We have sufficient knowledge and unprecedented access to global resources to dramatically reduce the transmission of HIV-1 from mothers to children [MTCT] worldwide. In industrialized and some middle-income countries, therapeutic and prophylactic antiretrovirals [ARVs], often with Caesarean section and always with breastfeeding avoidance, have reduced antepartum, intrapartum and postnatal transmission from 25 to 1–5%. Breastfeeding by HIV-positive women, which is unavoidable in poor communities, and inadequate infrastructure to deliver health interventions, are key reasons for lack of success in developing countries. We now have data that strongly suggest that breastfeeding transmission can be substantially decreased. There are recent reviews on this subject [1•,2].
Primary prevention of HIV-1 in women
The first steps in a comprehensive approach to this problem are: (i) to prevent unwanted pregnancies in all women living in high-HIV-prevalence regions; (ii) the acquisition of HIV in women of child-bearing age; and (iii) to prevent unwanted pregnancy and delivery in those already HIV infected. There has been interesting new evidence from Zimbabwe [3••], together with previous information from eight African countries  and Thailand , that suggests that this primary prevention strategy works. In Zimbabwe [3••], MTCT of HIV decreased from 8.2% in 2000 to 6.2% in 2005, predominantly attributable to declining maternal HIV prevalence rather than to the MTCT programme. Furthermore, a clue to the proportions of transmission affected was gained when calculations showed that, in 2005, 4.0% of infant infections were attributable to maternal seroconversion [new infections] and 32% to breast feeding. These studies are inherently difficult given the multiplicity of factors and temporal changes likely to have affected the outcome of transmission of HIV to infants.
Is it feasible, in principle, to eliminate mother-to-child transmission of HIV globally?
Over the last few years [1•,2] a better understanding of the risk factors  (see Fig. 1) and key advances in prevention of MTCT have led to interventions that reduce breastfeeding transmission. It is reasonable to assume that when the findings of some of the published studies described here are implemented in the public sector in poor countries, and the results of currently active studies alluded to below become available during 2009, the feasibility of virtually eliminating transmission of HIV-1 to infants throughout the world becomes realizable for the first time. The studies cited below are a mixture of well designed randomized control trials and single-arm observational studies. A well designed randomized trial in Zambia of counselling for exclusive breastfeeding, and early cessation versus continued breastfeeding, adds much new information on the advantages of exclusivity and the disadvantages of early cessation [7••]. Nearly, all the studies point in the direction of benefit. In brief, the evidence over recent years has shown the efficacy of ARVs given to the breastfeeding HIV-exposed infant, or highly active antiretroviral therapy (HAART) to the lactating mother, in reducing breastfeeding transmission. There are other beneficial outcomes, especially a decrease in infant mortality, that have been recorded in some studies. The presently available maternal HAART data are based on observational cohort studies, but the findings are consistent between different groups, and the results appear stunning.
The Zambian study [7••] asked the question whether a shortened period of breastfeeding would be of benefit in terms of reduced transmission of HIV and mortality. Abrupt weaning at 4 months was compared with continued breastfeeding in 958 HIV-infected mothers. The interventions included counselling, provision of formula and weaning cereal, and improved health services [growth monitoring, primary care]. In the control group the median duration of breastfeeding was 16 months and in the abrupt weaning group it was 5 months at most. At 24 months there was no difference in HIV-free survival between the group that had continued breastfeeding, and the group that had abruptly weaned.
Children who became HIV infected and had weaned abruptly, had a higher mortality by 24 months than the HIV-infected children who continued breastfeeding. It follows that in these children early, abrupt cessation of breastfeeding did not improve a composite index of ‘survival together with protection from HIV transmission’ by 2 years in children exposed to HIV through breast milk, and increased the death rate in HIV-infected children. Most importantly, they have subsequently described critical risk factors for transmission [8•]: women with low CD4 cell counts accounted for most of the transmissions, and much of the transmission had already occurred before weaning at 4 months of age.
Prophylactic antiretrovirals in breastfeeding infants
The utility of ARVs in infants to minimize breastfeeding transmission was assessed in the Milk-Borne Transmission of HIV-1C (MASHI) [Botswana] , Six Week Extended-dose Nevirapine (SWEN) [India, Ethiopia, Uganda] [10••], Post-Exposure Prophylaxis of Infants (PEPI)-Malawi [11••], and MITRA [Tanzania] [12•] studies.
MASHI  was a 2 × 2 factorial design trial with peripartum sdNVP versus placebo, and postpartum infant feeding either with formula or with breastfeeding covered by prophylactic ZDV. The index of ‘HIV infection rates or cumulative mortality’ at 18 months was similar in the formula and breastfed + ZDV groups. Exclusive breastfeeding rates were low: 57.1% at the first month, 31.3% at month 3, and 17.5% at month 5, and mixed breastfeeding rates were 21.7, 48.6, and 75% respectively. The 7-month HIV-infection rates were higher but mortality was lower in the breastfed + ZDV group. The association between CD4 cell counts and outcomes was difficult because of the use of HAART in some mothers. Accordingly, ZDV did not decrease breastfeeding transmission as effectively as formula, but was associated with a lower mortality.
SWEN [10••] was a randomized controlled trial (RCT) that compared 6 weeks of infant NVP to sdNVP in breastfeeding infants. There was a significant reduction in transmission at 6 weeks [relative risk 0.54, 95% CI 0.34–0.85], but not at 6 months, and significant decreases in the composite index of ‘HIV infection or death’ at both 6 weeks and 6 months. Breastfeeding frequencies dropped rapidly after 14 weeks. These findings suggest that breastfeeding transmission is not reduced after cessation of the drug and longer periods of prophylactic NVP, preferably to 6 months [the optimum period for exclusive breastfeeding] may be required to prolong the benefits of breastfeeding.
The PEPI-Malawi study [11••] implicitly addressed the specific question raised by the SWEN paper by extending ARV prophylaxis for 14 weeks after birth. Infants were randomized to one of three arms: sdNVP+ a week of ZDV [control]; control regimen + daily NVP; control regimen+ NVP + ZDV. There was no added benefit of the addition of ZDV to the second intervention arm, but there was a significantly higher rate of adverse events possibly related to the drug. At 9 months the transmission was 10.6% in the controls, compared with 5.2% in the extended NVP arm [P < 0.001], and 6.4% in the NVP + ZDV arm [P = 0.002]. In comparison to the control arm the extended NVP group had a lower combined index of ‘HIV infection or death’ at 15 months. Lower CD4 cell counts were significantly associated with the transmission rate and the index of ‘HIV infection or death’. Breastfeeding rates were very high for 6 months in all groups.
However, there are difficulties in interpretation of the PEPI-Malawi results in relation to the SWEN results. PEPI-Malawi was not designed to show superiority over SWEN, that is, 14 weeks of NVP are better than 6 weeks of NVP in reducing transmission. And a comparison of the two studies reveals complexities in assessing the degree of benefits achieved, the effects on mortality and transmission rates, and the temporal accumulation of infant HIV infections.
A nonrandomized, open-label, prospective cohort study from Tanzania [MITRA Study] [12•] employed antenatal ZDV + 3TC, and in the infants 3TC + ZDV for a week followed by 3TC alone for a maximum of 6 months. HIV infection at 6 months in those who remained uninfected at 6 weeks was 1.2% [95% CI 0.0–2.4]; the index of ‘HIV infection or death’ was 8.5% [5.7–11.4] at 6 months. These results showed a 50% reduction in transmission compared with historical controls. Low CD4 cell counts were associated with HIV transmission and with ‘HIV infection rates or death’. Median period for breastfeeding was 18 weeks. The same group has used a similar design [MITRA-PLUS] and shown a breastfeeding transmission rate as low as 0.9% at 6 months with the use of maternal HAART given to all HIV-positive pregnant women for the same period [13•].
Antiretrovirals to breastfeeding mothers to reduce mother-to-child transmission
The MITRA-PLUS [13•], KIBS [14•], AMATA [15•], DREAM [16•] and Kesho-Boro [17•] studies represent the use of maternal HAART in observational, open-label, one-arm trials, to decrease breastfeeding transmission of HIV-1.
The MITRA-PLUS [13•] study was quoted earlier. Further data were provided at the WHO consultation [8•]. The HIV transmission rates due to breastfeeding on maternal HAART [ZDV + 3TC + NVP] were: 0.9% at 6 months, 1.7% at 12 months, and 1.9% at 18 months; the index of ‘HIV infection or death’ was 2.7, 6.9 and 7.7 respectively. The CD4 cell count and percentage were unrelated to transmission, although viral load was significantly associated. There were an astonishing 1.6% of patients on ARVs who developed Stevens–Johnson syndrome, although in no-one was it fatal.
KIBS [14•] in Kenya used ZDV + 3TC + NVP/Nelfinavir [CD4 < 250 cells/> 250 cells/ul respectively] antenatally to 6 months postpartum, exclusive breastfeeding for 5.5 months, locally available foods for weaning and replacement, and nutritional supplement. The breastfeeding transmission rates were 1.5% at 6 weeks, 1.7% at 3 months, 2.6% at 6 months, and 3.5% at 12 months. There was no effect of CD4 cell counts on transmission, which was probably due to the effect of HAART. There was no hepatotoxic effect of NVP on those with CD4 cell counts more than 250cells/ul. Further data with lower rates of breastfeeding transmission and higher indices of ‘survival rates without HIV infection’ compared with historical controls have been reported at the WHO consultation [8•].
AMATA [15•] in Rwanda provided HAART to all HIV-positive women [n = 562] and showed an absence of breastfeeding transmission at 7 months, with no significant differences between formula feeders and breastfeeders for psychomotor development, morbidity, or mortality.
The DREAM study [16•] in Mozambique, is a development programme and includes HAART for all HIV-positive women, nutritional supplementation, intense counselling, and lifelong care [n = 341]. The breastfeeding transmission at 12 months was 1.3%, and there was a 50% reduction in infant mortality.
The Kesho-Boro study [17•] a multicentre project in Africa, offered HAART for life for women's own health if the CD4 cell count was less than 200 cells/ul or had symptoms, or short course HAART for MTCT for individuals with CD4 cell counts more than 500 cells/ul and who were asymptomatic; transmission at 12 months was 9% in the former and 6% in the latter. Interestingly, of the seven infants infected with HIV postnatally [between 6 weeks and 12 months] five were born to women on HAART [8•].
Studies which are current or planned
The rapid advances in this field are indicated by the number of, as yet, unpublished ongoing studies and those in the planning stage, presented at a recent WHO consultation on MTCT [8•]. The former include:
1. Kesho-Bora [17•], a multisite study in Africa. RCT of triple ARVs compared with short-course ZDV plus single-dose nevirapine [sdNVP] to mothers during pregnancy and breastfeeding.
2. Mma Bana in Botswana. RCT of two triple ARV regimens to mothers during breastfeeding.
3. The BAN study in Malawi. An RCT of triple ARVs to mothers or nevirapine to infants during breastfeeding.
4. HPTN 046 at four sites in Africa. RCT of NVP for 6 months compared with NVP for 6 weeks to breastfeeding infants.
The planned studies include an RCT of 3TC compared with placebo for breastfeeding infants [PROMISE-PEP] in African sites and an RCT with four different randomizations [PROMISE], antenatal, postpartum, during infancy, and for women postnatally, at multiple industrialized and developing country sites. The purpose is to answer questions on MTCT for women who do not qualify for HAART, and the benefit/safety of continuing against stopping HAART used for MTCT in these women. A central focus on infant and child health is to be achieved by determining the most effective strategy for reducing breastfeeding transmission and improving postweaning infant survival. The cost–effectiveness of each intervention is to be calculated. A study introducing ‘universal’ maternal HAART is also proposed although at this stage implementation of ‘universal’ HAART for pregnant women cannot be justified by the currently available evidence.
Resistance to antiretrovirals used in mother-to-child-transmission programmes
Resistance that develops to the ARVs used in MTCT programmes is an important issue for the subsequent management of the HIV-infected infant or mother. It is a subject that deserves more attention than can be provided in this commentary. It is important to note that in the studies described earlier, which used prophylactic infant NVP or maternal HAART, there was a high incidence of resistance to NVP [Non-Nucleoside Reverse Transcriptase Inhibitors]. In the infant NVP prophylaxis SWEN study there was 100% [7/7] genotypic resistance to NVP at 6 months of age [18•], and in the KIBS maternal HAART prophylaxis study there was a 67% [16/24] resistance to ARVs in the 14-week and 24-week blood specimens from infants [19•]. In the maternal HAART AMATA and DREAM studies [8•] the genetic resistance to NNRTIs was present in about a quarter of the infants at 6 months of age in the former, and in about 12% between 2 and 6 months after treatment interruption in DREAM.
The subsequent ARV management of children and women who had developed resistance to these drugs during participation in MTCT programmes is being investigated. There is a suggestion that the longer the duration between exposure to the prophylactic ARV employed in MTCT and initiation of HAART for progression of disease and the children's or women's own health, the better the response ; and that perinatal use of other ARVs reduces resistance in the women .
Access to health services
It is generally agreed that the lack of health resources, health service facilities and health personnel are major barriers to access to programmes in the public sector in developing countries. Globally, in 2007, only 18% of pregnant women received HIV testing, about a third received ARVs for prevention of MTCT, and just 12% of HIV-positive pregnant women were assessed whether their disease was so advanced that they required ARVs for their own health [22••]. A series of papers from Africa clearly demonstrated inadequacies in national PMTCT programmes: there were large health service inequities that directly affected HIV-free survival, a need to focus on management of early HIV infections in newborns, poor quality of counseling, and the need for appropriate policies and programmes for infant feeding [23•,24•]. Improving coverage is essential.
Integration of services may be particularly efficient for MTCT programmes (see Fig. 2) [25••,26], although the evidence for this is not firmly established  and much more operational research is necessary.
(see Table 1).
Current evidence on reducing postnatal transmission includes the use of prophylactic ARVs in breastfeeding infants or lactating mothers. Prevention of unwanted pregnancies in all women living in areas of high HIV prevalence, primary prevention of HIV-1 in women of child-bearing age, and health service upgrading, are part of a comprehensive approach. Further direct research on use of ARVs is required on the following:
1. Determine CD4 cell thresholds for initiation of treatment with HAART for the health of the pregnant women herself, rather than for prevention of transmission of HIV to the infant only. As much of the transmission is accounted for by women with low CD4 cell counts, this single measure has considerable potential for substantially reducing overall mother-to-child transmission.
2. Which is the more efficacious intervention to decrease breastfeeding transmission: maternal HAART or infant ARVs?
3. Issues of resistance to ARVs used in mTCT programmes.
References and recommended reading
Papers of particular interest, published within the annual period of review, have been highlighted as:
• of special interest
•• of outstanding interest
Additional references related to this topic can also be found in the Current World Literature section in this issue (pp. 338–339).
1• Coovadia H, Kindra G. Breastfeeding HIV transmission and infant survival: balancing pros and cons. Curr Opin Infect Dis 2008; 21:11–15.
2 Wilfert CM, Fowler MG. Balancing maternal and infant benefits and the consequences of breastfeeding in the developing world during the era of HIV infection. J Infect Dis 2006; 195:165–167.
3•• Dube S, Boily MC, Mugurungi O, et al
. Estimating vertically acquired HIV infections and the impact of the prevention of mother-to-child transmission program in Zimbabwe: insights from decision analysis models. JAIDS 2008; 48:72–81.
4 Sweat MD, O'Reilly KR, Schmid GP, et al
. Cost–effectiveness of nevirapine to prevent mother-to-child HIV transmission in eight African countries. AIDS 2004; 18:1661–1671.
6 Coovadia H. Breastfeeding in HIV-positive women: do the benefits outweigh the risks? Editorial. Pediatr Health 2007; 1:3–8.
7•• Kuhn L, Aldrovandi GM, Sinkala M, et al
. Effect of early, abrupt weaning for HIV-free survival of children in Zambia. N Engl J Med 2008; 359:130–141.
8• WHO Expert Consultation on new and emerging evidence on the use of antiretroviral drugs for the prevention of mother-to-child transmission of HIV. WHO, Geneva, Switzerland 2008. A recent expert consultation attended by many of the most active investigators in the field of MTCT globally.
9 Thior I, Lockman S, Smeaton SM, et al
. Breastfeeding plus infant zidovudine for 6 months vs formula feeding plus infant zidovudine for 1 month to reduce mother-to-child HIV transmission in Botswana: a randomized trial: The Mashi Study. JAMA 2006; 296:794–805.
10•• Six Week Extended Dose Nevirapine [SWEN] Study Team. Extended-dose nevirapine to 6 weeks of age for infants to prevent HIV transmission via breastfeeding in Ethiopia, India, and Uganda: an analysis of three randomized controlled trials. Lancet. 2008; 372:300–313. The first randomized trial to demonstrate the efficacy of extending nevirapine in breastfeeding infants born to HIV-positive mothers beyond birth for 6 weeks. Significant decrease of MTCT at 6 weeks. Improved HIV-free survival of infants at 6 months.
11•• Kumwenda N, Hoover DR, Mofensen LM, et al
. Extended antiretroviral prophylaxis to reduce breast-milk HIV-1 transmission. N Engl J Med 2008; 359:119–129.
12• Kilewo C, Karlsson K, Massawe A, et al
. Prevention of mother-to-child-transmission of HIV-1 through breastfeeding by treating infants prophylactically with lamivudine in Dares Salaam, Tanzania. The Mitra Study. J Acquir Immune Defic Syndrome 2008; 48:315–323. An observational study showing that ARVs in infants can reduce breastfeeding transmission; in this study 3TC was used instead of nevirapine which was tested in other studies noted earlier.
13• Kilewo C, Karlsson K, Ngarina M, et al.
Emerging strategies to curb HIV transmission. 4th IAS Conference on HIV Pathogenesis, Treatment and Prevention, Sydney, Australia 2007; Abstract no.: TUAX101. www.ias2007.org
. An observational study from the same group as in Reference 12 indicating a profound reduction of transmission of HIV-1 to breastfeeding infants through maternal HAART.
14• Thomas T, Masaba R, Ndivo R, et al.
Prevention of mother-to-child transmission of HIV-1 among breastfeeding mothers using HAART: The Kisumu Breastfeeding Study, Kisumu, Kenya, 2003–2007. 15th Conference on Retroviruses and Opportunistic Infections, Boston, Massachusetts, 2008. Abstract no.: 45aLB. http://www.retroconference.org/2008/
. As in reference no. 12, an observational single arm study showing the efficacy of maternal HAART in MTCT.
15• Arendt V, Ndimubanzi P, Vyankandondera J, et al.
AMATA study: effectiveness of antiretroviral therapy in breastfeeding mothers to prevent postnatal vertical transmission in Rwanda. Thomas T, et al.
4th IAS Conference on HIV Pathogenesis, Treatment and Prevention, Sydney, Australia. 22–25 July, 2007; Abstract no.: TUAX102. www.ias2007.org
. As in the above references, the efficacy of maternal HAART in MTCT. Informative data on additional benefits on growth and nutrition of infants of maternal HAART prophylaxis.
16• Palombi L, Marazzi MC, Voetberg A, et al
. Treatment acceleration program and the experience of DREAM program in prevention of mother-to-child transmission of HIV. AIDS 2007; 21:S65–S71.
17• de Vincenzi I and The Kesho Bora Study Group. HIV-free survival at 12 months among children born to HIV-infected women receiving antiretrovirals from 34-to-36 weeks of pregnancy. WHO 2-Tier Approach: Kesho Boro Observational Cohort. 15th Conference on Retroviruses and Opportunistic Infections, Boston, Massachusetts 2008. Abstract no.: 638. http://www.retroconference.org/2008/
. Some preliminary results of a trial that investigated the maternal and infant outcomes of maternal HAART given either for the advanced stage of the pregnant women's HIV disease (i.e., for life), or for MTCT (i.e., short course, discontinue treatment on cessation of breastfeeding).
18• Church JD, Omer SB, Guay LA, et al
. Analysis of nevirapine (NVP) resistance in Ugandan infants who were HIV infected despite receiving single-Dose (SD) NVP plus daily NVP up to 6 weeks of age to prevent HIV vertical transmission. J Infect Dis 2008; 198:1075–1082.
19• Zeh C, Weidle P, Nafisa L, et al.
Emergence of HIV-1 drug resistance among breastfeeding infants born to HIV-infected mothers taking antiretrovirals for prevention of mother-to-child transmission of HIV: The Kisumu Breastfeeding Study, Kenya. 15th Conference of Retroviruses and Opportunistic Infections, Boston, Massachusetts. 2008; Abstract no.: 84LB. http://www.retroconference.org/2008/
. Maternal HAART for preventing MTCT of HIV-1 in breastfeeding infants in the KIBS study [see above] also causes resistance to ARVs used in prophylactic regimen.
20 Lockman S, Shapiro RL, Smeaton LM, et al
. Response to antiretroviral therapy after a single, peripartum dose of nevirapine. N Engl J Med 2007; 356:135–147.
21 Chi BH, Sinkala M, Mbewe F, et al
. Single-dose tenofovir and emtricitabine for reduction of viral resistance to nonnucleoside reverse transcriptase inhibitor drugs in women given intrapartum nevirapine for perinatal HIV prevention; an open-label randomized trial. Lancet 2007; 370:1698–1705.
22•• WHO,UNAIDS,UNICEF. Towards Universal Access: Scaling Up Priority HIV/AIDS Interventions in the Health Sector. Progress Report 2008. Data provided by countries. http://www.who.int/hiv/pub/2008progressreport/en/index.html
[Accessed 30 December 2008] Important global data on access to services for prevention of mother-to-child transmission of HIV-1. This paper makes the central point for the way forward: that is, accesses to health services, including those for MTCT, are woefully inadequate in many developing countries.
23• Chopra, M. Prevention of mother to child transmission of HIV in Africa. Operational research to reduce postnatal transmission and infant mortality. PhD Thesis. Uppsala University. Acta Universitatis Upsaliensis Uppsala, Sweden 2008. Multiple deficiencies in programmes for prevention of mother-to-child transmission in Africa. Identifies specific deficiencies that can be targeted for action within an overall improvement plan.
24• Chopra M, Rollins N. Infant feeding in the time of HIV: rapid assessment of infant feeding policy and programmes in four African countries scaling up prevention of mother-to-child transmission programmes. Arch Dis Child 2008; 93:288–291.
25•• Delvaux T, Konan JP, Ake-Tano O, et al
. Quality of antenatal and delivery care before and after the implementation of a prevention of mother-to-child HIV transmission programme in Côte d'Ivoire. Trop Med Int Health 2008; 13:970–979.
26 Salomon JA, Hogan DR, Stover J, et al
. Integrating HIV prevention and treatment: from slogans to impact. PLoS Medicine 2005; 2:e16.
27 Briggs CJ, Garner P. Strategies for integrating primary health services in middle- and low-income countries at the point of delivery. Cochrane Database Syst 2006; 2:CD003318.
28 Coovadia HM, Rollins N, Bland R, et al
. Mother-to-child transmission of HIV-1 infection during exclusive breastfeeding in the first 6 months of life: an intervention cohort study. Lancet 2007; 369:1107–1116.