Normally breastfeeding is by far the best way to feed an infant, especially in the developing world. It is the simplest and most efficient way to deliver good nutrition to the infant. Breastmilk contains maternal antibodies and other factors that help the baby fight off infection, and breastfeeding also benefits the mother, by delaying the return to fertility, which helps with child spacing and reducing the burden of childbearing.

However, when the mother is infected with HIV, breastfeeding can transmit the virus to her child. That risk can be great in countries where prolonged breastfeeding is common, accounting for 42% of HIV infections in infants and young children in Africa according to a major meta-analysis of breastfeeding studies (The Breastfeeding and HIV International Transmission Study Group 2004). Overall, an estimated 5 to 20% of infants born to HIV-infected mothers are infected post-natally via breastfeeding.

In the developed world, both British and United States guidelines recommend that HIV-positive women should refrain from breastfeeding and use formula feed. Elsewhere, the World Health Organization (WHO) and the United Nations Childrens Funds (UNICEF) also recommend that HIV-infected mothers completely avoid breastfeeding when replacement feeding is acceptable, feasible, affordable, sustainable and safe.

However, this option is not available to all women in resource limited settings. Furthermore, some experts have been concerned that children would lose out on the other health benefits that breastfeeding offers. For example, children who are not breastfed are far more likely to suffer from malnutrition and to contract life-threatening diseases during the first year of life. According to a WHO meta-analysis, in the poorest countries, children who are not breastfed during the first two months of life are six times more likely to die from infectious diseases (World Health Organization Collaborative Study Team on the Role of Breastfeeding on the Prevention of Infant Mortality 2000).

Recent work has demonstrated that much of the risk associated with breastfeeding is actually due to mixed feeding: breastfeeding combined with giving the infant solid foods or other liquids such as formula, other animal milk, tea or juice. In the most recent of these studies, when infants were exclusively breastfed, only 1% became infected at six months compared to 4% whose infants were mixture fed.

In light of these findings, if complete avoidance of breastfeeding is not possible, the WHO and UNICEF now recommend exclusive breastfeeding during the first few months of life.

To minimise HIV transmission risk, breastfeeding should be discontinued as soon as feasible, taking into account local circumstances, the individual womans situation and the risks of replacement feeding, including the prevalence of community acquired infections and malnutrition. However, recent research has found that these messages are not getting through clearly to mothers, and the result is continued mixed feeding. There is an urgent need to find ways to support mothers who decide to either exclusively breastfeed or formula feed their infants.

Another focus of research is how to make exclusive breastfeeding safer by better understanding what factors are associated with an increased risk of transmission during breastfeeding. Researchers are also exploring whether specific interventions, such as antiretroviral therapy for the mother or infant could reduce the risk of transmission.

HIV transmission

The first evidence that HIV could be transmitted via breastmilk was a case report of the child of a previously healthy woman who was delivered by Caesarean section. Because of blood loss from the operation, the mother was given a blood transfusion after the delivery. The baby was breastfed for six weeks. It was later learned that a unit of the transfused blood had been contaminated with HIV. The mother and her infant were subsequently found to have both become infected (Ziegler 1985).

There have also been reports of infants of HIV-negative mothers becoming infected through exposure to HIV via wet-nursing or pooled breast milk (Nduati 1994).

Rates of transmission

Prior to the introduction of short-course antiretroviral interventions, prolonged breastfeeding was associated with less than a half of mother-to-child HIV transmissions in the developing world (De Cock 2000). However, treatment interventions to prevent mother-to-child transmission during pregnancy and childbirth are now commonly used. This means that despite a reduction in mother-to-child HIV transmissions, prolonged breastfeeding may become the leading cause of HIV infection in infants.

Overall, an estimated 5 to 20% of infants from HIV-positive mothers are infected postnatally and the risk increases as breastfeeding continues. A randomised clinical trial of 425 women in Nairobi reported that without short-course antiretroviral treatment, breastfeeding for two years or more doubles the overall risk of mother-to-child transmission of HIV to about 40% (Nduati 2000). Another study of pooled data from mother-to-child transmission studies found that 12% of over 1500 infants born to HIV-positive mothers contracted HIV infection via breastmilk (Read 2002). A Tanzanian study showed that infants who are not infected at birth have a 4% risk of contracting HIV at four months and an 18% risk of HIV infection at two years. This risk is increased by high maternal viral load and low CD4 cell counts (Fawzi 2002b).

Mixed versus exclusive breastfeeding

Two major studies now clearly show that there is a distinct difference between exclusive breastfeeding and the more common practice of mixed feeding: breastfeeding in combination with other foods. Mixed feeding is very common in African countries, and although solid food may be introduced as early as three months, breastfeeding will frequently continue alongside solid foods for up to two years after birth. In fact, the high rates of HIV transmission reported in the studies listed above were generally associated with mixed feeding.

The first study, conducted in South Africa, involved 551 women who were counselled on the risk of HIV transmission through breastfeeding, and offered formula at a subsidised price. After 15 months, transmission rates were 19% for formula fed infants, 25% in exclusively breastfed infants, and 35% in mixed fed infants. It is suggested that mixed feeding may have exposed babies both to allergens, causing inflammation and damage to gut mucosal barriers, and to HIV, which in turn led to a higher infection rate (Coutsoudis 2001).

The second, larger trial was carried out in Zimbabwe as part of Zvitambo, a study which was designed to assess the impact of vitamin A supplementation on maternal and infant health. It enrolled over 14,100 mothers at the time of delivery, of whom 32% were HIV-positive at the time of enrolment. Mixed feeding resulted in a fourfold greater risk of HIV transmission after six months, although this fell to threefold by 18 months after birth (Humphrey 2005).

These data present healthcare workers with a significant challenge, since mixed feeding is a very common practice in resource-poor nations. Only 26% of the women in the Durban study who chose to exclusively breastfeed their infants did so, while a separate study involving women in rural KwaZulu Natal found exclusive breastfeeding of very young infants to be uncommon, at a rate of 5% (Bland 2000).

Risk factors for transmission

The process of HIV transmission through breastmilk is poorly understood. However, HIV has been detected in breast milk, and animal studies suggest that HIV can cross mucosal tissue in the mouth and upper gastrointestinal tract to infect neonates. High viral loads and low CD4 cell counts have been highly associated with transmission, as these may lead to higher viral loads in the breastmilk (Pillay 2000; Rousseau 2003).

Certain factors, including longer breastfeeding and cracked nipples are thought to increase the risk of transmission, although research has not conclusively proven the association. It has been shown, however, that inflammation of the breast or 'mastitis' does increases the risk and that detectable viral load in breast milk is associated with transmission (Semba 1999). Infection with malaria and low birth weight have also been linked to the risk of HIV transmission during breastfeeding (Brahmbhatt 2002).

As described above, mixed feeding with breast milk and other foods appears to significantly increase the risk of transmission. However, giving colostrum, the early, antibody-rich secretion from the breast, does not appear to affect the risk of transmission.

Unsurprisingly, both the volume of milk ingested and the length of exposure have been found to be factors in transmission. Ingestion of one litre of breastmilk was estimated to pose a comparable risk to that associated with one episode of unprotected vaginal sex with an HIV-positive man (Richardson 2000).

Timing of HIV transmission

Transmission of HIV via breast milk may occur at any time and the risk increases with the duration of breastfeeding.

A number of studies report a transient increase in viral load after childbirth, which could be associated with a higher risk of transmission via breastfeeding during this period. Retrospective analysis of data has found evidence that babies are more at risk of contracting HIV from breast milk early in life (Dunn 1998). This was confirmed in a studies from Malawi and Kenya (Miotti 1999; Rousseau 2003). However, these findings should be interpreted with caution because current HIV testing technology makes it difficult to determine for certain whether an infant has been infected during delivery or by breastfeeding shortly after birth.

The Malawian study also found that women with fewer than four previous births were more likely to transmit HIV through breast milk than women who had four or more. Similarly, young mothers were more likely to transmit through breast milk than older mothers. The authors speculate that this may be due to 'less experienced' mothers being more likely to be affected by mastitis (Miotti 1999).

A meta-analysis of data from nine mother-to-child transmission studies involving over 5000 infants has found that 1509 infants were breastfed and that breastfeeding continued for an average of 6.8 months. The number of HIV infections definitely attributed to breastfeeding was 179, giving a rate of 12%. HIV transmission occurred before four weeks of age in 64% of cases (Read 2002).

Despite the evidence of an elevated risk of transmission shortly after childbirth, a meta-analysis of nine major studies of mother-to-child HIV transmission has shown that there is a significant and sustained risk of late HIV transmission to breastfed children. Of the 4085 children included in the analysis, 58% of the infected children acquired infection by day 28. The risk of late postnatal transmission continued throughout the breastfeeding period and was relatively constant. The risk for transmission was significantly higher among male infants and infants born to women with lower CD4 cell counts (Bulterys 2005; The Breastfeeding and HIV International Transmission Study Group 2004).

Effect of breastfeeding or formula feeding on infant health and survival

Although breastfeeding is usually beneficial for infants, data are less clear as to the benefit of breastfeeding by women with HIV. The contradictory data might be explained by a failure of the studies to stratify by the mothers health. Women with very advanced HIV are less likely to carry maternal antibodies, and if they are suffering from malnutrition, their breast milk may be less nutritious as well.

A meta-analysis of six studies investigating the effects of breastfeeding in developing countries found that bottle-fed children had a sixfold greater risk of death from infectious diseases in the first two months of life than breastfed children (World Health Organization Collaborative Study Team on the Role of Breastfeeding on the Prevention of Infant Mortality 2000). It has been suggested that cup- rather than bottle-feeding is preferable in resource-poor countries, due to the difficulty of sterilising feeding bottles. Access to clean water supplies is also crucial (De Cock 2000).

Impact of breastfeeding on the mothers health

One unexpected finding from a Kenyan study comparing breast and formula feeding was that breastfeeding poses a significant risk to maternal health. After 24 months, the maternal mortality rate in breastfeeding mothers was 11%, compared to 4% in the formula feeding arm. Infant mortality was strongly dependent on maternal mortality: infants whose mother died within the period of follow-up were eight times more likely to die subsequently than children whose mother survived (Nduati 2000a).

The researchers suggested that the demands of breastfeeding in HIV-infected mothers might accelerate the progression to HIV-related death. However, this does not accord with studies examining exclusive breastfeeding, which have not found higher death rates in breastfeeding women (Coutsoudis 2001a; Read 2003). Furthermore, prospective studies in Tanzania and Zambia found no evidence that breastfeeding was detrimental to the health of HIV-positive mothers (Kuhn 2005; Sedge 2004).

The investigators wrote that recent studies have found that breastfeeding may not deplete a womans energy reserves to the extent previously thought, and that a womans metabolic efficiency is improved during breastfeeding. They suggest that the association found between disease progression and breastfeeding in the Kenyan study could have been because it was observational in nature and failed to include adequate controls.

A recent meta-analysis has found no differences in mortality between HIV-positive mothers who ever breastfed compared to those who never did after 18 months. However, amongst the women who did start breastfeeding, the study found a lower risk of death among those who were still breastfeeding after 18 months. This is probably due to the fact that the women who are able to breastfeed for longer are the women who are healthier, rather than mortality being affected directly by the mother's choice of feeding method (Breastfeeding and HIV International Study Group 2005).

Interventions to prevent HIV transmission

In addition to the recommendation for exclusive breastfeeding for the first few months and to wean early, a number of other interventions have been proposed to further reduce transmission from breastfeeding in resource-poor settings. These include supplementation, treatment of milk and medical interventions.

Women who take a daily multivitamin excluding vitamin A can reduce death and prolong HIV-free survival significantly among their breastfed children. In contrast, vitamin A supplements have been associated with increased breastfeeding transmission of HIV (Fawzi 2002).

A number of methods of treating breastmilk to remove the HIV transmission risk are under investigation and may be an especially important option during periods of increased risk of transmission such as when the mother has cracked nipples or breast abscesses. All of these techniques require expression of breastmilk, which requires the women to be taught how to express breast milk to avoid mastitis, maintain proper hygiene and to sustain this process for long periods. Issues of stigma may also need to be addressed by counselling (Rollins 2004).

Free and cell-associated HIV levels are reduced substantially if subjected to heat at 62.5°C for 30 minutes (Orloff 1993). Solar Innovation in Denmark has developed a pasteurisation device that can be powered by electricity or solar energy (Kitinya 2002). However, simple heating methods including flash boiling and 'Pretoria pasteurisation' can be carried out by mothers at home, with good results in terms of reducing HIV levels significantly without destroying nutrients (Israel-Ballard 2005; Jeffery 2000).

However, a disadvantage of heat treatment is that it reduces the immunoglobulins and other protective components of breast milk. Several studies have reported that naturally occurring anti- HIV factors in milk can also inactivate the virus if it is left to stand at room temperature for 30 minutes, but these methods need to be evaluated prospectively to determine whether they reduce HIV transmission through breastfeeding.

Another possibility is the use of immunoglobulin in breastfeeding women. Immunisation with virus-specific antibodies has been demonstrated to prevent transmission of similar animal viruses to HIV in some studies. For example, ACTG 185 showed that immunisation with HIV antibodies throughout pregnancy, and to the infant at birth coupled with AZT (zidovudine, Retrovir) treatment, produced an unexpectedly low rate of transmission.

Antiretroviral prophylaxis in breastfeeding infants

Although antiretroviral treatment can prevent mother-to-child transmission of HIV during pregnancy and childbirth, little is known as to whether the risk of HIV transmission through breast milk can be minimised using antiretroviral regimens.

There are two possible approaches:

• Giving antiretroviral treatment to the mother to reduce viral load in plasma and breast milk.

• Providing prophylaxis to uninfected infants during the period of breastfeeding.

What little data are available are either preliminary or come from uncontrolled studies:

• Stopping Infection from Mothers-to-child via Breastfeeding in Africa (SIMBA) was a randomised open-label trial in Uganda and Rwanda, which assessed the effect of giving mothers AZT and ddI (didanosine, Videx / VidexEC) from week 36 of pregnancy until one week after the baby's birth. The 397 infants were randomised to receive either 3TC (lamivudine, Epivir) or nevirapine (Viramune) until four weeks after the end of breastfeeding. Similar rates of HIV transmission were seen in the two arms, with the majority occurring in pregnancy, labour or soon after birth through breastfeeding. The overall rate of late post-natal transmission was 1%, much lower that in trials not involving giving the infants prophylaxis, but should be interpreted with caution since the mothers' viral loads were low and the infants were weaned early (Vyankandondera 2003). Although the rate of transmission in the study was low, drug resistance was high in these children: twelve of the 13 children receiving nevirapine developed a resistance mutation, while the M184V 3TC mutation was present in nine of the 13 children who took it. Once off treatment, the mutation faded over time: only two of the infants still had the mutation detectable three to six months later (Giuliano 2005).

• MITRA was a similar study in Tanzania, testing AZT plus 3TC from week 36 until one week after birth in mothers, plus AZT and 3TC for one week after birth for the infants, followed by daily treatment with 4mg/kg 3TC throughout the breastfeeding period. HIV infection was significantly less frequent in 3TC-treated infants at six months compared to a historical control group of infants born to mothers who participated in the original PETRA study (Kilewo 2005).

• Mashi is an ongoing study in Botswana with a design that has evolved over time as evidence for treatments to prevent HIV transmission has accumulated. In its most recent guise, the study has provided evidence that six months' treament with AZT during mixed feeding provides as much protection against HIV transmission as formula feeding with only one month's treatment with AZT after birth. All of the babies also received a single dose of nevirapine at birth. However, mortality rates were higher in the formula feeding arm (Thior 2005). It is possible that giving more support to mothers to exclusively breastfeed and wean early could produce even greater benefits from this approach.

• The NVAZ study has shown that a single dose of nevirapine produces the same level of protection as a week of AZT and nevirapine given to infants during breastfeeding (Taha 2003).

Preliminary findings from recent studies show that treatment of HIV-positive breastfeeding mothers with antiretroviral therapy results in reduced viral loads in breast milk as well as drug levels in breast milk that should be sufficient to prevent HIV transmission.

For example, a sub-study of the Mashi trial found that the levels of HIV in the milk of infected women who had been taking a combination of nevirapine, 3TC and AZT for at least two months, were lower those in women who took AZT alone from week 34 of pregnancy plus a single dose of nevirapine during labour. All of the milk samples were taken from between two and five months after giving birth, when all of the mothers were still breastfeeding.

However, when investigators measured levels of cell-associated HIV (pro-viral DNA in HIV-infected cells) in the womens breastmilk they found no significant differences in the levels of cell-associated HIV between the two groups of women. It is unknown how much of a transmission risk cell-associated DNA poses to infants. No HIV transmissions were observed in any of the mother-infant pairs during this study (Shapiro 2005b).

However, treatment of the mother may also have an added protective effect by delivering prophylactic doses of antiretroviral therapy to the infant through her breastmilk. The investigators also measured the levels of nevirapine, 3TC and AZT in the breastmilk of 20 of the mothers receiving the three-drug combination (Shapiro 2005a). The drug levels were sufficient to produce blood levels of nevirapine 40 times the 50% inhibitory concentration (IC₅₀) of 24ng/ml in the infant's blood. The IC₅₀ is the drug concentration that has been shown to inhibit HIV replication by 50%. This concentration is similar to the level achieved by a single dose of 2mg/kg nevirapine, which is the dose commonly used for infant prophylaxis.

In contrast to nevirapine, 3TC levels were only 5% of the IC₅₀, at a median of 28ng/ml. Median AZT levels were 123ng/ml, 25 times this drugs IC₅₀, but all of the infants were also receiving AZT by mouth. Although there were no HIV transmissions in this study, the levels of all three drugs could put any infants who do become infected at risk of developing drug resistance. This could impede their future treatment options if they are not given full doses of antiretrovirals in addition to the low levels in their mothers' milk.

While AZT monotherapy given for the last two to four weeks of pregnancy and the first week after childbirth has been shown to reduce HIV levels in the breastmilk, stopping treatment can result in a burst of viral load once this is stopped. This can increase the subsequent risk of HIV transmission if breastfeeing is continued (Manigart 2004). Furthermore, single-dose nevirapine given at the onset of labour to prevent HIV transmission is more likely to decrease breastmilk viral loads than twice-daily AZT, coupled with a fourfold lower transmission of HIV to infants after six weeks. However, transmission rates were higher than in the Mashi study, at 7% (Chung 2005). Use of single-dose nevirapine can also lead to the presence of nevirapine-resistant HIV in the breastmilk, which can develop independently of resistance in the blood (Lee 2005a,b).

These findings indicate that sustained combination therapy, as in the Mashi study, is required to preventing HIV transmission in mothers who have no alternative to breastfeeding.

References

Bland RM et al. Longitudinal infant feeding study: constraints to exclusive breast feeding. Thirteenth International Conference on AIDS, abstract WeOrC497, Durban, 9-14 July 2000.

Brahmbhatt HPB et al. Determinants of mother to child transmission of HIV in the uterine/intrapartum period and during breastfeeding in Rakai, Uganda. Fourteenth International AIDS Conference, Barcelona, abstract ThPeC7428, 2002.

Breastfeeding and HIV International Transmission Study Group. Mortality among HIV-1-infected women according to children's feeding modality: an individual patient data meta-analysis. J Acquir Immune Defic Syndr 39: 430-438, 2005.

Bulterys M et al. Prevention of mother-to-child transmission of HIV-1through breastfeeding: Past, present and future. J Infect Dis 189: 2149-2153, 2004.

Bulterys M et al. Combination antiretroviral therapy in African nursing mothers and drug exposure in their infants: new pharmacokinetic and virologic findings. J Infect Dis 192: 709-712, 2005.

Chung MH et al. Breast milk HIV-1 suppression and decreased transmission: a randomized trial comparing HIVNET 012 nevirapine versus short-course zidovudine. AIDS 19: 1415-1422, 2005.

Coetzee D et al. Formula feeding is safe in a resource poor urban setting with potable water. Antivir Ther 8: S237, 2003.

Coutsoudis A et al. Are HIV-infected women who breastfeed at increased risk of mortality? AIDS 15: 653-655, 2001a.

Coutsoudis A. Method of feeding and transmission of HIV-1 from mothers to children by 15 months of age: prospective cohort study from Durban, South Africa. AIDS 15: 379-387, 2001b.

Coutsoudis A et al. Breastfeeding in women with HIV. JAMA 284: 956-957, 2000.

De Cock K et al. Prevention of mother to child HIV transmission in resource poor countries: translating research into policy and practice. JAMA 283: 1175-1182, 2000.

Dunn DT et al. Mother-to-child transmission of HIV: implications of variation in maternal infectivity. AIDS 12: 2211-2216, 1998.

Fawzi WW et al. Randomized trial of vitamin supplements in relation to transmission of HIV-1 through breastfeeding and early child mortality. AIDS 16: 1935-1944, 2002.

Fawzi WW et al. Transmission of HIV-1 through breastfeeding among women in Dar es Salaam, Tanzania. J Acquir Immune Defic Syndr 31: 331-338, 2002b.

Gaillard P et al. Use of antiretroviral drugs to prevent HIV-1 transmission through breastfeeding: from animal studies to randomized clinical trials. J Acquir Immune Defic Syndr 35: 178-187, 2004.

Giuliano M et al. Selection of resistance mutations in children receiving prophylaxis with lamivudine or nevirapine for the prevention of postnatal transmission of HIV. Twelfth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 99, 2005.

Humphrey J et al. Safer breastfeeding for babies born to HIV-positive mothers: part of the answer to a dilemma. Twelfth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 106, 2005.

Iliff PJ et al. Early exclusive breastfeeding reduces the risk of postnatal HIV-1 transmission and increases HIV-free survival. AIDS 19: 699-708, 2005.

Israel-Ballard K et al. Viral, nutritional, and bacterial safety of flash-heated and Pretoria-pasteurized breast milk to prevent mother-to-child transmission of HIV in resource-poor countries: a pilot study. J Acquir Immune Defic Syndr 40: 175-181, 2005.

Jeffery B et al. Pretoria pasteurisation: a potential method for the reduction of postnatal mother to child transmission of the human immunodeficiency virus. J Trop Pediatr 46: 219-223, 2000.

Kilewo C et al. Prevention of mother to child transmission of HIV-1 through breastfeeding by treating infants prophylactically with lamivudine in Dar es Salaam, Tanzania. Third International AIDS Society Conference on HIV Treatment and Pathogenesis, Rio de Janeiro, abstract TuPe5.3P01, 2005.

Kitinya W et al. Pasteurisation of expressed breast milk: the experience of HIV infected women. Fourteenth International AIDS Conference, Barcelona, abstract WePeB5947, 2002.

Kuhn L et al. Uptake of interventions to reduce mother-to-child HIV transmission: nevirapine and exclusive breastfeeding study (ZEBS). Fourteenth International AIDS Conference, Barcelona, abstract TuOrB1176, 2002.

Kuhn L et al. Prolonged breast-feeding and mortality up to two years post-partum among HIV-positive women in Zambia. AIDS 19: 1677-1681, 2005.

Lee EJ et al. Breast-milk shedding of drug-resistant HIV-1 subtype C in women exposed to single-dose nevirapine. J Infect Dis 192: 1260-1264, 2005a.

Lee E et al. Clonal analysis of drug-resistant mutations in plasma and breast milk following single-dose nevirapine. Twelfth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 801, 2005b.

Leroy V et al. International multicentre pooled analysis of late postnatal mother-to-child transmission of HIV-1 infection. Lancet 352: 597-600, 1998.

Manigart O et al. Effect of perinatal zidovudine prophylaxis on the evolution of cell-free HIV-1 RNA in breast milk and on postnatal transmission. J Infect Dis 190: 1422-1428, 2004.

Miller M et al. Breastmilk erythropoietin and mother-to-child HIV transmission through breastmilk. Lancet 360: 1246-1248, 2002.

Miotti PG et al. HIV transmission through breastfeeding. JAMA 282: 744-749, 1999.

Nduati R et al. Effect of breastfeeding and formula feeding on transmission of HIV-1: a randomized clinical trial. JAMA 283: 1167-1174, 2000a.

Nduati R et al. Effect of breastfeeding on mortality among HIV-1 infected women: a randomised trial. Lancet 357: 1651-1655, 2001.

Nduati R et al. Postnatal transmission of HIV-1 through pooled breast milk. Lancet 344: 1432, 1994.

Orloff SL, Wallingford JC, McDougal JS. Inactivation of human immunodeficiency virus type 1 in human milk: effects of intrinsic factors in human milk and of pasteurization. J Hum Lact 9: 13-17, 1993.

Phadke MA et al. High post-partum morbidity in replacement fed infants born to HIV-infected women in India, receiving short-course zidovudine. Fourteenth International AIDS Conference, Barcelona, abstract WePeB5920, 2002.

Pillay K et al. Cell-free virus in breast milk of HIV-1-seropositive women. J Acquir Immune Defic Syndr 24: 330-336, 2000.

Read JS et al. Breastfeeding and late postnatal transmission of HIV-1: an individual patients data meta-analysis (Breastfeeding and HIV International Transmission Study). Fourteenth International AIDS Conference, Barcelona, abstract TuOrB1143, 2002.

Read JS et al. Late postnatal transmission of HIV in breastfed children: an individual patient data meta-analysis (The Breastfeeding and HIV International Transmission Study). Tenth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 97, 2003.

Richardson B et al. Breastmilk infectivity of HIV-1 infected mothers. Thirteenth International Conference on AIDS, Durban, abstract WeOrC492, 2000.

Rollins N et al. Preventing postnatal transmission of HIV-1 through breastfeeding: modifying infant feeding practices. J Acquir Immune Defic Syndr 35: 188-195, 2004.

Rousseau CM et al. Longitudinal analysis of HIV type-1 RNA in breast milk and its relationship to infant infection and maternal disease. J Infect Dis 187: 741-46, 2003.

Ruprecht RM et al. Oral transmission of primate lentiviruses. J Infect Dis 179: S408-S412, 1999.

Sedge G et al. Breastfeeding and maternal HIV-1 disease progression and mortality. AIDS 18: 1043-1049, 2004.

Semba RD et al. Human immunodeficiency virus load in breast milk, mastitis, and mother-to-child transmission of HIV type I. J Infect Dis 180: 93-98, 1999.

Serunkuuma R et al. Differences in growth and illnesses between breast fed and formula fed infants born to HIV-positive women in Gombe Hospital, Mpigi district. Antivir Ther 8: S478, 2003.

Shapiro RL et al. Antiretroviral concentrations in breastfeeding infants of women in Botswana receiving antiretroviral treatment. J Infect Dis 192: 720-727, 2005a.

Shapiro RL et al. Highly active antiretroviral therapy started during pregnancy or postpartum suppresses HIV-1 RNA, but not DNA, in breast milk. J Infect Dis 192: 713-719, 2005b.

Taha TE et al. Short postexposure prophylaxis in newborn babies to reduce mother-to-child transmission of HIV-1: NVAZ randomised clinical trial. Lancet 362: 1171-1177, 2003.

Tene G et al. Growth of breast and formula-fed babies born to nevirapine treated HIV-1 positive mothers in Yaounde, Cameroon. Antivir Ther 8: S487, 2003.

Tess BH et al. Infant feeding and risk of mother-to-child transmission of HIV-1 in Sao Paulo State, Brazil. Sao Paulo collaborative study for vertical transmission of HIV-1. J Acquir Immune Defic Syndr Hum Retrovirol 19: 189-194, 1998.

The Breastfeeding and HIV International Transmission Study Group. The late postnatal transmission of HIV-1 in breastfed children: An individual patient data meta-analysis. J Infect Dis 189: 2154-2166, 2004.

Thior I et al. Breastfeeding with 6 months of infant zidovudine prophylaxis versus formula feeding for reducing postnatal HIV transmission and infant mortality: a randomized trial in Southern Africa. Twelfth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 75LB, 2005.

Vyankandondera J et al. Reducing risk of HIV-1 transmission from mother to infant through breastfeedingusing antiretroviral prophylaxis in infants (Simba study). Second International AIDS Society Conference on HIV Pathogenesis and Treatment, Paris, abstract LB7, 2003.

World Health Organization Collaborative Study Team on the Role of Breastfeeding on the Prevention of Infant Mortality. Effect of breastfeeding on infant and child mortality due to infectious diseases in less developed countries: a pooled analysis. Lancet 355: 451-455, 2000.

Ziegler JB et al. Postnatal transmission of AIDS-associated retrovirus from mother to infant. Lancet 1: 896-898, 1985.

Zijenah LS et al. Timing of mother-to-child transmission of HIV-1 (MTCT) and infant mortality in the first six months of life in Harare, Zimbabwe. Antivir Ther 8: S198, 2003.