Hi Marie I understand the visceral response that people might have when considering encouraging wet nursing in such a context - and that it will make wet nursing a difficult sell, both amongst donors and beneficiaries. There's something very primal in the idea of 'contaminated' mothers or mothers' milk that evokes a sense of revulsion (the 'ick' factor) which is difficult to ignore. That said, I think most carers and practitioners can understand the difference between a 30 in a thousand and a 3 in a thousand risk of death. As a practitioner, I have some difficulty with the notion that exposing infants to a well established, and high, risk can be considered less ethically difficult than exposing infants to a tiny and unknown risk of HIV transmission from a mother of unknown HIV status (a risk of a risk, if you will). If it were my child, I would choose wet-nursing by a mother of unknown status over artificial feeding in a resource poor setting.

Thanks Marie for the question, The balance between protecting infants from HIV infection and Malnutrition/mortality is never easy. The preffered option is HIV testing for wet nurses and ARV treatment if found to be HIV positive. However, where these services are not available in emergency settings, wet nursing may be an option depending on age of the infant (case by case basis). For example, if the infant is aged above 3 months and CSB or other nutritious porridges mixes for > 6yrs are availabe , one might have to choose between using a wet nurse and risk a child suffering with HIV infection for the rest of their life or provide CSB and risk non-compliance to IYCF guidelines. I know giving CSB to children aged <6 months is againist the IYCF guidelines (exclusive breastfeeding for the first 6 months) but it is worth exploring in this situation. In many rural areas,infants are introduced to complementary foods early and they tolerate it, soin this case CSB or other foods might be a 'good' evil than 'HIV'. As we make choices to ensure child survival, the quality of life for surviving infants should also be put into consideration. I hope that in the future, NGOs providing nutrition serices in emergencies will also consider having some supplies HIV test kits reserved for such cases.From my experience working with HIV infected children, my view is that all efforts should be made to avoid even one child getting infected with HIV

Thanks for your thoughtful response, Florence. I have to say, I would really worry about the risks and feasibility of meeting a 3 month old's nutrition requirements relying on a semi-solid food like CSB, as well as the risks of contamination and the need to meet fluid requirements safely as well. Early introduction of complementary foods to infants is common but does contribute to worldwide undernutrition in these children. As you say, it is a real challenge to balance the least risky option. For your information and others, the 'real life' cases that prompted this question were very young infants (4 or 5 weeks old) where the mother had died (hepatitis).

Florence poses an interesting question. Could the risk of significant under-nutrition be the lesser of two evils in such a case. I think it's worth remembering that, even if the wet nurse is HIV+, transmission to an exclusively breastfed infant is by no means assured.

I agree with both of you. Even though MTCT rates are not 100% where possible we should try to protect babies from both malnutrion and HIV. Heat treatment of breastmilk is also an option subject to availability of cooking and cleaning facilities. However, challenges of hygiene raised by Marie will still apply and even if suitable BMS are available. Heat treatment of breastmilk if viable should be done as a temporary intervention while awaiting to determine HIV status of wet nurse. I haven't worked in emergency settings, but I guess that even with availability of BMS such infants may still be at risk of malnutrition due to water and sanitation issues. Seems we are caught between a rock and hard place!

Back to WHO IYCF Algorithim (2010), the breastfeeding recommendation for women with unknown HIV status is the same as for HIV Negative women. So wet nursing in the above case fits this description. The best is to treat the wet nurse as a mother with unknown HIV status.I was trying to avoid the wet nursing option, but seems to be the best option as per current guideliness!

Below are some key considerations regarding HIV and wet nursing provided by Dr Nigel Rollins, Department of Maternal, Newborn, Child and Adolescent Health, WHO, in response to this question. Note this is not WHO guidance. The question of wet nursing in emergency settings when the HIV status of women who may possibly wet nurse is unknown, is a balance of risk of factors that will most likely result in HIV-free survival of the child. This will include consideration of the prevalence of HIV, the likely duration of wet nursing and other factors such as the risks of not breastfeeding and using commercial infant formula in settings where there is no clean water or effective health system. There is no evidence base (that I am aware of) of these competing risks in emergency situations. One can consider the probable magnitude of relevant risks and thereby make good judgements. Here are some considerations regarding wet nursing in emergency settings where HIV may be present but where women may not know their HIV status: • An infant being wet nursed by a woman of unknown HIV status would be at low risk of HIV transmission since: o In most settings, the large majority of women are HIV uninfected. This is especially true in low prevalence settings and if the woman does not engage with high risk behaviours e.g. intravenous drug use; o The overall duration of breastfeeding is likely to be short, weeks or a few months, and therefore the proportional risk of HIV infection is likely to be low – the risk of transmission is about 0.79% per month of breastfeeding in the absence of all antiretroviral drugs; o HIV-infected women who have higher risks of transmission are those with high viral load and low CD4 count. These women are likely to have been unwell and to have already presented with symptoms, and perhaps they are not even well enough to breastfeed; • In most high prevalence settings (and ideally in low prevalence settings as well), pregnant women will have had an HIV test. A mother who is volunteering to wet nurse an infant in an emergency setting may know her status from some years back. • There has been one report of a previously HIV uninfected mother/wet nurse appearing to be infected with HIV by an infant who was already HIV-infected; it is extremely rare. • (There is also the situation where a mother knows she has HIV infection and has been giving replacement feeds to her infant. In an emergency setting, it may be more appropriate for her to start breastfeeding if the conditions for providing safe clean replacement feeds are not present, and especially if she is already on antiretroviral drugs) By contrast, in emergency settings, the risks of diarrhoea and malnutrition are high and therefore the related likelihood of mortality is high. Safe replacement feeding will be very difficult, especially within the first days after an emergency occurs. There is minimal opportunity to sterilise feeding equipment, access clean boiled water, and provide a safe environment for preparation of feeds. Furthermore, if the infant becomes unwell, the local health services is not likely to be able to treat the child effectively: health facilities may have been affected by the emergency itself e.g. earthquake, military conflict or flooding, or may not be able to cope with large numbers of other sick/injured persons requiring care. As a result, the child who develops diarrhoea or malnutrition because of not breastfeeding will be at high risk of death. In such situations, the overall balance of risks for HIV-free survival of infant/child is very likely to be in favour of breastfeeding, either by the mother or by a wet nurse, even if their HIV status is unknown. None of these considerations detract from the seriousness of potential HIV transmission through breastfeeding. However, the probability of death from diarrhoea or malnutrition is likely to be higher. Possible ways to manage the risks (though not formal WHO guidelines) • Provide clear information to health workers that breastfeeding is very important for the health and survival of infants and young children in the emergency setting. Clarify that when the HIV status of a wet nurse is unknown, the overall risk of transmission through breastfeeding is small. • Confirm to health workers that where mothers or wet nurses do not know their HIV status that they should breastfeed their infants in the same way that mothers who know they are not infected with HIV. • In settings with high HIV prevalence consider providing HIV testing in the later phase of the emergency response. It will not be feasible or a priority to do this immediately after the emergency event when a women starts to wet nurse but it may be possible to offer within a few weeks. • If there is a major concern that the mother has been involved with high risk behaviour or if the woman is unwell, then prioritise HIV testing. In high prevalence settings, consider developing a simple screening tool about HIV risk behaviour and symptoms. Questions could be asked of women volunteering to wet nurse. • If a wet nurse is confirmed to be HIV infected and there are no other immediate options for safe infant feeding, then provide oral nevirapine (daily oral dose) to the infant for 4 weeks beyond the total period of breastfeeding and then test the HIV status of the child. Source: Dr Nigel Rollins, Medical Officer, Research and Development, Department of Maternal, Newborn, Child and Adolescent Health.