Bullshitometer: Women with HIV should not breastfeed their babies

Should HIV-positive mothers breastfeed their babies? Canvas a few different sources--the La Leche League (LLL), the American Association of Pediatrics (AAP) and your local mums' group--and you are apt to get several different replies. Who is right? Read on....

Full circle
The commonly-held feeling that "HIV-positive = Don't breastfeed" dates back to 1985, when the first case came to light of a baby infected with HIV from its mother's milk. In the years following, international agencies and non-governmental organizations (NGOs) around the world rolled out programs for preventing mother-to-child transmission which including advising HIV-positive women not to breastfeed while providing them with free formula milk. Given the extreme panic over what was virtually a death sentence, it seemed like mere common sense at the time.

As the years went on, however, concerns began to be heard about the effects of these policies, especially from NGOs and community leaders on the ground who were beginning to see what the rough end of formula feeding in the developing world actually looked like in practice. Reports began to come in of high mortality and morbidity rates among babies who were being bottle-fed in areas lacking clean water, refrigerators or facilities for sanitizing bottles and teats, and where erratic supplies of formula were resulting in mothers filling in the gaps with cow's milk or by over-diluting bottles.

This is what formula feeding in resource-poor environments looks like. 

Breastfeeding, as it happens, doesn't seem to be a very efficient vector of HIV--not least because it contains certain substances which reduce the risk of transmission; indeed, HIV-positive women produce certain immunoglobulins not found in the milk of HIV-negative women such as anti-HIV-IgG, -IgA and -IgM. As the data began to come in, it became clear that even in the worst-case scenarios (mothers who were taking no medications and who were mixing breastfeeding with formula and other foods), the majority of babies were still free of HIV by the time they weaned. In fact, the rate of babies who both survived and were free from HIV long-term tended to be higher for breastfed babies, because so many formula fed babies were dying of other things (1).

What's more, it was also becoming excitingly apparent that the rate of transmission could be beaten down still further by putting mothers on medications to control their viral loads (2); moreover, sticking exclusively to breastfeeding (rather than adding in formula/solids) for the full six months lowered the risk further still (3). It's not clear why mixed feeding increases risk: perhaps the other substances compromise the integrity of the intestinal mucosa, or perhaps the problem is that in developing countries, food is often premasticated (4) (resulting in tiny amounts of blood from gums mixing with the saliva and chewed food). This could also explain why a small percentage of exclusively formula fed infants also tend to wind up with HIV.

So, to bring the discussion full circle: what are the recommendations for HIV-positive women? Roughly speaking: Women in the developing world are now recommended to breastfeed exclusively for six months, and alongside solid foods until around 12 months old if possible. However, HIV-positive women in developed countries are still advised to choose formula--by the AAP (5), by BHIVA (6) and just about every medical body. Why the difference? Because in rich countries, formula feeding is basically safe/fine, and the risk of infection (while against the odds) is therefore more worrying in comparative terms. Oh, and another point which is often forgotten is that HAART medications themselves tend to get into your breastmilk; studies have indicated fairly low toxicity but also to an increased rate of anemia in infants breastfed by mothers on HAART (5). I wouldn't want my baby drinking HAART-laced milk; these are not very nice drugs at all.

Breastfeeding advocates deserve much applause for their tenaciousness in helping to bring to light the serious drawbacks of bottle-feeding in developing countries, a move that will probably save hundreds of thousands of lives. You would think that that would be enough; however, judging from the HIV-and-breastfeeding information displayed by sources such as Kellymom, LLL and Best For Babes, it appears that some lactivists cannot be content with showing the breastfeeding is the best option for HIV-positive mums in developing countries; rather, they insist on trying to make out that it is the best option for HIV-positive mums in developed countries as well. That, roughly speaking, is where the science ends and the woo starts.

Hard cases
In the UK, a large and growing proportion of HIV-positive mums consists of immigrants/asylum seekers from countries such as Zimbabwe and Somalia. They are often highly resistant to the suggestion to formula feed, because of the stigma of bottle-feeding in communities where breastfeeding is normalized, and because of concern that this could "out" their HIV-positive status. The danger is that if these women are threatened with child protection services for breastfeeding, they most likely will continue to breastfeed in secret anyway, and that secrecy will make it impossible to counsel them on risk reduction measures like avoiding combo-feeding, dealing with cracked nipples promptly and so on. They may also be more likely to avoid medical services altogether which will make it harder to vaccinate the kids and so on. For this reason, medics in many countries are now advising pragmatism--counsel mothers to bottle-feed, but if they absolutely refuse, then you support them to reduce risk as much as possible rather than threatening to call in the social workers. It is essentially a "harm reduction" measure, like needle exchange programs for drug addicts; peds are not endorsing HIV-positive breastfeeding any more than needle exchanges are endorsing the use of illegal drugs.

An even harder case is represented by asylum seekers, because if an HIV-positive woman is encouraged to formula feed only to have her asylum application fail, she'll be left up shit creek; she will be forcibly repatriated to her own country where formula feeding is not safe, while her milk will have dried up. Thankfully, this loophole has now been closed (in the UK, at least), with breastfeeding now encouraged for such women until it is clear that they have the right to stay in their new host country. These common-sense and compassionate exceptions will help to safeguard some very vulnerable mothers and babies, so it's a real shame to see that some lactivists have chosen to take advantage of these to try and create deliberate confusion, implying that doctors have now given the green light to HIV-positive breastfeeding for all women in developed countries. In fact, doctors have done nothing of the sort.

Take the World Alliance for Breastfeeding Action (WABA) paper for example (7). Under "Current infant feeding recommendations for resource-rich settings," the WABA paper says:

"Their [BHIVA and CHIVA's] current published Position Paper 48 recognises in paragraph 3 that an HIV-positive woman already receiving triple ART, with a repeated undetectable viral load at delivery may, after careful consideration, choose to exclusively breastfeed for the first six months of her baby’s life. In such a scenario, the current guidance recommends: Continuing maternal triple ART treatment and short-term infant prophylaxis. Exclusive breastfeeding for six months, Frequent follow-up, Careful monitoring of maternal adherence until week after weaning. Monthly checks on maternal viral load and infant HIV status."

This is what BHIVA actually says:

"...avoidance of breastfeeding is still the best and safest option in the UK to prevent mother-to-child transmission of HIV. BHIVA/CHIVA recognise that occasionally a woman who is on effective HAART and has a repeated undetectable HIV viral load by the time of delivery may choose, having carefully considered the aforementioned advice, to exclusively breastfeed...While not recommending this approach, BHIVA/CHIVA accept that the mother should be supported to exclusively breastfeed as safely, and for as short a period, as possible. Thus, 3. In the very rare instances where a mother in the UK who is on effective HAART with a repeatedly undetectable viral load chooses to breast feed, BHIVA/CHIVA concur with the advice from EAGA and do not regard this as grounds for automatic referral to child protection teams...Breastfeeding, except during the weaning period, should be exclusive and all breastfeeding, including the weaning period, should have been completed by the end of 6 months. The 6-month period should not be interpreted as the normal or expected duration of breastfeeding in this setting but as the absolute maximum, since exclusive breastfeeding is not recommended beyond this period under any circumstances. The factors leading to the maternal decision to exclusively breastfeed should be regularly reviewed and switching to replacement feeding is advocated as early as possible, whether this be after one day, one week or 5 months."

Bit of a difference in tone and emphasis, huh? The BHIVA and CHIVA people are very clear that breastfeeding is a less safe option, but that one might consider allowing an exception for an absolute bare minimum of time if Mum really, really wants to breastfeed. The WABA spin makes it sound as though the docs have changed their minds and are all gung-ho for HIV-positive mums to get the boobies out. For what it's worth, the WABA paper is mostly a decent document with a good run-through of why breastfeeding makes sense for HIV-positive mums in poor countries. The problem is that, like a lot of lactivist literature, it can't resist trying to take things a stage further by twisting the evidence to make it look like breastfeeding is now the recommended approach for rich-country mums too. It isn't.

Smoke and mirrors
"IBCLCs can now feel more confident than ever before in supporting HIV-positive clients who express a desire to breastfeed. With certain safe-guards, including maternal adherence to antiretroviral (ARV) regimens which are mandatory in developed countries, the risk of transmission of HIV through breastfeeding can be reduced to virtually zero" states international board certified lactation consultant (IBCLC) Pamela Morrison in New HIV and Breastfeeding Resource from the World Alliance for Breastfeeding Action (WABA) in early 2013, which discusses the WABA paper mentioned above.

Now, that WABA paper is a little coy about the exact HIV/breastfeeding transmission figures--it kind of hides them away in the middle of one of the files--but when I found the relevant figures and crunched the numbers, it works out as follows: if you do everything "right" (proper ARV regime, exclusive breastfeeding with no formula or solids), your chance of transmitting HIV by age six months is about 0.74%, or approximately 1 in every 135 babies. That is not "virtually zero." It's a decent level of risk for a mother in a Cambodian stilt village; for a mother in a developed country who can formula feed safely it is absolutely unjustifiable, considering that we're talking about an incurable viral infection which requires a lifetime on a pretty unpleasant drug regime.

Shockingly, when two HIV-positive mothers themselves ask for advice in the Comments section on Morrison's article, Morrison continues to recommend breastfeeding to them without enquiring as to whether they are living in an area where formula-feeding is feasible, and even (in one case) after the mother has herself stated that she lives in the United States. We even see her advising a mother on nursing to 12 months--not six--which significantly increases the risk of infection above that 1-in-135 figure. Morrison's spin about reducing the HIV infection risk to "virtually zero" was picked up by several lactivist blogs and pages, including Best for Babes on 9 January 2013 which repeated the statement uncritically: "Big news for lactation pros and HIV-positive mamas!"

Why you should always read the original paper
Best For Babes is apparently not alone in writing odd headlines for articles they link to. On 29 January of the same year, the Kellymom Facebook page made the following post:

"The US Academy of Pediatrics Committee on Pediatric AIDS has come out with a paper saying breastfeeding should be accepted by HIV-infected women in the US under certain conditions..." 

followed by a link to the AAP's latest guidelines on Infant Feeding and Transmission of Human Immunodeficiency Virus in the United States. Now, this is what the guidelines actually say:

"An HIV-infected woman receiving effective antiretroviral therapy with repeatedly undetectable HIV viral loads in rare circumstances may choose to breastfeed despite intensive counseling. This rare circumstance (an HIV infected mother on effective treatment and fully suppressed who chooses to breastfeed) generally does not constitute grounds for an automatic referral to Child Protective Services agencies. Although this approach is not recommended, a pediatric HIV expert should be consulted on how to minimize transmission risk, including exclusive breastfeeding... 

and the opinion of the good doctors is and remains that "in the United States, where there is access to clean water and affordable replacement feeding, the AAP continues to recommend complete avoidance of breastfeeding as the best and safest infant feeding option for HIV-infected mothers, regardless of maternal viral load and antiretroviral therapy" due to the risks from infection and from drug penetration into the mother's milk. Anyone who does not actually click on the link and comb through the article (=85% of people on Facebook) is going to come away with the impression that doctors in countries like the States have now basically given HIV-positive mothers the green light. (Kellymom is no doubt fully aware that this is what will happen, and spins articles this way for precisely this reason.)

More recently, Kellymom enthusiastically heralded the arrival of a new issue of La Leche League International (LLLI)'s online magazine with a feature on "Breastfeeding for HIV-Positive Mothers." This piece contains all the usual smoke-and-mirrors stuff common to other lactivist resources on HIV. It makes vague references about exclusive breastfeeding being protective, while being coy about the fact that EBF is only protective in comparison with mixed feeding--when compared with formula feeding, it is still more likely to pass on HIV. It cites the Courtoudis study, which is the one and only (small) study where the breastfed babies were no more likely to get HIV that the formula-fed ones, but then completely fails to cite all the other studies (such as the Mashi study (8)) which all show that formula fed babies are less likely to get HIV. It repeats the misleading spin on the AAP's and BHIVA/CHIVA's stance on HIV which I described above.

It's disappointing to see Kellymom linking to this kind of thing. Then again, what can you expect from a website whose own HIV page is such a mess? The "HIV/breastfeeding" stuff on Kellymom consists of the following (see here and here):
(1) an article on ARV regimes for women in East Africa (irrelevant to the vast majority of Kellymom's readership);
(2) an article on the increased risk of transmission with mixed feeding in Zambia (again, irrelevant) which Kellymom has, confusingly, chosen to subtitle "Prolonged Breastfeeding Protects Kids From HIV" (the article does not say this);
(3) a broken link which appears to have once led to yet another set of recommendations aimed at the developing world;
(4) the bizarre "AnotherLook at breastfeeding and HIV/AIDS" page, which contains a Mothering Magazine article by the AIDS denialist (and, for that matter, germ theory denialist) David Crowe.

I think my "favorite" bit of Crowe's article--which is a nest of pseudoscience and conspiracy theories--is the bit where he decides to raise a glass to one Christine Maggiore for her brave stance against the evil establishment: "Christine Maggiore defied the authorities and, perhaps because she was a public figure with a wide support network, was able to continue to publicly breastfeed her two children." Readers, the well-known AIDS denialist Christine Maggiore is dead. Of AIDS. So is her daughter Eliza--quite probably because she consumed her mother's infected milk. I don't know if Crowe included this reference as some sort of unpleasant joke or because he thinks that his readers don't know how to Google stuff, but I really do have to wonder why on earth Kellymom thinks that it is remotely acceptable to have links to this kind of stuff on their website.

Conclusion
The swing back towards recommending breastfeeding for most women in developing countries is a huge achievement for public health and child survival. Nevertheless, the risk of HIV transmission via breastmilk remains, as does concern about the impact of ART drugs which penetrate the milk. Breastfeeding is therefore not recommended for women in developed countries with the exception of asylum seekers of uncertain status.


Further reading:
(1) Early exclusive breastfeeding reduces the risk of postnatal HIV-1 transmission and increases HIV-free survival, Iliff et al, AIDS
(2) WHO Guidelines for PMTCT & Breastfeeding
(3) Influence of infant-feeding patterns on early mother-to-child transmission of HIV-1 in Durban, South Africa: a prospective cohort study, Coutsoudis et al, The Lancet

(4) Premastication: A Possible Missing Link? Mark J. DiNubile, Clinical Infectious Diseases
(5) Infant Feeding and Transmission of Human Immunodeficiency Virus in the United States,
COMMITTEE ON PEDIATRIC AIDS, Pediatrics
(6) Position statement on infant feeding in the UK, BHIVA/CHIVA Writing Group
(7) Understanding International Policy on HIV and Breastfeeding: a comprehensive resource, World Alliance for Breastfeeding Action (WABA)
(8) Breastfeeding plus infant zidovudine prophylaxis for 6 months vs formula feeding plus infant zidovudine for 1 month to reduce mother-to-child HIV transmission in Botswana: a randomized trial: the Mashi Study, Thior et al, JAMA