Another major, postnatal (after birth) risk factor for MTCT -- breast feeding by HIV-positive mothers (in the absence of antiretroviral therapy for either mother or baby) -- is believed to account for up to one-third of HIV infections among babies in sub-Saharan Africa. The recommendation to avoid breast feeding that is standard in the developed world cannot be readily translated to developing settings, and currently there are no universal guidelines on infant feeding practices (i.e., that can be applied uniformly worldwide). Some research is available to guide decision making, but each woman's circumstances must be factored into the process.
Results of scientific studies on the relative risks of different infant feeding methods (breast, formula, or mixed feeding strategies) are somewhat contradictory, and scientific data to answer precise questions about when the risk of breast milk transmission of HIV is greatest are still lacking. For instance, factors that increase risk include the mother's general state of health (e.g., risk is generally greater when a mother has a very high plasma viral load) and anatomic characteristics (e.g., risk is greater if the mother has inflamed breasts or cracked and bleeding nipples). Beyond the individual, critically important considerations include community resources and social norms. Is clean water available? Is infant replacement formula available? These and other factors all must be considered to determine the best practice for an individual woman.
The debate over scientific and clinical approaches to infant feeding loomed large, as did discussions of relevant social and cultural issues and public health barriers to optimal infant feeding practices. Infant feeding was featured as the first plenary session of the meeting and constituted one of the main abstract sessions, as well as several workshops. The topic's predominance reflected its importance and also the uncertainty provoked by the inconclusive data accumulated to date; there is a critical need to sort out these inconsistencies and to come up with practicable approaches to offer pregnant women now, pending more conclusive data. As Arthur Amman, M.D., president of Global Strategies for HIV Prevention and the conference chair said, "Recommendations regarding breast feeding versus formula feeding will be one of the most difficult issues confronting the international community in developing countries."
Since the beginning of the epidemic, over four million children have died of AIDS. New infections continue at a rapid pace in the developing world. Since the primary risk of HIV infection in children today is perinatal or vertical (i.e., via their mothers), global inequities are painfully magnified in the MTCT arena. Over the course of the 20 years of the HIV/AIDS epidemic, however, remarkable scientific advances have been made in overall prevention strategies that protect children, including blood product screening and PMTCT. What has happened in the U.S. typifies the sort of successes achieved.
In the U.S., where the first case of pediatric AIDS was reported in 1982, early reports attributed HIV/AIDS in children to perinatal transmission, blood transfusion, and the use of contaminated blood products. However, after effective screening protocols for blood products and donations were implemented, the proportion of children infected postnatally through transfusions and use of blood products decreased substantially. By 1996 nearly 90% of cumulative pediatric AIDS cases and nearly 100% of new HIV infections in children resulted from perinatal transmission. In 1996 an estimated 1,750 children in the U.S. were born infected with HIV, and AIDS was the fifth leading cause of death for all children under 15 years of age. That same year, AIDS was the leading cause of death for children 1-4 years of age, according to the National Center for Health Statistics (NCHS).
By the end of the 1990s, significant additional progress had been made. Between July 1999 and June 2000, only 224 cases of AIDS in children less than 13 years of age were diagnosed, down from a high of 959 reported in 1993. Perinatal infection continued to account for the majority of cases: of 202 children with a reported risk factor, 195 (97%) had a mother with HIV. New infections among U.S. children today occur for a limited number of reasons, including failure of women to seek prenatal care, failure of physicians to screen for and treat HIV infection, and failure to suppress maternal viremia (HIV in the bloodstream) due to drug resistance or poor adherence. If these missed opportunities for prevention can be identified, new cases in the U.S. could be virtually eliminated -- a goal that many experts consider attainable.
This remarkable success can be attributed to the combination of improved HIV testing among pregnant women, use of potent antiretroviral therapy during pregnancy, and more effective treatment of infected children. Unfortunately, the progress and gains made in the developed world have not been readily shared by most of the rest of the world.
Another critical aspect of HIV among women and children is the large number of orphans. Worldwide, an estimated 13.2 million children have been orphaned because of AIDS, with five million in sub-Saharan Africa alone. While this topic falls outside the scope of the present article, the problems posed by the growing number of orphans are vast. As the epidemic expands, the increasing number of orphans will overwhelm current resources. Problems range from the immediate care of orphans, to resources, to development problems for a given country.
In the absence of treatment for HIV but with general overall stable maternal health (e.g., with asymptomatic HIV disease and no opportunistic or other diseases such as cytomegalovirus [CMV] or malaria), mothers have on average a 25% chance of transmitting HIV to their infants in utero or during delivery; that is, without any intervention, one in four babies will acquire HIV from his or her mother while in the womb or during birth. Early studies showed a range of rates of transmission, varying from less than 14% in Europe to 45% in sub-Saharan Africa, where breast feeding was considered largely responsible for the higher transmission rates. While data that quantify relative risk to babies of HIV infection via breast feeding are not completely clear, nor is the exact timing of transmission, it is clear that risk increases with duration of breast feeding.
The greatest reduction in MTCT will occur only when prevention programs that integrate antiretroviral treatment and safe alternatives to breast feeding are implemented. In the case of women in the developing world, many barriers remain. As mentioned previously, the use of perinatal antiretroviral therapy is not available to many pregnant or nursing HIV-positive women. In the developed world, HIV-positive mothers are advised to use replacement feeding or formula for their newborns. But replacement formula, while much cheaper than antiretroviral therapy, still is prohibitively expensive or unavailable to many women in developing countries. In addition, the correct usage of formula is also complicated in developing settings by various factors including, for example, the need for a reliable supply of clean water. Another complication derives from general confusion about the merits of formula, owing in large part to publicity about an ongoing international boycott against Nestlé, whose aggressive promotion throughout the 1970s of formula in developing settings became associated with infant illness and mortality. (A good source document is the World Health Organization's "International Code on the Marketing of Breast Milk Substitutes.")
Nevertheless, some infant feeding strategies that may reduce risk of MTCT can be used in developing settings (social and cultural norms notwithstanding; see section below entitled "Barriers to Formula Feeding in the Developing World"). For example, if formula is available and if the environment supports its sustained use throughout the appropriate time period, then the strong recommendation is for women to formula feed. Women who choose to formula feed in developing countries where it is unusual will need special support if they choose not to breast feed, as will women who make another nontraditional choice to breast feed exclusively -- another strategy that has been proposed to reduce MTCT.
In their study, Dr. Coutsoudis and colleagues evaluated the relative risks of different types of infant feeding in Durban, where the standard developed-world recommendation that HIV-positive women simply avoid breast feeding is not relevant (for various reasons, discussed below). The investigators looked at patterns of breast feeding, comparing MTCT rates in three groups of infants who were exclusively breast fed (EBF), mixed fed (MF), and formula fed (FF). (Participants were already enrolled in a trial looking at the influence of vitamin A supplementation on MTCT.) The distinctions between the three groups are important; mixed feeding means that mothers gave their babies foodstuffs in addition to breast milk, while formula feeding means that infants were never breast fed. The investigators documented the infant feeding practices of 549 HIV-positive women and then compared the MTCT of HIV between the three groups at different intervals.
According to Dr. Coutsoudis, what makes their study unique (and more compelling than most others) is that they controlled for "exclusive" breast feeding. Most studies that refer to breast feeding as practiced normally by African women are actually evaluations of mixed feeding, as most African women who breast feed usually also provide weak teas or cereals and foods other than breast milk to their infants. Because of the intensive infant feeding and HIV counseling, education, and support provided to the women in this study, Dr. Coutsoudis says she is confident that women who made an informed choice to exclusively breast feed were aware of the importance of adhering to their chosen mode of infant feeding, and did so faithfully.
The investigators found that, at three months of age, 18% of babies who were never breast fed (FF) were HIV positive, compared with 21% of breast fed babies (the EBF and MF groups). Other data on the two groups of infants who were breast fed -- the EBF and MF groups -- were striking: 14% of babies who were exclusively breast fed were HIV positive, compared with 24% of the mixed feeding babies. The EBF babies also fared better than the FF babies, of whom 18% were HIV positive. Unexpectedly, it appears that exclusive breast feeding may be preferable to mixed breast and formula feeding, which could have enormous public health implications in developing countries, where many mothers currently provide mixed feeding to their infants.
Researchers hypothesize that while breast milk is known to contain HIV, it may also contain antiviral factors as well as protective antibodies that may help protect the infant against HIV and other pathogens. Formula, on the other hand, may introduce new allergens or contaminants, which then can increase inflammation and absorption in babies' intestinal tracts; similarly, those babies receiving teas and cereals may be harmed by the introduction of contaminants (as would have happened in the mixed feeding group). Such babies therefore may have had gastrointestinal inflammation and infections that increased their risk of being infected with HIV through breast feeding.
Updated results of data analysis through six months, presented at Durban, showed no difference in the MTCT rate between exclusive breast feeding and formula feeding. However, there continued to be a significant difference between exclusive breast feeding and mixed feeding, which still appeared far riskier than breast feeding alone. Dr. Coutsoudis said that a metanalysis (examination and analysis of results) of several studies involving EBF and MF groups showed that EBF babies did better wherever the study was conducted, even in settings as different from Durban as Sweden.
These guidelines suggest changes to common practice in many parts of Africa, where it is customary to breast feed for an average of 24 months and to give the infant additional foods (mixed feeding). Many at the conference found these tentative suggestions reasonable and compelling. They form the backbone of recommendations currently used by many clinicians for women who cannot or will not formula feed.
Source: Protecting, Promoting and Supporting Breast Feeding available on the WHO Web site at http://www.who.int/dsa/cat98/z10steps.htm.
To assess this query, a team of investigators including Dr. Gray assembled a group of 80 HIV-positive antenatal clinic attendees and evaluated their attitudes and beliefs. At CHBH, voluntary HIV counseling and testing is offered to all women attending the clinic. All clients receive perinatal support through group and peer education and are offered nevirapine, administered according to the HIVNET 012 trial that showed a 50% reduction in the rate of MTCT through the simple administration of two doses of the drug, one to the mother and one to the newborn.
Dr. Gray reported that 58% of women said they make their own decisions about infant feeding -- that is, they do not rely on the counsel of elder women family members or on community opinion. All women are provided formula and encouraged to rapidly wean their infants at 3-6 months. Dr. Gray noted that providers at CHBH emphasize care for the whole family, and that active monitoring and support are extremely important to successfully prevent MTCT. Mixed feeding is still a reality, she noted, and antiretroviral regimens for mothers and babies remain the best prevention tool.
For more information on The Baby-Friendly Hospital Initiative, visit http://home.onemain.com/~ct1008688/bfusa.htm
One of the largest, most significant MTCT studies to date, PETRA compared different lengths of courses of AZT/3TC or placebo. Over 1,750 mothers and 1,800 infants were enrolled in Uganda, Tanzania, and South Africa. Mothers had a median CD4 cell count of 482 cells/mm3; 32% had a cesarean delivery; and 69% breast fed (this study did not control for exclusivity).
Each of the four study arms had 480 mother/child pairs. In the first arm (A), with the longest regimen, mothers received 300 mg of AZT and 150 mg of 3TC beginning in the 36th week of pregnancy and during labor and delivery (intrapartum), and infants received therapy for one week postpartum. In the second arm (B), mothers received 300-600 mg of AZT/3TC during labor and delivery, and infants received one week of therapy. In the third arm (C), mothers received only intrapartum therapy and infants received no therapy. Everyone in the fourth arm (P) received placebo.
Arms A and B -- the two longer regimens in which infants also received therapy -- showed efficacy. Arm C (the intrapartum only group) showed no efficacy compared with placebo. The placebo arm was stopped in February 1998 when positive results from a Thai study of short-course AZT were released. The main conclusion was that giving AZT and 3TC at the onset of labor and for one week postpartum to both mother and child reduced the MTCT rate by 38%, even in areas where a majority of women breast feed.
At the meeting in Kampala, Dr. Lange pointed out that looking at the late efficacy results now available gives a more accurate view of the risk of MTCT through breast feeding. Efficacy was assessed for the three regimens at six weeks and 18 months after birth. At week six, transmission rates were as follows: 5.9% for arm A, 8.9% for arm B, 14.2% for arm C, and 15.3% for placebo. These figures were based on infants who at week 6 had either a positive HIV-1 DNA or RNA test, or both. When the endpoints of infant HIV infection and mortality (deaths) were combined, week 6 rates for the four arms were 7.0%, 11.6%, 17.5%, and 18.1%.
At 18 months, the rate of HIV infection among infants from arms A, B, C, and P were 14.9%, 18.1%, 20.2%, and 22.2%, respectively. When HIV and mortality rates were combined at 18 months, the results were 18.9%, 23.8%, 25.2% and 25.6%. When considering only children who were breast fed, the combined HIV and mortality rates at 18 months were 21.6%, 24%, 27.8%, and 28.2%.
Infant feeding patterns were as follows: 53% of South African and 97% of east African (Tanzanian and Ugandan) mothers ever breast fed, but for different lengths of time. At six weeks postpartum, there was an overall 11% MTCT rate. By 18 months, there was an overall transmission rate of 19%, suggesting an emerging pattern that Dr. Lange called a "catching-up phenomenon," referring to the steeply increasing number of infections postpartum. He concluded that combination AZT/3TC given from week 36 until delivery and then postpartum reduced MTCT of HIV by 61% in a breast feeding population. However, he said, "the benefits diminished considerably after 18 months. [Ed. note: They were almost entirely lost.] The way to maintain the early treatment effect is to target breast feeding and to try to minimize transmission via breast feeding."
However, he added, "breast feeding is only one issue and there are no simple solutions. Prevention of MTCT has to be addressed in the larger context of access to care in developing countries.... We still need better alternatives." Dr. Lange concluded with a personal admonition against dividing the HIV population into "innocent children and guilty adults," calling it "nonproductive."
Social and cultural factors also influence infant feeding choices. The stigma of HIV in developing countries (as well as in the West) continues to be so heavy that many women say they will not formula feed because doing so would be equivalent to disclosing their HIV status. Such women report that they fear recriminations ranging from social rejection and isolation to physical abuse and even life-threatening violence. Even if they do not fear physical violence, many women breast feed simply because it is something that "good mothers" do without fail, a norm that some women find overwhelmingly compelling. Finally, the importance of the psychological bond that is created between mothers and infants through breast feeding has been well described by mothers and researchers around the world. Such a bond is not easily sacrificed, particularly when social and economic pressures also apply.
Nutrition is also relevant to the discussion of infant feeding in developing countries. The type of nutrition provided the infant by breast milk is complete, and in the absence of other nutritional support such as specially formulated replacement foods available in the developing world, some infants fed only formula have failed to thrive. Breast milk promotes digestion; enzymes and other substances in breast milk make digestion somewhat easier and aid in the absorption of nutrients. Manufactured formulas imitate breast milk fairly well, but some of the more complex substances in breast milk are too difficult to manufacture, and still others have not yet been identified. (Formula is a blend of proteins, lipids, and sugars derived from natural substances such as cow's and/or soy milk. It also contains added vitamins and minerals, but there are no synthesized compounds.)
Breast milk also helps fight infection; antibodies passed from the mother help protect the baby from conditions such as ear infections, diarrhea, pertussis (whooping cough), and coughs due to pneumonia and other respiratory infections. Colostrum, the type of breast milk produced for the first few days after birth, is especially rich in antibodies. As a group, breast fed babies tend to have fewer infections than babies fed formula. breast feeding may also reduce allergies. Of course, breast milk is also cost-free, and it is free of bacteria (although the cups and bottles used for expressed breast milk may not be).
Thus for many reasons, breast feeding may be preferable for many mothers in developing countries.
Real-world impediments to exclusive breast feeding were a focal point of the discussion. During the course of the workshop, participants concurred that infant feeding counseling done on a regular basis was very beneficial in promoting healthy feeding methods, and therefore exclusive breast feeding and reduced MTCT. Sore nipples often cause mothers to provide other foods to their infants, but teaching mothers how to help their babies attach to the nipple ("latch") properly, or how and when simply to switch breasts, for example, can help mothers maintain adherence to exclusive breast feeding Ideally, counseling should begin during pregnancy (at or around 26 weeks) and then be provided twice a week during the infant's first two weeks of life to help with positioning and to prevent lapsing into mixed feeding patterns.
Two distinct sets of skills were identified as integral to breast feeding counseling. First, the counselor must be committed to listening and learning from the mother, instead of merely directing information at her. Clinicians and others with real-life experience (in the field) agreed that mothers with HIV reported finding all the information they are given to be overwhelming, and that the manner in which it is delivered is significant. Second, the counselor also must seek to foster confidence in the mother by giving support, not by criticism and didacticism, which undermine the mother's confidence and the ultimate goal of reducing MTCT. In this workshop as elsewhere at the conference, the importance of adequate training and ongoing education for counselors was emphasized.
One participant from Botswana suggested that other health-care workers and even other support workers in clinics be educated adequately so that everyone working in the clinic will understand why unusual measures, such as exclusive breast feeding, are being recommended. For instance, he said, the women who swept the floors were in an excellent position to observe the proceedings in the clinic as they went about their work; but, since their work duties fell outside the PMTCT realm per se, the clinic's procedures were not explained to them. When these women went home, or out into the community, they told people what they saw and gave their interpretations, often spreading confused or misleading information in the community and undermining the program's PMTCT efforts. Educating indirect program participants would help create a broad support network throughout the community at large.
Expressing breast milk (by hand or by pump into a clean glass or hard plastic storage container) for the baby's use when the mother cannot nurse is a new concept to most women. New mothers who have to be away from their babies in order to work often allow babies to be fed alternative foods by caretakers -- one reason for mixed feeding. Yet one participant reported success in a rural setting, where expressed milk was turned over to grandmothers providing child care who then cup fed while the mother was away working during the day. Breast milk should remain viable for 8-10 hours unrefrigerated, and should be stored covered. (If refrigerated, milk is viable for three days and if frozen can be kept for six months; once thawed, breast milk should not be refrozen.)
One participant described the successful use of lay counselors in a PMTCT program in KwaZulu-Natal, South Africa. Another, from India, described women's associations as a good recruitment source for counselors. In Kenya, healthcare workers recruit and train community members, who then work in communities teaching hygiene and family planning, and helping with immunization projects.
Ten years ago, the WHO/UNICEF launched the Baby-Friendly Hospital Initiative (BFHI) to improve breast feeding practices in hospitals. Each hospital must comply with ten steps for training and supporting mothers, pre- and postpartum. Under the BFHI, no donated or subsidized formulas are allowed, in an effort to promote breast feeding over formula feeding. Dr. Vallenas said that in the context of HIV and with the data so far, the question today is "Do we quit the Baby-Friendly Hospital Initiative or do we adopt it?" Under the BFHI, formula would not be available in hospitals. Yet, with respect to the vagaries of HIV policy and formula distribution, the WHO recognizes that formula offers a valuable alternative for some HIV-positive women. Thus, another question is, if formula were to be made available in hospitals and clinics for HIV-positive women, how could clinics and clinicians avoid "spill over"? Spill over, or the creation in the wider community of mixed messages regarding infant feeding choices, could result from misinterpretations of the fact that the formula would be intended for the use of HIV-positive but not HIV-negative mothers. How are individual hospitals managing these potentially confusing health messages?
At Kampala's Mulago Hospital, once pregnant women and new mothers are identified as HIV positive, they receive individualized, one-on-one counseling for as many times as the mother chooses. Follow-up is available to mothers at home, too, should problems develop, whether the mother is formula feeding or breast feeding. In Zambia, mothers who wish to conceal their HIV status mix formula in a private room in the hospital, and then take the formula back to their own rooms (although another participant pointed out that going to the "special room" is tantamount to disclosure of HIV status). A participant from Kenya pointed out that foster parents or women without enough milk often use other types of animal milk, such as cow's, buffalo's, or goat's milk; another participant said that using cow's milk as replacement food is also common and culturally acceptable in India. These observations could lead to a strategy that might help protect mothers' HIV status. This option is still problematic, however, since animal milk may give babies indigestion, sometimes carries disease, and does not have enough protein for human babies.
Another possible solution lies in pasteurizing or heat-treating breast milk whenever possible. At some sites in India and Thailand, heat treatment for breast milk is already used. Mothers there report finding heat-treating easier after the first six months of the baby's life, when they (mothers) are less tired and when family monitoring relaxes. Women report that they find heat-treating breast milk in a glass jar kept inside a bigger pot of water at 60° C practical. Ideally, HIV-positive mothers would use heat treatment within the first six months, from as close as possible to the day of delivery.
Dr. John said, "We try for global recommendations, but will probably never have equally useful solutions. It is incumbent upon us [health-care providers] to share information and strategies, and to adapt as appropriate." Dr. Vallenas added that "the community has to be made aware not to stigmatize women for either formula feeding or for being HIV positive," although she admitted that reducing stigma will probably result from work in other areas such as human rights. She concluded the workshop by suggesting that counselors and health-care workers regard their job as guiding an individual mother's choice, after risk assessment. "Start with what the mother has already thought about, and what she wants," Dr. Vallenas recommended.
Earlier that day, at the Monday plenary session, Dr. Magoni presented a study on the feasibility and effectiveness of a PMTCT program in an urban hospital in Kampala. At the hospital's MTCT service center, nearly 15% of HIV-positive mothers who opted for formula feeding (and no breast feeding) admitted to having also breast fed. Of those who opted for exclusive breast feeding, only 10% adhered. Thus, Dr. Magoni said, "it is clear that the majority of Ugandan women find it very impractical to either exclusively breast feed until six months, or to give only formula. Mixed feeding is a persistent reality." Dr. Magoni further remarked that the failure of women to follow infant feeding recommendations is undermining national efforts to reduce MTCT, and that active monitoring is necessary at PMTCT service centers.
Pope Kosalaraksa, M.D., from the department of Pediatrics at Khon Kaen University in Thailand, agreed that exclusive breast feeding or exclusive formula feeding is culturally "misfit" to many women. Still, hard work in Thailand was able to bring about some social change, and the options of exclusive breast feeding and exclusive infant formula feeding have become more acceptable. Strong leadership and a high level of commitment from the Thai government has made this possible, remarked Anupong Chitwarakorn, M.D., from the Thai health ministry.
A participant from Brazil said that there is "no hype" over breast feeding in Brazil and that it is more of a personal choice. However, since the situation in African countries appears to be very different, the solution for reducing the transmission of HIV to infants also will have to be different. She added that "the birth of a child heralds a variety of cultural, social, and personal ceremonies," many of which surround breast feeding or infant feeding. Stigma and discrimination surrounding HIV/AIDS only complicate the search for a possible solution to prevent HIV transmission from mother to child.
For comprehensive information on the conference, see the Global Strategies Web site at www.globalstrategies.org.
Leslie Hanna is the editor of BETA.
Back to the SFAF BETA Winter 2002 contents page.