For centuries, all around the globe, women steadfastly breastfed their infants.

But in a world of poverty and disease, HIV-positive mothers face gut-wrenching choices with the concept that “breast is best.”

The most difficult choice of all: breastfeed and risk HIV transmission to their infant, or withhold breastfeeding with its immune-building power and risk watching their child die of other diseases.

“If you choose breastfeeding, you would, of course, have HIV infections. No question about it,” Hoosen Coovadia, MD, said during the 14th Annual Retrovirus Conference, held in Los Angeles in February. Breastfeeding could account for 300,000 new infant HIV infections per year, according to an estimate from the World Health Organization (WHO).

But a different figure, from UNICEF, estimates that 1.5 million babies would die each year if breastfeeding is not provided.

“So it’s a balance of death vs. infection,” said Coovadia in a plenary session. “And on the balance of probability, for poor women in the developing world, there is no other choice.”

Although UNICEF takes into account all women regardless of HIV status, research continues to show the benefits of breastfeeding by HIV-positive women in the developing world.

Although HIV is present in breast milk, providing water to infants in areas where clean water is not available introduces parasites and other dangerous microbes into their tiny bodies.

Coovadia, of the University of KwaZulu-Natal in Durban, South Africa, outlined the current knowledge on breastfeeding in HIV-positive mothers in the developing world. His report included the latest information presented at this year’s CROI. Among the disturbing reports: a study finding that the risk of infant death increased sharply within three months after breastfeeding was stopped in order to avoid HIV infection to those babies. (In countries like the United States, formula feeding is recommended for infants of HIV-positive mothers.) Coovadia conceded that the findings around breastfeeding are confusing.

Yet, with few—if any—good choices in the developing world, Coovadia, a longtime leader in caring for positive women and children, made the best possibilities shine through the grimness.

Pro and cons

Why would the cessation of breastfeeding lead to disease? Although HIV is present in breast milk, providing water to infants in areas where clean water is not available introduces parasites and other dangerous microbes into their tiny bodies. It is these dangers that can kill them, by causing diarrhea with its accompanying malnutrition, and pneumonia.

Years ago researchers reported that exclusive breastfeeding by HIV-positive mothers in developing countries was safer than mixed feeding. Mixed feeding is the introduction of water and food into the baby’s diet along with the breast milk.

Tracey Creek and colleagues, on behalf of the U.S. Centers for Disease Control and Prevention (CDC), reported that the number one risk for death for infants admitted to an emergency room in Botswana was the lack of breastfeeding. The risk was higher than that of storing drinking water, overflowing latrines, stagnant water near the home, or unwashed hands of caregivers. All risks were adjusted for socioeconomic status, age, and mother’s HIV status.

“But does breastfeeding protect infants of infected moms?” Coovadia asked. “The answer to that question was not planned for in these studies, but they tell a very compelling story. We have new evidence.”

In that growing body of evidence is a report from the Lancet medical journal in 2003 which shows that breastfeeding is far and away the best protection against death before the age of five. Among the different protective factors, it accounts for 13% of all the deaths that could have been prevented by all the measures listed, higher even than the use of the HIV drug Viramune together with replacement feeding. (“This is not an either-or situation,” Coovadia said. “We need both.”) None of the other factors listed provided a double-digit protection.

What about the effects of withholding breastfeeding from HIV-positive infants? A report from the Mashi study from Botswana, reported in the Journal of the American Medical Association (JAMA) in 2006, found that early mortality was higher in formula-fed positive children than those who were breastfed and given AZT. At one month of age, death was 4.3% for formula-fed infants but 1.5% for the breastfed ones. At seven months, the difference was 9.3% vs. 4.9%. At 12 months, the figures came closer together: 10.9% vs. 9.5%.

“As the child grows older, the difference becomes less,” Coovadia pointed out. “But here’s the dilemma. At 18 months, more of the breastfed infants were infected, but, more formula-fed infants died.”

Coovadia said an answer to this dilemma comes from a late breaker report at the conference from Louise Kuhn and Moses Sinkala and colleagues for the Zambia Exclusive Breastfeeding Study (ZEBS). Since various programs recommend that HIV-positive mothers cease breastfeeding sooner rather than later, ZEBS looked at the effect of cessation at four months of age. Not only was there no benefit in HIV-free survival as anticipated, but the risk of mortality was actually higher for both the HIV-positive and the HIV-negative infants who had stopped breastfeeding at four months.

The conference’s abstract report from this study concluded that, “Our results caution against early cessation of breastfeeding for HIV-infected women living in low-resource settings. … Programs providing HIV diagnosis services should strongly encourage breastfeeding into the second year of life for infants found to be HIV-infected.”

“It simply means, for heaven’s sake, don’t give water or bits of food or whatever else,” he reported.

In another report, from Uganda, researchers found that early withholding of breast milk from HIV-negative infants caused greater risk of serious gastroenteritis, resulting in an increased rate of mortality within three months of cessation. The research was testing guidelines from the country’s Ministry of Health that recommend abrupt weaning between three and six months of age to avoid HIV infection among infants of positive mothers.

In a press conference, Carolyne Onyango, MD, of the Makerere University-Johns Hopkins University Research Collaboration that conducted the study, said early HIV-negative results for the infants, provided to the mothers at two weeks and at four weeks after birth, led to discontinuation of breastfeeding. “There are statistics [supporting breastfeeding], but for that one mother whose baby gets HIV, that’s a 100% result,” she said. She also noted that the findings are preliminary, and the collaboration plans further analysis looking at other factors, such as the use of HIV medications and the mother’s viral load. She and another doctor also brought up in passing simple things that might have helped the women and their infants achieve better health—a bar of soap to help keep hands clean, a container to keep water clean after it’s been boiled.

Coovadia mentioned new findings from his research team, which are in the process of being published, that along with other reports show that exclusive breastfeeding is best for infants of positive mothers. “It simply means, for heaven’s sake, don’t give water or bits of food or whatever else,” he reported. “Exclusive breastfeeding lowers the risk of transmission. That point is made time and again.”

For one study of thousands of children in Zimbawe, exclusive breastfeeding up to six months had an HIV transmission rate of 1.3%. “It’s as low as any antiretroviral,” Coovadia pointed out. There were higher rates of 3% for children who were predominantly breastfed and 4.4% for those who were partially breastfed. “This is an absolute and convincing dose effect of exclusive breastfeeding and transmission rate,” Coovadia told his audience. He discussed other studies which support exclusive breastfeeding in positive mothers. In one of them, adding a solid to the baby’s diet led to a 10-times higher rate of transmission. Providing infant formula led to a 2-times higher transmission rate. “So a solid is very dangerous,” he said.

Yet, he noted, women do not exclusively breastfeed, and further, doctors report that it’s “impossible” to change that fact. “I told you that 99.9% of women don’t exclusively breastfeed. So how do you change this? We did,” Coovadia said. “The point is we can change behavior.” The catch: it cost millions of rands (the South African monetary value) through a generous grant, and involved both counseling and home visits. Nevertheless, said Coovadia, the point of principle was made.

“For me and for my colleagues,” said Coovadia, “and for many people like us who work in the developing world, we can’t give up on women and breastfeeding. That to me is too easy, too glib an answer. Most of the efforts of my colleagues presented here are really trying to make breastfeeding safe for women where they are, how they live, and hopefully, within the next few years. [Use of infant] formula did not work going from the West to the developing world. We have to find our own solutions. We have to make exclusive breastfeeding socially acceptable.”

Author’s note: For a Webcast or Podcast of Coovadia’s presentation, visit www.retroconference.org.