The demand for banked human milk has been increasing, in large part due to the benefits associated with its use. Multiple levels of safeguarding built into the current milk banking process have resulted in an unblemished safety record for banked milk in North America. Holder pasteurization destroys many pathogens in human milk while preserving much of the biologically active content. Among the potential benefits of using banked human milk in the neonatal intensive care unit are promotion of breastfeeding, reduced incidence of necrotizing enterocolitis and sepsis, and possible support of long-term positive neurodevelopmental outcomes in very low- and extremely-low birthweight infants.
After completing this article, readers should be able to:
Delineate the safety guidelines promulgated by the Human Milk Banking Association of North America.
List the potential benefits of the use of banked human milk in the neonatal intensive care unit.
Neonatology and human milk have been linked since the inception of the medical specialty. Two major “breakthroughs” that revolutionized the care of low-birthweight babies in late 19th century Paris, usually attributed to Stèphane Tarnier and his student Pièrre Budin, were the incubator and the gavage feeding tube. At that time, the only “formula” available was human milk, generally provided by the infant's mother or wet nurses. The need for wet nurses resulted in the formation of human milk banks in Europe and then North America shortly after the turn of the century. During the 20th century, the incidence of breastfeeding for all gestational ages declined in the United States and Europe, and the practice of wet nursing disappeared almost completely.
By the later part of the 20th century, ongoing research continued to demonstrate the importance of human milk for the health and development of infants. Improvements in cognitive skills, behavior, neurodevelopment, and rates of infection all have been ascribed to feeding neonates human milk. With the growth of neonatology and the proliferation of neonatal intensive care units (NICUs), the number of milk banks in the United States grew until about 1980. At that time, concerns about cost and possible transmission of agents such as human immunodeficiency virus (HIV) by human milk, coupled with the availability of improved formulas for preterm infants, led to a dramatic decrease in the number of banks to a low of eight in North America. However, the demand for banked human milk has been increasing in recent years (Figure⇓), with a substantial increase in the number of liters being distributed annually. How can this be explained, and is this change a reasonable and appropriate response to the needs of patients?
The Human Milk Banking Association of North America (HMBANA), founded in 1985, comprises 10 banks in the United States and 1 in Canada. The initial goal of this association of not-for-profit milk banks was to promulgate standards for milk banking. Their “Guidelines for the Establishment and Operation of a Donor Human Milk Bank” have been widely accepted in North America and elsewhere as the model for how to run a safe, effective milk bank. Since the inception of the guidelines 20 years ago, the safety record of banked milk in North America has been unblemished. Multiple levels of safeguarding are built into the current milk banking process. Specifically, the following steps are taken by all milk banks that follow HMBANA standards:
Potential donors are cleared initially by their own physicians to assure that their health and welfare are protected.
Potential donors undergo a screening history similar to the one used to screen potential blood donors. Donor exclusion criteria include:
A positive blood test result for HIV, human T-cell lymphoma virus (HTLV), hepatitis B or C, or syphilis.
The donor or her sexual partner is at risk for HIV infection.
Use of illegal drugs.
Smoking or use of tobacco products.
An organ or tissue transplant or a blood transfusion in the last 6 months.
Regular consumption of more than two alcoholic drinks per day.
Residing in the United Kingdom for more than 3 months or in Europe for more than 5 years between 1980 and 1996.
Being born in or traveled to Gabon, Niger, Cameroon, Chad, Congo, or Equatorial Guinea.
Use of medication or herbal supplements (with the exception of progestin-only oral contraceptives or injections, levothyroxine, insulin, prenatal vitamins).
All donors undergo serologic screening, at milk bank expense, for:
HTLV I and II.
Hepatitis B and C.
Donors are taught carefully how to pump and collect their milk safely and cleanly and how to keep their pumps and collection systems sterile.
Donor milk is frozen in sterile containers immediately after collection and maintained in the frozen state until processed by the receiving bank.
Donor milk is cultured for bacterial contamination when received by the bank, and milk that has abnormal findings is not processed for distribution. Such findings include:
Milk that has a high degree of bacterial contamination (ie, > 106 colony-forming units).
Milk contaminated with specific problematic flora (eg, Staphylococcus aureus, Bacillus sp).
Donor milk that has passed all of the previous screening steps is then Holder pasteurized at 62.5°C for 30 minutes, a process demonstrated to eliminate known bacterial and viral pathogens.
Aliquots of milk are recultured after pasteurization to assure sterility; the presence of any bacterial growth at this point in the process requires discarding of the contaminated batch.
Years of experience with this method have demonstrated its safety both in vitro and in clinical use. Multiple levels of redundancy provide a greater degree of security. For example, after following the previously described historical screening, which is used by blood banks nationally, the Mothers’ Milk Bank of San José recently reported that about 3% of potential donors over a 6-year period were rejected on the basis of serologic testing. The Red Cross reports that about 2% of their blood donors also are rejected after serologic testing, so these findings are not surprising. These are screening tests, and based on the experience of the Red Cross with potential blood donors, it is likely that confirmatory testing would be negative for most of these cases. Nevertheless, the additional level of protection provided by pasteurization seems prudent.
Milk banks do not screen potential donors for cytomegalovirus (CMV). The prevalence of CMV seropositivity for pregnant women in North America ranges from about 40% to 60%. Thus, a large proportion of potential donors would be expected to be CMV-positive. A high percentage of CMV-positive women shed the virus in their milk at some time during lactation, with the peak incidence of virolactia occurring at about 1 month postpartum. Freezing has been proposed as a mechanism for preventing transmission of CMV through milk, but there have been reports of CMV transmission to infants by their own mother's frozen milk. New in vitro studies have confirmed the efficacy of pasteurization as a method of eliminating CMV infectivity in milk, while questioning the efficacy of freezing. Thus, rather than screening and eliminating a very high percentage of potential donors, milk banks have relied on pasteurization to protect against transmission of CMV and other viruses. Fortunately, the cases reported thus far of CMV acquired from the infant's own mother's milk have resulted in relatively mild infection, possibly due to passive transfer of antibodies to the infant both in utero and through the mother's milk. In terms of CMV, pasteurized donor milk actually may be safer for babies than fresh maternal milk.
Holder pasteurization has been demonstrated to provide a safe supply of human milk by destroying pathogens. The question remains, however, of what else is destroyed by this process. As noted in the Table⇓, although there are losses from pasteurization, much of the biologically active content of the milk is preserved. Most of the immunoglobulin E (IgE) and IgA remain available in banked milk after processing.
Ig=immunoglobulin, sIgA=secretory immunoglobulin A. Data from Tully DB, et al. J Hum Lact. 2001;17:152–155.
Component Function Level (%) IgA and sIgA Binds microbes in gastrointestinal tract, prevents passage 67 to 100 IgM Antibodies targeted against specific pathogens after maternal exposure 0 IgG Antibodies targeted against specific pathogens after maternal exposure 66 to 70 Lactoferrin Binds iron, retards bacterial growth 27 to 43 Lysozyme Destroys bacterial cell walls 75 Lipoprotein lipase Helps lipolysis of milk triglycerides to monoglycerides and free fatty acids 0 Bile salt-activated lipase Helps lipolysis of milk triglycerides to monoglycerides and free fatty acids 0 Monoglycerides Disrupts membranes of virus, protozoans 100 Free fatty acids Disrupts membranes of virus, protozoans 100 Linoleic acid (18:2n6) Essential fatty acid, precursor for prostaglandins and leukotrienes 100 Alpha-linolenic acid (18:3n3) Essential fatty acid, precursor for docosahexanoic acid; important for eye and brain development 100
Ig=immunoglobulin, sIgA=secretory immunoglobulin A. Data from Tully DB, et al. J Hum Lact. 2001;17:152–155.
The first potential benefit of banked human milk is philosophical. It seems somewhat quixotic to counsel a new mother on the importance of human milk for her preterm infant and encourage her to endure pumping, while simultaneously telling her that until her supply comes in, her baby will be fed formula. If we truly believe in the importance of human milk and wish to communicate that to mothers, we should inform them that in the potential absence of a mother's own milk, we recommend human milk from a second, safe source. This does not deny the superiority of the mother's own milk, but reinforces the importance of human milk. It is interesting that in Norway, where every NICU maintains its own milk bank, only 2% of NICU patients are fed formula. Norwegian data document a very low incidence of late-onset septicemia, very early establishment of full enteral nutrition, and a low incidence of necrotizing enterocolitis (NEC) compared with American data.
Additional data support the role of human milk, including banked milk, in reducing the incidence of NEC and sepsis. The data suggesting a protective role for human milk compared with formula in the pathogenesis of NEC go back to the very earliest studies in animals and humans. Few question the efficacy of fresh maternal milk in this regard, and most accept that freezing and pasteurizing potentially may decrease the protective effect of banked milk. In 1984, Narayanan found no significant difference in the incidence of infections among infants fed fresh human milk, pasteurized milk, or a combination thereof, but a significant increase in infection rates was noted after formula was added to pasteurized milk. These data suggest that human milk continues to provide a clinically significant degree of protection against infection, even when pasteurized.
Recent reviews continue to document a protective effect of donor milk against NEC. In the most recent single center trial, Schanler and associates could not demonstrate a protective effect of donor milk against NEC, but several potentially confounding circumstances were associated with the study. Because the initial feedings and about 50% of the total intake of patients in both the donor milk and the formula groups were fresh maternal (patient's mother's) milk, both groups may have been fed at least a partially “protective dose” of maternal milk. This likely decreased the risk of NEC in all groups. Although the difference in the rate of NEC between the formula-supplemented group (11%) was not significantly different from either the bank milk-supplemented group or the maternal milk-only group (both 6%), a much larger study is needed to detect a significant difference. Additionally, infants who were randomized to the banked milk group also were fed formula to improve weight gain and then were included (appropriately) in the banked milk group for the “intent-to-treat” analysis. This clinically realistic approach makes inclusion of the study in a meta-analysis problematic. The most recent meta-analysis by Boyd and colleagues reports that donor milk decreases the risk of NEC by about 79% (95% confidence interval, 24% to 94%). Most consider this result to be clinically significant.
The long-term impact of the use of banked milk in the neonatal period is of interest. Lucas and colleagues in England randomized neonates to different feeding regimens and subsequently performed detailed, long-term follow-up studies. Although the feeding regimens, which included term formula, currently are not recommended for preterm infants, results from their database provide an immense amount of information regarding the impact of early neonatal nutrition.
The study population was born between 1982 and 1985, and as the authors point out, nutritional practices have improved significantly since that time. The use of banked milk slowed early weight gain, and a peak alkaline phosphatase concentration of more than 1,200 IU was predictive of shorter height. Although the preterm infants were significantly smaller than the term infants by several measures, there was no difference in anthropometric measurements at long-term follow-up between infants fed maternal milk, banked milk, or various formulas. There was no difference in height among the different diet groups at 9 to 12 years of age, and growth after 18 months of age appeared to have been the most important determinant of eventual stature. Bone age, bone mineral content, and bone mineral density all tended to be higher in the banked milk group compared with the preterm formula group, although not significantly. A Danish study compared dual energy x-ray absorptiometry scans of preterm infants fed maternal milk who were supplemented with banked milk (at about 200 mL/kg daily) with those of infants supplemented with fortifier or preterm formula. The human milk-fed babies weighed less at term, but had equal bone mineral content and head circumferences at follow-up.
One of the big advantages touted for preterm formula over banked milk is better weight gain, but more rapid early weight gain may not always be positive. The same group in England compared human milk-fed and formula-fed small-for-gestational age infants and found that the infants fed preterm formula gained weight faster initially, although the difference disappeared by 6 to 8 years of age. However, accelerated early weight gain was associated with higher blood pressure at 6 to 8 years. The study did not include a banked milk group, but one possible implication of the findings is that slower growth with banked milk may be less concerning in light of the absence of a deleterious effect on size or bone status in the longer term. Indeed, other studies of cohorts of Swedish term and Dutch preterm infants have found that more rapid weight gain in early infancy is associated with obesity.
Meta-analysis of the effect of banked milk as compared with formula on growth in the neonatal period reveals significantly slower early weight gain with donor milk, whether provided as sole or supplemental feedings. Compared with formula, donor milk as the sole feeding resulted in decreased early linear growth in five trials, with no difference noted in two trials; as a supplemental feeding, donor milk did not show an effect on early linear growth. The studies available for this meta-analysis were completed in the 1970s and 1980s, and none includes fortified donor milk, which limits their applicability to current NICU practice. Most donor banked milk is term human milk, which contains less protein, calories, and minerals than very low-birthweight (VLBW) infants appear to need to achieve intrauterine growth rates. Fortification of maternal milk for VLBW infants in the neonatal period is a currently accepted standard of care due to the risk of poor neonatal growth and phosphate and micronutrient deficiency without fortification. Future trials comparing feeding of fortified donor milk with feeding of preterm formula may be able to clarify these issues.
Data from the English cohort study showed other potential advantages of using banked human milk instead of formula. Although the absolute difference is small, blood pressure at 13 to 16 years of age in the banked milk-fed group was significantly lower compared with the preterm formula-fed infants. Further, the amount of human milk (both maternal and banked) consumed as a neonate was related inversely to blood pressure at long-term follow-up. Additionally, infants fed banked milk had better lipoprotein profiles and lower C-reactive protein concentrations compared with formula-fed controls in this same cohort.
There has been great interest recently in the use of formulas containing modified proteins to decrease the incidence of allergy. One study has shown that the use of banked human milk rather than formula to supplement maternal milk decreased the incidence of allergy in preterm infants who had a family history of atopy. Additional research is needed.
The major as yet-unanswered question concerns neurodevelopmental outcome. Although controversy continues, a correlation has been demonstrated between human milk feeding and improved intelligence and neurodevelopmental outcomes for term and preterm infants. Trials specifically examining the effect of banked donor milk have not been performed. In a large multicenter follow-up study of extremely low-birthweight (ELBW) infants, Vohr and associates reported a dose-related improved outcome for human milk-fed versus formula-fed babies. Specifically, they found that “every 10 mL/kg per day of human milk contributed 0.53 points to the Bayley Mental Development Index.” Banked milk was not used in this trial, but even if banking reduces the benefit, it is likely that a significant gain would be achieved for an ELBW infant fed exclusively banked milk instead of formula during a hypothetical 3-month NICU course.
Lucas and Cole demonstrated the potential value of milk banks for the British National Health Service in 1990. Since then, multiple studies have reported a financial benefit from using banked human milk in the NICU despite its up-front cost, assuming current pricing in the range of $3.50/fl oz. These analyses examined the acute early impact of donor milk and have not yet factored in any long-term benefits, such as improvements in neurodevelopmental outcome and effects on obesity, blood pressure, and atopy. The value of banked human milk as an adjunct to aggressive support of the preterm infant's own maternal milk supply cannot be denied. When maternal milk cannot be provided, such as when maternal disease contraindicates breastfeeding, banked milk clearly offers value that should not be denied to high-risk preterm infants.
Dr Cohen is Medical Director of the Mothers’ Milk Bank of San Jose, Calif.
- Copyright © 2007 by the American Academy of Pediatrics
American Academy of Pediatrics, Section on Breastfeeding. Breastfeeding and the use of human milk. Pediatrics 2005;115:496–506
Arnold LDW. The cost-effectiveness of using banked donor milk in the neonatal intensive care unit: prevention of necrotizing enterocolitis. J Hum Lact. 2002;18:172–177
Boyd CA, Quigley MA, Brocklehurst P. Donor breast milk versus infant formula for preterm infants: a systematic review and meta-analysis. Arch Dis Child Fetal Neonatal Ed. published online 5 April 2006; doi:10. 1136/adc. 2005.089490v3
Cone TE Jr. History of the Care & Feeding of the Premature Infant. Boston, Mass: Little, Brown & Company; 1985
Ekelund U, Ong K, Linné Y, et al. Upward weight percentile crossing in infancy and early childhood independently predicts fat mass in young adults: the Stockholm Weight Development Study (SWEDES). Am J Clin Nutr. 2006;83:324–330
Euser AM, Finken MJJ, Keijzer-Veen MD, et al. Associations between prenatal and infancy weight gain and BMI, fat mass, and fat distribution in young adulthood: a prospective cohort study in males and females born very preterm. Am J Clin Nutr. 2005;81:480–487
Fewtrell MS, Prentice A, Cole TJ, et al. Effects of growth during infancy and childhood on bone mineralization and turnover in preterm children aged 8 –12 years. Acta Paediatr. 2000;59:148–153
Guidelines for the Establishment and Operation of a Donor Human Milk Bank. Raleigh, NC: Human Milk Banking Association of North America; 2006
Human Milk Banking Association of North America Web site. Available at: http://www.hmbana.org/
Jones F. History of North American donor milk banking: one hundred years of progress. J Hum Lact. 2003;19:313–318
Lucas A, Brooke AG, Morley R, et al. Early diet of preterm infants and development of allergic or atopic disease: randomised prospective study. BMJ. 1990;300:837–840
Maschmann J, Hamprecht K, Weissbrich B, et al. Freeze-thawing of breast milk does not prevent cytomegalovirus transmission to a preterm infant. Arch Dis Child Fetal Neonatal Ed. 2006;91:F288–F290
McGuire W, Anthony MY. Donor human milk versus formula for preventing necrotising enterocolitis in preterm infants: systematic review. Arch Dis Child Fetal Neonatal Ed. 2003;88:F11–F14
Meier J, Lienicke U, Tschirch E, et al. Human cytomegalovirus reactivation during lactation and mother-to-child transmission in preterm infants. J Clin Microbiol. 2005;43:1318–1324
Rønnestad A, Abrahamsen TG, Medbo S, et al. Late-onset septicemia in a Norwegian national cohort of extremely premature infants receiving very early full human milk feeding. Pediatrics. 2005;115:e269–e276. Available at: http://pediatrics.aappublications.org/cgi/content/full/115/3/e269
Schanler RJ, Lau C, Hurst NM, et al. Randomized trial of donor human milk versus preterm formula as substitutes for mother's own milk in the feeding of extremely premature infants. Pediatrics. 2005;116:400–406
Singhal A, Cole TJ, Fewtrell M, et al. Promotion of faster weight gain in infants born small for gestational age: is there an adverse effect on later blood pressure? Circulation. 2007;115:213–220
Tully DB, Jones F, Tully MR. Donor milk: what's in it and what's not. J Hum Lact. 2001;17:152–155
Vohr BR, Poindexter BB, Dusick AM, et al. Beneficial effects of breast milk in the neonatal intensive care unit on the developmental outcome of extremely low birth weight infants at 18 months of age. Pediatrics. 2006;118:e115–e123. Available at: http://pediatrics.aappublications.org/cgi/content/full/118/1/e115