UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effect of ultrasonication, lyophilization, freezing and storage on lipids and immune components of human… Dhar, Jyoti 1989

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1989_A6_7 D42.pdf [ 6.77MB ]
Metadata
JSON: 831-1.0097455.json
JSON-LD: 831-1.0097455-ld.json
RDF/XML (Pretty): 831-1.0097455-rdf.xml
RDF/JSON: 831-1.0097455-rdf.json
Turtle: 831-1.0097455-turtle.txt
N-Triples: 831-1.0097455-rdf-ntriples.txt
Original Record: 831-1.0097455-source.json
Full Text
831-1.0097455-fulltext.txt
Citation
831-1.0097455.ris

Full Text

Effect of Ultrasonication, Lyophilization, Freezing and Storage on Lipids and Immune Components of Human Milk by Jyoti Dhar B.Sc, G.B. Pant University of Agriculture and Technology, India, 1981. M.Sc, Panjab University, India, 1983. A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF ' THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Division of Human Nutrition School of Family and Nutritional Sciences We accept this thesis as conforming to the required standard THE UNIVERSITY OF'BRITISH COLUMBIA January, 1989 © Jyoti Dhar, 1989 / In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date DE-6 (2/88) A b s t r a c t Administration of expressed human milk (EHM) to ill and premature infants ne-cessitates tube feeding the babies using mechanical pumps or gravity flow. Tube feeding however, leads to separation and adhesion of milk fat to the tube walls and syringes causing incomplete delivery of fat to the baby. 'Ultrasonic Homogeniza-tion' dramatically prevents this fat separation and observed fat loss during tube delivery of milk. However, would the effect of ultrasonication persist after storage? Also, since ultrasonication leads to disruption of large molecules in milk, would the unique nutritional and immune properties of milk be subsequently comprised? There is so far no documentation on the above aspects. This study was designed to determine the effects of ultrasonic homogenization and storage on the following constituents of human milk; 1) free fatty acid (FFA) and peroxide levels, 2) fatty acid profile (FAP), 3) immunoglobulin A (IgA) levels and 4) recovery of fat and protein after infusion. The study was divided into four parts, In part 1, seven EHM samples wer obtained. Each sample was divided into three parts (Pasteurized, Ul-trasonicated, and Ultrasonicated after Pasteurization) and stored in both frozen and lyophilized forms for one and four months. Analysis of Variance (ANOVA) and further Scheffe's tests were performed on the FFA and peroxide data obtained both before and after storage. The levels of FFA rose signficantly (P < 0.01) in the non-pasteurized, ultrasonicated samples and the increase was higher with a longer storage time and lyophilized form of storage. Peroxide levels signficantly increased after storage (P < 0.001), however, the level of increase in the non-pasteurized ul-trasonicated group was significantly lower (P < 0.05) compared to the pasteurized groups with or without ultrasonication. In part 2 of the study; five EHM samples were analyzed for FAP both before and after storage in fresh frozen, ultrasoni-cated frozen and ultrasonicated lyophilized forms for 1 month. Results of Multiple Analysis of Variance (MANOVA) showed a significant decrease in the saturated fatty acids C12:0 and C18:0 (P < 0.001), and Long Chain Poly-Unsaturated Fatty Acids (LC-PUFA) C20:4n6 and C22:4n6 (P < 0.001) with ultrasonication and storage. The levels of C20:5n3 signficantly increased (P < 0.05), and the levels of essential fatty acids Cl8:2n6 and Cl8:3n3 remained unaffected with processing and storage. In part 3 of the study; seven EHM samples were analyzed for IgA in fresh, ultrasonicated and subsequently stored fractions (for 1 month in both frozen and lyophilized forms). Results of paired t tests indicated slight but signficant losses of IgA with ultrasonication (P < 0.05) and subsequent storage (P < 0.01). Losses were higher with lyophilized storage, however, the absolute amount of loss was still small (6.4%) compared to 33% loss reported with pasteurization. In part 4 of the study, eight EHM samples were obtained. Each sample was pasteurized, ultrasonicated and stored in both frozen and lyophilized forms for one and four ii months. Slow and rapid infusion of milk using mechanical pump and gravity flow was conducted after storage. Fat and protein recovery in the infusates was stud-ied. Milk stored frozen after ultrasonic homogenization resulted in negligible and insignificant loss of both fat and protein upon infusion. However, recovery of fat in the lyophilized milk was signficantly affected (P < 0.01). Lyophilization somehow appeared to reverse the effects of ultrasonication, resulting in fat globules with larger and varied sizes. On the other hand, ultrasonication of lyophilized milk just before infusion, resulted in much more stable delivery of both fat and protein. In conclusion, although ultrasonication and subsequent storage of milk results in slight loss of IgA and some LC-PUFA, these losses don't appear important when considering the possibility of massive loss of fat (and therefore energy) upon infusion of non-ultrasonicated milk. Moreover, the increased FFA levels after ul-trasonication and storage could have beneficial implications for the newborn. Also, the antioxidant effects of ultrasonication on peroxide levels of pasteurized milk appears important. iii C o n t e n t s A B S T R A C T ii LIST OF TABLES vii LIST OF FIGURES viii A C K N O W L E D G E M E N T S x 1 Introduction 1 2 Literature Review 6 2.1 Clinical Uses of Expressed Banked Human Milk (EHM) 6 2.2 Human Milk Banking System 8 2.2.1 History 8 2.2.2 Common Milk Bank Procedures 9 2.2.3 Processing and Storage of Expressed Human Milk 10 2.3 Effects of Processing on Lipids of Human Milk 12 2.3.1 Lipolysis in Processed and Stored E H M 13 2.3.2 Effect of Processing and Storage on the Polyunsaturated Fatty Acid Profile of E H M 16 2.3.3 Oxidative Instability in Processed and Stored E H M 18 2.4 Processing Effects on the Immune Components of Human Milk . . . 19 2.5 Fat and Protein Concentrations in Human Milk When Delivered Through the Infusion Pumps 22 3 Experimental Design and Methodology 25 iv 3.1 Experimental Design 25 3.1.1 Part I: 25 3.1.2 Part II: 27 3.1.3 Part III: v . 27 3.1.4 Part IV(a): 30 3.1.5 Part IV(b): 30 3.2 Experimental Procedures 30 3.2.1 Milk Collection and Standardization 30 3.2.2 Ultrasonic Treatment of Expressed Human Milk 33 3.2.3 Pasteurization 33 3.2.4 Infusion System For Tube Delivery of EHM 33 3.2.5 Lyophilization of Expressed Human Milk 34 3.3 Assay Procedures 35 3.3.1 Free Fatty Acid Analysis . 35 3.3.2 Determination of Peroxides 35 3.3.3 Immunoglobulin A Analysis . . 37 3.3.4 Fatty Acid Profile Analysis 37 3.3.5 Total Lipids and Protein Determination 41 4 Results 45 4.1 Part I: Free Fatty Acids and Peroxides 45 4.2 Part II: Fatty Acid Profile 53 4.3 Part III: Immunoglobulin A (IgA) 57 4.4 Part IV: a) Total Fat and Protein Recovery 57 4.5 Part IV: b) Total Fat and Protein Recovery 63 5 Discussion 68 5.1 Effect of Processing and Storage on the Free Fatty Acid and Peroxide Levels in Human Milk 68 5.1.1 Part la: Free Fatty Acids (FFA) 68 5.1.2 Part lb: Peroxide Levels . 70 v 5.2 Part II: Fatty Acid Profile in Processed and Stored Human Milk . . 72 5.3 Part III: Immunoglobulin A (IgA) Concentration in Process and Stored Human Milk 74 5.4 Part IV and Part IVa: Processing and Storage Effects on the Fat and Protein Contents of Human Milk Infused Mechanically and by Gravity Flow 76 6 Summary of Findings 80 7 Suggestions For Further Research 82 APPENDICES 83 REFERENCES 96 vi L i s t o f T a b l e s 4.1 Effect of ultrasonication, pasteurization, lyophilization and freezing on the free fatty acid content of fresh and stored human milk (group A, B, C) 47 4.2 Three way ANOVA on the increase in FFA and peroxides (F ratios). 48 4.3 Comparison of the increase in FFA and peroxides in the three treat-ment groups (A, B, C) after storage 48 4.4 Effect of ultrasonication, lyophilization, pasteurization and freezing on the peroxide content of fresh and stored human milk (group A, B, C) 52 4.5 Multiple analysis of variance on the fatty acids profile 56 4.6 Effect of ultrasonication, lyophilization, freezing and storage on IgA concentration of human milk 58 vii L i s t o f F i g u r e s 1.1 Changes in fat recovery during infusion of untreated and ultrasonic treated human milk and homogenized cow's milk 3 3.1 Experimental design: part 1 26 3.2 Experimental design: part II 28 3.3 Experimental design: part III 29 3.4 Experimental design: part IV(a) 31 3.5 Experimental design: part IV(b) 32 3.6 Free fatty acid analysis 36 3.7 Determination of peroxides 38 3.8 Calculation of peroxide value 39 3.9 Immunoglobulin A analysis 40 3.10 Fatty acids profile analysis 42 3.11 Total lipids and protein determination 43 4.1 Effect of ultrasonication, lyophilization and freezing on the FFA con-tent of fresh and stored human milk (mean of 7 samples). 46 4.2 Effect of ultrasonication, lyophilization and freezing on the peroxide content of fresh and stored human milk (mean of 7 samples) 51 4.3 Fatty acid profile in fresh, processed and stored (l month) human milk (mean ± SD of 5 samples) 54 4.4 Long chain polyunsaturated fatty acid profile in fresh, processed and stored (1 month) human milk (mean ± SD of 5 samples) 55 4.5 Mean (± SD) periodical and cumulative changes in fat content of ultrasonicated-frozen and ultrasonicated-lyophilized milk infused at a slow rate (10 ml/h) 59 viii 4.6 Mean (± SD) periodical and cumulative changes in fat content of ultrasonicated-frozen and ultrasonicated-lyophilized milk infused at a rapid rate (40 ml/h) 61 4.7 Mean (± SD) periodical and cumulative changes in protein content of ultrasonicated-frozen and ultrasonicated-lyophilized milk infused at a slow rate (10 ml/h) 62 4.8 Mean (± SD) periodical and cumulative changes in protein content of ultrasonicated-frozen and ultrasonicated-lyophilized milk infused at a rapid rate (40 ml/h) 64 4.9 Mean (± SD) changes in fat and protein content of milk ultrason-icated after lyophilization and storage and both before and after lyophilization and storage and infused at a slow rate using mechan-ical pump 65 4.10 Mean (± SD) changes in fat and protein content of milk ultrason-icated after lyophilization and storage and both before and after lyophilization and storage and infused at a rapid rate using gravity flow technique 66 5.1 Fat globules in fresh and fresh-lyophilized human milk 78 5.2 Fat globules in human milk ultrasonicated both before and after lyophilization 79 ix A c k n o w l e d g e m e n t s I wish to thank the following people for their contribution to this project. Dr. Susan Barr, Dr. George Davidson, Dr. Indrajit Desai, and Dr. Shuryo Nakai for their excellent guidance as research advisors and for their constant encourage-ment and support throughout the course of this project. Dr. Sheila Innis for giving her time and advice regarding the fatty acid profile analysis. Mr. Ping Mah for his assistance with the statistical analysis of the data. Mrs. Agi Radcliffe, Heather and Lisa for their expert technical help and good humoured companionship. My parents and my friends for their immeasurable advice and support through-out the project. This study was supported by Grant No. 5-56282-3220 to Dr. George Davidson from the British Columbia Medical Services Foundation, Canada. The financial support of the University Graduate Fellowship (1986) and the University Teaching Assistantship (1987-88) are also gratefully acknowledged. x To my parents, Gouri and Motilal Dhar. C h a p t e r 1 I n t r o d u c t i o n Human milk is the best source of nutrition for infants (Roy 1979). Several publi-cations have emphasized the uniqueness of its biochemical composition and partic-ularly of its immunochemical and cellular components (Roy 1979, Nutrition Com-mittee of Canadian Pediatric Society 1978). The promotion of breast feeding as an optimal foundation for infant nutrition is supported by the.Nutrition Committee of the Canadian Pediatric Society (1978) and the American Academy of Pediatrics (1979). A wide range of benefits have been associated with feeding human milk to hu-man babies. Most of these advantages of breast feeding also apply to feeding banked expressed human milk (EHM) which has been collected, stored and dispensed in a proper way. Properly banked EHM has been demonstrated to contain many of the protective components of fresh breast milk and is frequently better tolerated by the infant than is formula (Liebhaber 1983). Many centres that provide care for ill infants have established milk banks to handle and distribute donor EHM (Canadian Pediatric Society 1985), which is widely used to serve the needs of the premature, small for gestational age and normal term infants who may be ill and cannot suckle (Poskitt 1983). Administration of EHM to these infants is usually done by continuous nasogas-tric infusion, either by gravity flow or mechanical pulp. Recent research from this centre (Martinez et ai.1984) casts doubt on whether the infants actually receive the nutrients they were assumed to be getting during tube feeding. It has been known for some time that human milk fat separates and sticks to the sides of the gavage tube and syringes during tube feeding (Brooke and Barley 1978, Spencer and Hull 1981, Stocks 1985). Cumulative losses of about 47.4% of fat have been found when expressed human milk was infused at a slow rate and 16.8% when infused through a fast flowing open syringe (Martinez et al. 1987). This occurs in EHM on standing due to the varying and large sizes of fat globules (Ruegg and Blanc 1981). This 1 loss of fat is of real concern with possible repercussions on growth and development of infants (Fomon et al. 1977). There is concern not only about the loss of fat, but also about the significant variation in the fat content of different aliquots of untreated EHM collected throughout the infusion period (Figure l.l). The same study showed that a progressive reduction in fat content occurs during the first three and a half hours of infusion followed by a sharp rise in the last 30 minutes. The alteration in the fat content may also have unrecognized deleterious effects due to the varying protein/calorie balance of the untreated EHM, since the protein and lactose content of milk remain relatively constant during the entire infusion period (Martinez et al. 1987). In addition, many important nutrients usually present in breast milk are also found in significant amounts in the cream phase of breast milk. For example, the fat soluble vitamins, iron (31%), copper (15%), zinc (12%), calcium (10%), and magnesium (24%), are found in the cream phase (Fransson and Lonnerdal 1984). Martinez et al. (1987) suggest that the separation of fat and fat associated nutrients can be prevented by small scale ultrasonic homogenization of EHM. As can be seen in Figure 1.1, the recovery of fat in the ultrasonic treated EHM infusates was above 94%, which was similar to the recovery of fat seen in homogenized cow's milk and infant formula. The application of ultrasonic homogenization to the routine processing of EHM could be a very important advance in the use of EHM for infant feeding. How-ever, before the clinical effects of using ultrasonically homogenized EHM can be assessed, several factors need to be studied. For example, we do not know whether ultrasonicated milk stored for various lengths of time also maintains a high nutrient recovery after infusion. Since expressed breast milk is often stored for later usage it becomes important to study the effects of type, length and conditions of storage on the recovery of nutrients from milk when infused. Processing and storage of human milk has been associated with loss of certain nutritional (Williams et al. 1981, sunshine et al. 1980) and antimicrobial factors (Poskitt 1983, Frier et al. 1984). Evidence that pasteurization and other meth-ods of preservation not only eliminate the bacterial contaminants but also destroy many beneficial components of milk, including the immunoglobulins, lysozyme and lactobacillus bifidus factor (Evans et al. 1978, Ford et al. 1977, Gibbs et al. 1977), has led some to suggest that heating of donor milk samples is unnecessary and in-advisable (American Academy of Paediatrics 1980, Bjorksten et al. 1980, Sunshine 1980). However, reports of the transmission of salmonellosis (Ryder et al. 1977), streptococcal diseases (Kenny and Zedd 1977, Shriener et al. 1977), hepatitis B sur-face antigen, Cytomegalovirus (Nutrition Committee of Canadian Pediatric Society 1985) and HIV (Arnold 1988), through ingestion of infected breast milk necessitate a careful examination and maintenance of quality in milk banking procedures. 2 % 6 Oi 2 g f-< 4 H z IU o 2 o HUMAN MILK-UNTREATED HUMAN MILK-ULTHASONIC TflEATED IMF AHT FORMULA COWS MILK-HOMOGENIZEO ' p < 0 . 0 1 .J.-.J.,.I...I-.I..x.J lV~-i—I—i i *——if — : — i — i — i — T ' > - r r r r . /I 100 50 CUUULA TIVE RECOVERY (»> ( a ) INFUSION TIME (h) . onanges in the lat content and final recovery o l lat as a (unct ion o( l ime during the infusion of untrealed (long-da shed l ine) and u l t rason i c t reated ( sho r t -dashed l ine) human milk, infant formula (dot-dashed line), and homoge-n ized cow's milk (dotted line) using mechan ica l pump. The milk samples were infused cont inuous ly al a How rate of 10 ml/h for a per iod of 4 h. Values are expressed as mean * SD of ten trials, t t . differences from zero time concentrat ion statist ical ly s ignif icant (p < 0.01); "*. differences in cumula-tive recovery between untreated human milk and ultrasoni-cal ly treated human milk or homogen ized milk or infant lor-mula statistically signif icant (p < 0.01). XI cn 2 o f-< h-2 LU O 2 O o t -< HUMAN MILK-UNTREATED HUMAN MILK-ULTRASONIC TREATED INFANT FORMULA COW'S MILK-HOMOOENI2EO D ( 0 . 0 5 T-.4 / o / / / i 100 50 12 18 24 30 CUMULATIVE RECOVERY (X) (b) INFUSION TIME (min) Changes in the fat content and final recovery of lat as a lunct ion of t ime dur ing the in lus ion of untreated ( long-da shed l ine) and u l t r a son i c - t r ea t ed ( sho r t - da shed l ine) human milk, infant formula (dot-dashed line), and homoge -nized c o w s milk (dotted line) us ing gravity How techn ique. The milk samples were infused cont inuous ly at a rate o l 40 ml/h lor a per iod of 30 min. Va lues are expressed as mean r SO of eight trials, t . d i f ferences Irom zero time concen t ra -tion statist ical ly s igni f icant (p < 0.05); ", d i f ferences in c u m u -lative recovery between untreated human milk and ul t rasoni-cally treated human milk or homogen i zed c o w s mi lk or in-lant formula stat ist ical ly s igni f icant (p < 0.05). FIGURE 1.1: Changes in fat recovery during infusion of untreated and ultrasonic treated human and homogenized cow's milk. (From Martinez et al. 1987) in Ultrasonic treatment has been shown to be lethal for a variety of organisms (Davies 1959, Hughes and Nyborg 1962). However, all organisms do not show the same sensitivity (Alliger 1975). In the human milk banks, pasteurization is fre-quently used to obtain non-sporeforming pathogen free milk. However, any method adding to the killing effect of heat without being detrimental to the milk quality would be highly beneficial. Ultrasonic treatment has been shown to significantly enhance the bactericidal efficiency of heat treatment (Juan et al. 1987). However, whether ultrasonication further increases the losses of important nutritional and immune components in milk needs to be investigated. Immunoglobulin A (IgA) is one of the most important antimicrobial agents in breast milk providing instant passive immunity to the child against a wide array of enteric and respiratory bacterial and viral pathogens (Garza et al. 1987). It is therefore important to study if the process of ultrasonication would adversely affect the immunoglobulin A content o human milk. This would also provide an indication of whether ultrasonication allows the preservation of immune bodies in breast milk. The stability of milk lipids during processing and storage is another area of significant concern for human milk banks (Berkow et al. 1984, Hamosh et al. 1984). One factor that may alter the quality of stored milk is hydrolytic rancidity. This is due to the action of lipases (present in breast milk) on the milk lipids leading to release of fatty acids which in turn lead to undesirable flavour and aroma changes (Clark et al. 1984). There have only been a few studies which have examined the effect of storage time and temperature on the lipolytic activity in human milk: these report a progressive increase in lipolysis and an elevation in the level of unesterified fatty acids in quick frozen samples stored for longer durations (Hamosh et al. 1984, Clark et al. 1984, Friend et al. 1983). As storage of human milk becomes more common, the problem of increased lipolysis and possible rancidity in human milk will become more important. Since there are no published reports of the effect of ultrasonication on the lipolytic activity in milk, and since ultrasonication might increase the exposure of finely divided fat globules to the lipases in breast milk, it becomes important to study these effects. Frozen storage is generally used by milk banks to preserve expressed human milk for later use (Asquith et al. 1987). These milk banks frequently handle very large volumes of milk; for example, 32,400 ounces by Mother's Milk Bank California, in the year 1983 (Asquith and Lee 1985) and 32,000 ounces by Breast Milk Service at the Children's Hospital, Vancouver in the same year. These volumes have been increasing over the years: in 1988, the volume of EHM handled by the Children's Hospital Breast Milk Service, Vancouver has been 66,000 ounces. Handling of these large volumes of milk however, becomes an important problem due to the 4 space required for storing frozen milk and the inconvenience of transporting large volumes of frozen milk. Storage of milk in the freeze dried (lyophilized) form may provide an alternative to frozen storage. Since freeze dried foods are much lighter in weight and occupy less space than equivalent amounts of fresh or frozen foods, and can be shipped (transported) without refrigeration (Campbell et al. 1979), freeze drying would make handling of large volumes of milk easier. Other than one study which reports no significant loss of most immunological components post lyophilization in human milk (Evans et al. 1978), there is little information regarding effects of lyophiliza-tion on EHM. More information on this aspect would help in the evaluation of the suitability of lyophilization as a means of storage. In view of these considerations, the present study was designed to determine the effects of ultrasonication, pasteurization, lyophilization, freezing and storage on selected lipid and immune components of EHM and on the suitability of processed milk for tube feeding. The following hypotheses were proposed and tested in this study: 1. Ultrasonic homogenization will not affect the extent of lipolysis and peroxida-tion of fat in pasteurized EHM, stored frozen and/or lyophilized, for various lengths of time. 2. The extent of lipolysis in non-pasteurized and ultrasonicated EHM will be higher than in pasteurized ultrasonicated milk, both before and after storage. 3. The fatty acid profile of EHM will not be affected significantly by ultrasonic homogenization and subsequent storage in both frozen and lyophilized forms. 4. The concentration of Immunoglobulin A in EHM will not be lowered signifi-cantly by ultrasonic homogenization and storage in both frozen and lyophilized forms. 5. Fat and protein recovery in tube delivered ultrasonicated and pasteurized EHM will not be significantly affected by freezing, lyophilization and storage time. 5 C h a p t e r 2 L i t e r a t u r e R e v i e w Human milk is a substance with many unique nutritional properties (Poskitt 1983), and its use has been advocated for a wide variety of clinical conditions (Asquith 1987). However, because processing, storage and delivery of human milk may all affect its properties, it is important to characterize the effects of these processes and to attempt to develop techniques that will minimize deleterious changes while ensuring safety. This literature review will begin with a discussion of the clinical importance of expressed human milk and the practical aspects of human milk banking system. The focus of the review will be on what is and is not known about the effects of processing (i.e. pasteurization, ultrasonication, freezing, lyophilization) and storage on the stability of specific lipid constituents and on the retention fo specific immune components in expressed breast milk. Also reviewed is the important issue of energy and nutrient loss observed during delivery of expressed breast milk using mechanical infusion pumps. Finally, since processing might affect the concentration of several constituents in human milk, the possible effects of processing and storage on the recovery of fat and protein in breast milk when delivered through infusion pumps will be reviewed. 2.1 Clinical Uses of Expressed Banked Human Milk (EHM) Expressed banked human milk has been advocated for the treatment of numerous conditions in patients ranging in age from very premature neonates to young adults. Although in some cases, evidence of the efficacy of human milk as therapeutic agent appears anecdotal, there is nevertheless a considerable body of more substantial clinical evidence for its use in certain instances. Perhaps the most common condi-6 tion for which banked human milk is prescribed are those directly linked to infant nutrition (Pitt 1979, Pittard et al. 1985), such as allergy or intolerance to other formulae (Maclean and Fink 1980, Kretchner 1981) and failure to thrive (Maclean and Fink 1980). It has also been advocated for the alimentation of newborns pos-sessing various birth defects (Nayman et al. 1979), such as transient-neonatal ty-rosinemia (Goldman et al. 1974) and acrodermatitis enteropathica (Branski 1980). Patients recovering from surgical procedures for gastroschisis, intestinal obstruc-tion with bowel fistulae, and renal insufficiency have also benefitted from banked milk (Asquith et al. 1987). The protective value of human milk in providing immune cells as well as humoral factors is now well known, especially in relation to the neonate. However, human milk may also serve an important role in the treatment of allergic and immunologic disorders in the older child (Garza et al. 1987). Studies have shown that in the office practice of pediatrics a third of all visits are generated by a question of allergies (Liebhaber 1983). In case of strong allergies (usually genetic in origin) elimination of unnecessary exposure to known allergens becomes important. Human milk/colostral antibodies provide the immature gut of the infant with immunity, and perhaps protection, against sensitization through the binding of IgA antibody to foreign protein (Walker 1980). Other than providing passive protection, human milk is also non- allergenic and generally no antibody response occurs following its ingestion (Jelliffe and Jelliffe 1979). Banked breast milk has been fed to patients with known or suspected allergy to bovine milk or commercial milk substitutes (Davies et al. 1978, Jakobsson et al. 1978, Kwock et al. 1984). Since human milk prevents antigens from crossing the intestinal barrier in both physiologic and pathologic states (Walker 1975), banked breast milk has been pre-scribed for a wide range of infectious diseases such as chronic intractable diarrhea (Narayanan et al. 1982), gastroenteritis (Ste. Marie et al. 1974) and ulcerative colitis (Pittard et al. 1985), where it has been claimed to have produced dramatic results. Another possible therapeutic indication for the use of human milk has been in the treatment of enteric pathogens. Single donor, unpasteurized banked milk has been utilized in the treatment of infants with both serologic and secretory-IgA deficiency, who are chronically infected with intestinal pathogens such as Giardia lamblia, Rotavirus (Liebhaber 1983, Gillin et al. 1985) and Clostridium botulinum (Narayanan et al. 1981). It has also been prescribed to support infants with sepsis and pneumonia (Narayanan et al. 1981). Interest in the use of fresh banked human milk as a possible prophylactic agent 7 against necrotizing enterocolitis is also emerging as of particular current interest (Asquith et al. ,1987, Ste. Marie et al. 1974). Because human milk has been shown to contain numerous nutrients, biologically active substances and antimicrobial factors including lysozymes, host resistance factors, leukocytes, IgA, zinc-copper and iron-binding proteins, and, epidermal growth factor (McClelland et al. 1978, Lonnerdal et al. 1982), it can be presumed that banked human milk will be used by an increasing number of clinicians for the treatment of an ever-widening range of conditions. However, a fundamental problem to its universal acceptance by the scientific community remains the relative lack of objective evidence for many of the uses listed above. This situation will only be corrected by firstly a careful delineation of the properties of banked milk (single or multiple donor), and the effects of different types of processing on these properties and secondly, careful clinical controlled trials of the use of this product. 2.2 Human Milk Banking System 2.2.1 History The origin of a breast milk "service" dates back to Rome during the fourth century when breast feeding mothers would gather at the market place and offer to breast feed other infants. Modern breast milk banking dates back to the late 1920's, al-though these early banks were mostly on an informal basis without accurate record keeping. Interest in human milk feeding and banking decreased during World War II, (possibly due to socio-political reasons) and after the introduction of cow's milk based formulae capable of sustaining normal infant growth and development (American Academy of Pediatrics 1980). As a result, human milk banks stopped operating in North America. However, the milk bank tradition was never aban-doned in certain parts of Europe, especially in British hospitals. In Finland, as in Scandinavia generally, proprietary formulas never gained widespread acceptance in feeding premature infants (Siimes et al. 1979). For example at the Helsinki Children's Hospital, the milk banking service has been in continuous operation for over 50 years, starting in the 1930's (Siimes et al. 1979). Since the early 1970's there has been an increasing survival of smaller prema-tures and an increased incidence of necrotizing enterocolitis (Egan 1981). The existing cow's milk based formulae did not meet the needs of the prematures and therefore more adequate and better tolerated feeds were needed. Since research emphasized the nutritional immunological uniqueness of breast milk, human milk has come back into favour in many intensive care nurseries and consequently there 8 has been renewed interest in breast milk banking (American Academy of Pediatrics 1980). In Canada, the breast milk banks started during the mid-1970's. The Children's Hospital at Vancouver has a breast milk service, which includes a 'bank' function which has been in operation since September 1974. It is the largest human milk bank in Canada, in terms of the amount of milk processed and banked. Since 1982, 302,000 ounces of human breast milk have been processed in this centre. The amount has been increasing from 21,000 ounces in 1982 to 60,000 ounces in 1988. A figure showing the amount of milk processed by the Children's Hospital Milk Bank has been included in Appendix #3. At present, there are about 24 breast milk banks in Canada. A list of the names and addresses of these is presented in Appendix #1. In view of the unique, beneficial and irreplaceable properties of human milk, the HUMAN MILK BANKING ASSOCIATION OF NORTH AMERICA was es-tablished in 1985. The purposes of the association are: 1. To provide a forum for networking among experts in the field on issues related to human milk banking; 2. To provide information to the medical community on the benefits and appro-priate uses of banked human milk; 3. To develop guidelines for milk banking practices in North America, and 4. To encourage researchers interested in the unique properties of human milk. 2.2.2 Common Mi lk Bank Procedures A great variation exists in the current milk banking practices across Canada and overseas. There is as yet no consensus regarding what must be the standard milk banking protocol. However, there are certain procedures which are common to most human milk banks. These are as follows: 1. Donor Recruitment Breast milk donors are contacted through several routes, the major ones usually being prenatal classes and obstetrical units, some milk banks ob-tain donors through postnatal settings, community health units or with the assistance of the LaLeche league. Many banks have posters and brochures available and only a few use the news media to contact donors (Nutrition Committee of the Canadian Pediatric Society 1985). In most milk banks in 9 Canada donors are volunteers. However, in the United States, donors are paid in a few milk banks (Mcintosh 1985, Roy 1979). 2. Donor Criteria To help ensure that milk banked for clinical purposes is of the highest attain-able quality, donors are usually screened for a history of serum or infectious hepatitis, tuberculosis or any systemic disorder depending on the prevalence of these diseases in that area (Asquith 1987, Mcintosh 1985). AIDS testing of milk donors has been initiated by B.C. Children's Hospital breast milk service. It has been screening donors for AIDS since November 1985. An increasing number of other centres are now following suit. In general, donors must not be smoking or taking medications including birth control pills (Nu-trition Committee of the Canadian Pediatric Society 1985). 3. Donor Training and Collection of Expressed Human Milk In most milk banks, donors are given instruction in general cleanliness, clean expression of milk into sterile containers provided by the milk bank, labelling of containers and freezing of milk (Asquith 1987, Radcliffe 1982, Mcintosh 1985). The interest of donors in maintaining the bacteriologic quality of the milk expressed is usually fostered by frequent contacts and feedback by the bank staff. In some centres advice on diet, alcohol and caffeine consumption is also given by the milk bank coordinator. The donated milk is usually collected either by a volunteer driver or by the milk bank coordinator and delivered to the milk bank (Radcliffe 1982, Mcintosh 1985, Asquith 1987). 2.2.3 Processing and Storage of Expressed Human Mi lk In most milk banks, once the milk arrives, it is thawed at 4° C in the refrigerator (usually overnight) and then aseptically pooled. Due to the dangers of microbio-logical contamination pooled milk is usually pasteurized before use. There are no firm criteria for the degree of bacterial contamination that can be .considered safe in milk fed to an ill or premature infant or in unpasteurized mother's milk that is to be fed to her baby (Nutrition Committee of the Canadian Pediatric Society 1985). Although the microbiologic standards adopted by human milk banks to date, are empiric and unproven, milk that is free of pathogens and has a total bacterial colony count of less than 10 7 /£ at the time of sampling has been considered safe to use (Sauve et al. 1984). Pools containing more than that limit are usually thawed and pasteurized (Asquith 1987, Mcintosh 1985). At most centres, weekly pasteur-ization is carried out, mostly using a safeguard home pasteurizer which heats the 10 milk at 62.5° C for 30 minutes and is then rapidly cooled (Nutrition Committee of the Canadian Pediatric Society 1985). In the Children's Hospital at Vancouver, the milk is pooled for each 4000 ml batch. It is placed in 40, 100 ml bottles and pasteurized in an Oxford human milk pasteurizer for 30 minutes at 63° C followed by a rapid cooling cycle (Radcliffe 1982). Mi lk that is to be stored for longer periods is mostly kept in a standard deep freeze at -20° C with flasks placed on their sides to accommodate efficient storage (Radcliffe 1982). In most cases, 96.9% of pasteurized samples have indi-cated no growth of microorganisms such as staphylococcus species, bacillus species, E coli and klebsiella (Mcintosh 1985, Radcliffe 1982). Treatment, either by pro-longed freezing at -20° C or by pasteurization has been also shown to inactivate cytomegalovirus (CMV) (Friis and Anderson 1982, Stagno et al. 1980). On the other hand, certain nutrients and certainly many host resistance factors are altered to some extent when human milk has to be processed in order to prevent contamination with microorganisms (Sunshine et al. 1980). Pasteurization by holder technique (62.5° C for 30 minutes) has had signifi-cant adverse effects on the protective immunochemical constituents of human milk (Sunshine 1980). It also appears to reduce the absorption of fat mainly due to the inactivation of naturally occurring lipases present in milk (Sunshine 1980). The storage of milk by freezing is an alternative to heating for the preservation of optimal nutritional value and immunologic benefits (Committee on Nutrition, American Academy of Pediatrics 1980). The use of frozen storage however requires more attention to bacteriologic screening. Efforts are required towards collection of clean milk with minimal bacterial contamination and to store it immediately in a freezer until it is gently thawed and fed (Committee on Nutrition, American Academy of Pediatrics 1980). Long term storage of frozen milk usually tends to become inconvenient as it requires large storage space. Since milk banks handle large volumes of milk every year, provision of enough storage space for the incoming milk becomes a problem. Similarly transportation of frozen milk also is not very convenient. Freeze drying (lyophilization) of collected breast milk has been recently started in the human milk bank of the Children's Hospital in Lindenhof, Berlin (Bottcher 1987). In freeze drying, the food (in this case, breast milk), is frozen and exposed to a vacuum, so that ice from the food is removed as vapour without becoming a liquid. So little heat is applied that the food does not reach a high temperature (Campbell et al. 1979). Freeze dried foods are much lighter in weight than equivalent amounts of fresh or frozen food and can be stored and transported without refrigeration (Campbell et al. 1979). Although a few investigators have studied the effects 11 of lyophilization on the level of immunoglobulin (IgA, IgG, IgM) in human milk (Evans 1978, Liebhaber 1977) there still is no information about its effects on nutritional factors. More knowledge regarding the above stated factors is required in order to assess whether lyophilization of human milk is safe nutritionally and immunologically. While it would be of great benefit to be able to use freshly expressed human milk without having to process it for prematurely born and sick neonates especially those with gastrointestinal abnormalities, there are major risks to this approach. On the other hand, feeding trials have shown that effective use of expressed human milk in infant feeding does require proper processing, storage and administration in order to maintain its unique nutritional properties (Brooke and Barley 1978, Greer et al. 1984, Stocks 1985). As mentioned previously, separation of fat in the milk during tube feeding results in significant energy and nutrient losses and variation in the composition of milk delivered (Brooke and Barley 1978, Greer et al. 1984, Stocks et al. 1985, Martinez et al. 1987). In order to eliminate the loss of nutrients during administration, the process of homogenization by ultrasonic treatment has been found to be beneficial (Martinez et al. 1987). Small scale ultrasonic treatment has been shown to be a rapid and efficient means of preventing fat separation in human milk, this treatment is aimed at homogenizing human milk so that the size of fat globules would be similar to that of commercially homogenized cow's milk. The homogenization time is a function of the volume of milk and intensity of the ultrasonic vibration. The rise in temperature can be avoided, if necessary by using an ice bath (Martinez et al. 1987). Ultrasonic homogenization has not as yet been used for routine processing of expressed human milk in milk banks. The effects of ultrasonication on several enzymological, immunochemical and microbiological properties of human milk need to be studied before it can be used for routine processing of expressed human milk. In the following sections, specific issues regarding the effects of processing (i.e. pasteurization, lyophilization, freez-ing, ultrasonication) on the various nutritional and immunological components of human milk will be addressed. A brief presentation on the effects of processing on the administration of human milk by tube feedings will also be made. 2.3 Effects of Processing on Lipids of Human Milk The nutrition committee of the Canadian Pediatric Society and the American Academy of Pediatrics recommend human milk as the sole source of food for a healthy infant during the first 4-6 months of life. During this period of rapid body 12 growth, an infant accumulates approximately 1.4 - 1.7 kg of fat (Fomon 1967). This lipid accumulation serves not only for energy storage and thermic insulation but also has a structural function in all tissues. The composition of the infant's endogenous lipids in plasma, tissues, and cell membranes depends on the fatty acid pattern of milk lipids (Putnam et al. 1982, Koletzko et al. 1987). The role of long-chain polyunsaturated fatty acids (LC-PUFA) with chain lengths of 20 and 22 carbon atoms has received special attention because the supply of LC-PUFA during life may be of particular relevance for growth and maturation of the nervous system (Koletzko and Bremer 1987). A brain growth spurt normally occurs in humans between the second trimester of pregnancy and 18 months postpartum. Among the changes in composition that occur are an accumulation in the developing brain of the essential fatty acids, 18:2w6 and 18:3w3, plus the products produced therefrom by elongation-desaturation (Clandinin et al. 1981, Clandinin et al. 1981a, Heim 1983). These long chain polyunsaturated fatty acids are physiologically important as precur-sors for the synthesis of prostaglandins and structural lipids (Nutrition Reviews 1984). More specifically, they are required for infant's brain development, cell pro-liferation, myelination and retinal function (Hamosh et al. 1984). The biologic importance of the concentration and composition of fatty acids of human milk for an infant's growth and development has prompted extensive research work in this field (Lammi and Jensen 1984). This section will provide a brief review on the effects of processing and storage on the stability of lipid constituents in expressed human milk (EHM). 2.3.1 Lipolysis in Processed and Stored E H M The stability of milk fat during storage has been the subject of numerous studies in the dairy industry, because hydrolysis of fat leads to the formation of free caproic and capric fatty acids which become the cause of the odour of rancid dairy products (Small 1986). The information gained from these studies cannot be applied to human milk because the fatty acid profile as well as the nature of the lipase(s) (the enzymes that might hydrolyze milk triglycerides during storage), differs in human and bovine milk. In contrast to bovine milk, which contains mainly lipoprotein lipase (Jensen and Pital 1976), human milk contains two different lipases: bile salt stimulated lipase (BSSL) and lipoprotein lipase (LPL) (Hernell and Olivecrona 1974). In the human neonate, BSSL is activated by bile salts in the intestine and leads to com-plete hydrolysis of triglycerides (Fredrikzon et al. 1978, Hernell and Olivecrona 1974a). Serum activated LPL, however, has no known function in neonatal diges-13 tion (Hernell and Olivecrona 1974a, Hamosh 1981, Hernell et al. 1981). Numerous studies with bovine milk have suggested that the serum stimulated lipoprotein li-pase may be responsible for lipolysis and the subsequent rancidity encountered during storage (Schwartz and Parks 1974). Since 1964, it has been known that lipolysis of lipids in human milk occurs during refrigeration (Tarassuk et al. 1964). More recently it has been shown that lipolysis of human milk occurs even when milk is stored at temperatures far lower than the refrigeration temperatures (around -20° C) (Clark et al. 1984, Berkow et al. 1984, Hamosh et al. 1984, Lavine and Clark 1987). Friend et al. (1983a) conducted a study to evaluate the effects of several meth-ods of freezing and storage on lipid components of mature human milk. During storage, there appeared to be progressive increases in lipase activity and the level of unesterified fatty acids, indicating that spontaneous lipolysis was occurring in the milk. Evidence of lipolysis was more pronounced in the quickly frozen sam-ples stored for 3 months than in the slowly frozen samples. However, as there were considerable sample to sample variations, the differences observed were not significant. Clark et al. (1984) showed a significant increase in the serum independent lipolysis in human milk with storage time. Storage at -20° C resulted in an increase of 20% and interestingly enough, storage at temperatures as low as -70° C also resulted in an increase of 36% over 4 weeks. Berkow et al. (1984) on the other hand found that although LPL and BSSL remain fully active during frozen storage of human milk, milk fat is hydrolyzed at -20° C but not at -70° C. When milk specimens were stored for 5 months, the triglyceride concentration decreased by 10 and 13% in milk stored at -20° C, immediately or after repeated freeze thawing, respectively; there was no decrease in the triglyceride content of milks stored at -70° C. Concomitant with the decrease in milk triglyceride levels during storage at -20° C, the free fatty acid content increased to 10 and 13.6% in milk stored immediately or after three freezing and thawing cycles respectively; free fatty acid levels changed only minimally in milk specimens stored at -70° C (Berkow et al. 1984). In another study by Hamosh et al. (1984a) the effects of storage on the lipids in human milk were studied. Milk specimens were stored at -20° C or -70° C for 2 months. The data showed that significant hydrolysis of milk triglycerides occurred during storage at -20° C, whereas storage at -70° C preserved the original composition of human milk fat. Recently Lavine and Clark (1987) also found that there was no measurable lipolysis in milk stored at -70° C for 8 weeks. This was in marked contrast to the 14 significant (P < 0.05) increase in the free fatty acids in milk stored at -11° C for the same period. It has been reported that BSSL loses its dependency upon bile salts during storage at -20° C (Mehta et al. 1982). This might explain the observed increase in serum independent activity in milk stored at -20° C. However, the increase in lipolytic activity even at -70° C observed in a few studies (Friend et al. 1983a, Clarke et al. 1984) has been speculated by some investigators to be a result of measuring a non-specific lipase activity (Berkow et al. 1984). However, in any case, the point to be made from these observations is that milk which is stored frozen undergoes more lipolysis than freshly collected milk. In fresh milk the fat globule membrane may prevent lipases from acting on triglycerides. Serum stimulated lipase (LPL) is associated with the fat globule, but on the outer surface, while bile salt stimulated lipase is present in the aqueous phase (Hernell and Olivecrona 1974, Hernell and Olivecrona 1984a). The milk fat globule as such is completely resistant to lipolysis, as long as the structure remains intact (Berkow et al. 1984). However, if these, milk fat globules are damaged, triglycerides would be available as substrates for lipases, such damage to milk fat globules can occur on heating or freezing and thawing (Wardell et al. 1981). In a study of human milk collected for banking, heating at 62.5° C for 30 minutes and freezing and thawing resulted in hydrolysis of triglycerides (Wardell et al. 1981). Freezing and thawing caused disruption of fat globules and a greater hydrolysis of triglycerides than did the heating process. Another study reports extensive hydrolysis of triglycerides when the specimens were thawed twice and then stored at -20° C for 2-5 months, with accumulation of large amounts of free fatty acids and some partial glycerides (Hamosh et al. 1984). To date, there is no information on the effect of ultrasonication on the release of fatty acids as such and during storage of human milk, although it has been reported that in bovine milk rapid lipolysis is also induced by sonication (Berkow et al. 1984). It is important to study this effect since the change in milk lipids brought about by such processing and additional storage may have a significant impact on the digestion and absorption of human milk in the infants. 15 2.3.2 Effect of Processing and Storage on the Polyunsatu-rated Fatty Acid Profile of E H M Lipids account for the bulk of energy in human milk, however their constituents play other important roles in the metabolism of mammals as well. Linoleic (Cl8:2w6) and linolenic (Cl8:3w3) acids, for instance, are converted to long chain polyunsat-urated fatty acids (LC-PUFA). These metabolites of the w6- and w3- series have been shown to affect the fluidity and stability of biological membranes (Olegrad and Svennerholm 1971, Sinclair and Crawford 1972). The high need for long chain PUFA during postnatal myelination and their presence in visual cells may suggest their importance for the development of the newborn (Sinclair and Crawford 1972, Cuthbertson 1976, Harzer et al. 1983). Most previous studies report the release of total fatty acids in milk during storage without identifying individual fatty acids (Berkow et al. 1984, Hamosh et al. 1984, Lavine and Clark 1987, Wardell et al. 1981). Investigation of individual fatty acids released during storage is therefore needed with particular attention to long chain unsaturated fatty acids. Human milk contains 1.08 to 2.65 percent w-6 fatty acids and 1.09 to 1.47 percent w-3 fatty acids synthesized from C-18 precursors (Clandinin et al. 1981a). Infant formulas on the other hand have been found to contain little or none of the LC-PUFA, as per the limited information available (Nutrition Reviews 1984, Clandinin et al. 1981). The requirements for infants have been calculated on the basis of the rate of deposition of w-3 and w-6 fatty acids in the tissues in infants (Clandinin et al. 1981). If the net energy intake of the nursing infant is 120 kcal per kg body weight per day, then about 20 kcal of fat fed per kg (2.2g) is deposited per day. A minimum of 23.6-58.3 mg of w-6 and 23.9 to 32.3 mg of w-3 acids is deposited per day. Human milk contains more than enough of these LC-PUFA to meet the needs of neural tissue (Nutrition Reviews 1984). The requirements were calculated to be 43.1 mg/week for w-6 and 22.1 mg for w-3 LC-PUFA. The actual amounts of essential fatty acids in the 200 ml of milk consumed in mg per day were w-6, 850 and w-3, 140 and the observed fatty acid accretion rates for total tissue, in mg per day were : w-6, 552 and w-3, 57 (Nutrition Reviews 1984). Processing and storage of human milk has been shown to have some effect on the pattern of these fatty acids present in human milk. Lavine and Clark (1987), recently looked at the effect of frozen storage (at -11° C and -70° C for 1, 2, 4, 6 or 8 weeks) on the release of individual fatty acids in human milk. At -11° C it was found that the proportion of 12:0, 14:0, 16:0 and 18:0 in the total free fatty acid pool decreased with the length of storage while long chain unsaturated fatty acids increased. The distribution of free fatty acids present in milk remained unaltered 16 when stored at -70° C. Chappell et al. (1985) also reported a change in distribution of free fatty acids in milk stored without an inhibitor of lipoprotein lipase which was similar to the changes observed by Lavine and Clark. Quantitatively, the greatest change seen was in the proportion of free 16:0 and 18:2. In freshly collected milk these two fatty acids contribute equally to the free fatty acid pool. As storage progresses, the percentage of 18:2 greatly increases while the percentage of 16:0 decreases. This change can be explained by the distribution of fatty acids in milk triglycerides and preferential hydrolysis of primary position by lipoprotein lipase in breast milk (Wong et al. 1982). Besides these major fatty acids, the distribution of the other free fatty acids in stored milk resembles the distribution of fatty acids in the primary positions of milk triacylglycerols rather than total triacylglycerol. Chappell et al. (1985) did not provide data on fatty acids with greater retention time on the gas chromatography than 18:2. However, Lavine and Clark (1987) observed that the proportion of longer chain polyenoic acids in the free fatty acid pool increased with length of storage. Their increase in the free fatty acid pool may alter their availability to the infant. It is not known if these long chain fatty acids would be more available to the infant or less. The availability may depend on the conditions in the lumen of the intestine (Lavine and Clark 1987). Similarly, Hernell and Blackberg (1982) suggest that, when intraduodenal bile salts are low, free fatty acids are better absorbed than esterified fatty acids. This may have important implications for infant nutrition as these long chain polyenoic acids (Cl8:3n3, C20:4n6, C22:4n6, C20:5n3, C22:6n3, C22:5n3) are needed by the infant for the synthesis of prostaglandins and thromboxanes (Lavine and Clark 1987). These LC-PUFA are preferentially accreted in the structural lipids of growing neural tissues (Clandinin et al. 1980) and may be relevant for brain function (Lavine and Clark 1987). Considering the implications of the above findings for infant nutrition, it is im-portant to study if ultrasonicating human milk and further lyophilizing or freezing would bring about any more changes in the free fatty acid profile. To date there is almost no information regarding the effects of processing human milk on its fatty acid profile. One study by Wardell (1981), looked at the effect of heating as well as freezing and thawing (but not storage) on the distribution fo fatty acids in human milk collected for banking. Heating at 62.5° C for 30 minutes and freezing and thawing resulted in significant hydrolysis of triglycerides. A decrease in the percentage of polyunsaturated fatty acids, linoleic acid (C18:2) and linolenic acid (C18:3), was observed after freezing, thawing and heating, but the other fatty acids of human milk triglycerides were not affected. They suggest that 17 the availability of linoleic and linolenic acid in milk declines when these procedures are used in human milk banking. Although these results do not totally agree with the results of Lavine and Clark (1987) and Chappell et al. (1985), they do seem to suggest that any slight modi-fication in the processing procedure might have a different impact on the pattern of fatty acids present in the human milk. Since application of ultrasonication in milk banking will be more as an addi-tional rather than as an only procedure, it would be important to look at the combined effect of ultrasonication/freezing/lyophilization and storage time on the fatty acids profile in human milk. 2.3.3 Oxidative Instability in Processed and Stored E H M It has been speculated that the loss of linoleate (C18:2) and linolenate (C18:3) observed after heating, freezing and thawing in human milk could be a result of the autoxidation reactions occurring in the unsaturated fatty acids (Wardell et al. 1981). Oxidation of the milk lipids has always been of major importance to the dairy industry. Fat oxidation is considered to be an unwanted reaction in food, mainly due to the formation of hydroperoxides and their volatile degradation products, such as short chain aldehydes, ketones, acids and hydroxyl compounds (Camp-bell et al. 1979, Hall et al. 1985). An increase in the percentage of long chain-polyunsaturated fatty acids in the free fatty acids pool with storage (Lavine and Clark 1987, Chappell et al. 1985) might possibly increase autoxidation in human milk. Lipid oxidation could be a problem in freeze-dried (lyophilized) human milk as dehydrated foods are considered to be susceptible to oxidation even at very low partial pressures of oxygen (Goldblith 1969). Lyophilization as a means of preservation is used primarily to remove water, so that the food stuff is inert chemically as well as microbiologically (Goldblith 1969). This process has been successfully used for preparing protein and cream fractions from the donor's milk needed for fortification of human milk for premature babies (Garger et al. 1984), and has been used for preservation of collected breast milk in the human milk banks of the Children's Hospital in Lindenhof, Berlin since 1986 (Bottcher et al. 1987). So far there is no information regarding the effect of lyophilization on the possibility of lipid oxidation in human milk and the resultant loss of LC-PUFA or development of rancid and off flavours. On the other hand, frozen storage of milk is known to possibly result in hydrolytic rancidity (Clark 18 et al. 1984). This most certainly might affect the acceptability of milk for infant feeding. Acceptability for human consumption can be determined by analyzing the ox-idation of lipids in human milk using chemical tests such as detection of peroxide value, thiobarbituric acid test and determination of total and volatile carbonyl compounds (Small 1986). Similarly, in order to check the suitability of ultrasonicated and stored milk for newborn babies, it would therefore be important to determine the oxidation of fat in that milk. 2.4 Processing Effects on the Immune Compo-nents of Human Milk One of the primary reasons for the resurgence of interest in breastfeeding is the pro-tection against infection that has been believed to be associated with it (Sunshine 1980). Human colostrum and milk secreted in the early period of lactation not only provide adequate nutritional support to the newborn, but also supply a passive sup-plement for the immature immune system by transferring to the gastrointestinal tract substantial quantities of immunologically active molecules (Clinical Nutrition Review 1987). Larsen and Homer (1978) report that, of 107 infants hospitalized with acute gastroenteritis, only one was being breast fed. Fallot et al. (1980) found that 11% of 136 infants 0-3 months of age that were hospitalized during a 1 year period were breast fed. These authors also reported that none of those in the breast fed groups suffered bacterial infections, whereas 27 cases of bacterial infections were documented among 121 non breast fed infants. Fallot et al. (1980) concluded that exclusive breast feeding of infants less than 3 months of age significantly reduced the incidence of gastroenteritis leading to hospitalization. These investigations however do suffer from two methodological shortcomings: first, is the failure to control for demographic and clinical factors that might affect a child's susceptibility to infection; and second, the failure to use active surveillance to detect infections, because given certain symptoms, mothers who breast feed may seek medical care and hospitalization for their children at a different rate from that of mothers who formula feed infants. Recently, Leventhal et al. (1986) conducted a case control study in which after minimizing the potential surveillance bias and stratifying the case- control pairs by the severity of their medical condition at the time of hospitalization, found that 19 breast feeding protects infants from hospitalization rather than from infections. He concluded that failure to consider the problem of surveillance bias might lead to erroneous conclusions about the protective effect of breast feeding. Nevertheless, it is still believed that there are certain protective proteins in human colostrum and milk which are available to the newborn for passive immunity (Lewis et al. 1985, Reddy et al. 1977, Gross et al. 1981, Goldman et al. 1982, Goldman et al. 1983). The passive protection of the breast fed individual especially against the intestinal pathogens present in the maternal environment is considered assured (Cruz and Arenalo 1986). Among the components that are thought to contribute to the anti-infective properties of human milk are lactoferrin, lysozyme, bifidus factor, anti- staphylo-coccal factor, components of the complement system and immunoglobulins (Mata 1979, Goldman and Smith 1973). The most important of the soluble substances, in terms of protection against diarrheal disease in the breast-fed child, seems to be the secretory immunoglobulins of the IgA class (slgA) (Hanson et al. 1973, Glass et al. 1983). They are especially high in colostrum (2-4 mg/ml). Then after 2- 4 days, levels decrease to 1-2 mg/ml which persist for several months (Jelliffe and Jelliffe 1979). IgA is quite resistant to proteolysis and pH change, and is though to prevent infections by blocking adherence of pathogens to the intestinal mucosa (Vahlquist 1976). Considering the beneficial effects of the immunoglobulins and other immune components, their retention in banked human milk is important. However, some studies looking at the effects of processing on immunoglobulins show a decrease in the immunoglobulin levels, especially after heat treatment (Goldsmith et al. 1983, Evans et al. 1978, Liebhaber et al. 1977). These studies indicate that while holding the time of heating constant at 30 minutes, increasing the temperature resulted in decreased concentration of IgA. Evans et al. (1978) report that IgA survived with relatively little loss up to the temperature of 70° C (33.3% loss). They found that accurate pasteurization at 62.5° C for 30 minutes in a well regulated water bath produced no loss of IgA. On the other hand, Wills et al. (1982) showed losses of 33%, 23%, and 10% of IgA following heat treatment at 62.5° C for 30 minutes, 62.5° C for 5 minutes and 56° C for 15 minutes respectively. Liebhaber et al. (1977) also reported a 33% loss of IgA following pasteurization at 63° C for 30 minutes. Although the aim of heat treatment is to destroy pathogens with minimum loss of milk constituents, significant losses of immune constituents are still observed. 20 In the present study we propose to carry out ultrasonic homogenization of EHM before storing and/or administering milk to the babies. The process of ultrason-ication also involves some generation of heat. The combined effect of ultrasonic waves and heat treatment has been shown to enhance the bactericidal efficiency in milk (Juan et al. 1987). However, the effect of ultrasonication on the immune components of human milk has yet to be investigated. Little is known regarding the best method of storage of human milk for preser-vation of anti-infective properties. A number of milk banks use deep freezing as a means of long-term storage (Sunshine et al. 1980, Committee on Nutrition, American Academy of Pediatrics 1980) and recently there have been attempts to lyophilize and store human milk (Bottcher et al. 1987). Evans et al. (1978) and Liebhaber et al. (1977) have studied the effects of deep freezing and lyophilization on certain immune components in human milk. Liebhaber et al. (1977) found that freezing human milk specimens for up to 4 weeks at -23° C resulted in no significant alteration of the IgA content. Similarly, Evans et al. (1978) report that deep freezing human milk at -20° C for 3 months produced no appreciable loss of IgA. On the other hand, the effects of lyophilization did not appear to be similar in the two studies. Liebhaber et al. (1977) found that lyophilization caused a signif-icant decrease in immunoglobulin concentration (21% loss) (P < 0.001); whereas Evans et al. (1978) did not find any significant loss of IgA after lyophilization and reconstitution of human milk. In short, we do know that the most commonly used method for preserving the quality of human milk (i.e. pasteurization) has a somewhat detrimental effect on its IgA content. However, considering the importance of reducing the potential transmission of infection from contaminated milk, slight loss in IgA concentration is probably acceptable. Storage of human milk by deep freezing seems to be a very satisfactory pro-cedure with no appreciable loss of IgA. On the other hand, information regarding the effect of lyophilization on the IgA levels of human milk still appears to be inconsistent. 21 2.5 Fat and Protein Concentrations in Human Milk When Delivered Through the Infusion Pumps The question of what milk should be given to preterm babies has attracted greater attention than 'how' it should be delivered. Scant attention has been given to the effects on human milk of the various methods of milk delivery used in the inten-sive care nursery. Because of their convenience, and possibly improved tolerance, continuous nasogastric and nasojejunal feedings using mechanical infusion pumps have become popular (Benda 1979). They have also gained acceptance because of reported earlier achievements of 'full feeds' and earlier regaining of birth weight by premature infants (Heird 1973). Recently, several pediatricians using mechanical infusion pumps for delivery of small volumes of human milk at low infusion rates have become concerned about fat separation as well as the fat losses that appeared to occur in the apparatus used in these continuous feeding systems (Narayanan et al. 1984, Brooke and Barley 1978, Stocks et al. 1985, Greer et al. 1984). Studies describing the loss of energy or fat have highlighted the importance of delivery technique in tube feeding. Brooke and Barley (1978) recorded up to 24% energy loss in a continuous system using a burette pump. The authors speculate that milk fat separates rapidly after the infusion is set up. This results in the lower energy content of the two to four hour samples, which are presumably relatively low in fat. Thereafter, increasing amounts of the upper fatty layer appears in the samples and the final residuum contains a disproportionate amount of fat, both as the remains of the high-fat layer and as fatty scum adherent to the burette. They suggest that hourly agitation of the milk in the burette is important to ensure that it remains reasonably well mixed. Spencer and Hull (1981) showed that a substantial amount of fat is wasted (mean of 34%), when drip sets controlled by an IVAC infusion pump are used for continuous feeding. Smaller though still appreciable amounts are lost when using a vertically placed syringe pump (mean of 19%). Some at risk babies therefore receive milk with a much lower energy content than the recommended 100-150 kcal/kg (0.42-0.63 MJ/kg). In another study by Spencer et al. (1982), there appeared to be no relation between the amount of energy provided to the baby (as per calculations of energy content in milk) and weight gain. The possible reasons for this could be variation 22 in energy absorption, variation in metabolic rate or in the uptake of nutrients from the milk. Brooke et al. (1979) showed using a formula feed that absorption of nutrients varies greatly both from day to day and between different immature babies. Although they observed a substantial variation in the metabolic rate, there was a strong correlation between weight and the transfer of nutrient (mainly fat and therefore energy) from the milk present in the tubes and syringes to the baby which would possibly account for the large variation in weight gain observed by Spencer et al. (1982) despite relatively small variation in energy intake. This fat (and therefore energy) content lost in the syringe and tube walls during tube feeding could obscure the influence of feeding human milk on growth of low birth weight babies. Narayanan et al. (1984) also observed a rise in the fat concentration of human milk within the syringe towards the end of the continuous infusion. They described a method of reducing fat accumulation in the reservoir syringe by using eccentric nozzle syringes set at an angle. A recent study extends these investigations by showing that continuous infu-sions result in great losses compared with intermittent boluses and that slow flow rates were associated with greater fat loss, which is relevant to the lowest birth weight babies (Stocks et al. 1985). The mean fat losses at different feeding rates for bolus and continuous methods were 17% and 34%, respectively, and there was a significantly inverse correlation between fat loss and flow rates for both techniques (r2 = 0.865 and 0.922 respectively). In the case of proteins, although the mean losses were significant (P < 0.05) at 7% and 5% for bolus and continuous feeding respectively, there was no significant difference between the techniques nor any correlation of loss with flow rate (Stocks et al. 1985). Large cumulative fat losses and highly significant decreases in fat concentration in human milk during delivery using slow infusion rates were also reported by Greer et al. (1984). They attempted to look at the effect of homogenization on fat concentration in milk when infused at a slow rate. Milk samples were either shaken vigorously by hand until the sample appeared to be homogeneous or blended in a Waring blender at high speed for 2 minutes. The majority of the fat globules were between 2 and 5 /zm for blended milk and 15 to 20 [im for the shaken milk. The changes in fat concentration however, were not significantly altered by this treatment. These studies raise several concerns relevant to the feeding of human milk to low-birth-weight infants by continuous nasogastric infusion using a mechani-cal pump. First, a significant proportion of fat that separates from EHM during continuous tube feeding becomes unavailable to the baby. In as much as fat ac-23 counts for 50% of the calories of mature human milk (Garza et al. 1987), 15% to 20% of the total caloric value of milk is lost (unavailable) in these delivery systems. If tubing with residual milk is discarded after the syringe or burette are empty, an even larger amount of fat is wasted and becomes unavailable (up to 50% of the total fat and 27.5% of the intended total calories (Greer et al. 1984). The extent of losses of other substances that may be associated with the fat, for example, fat soluble vitamins and trace minerals is unknown. It is possible that protein/calorie ratio of human milk feedings would be altered by the loss of fat with continuous pump infusions (Martinez et al. 1987). The second factor of major concern is the marked decrease in fat concentration that occurs during the infusion of human milk using a mechanical pump, resulting in marked variation in fat concentration during the continuous infusion. Also of concern is the large fat bolus delivered at the end. Premature infants are known to absorb fat inefficiently, and such boluses of fat may lead to feeding intolerance and delay the advancement of enteral feedings in these infants (Alemi et al. 1981). An important clinical implication would be that these fat and protein losses will tend to invalidate any metabolic study of the influence of feeding human milk on the growth of low birth weight babies. This would mainly result from miscalculating the nutrients delivered to the baby on the basis of fat content of the milk before delivery rather than on the fat content of milk actually delivered to the baby. Similarly, nutrient losses may be a potential cause for this lower weight-gain shown by babies fed continuously by the naso-jejunal route compared with those fed with intermittent boluses (Whitfield 1982). Very recently it has been shown that homogenization by ultrasonic treatment prevented changes in fat concentration during infusion and essentially eliminated loss of this nutrient during administration (Martinez 1987). Unlike the study by Greer (1984), where milk was considered homogenized when the fat globules were between 2-5 \im in diameter, in this study EHM samples were considered ade-quately homogenized when most fat globules were uniform in size and less than 1.5 fim in diameter (Martinez 1987). Although this study shows that ultrasonic treatment of EHM significantly improves recovery of fat when delivered through tubing, we still do not know what would be the recovery of fat in ultrasonicated EHM after it has been stored, either frozen or lyophilized, for various lengths of time. Since breast milk is generally collected whenever it is available from a donor mother, processed and stored in the milk bank until used, it is necessary to study the percentage fat and protein recovery in ultrasonicated milk stored for various lengths of time. 24 C h a p t e r 3 E x p e r i m e n t a l D e s i g n a n d M e t h o d o l o g y The following chapter is divided into three sections. The first describes the study design for the various parts of the experiment. The second section describes the basic experimental procedures such as milk collection, pasteurization, lyophilization and so on. The third section deals with the assay techniques that are used for analyzing specific components of breast milk. 3.1 Experimental Design The entire study was conducted in four parts: 3.1.1 Part I: The purpose of this part of the study was to determine the effect of ultrasonication and heat treatment, singly and in combination, on the extent of lipolysis and lipid peroxidation in milk stored for various time periods. The experimental design for Part I of the study has been summarized in Figure 3.1. Details: Each of the seven expressed human milk (EHM) samples were divided into 3 equal parts and given three treatments: (a) Ultrasonicated only (b) Pasteurized only (c) Pasteurized and ultrasonicated A small portion of each sample was analyzed immediately after processing for the extent of lipolysis and peroxidation of milk fat. The rest was stored for one month and four months in two forms: 25 flssnv ULTRHSONICRTED F R O Z E N 1 M O N T H 4 M O N T H S PASTEURIZED + ULTRHSONICRTED L V O P H I L I Z E D 1 M O N T H 4 M O N T H S T H A W E D R S S R Y R E C O N S T I T U T E D flSSRV H S S f l V : 1. F R E E F A T T Y A C I D S 2. P E R O K I O E S FIGURE 3.1: Experimental design: part I. 26 (a) Frozen at -20° C (b) Stored under nitrogen gas after lyophilization Frozen samples were thawed and the lyophilized samples reconstituted and an-alyzed for the extent of lipolysis and peroxidation of fat after one month and four months of storage. 3.1.2 Part II: The purpose of this part of the study was to determine the effects of ultrasonication, lyophilization, freezing and storage time on the fatty acid profile (PUFA) of human milk. The experimental design for Part II is presented in Figure 3.2. Details: A fraction of each of five milk samples was analyzed for FAP imme-diately. The remainder of each sample was divided into two parts. One part was frozen immediately at -20° C without processing. The other part was ultrasonicated and stored in two forms: (a) Frozen at -20° C (b) Stored under nitrogen gas after lyophilization After one month of storage, both the frozen fractions (ultrasonicated and non-ultrasonicated), as well as the lyophilized fractions of all five samples were analyzed for the fatty acid profile following appropriate thawing and reconstitution. 3.1.3 Part III: The purpose of this part of the study was to determine the effects of ultrasonication, lyophilization, freezing and storage time on the immunoglobulin A (IgA) levels in human milk. The experimental design for part III is presented in Figure 3.3. Details: A fraction of each of seven EHM samples was analyzed for IgA in the beginning and the rest was ultrasonicated and divided into three parts. One part was analyzed for the IgA fraction immediately (within 24 hours) and the remaining two parts were stored either frozen at -20° C or under nitrogen gas after lyophilization for a period of one month. After one month the samples were thawed and reconstituted accordingly and analyzed for the IgA fraction. 27 GD s i U L T R f l S O N I C f l T E D F R O Z E N 1 M O N T H T H f l U J E D F A T T Y A C I D A N A L Y S I S zx L Y O P H I L I Z E D 1 M O N T H R E C O N S T I T U T E D F f l H V R C I D flNHLVSIS F R O Z E N 1 M O N T H T H f l U J E D F A T T Y H C I D flNHLVSIS F A T T Y R C I D A N A L Y S I S FIGURE 3.2: Experimental design: part II. 28 Ig f i flNHLVSIS I g f l flNHLVSIS F R O Z E N 1 M O N T H THf lUJED Ig f l flNHLVSIS L Y O P H I L I Z E D 1 M O N T H R E C O N S T I T U T E D Ig f l A N A L Y S I S F I G U R E 3.3: Experimental design: part III 29 3.1.4 Part IV(a): The purpose of this part of the study was to determine the effect of lyophiliza-tion, freezing and storage time on the fat and protein recovery in mechanically infused ultrasonicated milk. The experimental design for Part IV(a) of the study is presented in Figure 3.4. Details: Each of the eight milk samples was heat treated (pasteurized), ultra-sonicated and then divided into two parts. One part was lyophilized and the other frozen at -20° C . Both these portions were stored for one and four months time as shown in Figure 3.4. Samples were appropriately thawed and reconstituted and then analyzed for the recovery of fat and protein. Recovery was assessed following delivery of the sample at both a rapid rate (40 ml/h ) and slow rate (10 ml/h ) using laboratory models simulating tube feeding in clinical practice. 3.1.5 Part IV(b): This was a follow-up study of Part IV(a). The purpose was to determine if ul-trasonication of milk after it had been stored in the lyophilized form gives better recovery of fat and protein when infused. The experimental design for Part IV(b) of the study is presented in Figure 3.5. Details: Five milk samples were obtained. Each sample was divided into two parts. (a) One part was pasteurized, ultrasonicated and then lyophilized and stored. After one month it was reconstituted, re-ultrasonicated and then infused. (b) The other part was pasteurized, lyophilized and stored for one month, after which it was reconstituted and then ultrasonicated and infused. The recovery of fat and protein was assessed as described in Part IV(a) of the study. 3.2 Experimental Procedures 3.2.1 Milk Collection and Standardization Breast milk was available at "no cost" from the Children's Hospital breast milk service, operating under the Division of Biochemical Diseases, Department of Pae-30 i GOOD MOON MO LLCS. PASTEURIZED AND ULTRASONICATED LYOPH 1L1 ZED FR02 EN 1 MONTH 4 MONTHS TUBE DELIUERY 1 MONTH 4 M ONTHS TUBE OELIUERV y\ y\ /\ s< RECOUERV RAPID SLOUJ RECOUERV * RECOUERV * RECOUERV * FAT AND PROTEIN RECOUERV FOLLOUJING INFUSION RT RAPID ( 40 ml/hr ) AND SLOUJ ( l O m l / h r ) RATES FIGURE 3.4: Experimental design: part IV(a). 31 LYOPHILIZED 1 MONTH RECONSTITUTED QD LL1TD9 tn] § LD iKI D CC SlTrLl LU TUBE DELIUERED RAPID SLOUJ RECOUERY* 5TTE 00 0302111 + LYOPHILIZED 1 MONTH RECONSTITUTED i i i r « ( o ) K ] O O T [ i i TUBE DELIUERED RAPID SLOUJ RECOUERY* FIGURE 3.5: Experimental design: part IV(b). 32 diatrics, Children's Hospital, University of B.C. Pooling of milk samples was undertaken when necessary to increase the sample volume. The nutritional and hygienic qualities of milk samples were controlled by ac-cepting milk which met a set of criteria: 1. From mothers who delivered infants at term. 2. From mothers lactating between 2-8 months. 3. From mothers who had been thoroughly screened for infectious diseases by Breast Milk Service (BMS) Coordinator following standard practice. 4. From donors instructed in the technique of expression by BMS Coordinator. 3.2.2 Ultrasonic Treatment of Expressed Human Milk The objective of this treatment was to homogenize human milk so that the size of the fat globule would be similar to that of commercial homogenized cow's milk. The conditions for homogenization were optimized by Martinez et al. (1987) by subjecting EHM samples to ultrasonic treatment at different levels of intensity, and for different time periods. In the present study, approximately 65 ml of milk sample was ultrasonicated at one time at a relative intensity of 5 (on a scale of 1 to 10), over a period of 5 minutes. The apparatus used was Tekmar Sonic Disruptor Model TSD-P 250, 500 watts, 20 Hz (Tekmar Co., Cincinnati, Ohio). 3.2.3 Pasteurization Expressed human milk was pasteurized in the Oxford human milk pasteurizer at the Children's Hospital, Vancouver, for 30 minutes at 63° C and followed by a rapid cooling cycle. 3.2.4 Infusion System For Tube Delivery of E H M Laboratory models simulating tube feeding in clinical practice were used for this study. Two methods of infusion were tested: a slow infusion system using a me-chanical syringe pump and a fast infusion system using gravity flow from an open syringe. The slow infusion system consists of a mechanical pump (Model AS-5B, Auto-Syringe Inc., Hookset, New Hampshire), and a 50 ml plastic-syringe (Plastipak 33 BD) connected to an extension tube 152 cm long, 1.8 ml capacity (Meldon Inc., Burbank, California) and an 8 French Argyle feeding tube (Sherwood Medical, St. Louis, Missouri) 40.5 cm long. The syringe was maintained in the horizontal position with agitation over a period of 4 hours. The volume and rate of milk to be infused during this period was calculated on the basis of the amount which would be received by a 1600 g newborn at a daily rate of 150 ml/kg. This represents a volume of 40 ml of infusate in a 4 hour period and was delivered by adjusting the flow rate to 10 ml/h Aliquots of milk flowing out of the feeding tube were collected in glass test tubes at 30 minute intervals and stored frozen at -20° C for analysis within a period of one week. Precautions were taken to recover as much milk as possible from the feeding tube by continuing air infusion at the end of each trial. The rapid infusion system consists of a 30 ml open plastic syringe (Plastipak BD) connected to a 44 cm extension tube (MX 450, Medex Inc., Hilliard, Ohio) and a French Argyle feeding tube (Sherwood Medical, St. Louis, Missouri). The syringe was held vertically and 20 ml of milk was allowed to flow by gravity over a period of 30 minutes (rate 40 ml/h). This approximately simulates the condi-tions encountered when feeding newborn babies every two hours. Aliquots of milk delivered by the tube were collected at six minute intervals and stored frozen for analysis as described for the slow infusion system. 3.2.5 Lyophilization of Expressed Human Milk Human milk samples were lyophilized using the freeze driers in the Departments of Food Science and Chemistry, University of British Columbia. For Part I and Part IV of the study a "Lab-Conco" Model-18 freeze drier in the Food Science Department was used. Due to the unavailability of this freeze drier on occasions, milk samples for Part II and Part III of the study were freeze dried using the "DURA-DRY", Corrosion resistant freeze drier (FTS Systems Inc., Stone Ridge, New York) available in the Chemistry Department. In both cases milk samples were placed in containers with large surface area (to facilitate drying), frozen, covered with parafilm with a few small holes in it and then placed in the freeze drier for approximately 3 days. After the samples were dried, they were removed, flushed with nitrogen gas and sealed tightly for storage. 34 3.3 Assay Procedures In the following section, flow diagrams for all the assay methods described are presented in order to help understand the steps clearly. 3.3.1 Free Fatty Acid Analysis The 'Extraction-titration' method of Deeth (1975) was used for estimating lipolysis in human milk samples. Milk (2.5 ml) in a test tube was mixed with 5 ml of extraction mixture (isopropanokpetroleum ether: 4NH2S04, 40:10:1). Petroleum ether (3 ml) and water (2 ml) were added and the stoppered test tube shaken vigorously for 15 seconds. The two layers were allowed to settle (5-10 min) and an aliquot of the upper layer was withdrawn and transferred to a 50 ml conical flask. After addition of 6 drops of 1% methanolic phenolphthalein the solution was titrated with 0.02 N methanolic KOH. A blank, in which milk is replaced with water, was used to obtain the background titration. The free fatty acid content (// equivalent/ml) of milk was obtained from the formula • 103 where T is the net titration volume, N the normality of the methanolic KOH, P the proportion of the upper layer titrated (i.e. volume of aliquot withdrawn/total volume of upper layer) and V the volume (in ml) of milk. The method has been summarized in Figure 3.6. 3.3.2 Determination of Peroxides A modified peroxide test for the detection of lipid oxidation in milk was used (Stine et al. 1954). The fat layer from the milk was first separated using the extraction procedure (Deeth 1974) described above. Aliquots obtained were flushed with nitrogen gas to evaporate petroleum ether. Using a micropipetter, 100^ j£ of fat was removed, and placed in a 10 ml standard taper volumetric flask. A solution fo benzene:methanol (70:30) was added to the mark and the stoppered flask inverted several times to dissolve the fat. This was followed by the addition of one drop (0.04 ml) of ferrous chloride and one drop (0.04 ml) of ammonium thiocyanate reagent and the flask was then shaken vigorously to disperse the reagents and placed in a water bath at 50° C for exactly 2 minutes for colour development. The temperature of the coloured solution was lowered to approximately room temperature during about 2 minutes in an ice bath. Finally, the mixture was transferred to a cuvette and the light absorption was 35 M I L K ( 2.5 ml ) + 5 ml EHTRRCTION MIHTURE j~1S0PR0PAN0L : PETROLEUM ETHER: SULFURIC RCID |_ 40 10 1 ] 3 ml PETROLEUM ETHER 2 ml LURTER ALL CONTENTS MIRED ALIQUOTS TITRATED UJITH 0.02 N METHONOLIC KOH CALCULATIONS FREE FATTY ACIDS T . N P . U X io T = TITRATION U 0 L U M E N = NORMALITY OF KOH P = PAOPORTION OF UPPER LAYER TITRATED ( UOLUME ) U = UOLUME ( IN m l ) OF M I L K FIGURE 3.6: Free fatty acid analysis. 36 determined at 505 m/x. A fat blank was run on the sample. Oil (100 fil) was handled in precisely the same manner except no ferrous chloride reagent was added. Use of reagents kept for more than one month lead to a higher peroxide value than would be observed with a fresh reagent. Therefore, fresh reagents were pre-pared after every month. The peroxide value was expressed in terms of milli-equivalents of oxygen per kg of fat. The analysis and calculations of peroxide value are shown in Figure 3.7 and Figure 3.8 respectively. 3 .3 .3 Immunoglobulin A Analysis Immunoglobulin A levels were quantitated using the radial immunodiffusion method of Mancini (1965). The plates and standard reagents for the determination of total IgA were obtained from Behring Diagnostics, Behringwerke AG, Marburg. For the standard, the contents of flask containing low concentration IgA protein standard serum (LC-A) were reconstituted with 0.5 ml of distilled water. For preparing reference curves, three dilution stages (1:1, 1:3, 1:7) with isotonic sodium chloride solution were set up. 20 \it of these three different dilutions were pipetted into three corresponding wells on the LC partigen immunodiffusion plate using the LC partigen microdispenser. The plates were placed at room temperature for 2-3 days and the diameter of the precipitation zones was measured and a standard curve was prepared by plotting width vs. standard concentration. For the control, the vial containing the control serum for NOR-partigen assay was rotated gently and the content brought to room temperature. A 40 fold dilution (10:390 fj.£) of the control was made using isotonic sodium chloride as diluent. 20/j.l of the diluted control was applied to the plate and the diameter of the precipitation zone read after 2-3 days and the value plotted on the standard curve. The milk samples were diluted 9 times (1:7) with isotonic sodium chloride, shaken and then applied to the immunoplates (20/z£/well). The diameter of the precipitate was read as in the above cases and concentrations derived from the standard curve in mg/d£. The method has been summarized in Figure 3.9. 3 .3 .4 Fatty Acid Profile Analysis The human milk samples were analyzed for the fatty acid profile at the B.C. Chil-dren's Hospital Research Centre using gas chromatographic analysis. 37 M I L K ( 10 ml ) FAT SEPARATED BV EHTRACTION METHOD ( DEETH 1975 ) 0.1 ml ( OIL ) 2 BENZENE : METHf 1N0L ( 10 ml ) FERROUS CHLORIDE ( 0 .04 ml ) A M M O N I U M T H I O C /NATE ( 0 .04 ml ) SHAKEN Ul GOROUSLY COLOUR ( 5 0 5 MEASURED n m ) FAT BLANK : STEP " 3 " OMMITTED. AEAGENT BLANK : STEP " H " OMMITTED. FIGURE 3.7: Determination of peroxides. 38 CALCULATIONS PEROHIDE UALUE : MILLIEQIUHL5NTS OF 0 2 / K G FAT 7. TRRNSMITTANCE : CONUERTED TO ug OF Fe / 10 ml SOLUENT OF BLANKS ( njiJH STANDARD CURUE ) NET UALUE FOR UNKNOUJN ug Fe/10 ml SOLUENT— ug Fe ( FAT + REAGENT BLANK) PEROHIDE UALUE AS m. eq. 02/ K^  FAT : NET ug Fe/10 ml q OF FAT USED X 55.84 FIGURE 3.8: Calculation of peroxide value. 39 u PROTEIN STANDARD SERUM RECONSTITUTED U J ITH 0.5 ml H 2 0 DILUTED UJITH ISOTONIC SALINE <X2,X4,X8,) 20 ul APPLIED ON I M M U N O P L R T E KEPT AT ROOM TEMPERATURE ( 2-3 DAYS ) DIAMETER OF PRECIPITATION MEASUAED CONTROL SERUM HUMAN M I L K S A M P L E S DILUTED UJITH ISOTONIC SRLINE ( X40 ) DILUTED UJITH ISOTONIC SALINE I X3 1 REST S A M E AS IN STANOAAD FIGURE 3.9: Immunoglobulin A analysis. 40 Immediately before analysis, the samples were thawed, total lipids were ex-tracted from 100/i£ of milk with chloroform-methanol (2:1, vol/vol) (Folch et al. 1957). Six millilitres of chloroform, 3 ml of methanol and 2 ml of saline containing 3 mM EDTA were added to 100/x£ of milk. Two hundred and fifty micrograms of internal standard Cn was added, the contents vortexed and centrifuged ~ 2000 RPM for 5 minutes. The bottom layer was filtered and the rest re-extracted with 6 ml of chloroform at a centrifugation speed of 3000 RPM. The bottom layer was again filtered out and added to the 1st extraction. The chloroform present in the solution was evaporated by subjecting it to a continuous stream of nitrogen gas. Fatty acid methyl esters were prepared with methanolic HCl (HCl:MeOH 1:5), by adding 1 ml of this mixture (HCl:MeOH) to the fat layer obtained after evaporation and digesting it for 10 minutes at 100° C. The contents were brought to room temperature and saline:pentane (3:6) was added, the tubes were vortexed and the top layer removed and evaporated under nitrogen until methylesters of fatty acids were left. These esters were dissolved and injected in the gas chromatograph equipped with a flame ionization detector (VARIAN Model No. 6000 GC). Helium was used as a carrier gas. All peaks were identified by comparison of their retention times with appropriate standards and their areas quantified with a vista 402, digital integrator. The procedure has been summarized in Figure 3.10. 3.3.5 Total Lipids and Protein Determination The method of Nakai and Chi (1970) was used to determine total lipids and protein in milk simultaneously (Figure 3.11). Five millilitres of 97% acetic acid was added to 0.05 ml of milk in a test tube closed with a ground-glass stopper and shaken to ensure complete solution. The protein was calculated from the absorbance measured at 280 m/i with a 1 cm cell against a reagent blank in a Beckman DB spectrophotometer. To this solution, 2.5 ml of urea imidazole solution (20% urea and 0.2% im-idazole) was added, the contents mixed thoroughly and allowed to stand for 30 minutes. Total lipid concentration was calculated by measuring the absorbance at 400 m/x with a round cell in a spectrophotometer. Calculations: 1. Cumulative Recovery of Total Lipids: 41 TOTRL FATTY RCID EHTRRCTION AND METHYLHT1 ON 100 ul M I L K AND 2 5 0 ug C 1 ? EHTRHCTED UJITH C AND 2 m l SFI H C L 3 / M e O H [ 6:3 ] LINE / EDTH FILTERED £ CENTRIFUGED ( 2 0 0 0 R P M ) RE-EHTRRCTED UJITH 6 ml CHCL, FILTERED CENTRIFUGED ( 3 0 0 0 R P M J EUAPORATED UJITH NITROGEN DIGESTED UJITH HCL : M e O H ( 1:5 ) EURPORHTED UJITH NITROGEN EHTRACTED UJITH SALINE : PENTANE ( 3:6 ) R E M A I N I N G METHVL ESTERS DILUTED UJITH HEHHNE AND INJECTED IN THE GAS CHROMHTOGRflM FIGURE 3.10: Fatty acids profile analysis. 42 (? M i n i OKI 0.05 ml M I L K + 5 m l HCETIC RCID ( 97% ) TTCinriDIL (L01P00D8 + 2.5 m l UREfl IMIDAZOLE UORTEKED READ AT 2 8 0 n m UORTEHED INCUBATED FOR 30 MINUTES RERD RT 400 n m FIGURE 3.11: Total lipids and protein determination. 43 % of total lipid in original milk samples recovered from infused milk samples. 2. Total Lipid Recovery: Lipid content of all aliquots (concentration of fat g/d£ x volume of aliquot). 3. Cumulative % Lipid Recovery: Lipid content in initial milk — total milk fat recovery. 4. Recovery of protein was calculated as per the calculations of total lipid re-covery. 44 C h a p t e r 4 R e s u l t s 4.1 Part I: Free Fatty Acids and Peroxides Results pertaining to the effects of processing and storage on the free fatty acids (FFA), and peroxide levels of milk will be presented in this section. Part la: Free Fatty Acids Results of the free fatty acid analysis of human milk are presented diagram-matically in Figure 4.1, and in tabular form in Table 4.1, 4.2, and 4.3. Figure 4.1 summarizes the overall effects of ultrasonication, lyophilization and freezing on the free fatty acid content of fresh and stored human milk samples. Generally, an in-crease in the free fatty acid (FFA) content was observed in the ultrasonicated milk before storage as well as after storage in comparison to the other treatment groups (Pasteurized 'A', Pasteurized and Ultrasonicated 'C'). The increase in free fatty acid level was most pronounced in the ultrasonicated group 'B' after lyophilization and subsequent storage. Results of one way analysis of variance performed on the baseline FFA values of the three treatment groups indicated that there was a significant (P < 0.001) treatment effect in the level of FFA in fresh milk (Table 4.1). A nonparametric one way analysis of Kruskal Wallis Rank test was also con-ducted on the baseline FFA values since the variance was not homogeneous, and that also indicated a significant treatment effect in the baseline FFA values of the three groups (A, B, C). Results of Scheffe's test further indicated that the levels of FFA in the ultrasonicated group 'B' were higher significantly at P < 0.01 than group A and group C (pasteurized A, and pasteurized and ultrasonicated C). There was not difference between group A and C. 45 Legend G2 Fresh • i Frozen: 1 month CS! Frozen: 4 months • Lyophilized: 1 month E 3 Lyophilized: 4 months o ^ I HT U S HT & U S Treatment Groups FIGURE 4.1: Effect of ultrasonication, lyophilization and freezing on the FFA content of fresh and stored human milk (mean of 7 samples). HT - heat treated, US - ultrasonicated, HT &; US - heat treated and ultrasonicated. 46 TABLE 4.1: Effect of ultrasonication, pasteurization, lyophilization and freezing on the free fatty acid content of fresh and stored human milk (group A, B, C). Fatty Acid Content (fj, equil./m/) Milk A B C Ultrasonicated Sample Pasteurized Ultrasonicated & Pasteurized F Fr. L F Fr. L F Fr. L 1 0.89 1 mo. 1.37 1.07 8.73 12.12 17.58 1.57 1.14 1.50 4 mo. 1.11 1.66 13.34 24.96 1.55 1.87 2 0.67 1 mo. 1.60 1.07 8.73 10.98 16.50 1.12 0.91 1.28 4 mo. 1.11 1.45 12.89 18.92 1.33 1.66 3 1.08 1 mo. 1.39 1.05 7.34 9.51 15.72 1.08 1.62 1.05 4 mo. 1.41 1.11 11.76 21.79 1.41 1.33 4 1.65 1 mo. 1.39 1.88 6.19 10.44 21.80 2.06 1.39 2.09 4 mo. 2.09 1.77 11.53 25.13 1.88 2.66 5 2.72 1 mo. 2.89 3.72 10.22 17.12 40.11 3.13 3.33 4.16 4 mo. 3.14 4.00 17.18 37.80 3.14 4.45 6 1.14 1 mo. 1.11 1.75 9.60 14.45 18.41 1.37 2.00 1.75 4 mo. 1.50 1.78 14.36 28.91 1.50 2.00 7 2.45 1 mo. 2.09 2.09 12.28 13.41 14.04 3.07 2.09 2.09 4 mo. 2.78 3.33 17.85 33.36 2.57 3.11 1 mo. 1.69 1.80 12.57 20.59 1.78 1.98 ± ± ± ± ± ± Mean 1.514 0.60 0.94 *9.013 2.63 8.9 1.914 0.80 1.03 ± ± ± SD 0.793 4 mo. 1.87 2.15 1.97 14.03 27.26 0.87 1.91 2.44 ± ± ± ± ± ± 0.793 1.07 2.34 6.58 0.68 1.07 * significantly higher than baseline (fresh values of Group A &; C (at P < 0.01). F = Fresh Fr. = Frozen L = Lyophilized 47 TABLE 4.2: Three way ANOVA on the increase in FFA and peroxides (F ratios). EFFECT FFA PEROXIDES Time (l mo./4 mo.) 6.53 * 319.01 *** Storage (frozen/lyophilized) 26.52 * * * 6.93 * Ultrasonication (U.S.) 41.68 * * * 14.19 * * + Two Way Interaction Time — Storage 2.91 ns 31.22 Time —Ultrasonication 0.99 ns 1.90 ns Storage — Ultrasonication 15.04 *** 2.43 ns ns not significant P < 0.05 *** P < 0.001 TABLE 4.3: Comparison of the increase in FFA and peroxides in the three treat-ment groups (A, B, C) after storage. A B C Pasteurized Pasteurized Ultrasonicated & Ultrasonicated FFAa 0.329 1.104* 0.097 Peroxides4 1.915 1.495f 1.853 a Mean fractional increase from baseline (n=7) 4 Mean log transformed increase from baseline (n=7) * Significantly higher than group A and C at P < 0.01 * Significantly lower than group A and C at P < 0.01 48 Table 4.1 presents concentrations of FFA in each milk sample undergoing differ-ent treatments (pasteurization and/or ultrasonication), fresh and after storage in frozen and/or lyophilized forms for different time periods. As can be seen from Ta-ble 4.1, there is considerable variation in the concentration of FFA in each group of fresh samples at baseline as well as after storage. However, several consistent trends are apparent, and revealed by the results of the three way analysis of variance (3-way-ANOVA). These results show that the increase in FFA levels are significantly affected by storage time, storage form, and the treatment provided to the milk samples (Table 4.2). Since we looked at the increase in the FFA levels from the baseline, there were only two levels of comparison, for the factor 'time' (1 month increase and 4 month increase), and for the factor 'storage form' (increase with freezing and increase with lyophilization). However, for the factor 'treatment' (pasteurized only, ultrasonicated only, pasteurized and ultrasonicated both), there were three levels of comparisons and therefore Scheffe's test was done to find out which two groups were different. Results of 3-way ANOVA (Table 4.2) indicate that there is a significant differ-ence (P < 0.005) in the increase of FFA between 1 month and 4 months storage. Looking at the mean values in Table 4.1 and Figure 4.1, it appears that the level of increase in FFA was higher in the 4 months storage group. Similarly, results presented in Table 4.1, 4.2, and Figure 4.1 indicate that the increase in FFA was significantly greater (P < 0.001) with storage in lyophilized form compared to frozen storage. The process of ultrasonication was found to have a significant effect on the increase in the FFA levels after storage (P < 0.001, Table 4.2). Results of Scheffe's test (Table 4.3) indicate that the increase in FFA was significantly higher (P < 0.01) with ultrasonication only (group B) compared to group A and C. Results of the 3-way ANOVA (Table 4.2) also indicate a significant 2 way in-teraction between storage form and the treatment: that is, the increase in FFA with lyophilized and frozen storage was confined to samples which had only been ultrasonicated (and not pasteurized). However, from the overall results in Table 4.1 and Figure 4.1, the level of FFA appear to be highest in the ultrasonicated and lyophilized milk stored for four months. 49 Part lb: Peroxide Levels Results pertaining to the peroxide content of human milk are presented graph-ically in Figure 4.2, and in the tabular form in Table 4.2 to Table 4.4. Figure 4.2 presents the overall effects of ultrasonication, lyophilization and freezing on the peroxide content of fresh and stored human milk. In general, the baseline or fresh values of peroxide in all three treatment groups (Pasteurized 'A', Ultrasonicated 'B', Pasteurized and Ultrasonicated 'C') appear to be similar. An increase in the peroxide content was observed after storage in both frozen and lyophilized forms in all treatment groups. Results of one way analysis of variance performed on the baseline peroxide values of groups A, B and C indicate that there was no significant different in the peroxide concentration of fresh milk given any of the three treatments (Table 4.4). However, when the fresh values of pasteurized and ultrasonicated groups were compared using paired t analysis, the concentration of peroxides in the ultrasoni-cated group appeared significantly higher (P < 0.05). Results of the 3-way-ANOVA indicates that the increase in peroxide value was significantly affected by storage time, storage form and by the treatment given to the milk sample (Table 4.2). Similar to the analysis of FFA, we looked at the increase in peroxide levels from the baseline. Therefore there were only two levels of comparison for the factor 'time' (1 month increase and 4 month increase) and for the factor 'storage form' (increase with freezing and increase with lyophilization). However, for the factor 'treatment' (pasteurized only, ultrasonicated only, pasteurized and ultrasonicated both), there were three levels of comparison and therefore Scheffe's test was performed to find out which two groups were different. Results presented in Table 4.2 indicate that there is a significant difference (P < 0.001) in the increase of peroxides between 1 month and 4 months storage. Looking at the mean values in Table 4.4 and Figure 4.2, it appears that the level of increase in peroxides was higher in the 4 months storage group. Similarly, ANOVA results presented in Table 4.2 indicate that the increase in peroxide value was significantly different (P < 0.05) when milk was stored in frozen or lyophilized form. Looking at the mean values in Table 4.4 and Figure 4.2, it appears that the levels of peroxide were higher in the lyophilized group at 1 month storage time and lower at 4 months storage time than the frozen group. The process of ultrasonication was found to have a significant effect on the increase in the level of peroxides after storage (P < 0.001) as shown in Table 4.2. 50 Legend rZZ Fresh mW Frozen: 1 month O Frozen: 4 months • Lyophilized: 1 month 68S Lyophilized: 4 months ^ 2 0 - i cr> \ I N O 15-o Treatment Groups FIGURE 4.2: Effect of ultrasonication, lyophilization and freezing on the peroxide content of fresh and stored human milk (mean of 7 samples). HT - heat treated, US - ultrasonicated, HT & US - heat treated and ultrasonicated. 51 TABLE 4.4: Effect of ultrasonication, lyophilization, pasteurization and freezing on the peroxide content of fresh and stored human milk (group A, B, C). Peroxide Content (millieq 02/kg fat) Milk A B C Ultrasonicated Sample Pasteurized Ultrasonicated & Pasteurized F Fr. L F Fr. L F Fr. L 1 0.18 1 mo. 0.60 7.16 0.45 0.43 1.44 0.11 0.25 3.49 4 mo. 11.89 6.87 8.16 6.16 11.10 7.59 2 0.18 1 mo. 0.68 7.16 0.36 0.60 2.16 0.24 0.60 6.63 4 mo. 13.68 12.60 9.96 6.87 12.80 12.60 3 0.50 1 mo. 2.76 3.85 1.25 2.40 2.69 1.25 2.58 3.40 4 mo. 14.90 9.02 13.32 6.87 13.32 9.74 4 2.36 1 mo. 1.96 3.67 2.20 1.79 2.29 2.47 2.76 3.04 4 mo. 14.19 7.59 12.60 4.01 12.60 6.87 5 0.96 1 mo. 1.59 1.95 1.50 1.48 1.30 1.14 1.59 2.67 4 mo. 9.74 8.30 10.45 8.30 9.02 8.30 6 1.68 1 mo. 2.67 5.21 2.94 1.77 2.60 1.14 2.67 4.67 4 mo. 13.32 16.30 11.89 12.60 10.45 17.04 7 1.14 1 mo. 1.76 7.02 1.40 1.65 2.15 1.22 1.76 6.80 4 mo. 8.87 11.89 7.45 6.87 7.45 11.89 1 mo. 1.70 5.14 1.44 2.08 1.74 4.38 ± ± ± ± ± ± Mean 1.0 0.85 2.06 1.44 0.69 0.53 1.08 1.01 1.00 ± ± ± SD 0.80a 4 mo. 12.36 10.36 0.91 10.54 7.38 0.77 10.96 10.57 ± ± ± ± ± ± 2.29 3.37 2.21 2.63 2.16 3.00 ° No significant difference among the fresh values. F = Fresh Fr. = Frozen L = Lyophilized 52 Results of Scheffe's test (Table 4.3) indicate that the increase in peroxides was significantly lower in the ultrasonicated group (P < 0.01%) compared to group A and C. Results of the 3-way-ANOVA (Table 4.2) also indicate a significant 2 way inter-action between storage time and storage form: that is, the increase in the peroxide value in the milk with time was dependent on the form in which it had been stored (frozen or lyophilized). From the overall.results for this section in Table 4.4, the maximum increase in peroxide appearance to be in the pasteurized milk frozen for 4 months. 4.2 Part II: Fatty Acid Profile Results pertaining to the effects of processing and storage on the fatty acid profile of human milk are presented in Figure 4.3, 4.4 and Table 4.5. Figure 4.3 illustrates the concentration of C12:0, C16:0, C18:0, C18:2n6 and Cl8:3n3 in fresh and processed human milk stored for 1 month. Results of multiple analysis of variance (MANOVA) conducted on these fatty acids are presented in Table 4.5. As can be seen from Figure 4.3 and Table 4.5, ultrasonication and subsequent storage appears to have a statistically significant lowering effect on the concentration of saturated fatty acids C12:0 and C18:0 at P < 0.001 and P < 0.01 respectively. Since the number of our samples was small (5), further post hoc tests were not conducted due to the possibility of accumulating error. However, the results of MANOVA do indicate a significant effect of treat-ments (ultrasonication and storage) on the concentration of C12:0 and C18:0. The concentration of C16:0 appears to decrease after various treatments (Figure 4.3) however, the differences were slight and statistically non-significant (Table 4.5). The concentration of essential fatty acids Cl8:2n6 and Cl8:3n3 appeared to remain unaffected by ultrasonication and subsequent storage (Figure 4.3). Results of the multiple analysis of variance (Table 4.5) indicate no significant treatment effect on fatty acids Cl8:2n6 and C18:3n3. Figure 4.4 presents the profile of long chain polyunsaturated fatty acids (LC-PUFA) in fresh and processed milk stored for one month. Overall, from the results of MANOVA (Table 4.5), ultrasonication and sub-sequent storage did not seem to significantly affect the percentage of C20:3n6, C22:5n3, and C22:6n3. However, fatty acids C20:4n6 and C22:4n6 were found to be decreased significantly upon ultrasonication and storage (P < 0.001). The per-centage of fatty acid C20:5n3 was also significantly affected (P < 0.05) after the 53 FRESH FRESH-FROZEN U S—FROZEN U S -LYPOPHIUZED C12:0 C16:0 C18:0 C18:2n6 FATTY ACID PROFILE C18:3n3 FIGURE 4.3: Fatty acid profile in fresh, processed and stored (1 month) human milk (mean ± SD of 5 samples). Fatty acid content is expressed as % of total fatty acids. US - ultrasonicated. 54 2.5 C20:3n6 C20:4n6_ C20:5n3 C22:4n6 C22:5n3 C22:6n3 FATTY ACID PROFILE FIGURE 4.4: Long chain polyunsaturated fatty acid profile in fresh, processed and stored (1 month) human milk (mean ± SD of 5 samples). Fatty acid content is expressed as % of total fatty acids. US - ultrasonicated. 55 given treatments does not appear to decrease. On the other hand, the levels of C20:5n3 do appear to rise in the fresh frozen samples but on a purely observational basis. Overall, ultrasonication and storage of human milk appears to bring about a decrease in the LC-PUFA of w-5 series with four double bonds whereas the concentration of w-3 fatty acids remain mostly unaffected. Treatment of human milk (ultrasonication, freezing, lyophilization) per se, seems to affect only two out of the six LC-PUFA; C22:4n6 and C20:4n6. TABLE 4.5: Multiple analysis of variance on the fatty acids profile. Fatty Acid F Values C12:0 29.40 P < 0.001 C16:0 1.46 ns C18:0 7.09 P < 0.01 C18:2n6 0.31 ns C18:2n3 0.12 ns C20:3n6 2.11 ns C20:4n6 9.04 P < 0.001 C20:5n3 5.24 P < 0.05 (t) C22:4n6 15.95 P < 0.001 C22:5n3 0.42 ns C22:6n3 0.14 ns ns - not significant 56 4.3 Part III: Immunoglobulin A (IgA) Results pertaining to the effect of processing and storage on human milk IgA concentrations are presented in Table 4.6. Table 4.6 presents the results of ultrasonication, lyophilization, freezing and storage (for one month) on the IgA concentration of each milk sample as well as the average. The average (mean) concentration of IgA in fresh milk was 46.05 mg/d£ which decreased to 45.48 mg/d£ after ultrasonication. Results of paired t test analyses showed that the reduction fo 0.57 mg of IgA after ultrasonication was significant at P < 0.05 (Table 4.6). Storage of ultrasonicated milk in the frozen form further decreased the IgA concentration by an average of 0.91 mg. This change in the concentration of IgA brought about by freezing was found to be significant using paired t test analysis P < 0.01). Storage of ultrasonicated milk in the lyophilized form compared to the frozen form further decreased the IgA concentration by an average of 2.00 mg. This decrease was found to be significant at P < 0.001 (Table 4.6), using paired t test analysis. Overall concentration of IgA in ultrasonicated milk appears higher in all the samples when stored frozen rather than lyophilized for one month. 4.4 Part IV: a) Total Fat and Protein Recovery-Results pertaining to the fat and protein content in processed and stored human milk infused at slow and rapid rates are graphically presented in Figure 4.5 to Figure 4.10 of this section. All values appearing in the figures are mean and standard deviations of eight samples. Part IV: Fat Recovery Figure 4.5 graphically present the periodical and cumulative changes in fat con-tent of ultrasonicated, frozen and lyophilized milk stored for one and four months each and mechanically infused at a slow rate of 10 ml/h. Figure 4.5 presents the mean fat content and standard deviations of eight milk samples collected at thirty minute intervals during slow mechanical infusion. In general, the recovery of fat from ultrasonicated and frozen milk appears to be significantly more complete than in the ultrasonicated milk stored in lyophilized 57 TABLE 4.6: Effect of ultrasonication, lyophilization, freezing and storage on IgA concentration of human milk. IgA concentration (mg/d£) Changes ' US Changes US Changes Milk Fresh US with and with and with Sample US Freezing Freezing Lyoph. Lyoph. 1 40.4 38.8 1.6 38.8 0.0 37.6 1.2 2 46.8 46.8 0.0 45.6 1.2 44.4 1.2 3 48.0 46.8 1.2 45.6 1.2 41.6 4.0 4 38.8 38.8 0.0 37.6 1.2 34.8 2.8 5 60.0 60.0 0.0 58.8 1.2 57.6 1.2 6 28.4 28.4 0.0 26.4 1.6 24.4 2.4 7 60.0 58.8 1.2 58.8 0.0 57.6 1.2 Mean 46.05 45.48 0.57* 44.57 0.91** 42.57 2.0*** SD ± 11.47 ± 11.34 ± 0.72 ± 11.58 ± 0.64 ± 12.05 ± 1.10 P < 0.05 US = Ultrasonicated ** P < 0.01 Lyoph. = Lyophilization *** P < 0.001 58 Lyophilized Frozen STORED FOR 1 MONTH O o r— < N r rr-rTr! t— -z. LU O o Lyophilized Frozen STORED FOR 4 MONTHS i \ \ I T LU > o O LU CC 111 > I— O INFUSION TIME (h) FIGURE 4.5: Mean (± SD) periodical and cumulative changes in fat content of ultrasonicated-frozen and ultrasonicated-lyophilized milk infused at a slow rate (10 ml/h). * Differences from zero time concentration significant (P < 0.01). 59 form (for both one and four months). In ultrasonicated and lyophilized milk stored for one and four months (Fig-ure 4.5), fat content fell significantly in the first thirty minutes of infusion and kept declining to three and a half hours. During the final thirty minutes of infusion there was a significantly sharp rise in the fat concentration. On the other hand, in ultrasonicated milk stored frozen for both one and four months, the concentration of fat in all but the final few hours infusate appeared relatively constant. Cumulatively, 87.5% and 74.7% of the initial fat content was recovered in the ultrasonicated and lyophilized milk stored for one and four months respectively. On the contrary, the cumulative recovery of fat in ultrasonicated milk frozen for one and four months was 99.2% and 97.7% respectively. Figure 4.6 illustrates the mean periodical and cumulative changes in fat content of ultrasonicated frozen and ultrasonicated lyophilized milk during thirty minutes of rapid infusion (40 ml/h). A slight decline was observed in the fat concentration after 24 minutes (vs. 0 time) of infusion in ultrasonicated and lyophilized milk stored for four months. However, compared to slow infusion technique, delivery of fat in the rapidly infused milk samples (ultrasonicated and lyophilized for one and four months) appeared much more stable (Figure 4.6). In the frozen stored ultrasonicated milk there was no decline in the concentration of fat at any infusion time. Cumulatively 93.54% and 84.28% of initial fat content was recovered in the ultrasonicated and lyophilized milk stored for one and four months respectively. In the ultrasonicated frozen samples, 102.5% and 98.5% of the fat present in the initial aliquot was recovered after one and four months storage respectively (Fig-ure 4.6). The difference between the frozen and lyophilized samples was significant at both one and four months (P < 0.05). Part IV: Protein Recovery Figure 4.7 graphically presents the periodical and cumulative changes in protein content of ultrasonicated frozen and ultrasonicated lyophilized milk stored for one and four months each and infused mechanically at a slow rate (10 ml/h). In the frozen and lyophilized samples stored for one month there appeared to be more fluctuations in the recovery of protein compared to the samples infused after four months storage. However, none of the changes differed significantly from baseline values. 60 Lyophilized Frozen STORED FOR 1 MONTH <N -cc LU > o O LU cc LU > I -D O 1 2 1 8 2 4 30 Lyophilized frozen STORED FOR 4 MONTHS r--r ' i " ' " T i cc LU > o O LU CC LU > I--2 o 0 6 1 2 1 8 2 4 3 0 I N F U S I O N T I M E (min) FIGURE 4.6: Mean (± SD) periodical and cumulative changes in fat content of ultrasonicated-frozen and ultrasonicated-lyophilized milk infused at a rapid rate (40 ml/h). 61 5 UJ 1-•z. O O LU o cc a. IT) cvi o oi Lyophilized Frozen STORED FOR 1 MONTH T T o >-DC LU > o o LU DC LU > Z) o in d LU O o LU 1-o rx CL in oi o oi Lyophilized Frozen STORED FOR 4 MONTHS >-DC LU > o o LU cc LU > 5 ID ZD O m d 1 2 3 INFUSION TIME (h) FIGURE 4.7: Mean (± SD) periodical and cumulative changes in protein content of ultrasonicated-frozen and ultrasonicated-lyophilized milk infused at a slow rate (10 ml/h). 62 The cumulative recovery of protein in the samples stored for one month was slightly higher in the lyophilized section (94.5%) compared to frozen (91.9%). On the other hand, the cumulative recovery of protein was same (98.9%) for both frozen and lyophilized samples stored for four months. In general, the ul-trasonicated frozen samples appear more consistent in terms of protein recovery at various infusion intervals compared to the lyophilized samples, although the differences were not significant. Figure 4.8 illustrates the mean periodical and cumulative changes in protein content of ultrasonicated frozen and lyophilized milk during thirty minutes of rapid infusion of 40 ml/h. In general, both frozen and lyophilized milk samples show slight fluctuations in protein recovery at various intervals when infused after one month storage. After four months, the concentration of protein in all the infusates of both frozen and lyophilized form was extremely stable. The cumulative recovery of protein in rapidly infused milk was very similar for both frozen and lyophilized samples at one month and four month storage levels (93.9% and 94.5%, 99% and 101%) respectively. 4.5 Part IV: b) Total Fat and Protein Recovery Figure 4.9 presents the results pertaining to the recovery of fat and protein content in milk ultrasonicated after lyophilization and both before and after lyophilization and infused at a slow rate. Milk samples in both cases were stored for one month in the lyophilized state. The periodical recovery of both fat and protein appears similar and consis-tent throughout the infusion time of four hours for milk ultrasonicated once after lyophilization as well as for milk ultrasonicated twice: before and after lyophiliza-tion. In both these groups, the final recovery of fat as well as protein also appears similar (98.5% and 99.6%, 99.5% and 102.8%) respectively. Figure 4.10 illustrates the mean changes in fat and protein content of milk (ul-trasonicated after lyophilization,a nd ultrasonicated both before and after lyophiliza-tion) during thirty minutes of rapid infusion. As was observed in the slow infusion study, the periodical recovery of fat and protein appears similar throughout the infusion time of thirty minutes in milk ultrasonicated once after lyophilization and in milk ultrasonicated twice; before and after lyophilization. Both these groups (ultrasonicated once and twice) appear 63 Lyophilized Frozen STORED FOR 1 MONTH 5 r-z LU O o LU o QC CL O I X cc LU > o o LU CC LU > I— ZD ZD O LO O 12 24 30 5 LU (-O O LU I-o cr a. LO OJ o oi Lyophilized Frozen STORED FOR 4 MONTHS I -i-cc LU > o o LU CC LU > I— 5 ZD ZD O in d 6 12 18 24 30 INFUSION TIME (min) FIGURE 4.8: Mean (± SD) periodical and cumulative changes in protein content of ultrasonicated-frozen and ultrasonicated-lyophilized milk infused at a rapid rate (40 ml/h). 64 5 t --z. LU h-o o LU r-o cc Q. oi o oi d U S before & afler U S after  T 1 cc LU > O O LU cc LU > I— ZD O T3 LU 1-O o U S before & after U S after T 1 > cc LU > o o LU CC LU > 1-2 0 1 2 3 4 INFUSION TIME (h) FIGURE 4.9: Mean (± SD) changes in fat and protein content of milk ultrasoni-cated after lyophilization and storage and both before and after lyophilization and storage and infused at a slow rate using mechanical pump. 65 T3 LU I— z o o LU I-o DC CL oi o CM in d U S belore & alter U S after T -1 2 1 8 24 I 3 0 > GC LU > o U UJ GC LU > •a cn UJ O O 1-< U S before & alter U S after (• — — - — - . . I J o 6 1 2 1 8 2 4 3 0 INFUSION TIME (min) >-GC LU > o O LU GC LU > FIGURE 4.10: Mean (± SD) changes in fat and protein content of milk ultrasoni-cated after lyophilization and storage and both before and after lyophilization and storage and infused at a rapid rate using gravity flow technique. 66 to have similar cumulative recovery of fat (98% and 96%) and protein (97.6% and 99%) respectively (Figure 4.10). In general, ultrasonication of milk once after reconstitution of lyophilized milk appears to produce results (recovery of fat and protein) similar to when milk was ultrasonicated twice; both before and after lyophilization. 67 C h a p t e r 5 D i s c u s s i o n The purpose of this study was a) to determine the effects of processing (i.e. ul-trasonication, heat treatment, lyophilization) and storage on free fatty acid levels, fatty acid profile and IgA content in human milk and b) to determine the effects of processing and storage on the recovery of fat and protein in mechanically in-fused human milk. Each of the above issues will be discussed individually and then followed by a concluding summary. 5.1 Effect of Processing and Storage on the Free Fatty Acid and Peroxide Levels in Human Milk 5.1.1 Part la: Free Fatty Acids (FFA) In the present study a significant increase in the level of FFA was observed fol-lowing ultrasonication. Storage of ultrasonicated milk was associated with further increases in the FFAs. However, increase in FFAs observed with storage was limited to milk that was only ultrasonicated (i.e. not in milk that was both ultrasonicated and pasteurized. An increase in free fatty acids could be due to enzymatic lipolysis (Hall 1980, Hall 1982, Hernell et al. 1974 and Hernell et al. 1982) or to release of fatty acids from the fat globules upon agitation which might occur during freezing and thawing (Wardell et al. 1981, Berkow et al. 1984, Wardell et al. 1984). Vigorous agitation would also occur with ultrasonic vibration which creates high pressure areas in the milk thereby rupturing the fat globules (Martinez et al. 1987). Although disruption of the fat globules would possibly increase its exposure to the lipolytic enzymes present in breast milk thereby increasing the FFA levels, this observation 68 has been limited to nonpasteurized (non heat treated) milk (Wardell et al. 1981, Berkow et al. 1984). The above findings relate very well with the results of the present study where no increase in FFAs was found after storage of pasteurized milk, even though it went through the mechanical action of ultrasonication, lyophilization and freeze thaw cycle (Group 'C, Table 4.1), that would rupture the fat globule and increase its surface area. On the other hand, milk that was only ultrasonicated did give rise to more FFA immediately after storage (group 'B', Table 4.1, Figure 4.1). This suggests that enzymatic hydrolysis is the 'critical' factor for FFA formation. The mechanical action of ultrasonication does appear to render the triglyceride more susceptible to enzymatic lipolysis, however, if the enzymes are inactivated, which is known to occur upon heat treatment (Sunshine 1980, Wardell et al. 1984), lipolysis does not appear to happen. The increase in FFA upon ultrasonication and lyophilization (Figure 4.1) could be due to the increase in BSSL activity because of the uncontrolled moisture content of the samples and warm storage temperature. One important question that arises here is whether the increase in FFA observed with ultrasonication of milk in the absence of heat treatment is advantageous or disadvantageous to the newborn. The immaturity of pancreatic and hepatic function in the newborn and espe-cially the premature (Hamosh et al. 1981, Hernell et al. 1982) suggests that they may depend on extrapancreatic lipases for adequate fat digestion. Of special im-portance is the Bile Salt Stimulated Lipase (BSSL) supplied with the milk. BSSL hydrolyzes triglycerides to FFA and glycerol whereas the action of pancreatic lipase results in release of FFA and 2 monoglycerides. Of importance here is the intraluminal bile salt concentration which has been repeatedly shown to be much lower in newborn infants and especially preterm in-fants than the older infants or adults (Hernell and Blackberg 1983). an important function of the bile salts is to displace the products of lipolysis from the surfaces of the lipid droplets, to the site of action of pancreatic lipase. In the intraluminal wa-ter phase, bile salts then solubilize the products into mixed micelles. Therefore, fat absorption is considered to be dependent on bile salt concentration. The solubility of monoglycerides in an aqueous phase shows an especially strong dependency on bile salt concentrations (Hernell and Blackberg 1982). Since bile salt stimulated lipase releases glycerol, which is completely water soluble, it will rapidly, indepen-dent of bile salt concentration, be released in the aqueous phase for absorption and also won't be available for any esterification process (catalyzed by pancreatic 69 lipase) (Hernell and Blackberg 1983, Hamosh 1983). Since fatty acids are the major product formed by BSSL an essential question is whether they are less dependent on the bile salts for absorption than are the monoglycerides. That this in fact might be so is supported by the observation that under other conditions of low intraluminal bile salt concentration, such as bile fisula rats, fatty acids are more efficiently absorbed than monoglycerides (Morgan et al. 1969). This would also agree with the results of serial metabolic studies carried out in preterm infants which show that pasteurization (heat treatment) of human milk reduces fat absorption by a third (Atkinson et al. 1981, Williamson 1978). This reduction could be explained on the basis that heat treatment denatures BSSL thereby reducing the availability of FFA to the newborn's system. Another question that could be raised is whether FFA might bind to calcium or other components of milk thereby making them unavailable for absorption. This question has been studied by Lavine and Clark (1987). They found that the longer the milk was stored and the warmer the storage temperature, the more evidence of ionized fatty acid. However, these were always present in small quantities. The greatest quantity of ionized fatty acids observed was in milk stored for 8 weeks at -11° C. This milk on the average contained 2.2 mg/d£, which was ~ 1.4% of the total free fatty acids found in the milk. Therefore, the formation of ionized fatty acids was not considered a major factor (Lavine and Clark 1987). On the other hand, products of lipolysis, i.e. the free fatty acids produced by the naturally occurring enzymes in human milk, have been found to destroy Giardia lamblia, a protozoan parasite causing Giardiaris which is a frequent cause of diarrhea in children (Nutrition Reviews 1987), but not frequent in newborns. So far there is no evidence regarding what should be a safe upper limit of the FFA present in milk that has to be fed to the newborn. From this study it can be concluded that ultrasonication of human milk results in significant increase in the free fatty acid level. The increase persists during storage of ultrasonicated milk in any form if not pasteurized. This might be beneficial for the baby, however, further studies, particularly on feeding, are needed. 5.1.2 Part lb: Peroxide Levels Milk is especially susceptible to oxidative deterioration, and since antioxidants cannot be added to milk, the effect of processing on the oxidative stability of milk has received much attention in the dairy industry (Taylor et al. 1980). However, there is little information regarding oxidative stability in processed human milk stored in human milk banks. 70 In the present study the effect of processing (ultrasonication, pasteurization, lyophilization, freezing) and storage on the extent of oxidation in human milk was studied by measuring the peroxide levels (method of Stine et al. 1954). Peroxides are the main initial products of autoxidation. They can be measured by techniques based on their ability to oxidize ferrous to ferric ions (which has been used here). The reliability of this test depends mostly on the accuracy of the procedure. Good correlations are sometimes obtained between peroxide values and the development of off flavours but sometimes the results are also inconsistent. However, no single test can possibly measure all the oxidative events at once and no single test can be equally useful for all stages in the oxidative process (Small 1986). The results of this study suggest that the overall increase in peroxide level of breast milk tends to be affected mostly by the length of storage. Ultrasonication alone (without pasteurization) seems to have a significant lowering effect on the increase of peroxides, more specifically in the lyophilized milk. Since the FFA were highest in the ultrasonicated and lyophilized samples, it was expected that there would be a greater possibility of oxidation and therefore a higher peroxide level (Campbell et al. 1979). However, as seen in Figure 4.2, and Table 4.3, the peroxide level in the samples that were ultrasonicated (but not pasteurized) was significantly lower after storage compared to the samples that were pasteurized (Groups A and C). It is possible that ultrasonication might rearrange the components of milk such that the fat layer could be coated by a protein layer thereby preventing oxidation of fat and resultant increase in peroxides. In the study by Taylor et al. (1980), treatment of skim milk with ultrasonic vibration greatly increased its antioxidant activity in a linoleate emulsion. Sonication had the most effect on the casein fraction of milk and less on whey proteins. The effect of sonication was not due to changes in sulphydryl content or reactivity. It was suggested that sonication probably exerted its antioxidant effect by dis-rupting casein micelles, increase the effective concentration of casein resulting in physical protection of the linoleate emulsion droplets from lipid peroxidation. Un-sonicated casein was potent antioxidant in this system and the disruption of micelles could have accounted for the increased antioxidant activity of sonicated casein. It is possible that something similar to that effect was occurring in our samples. Irrespective of the processing, the increase in peroxide after one month of freez-ing was minimal, however after four months of freezing a significant increase in peroxides was seen. The ultrasonicated and lyophilized milk appeared to have the smallest increase in peroxides after four months storage. However, this lowering effect of ultrasoni-71 cation disappears when milk was also pasteurized. Therefore, it appears that heat treatment in some way interferes with the antioxidant effect of ultrasonication. The lyophilized milk samples appeared to show smaller increases in peroxide value than the frozen samples. It could be possible that storage of lyophilized milk under nitrogen gas was preventing the milk from being oxidized (Campbell et al. 1979). Overall ultrasonication did not appear to bring about an increase in the peroxide level greater than the conventional milk bank processes, such as pasteurization, but appeared instead to prevent excess peroxide formation. However, this protective effect of ultrasonication was not observed in milk undergoing both pasteurization and ultrasonication. It can be concluded that in pasteurized E H M , ultrasonication does not further increase the level of peroxides upon storage whereas in unpasteur-ized E H M , ultrasonication may actually protect against peroxide formation upon storage. 5.2 Part II: Fatty Acid Profile in Processed and Stored Human Milk The overall results of this study suggest that ultrasonication and subsequent storage of human milk in the frozen and/or lyophilized form for one month produces a significant lowering effect on the levels of saturated fatty acids; C12:0, and Cl8:0. A trend towards the decrease in the concentration of C16:0 (although not significant) was also observed after ultrasonication and freezing. The results partially agree with the study of Lavine and Clark (1987) where they observed a decrease in the proportion of C12:0, C14:0, C16:0 and C18:0 in the total free fatty acid pool with storage of 8 weeks at -11° C. Since in the present study total fatty acids were measured, the decrease in saturated fatty acids could be in the phospholipid, triglyceride or cholesterol fraction of the fatty acids. Wardell et al. (1982) have shown a decrease in C16:0 and C18:0 of the triglyceride section in human milk when frozen at -20° C and thawed (without storage), which might suggest that the decrease observed in our study could be due to a decrease in the triglyceride section of saturated fatty acids also. Nevertheless, the most important essential fatty acids, Cl8:2n6 and Cl8:3n3, were not adversely affected by ultrasonication and frozen or lyophilized storage, as per the results of this study. Since we looked at total fatty acid concentrations, it is not possible to directly compare our results with those of Lavine and Clark (1987) where the percentage of C18:2 in the free fatty acid pool greatly increased with 72 storage. However, since the results of Part la of the study suggest an increase in the concentration of free fatty acids after ultrasonication and storage, it is possible that an increase in the concentration of C18:2 could have contributed toward that. However, further studies are required to identify the individual fatty acids respon-sible for causing the observed increase in the concentration of free fatty acids after ultrasonication and storage (Part la). Wardell et al. (1982) have reported a loss of C18:2 from the milk triglycerides upon freezing and thawing which they speculate could be due to autoxidation. However, since our results do not show an overall loss in C18:2 (in total fatty acid fraction) it is possible that the essential fatty acid C18:2 was not actually lost due to autoxidation, but instead appeared in the free fatty acid pool which (as suggested by Hernell et al. 1982), might further facilitate its absorption in the newborn. In general, from the present study, ultrasonication and storage of ultrasonicated milk in both frozen and lyophilized forms appears safe in terms of conserving the fatty acids (C18:2n6 and Cl8:3n3) essential for development of infant brain and visual tissues. A number of biologically active long chain polyunsaturated fatty acids (LC-PUFA) are derived from these two main dietary essential fatty acids 18:2n6 and 18:3n3 (Koletzko et al. 1988). These LC-PUFA are: C20:4n6, C22:4n6, C20:3n6, C22:5n3, C22:6n3 and C20:5n3. These LC-PUFA are physiologically important as precursors for the synthesis of prostaglandins and structural lipids (Nutrition Reviews 1984). So far there is little information regarding the capability of infant tissue to elongate Cl8:2n6 and Cl8:3n3. The human fetus is believed to be dependent upon placental synthesis and transfer of LC-PUFA which are also not synthesized at maximal rates until five weeks postnatally (Nutrition Reviews 1984). Human milk or formulas then become the primary, perhaps the sole source of LC-PUFA for the young infant, and in particular the premature. Processing (ultrasonication) and storage of human milk for one month in frozen or lyophilized form appears to retain most of the LC-PUFA; C20:3n6, C20:5n3, C22:5n3 and specifically C22:6n3 which is a brain structural fatty acid considered very essential for the neonate (Putnam et al. 1982). A significant increase in the levels of C22:5n3; an important precursor for synthesis of structural lipids (Nutrition Review 1984) was observed in the unprocessed EHM frozen for 1 month. This partially agrees with the results of Lavine and Clarke (1987) where an increase in C22:5 was observed with frozen storage at -11° C for 1 month and more. On the other hand, C20:4n6, a precursor of the eicosanoids (Putnam et al. 73 1982), and C22:4n6, involved in the synthesis of structural lipids, (Nutrition Re-views 1984) were found to be reduced with ultrasonication and storage. It is nev-ertheless interesting to note that even the reduced concentrations of C20:4n6 and C22:4n6 were found to be within the normal range of these fatty acids observed in fresh mature human milk (Koletzko et al. 1988, Harzer et al. 1983). In general, a lot of variation in the concentration of individual fatty acids was observed in different milk samples. As suggested by Koletzko et al. 1988, the interindividual variation in LC-PUFA content could be due to the interindividual differences in the capacity of the desaturating and chain elongating pathway and/or incorpora-tion process of the LCP metabolites into milk lipids. Those observed variations in the fatty acid concentrations of different milk samples could also be explained by the differences in the dietary intake of mothers (Kneebone et al. 1985, Finley et al. 1985). However, there still is little work examining the effect of processing and storage on the majority of LC-PUFA now detected in human breast milk. This study presents a variety of processing and storage effects on a number of essential and long chain fatty acids in breast milk. Although more information would certainly be required on the above aspect, this study could be treated as a probe into the effects of processing and storage on important fatty acids in human milk. Overall, the results of the present study do suggest that ultrasonication and storage of milk in frozen or lyophilized form for one month does not result in any loss of essential fatty acids and also retains most of the LC-PUFA in human breast milk. 5.3 Part III: Immunoglobulin A (IgA) Concen-tration in Process and Stored Human Milk Of all Igs, the need for Immunoglobulin A (IgA) in the infants seems most apparent. The infant indeed lacks this immune factor and exposure to it soon after birth is essential to guard against certain bacterial and viral diseases (Packard 1982). Although all major classes of immunoglobulins exist in human milk, secretory IgA is found in highest concentrations in all phases of lactation (Ogra et al. 1979, Hanson et al. 1977). A large variation generally exists in the IgA concentration of breast milk obtained from different individuals (Lewis et al. 1985). This was also evident from the very high standard deviations present in fresh milk samples obtained in this study (Table 4.6). Processing of human milk has been shown to reduce the level of IgA in banked human milk (Packard 1982). Two investigators have reported a decrease of 22% 74 and 33% in IgA concentration in milk when it was pasteurized by heating to 62.5° C for 30 minutes (Ford et al. 1977, Liebhaber et al. 1977). So far there have been no reports published on the effects of ultrasonication on IgA levels in human milk. Our data suggests a 1.21% loss of IgA by ultrasonication. This loss was found to be statistically significant (P < 0.05), but the absolute amount of loss was small (1.21%) and much less than by pasteurization. With a retention of 98.8% of IgA, ultrasonication does not appear to be relatively safe processing technique in terms of conserving the most important of the immunoglobulins in human breast milk. Freezing the ultrasonicated milk for one month resulted in a statistically signif-icant loss of IgA (Table 4.5). However the absolute amount of IgA lost was again small (2.27%). Liebhaber et al. (1977) reported no loss of IgA after one month's frozen storage and Evans (1978) reported the same after 3 months frozen storage. Considering their results, it could be beneficial to freeze unprocessed milk and then thaw and ultrasonicate when needed. However, at the same time, we are faced with the possibility of infectious disease agents being present and multiplying in the un-processed milk. Pasteurization is seen as a realistic and necessary alternative by the Food and Drug Administration (FDA) to prevent the infectivity of disease agents in human milk, such as the AIDS virus (Wallingford 1987). However, the process of pasteurization also has very significant detrimental effects on the IgA levels in milk (33% loss, Liebhaber et al. 1977). This loss is much more than observed in the process of ultrasonication (1.2%), however, the effects of freezing, thawing, pasteurizing, length of storage and ultrasonication can be cumulatively high. At the same time, it has also been pointed out that formulas do not contain many of the unique factors destroyed by these processes, and even a reduced amount of the components such as IgA (in this case) is better than nothing (Arnold 1988). Results of lyophilization on IgA levels have been contradictory in the past lit-erature. Liebhaber et al. (1977) showed a 21% decrease in IgA with lyophilization whereas Evans et al. (1978) showed no significant change. Results of the present study indicated a significantly greater loss of IgA in ultrasonicated milk when it was lyophilized and stored compared to frozen storage. The decrease in IgA after lyophilization was significant in this study which was consistent with results of Liebhaber (1977). However, the intensity of loss of IgA was less, 6.39% compared with 21% loss as seen in the other study (Liebhaber 1977). These differences could be explained by the fact that in our study milk was lyophilized after ultrasonication whereas Liebhaber et al. lyophilized raw milk. However, the reasons for change in intensity of losses due to ultrasonication are not known. Since sonication of milk has been reported to result in a diversity of 75 physiochemical changes in macro-molecules, including enzyme inactivation (Azhar et al. 1979), and casein disruption (Taylor et al. 1980), it could possibly have an effect on rupturing in the IgA molecule. In conclusion, ultrasonication of human milk results in small but statistically significant loss of IgA. However, compared to the losses observed in existing pro-cesses such as pasteurization (33% loss, Liebhaber et al. 1977), these losses do not appear important. In terms of preservation of the IgA content, storage of ultrason-icated milk in the frozen form was found to be more beneficial than the lyophilized form. 5.4 Part IV and Part IVa: Processing and Stor-age Effects on the Fat and Protein Contents of Human Milk Infused Mechanically and by Gravity Flow It has been shown by several investigators (Stocks 1985, Narayanan et al. 1984, Brook et al. 1978, Spencer et al. 1982, Greer et al. 1984, Martinez et al. 1987) that continuous mechanical infusions at slow rates are associated with poor recovery of fat due to fat lost on the tube walls, syringe or burette. Although Martinez et al. (1987), in their studies, had shown that ultrasonica-tion of human milk prevents fat loss that occurred during tube feeding of untreated human milk (Figure 1.1), it was still now known whether storage of ultrasonicated milk would in any way affect the advantages of ultrasonication in terms of better fat recovery. Results of the present study suggest that storage of ultrasonicated milk in the frozen form for up to four months has least effect on fat recovery when infusing milk at a slow rate (using mechanical infusion pumps) or rapid rate (using gravity flow). On the other hand, storage of ultrasonicated milk in the lyophilized form for both one and four months appears to result in significantly lower fat recovery than with frozen stored milk. These results suggest that the advantages of ultrasonicating milk seem to be lost to some extent with lyophilized storage. It is possible that the process of lyophilization brings together the fat globules and other larger molecules in milk (i.e. reversing homogenization effects) and thereby leading to poor fat recovery. However, this problem was easily resolved by ultrasonicating milk after it had been stored in the lyophilized form and reconstituted for use. This clearly suggests that 76 reconstituted milk can be homogenized by ultrasonication just like fresh milk. The above mentioned effects are clearly shown in Figure 5.1 and Figure 5.2 illus-trating fresh (a), fresh-lyophilized (b), ultrasonicated-lyophilized (c), and lyophilized-ultrasonicated (d) milks. Milk that was ultrasonicated after lyophilization (d) had more numerous, more uniform and smaller fat globules/mm2 compared to milk that was ultrasonicated before lyophilization (c). This clearly suggests that for the maximum recovery of fat from lyophilized milk, it should be ultrasonicated after it has been stored and reconstituted for use. As was seen with the recovery of fat, frozen storage of ultrasonicated milk (for both one and four months) resulted in excellent recovery of protein. Lyophiliza-tion and subsequent storage of ultrasonicated milk, however, resulted in slight fluctuations in protein recovery during slow infusion. This effect could be due to some experimental error, but it is also possible that the coalescence of fat glob-ules brought about by lyophilization might also entangle some protein molecules thereby preventing complete protein recovery. Whether or not this happened, ul-trasonicating milk after storage in lyophilized form did result in 100% recovery of protein (particularly in slowly infused milk). Overall, it appears that if milk has to be stored in the lyophilized form, then ultrasonicating it just before use would give better fat recovery. On the other hand, if milk has to be frozen, then ultrasonicating it before storage results in excellent recovery of both fat and protein. 77 (a) Fresh Human Milk (80 fat/globules/mm2) (b) Fresh-Lyophilized Human Milk (64 fat/globules/mm 2) F I G U R E 5.1: Fat globules in fresh and fresh-lyophilized human milk. 78 (c) Ultrasonicated and Lyophilized Human Milk (106 fat globules/mm2) (d) Lyophilized and Ultrasonicated Human Milk (150 fat globules/mm2) FIGURE 5.2: Fat globules in human milk ultrasonicated both before and after lyophilization. 79 C h a p t e r 6 S u m m a r y o f F i n d i n g s Concluding statements regarding each part of the study are presented in this sec-tion: Part I: • Ultrasonication of human milk was found to increase the free fatty acid con-tent of unpasteurized milk. The increase persisted during storage of ultra-sonicated milk in both frozen and lyophilized forms for four months. This increase in the FFA levels of milk could be beneficial in the digestion and absorption of milk fat by the newborn. • Ultrasonication of human milk did not bring about an increase in peroxide value greater than the conventional processes, such as pasteurization. It may in fact protect against excess peroxide formation in pasteurized milk. Part II: • Ultrasonication and storage of human milk in frozen and/or lyophilized form for one month did not result in significant loss of essential fatty acids and also retained most of the long chain polyunsaturated fatty acids. Part III: • Ultrasonication and subsequent storage of human milk in frozen and/or lyophilized form resulted in small but statistically significant loss of IgA. However, com-pared to existing milk banking processes, such as pasteurization, these losses appeared extremely small. 80 Part IV(a): • The recovery of fat and protein in ultrasonicated milk infused by mechanical pump and gravity flow was not adversely affected by frozen storage for up to four months. • The recovery of fat in ultrasonicated milk was significantly decreased when infused after being stored in lyophilized form for both one and four months. Part IV(b): • The recovery of both fat and protein was more than 96% in samples that were ultrasonicated and infused after being stored in lyophilized form for one month. Overall, the process of ultrasonication appears reasonably safe in terms of pre-serving the IgA content, essential fatty acids and most of the LC-PUFA. The increase observed in FFA with ultrasonication could be beneficial for the immature digestive system of the newborn. Similarly, prevention of excess peroxide formation in milk by ultrasonication could be an added advantage. Although this protective effect of ultrasonication was not seem in the pasteurized milk, it still did not fur-ther increase level of peroxides in that milk. In addition, ultrasonication of human milk (if done at the appropriate time) resulted in excellent recovery of both fat and protein in milk infused after being stored for up to four months. 81 C h a p t e r 7 S u g g e s t i o n s F o r F u r t h e r R e s e a r c h In order to initiate the use of ultrasonicated human milk in feeding ill and prema-ture babies, future research should include feeding trials with ultrasonicated breast milk giving prime emphasis on weight gain and fat balance. Studies comparing the effects of feeding ultrasonicated/non-ultrasonicated/pasteurized milk, on the long term growth and development of babies need to be conducted. Si-multaneously, feeding trials with lyophilized and reconstituted human milk need to be conducted with special emphasis on digestibility and energy retention or weight gain. Effects of ultrasonication and lyophilization on certain other immune compo-nents in human milk such as Immunoglobulin G, Immunoglobulin M, and leuko-cytes also need to be studied. 82 Appendix #1 CANADIAN MILK BANK ADDRESSES BRITISH COLUMBIA Vancouver Children's Hospital Aggie Radcliffe, Coordinator Mother's Milk Bank 250 West 59th Avenue Vancouver, BC V5X 1X2 Royal Inland Hospital Kamloops, BC ALBERTA Calgary Mother's Milk Bank c/o Foothills Hospital 1403 29th Street N.W. Calgary, AB T2N 2T9 Holy Cross Hospital Attention: Anne Parter 2210 2nd Street S.W. Calgary, AB T2S 1S6 University of Alberta Hospital Adeline Nitska, Change Nurse, ICN 112 Street & 84th Avenue Edmonton, AB T6B 2B7 Royal Alexandra Hospital Mrs. MacMillan Mother's Milk Bank Kingsway Avenue & 103 Street Edmonton, AB T5H 3V9 83 A L B E R T A Edmonton General Hospital (con't) 1111 Jasper Avenue Edmonton, AB T5K 0L4 Misericordia Hospital 16940 87th Avenue Edmonton, AB T5R 4H5 S A S K A T C H E W A N Saskatoon City Hospital 701 Queen Street Saskatoon, SK S7K 0M7 Regina General Hospital 1400 14th Street Regina, SK S4P 0Y7 Holy Family Hospital 675 15th Street S.W. Prince Albert, SK S6V 3R8 Melfort Union Hospital 317 Bemister West Melfort, SK SOE 1A0 Yorkton Union Hospital 270 Bradbrooke Drive Yorkton, SK S3N 2K6 Weyburn Union Hospital 201 1st Avenue, N.E. Weyburn, SK S4Y 1P7 84 S A S K A T C H E W A N (con't) St. Joseph's Hospital 1401 1st Street Estevan, SK Nipawin Union Hospital Mother's Milk Bank c/o White Fox Kinettes Nipawin, SK M A N I T O B A University of Manitoba Dr. Jim Haworth Dept. of Paediatrics Faculty of Medicine 685 Bannatyne Avenue Winnipeg, MB St. Boniface General Hospital Mrs. Edith Parker, Maternity Ward 409 Tache Avenue Winnipeg, MB O N T A R I O Mr. Jim Nasso National Baby Formula Service of Canada, Ltd. 365 Midwest Road Scarborough, ON M1P 4T8 (Central Bank for Ontario) Q U E B E C Mother's Milk Bank St. Justine's Hospital 3175 Cote St. Catherine Montreal, PQ H3T 1C5 Mother's Milk Bank Montreal Children's Hospital 2300 Tupper Street Montreal, PQ H3H 1P3 85 Mother's Milk Bank Montreal General Hospital 1650 Cedar Avenue Montreal, PQ Mother's Milk Bank Royal Victoria Hospital 687 Pine Avenue West Montreal, PQ H3A 1A1 N O V A S C O T I A Izaak Waltor Killam Hospital for Children 5850 University Avenue P.O. Box 3070, NS B3J 3G9 Grace Maternity Hospital 5821 University Avenue, NS B3H 1W3 N E W F O U N D L A N D Janeway Health Centre Pleasantville St. John's NF A1A 1R8 N O R T H W E S T T E R R I T O R I E S Stanton Yellowknife Hospital Marjorie McClelland (RN), Coordinator Breast Milk Bank Program Box 10 Yellowknife, NT X1A 2N1 Nursing Mother's Group and La Leche League in Hay River and Inuvik also available for emergencies. Q U E B E C (con't) 86 OTHER MILK BANKS D E N M A R K Milk Bank Children's Hospital Drosselvej 57, 2000 F Copenhagen, Denmark E N G L A N D Kings College Hospital Denmark Hill, S.E. London, England F R A N C E Hospices Civils Lactarium Puericulture 67005 Strasbourg, France F I N L A N D Children's Hospital University of Helsinki Helsinki, Finland B R A Z I L Universidad Federal do Rio de Janeiro Ilha do Fundao Rio de Janeiro, Brazil U N I T E D S T A T E S OF A M E R I C A San Diego Mother's Milk Bank University Hospital 225 West Dickinson Street San Diego, CA USA 92103 Santa Barbara Medical Foundation Box 1200 Santa Barbara, CA (under organization) 87 Mother's Milk Bank Institute for Medical Research 751 South Bascom Avenue San Jose, CA USA 95128 Wilmington Milk Bank Wilmington Medical Center 501 West 15th Street Wilmington, DE USA 19899 Nursing Mothers Association Hilo Hospital 1190 Waianuenue Avenue Hilo, HI USA 96720 Mother's Milk Bank Norton Children's Hospital P.O. Box 35070 Louisville, KY USA 40232 Central Massachusetts Regional Milk Bank Hahnemann Hospital 281 Lincoln Street Worcester, MA USA 01605 Department of Maternal and Child Health Dartmouth Medical School Hanover, NH USA 03755 Piedmont Mother's Milk Bank 4025 Greenleaf Street Raleigh, NC USA 27606 UNITED STATES OF AMERICA (con't) 88 Jefferson Davis Hospital 1801 Allen Parkway Houston, TX USA 77019 Children's Hospital of Kings Daughter 609 Colley Avenue Norfolk, VA USA 23507 Children's Hospital of San Francisco P.O. Box 3805 San Francisco, CA USA 94119 Mother's Milk Bank Alisal Community Hospital Salinal, CA USA 93901 OR Mother's Milk Bank Salinas Valley Memorial Hospital Salinas, CA USA 93901 Mother's Milk Bank Hartford Connecticut Hospital 80 Seymour Street Hartford, CT USA 06115 Mother's Milk Bank Louise S. Childs, MD 222 Waliupe Circle Honolulu, HI USA 96821 U N I T E D S T A T E S OF A M E R I C A (con't) 89 Appendix #2 BREAST MILK SERVICE — PROCEDURES (August 17, 1982) 1. We do have a Breast Milk Service at Children's Hospital which has been in operation since September 1974. We process approximately 25,000 oz. of milk per year, most of which goes to SCN's in the lower mainland. 2. We get most of our donors by means of a brochure that is given to all new mothers by local hospitals, by P.H.N.'s and various childbirth associations. 3. Donating mothers must have general good health. In BC all new mothers are tested routinely for Hepatitis B Antigens and those who are positive are not accepted. We prefer donors to be non-smokers and not to be on any medications. If a mother must go on antibiotics etc. for a period of time, we ask that she not express for us for 24 hours after going off them. Our donors get no remuneration whatever for their donations of breast milk — theirs is truly a labour of love. Because their "work" is rewarded only by the satisfaction they receive from helping other infants, they seem to be exceptionally consciencious about following the guidelines we set. 4. Donors express at home using clean techniques as far as hands and nipples are concerned and sterile technique for utensils. They may express by hand, by handpump (Evenfio hand pump is recommended), or electric pumps. The mothers are encouraged to use bottle liners (Playtex, Evenfio, Curity) for storing the milk and to seal them with twist-ties. Small sterile jars are also acceptable. The milk must be refrigerated as soon as it is expressed and is frozen within 24 hours. We provide a pick-up service for transporting the milk to the hospital. 5. Milk from 6-8 different mothers is pooled for each 4000 ml batch and a sam-ple of this is sent to Bacteriology. The milk is placed in 40, 100 ml bottles and pasteurized in our Oxford Human Milk Pasteurizer for 30 minutes, at 63° C followed by a rapid cooling cycle. An unopened bottle is then sent for bacteriological testing. No milk is dispensed until this report comes back negative. 90 6. While we do not do any nutrient analysis at present, we will soon be doing crematocrit levels to determine the caloric value of the milk. Agi Radcliffe, R.N. Coordinator, Breast Milk Service B.C. Children's Hospital ARrjl 0645/Aug/82 91 Appendix #3 Fatty Acid Profile of Fresh and Processed Human Milk Fatty Acid Content (%) C10:0 C12:0 C14:0 C14:1 C16:0 C16:1 C18:0 F R E S H 1 1.08 5.38 7.62 0.64 20.45 3.70 7.53 2 1.13 7.41 7.37 0.62 20.14 2.67 10.67 3 0.72 5.73 7.60 0.64 22.07 2.99 0.61 4 0.25 7.40 7.97 0.46 16.10 1.57 8.00 5 0.11 4.40 5.55 0.37 15.45 1.54 8.36 MEAN 0.66 6.06 7.22 0.55 18.84 2.13 8.36 SD 0.470 1.320 0.960 0.125 2.902 0.935 1.429 F R O Z E N 1A 0.00 1.80 5.55 0.55 19.54 0.59 7.92 2A 0.22 4.41 6.58 0.53 20.61 0.52 11.24 3A 0.39 4.32 6.17 0.50 18.46 0.55 6.05 4A 0.00 2.75 6.31 0.24 15.02 0.75 8.60 5A 0.08 2.06 4.31 0.25 13.11 0.00 6.51 MEAN 0.14 3.07 5.78 0.41 17.51 0.64 8.22 SD 0.166 1.235 0.907 0.158 3.035 0.126 1.882 U L T R A S O N I C A T E D F R O Z E N IB 0.03 2.36 5.80 0.56 18.53 0.62 5.52 2B 0.28 4.35 5.13 0.89 14.23 0.77 5.23 3B 0.14 4.10 6.12 0.48 18.82 3.64 5.16 4B 0.03 4.35 6.27 0.23 14.41 1.88 5.61 5B 0.03 3.05 4.67 0.28 12.78 0.56 4.44 MEAN 0.10 3.64 5.60 0.49 14.75 1.49 5.13 SD 0.109 0.897 0.678 0.260 2.742 1.315 0.460 U L T R A S O N I C A T E D L Y O P H I L I Z E D IC 0.10 2.91 6.57 0.63 22.67 0.54 6.99 2C 0.19 4.25 6.67 0.52 20.59 0.53 7.36 3C 0.14 3.73 7.19 0.56 23.52 0.55 6.75 4C 0.08 4.29 6.06 0.29 17.03 0.45 6.81 5C 0.06 2.88 4.49 0.26 14.48 0.47 5.58 MEAN 0.11 3.61 6.36 0.45 19.66 0.51 6.71 SD 0.052 0.693 1.068 0.165 3.828 0.043 0.670 92 Fatty Acid Profile of Fresh and Processed Human Milk Fatty Acid Content (%) C18:1 Cl8:lt C18:2n6 C18:3n6 C20:0 Cl8:3n3 C20:l FRESH 1 30.09 2.20 12.63 0.13 0.35 1.27 0.48 2 28.26 2.07 12.65 0.10 0.45 1.47 0.40 3 31.39 2.34 14.53 1.10 0.39 1.15 0/27 4 29.38 2.09 19.44 0.08 0.27 1.28 0.29 5 33.48 2.34 22.07 0.16 0.00 3.38 0.52 MEAN 30.52 2.21 15.82 0.12 0.29 1.71 0.39 S D 2.006 0.131 5.035 0.031 0.175 0.940 0.110 FROZEN 1A 29.56 2.52 15.38 0.14 0.25 1.37 0.29 2A 35.17 2.32 12.01 0.10 0.31 1.14 0.41 3A 38.99 2.29 14.24 0.12 0.18 1.35 0.30 4A 34.74 1.69 22.18 0.11 0.37 1.38 0.41 5A 27.07 2.21 25.16 0.20 0.28 3.01 0.57 MEAN 37.11 2.21 17.80 0.13 0.28 1.63 0.39 S D 2.171 0.312 5.597 0.041 0.069 0.761 0.114 ULTRASONICATED FROZEN IB 43.89 2.05 14.42 0.14 0.11 1.29 0.81 2B 47.71 0.65 15.66 0.17 0.06 1.78 0.25 3B 41.06 2.00 13.43 0.11 0.08 1.27 0.03 4B 37.91 1.74 23.03 0.10 0.10 1.44 0.36 5B 39.06 2.02 25.64 0.21 0.25 3.47 0.52 MEAN 42.21 1.69 18.43 0.14 0.12 1.85 0.40 S D 3.784 0.595 5.519 0.146 0.074 0.928 0.302 ULTRASONICATED LYOPHILIZED 1C 42.24 0.72 12.24 0.16 0.00 1.46 0.33 2C 43.02 0.64 12.32 0.12 0.00 1.77 0.29 3C 40.69 0.47 12.20 0.13 0.00 1.46 0.43 4C 37.64 1.66 20.34 0.07 0.00 1.69 0.43 5C 40.39 2.06 22.60 0.20 0.00 3.38 0.70 MEAN 40.80 1.11 15.91 1.14 0.00 1.95 0.40 S D 2.072 0.704 5.109 0.048 0.000 0.810 0.175 93 Fatty Acid Profile of Fresh and Processed Human Milk Fatty Acid Content (%) 18:4 C20:2 C20:3N6 C20:4N6 C22:0 C22:l C20:5N3 F R E S H 1 0.13 0.31 0.46 1.76 0.00 1.40 0.28 2 0.06 0.17 0.30 1.87 0.50 0.41 0.08 3 0.04 0.21 0.34 1.75 0.29 0.38 0.18 4 0.04 0.27 0.30 2.13 1.00 0.56 0.24 5 0.03 0.26 0.29 0.64 0.64 0.07 0.06 MEAN 0.06 0.24 0.34 1.63 0.49 0.58 0.17 SD 0.039 0.053 0.070 0.571 0.376 0.492 0.096 F R O Z E N 1A 0.07 0.00 0.55 0.78 0.25 0.55 0.38 2A 0.00 0.03 0.35 0.79 0.80 0.72 0.53 3A 0.05 0.04 0.48 1.57 0.67 0.69 0.38 4A 0.00 0.00 0.29 0.52 0.93 0.82 0.42 5A 0.23 0.00 0.55 0.78 0.69 0.64 0.24 MEAN 0.07 0.01 0.44 0.87 0.67 0.69 0.39 SD 0.095 0.018 0.119 0.403 0.254 0.099 0.106 U L T R A S O N I C A T E D F R O Z E N IB 0.44 0.06 0.43 0.72 0.26 0.23 0.40 2B 0.02 0.06 0.43 0.85 0.23 0.21 0.16 3B 0.01 0.07 0.31 0.94 0.16 0.21 0.28 4B 0.02 0.04 0.26 0.56 0.34 0.31 0.11 5B 0.03 0.05 0.39 0.63 0.19 0.25 0.07 MEAN 0.10 0.06 0.36 0.72 0.24 0.24 0.20 SD 0.188 0.12 0.077 0.124 0.068 0.042 0.135 U L T R A S O N I C A T E D L Y O P H I L I Z E D IC 0.03 0.04 0.36 0.54 0.13 0.16 0.25 2C 0.02 0.02 0.31 0.48 0.12 0.11 0.11 3C 0.08 0.06 0.30 0.64 0.16 0.13 0.24 4C 0.04 0.02 0.28 0.42 0.17 0.23 0.09 5C 0.05 0.02 0.38 0.55 0.27 0.38 0.12 MEAN 0.05 0.03 0.32 0.52 0.17 0.20 0.16 SD 0.025 0.016 0.042 0.084 0.060 0.110 0.077 94 Fatty Acid Profile of Fresh and Processed Human Milk Fatty Acid Content (%) C24:0 C22:4N6 C22:4N3 C22:5N6 C22:5N3 C22:6N3 Total FFA F R E S H 1 0.00 0.35 0.16 0.32 0.55 0.73 100 2 0.03 0.43 0.00 0.20 0.19 0.25 100 3 0.00 0.43 0.03 0.25 0.18 0.37 100 4 0.17 0.13 0.00 0.21 0.10 0.30 100 5 0.00 0.07 0.40 0.05 0.17 0.18 100 MEAN 0.04 0.28 0.12 0.21 0.24 0.36 100 SD 0.073 0.173 0.172 0.102 0.177 0.217 F R O Z E N 1A 0.00 0.44 0.06 0.17 0.47 0.82 100 2A 0.00 0.50 0.07 0.23 0.25 0.17 100 3A 0.04 0.36 0.19 0.00 0.38 0.44 100 4A 0.12 0.62 0.25 0.00 0.32 0.33 100 5A 0.33 0.51 0.25 0.00 0.32 0.33 100 MEAN 0.10 0.49 0.16 0.08 0.30 0.38 100 SD 0.140 0.098 0.095 0.113 0.144 0.269 U L T R A S O N I C A T E D F R O Z E N IB 0.04 0.10 0.05 0.09 0.39 0.65 100 2B 0.08 0.11 0.05 0.07 0.31 0.25 100 3B 0.07 0.09 0.03 0.07 0.26 0.36 100 4B 0.09 0.23 0.02 0.08 0.16 0.33 100 5B 0.08 0.12 0.04 0.09 0.13 0.15 100 MEAN 0.07 0.13 0.04 0.08 0.25 0.35 100 SD 0.021 0.058 0.13 0.010 0.108 0.190 U L T R A S O N I C A T E D L Y O P H I L I Z E D 1C 0.01 0.05 0.04 0.04 0.29 0.50 100 2C 0.06 0.06 0.05 0.05 0.16 0.19 100 3C 0.05 0.06 0.06 0.05 0.19 0.13 100 4C 0.00 0.13 0.06 0.25 0.28 0.36 100 5C 0.17 0.07 0.09 0.09 0.12 0.15 100 MEAN 0.06 0.08 0.06 0.10 0.21 0.30 100 SD 0.067 0.033 0.018 0.091 0.076 0.141 95 References Alemi B, Hamosh M, Scanlon JW, Salzmann C, Hamosh P Fat Digestion in Very Low Birth Weight Infants: Effect of Addition of Human Milk to Low-Birth Weight Formula. Pediatrics 68:484-488 (1981). Alliger H Ultrasonic Disruption. Am Laboratory. 10:75-85 (1975). American Academy of Pediatrics, Committee on Nutrition. Commentary on breast feeding and infant formulas, including proposed standards for formu-las. Pediatrics 57: 278-285 (1976). American Academy of Pediatrics, Committee on Nutrition. Human Milk Banking. Pediatrics 65:854-857 (1980). American Academy of Pediatrics, Nutrition Committee of the Canadian Pe-diatric Society and the Committee on Nutrition of the American Academy of Pediatrics. Breast Feeding. A commentary on the celebration of the inter-national year of the child. 1979. Pediatrics 62:591-601 (1978). Arnold LDW Newsletter of the Human Milk Banking Association of North America. MMB, New York 3:3-8 (1988). Asquith MT, Lee D Ten Years of Human Milk Banking. J Perinatol 5:90 (1985). Asquith MT, Pedrotti PW, Stevenson DK, Sunshine P Clinical Uses, Collection and Banking of Human Milk. Clin Perinatol 14:173-185 (1987). Atkinson SA, Bryan MH, Anderson GH Human Milk Feeding in Premature Infants: Protein, Fat and Carbohydrate Balance in the First Two Weeks of Life. J Pediatr 99:617-624 (1981). Benda G M Modes of Feeding Low-Birth Weight Infants. Semin Perinatol 3:407 (1979). Berkow SE, Freed LM, Hamosh M, Bitman J, Wood DL, Happ B, Hamosh P Lipases and Lipids in Human Milk: Effect of Freeze Thawing and Storage. Pediatr Res 18:1257-1262 (1984). Bottcher VM, Mikkwitz HG, Schmidt C Uber die Lyophilisierung von Fravenmilch. Kinderarztl Praxis 55:403-405 (1987). Bjorksten B, Burman L, DeChateau P, Fredrikzon B, Gothefors L, Hernell O Collection and Banking of Human Milk: To Heat or Not to Heat? Br Med J 281:765-769 (1980). 96 Branski D The Prevalence of Disease in the First Year of Life: Human and Cow Milk Compared. In, Human Milk: Its Biological and Social Value. Freir S, Eidelman WL (eds). Excerpta Medica. Amsterdam. 209-213 (1980). Brooke OG, Barley J Loss of Energy during continuous Infusions of Breast Milk. Arch Dis Child 53:344-345 (1978). Brooke OG, Alvear J, Arnold M Energy Retention, Energy Expenditure and Growth in Healthy Immature Infants. Pediatr Res 13:215-220 (1979). Campbell AM, Penfield MP, Griswold R M The Experimental Study of Food. 2nd Ed. Houghton Mifflin, USA. 231-232 (1979). Canadian Pediatric Society Statement on Human Milk Banking. Can Med Assoc J 132:750-752 (1985). Chappell J E , Clandinin MT, Mevey MA, Chance GW Free Fatty Acid Content of Human Milk: Physiologic Significance and Artifactual Determi-nants. Lipids 20:216-221 (1985). Clandinin MT, Chappell J E , Heim T, Swyer PR, Chance GW Fatty Acid Utilization in Perinatal De Novo Synthesis of Tissues. Early Hum Dev 5:355-366 (1981). Clandinin MT, Chappell J E , Heim T Do Low Weight Infants Require Nu-trition With Chain Elongation — Desaturation Products of Essential Fatty Acids? Prog Lipid Res 20:901-904 (1981a). Clark RM, Kenneth H, Hundriser RS, Brown PD Effect of Tempera-ture and Length of Storage on Serum Stimulated and Serum-Independent Lipolytic Activities in Human Milk. J Pediatr Gastroenterol Nutr 3:567-570 (1984). Clinical Nutrition Review Banked Milk is Low in Immunologically Protective Proteins. Nutr Rev 45:44-46 (1987). Cruz JR, Arevalo C Levels of Human Milk Specific IgA Antibodies During Lactation. Pediatr Infec Dis 5:S148-S151 (1986). Cuthbertson WFJ Essential Fatty Acid Requirements in Infancy. Am J Clin Nutr 29:559-568 (1976). Davies DP, Evans J J Nutrition and Early Growth of Preterm Infants. Early Hum Dev 2:383-392 (1978). Davies R Observations on the Use of Ultrasound Waves for the Disruption of Microorganisms. Biochem Biophys Acta 33:481-493 (1959). 97 Deeth MC, Fitzgerald CM, Wood A F A Convenient Method for Determining the Extent of Lipolysis in Milk. Aus J Dairy Res 109-111 (1975). Egan E A Neonatal Necrotizing Enterocolitis. In Textbook of Gastroenterology and Nutrition in Infancy. Lebenthal E. (ed). Raven Press, New York (1981). Evans T J , Ryley MC, Neal LM, Dodge JA, Lewarne L M Effect of Storage and Heat on the Antimicrobial Proteins in Human Milk. Arch Dis Child 53:239-241 (1978). Fallot ME, Boyd JL, Oski FA Breast Feeding Reduces Incidence of Hospital Admissions for Infections in Infants. Pediatrics 65:1121-1124 (1980). Finley, DA, Lonnerdal B, Dewey KG, Grivetti L E Breast Milk Compo-sition: Fat Content and Fatty Acid Composition in Vegetarians and Non-Vegetarians. Am J Clin Nutr 41:787-800 (1985). Folch J , Lees M, Stanley GHS A Simple Method for the Isolation and Pu-rification of Total Lipids From Animal Tissues. J Biol Chem 226:497-509 (1957). Fomon S3 Body Composition of the Male Reference Infant During the First Year of Life. Pediatrics 5:863-870 (1967). Fomon SJ, Filer L J , Zeigler EF, Bergmann K E , Bergmann RL Skim Milk in Infant Feeding. Acta Paediatr Scand 66:17-30 (1977). Ford J E , Law BA, Marshall, VME, Reiter B Influence of the Heat Treat-ment of Human Milk on Some of its Protective Constituents. J Pediatr 90:29-35 (1977). Fransson GB, Lonnerdal B Iron, Copper, Zinc, Calcium and Magnesium in Human Milk Fat. Am J Clin Nutr 39:185-189 (1984). Fredrikzon B, Hernell O, Blackberg L, Olivecrona T Bile Salt Stimulated Lipase in Human Milk: Evidence of Activity in Vivo and of a Role in the Digestion of Milk Retinol Esters. Pediatr Res 12:1048-1052 (1978). Freir S, Faber J Loss of Immune Components During the Processing of Human Milk. In, Human Milk Banking. Williams AF, Baum JD. (eds). Nestle Nutrition. Raven Press, New York 5:123-132 (1984). Friend BA, Shahani KM, Long CA, Agel E N Evaluation of Freeze Dry-ing, Pasteurization, High Temperature Heating and Storage on Selected En-zymes, B Vitamins and Lipids of Mature Human Milk. I. Evaluation of Fresh Samples and Effects of Freezing and Frozen Storage. Pediatr Res 17:61-64 (1983a). 98 Friis H, Anderson HK Rate of Inactivation of Cytomegalovirus in Raw Banked Milk During Storage at -20° C and Pasteurization. Br Med J 285:1604-1605 (1982) . Garza C, Schanler RJ, Butte NF, Motil K J Special Properties of Human Milk. Clin Perinatol 14:11-32 (1987). Garga C, Schanler RJ, Goldman AS, Dill C, Nichols BL Preparation and Evaluation of Fortified Human Milk for Very Low Birth Weight Infants. In, Human Milk Banking. Williams AF, Baum JD. (eds). Nestle Nutrition. Raven Press, New York 5:101-111 (1984). Gibbs JH, Fisher C, Battacharya S, Goddard P, Baum JD Drip Breast Milk: Its Composition, Collection and Pasteurization. Early Hum Dev 1:227-245 (1977). Gillin FD, Reiner DS, Gault M J Cholate Dependent Killing of Giardia Lam-blia by Human Milk. Infect Immun 47:619-622 (1985). Glass RJ, Svennerholm AM, Stoll BJ Protection Against Cholera in Breast Fed Children by Antibodies in Breast Milk. N Eng J Med 308:1389-1392 (1983) . Goldblith SA The Problems of Storage of Lyophilized Food Products. In, Recent Developments in Freeze Drying. Tenth Commission on Freeze Drying. Paris (1969). Goldman AS, Garza C, Nichols BL, Goldblum R M Immunologic Factors in Human Milk During the First Year of Lactation. J Pediatr 100:563-567 (1982) . Goldman AS, Goldblum RM, Garza C Immunologic Components in Human Milk During the Second Year of Lactation. Acta Pediatr Scand 72:461-462 (1983) . Goldman AS Smith CW Host Resistance Factors in Human Milk. J Pediatr 82:1082-1090 (1973). Goldman HL, Goldman JS, Kaufman I, Liebman OB Late Effects of Early Dietary Protein Intake on Low Birth Weight Infants. J Pediatr 85:764-769 (1974). Goldsmith SJ, Dickson JS, Barnhart HM, Toledo RT, Eitenmiller RR IgA, IgG, IgM and Lactoferrin Contents of Human Milk During Early Lacta-tion and the Effect of Processing and Storage. J Food Protec 46:4-7 (1983). 99 Greer FR, McCormick A, Loker J Changes in Fat Concentration of Hu-man Milk During Delivery by Intermittent Bolus and Continuous Mechanical Pump Infusion. J Pediatr 105:745-747 (1984). Gross SJ, Buckley RH, Wakil SS, McAllister DC, David RJ, Faix RG Elevated IgA Concentration in Milk Produced by Mothers Delivering Preterm Infants. J Pediatr 99:389-393 (1981). Hall B, Muller DPR Studies of Lipase Activity in Human Breast Milk Using Human Milk as Source of Both Enzyme and Substrate. In, Human Milk: Its Biological and Social Value. Freir S, Eidelman AI. (eds). Excerpta Medica Amsterdam. 79-84 (1980). Hall B, Muller DPR Studies on Bile Salt Stimulated Lipolytic Activity of Human Milk Using Whole Milk as a Source of Both Substrate and Enzyme, I. Nutrition Implications. Pediatr Res 16:251-255 (1982). Hall G, Anderson J, Linquest M, Olafsson B Flavour Changes in Whole Milk Powder During Storage. J Food Quality 7:153-190 (1985). Hamosh M Physiological Role of Milk Lipases. In, Lebenthal E. (ed). Textbook of Gastroenterology and Nutrition. Raven Press. New York. 473-482 (1981). Hamosh M, Berkow S, Bitman J, Freed L, Happ B, Jones JB, Mehta NR, Wood DL, Hamosh P Handling and Storage of Human Milk Speci-mens for Research. J Pediatr Gastroenterol Nutr 3:284-289 (1984a). Hamosh M, Bitman J , Wood DL, Hamosh P, Mehta NR Lipids in Milk and the First Step in Their Digestion/Pediatrics 75:146-150 (1984). Hanson LA, Motas C, Barrett J Studies of the Main Immunoglobulin of Human Milk: Secretory IgA. Immunology of Reproduction. Bulg Acad Sc 687-692 (1978). Harzer G, Haug M, Dieterich BS, Gentner PR Changing Patterns of Hu-man Milk Lipids in the Course of the Lactation and During the Day. Am J Clin Nutr 37:612-621 (1983). Heim T Energy and Lipid Requirements for the Fetus and Preterm Infant. J Pediatr Gastroenterol Nutr 2:516-541 (1983). Heird WC Improved Methods of Non-Oral Neonatal Feedings: A Commentary. J Pediatr 82:963-965 (1973). Hernell O, Blackberg L Digestion of Human Milk Lipids: Physiologic Signif-icance of Sn-2 Monacylglycerol Hydrolysis of Bile Salt Stimulated Lipase. Pediatr Res 16:882-885 (1982). 100 Hernell O, Blackberg L, Olivecrona T Human Milk Lipases. In, Textbook of Gastroenterology and Nutrition. Lebenthal E. (ed). Raven Press. New York. 347-354 (1981). Hernell, O, Olivecrona T Human Milk Lipases, I. Bile Salt Stimulated Lipase. J Lipid Res 15:367-374 (1974). Hernell O, Olivecrona T Human Milk Lipases, II. Bile Salt Stimulated Lipase. Biochem Biophys Acta 369:234-244 (1974a). Hughes DE, Nyborg WL Cell Disruption by Ultrasound. Science 138:108-114 (1962). Jakobsson I, Lindberg T Cow's Milk as a Cause of Infantile Colic in Breast Fed Infants. Lancet 2:437-439 (1978). Jelliffe DB, Jelliffe E F P Human Milk in the Modern World. Oxford Press, New York. 96-101 (1979). Jensen RG, Pital R E Milk Lipoprotein Lipases: A Review. J Dairy Sc 59:1203-1204 (1976). Juan A, Ordonez M, Angeles AM, Garcia L, Sanz B Effect of Combined Ultrasonic and Heat Treatment (Thermoultrasonication) on the Survival of a Strain of Staphylococcus Aureus. J Dairy Res 54:61-67 (1987). Kenny J F , Zedd A J Recurrent Group B Streptococcal Disease in an Infant Associated with the Ingestion of Infected Mother's Milk. J Pediatr 91:158-159 (1977). Kneebone GM, Kneebone R, Gibson RA Fatty Acid Composition of Breast Milk from Three Racial Groups from Penang, Malaysia. Am J Clin Nutr 41:765-769 (1985). Koletzko B, Bremer HJ Long Chain Polyunsaturated Fatty Acids in Infant Formulae. Eur J Pediatr 146:93-94 (1987). Koletzko B, Mrotzek M, Bremer HJ Trans Fatty Acids in Human Milk and Infant Plasma and Tissue. In, Human Lactation 3: Effect on the Recipient Infant. Goldman AS. (ed). Plenum Press, New York. 323-333 (1987). Kwock RO, Keen CL, Hegenauer J , Saltman P, Hurley LS, Lonnerdal BO Retention and Distribution of Iron Added to Cow's Milk and Human Milk as Various Salts and Chelates. J Nutr 114:1454-1461 (1984). Lammi K C J , Jensen RG Lipids in Human Milk: A Review. 2: Composition and Fat Soluble Vitamins. J Pediatr Gastroenterol Nutr 3:172-198 (1984). 101 Larsen SA, Homer D R Relation of Breast Versus Bottle Feeding to Hospi-talization for Gastroenteritis in a Middle-Class U.S. Population. J Pediatr 92:417-418 (1978). Lavine M, Clark RM Changing Patterns of Free Fatty Acids in Breast Milk During Storage. J Pediatr Gastroenterol Nutr 6:769-774 (1987). Leventhal JM, Shapiro ED, Aten CB, Berg AT, Egerter SA Does Breast Feeding Protect Against Infections in Infants Less Than 3 Months of Age. Pediatrics 78:896-903 (1986). Lewis-Jones DI, Lewis-Jones MS, Connolly RC, Lloyd DC, West CR Sequential Changes in the Antimicrobial Protein Concentrations in Human Milk During Lactation and its Relevance to Banked human Milk. Pediatr Res 19:561-565 (1985). Liebhaber MI Human Milk: Uses in Allergy/Immunology in the Older Child. Milk Banking International News. MMB, California 2:2 (1983). Liebhaber MI, Lewiston NJ, Asquith MT, OldsArroyo L, Sunshine P Alterations of Lymphocytes and of Antibody Content of Human Milk After Processing. J Pediatr 91:897-900 (1977). Lonnerdal B, Hoffman B, Hurley LS Zinc and Copper Binding Protein in Human Milk. Am J Clin Nutr 36:1170-1176 (1982). Maclean WC, Fink BB Lactose Malabsorption by Premature Infants. Magni-tude and Clinical Significance. J Pediatr 97:383-388 (1980). Mancini G, Carbonara AO, Heremans JF Immunochemical Quantitation of Antigens Single Radial Immunodiffusion. Immunochem 2:235-254 (1965). Martinez FE, Desai, ID, Davidson AGF, Nakai S, Radcliffe A Ultrasonic Homogenization of Expressed Human Milk to Prevent Nutrient Loss During Tube Feeding. J Paediatr Gastroenterol Nutr 6:593-597 (1987). Mata LJ Breast Feeding: Main Promoter of Infant Health. Am J Clin Nutr 31:2058-2065 (1979). McClelland DBL, McGrath J, Samson RR Antimicrobial Factors in Human Milk. Studies of Concentration and Transfer to the Infant During Early Stages of Lactation. Acta Paediatr Scand Suppl 271:3-20 (1978). Mcintosh D A Mother's Milk Bank Manual. The Calgary Mother's Milk Bank 1-17 (1985). Mehta NR, Jones JB, Hamosh M Lipases in Preterm Human Milk. On-tongeny and Physiologic Significance. J Pediatr Gastroenterol Nutr 1:317-326 (1982). 102 Morgan RGH, Borgstrom B The Mechanism of Fat Absorption in the Bile Fistula Rat. Q J Exp Physiol 54:228-243 (1969). Nakai S, Chi A C Spectrophotometry Determination of Protein and Fat in Milk Simultaneously. J Dairy Sc 53:276-278 (1970). Narayanan I, Prakash K, Gujral V V The Value of Human Milk in the Pre-vention of Infection in the High-Risk Low Birth Weight Infant. J Pediatr 99:496-498 (1981). Narayanan I, Prakash K, Prabhakar A K A Planned Prospective Evaluation of the Anti Infective Property of Varying Quantities of Expressed Human Milk. Acta Paediatr Scand 71:441-445 (1982). Narayanan I, Singh B, Harvey D Fat Loss During Feeding of Human Milk. Arch Dis Child 59:475-477 (1984). Nayman R, Thomson ME, Scriver CR Observations on the Composition of Milk-Substitute Products for Treatment of Inborn Errors of Amino Acid Metabolism. Am J Clin Nutr 32:1279-1289 (1979). Nutrition Committee of the Canadian Pediatric Society and the Com-mittee on Nutrition of the American Academy of Pediatrics Breast-feeding. Pediatrics 62:591-601 (1978). Nutrition Committee of the Canadian Pediatric Society Statement on Human Milk Banking. Can Med Assoc J 132:750-752 (1985). Nutrition Reviews: Human Milk as a Source of Long Chain Polyunsaturated Fatty Acids for Preterm Human Infant Neural Tissues. 42:247-248 (1984). Olegrad R, Svennerholm I Effects of Diet on Fatty Acid Composition of Plasma and Red Cell Phosphoglycerides in 3 Month Old Infants. Acta Pae-diatr Scand 60:505-511 (1971). Pittard WB, Anderson DM, Cerutti ER Bacteriostatic Qualities of Human Milk. J Pediatr 107:240-243 (1985). Pitt J Breast Milk and the High Risk Baby: Potential Benefits and Hazards. Hosp Pract 41:81-86 (1979). Poskitt E M E Infant Feeding: A Review. Hum Nutr Appl Nutr 37:271-286 (1983). Putnam JC, Carlson SE, Devoe PW, Barness LA The Effect of Variations in Dietary Fatty Acid Composition of Erythrocyte Phosphatidylcholine and Phosphatidylethanolamine in Human Infants. Am J Clin Nutr 36:106-114 (1982). 103 Radcliffe A A Report: Breast Milk Service Procedures. Children's Hospital (1982). Reddy V, Bhaskaram C, Raghuramulu N, Jagadeesan V Antimicrobial Factors in Human Milk. Acta Paediatr Scand 66:229-232 (1977). Roy CC Human Milk Banking. A Report for the Committee on Nutrition. Canadian Pediatric Society. March (1979). Ruegg M, Blanc B The Fat Globule Size Distribution in Human Milk. Biochem Biophys Acta 666:7-14 (1981). Ryder RW, Crosby-Richie A, McDonough B, Hall WJ Human Milk Con-taminated with Salmonella Kottbus. A Cause of Nosocomial Illness in In-fants. J Am Med Assoc 238:1533-1534 (1977). Sauve RS, Buchan D, Clyne A N Mother's Milk Bank: Microbiologic Aspects. Can J Public Health 75:133-136 (1984). Schreiner RL, Coates T, Schackelford P G Possible Breast Milk Transmission of Group B Streptococcal Infection. J Pediatr 91:159 (1977). Schwartz DP, Parks OW The Lipids in Milk: Deterioration. In, Fundamentals of Dairy Chemistry. Webb BH, Johnson AH, Alford JA. (eds). Wesport, Ani Publishing Company 220-272 (1974). Siimes MA, Hallman N A Perspective of Human Milk Banking. J Pediatr 94:173-174 (1979). Sinclair A J , Crawford MA The Incorporation of Linoleic Acid and Docosa-hexanoic Acid Into Liver and Brain Lipids of Developing Rats. FEBS Lett 26:127-129 (1972). Small D M The Physical Chemistry of Lipids from Alkanes to Phospholipids. Plenum Press. New York. (1986). Spencer SA, Hull D Fat Content of Expressed Breast Milk: A Case for Quality Control. Br Med J 282:99-100 (1981). Spencer SA, Hendrickse W, Roberton D, Hull D Energy Intake and Weight Gain of Very Low Birth Weight Babies Fed Raw Expressed Breast Milk. Br Med J 285:924-962 (1982). Stagno S, Reynolds DW, Pass RF Breast Milk and the Risk of Cytomegalovirus Infection. Engl J Med 302:1073-1075 (1980). Ste-Marie MT, Lee EM, Brown WR Radioimmunologic Measurements of Naturally Occurring Antibodies Reactive with E. Coli or B. Fragilis in Breast Fluids and Sera of Mothers and Infants. Pediatr Res 8:815-819 (1974). 104 Stine CM, Harland MA, Coulter ST, Jeness R A Modified Peroxide Test For Detection of Lipid Oxidation in Dairy Products. J Dairy Sc 37:202- 208 (1954). Stocks RJ, Davies DP, Allen F, Sewell D Loss of Breast Milk Nutrients During Tube Feeding. Arch Dis Child 60:164-166 (1985). Sunshine P, Asquith MT, Liebhaber MI The Effects of Collection and Pro-cessing on Various Components of Human Milk. In, Human Milk: Its Bi-ological and Social Value. Freir S, Eidelman AI. (eds). Excerpta Medica Amsterdam. 187-190 (1980). Sunshine P Human Breast Milk: Storage and Safety Considerations: Protective Effects. West J Med 132:61-62 (1980). Tarassuk NP, Nickerson TA, Vaguchi M Lipase Action in Human Milk. Nature 201:298-299 (1964). Taylor MJ, Richardson T Antioxidant Activity of Skim Milk: Effect of Soni-cation. J Dairy Sc 63:1938-1942 (1980). Vahlquist B New Knowledge Concerning the Biological Properties of Human Milk. Bull Intl Pediatr Assoc 6:22-25 (1976). Walker WA Antigen Absorption From the Small intestine and Gastrointestinal Disease. Symposium on Gastrointestinal and Liver Disease. Pediatr Clin North Am 22:731-746 (1975). Walker WA Effects of Immunologic and Maturational Factors in Colostrum on Antigen Uptake. In, Human Milk: Its Social and Biological Value. Freir S, Eidelman AI. (eds). Excerpta Medica Amsterdam. 127-131 (1980). Wallingford J C Nutritional and Anti-Infective Consequences of Pasteurization of Breast Milk. Newsletter, Human Milk Banking Association of North Amer-ica. MMB, New York 3:4-5 (1988). Wardell JM, Hill CM, D'Souza SW Effect of Pasteurization and of Freezing and Thawing Human Milk on its Triglyceride Content. Acta Paediatr Scand 70:467-471 (1981). Wardell JM, Wright AJ, Bardsley WG, D'Souza SW Bile Salt-Stimulated Lipase and Esterase Activity in Human Milk After Collection, Storage and Heating: Nutritional Implications. Pediatr Res 18:382-386 (1984). Whitfield MF Poor Weight Gain of the Low Birth Weight Infant Fed Naso-jejunally. Arch Dis Child 57:597-601 (1982). 105 Williams F H , P i t ta rd W B Human Milk Banking: Practical Concerns for Feed-ing Premature Infants. J Am Dietet Assoc 79:565-568 (1981). Williamson S, Finucane E , Ellis H , Gamsu H R Effect of Heat Treatment of Human Milk on the Absorption of Nitrogen, Fat, Sodium, Calcium and Phosphorus by Preterm Infants. Arch Dis Child 53:555-556 (1978). Wills M E , Han V E M , Harris D A , B a u m J D Short Time Low Temperature Pasteurization of Human Milk. Early Hum Dev 7:71-80 (1982). 106 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097455/manifest

Comment

Related Items