- Email: [email protected]
Effects of methods of collection and storage on nutrients in human milk
Early Human Development, Elsevier Biomedical Press
6 (1982) 295-303
295
Effects of methods of collection and storage on nutrients in human milk Cutberto
Garza a, Carmen A. Johnson a, Ronald Buford L. Nichols a
Harrist b and
a Department
of Pediatrics, Baylor College of Medicine, Houston, Texas, and ’ University of Texas School of Public Health, Houston, Texas, U.S.A. Accepted
for publication
21 January
1982
Summary The effects of collection technique, storage container, and the duration and temperature of storage on selected nutrient concentrations in mature human milk were evaluated. Milk samples were collected during the fourth week of lactation from women 20-35 years of age by hand expression or suction. Greater volumes and fat concentrations were observed in milks collected by suction. Vitamin A, zinc, iron, copper, sodium, and protein nitrogen concentrations were not affected by storage of milk in either Pyrex or polypropylene containers for up to 24 h. The storage temperature had a significant effect on protein nitrogen and ascorbic acid concentrations. These findings indicate that collection methods and storage procedures used for comparatively brief periods will affect the concentrations of selected nutrients of mature human milk. Specific recommendations are made for the collection and storage of milk. milk collection
and storage;
nutrient
concentration;
human
milk
Introduction The nutritional content and functional properties of human milk are unique. The high bioavailability of its nutrients, its optimal nutrient concentrations and pattern, and its immunological components have been cited as reasons for preferring human milk in the feeding of low birthweight and normal term infants. However, in contrast to recommendations regarding the feeding of term infants, the appropriateness of feeding human milk to low birthweight infants is not viewed uniformly [8,25]. Two potential problems are cited often: (1) dependence on banked human milk is of 0378-3782/82/0000-0000/$02.75
0 1982 Elsevier Biomedical
Press
296
concern because of an incomplete understanding of changes taking place between collection time and feeding time, and (2) the nutrient requirements of low birthweight infants, though mostly unknown, are believed to be greater than those of normal term infants [27]. These concerns are closely interrelated. The issue of the comparability of banked milk and milk obtained directly from the breast is important. The bioavailability and apparent quality of several key nutrients in human milk are greater than in available substitutes [20,28,31]. This quality allows infants fed human milk to achieve normal growth at nutrient intake levels that are below the intakes of infants fed synthetic formulas [7]. Importantly, however, the relatively low nutrient concentrations in human milk also make it necessary that collection and storage conditions which adversely alter the concentrations, quality, or bioavailability of nutrients in banked milk be minimized. Little attention has been given to defining these conditions experimentally. Similarly, little attention has been focused on the effects of such conditions on milk obtained for investigational purposes. Milk composition is effected by the degree of breast emptying [6] because of differences between ‘fore’ and ‘hind’ milk. Collection methods, therefore, that differ in their breast emptying efficiency may yield milks of varying composition. This is well recognized. Few studies, however, have quantitated differences under controlled conditions between methods expected to result in disparate values such as hand expression and intermittent gentle suction applied by an electrical pump [5]. Although minimal differences would be expected from donors trained to use both methods, significant differences may be anticipated when samples are collected from ‘single-time’ or infrequently used donors. Storage containers interact differently with selected milk components. The differential adherence of milk cells to ‘plastic’ and Pyrex containers is recognized [23], but the effects may be highly time-dependent [lo]. Apparent differences in the adherence of non-cellular water soluble immunologic factors to commonly used storage materials, i.e. Pyrex, polyethylene, and polypropylene also have been obof nutrient-storage container interactions served [ 10,111. No similar evaluations have been published. Yet, the adherence of lipids to synthetic materials used in storage materials has been reported [4]. The association of key trace elements with the cream-fraction of milk [ 16,241 coupled with the possibility of binding interactions between storage materials and lipid and water soluble components of milk is of concern to the banking of milk for either clinical or strictly investigational purposes. Evaluations of possible interactions between commonly used storage materials and key nutritive components of human milk are therefore indicated. Evaluations of the effects of storage temperature have also been of interest. Most of these, however, have focused on immunologic factors. The effects of storage temperature on nutritive components that are sensitive to oxidation and heat have not been examined. Such an evaluation would aid in identifying ideal conditions for minimizing potentially adverse changes and would serve to document the extent of deterioration of potentially labile components. Ascorbic acid [19] and unsaturated fatty acids [3] may be particularly labile and may serve as good markers for the extent of oxidation in stored milks.
297
This study was undertaken to quantitate the effects of specific collection techniques and storage conditions on the concentrations of representative human milk nutrients. Vitamin A, zinc, iron, copper, sodium and protein nitrogen concentrations were not affected by storage of milk in either Pyrex of polypropylene containers for up to 24 h. The storage temperature had a significant effect on protein nitrogen and particularly ascorbic acid concentrations. These findings suggest specific recommendations for collecting and storing human milk for clinical and investigative purposes. This evaluation is part of a larger study assessing the effects of processing conditions on nutritional, immunologic, and enzymatic components of human milk [lO,ll].
Materials and Methods Subjects Women who were 20-35 years of age and were breast-feeding exclusively their first or second born infant were recruited from a population utilizing private hospital services. Mothers and infants were in good health. The diets of donors contained a wide assortment of foods. Donors reported no efforts to exclude or restrict specific items from the diet other than coffee, tea, and alcohol. The donors limited intakes of these beverages to one serving or less per day. The subjects did not use tobacco. All subjects denied the intake of any drugs or medications during the week before donation. Approvals from Institutional Human Experimentation Committees and informed consent from all subjects were obtained. Experimental
design
General procedures Milk saml.Jes were obtained in the fourth week of lactation between 8:30 a.m. and 11:30 a.m. Unless otherwise specified, the Egnell pump (Cary, IL) was used because it was found to produce a reliably low intermittent pressure effective in stimulating the flow of milk. Samples were collected 1.5-3 h after the previous nursing. One breast was emptied at each collection. Two samples from each donor were obtained 2-3 days apart, each from the same breast. Samples from each donor were collected at approximately the same intervals from the previous nursing and at the same time of day. Samples were stored at 4°C unless otherwise indicated. All collections were supervised by trained milk bank staff. Experiment I In this experiment the effects of collection by hand expression or negative pressure on fat and total nitrogen (N) concentrations in milks were compared. Eighteen women donated two samples, one collected by each of the two methods. No donor had prior experience with either collection procedure. The first collection method was randomly determined. Polyethylene bags were used to store the milk.
298
Experiment II The effects on Cu, Zn, Na, Fe, vitamin A, and protein N concentrations were investigated after storage in Pyrex and polypropylene containers for 4 and 24 h. A milk sample was obtained from each donor at each of two separate visits. The type of storage container for the first sample was randomly determined, and the alternate type of container was used on the second visit. Nutrient analyses were performed on aliquots of each sample after 4 and 24 h of storage. A minimum of 11 paired samples was used in making specific nutrient comparisons. Experiment III The effects of storage temperature on ascorbic acid, iodine number, and protein N were assessed in this experiment. All samples were stored in polypropylene containers. Aliquots of samples were stored at 37°C 4°C and -72°C. Aliquots stored at -72°C were analyzed after 24 h storage. Aliquots stored at 37°C and 4°C were analyzed after 4, 24, and 48 h storage. A minimum of 11 paired samples was used in making specific comparisons. Chemical methods
Na, Zn, Cu, and Fe were analyzed using atomic absorption spectrophotometry. Samples were prepared by wet ashing [ 11. Milk fat was analyzed by the Roese-Gottlieb method [l], vitamin A by the Carr-Price method [2], and ascorbic acid by a modification of the method by Loefflen and Proting [l]. Nitrogen was determined by the Kjeldahl procedure [14]. Protein N was determined by first precipitating milk proteins with 24% trichloroacetic acid (1 : 1 by volume), centrifuging for 90 min at 11000 X g, and then decanting the supernatant [ 131. The precipitate was dissolved in 0.1 M NaOH and then digested by the Kjeldahl procedure. All aliquots for analysis were obtained while the milk was stirred. All materials used for collection and analyses were acid washed and rinsed in deionized water. Either a plastic or Pyrex pipette was used, depending upon the container in which the sample was stored. All analyses were done in duplicate. Statistical
methods
Data were analyzed by the two sample t-test, paired t-test, or analysis of variance [29]. A three-way analysis of variance (ANOVA), a mixed model for random (individual) and fixed effects (time and container), was used to evaluate the results. This analysis allowed the effects due to variation between subjects to be separated from the effects of container and time. In the presentation of results, the mean (K) and S.D. were calculated when population variability was being described. Standard errors of the mean (S.E.M.) were calculated when treatment effects were compared.
299
Results Experiment I Total nitrogen concentrations were shown not to differ in milk collected by hand expression and by electric pump, 210 f 20 mg/dl (X * SD) and 220 k 20 mg/dl, respectively. The mean concentration of total fat in milk obtained by electric pump was 25% higher than that of hand expressed milk, 3.7 * 0.5 g% (x 2 SEM), and 3.0 *OS g%, respectively (paired t = 1.48, df= 9, P
Experiment II No effects on the concentrations of Na, Fe, Zn, Cu, and vitamin A due to storage in either Pyrex or polypropylene containers or to duration of storage were detected.
Experiment III Effects of storage temperature on the concentration of protein N and ascorbic acid were found to be significant. The overall mean protein N concentration (see Table I) was 3% higher in milk stored at 37’C than in milk stored at 4’C. Effects of storage time were not found to be significant. Effects of the duration of storage and storage temperature on ascorbic acid concentration were significant (see Table II). No significant changes were observed in iodine number.
TABLE I Protein
N concentrations
Storage
time
(mg N/d]) Storage
temperature
W 4oc
3lOC 4 24 48
- 72°C
187f8 18327
* ****
18127 17825
* *,**
189*8
*
178*5
*
18628
**
Mean&l SM(n=ll). * Effects due to temperature significant when samples stored at 37 and 4°C are compared F=9.3 [l,lO]). ** Comparison of storage temperature effects on samples stored only for 24 h not significant F = 1.4 [2,33]).
(PcO.05, (P>O.O5,
300 TABLE Ascorbic Storage
II acid concentrations time
(pg/dl)
Storage
temperature
(h)
4 24 48
37YI
4°C
- 72°C
4077f560 * 14472363 *.** 13792309 *
4864~416 * 2911*352*~** 1741*386*
4197*375
**
Mean* 1 S.E.M. (n = 11). * Effects due to temperature (P
Discussion Anticipated differences in the breast emptying efficiencies of hand expression and intermittent suction did result in the collection of significantly greater milk volumes when the latter method was used, Surprisingly, however, despite a 90% difference in milk volumes, statistically significant differences in the fat concentrations were not observed. This finding however must be interpreted cautiously. While expected differences in milk composition resulting from disparate collection methods may be minimized, the absolute differences observed in this experiment in volumes and fat concentrations indicate that statistically significant differences may occur even when collections are done under carefully controlled conditions. Representative water and lipid soluble nutrients were chosen to evaluate possible interactions between these components and storage container materials in Expt. II. Past observations that significant quantities of iron and copper are associated with the cream portion of milk [16] and more recent findings of the apparent adherence of water- [lO,ll] and lipid-soluble [4] components to container materials suggested these comparisons. Results obtained in this study indicate that there is no significant adherence of either water- or lipid-soluble nutrients to the materials tested. These observations suggest that interactions observed in other experiments may be either highly specific, as in the case of immunologic factors [lo], or to be of minimum significance. The effects of storage temperature on specific nutrient concentrations were quantitated in Expt. III. Statistically significant increases in protein N concentrations were found in samples stored at 37°C. The small absolute difference in concentrations between samples stored at 37°C and other temperatures suggests that this observation may be of little biological importance unless the difference is limited to specific proteins of functional significance. For example, in an extension of these studies it has been found that lysozyme concentrations in mature milk were 33% higher in samples stored at 37°C than in those stored at colder temperatures
301
(unpublished results). Comparable differences also were observed in measurements of lactoferrin, but not in IgA (unpublished results). These observations suggest that effects of storage temperature on proteins may be highly specific. The effects of storage temperature and time on constituents that may be oxidized easily were evaluated also. The levels of ascorbic acid fell markedly in milk stored at 37’C and 3-4’C for 24 and 48 h. Values for samples stored for 4 h at the higher temperatures were comparable to those obtained in milk stored for 24 h in solid CO,. More prolonged storage resulted in marked decreases in ascorbic acid. The apparent transient maintenance of ascorbic acid levels suggest that it is temporarily protected by other factors which are more easily oxidized or that it is temporarily compartmentalized in such a way to protect it. The decrease in ascorbic acid levels with more prolonged storage also suggests the potential for the significant oxidation of other labile components. One of the few studies assessing the effects of processing on human milk nutrients includes studies of ascorbic acid. Legge and Richards [ 171 measured ascorbic acid in milk pools before and after heating to boiling (total heating time 8 min). These investigators found that ascorbic acid levels decreased by 40% after heating. These reported losses are comparable to those observed after storage for 24 h at 3-4°C. The deterioration of oxidizable substances also was monitored by measuring iodine number. Iodine number has been used commonly as a semiquantitative measure of the unsaturated fatty acid content of foods. Significant changes in iodine number were not observed with storage for up to 48 h at any temperature. Neither were any correlations observed between the concentration of unsaturated bonds as reflected by iodine number and changes in this value. It is of interest to contrast these results with those of Legge and Richards [17] who evaluated changes in the lipid composition of pooled human milk pre- and post-heating as described earlier. The mean decreases (ranges) in oleic, linoleic, and linolenic acid concentrations were 10% (4-17%) 18% (3354%) and 10% (14-24%). Our results and those of Legge and Richards suggest that more sensitive evaluations of lipid degradation are needed. In that regard, it may be necessary to obtain peroxide, thiobarbituric acid number, carbonyl type, and acid determinations. In summary, the effects of collection and storage procedures on selected nutrient components of human milk have been evaluated. The markedly different volumes of milk obtained by hand expression and by intermittent gentle suction support the concern that the method chosen for sample collection may affect the composition of milk that is obtained. Importantly, however, potential differences in composition may be minimized if collections are done under closely supervised conditions. Both Pyrex and polypropylene containers were found not to interact with selected waterand fat-soluble nutrients. However, practical problems limiting the usefulness and possible safety of polyethylene bags were encountered. The differential effects of Pyrex, polypropylene, and polyethylene containers on immunologic factors should be kept in mind when choosing a container. Especially important is the reported adherence of SIgA antibodies to polyethylene and the time dependent nature of cell adherence to Pyrex and polypropylene [lo]. Evaluations of storage temperatures suggest that milk ideally should either be used within 4 h of collection or that it
302
should be frozen immediately to minimize changes brought about by oxidation. Refrigeration at 3 to 4°C for 24 h, however, maintains ascorbic acid levels at 60% of the value observed at 4 h after collection. More prolonged storage at this temperature results in greater losses. Experiments are in progress assessing the effects of these temperatures on specific immunologic components.
Acknowledgments Publication of this work is supported by a contract from the National Institutes of Child Health and Human Development (DHEW No. lHD-8-2828) and by the USDA/SEA Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine and Texas Children’s Hospital. The technical assistance of J. Sachen, D. Hines, T. Phipps and U. Shankur, the nursing assistance of E. Cherry, J. Jones, D. Mitchell, P. Perez, J. Smith, and J. Mouton, the secretarial help from J. Christian, K. Williamson, and C. Daniels and the editorial assistance of E.R. Klein and M. Boyd were greatly appreciated.
References 1 Horowitz, W. (Ed.) (1975): In Methods of Analysis. Association of Official Analytical Chemists, Edn. 12, Washington, DC. 2 Association of Vitamin Chemists, Inc., Methods Committee. (1966): In: Methods of Vitamin Assay. Interscience Publishers, New York. 3 Berk, 2. (1976): The Biochemistry of Foods, p. 219, Elsevier/North Holland, Inc., New York. 4 Dill, C.W., Herlick, .%A., Richter, R.L. and Davis, J.W. (1979): Fat test depression during chilled storage of milk samples in plastic containers for analysis by the micro-tester. J. Food Protect., 42, 314-316. 5 Egnell, E. (1956): The mechanics of different methods of emptying the female breast. J. Swed. Med. Ass., 40, l-8. 6 Emery, W.B., III, Canolty, N.L., Aitchinson, J. and Dunkley, W.L. (1978): Influence of sampling on fatty acid composition of human milk. Am. J. Clin. Nutr., 31, 1127- 1130. 7 Fomon, S.J. (1974): Infant Nutrition. 2nd edition. W.B. Saunders Co., Philadelphia, p. 359. 8 Fomon, S.J., Ziegler, E. and Vasguy, H. (1977): Human milk and the small premature infant. Am. J. Dis. Child., 13 1, 463-467. 9 Ford, J.E., Law, B.A., Marshall, V.M.E. and Reiter, B. (1977): Influence of. the heat treatment on human milk and some of its protective constituents. J. Pediatr., 90, 29-35. 10 Goldblum, R.R.M., Garza, C., Johnson, C.A., Nichols, B.L., Harrist, R. and Goldman, A.S. (1982): Effects of container upon immunologic factors in mature milk. Nutr. Res., in press. 11 Goldblum, R.R.M., Garza, C., Johnson, CA., Nichols, B.L., Harrist, R. and Goldman, A.S. (1981): Effects of containers upon immunologic factors on colostrum. Acta Pediatr. Stand., in press. 12 Goldman, A.S. and Smith, C.W. (1973): Host resistance factors in human milk. J. Pediatr., 82, 1082- 1090. 13 Hambraeus, L., Lonnerdal, B., Forsum, E. and Gebre-Medhim (1978): Nitrogen and protein components of human milk. Acta Pediatr. Stand., 67, 561-565. 14 Henry, R.J., Cannon, D.C. and Winkelman, J.E. (1974): Clinical Chemistry: Principles and Techniques. Harper and Row, New York. 15 Jones, C.L., Jennison, R.F. and D’Suza, S.W. (1979): Bacterial contamination of expressed breast milk. Br. Med. J., 2, 1320-1322.
303
16 King, R.L., Luick, J.R., Litman, 1.1. and Dunkley, W.L. (1959): Distribution of natural and added copper and iron in milk. J. Dairy Sci., 42, 780-790. 17 Legge, M. and Richards, K. (1978): Biochemical alterations in human breast milk after heating. Aust. Paediatr. J., 14, 87-90. 18 Liebhaber, M., Lewiston, N.J., Asquith, M.T. and Sunshine, P. (1978): Comparison of bacterial contamination with two methods of human milk collection. J. Pediatr., 92, 236-237. 19 Lund, D.B. (1979): Effect of commercial processing on nutrients. Food Technol., 33 (2), 28-34. 20 McMillan, J.A., Landaw, S.A. and Oski, F.A. (1976): Iron sufficiency in breast-fed infants and the availability of iron from human milk. Pediatrics, 58, 686-691. 21 Nichols, B.L. and Nichols, V.N. (1979): Lactation. Adv. Pediatr., 26, 137-161. 22 Nutrients in Processed Foods: Vitamins, Minerals. (1974): American Medical Association. Publishing Sciences Group, Inc., Acton, MA. 23 Paxson, CL., Jr. and Cress, C.C. (1979): Survival of human milk leukocytes. J. Pediatr., 94, 61-64. 24 Picciano, M.F. and Guthrie, H.A. (1976): Copper, iron and zinc contents of mature human milk. Am. J. Clin. Nutr., 29, 242-254. 25 Raiha, N.C.R., Heionen, K., Rassen, D.K. and Gaull, G.E. (1976): Milk protein quantity and quality in low-birthweight infants. I. Metabolic responses and effects on growth. Pediatrics, 57, 659-674. 26 Raptopoulou-Gigi, M., Marwick, E. and McClelland, D.B.L. (1977): Antimicrobial proteins in sterilized human milk. Br. Med. J., 1, 12-14. 27 Shaw, J.C.L. (1973): Parenteral nutrition in the management of sick low-birthweight infants. Pediatr. Clin. North Am., 20, 333-358. 28 Singer, E., Murphy, G.M., Elkins, S. and Anderson (1974): Role of bile salts in fat malabsorption of premature infants. Arch. Dis. Child., 49, 1744180. 29 Snedecor, G.W. and Cochran, W.G. (1967): In: Statistical Methods, edn. 6, The Iowa State University Press, Ames, IA. 30 Welch, J.K. and May, J.T. (1979): Anti-infective properties of breast milk. J. Pediatr., 94, 1-9. 31 Widdowson, E.M. (1965): Absorption and excretion of fat nitrogen, and minerals from filled milks by babies one week old. Lancet, 2, 1099- 1105.
6 (1982) 295-303
295
Effects of methods of collection and storage on nutrients in human milk Cutberto
Garza a, Carmen A. Johnson a, Ronald Buford L. Nichols a
Harrist b and
a Department
of Pediatrics, Baylor College of Medicine, Houston, Texas, and ’ University of Texas School of Public Health, Houston, Texas, U.S.A. Accepted
for publication
21 January
1982
Summary The effects of collection technique, storage container, and the duration and temperature of storage on selected nutrient concentrations in mature human milk were evaluated. Milk samples were collected during the fourth week of lactation from women 20-35 years of age by hand expression or suction. Greater volumes and fat concentrations were observed in milks collected by suction. Vitamin A, zinc, iron, copper, sodium, and protein nitrogen concentrations were not affected by storage of milk in either Pyrex or polypropylene containers for up to 24 h. The storage temperature had a significant effect on protein nitrogen and ascorbic acid concentrations. These findings indicate that collection methods and storage procedures used for comparatively brief periods will affect the concentrations of selected nutrients of mature human milk. Specific recommendations are made for the collection and storage of milk. milk collection
and storage;
nutrient
concentration;
human
milk
Introduction The nutritional content and functional properties of human milk are unique. The high bioavailability of its nutrients, its optimal nutrient concentrations and pattern, and its immunological components have been cited as reasons for preferring human milk in the feeding of low birthweight and normal term infants. However, in contrast to recommendations regarding the feeding of term infants, the appropriateness of feeding human milk to low birthweight infants is not viewed uniformly [8,25]. Two potential problems are cited often: (1) dependence on banked human milk is of 0378-3782/82/0000-0000/$02.75
0 1982 Elsevier Biomedical
Press
296
concern because of an incomplete understanding of changes taking place between collection time and feeding time, and (2) the nutrient requirements of low birthweight infants, though mostly unknown, are believed to be greater than those of normal term infants [27]. These concerns are closely interrelated. The issue of the comparability of banked milk and milk obtained directly from the breast is important. The bioavailability and apparent quality of several key nutrients in human milk are greater than in available substitutes [20,28,31]. This quality allows infants fed human milk to achieve normal growth at nutrient intake levels that are below the intakes of infants fed synthetic formulas [7]. Importantly, however, the relatively low nutrient concentrations in human milk also make it necessary that collection and storage conditions which adversely alter the concentrations, quality, or bioavailability of nutrients in banked milk be minimized. Little attention has been given to defining these conditions experimentally. Similarly, little attention has been focused on the effects of such conditions on milk obtained for investigational purposes. Milk composition is effected by the degree of breast emptying [6] because of differences between ‘fore’ and ‘hind’ milk. Collection methods, therefore, that differ in their breast emptying efficiency may yield milks of varying composition. This is well recognized. Few studies, however, have quantitated differences under controlled conditions between methods expected to result in disparate values such as hand expression and intermittent gentle suction applied by an electrical pump [5]. Although minimal differences would be expected from donors trained to use both methods, significant differences may be anticipated when samples are collected from ‘single-time’ or infrequently used donors. Storage containers interact differently with selected milk components. The differential adherence of milk cells to ‘plastic’ and Pyrex containers is recognized [23], but the effects may be highly time-dependent [lo]. Apparent differences in the adherence of non-cellular water soluble immunologic factors to commonly used storage materials, i.e. Pyrex, polyethylene, and polypropylene also have been obof nutrient-storage container interactions served [ 10,111. No similar evaluations have been published. Yet, the adherence of lipids to synthetic materials used in storage materials has been reported [4]. The association of key trace elements with the cream-fraction of milk [ 16,241 coupled with the possibility of binding interactions between storage materials and lipid and water soluble components of milk is of concern to the banking of milk for either clinical or strictly investigational purposes. Evaluations of possible interactions between commonly used storage materials and key nutritive components of human milk are therefore indicated. Evaluations of the effects of storage temperature have also been of interest. Most of these, however, have focused on immunologic factors. The effects of storage temperature on nutritive components that are sensitive to oxidation and heat have not been examined. Such an evaluation would aid in identifying ideal conditions for minimizing potentially adverse changes and would serve to document the extent of deterioration of potentially labile components. Ascorbic acid [19] and unsaturated fatty acids [3] may be particularly labile and may serve as good markers for the extent of oxidation in stored milks.
297
This study was undertaken to quantitate the effects of specific collection techniques and storage conditions on the concentrations of representative human milk nutrients. Vitamin A, zinc, iron, copper, sodium and protein nitrogen concentrations were not affected by storage of milk in either Pyrex of polypropylene containers for up to 24 h. The storage temperature had a significant effect on protein nitrogen and particularly ascorbic acid concentrations. These findings suggest specific recommendations for collecting and storing human milk for clinical and investigative purposes. This evaluation is part of a larger study assessing the effects of processing conditions on nutritional, immunologic, and enzymatic components of human milk [lO,ll].
Materials and Methods Subjects Women who were 20-35 years of age and were breast-feeding exclusively their first or second born infant were recruited from a population utilizing private hospital services. Mothers and infants were in good health. The diets of donors contained a wide assortment of foods. Donors reported no efforts to exclude or restrict specific items from the diet other than coffee, tea, and alcohol. The donors limited intakes of these beverages to one serving or less per day. The subjects did not use tobacco. All subjects denied the intake of any drugs or medications during the week before donation. Approvals from Institutional Human Experimentation Committees and informed consent from all subjects were obtained. Experimental
design
General procedures Milk saml.Jes were obtained in the fourth week of lactation between 8:30 a.m. and 11:30 a.m. Unless otherwise specified, the Egnell pump (Cary, IL) was used because it was found to produce a reliably low intermittent pressure effective in stimulating the flow of milk. Samples were collected 1.5-3 h after the previous nursing. One breast was emptied at each collection. Two samples from each donor were obtained 2-3 days apart, each from the same breast. Samples from each donor were collected at approximately the same intervals from the previous nursing and at the same time of day. Samples were stored at 4°C unless otherwise indicated. All collections were supervised by trained milk bank staff. Experiment I In this experiment the effects of collection by hand expression or negative pressure on fat and total nitrogen (N) concentrations in milks were compared. Eighteen women donated two samples, one collected by each of the two methods. No donor had prior experience with either collection procedure. The first collection method was randomly determined. Polyethylene bags were used to store the milk.
298
Experiment II The effects on Cu, Zn, Na, Fe, vitamin A, and protein N concentrations were investigated after storage in Pyrex and polypropylene containers for 4 and 24 h. A milk sample was obtained from each donor at each of two separate visits. The type of storage container for the first sample was randomly determined, and the alternate type of container was used on the second visit. Nutrient analyses were performed on aliquots of each sample after 4 and 24 h of storage. A minimum of 11 paired samples was used in making specific nutrient comparisons. Experiment III The effects of storage temperature on ascorbic acid, iodine number, and protein N were assessed in this experiment. All samples were stored in polypropylene containers. Aliquots of samples were stored at 37°C 4°C and -72°C. Aliquots stored at -72°C were analyzed after 24 h storage. Aliquots stored at 37°C and 4°C were analyzed after 4, 24, and 48 h storage. A minimum of 11 paired samples was used in making specific comparisons. Chemical methods
Na, Zn, Cu, and Fe were analyzed using atomic absorption spectrophotometry. Samples were prepared by wet ashing [ 11. Milk fat was analyzed by the Roese-Gottlieb method [l], vitamin A by the Carr-Price method [2], and ascorbic acid by a modification of the method by Loefflen and Proting [l]. Nitrogen was determined by the Kjeldahl procedure [14]. Protein N was determined by first precipitating milk proteins with 24% trichloroacetic acid (1 : 1 by volume), centrifuging for 90 min at 11000 X g, and then decanting the supernatant [ 131. The precipitate was dissolved in 0.1 M NaOH and then digested by the Kjeldahl procedure. All aliquots for analysis were obtained while the milk was stirred. All materials used for collection and analyses were acid washed and rinsed in deionized water. Either a plastic or Pyrex pipette was used, depending upon the container in which the sample was stored. All analyses were done in duplicate. Statistical
methods
Data were analyzed by the two sample t-test, paired t-test, or analysis of variance [29]. A three-way analysis of variance (ANOVA), a mixed model for random (individual) and fixed effects (time and container), was used to evaluate the results. This analysis allowed the effects due to variation between subjects to be separated from the effects of container and time. In the presentation of results, the mean (K) and S.D. were calculated when population variability was being described. Standard errors of the mean (S.E.M.) were calculated when treatment effects were compared.
299
Results Experiment I Total nitrogen concentrations were shown not to differ in milk collected by hand expression and by electric pump, 210 f 20 mg/dl (X * SD) and 220 k 20 mg/dl, respectively. The mean concentration of total fat in milk obtained by electric pump was 25% higher than that of hand expressed milk, 3.7 * 0.5 g% (x 2 SEM), and 3.0 *OS g%, respectively (paired t = 1.48, df= 9, P
Experiment II No effects on the concentrations of Na, Fe, Zn, Cu, and vitamin A due to storage in either Pyrex or polypropylene containers or to duration of storage were detected.
Experiment III Effects of storage temperature on the concentration of protein N and ascorbic acid were found to be significant. The overall mean protein N concentration (see Table I) was 3% higher in milk stored at 37’C than in milk stored at 4’C. Effects of storage time were not found to be significant. Effects of the duration of storage and storage temperature on ascorbic acid concentration were significant (see Table II). No significant changes were observed in iodine number.
TABLE I Protein
N concentrations
Storage
time
(mg N/d]) Storage
temperature
W 4oc
3lOC 4 24 48
- 72°C
187f8 18327
* ****
18127 17825
* *,**
189*8
*
178*5
*
18628
**
Mean&l SM(n=ll). * Effects due to temperature significant when samples stored at 37 and 4°C are compared F=9.3 [l,lO]). ** Comparison of storage temperature effects on samples stored only for 24 h not significant F = 1.4 [2,33]).
(PcO.05, (P>O.O5,
300 TABLE Ascorbic Storage
II acid concentrations time
(pg/dl)
Storage
temperature
(h)
4 24 48
37YI
4°C
- 72°C
4077f560 * 14472363 *.** 13792309 *
4864~416 * 2911*352*~** 1741*386*
4197*375
**
Mean* 1 S.E.M. (n = 11). * Effects due to temperature (P
Discussion Anticipated differences in the breast emptying efficiencies of hand expression and intermittent suction did result in the collection of significantly greater milk volumes when the latter method was used, Surprisingly, however, despite a 90% difference in milk volumes, statistically significant differences in the fat concentrations were not observed. This finding however must be interpreted cautiously. While expected differences in milk composition resulting from disparate collection methods may be minimized, the absolute differences observed in this experiment in volumes and fat concentrations indicate that statistically significant differences may occur even when collections are done under carefully controlled conditions. Representative water and lipid soluble nutrients were chosen to evaluate possible interactions between these components and storage container materials in Expt. II. Past observations that significant quantities of iron and copper are associated with the cream portion of milk [16] and more recent findings of the apparent adherence of water- [lO,ll] and lipid-soluble [4] components to container materials suggested these comparisons. Results obtained in this study indicate that there is no significant adherence of either water- or lipid-soluble nutrients to the materials tested. These observations suggest that interactions observed in other experiments may be either highly specific, as in the case of immunologic factors [lo], or to be of minimum significance. The effects of storage temperature on specific nutrient concentrations were quantitated in Expt. III. Statistically significant increases in protein N concentrations were found in samples stored at 37°C. The small absolute difference in concentrations between samples stored at 37°C and other temperatures suggests that this observation may be of little biological importance unless the difference is limited to specific proteins of functional significance. For example, in an extension of these studies it has been found that lysozyme concentrations in mature milk were 33% higher in samples stored at 37°C than in those stored at colder temperatures
301
(unpublished results). Comparable differences also were observed in measurements of lactoferrin, but not in IgA (unpublished results). These observations suggest that effects of storage temperature on proteins may be highly specific. The effects of storage temperature and time on constituents that may be oxidized easily were evaluated also. The levels of ascorbic acid fell markedly in milk stored at 37’C and 3-4’C for 24 and 48 h. Values for samples stored for 4 h at the higher temperatures were comparable to those obtained in milk stored for 24 h in solid CO,. More prolonged storage resulted in marked decreases in ascorbic acid. The apparent transient maintenance of ascorbic acid levels suggest that it is temporarily protected by other factors which are more easily oxidized or that it is temporarily compartmentalized in such a way to protect it. The decrease in ascorbic acid levels with more prolonged storage also suggests the potential for the significant oxidation of other labile components. One of the few studies assessing the effects of processing on human milk nutrients includes studies of ascorbic acid. Legge and Richards [ 171 measured ascorbic acid in milk pools before and after heating to boiling (total heating time 8 min). These investigators found that ascorbic acid levels decreased by 40% after heating. These reported losses are comparable to those observed after storage for 24 h at 3-4°C. The deterioration of oxidizable substances also was monitored by measuring iodine number. Iodine number has been used commonly as a semiquantitative measure of the unsaturated fatty acid content of foods. Significant changes in iodine number were not observed with storage for up to 48 h at any temperature. Neither were any correlations observed between the concentration of unsaturated bonds as reflected by iodine number and changes in this value. It is of interest to contrast these results with those of Legge and Richards [17] who evaluated changes in the lipid composition of pooled human milk pre- and post-heating as described earlier. The mean decreases (ranges) in oleic, linoleic, and linolenic acid concentrations were 10% (4-17%) 18% (3354%) and 10% (14-24%). Our results and those of Legge and Richards suggest that more sensitive evaluations of lipid degradation are needed. In that regard, it may be necessary to obtain peroxide, thiobarbituric acid number, carbonyl type, and acid determinations. In summary, the effects of collection and storage procedures on selected nutrient components of human milk have been evaluated. The markedly different volumes of milk obtained by hand expression and by intermittent gentle suction support the concern that the method chosen for sample collection may affect the composition of milk that is obtained. Importantly, however, potential differences in composition may be minimized if collections are done under closely supervised conditions. Both Pyrex and polypropylene containers were found not to interact with selected waterand fat-soluble nutrients. However, practical problems limiting the usefulness and possible safety of polyethylene bags were encountered. The differential effects of Pyrex, polypropylene, and polyethylene containers on immunologic factors should be kept in mind when choosing a container. Especially important is the reported adherence of SIgA antibodies to polyethylene and the time dependent nature of cell adherence to Pyrex and polypropylene [lo]. Evaluations of storage temperatures suggest that milk ideally should either be used within 4 h of collection or that it
302
should be frozen immediately to minimize changes brought about by oxidation. Refrigeration at 3 to 4°C for 24 h, however, maintains ascorbic acid levels at 60% of the value observed at 4 h after collection. More prolonged storage at this temperature results in greater losses. Experiments are in progress assessing the effects of these temperatures on specific immunologic components.
Acknowledgments Publication of this work is supported by a contract from the National Institutes of Child Health and Human Development (DHEW No. lHD-8-2828) and by the USDA/SEA Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine and Texas Children’s Hospital. The technical assistance of J. Sachen, D. Hines, T. Phipps and U. Shankur, the nursing assistance of E. Cherry, J. Jones, D. Mitchell, P. Perez, J. Smith, and J. Mouton, the secretarial help from J. Christian, K. Williamson, and C. Daniels and the editorial assistance of E.R. Klein and M. Boyd were greatly appreciated.
References 1 Horowitz, W. (Ed.) (1975): In Methods of Analysis. Association of Official Analytical Chemists, Edn. 12, Washington, DC. 2 Association of Vitamin Chemists, Inc., Methods Committee. (1966): In: Methods of Vitamin Assay. Interscience Publishers, New York. 3 Berk, 2. (1976): The Biochemistry of Foods, p. 219, Elsevier/North Holland, Inc., New York. 4 Dill, C.W., Herlick, .%A., Richter, R.L. and Davis, J.W. (1979): Fat test depression during chilled storage of milk samples in plastic containers for analysis by the micro-tester. J. Food Protect., 42, 314-316. 5 Egnell, E. (1956): The mechanics of different methods of emptying the female breast. J. Swed. Med. Ass., 40, l-8. 6 Emery, W.B., III, Canolty, N.L., Aitchinson, J. and Dunkley, W.L. (1978): Influence of sampling on fatty acid composition of human milk. Am. J. Clin. Nutr., 31, 1127- 1130. 7 Fomon, S.J. (1974): Infant Nutrition. 2nd edition. W.B. Saunders Co., Philadelphia, p. 359. 8 Fomon, S.J., Ziegler, E. and Vasguy, H. (1977): Human milk and the small premature infant. Am. J. Dis. Child., 13 1, 463-467. 9 Ford, J.E., Law, B.A., Marshall, V.M.E. and Reiter, B. (1977): Influence of. the heat treatment on human milk and some of its protective constituents. J. Pediatr., 90, 29-35. 10 Goldblum, R.R.M., Garza, C., Johnson, C.A., Nichols, B.L., Harrist, R. and Goldman, A.S. (1982): Effects of container upon immunologic factors in mature milk. Nutr. Res., in press. 11 Goldblum, R.R.M., Garza, C., Johnson, CA., Nichols, B.L., Harrist, R. and Goldman, A.S. (1981): Effects of containers upon immunologic factors on colostrum. Acta Pediatr. Stand., in press. 12 Goldman, A.S. and Smith, C.W. (1973): Host resistance factors in human milk. J. Pediatr., 82, 1082- 1090. 13 Hambraeus, L., Lonnerdal, B., Forsum, E. and Gebre-Medhim (1978): Nitrogen and protein components of human milk. Acta Pediatr. Stand., 67, 561-565. 14 Henry, R.J., Cannon, D.C. and Winkelman, J.E. (1974): Clinical Chemistry: Principles and Techniques. Harper and Row, New York. 15 Jones, C.L., Jennison, R.F. and D’Suza, S.W. (1979): Bacterial contamination of expressed breast milk. Br. Med. J., 2, 1320-1322.
303
16 King, R.L., Luick, J.R., Litman, 1.1. and Dunkley, W.L. (1959): Distribution of natural and added copper and iron in milk. J. Dairy Sci., 42, 780-790. 17 Legge, M. and Richards, K. (1978): Biochemical alterations in human breast milk after heating. Aust. Paediatr. J., 14, 87-90. 18 Liebhaber, M., Lewiston, N.J., Asquith, M.T. and Sunshine, P. (1978): Comparison of bacterial contamination with two methods of human milk collection. J. Pediatr., 92, 236-237. 19 Lund, D.B. (1979): Effect of commercial processing on nutrients. Food Technol., 33 (2), 28-34. 20 McMillan, J.A., Landaw, S.A. and Oski, F.A. (1976): Iron sufficiency in breast-fed infants and the availability of iron from human milk. Pediatrics, 58, 686-691. 21 Nichols, B.L. and Nichols, V.N. (1979): Lactation. Adv. Pediatr., 26, 137-161. 22 Nutrients in Processed Foods: Vitamins, Minerals. (1974): American Medical Association. Publishing Sciences Group, Inc., Acton, MA. 23 Paxson, CL., Jr. and Cress, C.C. (1979): Survival of human milk leukocytes. J. Pediatr., 94, 61-64. 24 Picciano, M.F. and Guthrie, H.A. (1976): Copper, iron and zinc contents of mature human milk. Am. J. Clin. Nutr., 29, 242-254. 25 Raiha, N.C.R., Heionen, K., Rassen, D.K. and Gaull, G.E. (1976): Milk protein quantity and quality in low-birthweight infants. I. Metabolic responses and effects on growth. Pediatrics, 57, 659-674. 26 Raptopoulou-Gigi, M., Marwick, E. and McClelland, D.B.L. (1977): Antimicrobial proteins in sterilized human milk. Br. Med. J., 1, 12-14. 27 Shaw, J.C.L. (1973): Parenteral nutrition in the management of sick low-birthweight infants. Pediatr. Clin. North Am., 20, 333-358. 28 Singer, E., Murphy, G.M., Elkins, S. and Anderson (1974): Role of bile salts in fat malabsorption of premature infants. Arch. Dis. Child., 49, 1744180. 29 Snedecor, G.W. and Cochran, W.G. (1967): In: Statistical Methods, edn. 6, The Iowa State University Press, Ames, IA. 30 Welch, J.K. and May, J.T. (1979): Anti-infective properties of breast milk. J. Pediatr., 94, 1-9. 31 Widdowson, E.M. (1965): Absorption and excretion of fat nitrogen, and minerals from filled milks by babies one week old. Lancet, 2, 1099- 1105.