ELF Feeds: a fairy story analysed: part 1

Maureen MinchinAllergies, Bottle Feeding, breastmilk, Child Health, Commentary, Complementary Feeding, General News, Immunology, Infant Formula, Infant Health, Lactation, Microbiome, Reviews by MM


[Another post will comment on this group’s newest published research asap.] 

… a tiny study of just forty mothers336 may have persuaded some (who didn’t read it carefully) that infant formula is harmless, even beneficial, because it supports maternal breastfeeding…. The study authors’ conclusion: giving early limited formula (ELF) after every breastfeed up until the time of mature milk production means more breastfeeding for longer, as more mothers were solely breastfeeding at three months.

This widely publicised ELF study looked at the effects of giving 10 ml of expensive extensively hydrolysed infant (E-HF) formula by syringe after every breastfeed, within twenty-four to forty-eight hours after birth, to infants who had lost 5 per cent or more from birth weight under thirty-six hours old. The supplements were to be continued ‘until the onset of mature milk production’, likely to be day 4 at latest. A control group of mothers were instructed in ‘soothing techniques’, and advised to breastfeed exclusively for the duration of their hospital stay. Follow-up at one week and three months showed that the deliberately supplemented babies were much more likely to be fully breastfed than those advised to breastfeed exclusively, but free to feed their babies as they chose.

Those given the study’s 10 ml feeds drank much less infant formula in their first week of life (an average of 116 ml, at most 226 ml) than the other mothers advised but not supported to breastfeed exclusively. The group advised to breastfeed exclusively in fact fed an average of 262 ml (at most 673 ml) of various non-hydrolysed formulas to their babies in the first week. It was not stated whether the no comps/supposed to be ‘exclusively breastfeeding’ control group of mothers gave this formula in hospital, or only in the days after discharge; nor was the day of discharge given for either group. Not surprisingly, the supposedly exclusively breastfed group of babies actually fed the greater volume of infant formula in the first week of life were much less likely to be breastfeeding at three months of age.

True exclusive breastfeeding, according to the global definition, involves nothing but human milk from birth. Advising mothers to breastfeed and leaving them free to provide formula does not create an exclusively breastfed control group. In fact, when the figures are examined, it seems that of the forty mothers who entered the study, at most eight babies of mothers in the control group could have been genuinely exclusively breastfed at three months (overall 20 per cent, compared with the stated national figure of 30 per cent).337 Should the researchers be happy about that?

The conclusions that I draw from this study are:

•• using only tiny amounts of an extensively-hydrolysed brand for only 2–3 days does less harm than allowing mothers free use of both bottles and normal allergenic infant formula for the first week;

•• the more infant formula in total given by mothers in the first week, the lower the chance of continuing to breastfeed;

•• maternal anxiety about their initial milk supply needs to be addressed with effective education and support after hospital discharge and if need be small quantities of donated breastmilk (not gut-altering infant formula);

•• health professionals are unaware of the power of their modelling the use of infant formula, and are ignoring the WHO-preferred option of using donated human milk where necessary.

None of those conclusions is surprising or new. There were other aspects of this study that deserved comment:

•• The study was of infants who had lost 5 per cent or more from birth weight under thirty-six hours. There was no discussion of the difficulty of establishing true infant birth weight when IV fluids have been given in labour, so that ‘weight losses’ may have been real in some infants, but simply a reduction of fluid overload in others. Unless we know this, we can’t judge whether the intervention was either needed or useful in preventing further weight loss, though we can know that it altered the infant microbiome.

•• The higher parity of mothers in the intervention group, and the lower gestational age of infants in the control group, recorded in the study, may well have influenced outcomes, making continued breastfeeding more likely in the restricted comps group;

•• The apparently high stated rate (eleven of forty mothers) of delayed onset of lactation across both groups strongly suggests sub-optimal obstetric and postnatal practices, which undermine breastfeeding initiation.

Allaying mothers’ fears about the adequacy of their milk supply (by giving babies very small doses of formula by syringe) did indeed result in less formula use overall in the first week. And so greater breastfeeding success at three months. But was giving formula the best way to achieve that? Would, for example, ‘kangaroo care’ – keeping babies skin to skin and allowing frequent feeding – have achieved the same ends without the potential for altering infant gut development? And what about using donor human milk, and thus proclaiming by actions, not words, the second-rate nature of infant formula? It is sad that any infant formula was even considered for these seemingly quite healthy babies, without any research into consequences for gut development.

Yes – surprisingly in a study done so recently, there was no documentation of the unavoidable effects on gut microbiota, and the possibility that the effects of derangement of the microbiome might be subtle and very long-term. The authors were aware of the possibility that giving formula so soon after birth could have adverse health results for some babies, but wondered if this might be balanced by greater breastfeeding duration. Is it ethical to experiment without at least attempting to document results other than duration of breastfeeding? (It was pleasing that apparently many mothers had the sense to refuse to participate in the trial – maybe because they had been clearly advised that this might not be safe.) For example, given the research interest in the microbiota, could not the researchers have arranged for studies of the evolution of gut flora in both groups? Nor was there any apparent awareness of the powerful psychological impact of official medical approval of formula complementation, which may have influenced the rate of breastfeeding in both groups. Did it also influence formula or brand choice post discharge? Did anyone look? Physicians who recommend such practices should take care to think of the likely results of their actions.

Nor was there discussion of strategies (besides giving formula) to create greater maternal confidence that unsettled infant behaviour was not simply hunger; or if it was, strategies for meeting that infant need for suckling and sustenance. Feed frequencies of 8-12 times a day are mentioned as normative in the article, limiting formula intake to 80-120 mL per day. But  in the first days of life babies are at the breast much more often if access is unrestricted. Gastric emptying times of ninety minutes make two to three-hour intervals unnatural.

This study fits neatly into the Western mindset, evidenced in so many actions and books written by health professionals until recently: ‘When breastfeeding is a problem, formula is the answer.’ But is it? There may well be serious consequences for health professionals who cause babies to be exposed to artificial feeding unnecessarily. And necessity has been clearly outlined in the Baby-Friendly Hospital Initiative’s Global Criteria; it is no longer a matter of one staff person’s opinion, however high in the hierarchy.

[The next pages discuss the real and potential legal liability of those who order or give infant formula to infants without fully informed parental consent.]

336 Flaherman VJ, Aby J, Burgos AE, Lee KA, Cabana MD, Newman TB. Effect of Early Limited Formula on Duration and Exclusivity of Breastfeeding in At-Risk Infants: An RCT. Pediatrics; originally published online May 13, 2013; DOI: 10.1542/peds.2012-2809.

337 Maybe not even 8. Only 10 of 20 babies in the so-called EBF group were not getting formula at the end of the first week. The 8 women who were (dubiously) classified as exclusively breastfeeding (EBF)/not giving formula when asked at one month, may not be the same women as those classified as EBF at one week, or 2 or 3 months. Not surprisingly, getting no formula at the end of the first week was the best predictor of solely breastfeeding on the later dates.