Higher urinary excretion of essential amino acids in preterm infants fed protein hydrolysates.

AIM: Protein hydrolysates have been introduced in preterm formulae, but it is not clear whether they are needed for the feeding of preterm infants. We designed a randomized, controlled trial to test the effects of a preterm formula with hydrolysed cow's milk proteins on short-term growth and urinary and plasma amino acids levels. METHODS: Infants with a birthweight < or = 1750 g and gestational age < or = 34 wk fed a conventional preterm infant formula (formula B) or a hydrolysed formula (formula A). Weight was measured daily; length, head circumference, mid-arm circumference and total skinfold thickness were measured weekly. Blood and urine were analysed for amino acid concentrations at start, 14 and 28 d. RESULTS: Twenty-one infants met the criteria for randomization. The daily feeding volumes were: formula A 172.8 +/- 5.6 vs formula B 170.1 +/- 2.8 ml/kg/d. Infants fed with formula A showed slower weight gain (17.4 +/- 3.4 vs 20.5 +/- 3.3 g/kg/d; p = 0.045) and lower mean change in Z-scores for weight (-0.18 +/- 0.16 vs 0.00 +/- 0.09; p = 0.009) and for head circumference (-0.06 +/- 0.13 vs 0.06 +/- 0.13; p = 0.049). After 14 d, infants receiving formula A had statistically significant higher urinary levels of essential amino acids compared to infants receiving formula B. CONCLUSION: Our results support the hypothesis of less nutritional value of hydrolysed versus conventional preterm formulae. Higher renal excretion of essential amino acids may be one of the mechanisms involved. These findings must be confirmed by further studies with larger sample sizes and protein hydrolysates with different degrees of hydrolysis.

An infant formula free of glycomacropeptide prevents hyperthreoninemia in formula-fed preterm infants.

BACKGROUND: Hyperthreoninemia is a well-known phenomenon in infants fed a whey protein-predominant formula. Sweet whey is commonly used for the production of these whey-predominant infant milk formulas. Sweet whey contains not only whey proteins but also the threonine-rich glycomacropeptide (GMP). In the current study, an experimental formula based on acid whey without GMP and a formula based on sweet whey with GMP (threonine content 17.2% higher than in the experimental formula) but otherwise with identical composition were tested with particular respect to threonine metabolism. METHODS: Fourteen preterm infants appropriate for gestational age were enrolled in this randomized cross-over study. After a feeding period of at least 7 days, the nutrition of each infant was switched to the other formula for the second feeding period. At the end of each feeding period, the concentrations of creatinine and amino acids in the plasma and in the urine were measured. RESULTS: In the plasma, the threonine concentration was significantly lower in the group fed the experimental GMP-free formula than in the group fed the sweet whey formula (P < 0.001). Renal excretion of all essential amino acids was generally very low and less than 2% of the intake, indicating that the kidneys had no marked homeostatic function with respect to plasma amino acid. The plasma concentrations of the threonine metabolites glycine and serine, and that of urea were not influenced by diet. CONCLUSION: Feeding a whey protein-predominant bovine milk produced from acid whey protein reduces significantly the hyperthreoninemia commonly found in formula-fed preterm infants. Thus, acid whey formulas should be recommended for feeding preterm infants.

Mass spectrometric investigations of human milk oligosaccharides.

Human milk oligosaccharides (OS) have been fractionated by gel permeation chromatography (GPC) and analyzed by mass spectrometry (MS). Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF/MS) and electrospray ionization (ESI) ion trap/MS were used. Using a large human milk pool, the MALDI-TOF/MS analysis of high-molecular-mass GPC fractions showed that complex, multiply fucosylated and/or sialylated OS are present over a larger mass range than described previously Acidic oligosaccharides could be detected from low-retention-time GPC fractions with masses up to 4000 Da, which has not been reported before.

Detection of four human milk groups with respect to Lewis-blood-group-dependent oligosaccharides by serologic and chromatographic analysis.

Oligosaccharides from human milk samples obtained from individual donors were analyzed using high-pH anion-exchange chromatography. Three patterns of neutral oligosaccharides were detected corresponding to milk groups already described. These oligosaccharide groups correspond to the Lewis blood types Le(a-b+), Le(a+b-), and Le(a-b-). A new carbohydrate pattern was detected in a milk sample from a Le(a-b-) person in which only nonfucosylated oligosaccharides and compounds bearing alpha1,3-linked fucosyl residues were found. This finding led to the hypothesis that there exist 4 different oligosaccharide milk groups that fit well to the genetic basis of the Lewis blood group system.

Human milk oligosaccharides are resistant to enzymatic hydrolysis in the upper gastrointestinal tract.

BACKGROUND: Human milk oligosaccharides (HMOs) show a complexity and variety not found in milk of any other species. Although progress has been made in the past 3 decades with regard to identification and structural characterization of HMOs, not much is known about the physiologic functions of HMOs. OBJECTIVE: As a prerequisite for biological activity in infant metabolism, HMOs have to resist enzymatic hydrolysis in the gastrointestinal tract. To assess the extent to which selected HMOs are hydrolyzed, we carried out in vitro digestion studies using enzyme preparations of human and porcine pancreas and intestinal brush border membranes (BBMs). DESIGN: Fractions of HMOs, including structurally defined isolated oligosaccharides, were digested for up to 20 h with human pancreatic juice and BBMs prepared from human or porcine intestinal tissue samples. HMOs were incubated by using a porcine pancreatic homogenate and BBMs as enzyme sources. HMOs and digestion products were identified by mass spectrometry and anion-exchange chromatography. Additionally, free D-glucose, L-fucose, and N-acetylneuraminic acid were determined enzymatically. RESULTS: Whereas maltodextrin (control) was rapidly and completely hydrolyzed, neutral and acidic HMOs showed a profound resistance against pancreatic juice and BBM hydrolases. However, cleavage of most of the HMOs was achieved by using a pancreatic homogenate containing intracellular, including lysosomal, enzymes in addition to secreted enzymes. CONCLUSIONS: The results of this study strongly suggest that HMOs are not hydrolyzed by enzymes in the upper small intestine. Although intact HMOs may be absorbed, we postulate that a majority of HMOs reach the large intestine, where they serve as substrates for bacterial metabolism. Therefore, HMOs might be considered the soluble fiber fraction of human milk.

Postprandial plasma amino acids in preterm infants: influence of the protein source.

Preprandial plasma amino acid concentrations have been used extensively as a marker of the nutritional value of dietary proteins in preterm infants. This study investigated the postprandial plasma amino acid profiles of preterm infants fed with different dietary proteins at similar protein intakes during the first weeks of life. In 12 preterm infants, pre- and postprandial plasma amino acid concentrations were measured before the removal of an indwelling central venous catheter placed for parenteral nutrition. All infants received breast milk until the time of study. At the start day of the study, infants were randomized to receive a test meal of 10 ml/kg, either of breast milk fortified with breast milk protein to reach a protein content of 2.0 g/dl or of a bovine milk preterm formula with a protein content of 2.0 g/dl (whey/casein ratio 60/40). Five samples of 100 microl blood were obtained immediately before and 15, 30, 45 and 60 min after the test meal. The plasma amino acid analysis was performed by a reversed-phase high-performance liquid chromatography based on o-phthaldialdehyde/2-mercaptoethanol pre-column derivatization. In both groups, the plasma amino acid concentrations increased within the first 30 min and the levels did not return to the preprandial baseline during the observation period. Fifteen minutes after the test meal, the plasma levels of all essential amino acids with the exception of histidine were higher in the bovine milk formula fed infants than in the fortified breast milk fed infants. The sum of plasma essential amino acid levels found in the formula fed infants were significantly (p < 0.05) higher than the levels found in the fortified breast milk fed infants at 15, 30 and 45 min. The kinetics of individual amino acids were influenced by the different quality of the protein even when the intakes in the groups were similar, as demonstrated for histidine and phenylalanine. The data indicate that postprandial plasma amino acid concentrations depend significantly on the dietary amino acid source and cannot simply be calculated from the amino acid composition of dietary proteins. Therefore, postprandial plasma amino acid concentrations should be included in the nutritional evaluation of dietary proteins in preterm infants.

Analysis of high-molecular-weight oligosaccharides from human milk by liquid chromatography and MALDI-MS.

Pooled human milk oligosaccharides were fractionated by anion-exchange chromatography on AG 1-X2 and by an improved gel filtration procedure that allowed the separation of large oligosaccharides on Toyopearl HW 40 (S) and Bio-Gel P-6 columns, respectively. The analysis of the resulting nonderivatizated fractions by matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) revealed several neutral and acidic high-molecular-weight oligosaccharides. So far unknown acidic oligosaccharides containing up to 20 monomers were detected in a molecular mass range of 2094-3626 Da. Furthermore, neutral structures containing up to 35 monosaccharides were identified after fractionation on Toyopearl HW 40 (S) and subsequent P-6 fractionation, demonstrating the suitability of the applied method for the preparation of oligosaccharides in this high-molecular-mass range. The composition of the detected oligosaccharides was found to be the same as those previously identified in oligosaccharides of lower masses. However, an enormous structural heterogeneity was observed when acidic and neutral fractions were characterized by high-pH anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD). From our analysis we may conclude that each molecular mass identified by MALDI-MS corresponds to a variety of isomeric structures. The total number of oligosaccharides occurring in human milk may consequently be much higher than estimated before.

Matrix optimization for matrix-assisted laser desorption/ionization mass spectrometry of oligosaccharides from human milk.

Neutral and acidic oligosaccharides from human milk were analyzed by matrix-assisted laser desorption/ionization mass spectrometry (MALDI/MS). These experiments require suitable matrices; their selection and particularly their preparation protocols must be optimized. Important criteria are sensitivity, reproducibility, tolerance against impurities and resolution over a wide mass range. For analytical investigations of these oligosaccharides, containing labile fucosylated and sialylated components, another property of a matrix becomes a significant factor, namely the influence on ion stability and the extent of (metastable) fragmentation. The experience gained with the MALDI/MS of neutral and acidic oligosaccharides is summarized taking into account different intentions of measurement and typical problems, such as impurities after enzymatic treatment. For a rapid screening of an oligosaccharide sample, superior results were obtained with a new preparation technique using 5-chloro-2-mercaptobenzothiazole (CMBT) as the first layer for 2,5-dihydroxybenzoic acid. For structural analysis by post-source decay, CMBT as the first layer for 3-aminoquinoline is a favoured preparation protocol, because extensive fragmentation is achieved. For acidic oligosaccharides, a special preparation protocol makes it possible to determine the number of sialic acids by inducing highly effective cationization. Matrix-assisted laser desorption/ionization mass spectrometry; matrices; oligosaccharides; post-source decay.

Effect of increasing dietary threonine intakes on amino acid metabolism of the central nervous system and peripheral tissues in growing rats.

The threonine content of most of the infant formulas currently on the market is approximately 20% higher than the threonine concentration in human milk. Due to this high threonine content the plasma threonine concentrations are up to twice as high in premature infants fed these formulas than in infants fed human milk. To study the effect of different threonine intakes on plasma and tissue amino acid concentrations, 24 young male Wistar rats were fed three experimental diets based on a mixture of bovine proteins with a whey protein/casein ratio of 60/40 with different threonine contents [group A, 0.86 g of threonine/100 g (n = 8); group B, 1.03 g of threonine/100 g (n = 8); group C, 2.21 g of threonine/100 g (n = 8)]. Eight animals were fed a typical rat diet based on bovine casein as controls. After a feeding period of 15 d, amino acids were measured in plasma and in homogenates of the cerebral cortex, brain stem, liver, and muscle. There was a significant correlation between threonine intake and plasma threonine levels (r = 0.687, p < 0.001). The plasma threonine concentration correlated significantly with the threonine concentration in the cortex (r = 0.821, p < 0.01) and the brain stem (r = 0.882, p < 0.01). There was a positive significant correlation between threonine and glycine concentrations in the cortex (r = 0.673, p < 0.01), and the brain stem (r = 0.575, p < 0.01), whereas the glycine concentration decreased with increasing threonine intakes in the liver and muscle. The presented data indicate that increasing the threonine in plasma leads to increasing brain glycine and thereby affects the neurotransmitter balance in the brain. This may have consequences for brain development during early postnatal life. Therefore, excessive threonine intake during infant feeding should be avoided.

Experimental milk formulae with reduced protein content and desialylated milk proteins: influence on the faecal flora and the growth of term newborn infants.

We have assessed the growth, tolerance and the faecal flora composition in healthy infants on different feeding regimens. Four groups of infants were fed exclusively on mother's milk, a standard formula and two experimental formulae. The first experimental formula consisted of a milk with a reduced protein content (1.2 g/100 ml), the second in a formula with the same protein content and with milk proteins desialylated by mild acid hydrolysis. The aim of the study was to test whether lowering the protein content and/or modifying the proteins by desialylation would favour the development of a bifidus flora. A bifidus flora was detected in 60% of breastfed infants at 1 month of life. All formulae employed during the study failed to induce a prevalence of colonization with bifidobacteria at 1 month of age. The two experimental milk formulae were well tolerated, but the infant growth rate was slightly lower as compared to the breastfed infants and the infants fed the standard formula. The presence in milk formulae of pre-digested and desialylated proteins can offer some advantages in term of digestibility and mimic a physiological intestinal mechanism of the infant.

[3-13C] gamma-linolenic acid: a new probe for 13C nuclear magnetic resonance studies of arachidonic acid synthesis in the suckling rat.

Our objective was to develop a suitable probe to study metabolism of polyunsaturated fatty acids by 13C nuclear magnetic resonance (NMR) in the suckling rat pup. [3-13C] gamma-Linolenic acid was chemically synthesized, and a 20 mg (Experiment 1) or 5 mg (Experiment 2) dose was injected into the stomachs of 6-10-day-old suckling rat pups that were then killed over a 192 h (8 d) time course. 13C NMR showed that 13C in gamma-linolenate peaked in liver total lipids by 12-h post-dosing and that [5-13C]-arachidonic acid peaked in both brain and liver total lipids 48-96 h post-dosing. 13C enrichment in brain gamma-linolenic acid was not detected by NMR, but gas chromatography-combustion-isotope ratio mass spectrometry showed that its mass enrichment in brain phospholipids at 48-96 h post-dosing was 1-2% of that in brain arachidonic acid. 13C was present in liver and brain cholesterol and in perchloric acid-extractable water-soluble metabolites in the brain, liver and carcass. We conclude that low but measurable amounts of exogenous gamma-linolenic acid do access the suckling rat brain in vivo. The slow time course of [5-13C] arachidonic acid appearance in the brain suggests most of it was probably transported there after synthesis elsewhere, probably in the liver. Some carbon from gamma-linolenic acid is also incorporated into lipid products other than n-6 long-chain polyunsaturated fatty acids.

Detection of four human milk groups with respect to Lewis blood group dependent oligosaccharides.

Neutral oligosaccharides in human milk samples from approximately 50 women were analysed applying a recently developed high-pH anion-exchange chromatographic method. Three different oligosaccharide patterns could be detected in accordance with milk groups that had been already described. These oligosaccharide groups correspond to the Lewis blood types Le(a-b+), Le(a+b-) and Le(a-b-). In addition to these oligosaccharide patterns, a new carbohydrate pattern was detected in a milk sample from a Le(a-b-) individual. Here, only nonfucosylated oligosaccharides and compounds bearing alpha1,3 linked fucosyl residues were found, whereas structures with alpha1,2 and alpha1,4 fucosyl linkages were missing. This finding led to the hypothesis that there are four different oligosaccharide milk groups that fit well to the genetic basis of the Lewis blood group system.

Influence of the derivatization procedure on the results of the gaschromatographic fatty acid analysis of human milk and infant formulae.

Many different analytical procedures for fatty acid analysis of infant formulae and human milk are described. The objective was to study possible pitfalls in the use of different acid-catalyzed procedures compared to a base-catalyzed procedure based on sodium-methoxide in methanol. The influence of the different methods on the relative fatty acid composition (wt% of total fatty acids) and the total fatty acid recovery rate (expressed as % of total lipids) was studied in two experimental LCP-containing formulae and a human milk sample. MeOH/HCl-procedures were found to result in an incomplete transesterification of triglycerides, if an additional nonpolar solvent like toluene or hexane is not added and a water-free preparation is not guaranteed. In infant formulae the low transesterification of triglycerides (up to only 37%) could result in an 100%-overestimation of the relative amount of LCP, if these fatty acids primarily derive from phospholipids. This is the case in infant formulae containing egg lipids as raw materials. In formula containing fish oils and in human milk the efficacy of esterification results in incorrect absolute amounts of fatty acids, but has no remarkable effect on the relative fatty acid distribution. This is due to the fact that in these samples LCP are primarily bound to triglycerides. Furthermore, in formulae based on butterfat the derivatization procedure should be designed in such a way that losses of short-chain fatty acids due to evaporation steps can be avoided. The procedure based on sodium methoxide was found to result in a satisfactory (about 90%) conversion of formula lipids and a reliable content of all individual fatty acids. Due to a possibly high amount of free fatty acids in human milk, which are not methylated by sodium-methoxide, caution is expressed about the use of this reagent for fatty acid analysis of mothers milk. It is concluded that accurate fatty acid analysis of infant formulae and human milk requires a careful and quantitative derivatization of both polar and nonpolar lipid classes. Sodium methoxide seems to be a reliable and time-saving method for routine fatty acid analysis of infant formulae, which should be validated by interlaboratory comparison. Anhydrous procedures based on methanolic hydrogen chloride including an additional nonpolar solvent are also suitable for infant formulae but seem to be preferable for human milk samples.

Quantification of individual oligosaccharide compounds from human milk using high-pH anion-exchange chromatography.

A quantification method suitable for determination of individual oligosaccharide compounds from human milk has been established. The crude milk oligosaccharide fraction was separated into acidic oligosaccharides, neutral oligosaccharides, and lactose by gel permeation chromatography. After this separation step neutral and acidic oligosaccharides were analyzed separately by high-pH anion-exchange chromatography with pulsed amperometric detection. The concentrations of 14 neutral oligosaccharides, of 6 acidic oligosaccharides, and of N-acetylneuraminic acid were determined on the internal standards stachyose and galacturonic acid, respectively. Thus, previously applied quantification methods for milk oligosaccharides based on gel permeation chromatography have been decisively improved.

Influence of two infant formulas and human milk on the development of the faecal flora in newborn infants.

The establishment of the faecal flora of 39 full-term infants fed exclusively on breast milk (n = 20) or with two different modern adapted cow's milk formulas (n = 19) was studied during the first 3 months of life. One formula investigated was based on 100% bovine casein as the protein source whereas the other formula contained bovine milk proteins with a whey/casein ratio of 60:40. A faecal flora rich in bifidobacteria was found in all study groups; the growth of putrefactive bacteria (especially Bacteroides spp.), however, was limited. In formula-fed infants, significantly higher bacterial counts of enterococci and clostridia were detected compared to breast milk-fed infants. Similarities and differences due to the feeding regimen were particularly reflected in the pattern of the anaerobic bacterial species. Bifidobacterium bifidum, B. infantis and B. breve constituted the majority of the bifidobacterial flora independent of the type of milk feeding. Other bifidobacterial species such as B. longum, B. adolescentis, B. parabifidum and B. pseudo-catenulatum were detected in high numbers and at low frequencies in breastfed infants. The latter three were observed in infants fed the whey/casein formula as well. It seems that infants fed a casein formula develop a faecal flora more like that of breastfed infants concerning Lactobacillus spp. (especially L. fermentum and L. brevis).

Vitamin E status of low birthweight infants fed formula enriched with long-chain polyunsaturated fatty acids.

It is recommended in Europe that low birthweight infants (LBWI) who do not receive human milk (HM) should be fed formula enriched with long-chain polyunsaturated fatty acids (LCP). The question has been raised whether LCP supplementation to LBWI formula may have adverse effects on antioxidant status in the recipient infant, particularly on the major lipid-soluble antioxidant vitamin E (alpha-tocopherol, alpha-toc.). We studied alpha-toc. status in LBWI fed HM (n = 15) or formula either without (f, n = 8) or with LCP derived from egg lipids and fish oil (LCP-F, n = 9) on days 4 and 21 of life. Plasma alpha-toc. concentrations increased significantly in infants fed HM [d. 4: 4.53 (1.31), d. 21: 6.35 (2.18), mg/l, mean (SD), p < 0.01], whereas there were no changes in infants fed F of LCP-F. Plasma alpha-toc./total lipid ratios and erythrocyte membrane alpha-toc. concentrations did not change significantly between days 4 and 21 in either group. In contrast, erythrocyte membrane alpha-toc./total lipid ratios decreased significantly in infants fed LCP-F [0.42 (0.13) vs. 0.31 (0.06), p < 0.05], whereas no change occurred in the other two groups. We conclude that an LCP supplementation based on egg lipids and fish oil to formula may induce an early postnatal decrease of alpha-toc./total lipid ratios in erythrocyte membranes of LBWI. Therefore, the effects of different forms of LCP supplementation to infant formula on infantile vitamin E status should be carefully evaluated.

Oligosaccharides from human milk as revealed by matrix-assisted laser desorption/ionization mass spectrometry.

In this study neutral and acidic oligosaccharide fractions prepared from human milk have been investigated using matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). The fraction of neutral oligosaccharides was separated by gel permeation chromatography (GPC) and the resulting subfractions were analyzed by MALDI-MS using the positive ion mode. Several low-molecular-weight glycans (degree of polymerization up to 13) were observed whose structures have already been elucidated. In addition, a variety of so far unknown large-sized carbohydrates was detected whose molecular weights range from M(r) 2242 to 8000. The large-sized glycans which possess a low abundance appear to be composed of both lactosamine and fucose residues attached to the lactose unit at the reducing end of the sugar chains with a highly variable stochiometry. Following subfractionation by GPC, acidic (i.e., containing sialic acid) glycans were analyzed by MALDI-MS using both positive and negative ion mode. Because of the inferior stability of acidic glycans, various matrices were applied and compared with respect to signal intensity, resolution, and analyte stability.

Pitfalls in the design and manufacture of infant formulae.

The composition of modern infant formulae is basically oriented on the "golden standard" human milk and influenced by several official regulations and recommendations (EC, ESPGAN, etc.). This article will focus on two recent improvements in the field of long-chain polyunsaturated fatty acids (LCPs) and protein hydrolysates. The addition of LCPs for preterm formulae was recommended recently by ESPGAN (the European Society for Pediatric Gastroenterology and Nutrition). Our research has focused on this problem for many years and we have found a good source of LCPs using specially prepared egg-yolk lipids. Others have used fish oils and run into the problem of growth retardation. Therefore, the possible sources of LCPs have to be discussed very critically and the alternatives will be shown. Also, new developments like the use of single-cell oils will be discussed. Second, the use of protein hydrolysates have been introduced for the so-called hypoantigenic or hypoallergenic formulae. Hypoantigenic formulae for preventive use have to be differentiated clearly from hypoallergenic formulae for treatment of proved cows' milk protein allergy. The problems of designing suitable hydrolysates that are low in antigenicity and good in taste will be outlined. The determination of the molecular weight distribution by gel chromatography will be compared critically with the newer techniques. The ELISA technique for testing the antigenicity is recommended before any in vivo evaluation. So far, the anaphylactic guinea-pig model is the most sensitive in vivo testing method. Summing up, modern infant formulae manufacture is much more dependent on modern laboratory techniques, which have to be chosen critically and must be adapted to the newest state of the art.

Diet and the essential fatty acid status of term infants.

Long-chain polyunsaturated fatty acids with 20 and 22 carbon atoms (LCPs) seem to play an important role during the rapid development of the infant brain in the late fetal and early postnatal period. These LCPs are integral constituents of biological membranes and they are involved in the regulation of functional properties like fluidity, permeability and activity of membrane-bound enzymes. Human milk contains LCPs in an amount of 0.5-3 wt% of total fatty acids, whereas commercially available infant formulae are almost free of them. Recently, several clinical trials, primarily with preterm infants, have reported that the content of LCPs in the blood and a functional parameter like visual acuity correlate with the content of LCPs in the diet. In this clinical trial we studied the effect of different diets on the fatty acid pattern of plasma and erythrocyte lipids of healthy term infants during the first 3 months of life. Breast-fed infants were compared with formula-fed babies who received a commercially available formula without LCPs or a new experimental formula enriched with LCPs that was similar to human milk. The results indicate that the introduction of milk feeding leads to marked differences in the blood lipid composition during the first months of life, independent of the feeding regimen. Secondly, the supplementation of a formula with LCPs seems to result in a blood lipid composition similar to infants fed with human milk. This supports the hypothesis that the newborn term infant has a limited desaturating capacity and depends on an exogenous supply of LCPs during the first months of life.