infection in women has the addition benefit of preventing perinatal transmission. However, many women are already infected, and significant numbers will continue to be infected despite current intervention programs. For these women, other strategies are needed to reduce transmission of HIV to their newborns; several such strategies are described in the remainder of this section, although there are problems with each that limit their applicability in sub-Saharan Africa.
One approach may be the avoidance of breastfeeding by HIV-infected mothers. There are two problems with this strategy: large numbers of pregnant women would have to be tested and counseled for HIV infection and counseled as well in bottle feeding, and safe and affordable alternatives to breast milk would have to be easily available. This approach is not feasible in most countries in sub-Saharan Africa except for an elite urban minority, for economic and sociocultural reasons. The current recommendation by WHO/UNICEF is that all countries, irrespective of HIV infection rates, should continue to promote and support breastfeeding because of its impressive nutritional, immunological, and child-spacing benefits (World Health Organization, 1992a). In particular, WHO (1992a) recommends that in areas such as sub-Saharan Africa, where the primary causes of infant mortality are infectious disease and malnutrition, breastfeeding should be encouraged for all women, regardless of HIV status. The rationale for this recommendation is provided by studies showing that an infant's risk of becoming HIV infected through breast milk is lower than its risk of dying of other causes if deprived of breast milk.
Another approach to prevention of perinatal transmission is the provision of voluntary and confidential counseling and testing services to women who may then choose to avoid or terminate pregnancy. This alternative is unlikely to be feasible in African countries where access to HIV testing is poor, abortion is often illegal and frequently unsafe, and fertility is highly valued.
A third approach is the delivery by caesarean section of infants of HIV-infected mothers in order to minimize the risk of neonatal infection during vaginal delivery. While caesarean section delivery has proven effective in prevention of other viral infections (Minkoff and Duerr, 1994), no definite conclusions can be drawn on the basis of available data concerning its potential protective effect with regard to perinatal HIV transmission. Furthermore, even if the efficacy data were to point to caesarean section as an effective prevention strategy, the increased risk and high cost associated with the procedure render it an infeasible alternative in all but a handful of teaching hospitals in the region.
Another possible approach to reducing the risk of infection during vaginal delivery is vaginal lavage, or washing with a microbicidal agent during delivery. Vaginal lavage has been shown to lower rates of neonatal group B streptococcal infection in newborns and may prove to be helpful in the prevention of HIV transmission during delivery (Minkoff and Duerr, 1994). Potential problems, however, include fetal toxicity and genital mucosal irritation. More research is