Photo by Michael Chung
An infant breastfeeding on a normal schedule for three days on milk infected with HIV has as much risk of acquiring the virus as an adult who engages in one unprotected heterosexual act with an infected partner.
This statistic emerged from one of two recently published studies of HIV (human immunodeficiency virus) transmission through breast milk that will bolster the development of strategies to prevent such risk to newborns.
The studies - which aim to help resolve the quandary of how best to reduce HIV transmission to infants - were conducted by Dr. Barbra Richardson, a biostatistician in the Public Health Sciences Division, along with colleagues in Seattle and Nairobi, Kenya, where the work took place.
The analysis was the first to quantify the chances that a child will become infected with HIV based on the volume of ingested breast milk.
Researchers also found that women with advanced disease were more likely to transmit HIV to their newborns through breast milk than healthier women infected with the virus.
An accompanying study, led by Dr. Christine Rousseau, who worked as a postdoc in Dr. Julie Overbaugh's lab in the Human Biology Division, revealed that levels of virus in breast milk are highest immediately after birth and that high HIV levels are linked with greater risk of viral transmission.
The amount of virus in breast milk declines as time passes after a child's delivery. However, Richardson observed that an infant's risk of infection via breastfeeding is similar before and after four months of age, possibly because older children ingest more milk per feeding.
This discovery makes it unlikely that effective preventive strategies such as antiviral drug therapy can be targeted to a particular window of time. This approach has been successful at reducing transmission of HIV during childbirth.
The studies are published in the March 1 edition of the Journal of Infectious Diseases. Coauthors at other institutions included Dr. Grace John-Stewart, Dr. James Hughes, Matthew Steele and Dr. Joan Kreiss at the University of Washington; and Drs. Ruth Nduati and Dorothy Mbori-Ngacha at the University of Nairobi. A third paper by Richardson and Hughes to appear in the March edition of Biometrics describes the statistical methodology.
Richardson said the risk of an infant contracting HIV through breastfeeding is about one infection per 1,500 liters (or 330 gallons) of ingested breast milk.
"That number sounds low, and that is why it's important to put it in the context of other known risks, such as that posed by engaging in unprotected heterosexual sex with an HIV-infected partner," she said.
"You have to make that point for the public to recognize that viral transmission through breast feeding poses a considerable risk in countries where breast-feeding is the norm and access to the safe feeding of infant formula is unaffordable or unavailable."
An HIV-infected woman in the United States would be advised by her doctor to feed her infant with formula. But in underdeveloped countries that lack clean water to reconstitute formula powder, formula feeding poses serious health risks for newborns, including diarrheal diseases caused by bacterial infections.
After aggressive marketing of infant formula 30 years ago led to widespread infant death in Africa, the United Nations Children's Fund disallowed the promotion of formula there, and its use is still discouraged today. Moreover, formula feeding is unaffordable for many women in developing countries.
From work in Nairobi between 1992 and 1998, the researchers had determined that breast-milk transmission of HIV occurred in 16 percent of infants who were breastfed by infected Kenyan mothers. In that project, women were assigned randomly to a breastfeeding group or to a formula-feeding group. Up to seven sequential breast-milk samples were collected from women, and infants were followed for up to two years after their delivery to determine infection rates.
The current studies, which evaluated breast milk samples and infection-rate data from the randomized trial - resources that the investigators described as unmatched in the world - were initiated to determine probability of transmission and factors that influence breast-milk infectivity.
Such information will be essential to develop successful preventive strategies, Richardson said.
"The hope early on was that drug therapy could be targeted to the window of time right after birth, when infants may be most vulnerable to infection," she said. "But because our study shows that the risk of transmission is high both before and after four months of age, it's not likely that a one-time drug dose will be fully successful in preventing breast-milk transmission of HIV."
An earlier study involving researchers at Fred Hutchinson and UW showed that a single dose of the antiviral drug nevirapine given at the onset of labor nearly halves the risk of mother-to-infant HIV transmission during childbirth and the early breastfeeding period. The drug also was given prophylactically to babies in the first few hours of life.
The success of this strategy in reducing infant infection suggests that the early breastfeeding period, during which nevirapine was present in the newborns' systems, may be a time when drug therapy can have a significant, although incomplete, preventive effect.
This idea is supported by Rousseau's findings that virus levels were highest in breast milk produced during the first 10 days after birth. Further, she found that the risk of virus transmission doubles for every 10-fold increase in the level of virus in a mother's breast milk.
"Taken together, these results suggest that the risk that a woman will transmit HIV to her child through breast milk may be highest immediately after birth, since viral load in breast milk is highest at this time," said Rousseau, now a fellow at the Veterans Affairs Puget Sound Health Care System in Seattle, where she is completing a master's degree in epidemiology through UW.
Richardson's analysis did not aim to compare the infection rate in the first few weeks after birth with later time periods.
Another key finding from the studies was the correlation between a woman's health status and the infectivity of her breast milk.
"Women who are sicker and more immunocompromised as defined by high levels of virus in their blood and low levels of CD4 cells - the immune-system cells that HIV infects - are significantly more likely to transmit the virus to their infants," Richardson said.
She said treating women with what is known as HAART, a cocktail of antiretroviral therapy used to treat many AIDS patients in the United States and other developed nations, might be an effective preventive approach. Such drugs have not been widely available or affordable in Africa, where nearly 30 million people have the disease.
Overbaugh said ongoing studies in her research group seek to examine how different short-course drug therapies affect the levels of virus in breast milk and genital secretions.
"These approaches may help us find the best balance between what is a practical intervention in developing countries and what is most effective in preventing mother-to-child transmission," she said.
Despite social and economic barriers that have fueled AIDS-infection rates in African and other developing countries, Richardson remains optimistic that international recognition of the crisis is increasing. She noted the recent pledge by President Bush to spend $15 billion for drugs and infrastructure to combat the disease in countries hardest hit by the epidemic.
She also hopes the fruits of her research collaborations will have an impact on the AIDS crisis.
"Every day, I wake up and feel that the work is exciting and important," she said. "What motivates me is the hope that we're making strides in improving people's lives."