Article Text

Cochrane systematic review
There is an absence of randomised trials investigating the use of preterm banked milk compared to term banked milk for very low birthweight infants
  1. Dorothy Dougherty
  1. Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
  1. Correspondence to Dorothy Dougherty
    Sunnybrook Health Sciences Centre, Women and Babies Program, Neonatal Intensive Care Unit, Toronto, Ontario M4N 3N5, Canada; dorothy.dougherty{at}

Statistics from

Commentary on:

Sharing breast milk has always had a place in the history of infant feeding. However in 1985, it was established that HIV can pass from mother to child via breast milk ingestion, and there was doubt that the pasteurisation process could render donor breast milk free of the virus.1 This uncertainty decreased breast milk sharing and forced the closure of almost all North American milk banks. In response, the Human Milk Bank of North America association (HMBANA) was created. It galvanised policy creation, regulated operational standards and optimised screening of potential breast milk donors.2 Since then, research has shown that HIV along with hepatitis A, B, C and human T lymphotropic virus are eliminated during the pasteurising process.3 Pasteurised human donor milk (PHDM) has since been endorsed by the WHO, the American Academy of Pediatrics and the Canadian Pediatric Society as the second feeding choice for all infants in the absence of their own mother's milk, which is the first choice, with the rare exception.

The USA has 10 operational milk banks; there is one in Canada and 180 in Brazil, with others located throughout Britain, Europe and Australia. In 2009, HMBANA statistics showed that US milk banks distributed 1400 000 ounces of PHDM, the greater part of that milk being used for the populations in neonatal intensive care units. The widespread use of PHDM is supported by research studies that derive results from comparing PHDM to formula use or a combination of PHDM and mothers' own milk in comparison to formula.

It is known that preterm breast milk (preemie milk) is different from term breast milk. Unpasteurised preterm colostrum and breast milk has higher levels of secretory immunoglobulin antibodies, higher energy, calcium, phosphorus, magnesium, zinc, sodium, chloride and protein than unpasteurised term colostrum and breast milk. However, this is not a sustained difference and preterm breast milk values drop closer to term levels at roughly 3–4 weeks post preterm birth.4 ,5 Looking at the variation of pasteurised and unpasteurised term breast milk components, there are minimal nutritional and immunological differences. The notable difference is lower levels lactoferrin and IgG.5

Dempsey and Miletin present a well-written, methodical review achieving the stringent criteria necessary for acceptance by the Cochrane collaboration. Evaluation of the studies that met inclusion criteria resulted in no supporting evidence for their original query. They suggested in the conclusion that a randomised control trial looking at exclusively using either pasteurised preterm or term milk as a supplement to mothers' own milk or formula would be warranted as for further study. And yes, this is an appropriate proposition for further study, but even with these relaxed criteria, significant obstacles are likely to be encountered when gathering pasteurised ‘preemie’ donor milk, which is breast milk received from mothers whose infants are less than 36 gestational weeks.5 From experience, a consistent supply of ‘preemie’ donor milk is difficult to achieve. There is also a limited window of opportunity to collect breast milk that is still in its ‘preemie’ composition state before it transitions into a term breast milk composition.

There is strong support as shown in the studies examined in this review, that PHDM has superior benefits, including better tolerance of feeds and possible protection from necrotizing enterocolitis, when compared to formula use in the preterm population.6 ,7 It is likely that this body of evidence will support the current standard of care using pooled PHDM. The ability to engineer PHDM for specific infant populations, preterm versus term, is one that should be pursued, but it is important to remember that it is mothers' own milk that is the optimal infant food and needs to be supported as vigorously as PHDM research.


View Abstract


  • Competing interests None.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.