Since 1994, in the USA and Europe, antiretroviral and obstetric interventions and the avoidance of breastfeeding have dramatically reduced mother-to-child HIV-1 transmission [1,2]. However, most of the 17.6 million HIV-1 infected women live in resource-poor countries and do not have access to antiretroviral medication or safe alternatives to breastfeeding [3,4]. The United Nations Program on HIV/AIDS (UNAIDS) estimates that more than 800 000 children continue to be infected with HIV-1 each year, with more than one-third of these infections attributed to breastfeeding transmission [3,5]. Sub-Saharan Africa has been most severely affected by the dual burdens of the HIV-1 pandemic and poverty.
The risk of mother-to-child HIV-1 transmission in predominantly breastfed populations is estimated to range from 25% to 48% . In Africa, where prolonged breastfeeding into the second year of life is the norm, the HIV-1 pandemic has created a vexing policy dilemma concerning infant feeding . In poor, unsanitary environments the risk of HIV-1 transmission through breastfeeding must be weighed against the substantial benefits of breastfeeding, including health and survival benefits to the child as well as contraceptive, economic and social benefits to the mother [7,8]. Critical research gaps remain regarding the risk and mechanism of HIV-1 transmission through early and prolonged breastfeeding . This information is vital for the development of interventions to reduce HIV transmission through breastfeeding and promote child survival. It would also assist policymakers, health professionals, and HIV-infected women to make context-specific decisions about breastfeeding .
Recent randomized clinical trials have assessed the efficacy of a number of relatively simple and inexpensive perinatal antiretroviral interventions as well as helped clarify the rate and timing of HIV-1 transmission attributable to breastfeeding. The aim of this review is to analyze their collective findings qualitatively and quantitatively, and use these data to suggest ways to strengthen global perinatal HIV-1 prevention efforts and direct perinatal research priorities. As the trial interventions differ, it is not appropriate to pool the trials’ results and use statistical methods of analysis; we report the randomized trial results but make observational comparisons between the trials.
We reviewed data from randomized clinical efficacy trials of antiretroviral or infant feeding interventions to prevent mother-to-infant HIV-1 transmission among breastfeeding African populations. In addition to data derived from peer reviewed publications we included the most recent trial data presented at international HIV conferences as we felt the research and public health implications of these data are timely and compelling. We identified eligible trials by searching the MEDLINE, AIDSLINE, and International HIV Conference databases for the following keywords: disease transmission, vertical; perinatal; HIV infections; and breastfeeding. Titles and abstracts were scanned to identify eligible clinical trials, and a further search was conducted using authors’ names.
Study methods and variables known to be associated with HIV-1 transmission such as maternal CD4 count, duration of membrane rupture, mode of delivery, and breastfeeding were described for each trial population [4,10–13]. Unfortunately insufficient data were collected to accurately define breastfeeding practices as exclusive or mixed feeding but the literature suggests that mixed breastfeeding would have been the predominant cultural practice. These factors should be considered when assessing findings and comparisons within and across studies.
Trial endpoints included cumulative HIV-1 transmission rates and mortality. The relative efficacy of interventions for reducing early and late HIV transmission risk and/or increasing HIV-free survival is determined from these endpoints. Relative efficacy of an intervention is defined as the preventive fraction in the group exposed to an intervention compared to the referent group (exposed to a placebo or to another intervention), where RExp is the estimate of the rate of interest in the exposed group and RRef that in the referent group: E = 1 – (RExp/RRef). Incremental HIV-1 transmission rates for each trial arm were defined as the additional cumulative transmission reported within each testing interval.
Infant HIV-1 infection status was determined by sensitive virologic tests at birth and at multiple subsequent time points. Some data suggests that current RNA PCR tests may have slightly higher sensitivity than DNA PCR tests in detecting infant HIV infection at birth [14,15]. Nucleic acid amplification HIV testing techniques at 6 weeks or more after birth should detect all infant HIV-1 infection due to intrauterine and intrapartum exposures . Thus, cumulative HIV-1 transmission at 6 weeks among breastfed infants was interpreted to represent infections due to intrauterine and intrapartum exposures as well as those due to very early breastfeeding exposure.
Crude estimates of the average monthly hazard of infant HIV infection related to breastfeeding transmission were derived from the available data. The estimates of very early breastfeeding transmission are derived from the difference in incremental HIV-transmission between breastfed and non-breastfed infants in the first postnatal period (between birth and 6–8 weeks) as infants already infected at birth are not at risk of infection through breast-milk exposure. Based on the assumption that almost all new infections after the first 6 weeks were attributable to breastfeeding transmission, crude estimates of the average monthly hazard after 6 weeks were calculated by dividing incremental HIV transmission by the testing interval in months.
Six randomized clinical efficacy trials evaluating interventions to prevent mother-to-child HIV-1 transmission in breastfeeding populations were identified. These were all conducted in urban African settings. One trial evaluated infant feeding practices (breastfeeding versus formula feeding) in a Kenyan population and published results to 24 months . The other five trials evaluated oral antiretroviral interventions. Placebo-controlled zidovudine (ZDV) trials were completed in west Africa by the RETRO-CI and DITRAME trial groups with published individual and pooled results to 24 months [18–21]. The HIVNET-012 trial group evaluated nevirapine (NVP) versus an ZDV regimen in Uganda with published early results and presented late results through 18 months [22–24]. Two trials evaluated ZDV/lamivudine (3TC) regimens: the PETRA trial group evaluated the timing of various ZDV/3TC regimens against placebo in Tanzania, Uganda, and South Africa with follow up data through 18 months; [25,26] and the SAINT trial group evaluated NVP versus an ZDV/3TC regimen in South Africa and has presented preliminary results through 8 weeks .
The five-antiretroviral trials differed from each other in terms of antiretroviral drugs, dosage and/or timing (Table 1). Three oral antiretroviral regimens were evaluated: ZDV alone, NVP alone, and the combination of ZDV and 3TC. RETRO-CI, DITRAME, and PETRA Arm-A included an antenatal component from 36 weeks gestation or later. All interventions arms included an intrapartum component. PETRA Arm-C was exclusively an intrapartum intervention, while HIVNET-012, PETRA Arm-B and SAINT included intrapartum and postpartum prophylaxis. DITRAME and PETRA Arm-A included antenatal, intrapartum and postpartum components. The infant feeding trial was conducted in Nairobi, Kenya, among HIV-1 infected women with access to safe water who had agreed to be randomly assigned to either breastfeed or to formula feed their baby.
All sites used sensitive nucleic acid amplification tests with primers to detect regional HIV-1 subtypes to detect early infant infection. Most sites used nested DNA PCR; however, HIVNET 012 used RNA PCR exclusively. The five-antiretroviral trials used Kaplan–Meier survival analysis to calculate HIV-1 transmission rates; the Kenya breastfeeding trial described a modified analytic technique, and the pooled west African analysis incorporated weaning as a competing event. The different analytic treatment of weaning means that the Kenyan, Ugandan HIVNET and PETRA analyses estimate late transmission rates in the whole study population with their prevalent infant feeding practices, whereas the west African analysis describe late transmission rates in the study population at risk of late HIV-1 transmission (that is among infants exposed to breast-milk) .
Table 2 summarizes the clinical and demographic characteristics of the trial cohorts. There were substantive differences between the trial populations in key risk factors for mother-to-child HIV transmission such as immunologic status, mode of delivery, duration of membrane rupture and breastfeeding practices; these differences were most marked between the South African participants (of PETRA and SAINT) and the other study populations. Not all data points have been reported for the PETRA and SAINT trials.
Median CD4 cell count at enrollment ranged between 407 × 106/l and 549 × 106/l. The proportion of women with CD4 cell counts < 200 × 106/l ranged from 6% to 16%. Generally, the two west African trial populations (DITRAME and RETRO-CI) had less advanced disease than did the populations in the other trials, according to immunologic markers. The proportion of Cesarean deliveries varied greatly and ranged from 1.5% to 32%. The proportion of women with duration of ruptured membranes > 4 h was similar between the two west Africa trials and the Kenya trial (range, 27–29%) but was lower among Ugandan HIVNET participants (14%).
Breastfeeding patterns also varied markedly by region. Breastfeeding was highly prevalent and generally continued into the second year of life at the west and east Africa sites. Breastfeeding was initiated by more than 96% of the women participating in the breastfeeding arm of the Kenya trial and in the antiretroviral trials at the west and east Africa sites. In contrast, few South African trial participants initiated breastfeeding (53% PETRA and 40% SAINT), and did so for a much shorter time. In the Kenya trial, 30% of the women randomized to formula were non-adherent and breastfed their infants.
Fig. 1 shows HIV-1 transmission detected by 6–8 weeks across the trials, reflecting cumulative intrauterine, intrapartum, and early breastfeeding transmission and the time when the antiretroviral interventions are targeted to reduce HIV transmission. As characteristics of the study populations, the ZDV regimens and the HIV transmission rates were very similar between the two west African trials (DITRAME and RETRO-CI), we will report the pooled results . Similar early transmission rates were measured across trials for the west Africa pooled placebo arm, the Kenya breastfeeding arm, and the Uganda HIVNET ZDV arm (intrapartum and 1-week neonatal ZDV), with point estimates ranging from 19.9% to 24.8%. The PETRA study placebo arm had a lower point estimate of 15.3%, consistent with a probable reduction due to fewer vaginal deliveries and reduced breastfeeding exposure.
Two antiretroviral interventions did not demonstrate measurable early efficacy: intrapartum ZDV/3TC alone (PETRA-C) resulted in an HIV-1 transmission rate similar to that of the PETRA placebo arm (P = 0.74), and the HIVNET intrapartum and 1-week neonatal ZDV arm was significantly inferior to the NVP arm (P = 0.0063). While the cumulative HIV transmission rate documented in the HIVNET ZDV arm was very similar to that of the pooled west African placebo arm and the Kenyan breastfeeding arm, in the absence of a placebo comparison, the efficacy of the HIVNET ZDV regimen cannot be determined. If this ultra-short ZDV regimen also reduced transmission, then the risk reduction attributable to NVP would be underestimated. For simplicity, due to the very early discontinuation of the placebo arm in the HIVNET trial and the relative lack of efficacy of the HIVNET ZDV arm we will refer to this arm as a referent arm, together with other trials’ placebo arms and the Kenya breastfeeding arm.
The randomized trial data showed that six interventions were efficacious in preventing early HIV-1 transmission; the magnitude of the efficacy estimates were substantial and ranged from 41% to 63% with statistical significance at the P = 0.05 level. Efficacy of the antiretroviral intervention arms at 6 weeks was compared with placebo for the pooled west Africa trials [41%; 95% confidence interval (CI), 17–58%], PETRA-A (63%; P = 0.001) and PETRA-B (42%; P = 0.025). The early efficacy of the HIVNET NVP regimen was 42% relative to the intrapartum and neonatal ZDV arm (95% CI, 13–62%). The Kenyan formula arm's efficacy was 51% relative to the breastfeeding arm in the Kenya infant feeding trial but this result is difficult to interpret given the unexpectedly low rate in the formula arm. The early transmission rates reported for the SAINT trial are consistent with the Ugandan and PETRA data and suggest the short intrapartum and 1-week postpartum NVP and ZDV/3TC regimens have similar efficacy .
Fig. 2 shows incremental HIV-1 transmission as a proportion of cumulative transmission for the trials with late HIV transmission data. At birth, HIV-1 transmission point estimates generally ranged from 6.6% to 10.3%; however, there was an unusually low rate of 3.1% in the formula arm of the Kenyan trial. Statistically significant differences within the trials attributable to effective interventions were measurable by 6 weeks. Between 6 weeks and 3 months, incremental transmission was < 2.1% in the pooled west African and HIVNET 012 trial arms and was 3.5% in the formula arm and 4.6% in the breastfeeding arm of the Kenyan trial. Between 3 and 6 months, incremental HIV transmission remained low despite high breastfeeding prevalence; these were between 0 and 3.5% for the pooled west Africa and Kenya trial arms (data not shown). Of note, much of the HIV transmission in the Kenya formula arm occurred after 6 weeks and is presumed to reflect non-adherence to formula and transmission due to breastfeeding.
All trial arms documented substantial postnatal HIV transmission after 3 months; this accounted for one-quarter of the total 2-year HIV-transmission in the west Africa pooled ZDV arm and one-third in the Kenyan breastfeeding arm (Fig. 2). In the Ugandan HIVNET trial, late transmission was lower in the NVP arm than in the ZDV arm, hence the rate difference and relative efficacy estimates were maintained over time; the rate difference was 8.2% at 6 weeks and 10.1% at 18 months and relative efficacy was 41% (95% CI, 16–59%). However, in the pooled west African and PETRA trials the opposite pattern was observed with increased late transmission in the efficacious intervention arms compared to placebo. For the west African pooled ZDV trial the rate difference estimate decreased from 10.1% at 6 weeks to 7.8% at 2 years, resulting in a 2-year efficacy estimate of 26% (95% CI, 2–44%). For PETRA-A the rate difference estimate decreased from 9.6% at 6 weeks to 7.3% at 18 months, and from 6.4% to 4.1% for PETRA-B which resulted in a decrease in relative efficacy at 18 months that was not statistically significant at the 0.05 level. The SAINT trial was not designed to assess late efficacy.
We have summarized crude estimates of transmission from breastfeeding as the average monthly hazard of new infant infections (Fig. 3). There is very limited data from which to estimate early breast-milk associated transmission; the Kenyan infant-feeding trial represents the only randomized trial data but 30% of women randomized to formula-feed were non-compliant. In the SAINT trial the investigators reported the infant feeding practices of participating women and presented an observational comparison across the two intervention arms. This limited data support a higher rate of breastfeeding transmission during the first several weeks of life than after 3 months of life, however no randomized data is available which quantifies early breast-milk associated transmission in the presence of an effective antiretroviral intervention.
The composite outcome of HIV-free survival is a superior outcome measure as, in addition to HIV-1 transmission, it also captures the adverse effects of formula use and/or early weaning on infant survival. Late HIV-free survival results are similar to the HIV transmission results; in the HIVNET study, 18-month HIV-free survival was 79.3% in the NVP arm and 69.3% in the ZDV study arm (P = 0.0048) whereas for PETRA, 18-month HIV-free survival showed no statistically significant difference between intervention arms and placebo [25,26]. The results of the Kenya breastfeeding trial suggest that formula use has both risks and benefits. The probability of HIV-free survival for 2 years was 58% in the breastfeeding arm and 70% in the formula-feeding arm (P = 0.01). However, 2-year cumulative child death rates were disturbingly high in both arms: 20.0% in the formula-feeding arm and 24.4% in the breastfeeding arm (P = 0.3). Much of the infant morbidity and many of the deaths in the formula-feeding arm occurred early; 19% of the deaths occurred in the first 6 weeks and 54% by 6 months. In the breastfeeding arm, these proportions were 4% and 36%, respectively [28,29].
Consistent with previous studies, advanced HIV disease (as measured by CD4 cell count and viral load) were found to be strong independent risk factors for vertical transmission [21,23,24,26]. The group of pregnant women with advanced disease includes a substantial minority of trial participants; one-third of the HIVNET participants and approximately one-quarter of the west African participants had CD4 cell counts <350 × 106/l and more than half the vertical HIV transmission occurred from this subset of women. Further, CD4 cell count at baseline was a strong predictor of postnatal transmission; Fig. 4 shows that among women in the west African trials with CD4 cell counts > 500 × 106/l little late postnatal breastfeeding transmission occurred (approximately 1.4% in the pooled ZDV arm) but this was substantial in those with CD4 counts < 500 × 106/l (approximately 14.0% in the pooled ZDV arm) .
These HIV intervention trial results have profound and immediate implications for international child survival efforts and future research priorities. In Africa and other resource-constrained settings, current HIV-1 transmission rates are unacceptable when the combination of a safe, efficacious, inexpensive antiretroviral such as NVP and earlier weaning is likely to decrease vertical transmission to approximately 15% at 2 years. Further, as advanced maternal HIV disease is associated with poor maternal outcomes and the highest risk of vertical HIV transmission, the provision of comprehensive HIV care with combination antiretroviral therapy to this subgroup would be of dual benefit for maternal and child health.
The ultra-short regimens of NVP or ZDV/3TC (intrapartum and 1 week postpartum) were found to have a similar efficacy to those of ZDV or ZDV/3TC initiated at 36 weeks gestation. Interventions begun intrapartum, particularly NVP, have advantages of feasibility and cost as well as being accessible for women presenting late in pregnancy . These interventions target the critical period around labor and delivery when the highest rate of mother-to-child HIV-1 transmission is known to occur .
The HIVNET, PETRA and SAINT trial results provide strong evidence that intrapartum and early postpartum antiretroviral prophylaxis with either ZDV/3TC or NVP reduce perinatal HIV-transmission substantially among breastfeeding populations. Because these regimens included 1-week postpartum therapy to the neonate and to the mother (NVP because of the prolonged half life of the maternal dose during labor) [31,32] the relative importance of the neonatal and maternal components cannot be determined. The substantial decrease in maternal viral load, and/or changes in viral infectivity in early breast-milk may plausibly decrease the risk of transmission during early breastfeeding.
Observational data from non-breastfed US HIV-infected cohorts showed a trend between prolonged neonatal ZDV prophylaxis given soon after birth and reduced HIV-transmission [33,34]. However, no trial has been conducted to evaluate the efficacy of postpartum ZDV (as the designs of the HIVNET-012 trial and the Thai ZDV equivalency trial were modified after interim reviews) [22,35]. The DITRAME ZDV regimen (with an additional 1-week maternal ZDV component) resulted in a very similar transmission profile to that for the RETRO-CI ZDV regimen. The HIVNET ZDV regimen (intrapartum and 1-week neonatal ZDV) was inferior to NVP and resulted in very similar rates to those reported for the west African placebo arms and the Kenyan breastfeeding arm. These observational comparisons, while not conclusive, suggest that no substantial effect is achieved through the addition of short maternal postpartum ZDV prophylaxis, or ZDV in the intrapartum and early neonatal period.
Providing effective antiretroviral prophylaxis for HIV-1 infected pregnant women during labor and after delivery should be a minimum goal in all settings. Further trials will also evaluate interventions to reduce breastfeeding transmission, such as extended postpartum antiretroviral prophylaxis to the mother and/or infant. However, to maximally decrease early HIV-1 transmission it is also necessary to target the late antenatal period. The transmission rates at birth in the HIVNET trial based on highly sensitive RNA PCR [14,15] suggest the risk for in-utero transmission may be higher than previous estimates based on DNA PCR results [12,16,36]. Recent trial data from Thailand and France suggests that earlier antiretroviral prophylaxis (between 28 and 36 weeks gestation) further reduced HIV-1 transmission and highlights the high-risk subgroup of women who deliver prematurely and who will receive no or minimal antiretroviral prophylaxis if prophylaxis is begun in late pregnancy [35,37].
To build on current knowledge, further clinical trials should investigate the safety and efficacy of potent antiretroviral drugs to suppress viral load in the peripartum period and maximally reduce early transmission to < 5%. However, the type and duration of antiretroviral prophylaxis must balance the reduction of vertical HIV-transmission against the potential risks to the mother and infant associated with drug toxicity or the induction of resistance. A two-tiered approach could optimize benefits: HIV infected women who do not meet criteria for antiretroviral treatment for their own health (based on clinical criteria or CD4 cell count), could receive short course potent antiretroviral prophylaxis (e.g., ZDV/3TC) in the peripartum period with short infant prophylaxis. The goal would be to reduce viral load around delivery and the risk of perinatal HIV transmission while minimizing the risk of drug-related toxicity. (We refer the interested reader to a comprehensive review of the safety and toxicity of antiretroviral drugs used during pregnancy and the postpartum period .) The subgroup of pregnant women with advanced HIV disease should commence highly active combination therapy in the antenatal period. This would provide antiretroviral therapy rather than prophylaxis to those women at highest risk of maternal–child HIV transmission, maternal disease progression and of antiretroviral drug resistance with dual health benefits to mother and infant. To target these resources effectively, the efficacy, optimal timing, feasibility, and marginal cost of these complementary prevention and care strategies should be determined . In Africa, use of combination antiretroviral drugs is increasingly feasible in a dynamic world of changing prices and monitoring approaches as well as expanding experience and international resources.
The implications of antiretroviral resistance require consideration in light of emerging data,  such as the reports of the induction of major genotypic mutations among subgroups of women and infants after intrapartum NVP prophylaxis (K103N and Y181C) [41,42] and extended (> 4 weeks) antenatal ZDV/3TC prophylaxis (M184V) . As advanced HIV disease and high viral load is associated with the highest rate of resistance mutations the provision of combination antiretroviral treatment to women with advanced HIV disease would be expected to reduce the incidence of resistance. Further research is required to determine the effects on future treatment and perinatal prophylaxis options. In light of this and other data, UNAIDS has conducted expert consultations to assess the risks and benefits of antiretroviral prophylaxis and recommends immediate implementation of effective NVP, ZDV, or ZDV/3TC interventions to decrease perinatal HIV-1 transmission in resource-limited settings [43,44].
Of concern, prolonged breastfeeding contributed substantially to total HIV-1 transmission, even in the presence of an early effective intervention and ongoing intensive counseling and support from research staff. The limited trial data is consistent with observational data from Malawi and suggests that, without antiretroviral drugs the rate of transmission attributable to breastfeeding is highest in the first few weeks after birth and then declines to a steady low rate or a declining rate . Further analysis of the west African data suggest that maternal disease stage is an important modifier of the risk of postnatal transmission; the risk of late postnatal transmission was very low (< 1.5%) for women with baseline CD4 cell counts > 500 × 106/l but was substantial for the subset of women with depressed CD4 cell counts . This is an important empiric finding and is consistent with published data showing a strong relationship between maternal disease stage and breast-milk viral load. If substantiated by further analysis of the virologic and immunologic data from the trial cohorts  it would allow the subgroup of women at highest risk for breastfeeding transmission to be targeted for breast-milk replacement; and/or treatment during breastfeeding with potent combination antiretroviral drugs. This targeted strategy would then allow the continued promotion and support of good breastfeeding practices for all women (including most HIV-infected women) and reduce the morbidity and mortality associated with formula use. Previous research has identified risk factors for HIV transmission that may be modified through good breastfeeding practices such as cracked nipples, mastitis, and early mixed breastfeeding [4,46–51]. All women who elect to breastfeed should be encouraged to exclusively breastfeed during their infant's first 6 months of life, as exclusive breastfeeding is associated with decreased infant morbidity and improved birth spacing . The trial results suggest that targeted breast-milk replacement and early weaning strategies could be combined with effective antiretroviral interventions to decrease total mother-to-infant HIV-1 transmission and promote child survival.
There are inextricable links between maternal and child health and between HIV-related prevention, care and treatment. Large-scale implementation of programs to prevent pediatric HIV infections is now a global public health priority. Implementation will require a renewed commitment to increasing access to quality maternal and child health services and to expanding these services to integrate voluntary counseling, HIV testing, perinatal antiretroviral prophylaxis, infant feeding support and comprehensive cost-effective HIV care. The public health imperative is clear: large-scale efforts can prevent hundreds of thousands of pediatric HIV infections, provide hope for millions of HIV-affected families and reverse the recent precipitous declines in child survival in sub-Saharan Africa.
1.Cooper ER, Charurat M, Burns DN, Blattner W, Hoff R. Trends in antiretroviral therapy and mother-infant transmission of HIV. The Women and Infants Transmission Study Group. J Acquir Immune Defic Syndr 2000, 24:45–47.
2.Garcia PM, Kalish LA, Pitt J,for the Women and Infants Transmission Study Group. Maternal levels of plasma HIV type 1 RNA and the risk of perinatal transmission. N Engl J Med 1999, 341:394–402.
5.DeCock, K, Fowler, MG, Mercier E, et al. Prevention of mother-to-child transmission of HIV-1 in resource poor countries: translating research into policy and practice. JAMA 2000, 283:1175–1182.
6.Nicoll A, Newell M-L, Van Praag E, Van de Perre P, Peckham C. Infant feeding policy and practice in the presence of HIV-1 infection. AIDS 1995, 9:107–119.
7.WHO Collaborative Study Team on the Role of Breastfeeding on the Prevention of Infant Mortality. Effect of breastfeeding on infant and child mortality due to infectious diseases in less developed countries: a pooled analysis. Lancet 2000, 355: 451–455.
8.Fowler, MG, Bertolli J, Nieburg P. When is breastfeeding not best. JAMA 1999, 282:781–783.
9.WHO/UNAIDS/UNICEF. HIV and Infant Feeding. Guidelines for Decision Makers. Geneva: WHO; 1998.
10.The European Mode of Delivery Collaboration. Elective caesarean-section versus vaginal delivery in prevention of vertical HIV-1 transmission: a randomised clinical trial. Lancet 1999, 353:1035–1039.
11.The International Perinatal HIV Group. The mode of delivery and the risk of vertical transmission of human immunodeficiency virus type 1. A meta-analysis of 15 prospective cohort studies. N Engl J Med 1999, 340:977–987.
12.Dunn DT, Newell ML, Ades AE, Peckham CS. Risk of human immunodeficiency virus type 1 transmission through breastfeeding. Lancet 1992, 340:585–588.
13.Leroy V, Newell M-L, Dabis F, et al.for the Ghent International Working Group on Mother-to-Child Transmission of HIV 1 Infection. International multi-center pooled analysis of late postnatal mother to child transmission of HIV-1 infection. Lancet. 1998, 352:597–600.
14.Mofenson L, Harris R, Steihm ER, et al. Performance characteristics of HIV-1 culture, DNA PCR, or quantitative RNA for early diagnosis of perinatal HIV-1 infection. Seventh Conference on Retroviruses and Opportunistic Infections. San Francisco, January 2000 [abstract 713].
15.Reisler R, Thea D, Pliner V, et al. Early detection of reverse transcriptase activity in plasma of neonates infected with HIV-1: A comparative analysis with RNA-based and DNA-based testing using polymerase chain reaction. J Acquir Immune Def Syndr 2001, 26:93–102.
16.Dunn DT, Brandt CD, Krivine A, et al. The sensitivity of HIV-1 DNA polymerase chain reaction in the neonatal period and the relative contributions of intra-uterine and intra-partum transmission. AIDS 1995, 9:F7–F11.
17.Nduati R, John G, Mbori-Ngacha D, et al. Effect of breastfeeding and formula feeding on transmission of HIV-1. A randomised clinical trial. JAMA 2000, 283:1167–1174.
18.Wiktor S, Ekpini E, Karon J, et al. Short course oral zidovudine for prevention of mother-to-child transmission of HIV-1 in Abidjan, Cote d'Ivoire: a randomised trial. Lancet 1999, 353:781–785.
19.Dabis F, Msellati P, Meda N, et al. 6-month efficacy, tolerance, and acceptability of a short regimen of oral zidovudine to reduce vertical transmission of HIV in breastfed children in Cote d'Ivoire and Burkina Faso: a double-blind placebo controlled multicentre trial. Lancet 1999, 353:786–792.
20.DITRAME ANRS 049 Study Group. 15-month efficacy of maternal oral zidovudine to decrease vertical transmission of HIV-1 in breastfed African children. Lancet 1999, 354:2050–2051.
21.Leroy V, Karon J, Alioum A, et al. 24-month efficacy of a maternal short-course zidovudine regimen to prevent mother-to-child transmission of HIV-1 in West Africa: A pooled analysis of two randomized clinical trials. AIDS 2002, 16:631–641.
22.Guay L, Musoke P, Fleming T, et al. A randomised trial of single dose nevirapine to mother and infant versus azidothymidine in Kampala, Uganda for prevention of mother to infant transmission of HIV-1 (HIVNET 012). Lancet 1999, 354:795–802.
23.Owor M, Deseyve M, Duefield C, et al. The one year safety and efficacy data of the HIVNET-012 trial. XIII International Conference on AIDS. Durban, July 2000 [abstract LbOr1].
24.Fowler MG, Mwatha A, Guay L, et al. for the HIVNET-012 Study Team. Effect of Nevirapine for perinatal HIV prevention appears greatest among women with most advanced disease: subgroup analyses of HIVNET-012. Ninth Conference on Retroviruses and Opportunistic Infections. Seattle, February 2002 [abstract 120].
25.Lange JM, on behalf of the PETRA trial study team. Early and late efficacy of three short ZDV/3TC combination regimens to prevent mother-to-child transmission of HIV-1. Third Conference on Global Strategies for the Prevention of HIV Transmission From Mothers to Infants. Kampala, Uganda, September 2001 [abstract 017].
26.The PETRA study team. Efficacy of three short-course regimens of zidovudine and lamivudine in preventing early and late transmission of HIV-1 from mother to child in Tanzania, South Africa and Uganda (PETRA study): a randomised, double-blind, placebo-controlled trial. Lancet 2002, 359:1178–1186.
27.Moodley D, for the SAINT Study Team. The SAINT trial: Nevirapine (NVP) versus zidovudine (ZDV) + lamivudine (3TC) in prevention of peripartum transmission. XIII International Conference on AIDS. Durban, July 2000 [abstract LbOr2].
28.Mbori-Ngacha D, Nduati R, John G, et al. Morbidity and mortality in breastfed and formula fed infants of HIV-1 infected women: results of a randomized clinical trial. XIII International Conference on AIDS. Durban, July 2000 [abstract WeOrC494].
29.Nduati R, Richardson B, John G, et al. Effect of breastfeeding on mortality among HIV-1 infected women: a randomized trial. Lancet 2001, 357:1651–1655.
30.Marseille E, Kahn JG, Mmiro F, et al. Cost effectiveness of a single dose nevirapine regimen to mother and infant to reduce vertical HIV transmission in sub-Saharan Africa. Lancet 1999, 354:803–809.
31.Mirochnick M, Fenton T, et al.,for the Pediatric AIDS Clinical Trials Group Protocol 250 Team. Pharmacokinetics of nevirapine in HIV type 1 infected pregnant women and their neonates. J Infect Dis 1998, 178:368–374.
32.Musoke P, Guay LA, Bagenda D, et al. A phase 1/11 study of the safety and pharmacokinetics of nevirapine in HIV-1 infected pregnant Ugandan women and their neonates. AIDS 1999, 13:479–486.
33.Wade N, Birkhead G, Warren B, et al. Abbreviated regimens of zidovudine prophylaxis and perinatal transmission of the human immunodeficiency virus. N Engl J Med 1999, 340:1409–1414.
34.Bulterys M, Orloff S, Abrams E, et al. Impact of zidovudine post-perinatal exposure prophylaxis on vertical HIV-1 transmission: a prospective cohort study in 4 US cities. Second Conference on Global Strategies for the Prevention of HIV Transmission for Mothers to Infants. Montreal, September 1999 [abstract 015].
35.Lallemant M, Jourdain G, Le Couer S, et al. A trial of shortened zidovudine regimens to prevent mother to child transmission of HIV type-1. N Engl J Med 2000, 343:982–991.
36.Bertolli J, St Louis ME, Simonds RJ, et al. Estimating the timing of mother to child transmission of human immunodeficiency virus in a breast feeding population in Kinshasa, Zaire. J Infect Dis 1996, 174:722–726.
37.Mandelbrot L, Landreau-Mascaro A, Rekacewicz, et al. Lamivudine-zidovudine combination for prevention of maternal-infant transmission of HIV-1. JAMA 2001, 285:2083–2093.
38.Centers for Disease Control and Prevention. Public Health Service Task Force recommendations for the use of antiretroviral drugs in pregnant HIV-infected women for maternal health and interventions to reduce perinatal HIV-1 transmission in the United States. MMWR
:1–30 (most recent revisions available at http://www.hivatis.org
39.Dabis F, Newell ML, Fransen L, et al.for the Ghent working group on mother-to-child transmission of HIV. Prevention of mother-to-child transmission of HIV in developing countries: recommendations for practice. Health Policy and Planning 2000, 15:34–42.
40.Hirsch MS, Brun-Vezinet F, D'Aquila RT, et al. Antiretroviral drug resistance testing in adult HIV-1 infection. Recommendations of an international AIDS Society-USA panel. JAMA 2000, 283:2417–2426.
41.Jackson JB, Becker-Pergola G, Guay LA, et al. Identification of the K103N resistance mutation in Ugandan women receiving nevirapine to prevent HIV-1 vertical transmission. AIDS 2000, 14:F111–F115.
42.Eshleman S, Mracna M, Guay L, et al. Selection and fading of resistance mutations in women and infants receiving Nevirapine to prevent HIV-1 vertical transmission (HIVNET 012). AIDS 2001, 15:1951–1957.
43.WHO Technical Consultation on behalf of the UNFPA/UNICEF/WHO/UNAIDS Interagency Task Team on Mother-to-Child HIV Transmission of HIV. New Data on the Prevention of Mother-to-Child Transmission of HIV and their Policy Implications: Conclusions and recommendations.
Geneva: WHO; October 2000. http://www.unaids.org/publications/documents/mtct/MTCT_
44.WHO, Department of Reproductive Health and Research. Use of Nevirapine to Reduce Mother-to-Child Transmission of HIV. WHO Review of Reported Drug Resistance. Geneva: WHO; 2000.
45.Miotti PG, Taha TE, Kumwenda NI, et al. HIV transmission through breastfeeding. A study in Malawi. JAMA 1999, 282:744–749.
46.Embree JE, Njenga S, Datta P, et al. Risk factors for postnatal mother-child transmission of HIV-1. AIDS 2000, 14:2535–2541.
47.Ekpini E, Ehounou R, Wiktor SZ, et al. Late postnatal mother-to-child transmission of HIV-1 in Abidjan, Cote d'Ivoire. Lancet 1997, 349:1054–1059.
48.Van de Perre P, Hitimana D, Simonon A, et al. Postnatal transmission of HIV-1 associated with breast abscess. Lancet 1992, 339:1490–1491.
49.Semba R, Kumwenda N, Hoover D, et al. Human immunodeficiency virus load in breast milk, mastitis and mother-to-child transmission of HIV Type 1. J Infect Dis 1999, 180:93–98.
50.John G, Nduati R, Mbori-Ngacha D, et al. Correlates of mother-to-child human immunodeficiency virus type-1 transmission: Association with maternal plasma HIV-1 RNA load, genital HIV-1 DNA shedding and breast infections. J Infect Dis 2001, 183:206–212.
51.Coutsoudis A, Pillay K, Spooner E, et al. Influence of infant feeding patterns on early mother to child transmission of HIV-1 in Durban, South Africa. Lancet 1999, 354:471–476.
© 2002 Lippincott Williams & Wilkins, Inc.