Breast milk is universally recognized as the optimal feeding choice for term and preterm newborns and infants (1,2). Under several circumstances, at home or in hospital, milk collection and storage may be necessary. Depending on the length and type of storage, milk may lose some important nutritional and functional properties (3–6). Because of the limited knowledge on which common advice on human milk (HM) storage is based, present recommendations concerning safe HM conservation are far from being univocal (7–9), as recently underlined by Davanzo et al (10). In an attempt to clarify some of the uncertainties existing on the storage of HM, we designed a comprehensive study to investigate the effects of prolonged refrigeration on the nutritive and nonnutritive characteristics of HM. The present experiment is the first study providing extensive information on the biochemical profile of HM, as affected by prolonged refrigeration.
An important nutritional attribute of breast milk lies in its lipid profile and, in particular, in its n-6 and n-3 long-chain polyunsaturated fatty acid (LC-PUFA) content. Arachidonic acid and docosahexanoic acid are essential to normal infant growth in the first year (11), and to retina and brain development (12).
The modifications to the HM lipid profile during prolonged storage have been studied in the past with different methodological approaches and study design, making it difficult to compare the results, leading to uncertainty in recommendations for safe milk storage (13–16). Lipolysis was frequently described in HM as freeze-stored or refrigerated; however, no direct measurements of lipase activity in refrigerated HM have ever been carried out (13,16,17). HM contains 2 lipases, that is, bile salt–dependent lipase (BSDL) and lipoprotein lipase or serum-stimulated lipase (SSL) (18,19). BSDL is known to improve fat digestion and absorption, and plays a role in the digestion of retinol esters, which are a source of vitamin A for the infant (20). Lipoprotein lipase has no reported known function in neonatal digestion, but a role in the uptake of circulating fatty acids into the mammary gland and in the construction of the HM fat pool has been described (21,22).
Conflicting results have been reported in the past concerning the occurrence of lipid peroxidation and the production of dangerous final products, such as malondialdehyde (MDA) and 4-hydroxy-nonenal, during milk refrigeration (23–26), probably because of the different methodologies employed.
The study was carried out to evaluate the complete lipid profile, the lipid-associated enzymatic activities, and oxidative status of refrigerated HM every 24 hours, up to 96 hours of conservation.
Sample Collection and Preparation
HM samples were collected on December 13, 2010 (pool 1), January 24, 2011 (pool 2), and May 2, 2011 (pool 3) at the neonatal intensive care unit (NICU) of the Department of Pediatric and Adolescence Science, University of Turin, Turin, Italy, from 4, 8, and 5 healthy mothers of preterm newborns, respectively. The mean delivery date was 1 month before milk collection. Pooled samples were used to overcome the limitations caused by the milk volumes required by the wide range of biochemical analyses, which could have not been performed on milk from single individuals.
The milk specimens were obtained by complete emptying of 1 breast using an electric breast pump. Immediately after collection, the HM samples were pooled in a sterile bottle (300 mL total volume) and divided into 5 aliquots. One aliquot (0 hour) was analyzed within 3 hours of collection, whereas the others (24 hours, 48 hours, 72 hours, 96 hours) were stored in the refrigerator of the NICU for 24, 48, 72 and 96 hours, respectively. The replicate pools were kept separated and analyzed independently.
The temperature of the refrigerator was constantly monitored by 2 mini data loggers equipped with internal probes (Testo 174T, Lenzkirch, Germany) placed on the bottom and top shelves of the NICU fridge, and programmed to record the temperature every 5 minutes.
Forty milliliters from each HM aliquot were stored at −80°C until used for lipid extraction and analysis. Ten additional milliliters of each aliquot were used for pH determination, thiobarbituric acid reactive substances (TBARS) determination, and MDA quantification by high-performance liquid chromatography. One milliliter of each aliquot was skimmed by centrifugation at 2000g for 30 minutes at 4°C, and used for total antioxidant capacity (ToAC) assay.
The milk samples were thawed in a thermostatic bath at 30°C in the dark, vortexed for 30 seconds, and aliquots were immediately collected for all successive analyses. Before extraction, the samples were flushed with nitrogen to minimize oxidation. Each milk sample was transferred to a screw-capped centrifuge glass tube and 6 mL of solvent solution (chloroform/methanol 2:1, v/v) was added. The mixture was shaken for 30 seconds and left to stand for 5 minutes; then 1 mL of 7% NaCl (w/v) aqueous solution was added, the sample was shaken, and centrifuged at 1500g for 10 minutes. The resulting organic phase was collected in 10-mL tubes and the solvent was removed under nitrogen flow.
Fatty Acids Profile
The methylation procedure proposed by Morrison and Smith (27) was followed, with minor changes: 0.5 mL of each milk sample was spiked with 75 μL of a mix of heptanoic, tridecanoic, and tricosanoic acids as internal standards (400 μg/mL of each component). Lipids were extracted as described above and the resulting residue was dissolved in 4 mL of boron trifluoride methanol complex solution (14% w/v, Sigma-Aldrich, St Louis, MO) and allowed to react at 80°C for 30 minutes to allow a complete transesterification of fatty acids. The fatty acid methyl esters were extracted using 1 mL of hexane. Three independent determinations were performed on each sample.
Free Fatty Acids Determination
Lipid extraction and isolation of free fatty acids (FFAs) were based on the method reported by Huerta-Gonzalez et al (28). Briefly, to 1 mL of sample, spiked with internal standards (25 μL of a solution of a mix of heptanoic, tridecanoic, and tricosanoic acids, 400 μg/mL each), 100 μL of 2.5 mol/L H2SO4 was added. The sample was extracted with 3 mL of diethyl ether/n-hexane (1:1, v/v) in a screw-capped centrifuge tube and then centrifuged for 10 minutes at 2500g. The extraction was repeated twice with 2-mL aliquots of the solvent mixture and all the extracts were combined and applied to a solid-phase extraction column (Bond Elut Carbon/NH2, Agilent Technologies, Santa Clara, CA), previously conditioned with 10 mL of hexane. The neutral lipids were extracted with 16 mL of chloroform/2-propanol (2:1, v/v) and then the FFAs were eluted from the column by addition of 5 mL of 2% v/v formic acid in diethyl ether. The FFA extract was dried under nitrogen flow and the residue was treated with boron trifluoride methanol complex solution as described above. Two independent determinations were carried out for each sample. The procedure and elution volumes were optimized using a standard mix of triacylglycerols and fatty acids.
Determination of Lipid Peroxides
The method applied to this analysis was based on the study by Lagarda et al (29) with some modifications.
Two milliliters of methanol was added to the lipid residue from fat extraction and sonicated for 15 minutes. It was then transferred to a round-bottom flask containing 50 mg of palladium on activated charcoal (10% Pd basis) (Sigma-Aldrich) and, after vacuum air removal, the suspension was stirred under hydrogen in a heated water bath (60°C) for 60 minutes. Then 2 mL of chloroform was added and the palladium/charcoal was removed by filtration through Celite (CeliteS, Sigma-Aldrich). The filtrate was collected in a glass tube, the solvent was removed under nitrogen flow, and the residue was treated with boron trifluoride methanol complex solution as described above. Acetylation of hydroxyl groups was performed as described (29). Standard solutions of 12-hydroxystearic acid were used to evaluate the yield of the procedure.
Gas Chromatography Analysis
Gas chromatography-mass spectrometry determination was carried out on an Agilent 6890 gas chromatograph equipped with an Agilent 7683 B autosampler and an Agilent 5973 mass selective detector. A Varian CP WAX 52CB fused silica capillary column (30-m long, 0.25-mm inner diameter, and 0.25-μm film) was used (Varian Medical Systems Inc, Palo Alto, CA). Both injector and transfer line temperatures were set to 250°C. The samples (1 μL each) were injected in splitless mode for 50 seconds, after which the split ratio was set to 25:1. The oven was programmed as follows: 50°C for 1 minute, increased at 15°C/min to 165°C, held constant for 1 minute, increased at 3°C/min to 211°C, held constant for 5 minutes, increased at 0.5°C/min to 220°C and held constant for 5 minutes. Helium was used as carrier gas at a flow rate of 1.0 mL/min. The detector response factors for different fatty acid methyl esters were calculated by analyzing known quantities of authentic standards (37 component FAME mix, Supelco Inc, Bellefonte, PA). Results are expressed as grams per 100 g total fatty acids.
Determination of Conjugated Dienes
Conjugated dienes were measured, as they are the final products of advanced peroxidation on PUFAs. The fat extracted from 1 mL of each milk sample was dissolved in 5 mL of isooctane. After dilution with 10 volumes of isooctane, 1.5 mL of this solution was transferred into a quartz cuvette (1.0-cm path length) and the absorbance was measured at 233 nm versus blank using a U-2000 spectrophotometer (Hitachi Ltd, Tokyo, Japan) (30). The analyses were performed in triplicate.
Thiobarbituric Acid Test
TBARS are considered to be early indicators of lipid peroxidation. The method for determining TBARS in milk according to King (31) was used. 2.5 mL from each HM sample was mixed with 184 μL of trichloroacetic acid (1 g/mL) and 138 μL of ethanol, vortexed, left to stand for 5 minutes, and centrifuged at 8000 g for 5 minutes. The upper phase (1.5 mL) was transferred into a screw-capped glass tube and 375 μL of 0.05 mol/L thiobarbituric acid in ethanol was added. The solution was then heated at 70°C for 60 minutes in a thermostatic bath and, after cooling, the absorbance values at 532 and 570 nm were measured using a Hitachi U-2000 spectrophotometer and the difference calculated. The TBARS were determined based on a standard curve prepared with different dilutions of MDA (micromoles). The analyses were performed in triplicate.
Quantification of ToAC and MDA Concentration
ToAC, which gives an indication of the oxygen-scavenging capacity of hydrosoluble enzymes and vitamins, was measured in skimmed milk. The Antioxidant Assay Kit (Sigma-Aldrich) was used according to the manufacturer's instructions on 1:2 diluted skimmed samples in triplicate.
The MDA concentrations were assessed using the HPLC method (32) and considered, as TBARS, as indicators of early lipid peroxidation. Fifty microliters of each unskimmed breast milk sample was used for the determinations. 0.21 to 5 μmol/L tetramethoxypropane in water was used to calculate the standard curve. Chromatographic separations were performed on a Dionex UltiMate 3000 HPLC system (Thermo Fisher Scientific Inc, Waltham, MA) equipped with a Shimatzu RS-10 fluorescence detector (Shimatzu, Kyoto, Japan) and a 150 × 4.6-mm Phenomenex Gemini C18 column (Phenomenex Inc, Torrance, CA) with 5-μm particles, equipped with a Phenomenex SecurityGuard Gemini C18 TWIN technologies guard column (4 × 3 mm).
Measurement of the Lipase Activities
The lipase activities were measured in duplicate in each unskimmed HM sample. Three lipase activity assays were performed for each sample: 1 with the addition of sodium cholate, to measure BSDL activity; 1 without sodium cholate, to measure total lipase activity (serum-independent lipase [SIL]); 1 without sodium cholate at 4°C, to measure the SIL activity in refrigerated conditions (refrigerated lipase [RL]). The method used was described by Iijima et al (33), with modifications. Five hundred microliters of substrate solution (25 mol/L Tris-HCl pH 9.0, 0.25 mmol/L 2-methoxyethanol, 0.53 mmol/L p-nitrophenyl myristate in dimethyl sulfoxide, plus 5 mmol/L sodium cholate for BSDL) was added to 2, 5, or 10 μL of diluted HM sample (1:10 for SIL and RL activities, 1:100 for BSDL activity), mixed, and left for 15 minutes to incubate at room temperature (or on ice for 4°C lipase activity). Then, 700 μL of stop solution (acetone/n-heptane 5:2) was added and immediately mixed by inversion. After centrifugation for 2 minutes at 6000g at 4°C, 800 μL of the lower aqueous phase was collected and their absorbance read at 405 nm. The blanks were prepared as samples, but the stop solution was added immediately after the substrate solution.
Data on lipid composition, FFAs, oxidative status, and lipase activities were analyzed with 1-way analysis of variance at significance below 0.05 to assess significant differences using the KyPlot 2.0 statistical software (Kyens Lab Inc, Tokyo, Japan). When the calculated values of F were significant, Tukey post-hoc analysis was used to interpret any significant differences between the mean values.
The temperature in the upper and lower part of the NICU refrigerator was monitored every 5 minutes, using 2 mini data loggers equipped with internal probes. The mean temperatures recorded for the 3 replicate pools of the experiment, with their standard deviations, are reported in Table 1.
Total Fatty Acid Analysis
Table 2 and Supplemental Table 1 (http://links.lww.com/MPG/A182) show the percentage of fatty acids in the HM samples at different times of refrigeration. Mean values and standard deviations were reported for all the fatty acids in Supplemental Table 1. In agreement with another study (34), some fatty acids showed considerable sample-to-sample variation, as indicated by high standard deviations with respect to their means.
No significant differences in fatty acid composition can be observed during prolonged refrigerated storage. In particular, the saturated/unsaturated fatty acid (SFA/UFA) ratios were constant, thus indicating that negligible oxidation occurred (Table 2).
The concentrations of FFAs are reported in Table 3 and Supplemental Table 2 (http://links.lww.com/MPG/A183). During prolonged refrigerated storage the concentration of FFAs in HM significantly increased with storage from 1.1% to 2.4% of total fat (data not shown). With increasing FFAs, a corresponding decrease in milk pH was observed.
Significant differences were found mainly for UFAs, whereas for SFA, the increase occurred to a lesser extent. To evaluate the trends in FFA release, the ratio between the SFA, monounsaturated fatty acids (MUFA), and PUFA concentrations at each conservation time and their starting concentration in milk at time 0 hour are reported in Figure 1. The percentages of free SFA/total SFA, free PUFA/total PUFA, and free MUFA/total MUFA increased from 1.4%, 1.1%, and 0.8% in milk at time 0 hour to 2.5%, 3.0%, and 1.9%, respectively, after 96 hours of refrigerated storage (data not shown).
The SFA/UFA ratio (Table 3) showed significant variation during HM conservation; however, its mean value was considerably greater than in total fatty acid composition (Table 2). Because free SFA/UFA ratio values are close to 1, approximately half of FFAs were saturated.
Oxidative Status of Breast Milk
The oxidative balance in breast milk as affected by refrigerated storage was characterized by determination of ToAC of skimmed milk, conjugated dienes, TBARS, and MDA concentration of whole breast milk. None of the oxidative parameters considered in the study varied with prolonged refrigeration of HM (Table 4).
No acetoxymethyl esters were found in the chromatograms of the hydrogenated samples. This means that the hydroperoxyacids, where present, were below the limit of detection (1 μg/L in milk).
Lipase Activities in Breast Milk During Refrigerated Storage
The activities of lipolytic enzymes in specific conditions were assessed by measuring the liberation of p-nitrophenol from p-nitrophenyl myristate (C14:0). The analyzed conditions were enzymatic activity at room temperature (25°C) (SIL); enzymatic activity on ice (4°C) (RL); enzymatic activity at room temperature in presence of sodium cholate bile salt (BSDL). As reported in Table 5, no significant difference was noticed in any lipolytic activity at any time of storage. The lipase RL and SIL activities were more than 20-fold lower than BSDL activity, which could mainly be ascribed to the activity of 1 specific lipase, that is, the HM BSDL. BSDL activity showed a decreasing trend, although not significant, during the course of refrigerated storage.
The use of refrigerated breast milk is a common practice for mothers at home and a common clinical practice in NICUs. The maximum refrigeration time for HM ranges between 24 hours and 8 days, according to present advice. Such variability reflects the heterogeneity of the scientific sources, which is attributable to differences in the study design and in the methodological approaches (7,10). Based on this observation, we designed an experiment aimed at generating extensive information on the effects of prolonged refrigeration of HM, by evaluating the changes in the lipid profile, in the lipid-associated enzymatic activities, and in the oxidative status every 24 hours, for up to 96 hours of conservation, in 3 replicated trials.
The continuous temperature monitoring of the refrigerator, during the experimental period in the 3 replicated trials, showed a mean value of 6.8°C ± 1.1°C, even though the refrigerator thermostat was set at 5°C. This observation is in accordance with the findings of a survey showing that the temperature measured using a thermometer at a given moment does not necessarily represent the true operating conditions of the refrigerator, which should be taken into account by both health workers and families, who use domestic refrigerators for milk storage (35).
Refrigerated storage for 96 hours did not affect the total fatty acid profile of fresh milk, confirming a previous report on the issue (14). The SFA/UFA ratio also remained constant, thus indicating a lack of degradation and/or oxidation processes during prolonged cold storage in a NICU refrigerator. The lipid asset of milk was preserved throughout the refrigeration period. It is interesting to note that medium-chain SFAs (C10–C14), considered of primary importance for infant nutrition, and especially for preterm newborns, as they are more easily absorbed than long-chain fatty acids (36–38), were also stable throughout the conservation period. Neither was the level of LC-PUFAs affected by prolonged refrigeration. This finding is particularly relevant for the nutrition of premature infants, who may experience reduced availability of arachidonic acid and docosahexanoic acid to a greater extent than term infants, as 80% of these PUFAs accumulate during the last 3 months of pregnancy (39), and therefore have higher LC-PUFA requirements.
The constant SFA/UFA ratio throughout the refrigeration period indicated a lack of degradation/oxidation. To back up these findings, analyses of the oxidative status of HM were made by measuring the ToAC, TBARS, MDA content, and the level of conjugated dienes. We found that preserving milk by refrigerated storage for up to 96 hours did not affect its oxidative status, as shown by the nonsignificant increase in the oxidation products and the stable level of the antioxidant capacity. Previous studies indicated that cold storage of HM may decrease its antioxidant capacity (40,41) and increase its MDA content (23). This discrepancy may be attributable to the different analytical methods used. In the present study, the values of early and final peroxidation products (TBARS, MDA, and conjugate dienes) throughout the experiment did not indicate any evidence of lipid peroxidation, consistently with the recent report by Michalski et al (25). This finding is particularly relevant for preterm infants, who have a reduced antioxidant capacity and are often exposed to oxidative stress caused by infection, oxygen, mechanical ventilation, intravenous nutrition, and blood transfusions (42). Many of the disorders of preterm infants are thought to be because of this imbalance between antioxidant capacity and oxidative stress (40).
Several previous studies have evidenced the occurrence of lipolysis during cold storage of HM (13,14,16). Our results confirmed their findings, with a 2.2-fold increase in the total FFA content in 96 hours being observed. The increase in FFAs during refrigeration is consistent with the significant decrease in pH. The significant decrease in the SFA/UFA ratio is because of the higher quantity of UFAs released during storage. Among free UFAs, the highest release was seen for LC-PUFAs, already observed by Lavine and Clark (13), and, in particular, for those in the n-6 series (almost 4-fold increased). The n-6 LC-PUFAs are known to be preferentially located in the outer position of HM triglycerides (43). Preferential lipolysis in primary positions in triglycerides is a common feature of some lipases, including HM lipoprotein lipase (44). Limited lipolysis in HM has been described as a means for feeding the infant with more available lipids, thus helping a newborn's immature digestive system (36). The secretion of endogenous lipases in HM, called dietary lipases, enforces this hypothesis (45).
This is the first report on the assessment of lipase activity in HM during prolonged refrigeration. Our findings about lipase activities in HM during prolonged refrigerated storage indicated that they were not affected by storage conditions, as stable activity was observed both in the absence and in the presence of bile cofactors, when samples were incubated at room temperature. To simulate the conditions for lipolysis in refrigerated milk, the enzymatic assay was conducted at 4°C, without the addition of bile salts. The basal activity in these conditions was approximately 40% lower than that the SIL activity at room temperature and was not affected by storage times. The addition of bile salts to HM at room temperature caused a >20-fold increase in lipase activity, regardless of the time of storage, because of specific induction of BSDL in HM, as already described (46). The overall stability of BSDL activity was consistent with the constant intensity of the corresponding band in protein electrophoresis, as described by Giribaldi et al (unpublished data); however, the slight decline in BSDL activity over storage, although not significant, warrants further investigation to evaluate its potential clinical relevance. As BSDL is thought to be inactive in refrigerated HM (22), the lipolytic activity found in HM at 4°C should be mainly ascribed to SSL (13,47). SSL was reported to be responsible for rancidity in both bovine (48) and HM (49). These findings demonstrate that infants who receive expressed milk that has been stored for up to 96 hours receive essentially the same amount of lipase activity as infants fed directly at the breast.
The results of the present study confirm that HM can be stored in controlled refrigerated conditions for as long as 96 hours, without its nutritive and nonnutritive components being compromised. The study provides a more detailed insight into lipolysis, and novel evidences on the absence of oxidation processes and on the preservation of lipase activities, thus supporting safe refrigeration of HM for up to 96 hours.
1. Agostoni C, Braegger C, Decsi T, et al. Breast-feeding: a commentary by the ESPGHAN Committee on Nutrition. J Pediatr Gastroenterol Nutr 2009; 49:112–125.
2. Agostoni C, Buonocore G, Carnielli VP, et al. Enteral nutrient supply for preterm infants: commentary from the European Society for Paediatric Gastroenterology, Hepatology, and Nutrition Committee on Nutrition. J Pediatr Gastroenterol Nutr 2010; 50:85–91.
3. Williamson MT, Murti PK. Effects of storage, time, temperature, and composition of containers on biologic components of human milk. J Hum Lact 1996; 12:31–35.
4. Lawrence RA. Storage of human milk and the influence of procedures on immunological components of human milk. Acta Paediatr Suppl 1999; 88:14–18.
5. Tully DB, Jones F, Tully MR. Donor milk: what's in it and what's not. J Hum Lact 2001; 17:152–155.
6. Heiman H, Schanler RJ. Benefits of maternal and donor human milk for premature infants. Early Hum Dev 2006; 82:781–787.
7. Arslanoglu S, Bertino E, Tonetto P, et al. Guidelines for the establishment and operation of a donor human milk bank. J Matern Fetal Neonatal Med 2010; 23:1–20.
8. Academy of Breastfeeding Medicine. ABM Clinical Protocol #8: human milk storage information for home use for full-term infants. Breastfeed Med 2010;3:127–30.
10. Davanzo R, Travan L, Demarini S. Storage of human milk: accepting certain uncertainties. J Hum Lact 2010; 26:233–234.
11. Carlson SE, Werkman SH, Peeples JM, et al. Arachidonic acid status correlates with first year growth in preterm infants. Proc Natl Acad Sci USA 1993; 90:1073–1077.
12. Neuringer M, Anderson GJ, Connor WE. The essentiality of n-3 fatty acids for the development and function of the retina and brain. Annu Rev Nutr 1988; 8:517–541.
13. Lavine M, Clark RM. Changing patterns of free fatty acids in breast milk during storage. J Pediatr Gastroenterol Nutr 1987; 6:769–774.
14. Romeu-Nadal M, Castellote AI, Lopez-Sabater MC. Effect of cold storage on vitamins C and E and fatty acids in human milk. Food Chem 2008; 106:65–70.
15. Tacken KJ, Vogelsang A, van Lingen RA, et al. Loss of triglycerides and carotenoids in human milk after processing. Arch Dis Child Fetal Neonatal Ed 2009; 94:F447–F450.
16. Slutzah M, Codipilly CN, Potak D, et al. Refrigerator storage of expressed human milk in the neonatal intensive care unit. J Pediatr 2010; 156:26–28.
17. Morera-Pons S, Castellote Bargalló AI, López-Sabater MC. Evaluation by high-performance liquid chromatography of the hydrolysis of human milk triacylglycerides during storage at low temperatures. J Chromatogr A 1998; 823:467–474.
18. Hernell O, Olivecrona T. Human milk lipases. I. Serum-stimulated lipase. J Lipid Res 1974; 15:367–374.
19. Hernell O, Olivecrona T. Human milk lipases. II. Bile salt-stimulated lipase. Biochim Biophys Acta 1974; 369:234–244.
20. Manson WG, Weaver LT. Fat digestion in the neonate. Arch Dis Child Fetal Neonatal Ed 1997; 76:F206–F211.
21. Hamosh M, Hamosh P. In: Sanderson IR, Walker WA, eds. Development of Digestive Enzyme Secretion. Toronto: BC Decker; 2000:261–78.
22. Clark RM, Hundrieser KE, Brown PB. The effect of temperature and length of storage on bile salt-stimulated lipase and esterase in human milk. Nutr Res 1984; 4:957–960.
23. Miranda M, Muriach M, Almansa I, et al. Oxidative status of human milk and its variations during cold storage. Biofactors 2004; 20:129–137.
24. Turoli D, Testolin G, Zanini R, et al. Determination of oxidative status in breast and formula milk. Acta Paediatr 2004; 93:1569–1574.
25. Michalski MC, Calzada C, Makino A, et al. Oxidation products of polyunsaturated fatty acids in infant formulas compared to human milk—a preliminary study. Mol Nutr Food Res 2008; 52:1478–1485.
26. Elisia I, Kitts DD. Quantification of hexanal as an index of lipid oxidation in human milk and association with antioxidant components. J Clin Biochem Nutr 2011; 49:147–152.
27. Morrison WR, Smith LM. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron fluoride-methanol. J Lipid Res 1964; 5:600–608.
28. Huerta-Gonzalez L, Wilbey RA. Determination of free fatty acids produced in filled-milk emulsions as a result of the lipolytic activity of lactic acid bacteria. Food Chem 2001; 72:301–307.
29. Lagarda MJ, Mañez JG, Manglano P, et al. Lipid hydroperoxides determination in milk-based infant formulae by gas chromatography. Eur J Lip Sci Technol 2003; 105:339–345.
30. Wanasundara UN, Shahidi F. Stabilization of canola oil with flavonoids. Food Chem 1994; 50:393–396.
31. King RL. Oxidation of milk fat globule membrane material. I. Thiobarbituric acid reaction as a measure of oxidized flavor in milk and model systems. J Dairy Sci 1962; 45:1165–1171.
32. Seljeskog E, Hervig T, Mansoor MA. A novel HPLC method for the measurement of thiobarbituric acid reactive substances (TBARS). A comparison with a commercially available kit. Clin Biochem 2006; 39:947–954.
33. Iijima N, Tanaka S, Ota Y. Purification and characterization of bile salt-activated lipase from the hepatopancreas of red sea bream, Pagrus major. Fish Physiol Biochem 1998; 18:59–69.
34. Yuhas R, Pramuk K, Lien EL. Human milk fatty acid composition from nine countries varies most in DHA. Lipids 2006; 41:851–858.
35. Laguerre O, Derens E, Palagos B. Study of domestic refrigerator temperature and analysis of factors affecting temperature: a French survey. Int J Refrig 2002; 25:653–659.
36. Bitman J, Wood L, Hamosh M, et al. Comparison of the lipid composition of breast milk from mothers of term and preterm infants. Am J Clin Nutr 1983; 38:300–312.
37. Gastaldi D, Bertino E, Monti G, et al. Donkey's milk detailed lipid composition. Front Biosci 2010; 2:537–546.
38. Moltó-Puigmartí C, Castellote AI, Carbonell-Estrany X, et al. Differences in fat content and fatty acid proportions among colostrum, transitional, and mature milk from women delivering very preterm, preterm, and term infants. Clin Nutr 2011; 30:116–123.
39. Fleith M, Clandinin MT. Dietary PUFA for preterm and term infants: review of clinical studies. Crit Rev Food Sci Nutr 2005; 45:205–229.
40. Hanna N, Ahmed K, Anwar M, et al. Effect of storage on breast milk antioxidant activity. Arch Dis Child Fetal Neonatal Ed 2004; 89:F518–F520.
41. Sari FN, Akdag A, Dizdar EA, et al. Antioxidant capacity of fresh and stored breast milk: is −80°C optimal temperature for freeze storage? J Matern Fetal Neonatal Med 2012; 25:777–782.
42. Thibeault DW. The precarious antioxidant defenses of the preterm infant. Am J Perinatol 2000; 17:167–181.
43. Martin JC, Bougnoux P, Antoine JM, et al. Triacylglycerol structure of human colostrum and mature milk. Lipids 1993; 28:637–643.
44. Wang CS, Kuksis A, Manganaro F. Studies on the substrate specificity of purified human milk lipoprotein lipase. Lipids 1982; 17:278–284.
45. Bernbäck S, Bläckberg L, Hernell O. The complete digestion of human milk triacylglycerol in vitro requires gastric lipase, pancreatic colipase-dependent lipase, and bile salt-stimulated lipase. J Clin Invest 1990; 85:1221–1226.
46. Wang CS. Human milk bile salt-activated lipase. Further characterization and kinetic studies. J Biol Chem 1981; 256:10198–10202.
47. Freed LM, Berkow SE, Hamosh P, et al. Lipases in human milk: effect of gestational age and length of lactation on enzyme activity. J Am Coll Nutr 1989; 8:143–150.
48. Deeth HC. Lipoprotein lipase and lipolysis in milk. Int Dairy J 2006; 16:555–562.
49. Castberg HB, Hernell O. Role of serum-stimulated lipase in lipolysis in human milk. Milchwissenschaft 1975; 30:721–724.