Strategies for Preventing Late-Term Vertical HIV Transmission

Jill Cadman

The significant reduction in the rate of transmission of HIV from mother to child (vertical transmission) in developed countries is one of the true success stories of the AIDS epidemic. In many industrialized nations, the rate of vertical transmission has fallen below 5%. In the U.S., the number of HIV positive infants born each year has decreased from 1,000-2,000 in the early 1990s to less than 500 in 1997. Unfortunately, such progress is lacking in developing countries, where over 1,500 children become infected daily through vertical transmission and breast-feeding.

The widespread implementation of the three-part AZT protocol, proven to reduce transmission by two-thirds in AIDS Clinical Trials Group (ACTG) study 076, is the primary reason for the low number of HIV positive children born in industrialized nations. The 076 regimen involves treatment of the mother with AZT beginning at 14 weeks of gestation, intravenous AZT during labor and delivery, and treatment of the newborn with AZT for six weeks (for a thorough report and references, see BETA, September 1994).

Because the 076 regimen is too complex and costly for use in developing countries, an abbreviated, two-part course of AZT preventive therapy was studied overseas and proven quite effective in a Thai clinical trial, in which vertical transmission was reduced by half. A recent report from the New York State Department of Health suggests that even shorter regimens may also be viable alternatives.

The Joint United Nations Programme on HIV/AIDS (UNAIDS) and other public health agencies are working to bring adequate prenatal care, including HIV testing and counseling and supplies of discounted AZT, to women in the developing world. This is a mammoth undertaking in countries where seroprevalence in pregnant women ranges from 10% to over 30%, where 32% of women receive no prenatal care, and where 43% give birth without an attendant, much less a physician.

Efforts to address other risk factors associated with vertical transmission have yielded impressive results, but again, these primarily benefit women living in industrialized countries. Encouraging reports on the effectiveness of cesarean sections (C-sections) were presented in June 1998 at the 12th World AIDS Conference in Geneva. This type of intervention is simply not feasible in low-income countries. Additionally, in countries with no safe alternatives to breast-feeding, the impact of transmission through breast milk must be considered. Estimates from the Centers for Disease Control and Prevention (CDC) suggest that at least 273,000 infants worldwide are infected through breast-feeding each year.

Targeted Interventions at the Time of Delivery

The original 076 protocol was designed before researchers had pinpointed the most likely time for vertical transmission to take place. The rationale of the trial was to interrupt transmission at each stage at which it might occur -- during pregnancy (in utero), during delivery (intrapartum), and after birth (postpartum). Since many studies now indicate that roughly 60% of vertical transmissions occur at or shortly before birth, it would seem that interventions at this time would have the best chance of success.

Confirmation of the efficacy of such targeted interventions came from the recently completed Thai trial of short-course AZT therapy, in which the median length of treatment of the mothers was 25 days, as opposed to 14 weeks in the 076 protocol. The abbreviated AZT regimen proved to have a potent preventive effect, although it was somewhat less effective than the longer course of AZT employed in the 076 protocol.

It may be possible to delay AZT treatment even longer and still achieve significant reductions in the rate of vertical transmission. In the November 12, 1998 issue of the New England Journal of Medicine, Nancy Wade and colleagues report the results of a chart review of 939 HIV-exposed infants conducted by the New York State Department of Health. Data on the timing of perinatal AZT treatment were collected and analyzed. Mothers who received abbreviated regimens rather than the recommended 076 regimen did so because of limited prenatal care or choice.

Even when treatment did not begin until labor, transmission of HIV was often prevented. More interesting, the rate of vertical transmission was also significantly reduced when AZT was administered only to the infants, beginning within 48 hours after birth (this effect was not seen in infants who started AZT therapy more than 48 hours after birth). When treatment was begun in the prenatal period, the rate of vertical transmission was 6.1%; when begun intrapartum, the rate was 10%; and when begun within the first 48 hours of life, the rate was 9.3%. In the absence of AZT treatment, the rate of vertical transmission was 26.6%.

If treatment of the baby alone is confirmed to reduce vertical transmission, such information would have important public health implications in parts of the world where the high cost of antiretroviral agents remains an obstacle to their use. The drug-related cost of such prophylactic treatment of infants would be a fraction of that for even a short course of maternal treatment. In the U.S., the standard of care remains the full three-part 076 regimen, which has the greatest potential to reduce vertical transmission. However, the results of this study indicate that it still worthwhile to initiate AZT therapy during the intrapartum period or immediately after birth.

Elective Cesarean Sections

Researchers and clinicians are looking beyond drug therapy to surgical interventions targeted at the crucial period of labor and delivery. It has long been suggested that HIV positive women who deliver by elective (non-emergency) C-section are less likely to transmit the virus to their newborns than women who deliver vaginally. Such C-sections must be performed prior to the rupture of uterine membranes (breaking of the "bag of waters"). As long as the membranes are intact, pathogens such as HIV may have a more difficult time entering the uterus and infecting the fetus. The virus gains greater access to the infant after the membranes rupture and labor begins. Studies have shown that the risk of vertical transmission nearly doubles when the membranes rupture more than four hours before delivery. Elective C-sections, performed while the membranes are still intact, might prevent the infant from being exposed to maternal blood and secretions while passing through the birth canal.

Until now, controversy has surrounded studies attempting to demonstrate the effectiveness of elective C-sections in reducing vertical transmission, due to various confounding factors such as concurrent use of antiretroviral therapy and rupture of membranes prior to C-section. Numerous presentations at the Geneva AIDS conference unveiled the most convincing data so far on enhanced reduction in vertical transmission in women who were both taking antiretroviral treatment and who chose elective C-sections before their membranes ruptured. The benefit of this intervention occurred in the absence of AZT therapy, and it conferred additional protection when combined with AZT.

European investigators reported some impressive results. A large French cohort study found that among 902 women who were treated with AZT, elective C-section resulted in a rate of vertical transmission of only 0.8%. In comparison, emergency C-section resulted in a 11.4% rate of transmission, and the rate for normal vaginal delivery was 6.6%. The Swiss Neonatal Group presented data from an ongoing nationwide prospective study. Among 45 women who completed the full 076 regimen and had elective C-sections, there were no cases of vertical transmission.

The German Perinatal Cohort study of 255 mother-child pairs reported that since 1994, elective C-sections were performed in addition to AZT treatment in 80 HIV positive pregnant women. The rate of transmission in this group was reduced to 2.5%. For women who had elective C-sections but did not take AZT, the transmission rate was 10.8%. Transmission was 7% in AZT-treated women who had vaginal deliveries. From Italy, results were reported of a five-year international trial of randomizing pregnant HIV positive women taking similar antiretroviral regimens to either elective C-section at 38 weeks or spontaneous vaginal delivery. Of the 133 children delivered by C-section, 3% contracted HIV compared to 10.3% of 132 infants delivered vaginally.

Finally, a large survey funded by the National Institutes of Health (NIH) analyzed data from five European and ten North American prospective studies on a total of 8,533 mother-child pairs. After adjusting for use of antiretroviral therapy (i.e., AZT), maternal disease progression, and birth weight, risk of vertical transmission was reduced by over 50% with elective C-section compared to other modes of delivery (vaginal delivery and emergency C-section). In women who received AZT therapy, the rate of transmission was 2% with elective C-section and 7.3% with other modes of delivery. In women who were not taking AZT, the transmission rates were 10.4% with elective C-section and 19% with other modes of delivery.

C-sections performed after labor has begun have not been shown to reduce vertical transmission. For this reason, HIV positive women must decide what do before labor so that the procedure can be scheduled for an appropriate time at the end of gestation (pregnancy) but before the onset of labor. Pregnant HIV positive women and their primary care providers will have to consider a number of factors before deciding what mode of delivery to choose.

Women who are taking antiretroviral drugs, especially those on maximally suppressive multidrug therapy, already have a low risk of vertical transmission. The additional reduction in transmission that was seen when C-sections were added to AZT may be outweighed by the added discomfort and potential complications associated with this procedure. According to Lynne Mofenson, MD, of the NIH, "My conclusion is that the data are important to present to patients. However, if I had a woman with a CD4 count of 500 cells/mm³ and an undetectable viral load who was on at least AZT, if not combination therapy, I think that the benefit of elective C-section would be minimal compared to the potential risk of operative complications. I think that rather than making a global recommendation based on this data, there needs to be an individualized risk assessment and discussion with each woman. I would not take the data and say, universally, that this means every woman needs to be sectioned." The maximum benefits from C-sections were seen only in women taking AZT; those not taking AZT who had elective C-sections were still at elevated risk for transmitting HIV to their infants.

While C-sections are generally quite safe in industrialized countries, some studies have found that HIV positive women have an increased risk of post-operative complications. The risk for pregnant women in developing countries may be even greater than in more affluent nations. Furthermore, surgical procedures may not be an option for the overwhelming majority of women in resource-poor countries with limited healthcare infrastructures and budgets.

Mode of Infant Feeding

The average rate of vertical transmission without any type of intervention is around 25%. However, rates of transmission differ significantly between developed and developing countries. Rates vary from less than 14% in Europe to 45% in sub-Saharan Africa. In developing countries, breast-feeding may be largely responsible for the higher rates. Studies indicate that more than one-third of HIV positive infants are infected through nursing in developing countries. These studies suggest an average risk of late postnatal transmission through breast-feeding of one in seven children.

At the Geneva AIDS conference, an international multicenter metanalysis conducted by the Ghent International Working Group evaluated the extent of the risk of postnatal transmission of HIV in industrialized and developing countries. Data was pooled from eight prospective studies that included breast-fed and formula-fed infants. The children at risk for postnatal transmission were uninfected and between the ages of 2.5 and 18 months. Fewer than 5% of 2,807 children from industrialized countries were breast-fed, and no cases of late postnatal transmission occurred. In contrast, there were 49 cases of late postnatal transmission in 902 children from developing countries, where breast-feeding is the established and accepted method of infant feeding. This amounts to an additional risk of vertical transmission due to breast-feeding of 3% per child-year of breast-feeding, constituting a significant risk to the nursing infant.

Information on the exact timing of infection was available for 20 of the 49 infants. The risk of transmission clearly increased with the duration of breast-feeding. Postnatal transmission rates by age were less than 1% between two and six months, 2.5% at 12 months, 6.3% at 18 months, 7.4% at 24 months, and 9.2% at 36 months. Given these rates, postnatal transmission in the study would have occurred in a minimum of none or a maximum of two cases if breast-feeding had ceased at four months, and three or four cases if breast-feeding had ceased at six months of age. (The absolute number would be larger when applied to larger numbers of infants.)

While the ability of short-course AZT to reduce vertical transmission provides hope for women in developing countries, the Thai study does not address the efficacy of the regimen among women who breast-feed. This is currently being investigated in other studies. It is likely that AZT will provide some degree of protection, but it will probably be less than the protection the drug provides to infants who are not breast-fed. The margin of difference in transmission rates between those treated and those not treated will narrow in cases where women breast-feed their infants.

The greatest reduction in vertical transmission can only occur when an integrated prevention program that combines AZT and safe alternatives to breast-feeding is implemented. According to UNAIDS, it may be impractical in some countries to simultaneously implement access to AZT and access to safe alternatives to breast-feeding. In these situations, the implementation of one prevention component should not be delayed until the other is feasible. Furthermore, if a woman chooses not to use both AZT and safe alternatives to breast-feeding, she should still have access to the intervention of her choice, and be supported to carry out the use of this intervention safely and effectively.

In certain societies where breast-feeding is the established method of infant feeding, not breast-feeding may be interpreted as an indication of a woman's HIV positive serostatus. Mothers in these cultures will need particular support if they choose not to breast-feed. Healthcare workers will need to be trained to provide appropriate guidance on how to prepare other foods properly and as safely as possible, how to clean utensils, and how to introduce complementary foods. Complementary foods and micronutrient supplements may be needed up to two years of age. Cup-feeding is considered safer and more hygienic than bottle-feeding.

If infant formulas are provided, strict controls must be maintained according to the provisions of the 1981 International Code of Marketing of Breast-Milk Substitutes. This mandate is designed to prevent commercial pressure for artificial feeding, including protecting parents from inappropriate promotion of breast-milk substitutes by manufacturers and distributors of formula. In the 1970s, the misuse of infant formula in developing countries with poor sanitation was blamed by opponents for killing one million babies per year. Advocates for breast-feeding charged that manufacturers of formula, in an effort to promote their products, disregarded dangers such as contaminated water, or the possibility that poor families might over-dilute the products, leaving infants malnourished

At an oral session entitled "Mother-to-Child HIV Transmission: Infant Feeding," delivered at the Geneva AIDS conference, several presenters stressed the need to prevent a "spillover" effect of replacement feeding which might undermine breast-feeding among HIV negative women. Felicity Savage-King of the World Health Organization (WHO) stated that formula should be regarded as a medicinal product. If governments provide breast-milk substitutes, supplies should be procured centrally and made available by prescription only. HIV positive mothers who choose not to breast-feed must be assured of breast-milk substitutes for at least six months. Distribution of the formula should be linked to follow-up care. The growth of formula-fed children must be monitored on an ongoing basis. Even if they are not HIV positive, such children are likely to have a higher incidence of diarrhea and respiratory infections. Infant feeding should be part of a continuum of care and support services for HIV positive women and their families.

Glenda Gray, MD, a pediatrician from Baragwanath Hospital in Soweto, South Africa, presented information from the PETRA study of over 200 HIV positive mothers receiving care in the hospital clinic. In this oral presentation, Gray did not discuss transmission data, which is currently being analyzed and is expected by the end of the year. Rather, she focused on the sorts of infant feeding choices that a group of pregnant women made when offered subsidized formula and counseling on options. Compared to women in the clinic who were not enrolled in the study, participants were more likely to formula-feed. Participants who received more than one counseling session were more likely to opt for exclusive formula-feeding. The availability of subsidized formula increased the likelihood that women would not breast-feed, and also decreased the likelihood of mistakes in mixing the formula.

Of the women who chose to formula-feed, 33% mixed the formula perfectly, 25% demonstrated some discrepancy in formula preparation, and 10% demonstrated major discrepancies. Frequency of incorrect preparation of formula and its effect on the infant's growth warrants attention and further study. Gray speculated that if women have access to an ongoing, reliable source of formula, they are less likely to water it down in an effort to make it last longer. Other presenters stressed the need for easy-to-understand instructions in local languages, targeted at those with low literacy levels. Finally, in low-income countries where formula is not subsidized, its cost may be prohibitive for the majority of the population.

Other infant feeding options were discussed during the oral session. Realistic and sustainable options in many settings may eventually include the use of home-prepared formulas made from animal milks, typically from cows, goats, buffalo, or sheep. The composition of animal milk is different from that of human milk, may lack micronutrients (especially iron), and should be modified for infants. Expressed milk (breast milk that has been pumped by the mother) can be heat-treated to kill HIV. Heat-treated breast milk is nutritionally superior to animal milks, but heat treatment reduces the levels of anti-infective factors (antibodies). To pasteurize milk, it should be heated to 62.5° C (144° F) for 30 minutes. Alternatively, it can be boiled and then cooled immediately. Expressing and heat-treating breast milk is time-consuming. In addition, more studies are necessary to determine exactly how the heat treatment affects the nutritional value of the milk for the infant.

Another option, discussed by Savage-King, originating not from research but "common sense" brainstorming, is surrogate wet nursing by older relatives who have tested HIV negative. Older relatives who are HIV negative, the theory goes, may be less likely than younger relatives to engage in behaviors that might put them at risk for HIV. Finally, early cessation of breast-feeding reduces the risk of postnatal transmission by reducing the length of time the infant is exposed to HIV. The optimum time for early weaning is unknown. However, the most risky time for artificial feeding in environments with poor hygienic conditions is the first two months of life.

Breast-feeding is the preferred way to feed an infant in many developing regions. However, if the mother is HIV-infected, it may be preferable to replace breast milk. Given the importance of breast-feeding to infant health, but recognizing the part breast milk plays in vertical transmission, UNAIDS, the WHO, and the United Nations International Children's Education Fund (UNICEF) recommend that appropriate alternatives to breast-feeding be made available and affordable for women who are HIV positive. Furthermore, they recommend that efforts continue to promote and support breast-feeding by women who are HIV negative or of unknown status. All pregnant women should have access to voluntary HIV testing and counseling that includes information on vertical transmission and infant feeding. The risk of replacement feeding should be less than the potential risk of HIV transmission through infected breast milk, so that infant illness and death from other causes do not increase. Otherwise, there is no advantage in replacement feeding.

Vertical transmission is considered by some to be a preventable occurrence. In order for all women to benefit from the advances that have been made in this field, major initiatives are needed to provide education, health care, treatment, and empowerment to underprivileged women.

Jill Cadman is a Research Associate and medical writer at the Bentley-Salick Medical Practice in New York City. She is the former Associate Editor of GMHC's Treatment Issues and serves on the board of directors of New York City's Community Research Initiative on AIDS.

References

Gray, G. Attitudes and behavior of HIV-infected women on infant feeding. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Oral session 314.

HIV and infant feeding: guidelines for health care managers and supervisors. Prevention of Mother to Child Transmission. Jointly issued by UNICEF, UNAIDS, and WHO. Page 30. Revised June, 1998.

Landesman, S. and others. Obstetrical factors and the transmission of human immunodeficiency virus type 1 from mother to child. N Engl J Med 1996 Jun 20;334(25):1617-23.

Leroy, V. and others. Late post-natal mother-to-child transmission (LPT) of HIV-1: international multicentre pooled analysis. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Abstract 456/23266.

Lutz-Friedrich, R. and others. Combining ZVD treatment and elective cesarean section reduces the vertical transmission of HIV 1 below 3% in the German perinatal cohorts. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Abstract 23291.

Mandelbrot, L. and others. Decreased perinatal HIV-1 transmission following elective cesarean delivery with zidovudine treatment. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Abstract 23272.

Savage-King, F. Breast-feeding programmes and HIV questions. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Oral session 310.

Semprini, A.E. and others. The incidence of complications after cesarean section in 156 HIV positive women. AIDS 1995 Aug;9(8):913-7.

Wade, N.A. and others. Abbreviated regimens of zidovudine prophylaxis and perinatal transmission of the human immunodeficiency virus. N Engl J Med 1998 Nov 12;339(20):1409-14.

DT 19990110
DOCN BE990107