Background: Decreased nitrogen levels, calcium intestinal absorption rates, and plasma amino acid imbalances were reported for preterm infants who were fed partially hydrolyzed preterm formulas. In this pilot study, we evaluated a new formula with modified nitrogen and calcium sources.
Methods: During their second week of life, 16 preterm infants were randomly assigned to one of two groups: 9 were fed the new partially hydrolyzed formula and 7 were fed a conventional formula. Nutrient balance was performed at the end of the first month of life. Amino acid concentrations and anthropometric parameters were measured at theoretical term.
Results: Birth weight and gestational age (mean ± SD) were similar in the two groups (28.9 ± 7.0 weeks and 1183 ± 242 g vs. 27.7 ± 1.0 weeks and 1139 ± 162 g). Median nitrogen absorption rates (85% vs. 89%;P = 0.03) and biological values (59% vs. 69%;P = 0.13) were lower for infants who were fed the new formula than for those fed the conventional formula. After correction for difference in nitrogen intake, there was no significant difference in nitrogen retained between the two groups (P = 0.11). Plasma amino acid concentrations were also similar in the two groups. Median calcium absorption tended to be higher in the new-formula group than in the conventional-formula group (54% vs. 45%, P = 0.19). At theoretical term, infants fed the conventional formula were heavier than infants fed the new formula (3559 ± 362 g vs. 3193 ± 384 g, P = 0.04).
Conclusions: Because nitrogen content is 10% higher in hydrolyzed-protein formula than in entire-protein formula, appropriate nitrogen retention, plasma amino acid profile, and mineral use can be achieved with the new partially hydrolyzed formula. Further studies with larger groups are needed to evaluate the effect on growth.
Neonatology and Human Nutrition Research Center, Edouard Herriot Hospital, Claude Bernard University, Lyon, France; and Neonatology Department, Hopital de la Citadelle, University of Liege, Belgium
Received March 17, 2000; accepted January 11, 2001.
Supported by the Nestlé Company (Vevey, Switzerland) and Nestlé France, Marne la Valee, France.
Address correspondence and reprint requests to Dr. Jean-Charles Picaud, Neonatology, Edouard Herriot Hospital, Claude Bernard University, Place d'Arsonval, 69437 Lyon Cedex 03, France (e-mail: firstname.lastname@example.org).
Protein-hydrolysate preterm formulas are widely used in European neonatal units to reduce transit time, avoid intestinal stasis, improve gastrointestinal tolerance (1–3), and reduce the risk of bovine-milk–protein sensitization (4–6). However, the nutritional value of protein-hydrolysate formulas has been questioned (7–9). When compared with infants fed human milk or entire-protein formulas, infants fed hydrolyzed formula presented reduced nitrogen and mineral intestinal absorption rates and plasma amino acid concentration imbalances (7–11). Such amino acid imbalances may represent a risk for brain development during early infancy (12).
A new partially hydrolyzed preterm formula (PHPF) with a modified nitrogen profile and higher mineral content was manufactured to improve the nutritional value of available formulas. The current pilot study with very low birth-weight infants tested for differences in nitrogen, amino acid, and mineral balances between this new formula and a standard entire-protein preterm formula (SPF). During the study period, clinical tolerance and growth were also monitored.
MATERIALS AND METHODS
The current study was carried out in a single tertiary neonatal center: Edouard Herriot Hospital, University of Lyon. Neonates were included after written informed consent was obtained from their parents. The criteria for participation included a birth weight of 1500 g or less, age of 15 days or less at start of enteral feeding, and formula feeding exclusively. Initial exclusion criteria included breast-feeding, diabetic mother, intra-uterine growth retardation (13), major congenital anomalies, intra-uterine infection, and major clinical problems (e.g., oxygen dependency and necrotizing enterocolitis).
Eighteen very low birth-weight infants, all appropriate for gestational age, were included in the study between May 1995 and March 1996. During that period, 41 infants met the criteria to participate. However, the mothers of 15 chose to breast-feed and the parents of 8 refused to participate in the study. Later, two of the subjects fed SPF were excluded before the nutrient balance study. The first, a girl, showed signs of renal venous thrombosis 14 days after inclusion. The second, a boy, showed signs of grade IIIB necrotizing enterocolitis 20 days after inclusion (14). Sixteen low birth-weight infants completed the study: nine were fed PHPF (six boys) and seven were fed SPF (four boys).
Both diets contained the same protein source, a skim milk mixture of acid and rennet whey, and therefore had comparable amino acid profiles. According to the manufacturer, 64% of the peptides in the PHPF had a molecular weight below 1500 D and new technology was used to reduce Maillard reaction. Calcium glycerophosphate was the main mineral source and represented 35% of the calcium content and 44% of the phosphorus content in both formulas.
The Nutrition Laboratory in Liege controlled macronutrient and amino acid composition of the formulas. The two formulas were similar in every respect except for unexpected differences in nitrogen and calcium contents (see Statistical analysis and results). Routine Vitamin D supplements were similar in the two feeding groups (800 IU/day).
The current study is a double-blind randomized trial. The infants were randomly assigned to one of two feeding groups by means of cards in sealed envelopes. Each infant received either PHPF or the native whole-protein SPF. The medical personnel involved were obviously unaware of which formula was fed to each infant.
Initially, all subjects were orally fed pasteurized, banked human milk within 4 days of birth. Subsequently, infants were either fed their own mother's fortified milk or weaned onto a formula. Milk intake was increased daily, depending on clinical tolerance, until adequate volumes were reached (160–170 mL/kg/day). The infants were fed the trial formulas until they reached theoretical term. During the whole study period, even after the infants were discharged from hospital, the formulas were still provided to the mothers each time they visited the outpatient department. The mothers had to record feeding data in special diaries.
Clinical Tolerance and Transit Time
Instances of vomiting or gastric residues were recorded for each infant on a daily basis between inclusion and discharge. Transit time was measured using carmine red at the beginning and end of the nutrient balance study. Mean values were computed.
Body weight, crown-heel length, and head circumference were measured by the same investigator (B.R.) at inclusion, discharge, and theoretical term. Weight gain (g/kg/day) was calculated during the nutrient-balance study as previously described (15,16). Weight gain during the study was calculated (g/day) over two periods: before and after discharge from the neonatal unit. Anthropometric parameters were compared to intrauterine reference curves (13). Standard deviation (SD) scores and mean ± SD were computed.
Plasma Amino Acid Concentrations
Samples of preprandial peripheral venous blood were obtained at the end of the study period (theoretical term) for determination of plasma amino acid concentrations. Amino acid concentrations were measured in the Nutrition Laboratory in Liege using liquid chromatography as previously described (12). The results were compared with the reference values taking into account the mean ± SD of the combined values found in cord blood and in preterm infants fed human milk supplemented with human-milk proteins (17).
The nutrient-balance study was performed at the end of the first month of life, once the subjects were growing steadily. The infants were placed on a metabolic bed in an incubator and fed by nasogastric tube every 3 hours. Precise measurements of intakes and excreta over a 3-day period were taken as previously described (15,16). Nitrogen, calcium, and phosphorus levels were determined from aliquots of milk, fecal homogenate, and urine using the micro-Kjeldahl method, atomic absorption spectroscopy after ashing at 550°C for 14 hours, and the Brigg method after digestion with H 2 SO 4 , respectively. Fat content was determined from milk and feces using the gravimetric method. Amino acid concentrations in milk were determined using a chromatography auto-analyzer (LKB; Cambridge, U.K.) after acid hydrolysis at 100°C for 24 hours with 6N HCl at the Nutrition Laboratory in Liege.
Energy expenditure was measured using the doubly labeled–water method (18–21). After saving a control urine sample, a dose of doubly labeled water (Eurisotope 91194; Saint-Aubin, France) was given on day 1 of the nutrient-balance study. Doubly labeled water was administered orally with an accurately weighed dose of isotope solution, providing 0.25 g oxygen-18 and 0.13 g deuterium per kg of body weight. Urine samples were subsequently collected 4 and 6 hours after the oral dose of doubly labeled water. The other samples for isotope analysis were collected on a daily basis in the early morning for 6 days to measure isotopic enrichment decay during the nutrient-balance period. Isotope analysis was performed using an isotope-ratio mass spectrometer (Optima; Fisons, Rochester, NY) at the Human Nutrition Research Center in Lyon. The mean reproducibility of isotopic enrichment decay measurement was 1.4% for oxygen-18 and 6.2% for deuterium (22). Total energy expenditure was calculated as previously described (20). Metabolizable (absorbed) energy was calculated from nutrient-balance data and total energy storage was calculated as the difference between metabolizable energy and total energy expenditure.
Weight Gain Composition
Weight-gain composition was determined as previously reported (23). The amounts of protein and energy stored were known from nitrogen and energy-balance studies, respectively. Energy stored as protein was assessed; the remaining nonprotein energy stored consisted mainly of fat. Protein and energy stored were expressed as percent of weight gain.
Based on results from previous nutrient-balance studies (15,16), we calculated that a study group of 14 neonates (7 in each feeding group) would be necessary to detect a clinically relevant difference of 5 percentage points in the nitrogen-absorption rate between the PHPF and the SPF groups, with a P value of 0.05 and a power of 80%.
Retested macronutrient and amino acid contents of the two formulas were expressed as means ± SD and compared using Student t-test. Means ± SD and a two-tailed Student t-test were also used to express and compare normal distributions of the main clinical characteristics of the 16 participating preterm infants at birth and different stages of the trial.
For each group, non–normally distributed ingested, absorbed, and retained nitrogen quantities were expressed as medians (25th–75th percentiles). Differences between means with a 95% confidence interval (CI) were computed. The same was done with calcium and phosphorus quantities. Absorption rates were defined as nitrogen absorbed divided by nitrogen intake, utilization rates as nitrogen retained divided by nitrogen absorbed, and biological values as nitrogen retained divided by nitrogen intake. Absorption and utilization rates were expressed as percentages and compared using the Mann-Whitney U test. The same definitions and tests were used for calcium and phosphorus.
Because part of the differences in absorbed and retained nitrogen, calcium, and phosphorus quantities could be attributable to initial nitrogen, calcium, and phosphorus intakes, proportional corrections were used to control for differences in formula content. New differences were calculated between means as if nitrogen, calcium, and phosphorus intakes were initially identical in the two groups (i.e., if nitrogen content in the PHPF was identical to the mean diet content in the SPF group). Precisely, the mean SPF nitrogen, calcium, or phosphorus value was considered as constant intake and individual absorption rates were used to obtain adjusted individual absorbed quantities and, subsequently, adjusted individual retained quantities. Differences between the new means and corresponding 95% CIs were calculated to express absorption and retention differences. We used Statview software SE V1.04 (Abacus Concepts; Berkeley, CA).
The study protocol was approved by the Ethics Committee of Lyon University and authorized by the French Minister of Health. An informed parental consent was obtained for each participant.
Tests carried out in the Nutrition Laboratory in Liege revealed that energy, fat, phosphorus, and magnesium contents were similar in the two formulas; however, we found unexpected differences in nitrogen and calcium contents (Table 1). According to our measurement of nitrogen content from amino acids in the two study formulas, the nonprotein nitrogen content of PHPF and SPF represented 14% and 17% of total nitrogen content, respectively.
At birth and at time of inclusion, we did not find significant differences between the baseline characteristics of the two groups (Table 2). However, at time of nutrient balance, gestational age was slightly lower in the SPF group than in the PHPF group (Table 3). The infants diets were switched from banked human milk to mother's milk or formula at a postnatal age of 12 ± 3 days in the PHPF group and 15 ± 3 days in the SPF group (i.e., the time of inclusion). Adequate volumes of oral feeding were reached at the same postnatal age in the two groups (PHPF: 16 ± 8 days; SPF: 17 ± 7 days).
Because nitrogen content of PHPF was lower than that of SPF (Table 1), nitrogen intake was lower in the PHPF group than in the SPF group (Table 4). PHPF–fed infants exhibited a significantly lower nitrogen absorption rate than SPF–fed infants did (P = 0.03), and nitrogen absorbed was significantly lower in PHPF–fed infants than in SPF–fed infants (514 [500—534] mg · kg −1 · day −1 vs. 606 [537—612] mg · kg −1 · day −1 , P = 0.03). After proportional correction for the differences in nitrogen intakes, the difference in mean nitrogen absorbed quantities was 32.8 mg · kg −1 · day −1 (95% CI: 9.9; 55.9) and still significant (P = 0.03). Significant differences in nitrogen utilization rate between the two groups were not observed (P = 0.63), but mean retained nitrogen quantities (P = 0.11) and biological value (P = 0.13) tended to be lower in infants fed PHPF than in infants fed SPF. The difference in mean nitrogen retained was 66.8 mg · kg −1 · day −1 (95% CI: -2.0; 135.5, P = 0.11) decreasing to 36.8 mg · kg −1 · day −1 (95% CI: -11.1; 81.9, P = 0.11) after proportional correction for the differences in nitrogen intakes.
The calcium absorption rate was slightly higher in infants fed PHPF than in infants fed SPF but the difference was not statistically significant (Table 4). Because calcium intake was significantly higher in infants fed PHPF, the amounts of calcium absorbed were higher in the PHPF group than in the SPF group (P = 0.02). After correction for differences in calcium intakes, the difference in mean calcium absorbed was −11.4 mg · kg −1 · day −1 (95% CI: −28.9; 6.2), which was not significant (P = 0.19). Significant differences in calcium utilization rate between the two groups were not observed (P = 0.22) (Table 4), but retained calcium remained higher in infants fed PHPF than in infants fed SPF (P = 0.02). The difference in means after correction was −10.2 mg · kg −1 · day −1 (95% CI: −27.0; 6.7), which was not significant (P = 0.24). There was no significant difference in phosphorus intestinal absorption rates between the two feeding groups but significantly more phosphorus was retained in infants fed PHPF than in infants fed SPF (P = 0.02) (Table 4), which was still significant after proportional correction for differences in phosphorus intake and nitrogen retained (P = 0.01).
There were no differences in plasma amino acid concentrations between the two groups. Amino acid values remained within the normal ranges (Fig. 1). In particular, threonine, histidine, and tryptophan plasma concentrations were similar in PHPF– and SPF–fed infants (P = 0.19, P = 0.87, and P = 0.75, respectively).
Clinical tolerance was evaluated on a daily basis from inclusion until discharge (37 ± 15 days for the PHPF group vs. 33 ± 6 days for the SPF group). Instances of vomiting or gastric residues were similar in the two feeding groups (2.4 ± 3.0 vs. 1.4 ± 0.8, respectively;P = 0.74). During the nutrient balance study, the transit time was significantly shorter in PHPF–fed infants than in SPF–fed infants (18 ± 4 hours vs. 25 ± 10 hours, respectively;P = 0.02).
Energy intake, fat and energy absorption rates, energy expenditure, and nonprotein energy retained were not significantly different between the two groups. The weight gains calculated over a 7-day period at the time of nutrient-balance were similar in the two feeding groups (21.9 ± 2.1 g · kg −1 · day −1 in the PHPF group vs. 23.0 ± 2.0 g · kg −1 · day −1 in the SPF group, P = 0.24) and their compositions were not significantly different. When expressed as percentages of weight gained, protein storage was not significantly different between the two feeding groups (10% [9.8–11.3] vs. 12% [9.5–14.8], respectively;P = 0.31).
Anthropometric measurements at the time of nutrient balance (Table 3) and at discharge from the neonatal unit were similar in the two feeding groups. At theoretical term, body weight was significantly lower in the PHPF group than in the SPF group, whereas the other anthropometric parameters were similar in the two feeding groups (Table 3). That difference was the result of a significantly better weight gain, from inclusion to discharge, in infants fed SPF than in those fed PHPF (32.8 ± 2.4 g/day vs. 28.1 ± 5.1 g/day, P = 0.03). There was no difference in weight gain between the two groups afterwards, from discharge to the end of the study period. Between inclusion and theoretical term, the mean change in weight standard deviation scores was similar in the two feeding groups (−0.3 ± 1.0 for the PHPF group vs. −0.4 ± 0.9 for the SPF group, P = 0.87).
A lower nitrogen intestinal absorption rate was observed in infants fed PHPF compared with infants fed SPF, which is consistent with our previous results (10,11) and in agreement with the results of other authors. Although there is controversy about the effect of protein hydrolysis on nitrogen absorption in animals (24,25), it has been shown that, in the normal human jejunum, increasing peptide chain length has a negative effect on nitrogen absorption (26). This could be explained by the finding that, in the small intestine, absorption of dipeptides and tripeptides depends on size-specific systems, which are faster than amino acid uptake and transport systems (27,28).
Optimal absorption of nutrients requires adequate contact time with the intestinal mucosa. In animals, intestinal transit is influenced by the amount of protein ingested and by the degree of protein hydrolysis. Proteins have been shown to inhibit intestinal transit more often when in their intact form than in their hydrolyzed form (24). A correlation between intestinal transit time and absorption of nutrients in humans has been reported (29,30). In the current study, a shorter transit time was observed with PHPF than with SPF. The reduced contact time of PHPF with the intestinal mucosa could explain the lower nitrogen absorption rate. Another explanation could be that there is a more intense Maillard reaction in infants fed hydrolyzed protein than in those fed entire protein. This reaction occurs between reducing carbohydrates (e.g., lactose) and the free amino group of amino acids that results from hydrolysis. The resulting complexes are no longer available for metabolism (31). The difference in nitrogen content between the two formulas was partially related to a difference in nonprotein nitrogen between PHPF and SPF. Because of the small number of subjects, the difference in nitrogen retained between the two feeding groups did not reach statistical significance before or after correction for the difference in nitrogen intake. Although the difference in biological value did not reach statistical significance, our results suggest that nitrogen intake should be 10% higher than in entire-protein formula each time a partially hydrolyzed protein source is used in preterm formula.
Similar plasma amino acid concentrations were observed in the two feeding groups without previously observed imbalances such as high plasma threonine levels (≅50 μmol/L) and low histidine levels (≅3 μmol/L) (10,11). In preterm infants fed whey-predominant formula, the change in protein source from sweet-whey to acid-whey proteins resulted in lower plasma threonine levels caused by the absence of threonine-rich glycomacropeptide (32). The present study suggests that similar results can be obtained with acid-whey hydrolyzed proteins. After histidine supplementation in PHPF, plasma histidine levels were increased when compared with those of our previous studies and were in the normal range in the two feeding groups (10,11). There are few randomized studies about plasma amino acid homeostasis in preterm infants fed PHPF. Using a partially hydrolyzed, whey-predominant preterm formula, Mihatsch et al. (33) observed plasma amino acid levels similar to those in control participants who were fed SPF.
Because calcium absorption proceeds by a nonsaturable route and is a linear function of calcium intake in preterm infants (34), the amount of calcium absorbed in the PHPF–fed group was higher than in the SPF–fed group. However, calcium intestinal absorption rates tended also to be higher in the PHPF–fed group when compared with the SPF–fed group, although the difference was not significant because of the small number of subjects. In subjects fed the new PHPF, calcium absorption rates were even higher than they had been in subjects fed experimental protein-hydrolysate preterm formulas in previous studies (about 40%) (11). Such an improvement may be because calcium glycerophosphate, a more soluble form of calcium and phosphorus, represented a greater proportion of the calcium content than it did in previous experimental formulas (11,35). That improvement was also observed in phosphorus balance. After correction for differences in phosphorus intake and nitrogen retention, phosphorus retained remained higher in infants fed PHPF than in infants fed SPF and could be related to the difference in calcium retention.
Clinical tolerance was excellent in the two feeding groups and no perianal dermatitis was observed as was previously reported (10).
When they reached an age equivalent to theoretical term, the subjects fed SPF were heavier than the subjects fed PHPF, but the weight difference was not associated with differences in crown-heel length or head circumference. Body weights in the two feeding groups were within the range of the reference values (13). The variation in body weight between the two groups could be related to the lower amount of nitrogen retained in PHPF–fed infants than in SPF–fed infants. Although the difference in nitrogen retained (≅95 mg · kg −1 · day −1 ) was not statistically significant, it could have been clinically relevant over a 2-month period. In the latter case the difference in growth could have been related to the unexpected difference in nitrogen content between the two study formulas rather than to the difference in nitrogen absorption rates, and could have disappeared in subjects fed formulas with the same nitrogen content. Therefore, further randomized studies with larger groups are required to assess the long-term effects on growth of PHPF feeding on premature infants.
In conclusion, compared with entire-protein SPF, PHPF feeding of preterm infants induced a lower nitrogen intestinal absorption rate. This lower rate suggests that nitrogen intake should be 10% higher than with entire-protein formula each time a partially hydrolyzed protein source is used in preterm formula. In addition, the use of hydrolyzed acid and rennet-whey protein and the increase in calcium glycerophosphate content of the experimental formula resulted in improved plasma amino acid profiles and mineral utilization, respectively. Because the lower weight gain observed in infants who were fed PHPF during the first weeks of life may be related to the unexpected difference in nitrogen intake, further studies with larger groups are needed to evaluate the effects on growth before longitudinal studies to investigate the non-nutritional benefits of PHPF in very low birth-weight infants can be performed.
The authors thank Geraint Jones, Jean Iwaz, and René Ecochard for editorial assistance.
1. Billeaud C, Guillet J, Sandler B. Gastric emptying in infants with or without gastro-oesophageal reflux according to the type of milk. Eur J Clin Nutr 1990; 44: 577–83.
2. Lucas A, McLaughlan P, Coombs RRA. Latent anaphylactic sensitization of infants of low-birth weight to cow's milk proteins. BMJ 1984; 289: 1254–6.
3. Moughan PJ, Cranwell PD, Smith WC. An evaluation with piglets of bovine milk, hydrolyzed bovine milk, and isolated soybean proteins included in infants milk formulas. II. Stomach-emptying rate and the postprandial change in gastric pH and milk-clotting enzyme activity. J Pediatr Gastroenterol Nutr 1991; 12: 253–9.
4. Chandra RK, Singh G, Shridara B. Effect of feeding whey hydrolysate, soy and conventional cow milk formulas on incidence of atopic disease in high risk infants. Ann Allergy 1989; 63: 102–6.
5. Lucas A, Brooke OG, Morley R, et al. Early diet of preterm infants and development of allergic or atopic disease: randomized prospective study. BMJ 1990; 300: 837–40.
6. Savilahti E, Tuomikoski-Jaakkola P, Järvenpää AL, et al. Early feeding of preterm infants and allergic symptoms during childhood. Acta Paediatr 1993; 82: 340–4.
7. Rigo J, Verloes A, Senterre J. Plasma amino acid concentrations in term infants fed human milk, a whey-predominant formula, or a whey hydrolysate formula. J Pediatr 1989; 115: 752–5.
8. Vandenplas Y, Hauser B, Blecker U, et al. The nutritional value of a whey hydrolysate formula compared with a whey-predominant formula in healthy infants. J Pediatr Gastroenterol Nutr 1993; 17: 92–6.
9. Rigo J, Salle BL, Putet G, et al. Nutritional evaluation of various protein hydrolysate formulas in term infants during the first month of life. Acta Paediatr Suppl
10. Rigo J, Salle BL, Picaud JC, et al. Nutritional evaluation of protein hydrolysate formulas. Eur J Clin Nutr 1995; 49: S26–38.
11. Rigo J, Senterre J. Metabolic balance studies and plasma amino acid concentrations in preterm infants fed experimental protein hydrolysate preterm formulas. Acta Paediatr Suppl
12. Boehm G, Cervantes H, Georgi G, et al. Effect of increasing dietary threonine intakes on amino-acid metabolism of the central nervous system and peripheral tissues in growing rats. Pediatr Res 1998; 44: 900–6.
13. Usher R, Mc Lean F. lntrauterine growth of liveborn Caucasian infants at sea level: standards obtained from measurements in 7 dimensions of infants born between 25 and 44 weeks of gestation. J Pediatr
14. Walsh MC, Kliegman RM. Necrotizing enterocolitis: treatment based on staging criteria. Pediatr Clin North Am 1986; 33: 179–201.
15. Putet G, Senterre J, Rigo J, et al. Nutrient balance, energy utilization and composition of weight gain in very-low-birth-weight infants fed pooled human milk or a preterm formula. J Pediatr 1984; 105: 79–85.
16. Picaud JC, Putet G, Rigo J, et al. Metabolic and energy balance in small and appropriate for gestational age very low birth weight infants. Acta Paediatr 1994; 405 (suppl): 54–9.
17. Rigo J, Putet G, Picaud JC, et al. Nitrogen balance and plasma amino acids in the evaluation of protein sources for extremely low birth weight infants. In: Ziegler EE, Lucas A, Moro GE, eds. Nutrition of Very Low Birth Weight Infants. Philadelphia: Lippincott Williams & Wilkins, 1999: 139–49.
18. Westerterp KR, Lafeber HN, Sulkers EJ, et al. Comparison of short term indirect calorimetry and doubly labeled water method for the assessment of energy expenditure in preterm infants. Biol Neonate 1991; 60: 75–82.
19. Wauben I, Westerterp K, Gerver WJ, et al. Effect of varying protein intake on energy balance, protein balance and estimated weight gain composition in premature infants. Eur J Clin Nutr 1995; 49: 11–6.
20. Roberts SB, Coward WA, Schlingenseipen K, et al. Comparison of the doubly labeled water method with indirect calorimetry and a nutrient balance study for simultaneous determination of energy expenditure, water intake, and metabolizable energy intake in preterm infants. Am J Clin Nutr 1986; 44: 315–22.
21. Jones PJM, Winthrop AL, Schoeller DA, et al. Validation of doubly labeled water for assessing energy expenditure in infants. Pediatr Res 1987; 21: 242–6.
22. Picaud JC, Normand S, Pacchiaudi C, et al. Natural isotopic abundance in oxygen 18 and deuterium in very low birth weight infants: influence on measurement of energy expenditure by doubly labeled water method (abstract). Pediatr Res
23. Putet G. Energy. In: Tsang RC, Lucas A, Uauy R, et al, eds. Nutritional Needs of the Preterm Infant: Scientific Basis and Practical Guidelines. New York: Caduceus Medical Publishers, 1993: 15–28.
24. Zhao XT, McCamish MA, Miller RH, et al. Intestinal transit and absorption of soy-protein in dogs depend on load and degree of protein hydrolysis. J Nutr 1997; 127: 2350–6.
25. Cezard JP, Tran TA, Macry J, et al. Effects of two protein hydrolysates on growth, nitrogen balance, and small intestine adaptation in growing rats. Biol Neonate 1994; 65: 60–7.
26. Grimble GK, Keohane PP, Higgins BE, et al. Effect of peptide chain length on amino acid and nitrogen absorption from two lactalbumin hydrolysates in the normal human jejunum. Clin Sci 1986; 71: 65–9.
27. Keohane PP, Grimble GK, Brown B, et al. Influence of protein composition and hydrolysis on intestinal absorption of protein in man. Gut 1985; 26: 907–13.
28. Gropper SS, Grooper DM, Acosta PB. Plasma amino acid response to ingestion of l -Amino acids and whole protein. J Pediatr Gastroenterol Nutr 1993; 16: 143–50.
29. Lin H, Zhao XT, Wang LJ, et al. Fat absorption is not complete by midgut but dependant on load of fat. Am J Physiol 1996; 271: G62–7.
30. Holgate AM, Read NW. Relationship between small bowel transit time and absorption of a solid meal: influence of metoclopramide, magnesium, sulfate, and lactulose. Dig Dis Sci 1983; 28: 812–9.
31. Finot PA, Magnenat E. Metabolic transit of early and advanced Maillard products. Prog Food Nutr Sci 1981; 5: 193–207.
32. Rigo J, Boehm G, Georgi G, et al. Effect of an infant formula free of glycomacropeptide on threonine metabolism in preterm infants . J Pediatr Gastrenterol Nutr
33. Mihatsch WA; Pohlandt F. Protein hydrolysate formula maintains homeostasis of plasma amino acids in preterm infants. J Pediatr Gastroenterol Nutr 1999; 29: 406–10.
34. Bronner F, Salle BL, Putet G, et al. Net calcium absorption in premature infants: results of 103 metabolic balance studies. Am J Clin Nutr 1992; 56: 1037–44.
35. Hanning RM, Atkinson SA, Whyte RK. Efficacy of calcium glycerophosphate vs. conventional mineral salts for total parenteral nutrition in low birth weight infants: a randomized clinical trial. Am J Clin Nutr 1991; 54: 903–8.
Keywords:© 2001 Lippincott Williams & Wilkins, Inc.
Preterm neonates; Nutrition; Nitrogen; Calcium; Amino acids; Growth