postpartum period may relate to cutaneous synthesis or abbreviated gestational stage. Based on the vitamin D sterols measured in this study, the total vitamin D activity in human milk in the first month of lactation averaged 80±9 IU/L for preterm milk and 60±7 IU/L for term milk, both values which are in excess of the often quoted value for human milk vitamin D of 22 IU/L.

Clinical observations of bone pain, abnormal gait, and unusual fractures during remission of leukemia led us to assess mineral status in a cohort of 16 children with acute lymphoblastic leukemia treated with intensive chemotherapy. During maintenance and 6 months after the completion of therapy, blood and urine were analyzed for calcium and magnesium and blood for osteocalcin, vitamin D, and parathyroid hormone. Bone mineral content and bone width of the distal one third of the radius of the nondominant arm was measured by single-photon absorptiometry. During therapy, mild ionic hypocalcemia (less than 1.19 mmol/L) and hypomagnesemia (less than 0.77 mmol/L) were demonstrated in 9 and 8 of 16 children, respectively; hypercalciuria (8/16) and hypomagnesiuria (12/16) were also observed. Plasma osteocalcin values correlated with plasma magnesium levels (r = 0.54; p less than 0.05). Oral magnesium supplements normalized plasma magnesium, calcium, and osteocalcin levels, all of which were normal at the postchemotherapy study. Plasma 1,25-dihydroxyvitamin D levels were nondetectable (less than 8 ng/ml) in 12 of 13 patients receiving therapy and in 7 of 14 patients not receiving therapy; alkaline phosphatase activity increased significantly after therapy (179 +/- 86 to 340 +/- 101 units/L), and parathyroid hormone levels were normal in both studies. Bone mineral content/bone width ratio was less than 1 SD below the mean for age- and sex-related population standards in 70% of patients. These data indicate that alterations in magnesium, calcium, and vitamin D metabolism in children treated for acute lymphoblastic leukemia may be instrumental in inducing or sustaining altered bone turnover during chemotherapy.

The nutritional adequacy of the premature infant's own mother's milk was assessed during the first two weeks of life. Studies were carried out in three groups (n = 8) of infants of less than 1,300 gm birth weight, matched for gestational age and weight, and fed either pooled breast milk, their mothers own milk, or infant formula (SMA20 or SMA24). Macronutrient balances at the end of the first and second postnatal weeks demonstrated differences in nitrogen and lipid absorption and retention between groups. Nitrogen retentions (mg/kg/day) were similar to normal fetal accretion rates only in the PT and SMA24 groups. Fat absorption was poorest from the heat-sterilized PBM (average of 64.0% of intake) when compared to PT (88.2%) and SMA groups (83.3%). Average gross and metabolizable energy intakes were similar among groups. Nutritional status as measured by plasma total protein and albumin concentrations and weight gain tended to be poorest in the PBM-fed infants. It was concluded that either PT milk or infant formula of a composition similar to SMA24 are more appropriate than pooled banked milk for feeding the premature infant during the first two weeks of life.

Human milk is thought to be limited in calcium, phosphorous, sodium and possibly other minerals relative to estimated requirements for growing low birthweight infants. This study examined the efficacy of early supplementation of premature infants' mothers' milk (PTM) with calcium lactate (58 to 70 mg calcium/kg/24 hr) with respect to absorption of calcium, phosphorous, magnesium and dietary lipid and renal handling of major cations. Seventy-two hour metabolic balances were carried out in the first, second and fourth postnatal weeks in 14 infants born at 29.8±0.3 (mean±SEM) wk and 1286±35 g birthweight. Feeding of PTM with or without calcium supplementation (as calcium lactate) was associated with hypercalciuria in 83% of the infants studied. Supplementation with calcium resulted in greater calcium absorption but also greater urinary losses. Fecal lipid losses also increased with calcium supplementation. Absorption and excretion of phosphorous and magnesium did not appear to be altered by administration of calcium lactate. Supplementation of calcium as calcium lactate cannot be recommended for human milk fed infants. The hypercalciuria which occurred may place the developing kidney at risk of nephrocalcinosis, although this may be ameliorated by providing concurrent phosphorous therapy and conservative dietary sodium intakes. The amount and form(s) of calcium salt which can be maximally absorbed and retained with minimal interference with absorption of dietary lipid and/or other macro and trace elements from human milk has yet to be delineated.

Total nitrogen concentration of human milk during the first 4 weeks of lactation is higher in milk obtained from mothers giving birth prematurely than in milk from mothers giving birth at term. In the present study, the relative composition of total nitrogen was found to be similar in premature and full-term milk except for some minor differences in small molecule components. Total nitrogen is comprised of an average of 82% protein, 8% of the nonprotein nitrogen components--urea, free amino acids, uric acid, and ammonia--and the remaining 10% is predicted to be peptide nitrogen.The purpose of this study is to determine and compare the composition of total nitrogen (N) in milk produced during the 1st month of lactation by mothers giving birth either prematurely or at term. Total N was divided into protein, nonprotein, and its components: urea, free amino acids, uric acid, ammonia, and creatinine. At specific day intervals from 2-29 days postpartum, complete 24 hour milk collections were obtained from 7 mothers giving birth at 38-40 weeks of gestation. At the completion of each 24-hour period the total daily voume of milk was well mixed and measured before an aliquot was removed and stored at -20 degrees celsius for analysis. Linear regression analyses of the pooled data for total N demonstrated a decrease in N concentration at a similar rate in both the full term (FT) and premature (PT) milks over the 1st 29 days of lactation. The PT milk was shown to be significantly higher (P0.001) in total N. In the PT milk, protein N accounted for 82% of the total, 8% was nonprotein N components, and 10% is predicted to be peptide N. The absolute concentration of protein N was higher in the PT milk over the 4 weeks of lactation observed (P0.005) and the concentration of N derived from urea, free amino acids, lysine, and theorine were present in highest concentrations but showed no differences with lactational stage or gestational age. Valine, leucine, isoleucine, and histidine were higher in the PT milk than FT milk at the 15-29 day period only (P0.05). The data provide evidence that the N composition of milk of mothers giving birth prematurely is similar to milk of mothers at full term. However the higher N content of PT mothers' milk will provide the premature infant with higher amounts of protein N, nonprotein N, urea, and fre amino acids. During the 1st month of lactation expressed PT milk is 10-20% higher in energy density than FT milk due to a 30% higher fat concentration in PT milk, also advantageous for the premature infant.

We measured serum osteocalcin concentrations in 82 pregnant and 21 nonpregnant women. Osteocalcin values declined in the second trimester, but returned to nonpregnant levels late in the third trimester. The mean serum osteocalcin concentration in 36 women during pregnancy (mean gestation, 26 weeks) of 2.8 ng/mL was significantly lower than that in nonpregnant women (6.4 ng/mL; P less than 0.001) or term pregnant women at delivery (6.1 ng/mL; n = 46). Serum immunoreactive PTH (iPTH) levels were significantly higher during pregnancy than in nonpregnant women [97 +/- 5 vs. 56 +/- 4 ng/L (mean +/- SE); P less than 0.001]. No significant correlations were found between maternal osteocalcin concentrations and serum phosphorus, alkaline phosphatase, or iPTH, but significant negative correlations were found between osteocalcin and total calcium or total protein. Osteocalcin concentrations in midtrimester amniotic fluid were very low (mean, 0.3 +/- 0.1 ng/mL; n = 11). In 29 lactating mothers, the mean serum osteocalcin level was 9.5 +/- 1.5 ng/mL, significantly higher than in any of the other groups (P less than 0.05), but their serum calcium and iPTH levels were normal. There was no correlation between serum osteocalcin and calcium or iPTH concentrations in lactating women. These changes are compatible with a sequence in which bone turnover is reduced during early pregnancy, rebounds in the third trimester, and increases in postpartum lactating women.

The present study examined the effects of training status (endurance exercise or body building) on nitrogen balance, body composition, and urea excretion during periods of habitual and altered protein intakes. Experiments were performed on six elite bodybuilders, six elite endurance athletes, and six sedentary controls during a 10-day period of normal protein intake followed by a 10-day period of altered protein intake. The nitrogen balance data revealed that bodybuilders required 1.12 times and endurance athletes required 1.67 times more daily protein than sedentary controls. Lean body mass (density) was maintained in bodybuilders consuming 1.05 g protein.kg-1.day-1. Endurance athletes excreted more total daily urea than either bodybuilders or controls. We conclude that bodybuilders during habitual training require a daily protein intake only slightly greater than that for sedentary individuals in the maintenance of lean body mass and that endurance athletes require daily protein intakes greater than either bodybuilders or sedentary individuals to meet the needs of protein catabolism during exercise.

• An abnormally low concentration of zinc in mother's milk has been associated with clinical nutritional zinc deficiency in premature and term newborns. Defective mammary gland secretion of zinc was suggested as the causative factor. We investigated whether low milk zinc concentration might be reflected in an abnormality of the distribution of zinc between casein, whey, and fat components of the milk, in a reduction in the levels of milk citrate, a zinc-binding ligand, or in abnormal levels of copper and iron. Milk from mothers of four low-birth-weight infants was identified as being deficient in zinc content for the lactational stage. One infant had clinical signs of nutritional zinc deficiency. For two of the three infants with subclinical deficiencies in whom balance study data were available, the apparent zinc balance was negative (−5.7 and −6.0 μmol/kg per day). These milks were examined for content and distribution of zinc, copper, and iron, and citrate concentration. Although total zinc concentration (4.5 to 17.2 μmol/L) was below the 95% prediction interval (15.2 to 59.7 μmol/L) for the lactational stage, the distribution of zinc was normal between milk components. Levels of citrate (1.46 to 5.13 mmol/L) were also within normal limits, as were levels of copper (0.27 to 0.85 μmol/L and iron (0.41 to 1.13 μmol/L). We concluded that there was no indirect evidence of a secretory abnormality in milk zinc deficiency, but speculated that there might instead be an abnormality of zinc uptake by the mammary gland from the plasma. ( AJDC . 1989;143:608-611)

A study was carried out measuring the nitrogen concentration of milk in mothers of PT (premature) infants and mother of FT (fullterm) infants. Linear regression analysis of the data demonstrated that nitrogen concentration decreased significantly during the time of lactation at a similar rate in both groups. The PT mothers milk had however a significantly higher nitrogen concentration than the FT mothers milk at the start of lactation. The volume of milk produced was similar for both groups of mothers increasing in an individual mother by approximately 200 ml from week 1 to week 4. This information on volume of milk produced proves that premature delivery does not affect the ability to lactate. The observed differences in nitrogen concentration were not a function of differences in daily milk volume production or collection day.

• Although nitrogen balance studies have been carried out in low-birth-weight infants, few have partitioned the nitrogen into its components. In this study, 72-hour balance studies were conducted in 24 low-birth-weight infants (gestational age, 30.7±1.6 weeks; birth weight 1.36±0.25 kg) fed their mothers' milk (preterm milk) or 50% preterm milk and 50% formula. Total nitrogen, nonprotein nitrogen, and whey protein intake and excretion were measured. Total nitrogen intake (preterm milk group, 452 ± 138 mg/kg per day; preterm+formula group, 406±93 mg/kg per day), absorption (85%), and retention (71%) were not significantly different between groups. Intact and fragments of secretory IgA and lactoferrin were detected in soluble fecal extracts, and represented 25% and 9% of intake, respectively. Feeding preterm milk allows for nitrogen accretion similar to intrauterine growth rates for 5 weeks postnatally, and provides potentially functional proteins for the low-birth-weight infant. (AJDC. 1989;143:1485-1491)

There has long been controversy over the optimal form of enteral feeding for the premature infant. It is generally accepted that human milk is the desired form of nutrition for the term infant, but it has been proposed that human milk, especially banked mature human milk, is not suitable for the premature infant1,5. Currently, attention has been focused on assessing the adequacy of feeding the preterm infant milk from his/her own mother2–8. Atkinson et al.5 have suggested that the use of preterm milk supports significantly better nitrogen retention and protein nutritional status than pooled mature milk, especially during the first two weeks of life, due to its higher protein content. Therefore, it is thought that the use of “preterm milk” would more closely fulfill the needs of the preterm infant without causing the metabolic abnormalities that have been observed in infants fed formula with higher protein content9. Previous studies have all measured only total nitrogen of milk and excreta when determining nitrogen balance in premature infants. However, when assessing nitrogen utilization and protein requirements in infants fed human milk one has to consider the possible contribution of the non-protein nitrogen (NPN) fraction of human milk to the utilizable N pool provided to the recipient infant.

The central theme of this workshop was the utilization of specific minerals and vitamins from human milk and an evaluation of the adequacy of these nutrients in milk to support growth and development of normal term and premature infants. The trace elements iron, zinc, copper, manganese and selenium; the minerals calcium and phosphorus; and a water soluble vitamin, folic acid were discussed.

The dynamic changes that occur in milk composition associated with duration of lactation have been recognized for over a century as documented in several early publications some dating back to 1874 (1–3). In the mid 1900s Macy (4) reviewed 1500 publications on milk composition from many countries and selected 268 papers based on “the characteristics of subjects, methods of collecting samples, analytical techniques, diet and environmental factors”. From the data surveyed, they compiled a table of composite values for more than 50 components in human milk at three lactational stages: colostrum, representing one to five days postpartum; transitional milk, representing days six to ten postpartum; and mature milk, representing day 15 to 15 months postpartum. Since that time and particularly since the mid 1970s, a plethora of information on milk composition has emerged providing further characterization of important biological components in human milk and knowledge of those factors which are associated with alterations in the composition and volume of human milk produced by the average healthy lactating woman (Table 1).

Acrodermatitis, diarrhea, hyperirritability and delayed growth in association with low plasma zinc levels have been documented in case reports of premature infants exclusively fed human milk (1–3). In all cases the etiology of the nutritional zinc deficiency was attributed to levels of zinc in the infants‘ mothers’ milk that were below the usual range of values reported for milk at similar lactational stages. However, there has only been speculation as to whether this is caused by defective mammary gland zinc secretion or whether it simply represents the lower end of a normal spectrum (3). In the present report, we describe the quantitation of citrate, a major zinc binding ligand in human milk (4), and the distribution of zinc, copper and iron in the casein, whey, fat and whole milk fractions of milk which we determined to be “zinc deficient.”

The objectives of this workshop were two fold: first, to consider the effects on growth and body composition of feeding human milk or proprietary formulas in both term and premature infants. Secondly, to examine the interrelationships between energy intakes and utilization of the major energy source in milk (i.e., fat) during early developmental stages in infants.

Human milk whey contains proteins thought to have functional and nutritional roles for the infant. These proteins include lysozyme(Lys), lactoferrin(Lf), secretory IgA(sIgA), serum albumin(SA) and α-lactalbumin(α-LA). SA and α-LA primarily have nutritional importance as sources of amino acids. Lys, Lf and slgA may have both nutritional and physiological functions. Lys and Lf have been suggested to have bacteriostatic effects within the infant’s gastrointestinal (GI) tract. SlgA acts immunologically within the gut of the infant. For these proteins to exert their functions, they must remain largely intact during passage through the upper GI tract. We have demonstrated (AJCN 41: 852(1985)) that LF and slgA are detected in feces of exclusively breast-fed term infants. We were interest in quantitating these proteins in feces of preterm infants for two reasons. First, preterm milk contains higher quantities of these whey proteins. Secondly, the proteins may serve a greater protective role for preterm infants, due to immature immune function in such infants.

The premature infant's own mother's milk (preterm milk) and modified infant formula (SMA, 67 and 80 kcal/dl) were fed to paired groups of seven infants, all of whom were of very low birth weight (VLBW) (less than 1.3 kg) and were studied during the first month of life. Sodium, potassium, magnesium, calcium, and phosphorus status was compared. The apparent retention of sodium from their mother's milk paralleled intrauterine retention rates and was greater than retention from SMA formula (P less than 0.01) during the first two weeks of life. However, when the formula was supplemented with NaHCO3 to intakes of 2.7 mmol Na/kg/24 hr after week 2, the infants retained adequate amounts of sodium. Potassium retention was similar to intrauterine retention rates in both groups throughout the four postnatal weeks. Magnesium intake, but not retention, was consistently higher in the group fed SMA (P less than 0.01), and intrauterine retention rates were achieved only in the group given formula. Calcium and phosphorus intakes from SMA were also higher (P less than 0.01) than from human milk. However, retention of calcium and phosphorus in both groups did not meet intrauterine retention rates, and hypophosphatemia developed in infants who received their mothers' milk. Growth in length and head circumference in both groups approximated intrauterine growth rates. If it is assumed that body composition of the growing VLBW infants should be similar to the composition of the fetus at corresponding gestational ages, then their nutrient requirements should be based on knowledge of intrauterine nutrient accretion rates. Based on this premise, we conclude that, for the growing VLBW infant, early maternal milk provided for sufficient retention of sodium, chloride, and potassium during the first four postnatal weeks. Neither human preterm milk nor SMA supplied adequate calcium and phosphorus for the growing VLBW infant.

Determination of the biochemical composition of milk from mothers giving birth prematurely (preterm milk) by Atkinson et al, (1–3) and subsequently by others (4–9) demonstrated that preterm milk had greater concentrations of nitrogen, lipid, medium chain fatty acids, energy and sane minerals when compared to term milk at similar lactational stages. Evidence for these compositional differences is not universal and such lack of agreement amongst investigators may be attributed to differences in sample collection methodology, the inclusion of wide ranges of gestational stage and the greater degree of inter-individual variability in milk composition in preterm compared to term milk. For a more complete discussion of these areas, see chapter by Atkinson and Kaufman in this book.

Bioactives can be defined as: "Constituents in foods or dietary supplements, other than those needed to meet basic human nutritional needs, which are responsible for changes in health status" (Office of Disease Prevention and Health Promotion, Office of Public Health and Science, Department of Health and Human Services in Fed Reg 69:55821-55822, 2004). Although traditional nutrients, such as vitamins, minerals, protein, essential fatty acids and essential amino acids, have dietary reference intake (DRI) values, there is no such evaluative process for bioactives. For certain classes of bioactives, substantial scientific evidence exists to validate a relationship between their intake and enhanced health conditions or reduced risk of disease. In addition, the study of bioactives and their relationship to disease risk is a growing area of research supported by government, academic institutions, and food and supplement manufacturers. Importantly, consumers are purchasing foods containing bioactives, yet there is no evaluative process in place to let the public know how strong the science is behind the benefits or the quantitative amounts needed to achieve these beneficial health effects. This conference, Bioactives: Qualitative Nutrient Reference Values for Life-stage Groups?, explored why it is important to have a DRI-like process for bioactives and challenges for establishing such a process. Fil: J. R. Lupton. Fil: S. A. Atkinson. Fil: N. Chang. Fil: Fraga, César Guillermo. Consejo Nacional de Investigaciones Científicas y Técnicas. Oficina de Coordinación Administrativa Houssay. Instituto de Bioquímica y Medicina Molecular. Universidad de Buenos Aires. Facultad Medicina. Instituto de Bioquímica y Medicina Molecular; Argentina. Universidad de Buenos Aires. Facultad de Farmacia y Bioquímica. Departamento de Química Analitica y Fisicoquímica. Cátedra de Fisicoquímica; Argentina Fil: J. Levy. Fil: M. Messina. Fil: D. P. Richardson. Fil: B. van Ommen. Fil: Y. Yang. Fil: J. C. Griffiths. Fil: J. Hathcock.

Background A web-based inventory was developed as a voluntary registry of Canadian pregnancy and birth cohort studies, with the objective to foster collaboration and sharing of research tools among cohort study groups as a means to enrich research in maternal and child health across Canada. Description Information on existing birth cohort studies conducted in Canada exclusively or as part of broader international initiatives was accessed by searching the literature in PubMed and PsychInfo databases. Additional studies were identified by enquiring about the research activities of researchers at Canadian universities or working in affiliated hospitals or research centres or institutes. Of the fifty-eight birth cohort studies initially identified, forty-six were incorporated into the inventory if they were of a retrospective and/or prospective longitudinal design and with a minimum of two phases of data collection, with the first period having occurred before, during, or shortly after pregnancy and had an initial study sample size of a minimum of 200 participants. Information collected from each study was organized into four main categories: basic information, data source and period of collection, exposures, and outcome measures and was coded and entered into an Excel spreadsheet. The information incorporated into the Excel spreadsheet was double checked, completed when necessary, and verified for completeness and accuracy by contacting the principal investigator or research coordinator. All data collected were then uploaded onto the website of the Institute of Human Development Child and Youth Health of the Canadian Institutes of Health Research. Subsequently, the database was updated and developed as an online searchable inventory on the website of the Maternal, Infant, Child and Youth Research Network. Conclusions This inventory is unique, as it represents detailed information assembled for the first time on a large number of Canadian birth cohort studies. Such information provides a valuable resource for investigators in the planning stages of cohort studies and identifying current research gaps.

Several studies have found that caffeine improved endurance exercise performance, but the factors which are responsible for this are not fully understood. Possibilities include an increased free fatty acid (FFA) oxidation and a resultant sparing of muscle glycogen as well as an enhancement of neuromuscular function during exercise. In order to further examine these factors, six varsity level runners (VO2max = 63.3 ml.kg-1.min-1) were studied over 90 min of treadmill running (70% VO2max) in a thermoneutral environment in order to determine the metabolic and neuromuscular effects of caffeine (6 mg.kg-1) administered in a randomized, crossover, double-blind manner. Subjects were habitual caffeine consumers (200 mg.d-1) and were given identical diets during each 3-d testing period. Caffeine administration, 60 min prior to exercise, significantly (P less than 0.05) increased plasma FFA levels both prior to and during exercise. Caffeine administration did not alter any of the other variables examined: VO2, HR, RER, rating of perceived exertion; plasma levels of glucose, lactate, epinephrine, and norepinephrine; or neuromuscular function (maximal voluntary strength, peak twitch torque, and motor unit activation). We conclude that caffeine administration (6 mg.kg-1) in athletic, habitual caffeine consumers increased plasma FFA levels but had neither metabolic nor neuromuscular effects that would be of potential ergogenic benefit in endurance running.

Complete 24-hour expressions of human breast milk were collected serially from 7 mothers giving birth at 26-33 wk gestation (Premature,PT) and 8 mothers at 38-40 wk gestation (Full-term,FT) during the first 29 days postpartum. Individual milk samples (69) were analyzed for total N concentration. The regression lines describing the change in N concentration (y) with time in days (d) were y=368 - 4.36d (r=-0.60,p

The increasing survival of VLBW premature infants requires a precise knowledge of utilization of essential nutrients (P=protein F=fat, C=carbohydrate) in order to design suitable dietary regimes Energy balance (EB), substrate utilization for oxidation (0) and tissue deposition studies were performed on VLBW premature infants (n=10; gest. age 27-31 weeks; birth weight 940 - 1280 g; postnat. age 1 - 4 weeks) fed by own mother's milk (6 studies on 4 infants) or humanized milk (SMA 20/24 Wyeth; 13 studies on 6 infants). EB for growth was determined by the equation: EB = ME - 0 (P+F+C). By increasing the net energy intake (NEI) from 50 to 150 Kcal/kg/day, resting metabolic rate (RMR) increased from 38 to 58 Kcal/kg/day in the formula fed infants (FFI) and from 45 to 58 in the breast fed infants (BFI). In a range of 50-110 Kcal/kg/day NEI the RMR was consistently higher in BFI. The latter deposited more P than FFI at the same level of NEI. At a NEI of 100 Kcal/kg/day FFI deposited 1 g P/day in contrast to the 2 g/kg/day P deposition observed in the BFI. Tissue deposition of F increases with the enhancement of NEI in both FFI and BFI but the relationship is the reverse to that observed for P deposition. At a 100 Kcal/kg/day NEI 18 Kcal/kg/day energy is deposited as F in the BFI in contrast to the 28 Kcal/kg/day deposition in FFI. It is concluded that the quality of growth (P v.F deposition) in response to a specific diet composition may be defined by our investigative approach.