HIV and Breastfeeding

World AIDS Day, 01 December 2022

Dec 01, 2022

In the early 1980s it was discovered that HIV, the virus that causes AIDS, could be passed through a mother’s milk to her baby. Almost overnight in the industrialised countries, and later in the African countries most ravaged by HIV, breastfeeding became an endangered practice. But in the rush to reduce transmission of HIV, everything we already knew about breastfeeding’s life-saving effects was overlooked, with devastating consequences for mothers and babies.

In HIV and Breastfeeding – the untold story, former IBCLC Pamela Morrison, an acknowledged authority on HIV and breastfeeding, reveals how women in the world’s most poverty-stricken areas were persuaded to abandon breastfeeding as part of a short-sighted and deadly policy that led to an humanitarian disaster.

The dilemma that breastfeeding, an act of nurturing which confers food, comfort and love, could be at once life-saving yet lethal, has been called ‘the ultimate paradox’. This critical account reveals how vital breastfeeding is, even in the most difficult of circumstances, and examines the lessons that can be learned from the mistakes of the past – which is particularly relevant as we deal with the consequences for mothers and babies of another global pandemic, Covid-19.

With detailed information for HIV-positive mothers and their caregivers, and success stories from mothers themselves, this book is essential reading for anyone involved in protecting and supporting breastfeeding, or with a need for evidence-based information about breastfeeding and HIV.

HIV in breastmilk

The virus that led to development of AIDS was first found in specimens of cell-free breastmilk in 1985.1,2 In 1988 it was detected by electron microscope, which further supported the hypothesis that breastmilk could carry the virus.3

Much of the very early research on HIV in breastmilk was conducted at the University of Nairobi in Kenya by expatriate researchers and former Kenyan students who had studied and qualified at overseas universities. The first study began in 1986,4 the year after the first Lancet report appeared showing that HIV could be passed through breastfeeding. Financial assistance for the early work was received from several national research groups in Canada, the National Institutes of Health in the US and the European Community in Brussels. The Nairobi group was headed by Dr Ruth Nduati. Other members included Peter Piot, Marleen Temmerman, Grace John, Paul Lewis, Julie Overbaugh, Barbara Richardson, Joan Kreiss, Pratibha Datta, Joanne Embree, Dorothy Mbori-Ngacha and others. The group also had close links to the University of Washington in Seattle, where Ruth Nduati had studied and met Peter Piot, who subsequently also spent time at the University of Nairobi. The Nairobi group was hugely influential in the AIDS community. Peter Piot was appointed the first Director of UNAIDS in 1994 and Ruth Nduati went on to write the first background document which underpinned the 1996–97 WHO change in policy on HIV and breastfeeding (explored in Chapter 12).

In 2003, Barbara Richardson and colleagues in the Nairobi group reached the startling conclusion that the probability of infection per litre of breastmilk ingested by the baby of an HIV-infected mother is equivalent to the amount of virus passed in one act of unprotected sex.5 Richardson found that the probability of infection was .00064 per litre of breastmilk, or .00028 per day of breastfeeding (vs .0003–.0015 per sex act). While this seemed alarming, to put it in perspective, it would actually correspond to approximately one infection per 1,500 litres of breastmilk ingested. The average baby consumes roughly 750–900ml of breastmilk per day, so this would be the amount of milk consumed by one baby in about five years.

In 1992, WHO had recommended that HIV-positive mothers in developing countries should continue to breastfeed their babies because the risk of death due to breastfeeding was likely to be less than the risk of death when breastfeeding was withheld.6 However, when bottle-feeding of formula in the West seemed like such an easy solution, HIV clinicians were not persuaded and more research showing the risks of breastfeeding continued to be published.

Breastfeeding as a risk factor for MTCT

A 1993 paper found HIV-infected cells in breastmilk at 15 days postpartum; a defective maternal IgM response was the strongest predictor of infection of the infant.7 After ingestion of virus in breastmilk, infant gut mucosal surfaces were the most likely site of transmission.
In research published in 1994 to see whether cell-free virus could traverse intact neonatal mucosal surfaces, researchers administered simian immunodeficiency virus (SIV) to four rhesus monkey neonates within an hour of caesarean section delivery, thus ruling out the possibility of infection occurring during labour and delivery. They found that the monkeys could be infected orally and that tonsillar cells were capable of viral replication.8 All neonates developed viraemia and subsequently tested HIV positive.
In 1995, an analysis to define the prevalence, concentration and determinants of virus-infected cells (HIV DNA) in breastmilk from which the lipid layer had been removed (which inhibited antiviral activity) found that during the first few days of life an infant would ingest a daily dose of 25,000 HIV-infected cells, and at six months 24,000 HIV-infected cells per day. Researchers asserted that ‘there is compelling evidence that breastfeeding contributes substantially to vertical transmission’.9
The concentration of virus in breastmilk was considerably lower than in blood. If no antiretroviral drugs were used, the plasma of most HIV-infected individuals contains 10,000–1,000,000 copies of HIV/ml.10 Although breastmilk samples contained only 240–8,100 copies/ml,11 the baby would consume a large volume every day.
Even though technically it was impossible to distinguish transmission occurring during labour and birth from transmission occurring a day or two later, one paper said that ‘colostrum would provide an effective route for transmission of HIV’.12 Another suggested that viral shedding in mature milk does not support of the concept of withholding colostrum from infants;10 there would be no point – although almost 40% of cell-free breastmilk specimens contained HIV RNA, prevalence did not decrease as the milk supply matured.
Procedures which could cause damage or lesions in the infant’s mouth, e.g. oral suction, or frenulotomy of the baby’s tongue, should be identified as factors which could increase transmission risk, by allowing contact between virus in breastmilk and the infant’s bloodstream.13,14
While studying the relationship between viral shedding of HIV in the cervix, vagina and in breastmilk, Grace John suggested that simultaneous shedding from multiple sites may be due to a high total virus burden.

Cell-associated vs cell-free virus

There was a lot of research looking at the likely infectivity of cell-associated HIV compared to cell-free HIV, a lot of it conflicting:

Cell-associated virus refers to HIV which lives inside the cell, measured as HIV DNA Cell-free virus refers to parts of the virus (virions) not associated with a cell, measured as HIV RNA

There was a strong correlation between maternal immunosuppression (CD4 cell count <400/ml3 ) and prevalence and quantity of breastmilk HIV DNA. Aboveaverage viral load in plasma meant a 2–4-fold increased risk of viral shedding in blood, genital secretions and breastmilk, leading to a higher risk of infecting the baby by all routes, including breastfeeding.16,17
In a Ugandan study detection of HIV-infected cells in breastmilk was not associated with transmission. Although 80% of transmitting mothers had detectable virus in their breastmilk, so too did 72% of non-transmitting mothers; the presence of HIV DNA in the cells of breastmilk did not result in increased late breastmilk transmission.18
Cell-free and cell-associated virus varied in their prediction of HIV transmission at early and late lactation stages.19 High levels of cell-free virus in maternal plasma and breastmilk led to a high risk of HIV transmission during breastfeeding, and with cell-associated virus in breastmilk, suggesting that both cell-free and cell-associated virus were involved.20
A South African group, which included the highly respected researchers Anna Coutsoudis and Nigel Rollins, quantified the relationship between HIV RNA shedding in breastmilk, cumulative RNA exposure, and postnatal transmission, relating timing of infection in the infant to estimated total volume of milk exposure.21 Cumulative exposure to RNA particles in breastmilk significantly increased the risk of HIV acquisition postnatally, independently from maternal antenatal CD4 cell count, plasma HIV load, and duration of breastfeeding.
Previous research had shown that while ART leads to undetectable levels of cell-free virus in breastmilk, cell-associated virus was still detected.22,23 Cell-associated virus per ml was more important than cell-free virus for early postpartum transmission at six weeks. Each 10-fold increase was significantly associated with transmission. However, at six months, cell-free rather than cell-associated virus posed the higher risk. 24
Certain factors in breastmilk blocked the transmission of HIV from mother to child during breastfeeding when the virus was outside of cells, but had little effect on the transmission of the virus once it had entered them.25 However, the inhibitory effect on cell-free virus was lost during extended culture, while the effect on cell associated virus was not. Antiretroviral therapy (ART) made no difference to virus in cell-associated milk and direct co-culture of HIV-infected CD4 T-lymphocytes with susceptible target cells revealed that breastmilk was ineffective at blocking cell associated HIV infection.

Pinter & Martin

Discussion about this post