SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Background

The majority of HIV-infected children acquire the infection from their mothers. Preventing mother-to-child transmission (MTCT) of HIV is a key step in reducing the number of paediatric HIV infections.

Objective

To summarize the Cochrane reviews which have assessed the efficacy of interventions to prevent MTCT of HIV.

Methods

The Cochrane Database of Systematic Reviews was searched for any intervention to prevent MTCT of HIV. Data were extracted in duplicate and the GRADE assessment tool used to analyze the quality of extracted data.

Main Results

Five reviews met our inclusion criteria, and were of medium to high quality. One review examined the efficacy of antiretroviral prophylaxis for preventing MTCT of HIV. Each regimen used, ranging from the two-dose nevirapine regimen to combination regimens, such as zidovudine and lamivudine, reduced the risk of MTCT of HIV. A meta-analysis was not conducted as different regimens were used in the different trials. Caesarean section before labour and before ruptured membranes reduces the risk of MTCT of HIV (risk ratio [RR] 0.34, 95% confidence interval [CI] 0.14 to 0.8) compared to vaginal delivery. There was no evidence that vaginal disinfection (RR 0.94; 95%CI: 0.71 to 1.25), vitamin A supplementation (RR 1.10; 95%CI: 0.95 to 1.26), or hyperimmune HIV immunoglobulin (RR 0.67; 95%CI: 0.29 to 1.55) are efficacious interventions for the prevention of MTCT of HIV. There was no review on alternatives to breastfeeding.

Authors' Conclusions

Antiretroviral prophylaxis and caesarean section before labour and before ruptured membranes are efficacious in reducing the risk of MTCT of HIV. However, there is a need for research directed at (1) simplifying antiretroviral regimens to facilitate adherence and to minimize toxicity for both mothers and infants, (2) assessing the long-term impact of drug resistance on future treatment of HIV-infected mothers and infants, and identifying regimens that are less likely to induce resistance, and (3) assessing the efficacy of caesarean section among HIV-infected women with undetectable viral loads or who are receiving combination antiretroviral regimens. Copyright © 2007 John Wiley & Sons, Ltd.


Background

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Description of the condition

By the end of 2005, an estimated 2.3 million children under the age of 15 years (80% of them resident in sub-Saharan Africa) were living with human immunodeficiency virus (HIV) infection or acquired immunodeficiency syndrome (AIDS; 1). In 2006 alone, approximately 530 000 children were newly infected with HIV and 380 000 died of AIDS1. The vast majority of HIV-infected children acquire the infection through mother-to-child transmission (MTCT) in utero, around the time of labour and delivery, or postnatally through breastfeeding.

Description of the interventions

Several interventions for the prevention of MTCT of HIV have been evaluated, but only three have been found to be efficacious in preventing MTCT of HIV:

  • Antiretroviral (ARV) prophylaxis. ARVs given to HIV-infected pregnant women and/or to their infants can prevent MTCT of HIV.

  • Caesarean delivery before labour and before ruptured membranes (‘elective’ caesarean section, or ECS). Performing a caesarean section before labour and before ruptured membranes can prevent MTCT of HIV.

  • Complete avoidance of breastfeeding. Complete avoidance of breastfeeding can prevent MTCT of HIV.

Other interventions that have been evaluated include:

  • Hyperimmune HIV immunoglobulin. Hyperimmune HIV immunoglobulin administered to women receiving zidovudine (ZDV) prophylaxis has been proposed as an intervention to prevent MTCT of HIV.

  • Vaginal and/or infant surface decontamination. Vaginal or infant skin decontamination has been proposed as an intervention to prevent MTCT of HIV.

  • Vitamin A supplementation. Vitamin A supplementation during a woman's pregnancy and immediately postpartum has been suggested as a low-cost intervention for preventing MTCT of HIV.

How the interventions might work

ARV prophylaxis

ARVs reduce MTCT either by lowering viral load (of plasma and/or cervicovaginal fluid) in HIV-infected pregnant women, or through pre-exposure prophylaxis in their infants.

Caesarean section before labour and before ruptured membranes

The majority of cases of MTCT of HIV occur around the time of labour and delivery2. Possible mechanisms include transfusion of the mother's blood to the foetus during labour contractions, infection after the rupture of membranes, and direct contact of the foetus with infected secretions or blood from the maternal genital tract3, 4. Therefore, performing a caesarean section before the onset of labour and the rupture of membranes could decrease the risk of MTCT of HIV.

Complete avoidance of breastfeeding

An estimated 24% to 42% of MTCT of HIV occurs postnatally through breast milk transmission5. Complete avoidance of breastfeeding as an intervention to prevent MTCT of HIV has demonstrated efficacy6. If complete avoidance of breastfeeding is not possible or culturally acceptable—in many cultures, there can be stigma attached to a woman who does not breastfeed her child7—other proposed interventions include: early weaning (e.g. at 6 months of age), heat or chemical treatment of the milk to decrease viral load in the milk (thus decreasing the risk of MTCT), ARV prophylaxis to mothers and/or infants while breastfeeding is ongoing, and exclusive breastfeeding8.

Observational studies suggest that infants who are exclusively breastfed for at least the first three months have a lower risk of acquisition of HIV infection than those who are also given other foods9, 10. There is not yet a Cochrane review on feeding interventions for prevention of MTCT, though one is in preparation. According to current World Health Organization recommendations, infants should be exclusively breastfed for the first six months of life to achieve optimal growth, development and health. Thereafter, infants should receive nutritionally adequate and safe complementary foods while breastfeeding continues up to 24 months or beyond. Given the need, however, to reduce the risk of HIV transmission to infants, while minimizing the risk of other causes of morbidity and mortality, the guidelines recommend the following for nursing mothers with HIV: When replacement feeding is acceptable, feasible, affordable, sustainable and safe (AFASS), avoidance of all breastfeeding by HIV-infected mothers is recommended. Otherwise, exclusive breastfeeding is recommended during the first months of life, and should then be discontinued as soon as it is feasible11.

Hyperimmune HIV immunoglobulin

If in utero MTCT of HIV is the result of transportation of virus-antibody complexes across the placental barrier by means of Fc receptors on trophoblastic cells12, 13, 14, then immunoglobulin could non-specifically block placental Fc receptors (and thus prevent MTCT). Inflammatory cytokines such as interleukin-6 are produced in greater quantities in the placentas of HIV-infected women and may play a role in MTCT15. Intravenous immunoglobulin down-regulates proinflammatory cytokine production, especially interleukin-6, and thus administration of immunoglobulin could decrease transplacental transmission16. A Cochrane review of hyperimmune HIV immunoglobulin found that the addition of this treatment to ZDV offered no additional advantage over non-specific immunoglobulin against the MTCT of HIV17.

Vaginal and/or infant surface decontamination

Several disinfectants can neutralize HIV and, in low concentrations, are generally well tolerated. However, with some of these, especially at higher concentrations, there can be significant side effects, such as dermatological hypersensitivity reactions and anaphylactic shock.

Vitamin A supplementation

Vitamin A is the generic name for a group of fat-soluble compounds. There is increasing evidence that vitamin A is essential for normal immune function18, 19. More than 6 million pregnant women are deficient in vitamin A worldwide every year20. Vitamin A deficiency in pregnant women is associated with night blindness, severe anaemia, wasting, malnutrition, reproductive and infectious morbidity21 and increased risk of mortality in the 1 to 2 years following delivery22. Observational data suggested an association between maternal vitamin A deficiency and MTCT of HIV23. Low vitamin A intake has been associated with increased risks of disease progression to AIDS24, 25. It was postulated that vitamin A supplementation would slow the rate of disease progression among HIV-infected pregnant women, and thus decrease the risk of MTCT.

Why it is important to do this review

HIV infection is contributing substantially to infant and child mortality in resource-poor settings, particularly in sub-Saharan African countries. MTCT of HIV is reversing hard-won gains in child survival, achieved over the course of decades. Interventions aimed at reducing the risk of MTCT are therefore a priority. Cochrane reviews of these interventions have shown the evidence for what works, what does not work, what could be harmful, and where more research is needed. This ‘umbrella review’ will provide a necessary overview of what is currently known about preventing mother-to-child HIV transmission. In addition, this review examines six other outcomes affecting the mother and infant (chosen because each was examined in more than one review and for clinical importance).

Methods

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Search Strategy

We searched the Cochrane Database of Systematic Reviews Issue 3, 2006 for all systematic reviews examining any intervention for the prevention of mother-to-child transmission of HIV. The term ‘mother-to-child transmission of HIV’ was entered and restricted to the record title, abstract or keyword. This resulted in five systematic reviews and one protocol.

Of the five reviews, all were selected for this umbrella review. One review, on ARVs for reducing the risk of transmission, was in the process of being updated; the Cochrane HIV/AIDS Group provided us with the text of this update prior to its publication in Issue 1, 2007 of The Cochrane Library.

The review of interventions to decrease the risk of mother-to-child transmission of HIV through breast milk had not been completed by the time this umbrella review was prepared.

Selection of reviews

Reviews for this umbrella review were selected by all authors in consultation with the Cochrane HIV/AIDS Group, reviewing the objectives and methods in each Cochrane review to determine inclusion.

Data extraction and management

One author, Ken Bond, was responsible for the extraction of outcome data from the five reviews. Accuracy of extraction specifically for the ARV data was reviewed by a biostatistician and author of the ARV review, Lize van der Merwe.

Assessment of quality of evidence

Methodological quality was assessed by one author, Ken Bond, using the review authors' descriptions of studies and the GRADE assessment tool26. Quality ranged from moderate to high in all studies described.

Data synthesis

Data were extracted directly from the five systematic reviews included in this umbrella review. In one case27 the efficacy measures were converted to relative risk (RR) to ensure uniform reporting across all reviews. This conversion was undertaken by subtracting the efficacy from one. No additional statistical analysis was completed.

Involvement of authors of underlying Cochrane reviews

The team of authors for this umbrella review includes Charles Wiysonge, lead author on the vaginal disinfection and Vitamin A reviews, and Jennifer Read, lead author on the caesarean delivery review. Gail Kennedy, Tara Horváth, Nandi Siegfried and George Rutherford, of the Cochrane HIV/AIDS Group, were also involved as authors or advisers on this umbrella review.

Description of included reviews

Five reviews are included in this umbrella review; the characteristics of these reviews are presented in Table 1. With one exception, all reviews were published or substantively updated in 2005 or later. The exception is ‘Interventions for reducing the risk of mother-to-child transmission of HIV infection’17, which is gradually being superseded by several smaller reviews. The authors of the ARV prophylaxis review27 included only randomized controlled trials, and assessed quality according to the following criteria: the method used for generating the randomization sequence, the adequacy of allocation concealment, the adequacy of blinding of participants, providers and assessors, and the differential loss to follow-up in the comparison groups. The review on caesarean delivery28 included analytic epidemiological studies, both observational (case control and cohort studies) and interventional (clinical trials). The latter were evaluated according to a validity checklist for clinical trials29. The observational studies were evaluated according to the Newcastle-Ottawa Quality Scale30. Although the Brocklehurst review, which only included randomized controlled trials, presented data on ZDV prophylaxis, nevirapine (NVP) prophylaxis, caesarean section before labour and before ruptured membranes, and hyperimmune HIV immunoglobulin, this umbrella review draws only on the data presented for hyperimmune HIV immunoglobulin plus ZDV compared to non-specific immunoglobulin plus ZDV. The other, more up-to-date reviews, present current data on the other interventions. The review on vaginal disinfection31 included randomized and controlled clinical trials. The review on vitamin A supplementation32 included only randomized controlled trials. The authors of these reviews assessed quality based on the criteria in the Cochrane Handbook for determining if allocation is adequate, unclear or inadequate33.

Table 1. Characteristics of included reviews
TitleAuthorsDate Last UpdatedPatientsInterventionComparisonOutcomes for which data was found
Antiretrovirals for reducing the risk of mother-to-child transmission of HIV infectionVolmink J Siegfried NL van der Merwe L Brocklehurst P2006Adults•ZDV•ZDV vs. placeboPrimary Outcomes:
   Infants•Nevirapine (NVP)•Longer vs. shorter regimens•HIV infection status at birth, at 2 weeks, at 4 to 8 weeks, at 3 to 4 months, and at 6, 12 and 18 months
    •Lamivudine (3TC)•Regimens of different ARVs and duration•HIV or death at 2 weeks, at 4 to 8 weeks, at 3 to 4 months, and at 6, 12 and 18 months
 Secondary Outcomes:
 •Death at 2 weeks, at 4 to 8 weeks, at 3 to 4 months, and at 6, 12 and 18 months
 •Stillbirth
 •Birth weight (<2500 g)
 •Preterm birth (as defined by the authors)
Efficacy and safety of caesarean delivery for prevention of mother-to-child transmission of HIV-1Read JS2005AdultsCaesarean section before labour and before ruptured membranesVaginal delivery or non-elective caesarean sectionPrimary
 Newell ML Infants  •HIV infection status of child
 Secondary
 •Side effects in child (respiratory morbidity, skin lacerations)
 •Side effects in mother (febrile morbidity, endometritis, haemorrhage or severe anaemia, pneumonia, urinary tract infections)
Interventions for reducing the risk of mother-to-child transmission of HIV infectionBrocklehurst P2002AdultsHyperimmune immunoglobulin + ZDVHyperimmune immunoglobulin + ZDV against non-specific immunoglobulin + ZDVPrimary Outcomes:
   Infants  •Neonatal infection status and survival.
 Secondary Outcomes:
 •Severe acute morbidity
 •Admission to neonatal unit
Vaginal disinfection for preventing mother-to-child transmission of HIV infectionWiysonge CS Shey MS Shang JD Sterne JAC Brocklehurst P2005AdultsVaginal disinfectionNo disinfectionPrimary Outcomes
   Infants  •HIV infection status of child
 Secondary Outcomes
 •Death of child within 15 months
 •Side effects in child (irritation of skin, mucosae, or eyes)
 •Side effects in mother (reproductive tract syndromes and signs in women)
Vitamin A supplementation for reducing the risk of mother-to-child transmission of HIV infectionWiysonge CS Shey MS Sterne JAC Brocklehurst P2005AdultsVitamin A supplementationNo treatmentPrimary Outcomes
   Infants  •HIV infection status of child
 Secondary Outcomes
 •Stillbirth
 •Infant gestational age < 34 weeks
 •Infant gestational age < 37 weeks
 •Birth weight < 2500 g
 •Birth weight < 2000 g
 •Birth weight
 •Maternal death
 •Postpartum CD4 count
 •Infant Death
 •Death of child by 24 months
 •Later death of child
 

Results

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Five systematic reviews examining interventions of HIV and mother to child transmission (MTCT) were identified within the Cochrane Database of Systematic Reviews (Table IIa). Interventions examined within these reviews included: ARV prophylaxis, caesarean section prior to labour and membrane rupture, hyperimmune immunoglobulin plus ZDV, vaginal disinfection, and vitamin A supplementation. The five reviews encompass 25 trials examining seven outcomes. The primary outcome for all reviews was MTCT of HIV. Secondary outcomes included: neonatal mortality, postneonatal mortality, maternal mortality, stillbirth, low birth weight, and preterm birth. One review examining ARV prophylaxis identified 18 unique trials but a meta-analysis was not possible because no two trials assessed identical drug regimens. This review also examined the outcome ‘HIV infection or death’. However, as no other reviews reported this outcome, no comparisons could be made and this outcome has not been reported in this umbrella review.

Table 2a. Mother-to-child transmission of HIV
Author/YearIntervention and Comparison InterventionNumber Participants (# Trials)Control Group RiskRisk DifferenceRelative Risk (95% CI) (Random Effects Model)Quality of EvidenceComments
Volmink et al. 2007Various ARV regimens 

No meta-analysis was conducted as no 2 trials assessed identical drug regimes. Please see specific trials in table IIb

.
Read et al. 2005Elective caesarean section (ECS) vs. vaginal delivery*408 (1)11%− 0.070.34 [0.14, 0.8]HighWomen enrolled from 34 to 36 gestation
 No women breastfed
 Children followed until 18 months after birth
 Stratified by use of ZDV during pregnancy
 ECS vs. vaginal delivery: ZDV during pregnancy*13319%0.64 [0.13, 3.09]High 
 ECS vs. vaginal delivery: no ZDV during pregnancy*2363%0.36 [0.12, 1.03]High 
Brocklehurst 2002Hyperimmune immunoglobulin + ZDV vs. non-specific immunoglobulin + ZDV454 (1)6%− 0.020.67 [0.29, 1.55]Moderate—Women enrolled from 20 to 30 weeks' gestation
 adequacy of allocation concealment unclear 
Wiysonge et al. 2005Vaginal disinfection vs. no disinfection708 (2)22% to 23%− 0.010.94 [0.71, 1.25]HighWomen enrolled before 36 weeks' gestation
Wiysonge et al. 2005Vitamin A supplementation vs. no supplementation2022 (3)22% to 33%0.031.10 [0.95, 1.26]HighWomen enrolled from 12 to 39 weeks' gestation

MTCT of HIV (Table IIb)

ARV prophylaxis

One review examined the efficacy of ARV prophylaxis for the prevention of MTCT of HIV in 18 trials that included 14 398 participants in 16 countries. The majority of trials were conducted in a single country and all trials but one were conducted in resource-limited settings (Côte d'Ivoire, Kenya, Malawi, South Africa, Thailand, Uganda and Zimbabwe). PACTG 076 ZDV is the one trial conducted in in a resource rich setting, being a multi-centre study conducted in the United States and France. Eight trials, including the trial conducted in a resource-rich setting, were conducted with non-breastfeeding mothers. Seven trials examined breastfeeding mothers in Africa. The review found that various ARV regimens are efficacious in preventing MTCT of HIV, and (at least in the short-term) are generally safe.

Table 2b. Mother-to-child transmission of HIV, ARV trials
Author/YearIntervention and Comparison InterventionNumber Participants (# Trials)Control Group RiskRisk DifferenceRelative Risk (95% CI) (Random Effects Model)Quality of EvidenceComments
  1. *Based on actual mode of delivery and stratified by use of ARVs during pregnancy.

  2. Maximum cut-off for each trial reported.

  3. IRA = inadequate reporting of allocation concealment, IRB = inadequate reporting of blinding, IRR = inadequate reporting of method of randomization, NVP = nevirapine, ZDV = zidovudine.

ZDV vs. Placebo
Breastfeeding
RETRO-CIMaternal ZDV from 36 weeks gestation and in labour, no prophylaxis to the infant264 (1)0.63 [0.30, 0.96]Moderate—IRBHIV infection status assessed at 3 to 4 months
PETRAaMaternal ZDV + 3TC from 36 weeks' gestation through labour and ZDV + 3TC to mother and infant for first 7 days after delivery819 (1)0.67 [0.23, 1.12]Moderate—IRB, IRRHIV infection status assessed at 18 months
PETRAbMaternal ZDV + 3TC from onset of labour, and ZDV + 3TC to mother and infant first 7 days after delivery819 (1)0.82 [0.37, 1.27]Moderate—IRB, IRRHIV infection status assessed at 18 months
PETRAcMaternal ZDV + 3TC during labour only820 (1)0.9 [0.39, 1.41]Moderate—IRB, IRRHIV infection status assessed at 18 months
DITRAMEMaternal ZDV from 36 to 38 weeks' gestation, during labour, and 7 days after delivery391 (1)0.7 [0.43, 0.97]Moderate—IRBHIV infection status assessed at 18 months
Non-breastfeeding
Thai-CDCMaternal ZDV from 36 weeks' gestation and in labour with no prophylaxis to infant395 (1)0.5 [0.13, 0.87]HighHIV infection status assessed at 4 to 8 weeks
Limpongsanurak 2001Maternal ZDV from 38 weeks' gestation and in labour with no prophylaxis to infant182 (1)0.91 [0.56, 1.26]Moderate—IRA, IRRHIV infection status assessed at 6 months
PACTG 076Maternal ZDV from 14 to 34 weeks' gestation, through labour and to infant for first 6 weeks after birth402 (1)0.34 [0.03, 0.65]Moderate—IRA, IRB, IRRHIV infection status assessed at 18 months
Longer vs. shorter ARV regimens
Breastfeeding
Thistle 2004Maternal ZDV in labour only and to infant for first 3 days after birth vs. ZDV to mothers from 36 weeks, through labour with no prophylaxis for infant222 (1)0.91 [0.48, 1.34]Moderate—IRBHIV infection status assessed at 12 months
Non-breastfeeding
Bhoopat 2005Maternal ZDV given 62 to 92 days before and through labour vs. given 14 to 35 days before and through labour50 (1)0 [0, 1.14]Moderate—IRA, IRB, IRRHIV infection status assessed at 3 to 4 months
PHPT-1aMaternal ZDV from 28 weeks' gestation, through labour and to infant for first 6 weeks vs. given from 35 weeks' gestation, through labour and to infant for first 6 weeks739 (1)0.76 [0.31, 1.21]Moderate—IRB, IRRHIV infection status assessed at 6 months
PHPT-1bMaternal ZDV from 28 weeks' gestation, through labour and to infant for first 3 days vs. given from 35 weeks' gestation, through labour and to infant during first 6 weeks678 (1)0.55 [0.12, 0.98]Moderate—IRB, IRRHIV infection status assessed at 6 months
PHPT-2Single dose of maternal NVP at labour and to infant shortly after birth vs. single dose of maternal NVP at labour (all mothers receive ZDV during third trimester and infants receive 1 week of ZDV after birth)1365 (1)0.71 [0.16, 1.26]Moderate—IRRHIV infection status assessed at 6 months
Regimens of different ARVs and duration
Breastfeeding
Taha 2004Infant single dose of NVP + ZDV for 1 week vs. infants single dose NVP alone (all mothers received single dose of NVP at onset of labour)889 (1)0.87 [0.58, 1.16]Moderate—IRBHIV infection status assessed at 4 to 8 weeks
Taha 2003Infant NVP (single dose) immediately after birth and ZDV for first week vs. infant NVP (single dose) given immediately after birth865 (1)0.63 [0.3, 0.96]Moderate—IRBHIV infection status assessed at 4 to 8 weeks
HIVNET 012Maternal NVP (single dose) at onset of labour + infant NVP (single dose immediately after birth) vs. maternal ZDV during labour and to infant for 1 week after birth680 (1)0.59 [0.30, 0.88]Moderate—IRB, IRRHIV infection status assessed at 18 months
Non-breastfeeding
PACTG 316Maternal NVP (single dose) during labour + single dose NVP to infants immediately after birth vs. placebo (all women receiving one or more ARVs as part of clinical care)1268 (1)0.87 [0, 1.83]Moderate–IRA, IRB, IRRHIV infection status assessed at 4 to 8 weeks
SAINTMaternal ZDV + 3TC during labour and maternal and infant ZDV + 3TC for 1 week after delivery vs. maternal NVP during labour and a single dose of NVP given to mothers and infants immediately after delivery/birth944 (1)0.76 [0.47, 1.05]Moderate—IRA, IRBHIV infection status assessed at 4 to 8 weeks
Kiarie 2003Thai-CDC vs. HIVNET 012110 (1)0.42 [0, 1.05]Moderate—IRA, IRB, IRRHIV infection status assessed at 4 to 8 weeks
Gray 2005Single dose of NVP given to infants immediately after birth vs. ZDV given to infants for first 6 weeks after birth964 (1)0.6 [0.19, 1.01]Moderate—IRA, IRB, IRRHIV infection status assessed at 3 to 4 months
Gray 2006aMaternal stavudine from 34 to 36 weeks' gestation, through labour and to infants for first 6 weeks vs ZDV in the same regimen183 (1)2.16 [0.51, 3.81]Moderate—IRA, IRBHIV infection status assessed at 6 months
Gray 2006bMaternal didanosine from 34 to 36 weeks' gestation, through labour, and to infants for first 6 weeks vs ZDV in the same regimen186 (1)1.89 [0.3, 3.48]Moderate—IRA, IRBHIV infection status assessed at 6 months
Gray 2006cMaternal stavudine + didanosine from 34 to 36 weeks' gestation, through labour, and to infants for first 6 weeks vs ZDV in the same regimen180 (1)0.82 [0, 2.19]Moderate—IRA, IRBHIV infection status assessed at 6 months
 Relative Risk 
Author/YearIntervention and Comparison InterventionNumber Participants (# Trials)Control Group RiskRisk Difference(95% CI) (Random Effects Model)Quality of EvidenceComments

Three main comparison groups were examined within the review: ZDV vs. placebo, longer vs. shorter ARV regimens, and regimens of different ARVs and duration.

ZDV vs. placebo: Three trials (RETRO-CI, PETRA, and DITRAME) assessed different regimens of ZDV or ZDV plus 3TC vs. placebo in breastfeeding women and found a significant reduction in the risk of HIV infection from 4 to 8 weeks. In RETRO-CI and DITRAME, where ZDV was administered from 36 to 38 weeks' gestation until delivery, this reduction in MTCT of HIV remained significant at 3 to 4 months, and at 18 months in DITRAME where ZDV was additionally given to the mothers for the first seven days after labour. In PETRA-a and PETRA-b, where ZDV plus 3TC was given to mothers either at 36 weeks' gestation until delivery, or beginning at labour and to the mothers and infants for the first seven days after delivery, the initial significant reduction in MTCT of HIV was not maintained at 18 months. In PETRA-c no significant reduction in MTCT of HIV was seen when ZDV plus 3TC prophylaxis was given to mothers exclusively during labour. Three trials (Thai-CDC, PACTG 076, and Limpongsanurak 2001) assessed non-breastfeeding mothers. In PACTG 076 (in which ZDV was given from 14 to 34 weeks' gestation until delivery and to infants for the first 6 weeks) and Thai-CDC (in which ZDV was administered to mothers from 36 weeks' gestation until delivery) both found a reduction in the risk of HIV infection at 4 to 8 weeks, 6 months, or 18 months. In Limpongsanurak 2001, where ZDV was administered from 38 weeks' gestation until delivery, no significant reduction in MTCT of HIV was observed.

Longer vs. shorter ARV regimens: One trial in breastfeeding mothers (Thistle 2004) compared a regimen like the Thai-CDC regimen to administering ZDV to mothers during labour only and to the infants for the first 3 days of life, and found no significant difference in the risk of HIV infection at birth, 4 to 8 weeks, 3 to 4 months, 6 months, or 12 months. However, in non-breastfeeding mothers, PHPT-1b found that ZDV given to mothers from 28 weeks' gestation and to infants for the first 3 days significantly reduced the HIV infection rate of infants at 6 months when compared to a regimen of ZDV given to mothers from 35 weeks' gestation and to infants for the first 6 weeks after birth. Bhoopat 2005 did not find a significant difference in rates of MTCT of HIV between two regimens where ZDV was administered to mothers either beginning 62 to 92 days or 14 to 35 days before delivery, both continuing prophylaxis during labour. PHPT-2 compared regimens in which mothers receiving ZDV where additionally given a dose of NVP in labour with one arm giving NVP to the infants soon after birth. Although no significant reduction in HIV infection was observed between these two arms, a comparison between the NVP regimens and a ZDV-only regimen showed that adding NVP to the prophylactic regimen significantly reduces MTCT of HIV.

Regimens of different ARVs and duration: Three trials (HIVNET 012, Taha 2003, and Taha 2004) compared regimens of different ARVs and duration in breastfeeding mothers. The HIVNET 012 regimen of one maternal and one newborn dose of NVP did result in a lower rate of HIV infection at 4 to 8 weeks until 18 months. Taha 2003 found that giving infants ZDV for the first week of life in addition to a single dose of NVP at birth resulted in a significant reduction in HIV infection at 4 to 8 weeks, but that this reduction was not maintained. Taha 2004 also found that giving ZDV to the infants for one week in addition to the HIVNET 012 regimen did not result in a significant reduction in HIV infection at 4 to 8 weeks compared to the HIVNET 012 regimen alone. Five trials (Gray 2005, Gray 2006, Kiarie 2003, PACTG 316, and SAINT) assessed non-breastfeeding mothers. PACTG 316 compared two regimens where mothers receiving standard ARV prophylaxis and a single dose of NVP during labour with one arm that additionally gave the infants a dose of NVP immediately after birth. SAINT compared a regimen of ZDV given to the mother during labour and to mothers and infants for one week after delivery against giving the mother a dose of NVP during labour and an additional dose to mothers and infants immediately after delivery. Kiarie 2003 compared a ‘Thai-CDC’ regimen with a ‘HIVNET 012′ regimen, but the trial lacked sufficient power to detect a difference in transmission between the two regimens. Gray 2005 compared a program of ZDV administered to infants for the first 6 weeks of life compared to a single dose of NVP given to infants immediately after birth. Finally, Gray 2006 compared three regimens of the nucleosides stavudine, didanosine, and stavudine plus didanosine against regimens of ZDV. None of the above five trials found significant differences in the rates of MTCT of HIV between compared regimens.

Caesarean section before labour and before ruptured membranes

A Cochrane review of caesarean section before labour and before ruptured membranes (elective caesarean section, or ECS) found it to be efficacious in preventing MTCT28. However, only one multi-centre trial has been performed, and this trial included 408 women from Italy, France, United Kingdom, Spain, Switzerland, and Sweden. Ninety-two of the women (40 in the caesarean arm and 52 in the vaginal arm) received ZDV prophylaxis. The rate of MTCT at 18 months differed significantly according to mode of delivery with women who delivered by caesarean section being 20% less likely to transmit HIV to their children than women who delivered vaginally. This clinical trial was conducted in the early-to-mid 1990s, when HIV-infected women received either no ARVs during pregnancy or zidovudine (ZDV) only, and when plasma viral load assays were not routinely available34. Therefore, the efficacy of ECS is unknown among HIV-infected women with low plasma viral loads (e.g. less than 1000 copies/mL). In addition, the efficacy of this intervention among women receiving combination ARV regimens, including highly active ARV therapy (HAART) is unknown. Although the risk of postpartum morbidity among women undergoing caesarean delivery is higher among HIV-infected women (as it is, also, in the general population of women), the available data indicate that the potential benefits of caesarean section for prevention of MTCT outweigh the potential detrimental effect35.

Hyperimmune HIV Immunoglobulin

Brocklehurst 2002 reviewed one trial that compared hyperimmune HIV immunoglobulin plus ZDV against non-specific immunoglobulin (IVIG) plus ZDV. The trial included 501 women from the United States and Puerto Rico. The addition of hyperimmune immunoglobulin to ZDV did not have any additional effect on the risk of MTCT.

Vaginal disinfection

This Cochrane review of vaginal disinfection found no evidence of efficacy for prevention of MTCT of HIV. Vaginal disinfection during labour was examined in two randomized controlled trials that consisted of 708 women from Abidjan, Côte d’Ivoire; Bobo-Dioulasso, Burkina Faso; and Mombasa, Kenya. The trials used either a solution of 0.25% chlorhexidine or 1% benzalkonium chloride. The combined trials had only 78% power to detect a 30% reduction in the risk of MTCT of HIV, and less than 10% power to detect a 6% reduction in the risk of MTCT of HIV, from a baseline of 30%.

The authors point out the need for a large, well-designed and well-conducted randomized controlled trial to assess the additive effect of vaginal and/or infant surface decontamination on the risk of MTCT of HIV among women receiving ARVs31. Such a trial could incorporate higher concentrations of chlorhexidine, for which there are supportive data36.

Vitamin A supplementation

Vitamin A supplementation was examined in four trials that enrolled a total of 3033 HIV-infected pregnant women. Participants were drawn respectively from KwaZulu-Natal, South Africa; Dar es Salaam, Tanzania; Blantyre, Malawi; and Harare, Zimbabwe. Three trials reported outcome data on MTCT, with two finding no evidence that the effect of vitamin A supplementation was different from placebo (at 3 and 24 months, respectively), while the remaining trial found evidence that vitamin A supplementation increased the risk of MTCT of HIV at 24 months. The Cochrane review of vitamin A supplementation for prevention of MTCT concluded that the available evidence did not support the use of vitamin A supplementation of HIV-infected pregnant women33. The most recent vitamin A trial results37, 38, which were released after publication of Wiysonge 2005a, are consistent with the conclusions of this review39.

Neonatal mortality (Table 3)

Of the interventions examined that reported on neonatal mortality, none found a significant difference on neonatal mortality between comparison groups. The single trial that assessed hyperimmune HIV immunoglobulin plus ZDV versus IVIG plus ZDV, found no evidence that the addition of hyperimmune HIV immunoglobulin reduces the risk neonatal mortality. Only one trial examining vaginal disinfection and assessing neonatal death and death after neonatal period was included. Mandelbrot 2002, which included 108 women, found there was no evidence of effect on neonatal death. Of two trials that compared ZDV to placebo for the risk of neonatal death, RETRO-CI showed a reduction of risk of 0.13 for those women receiving ZDV. HIVNET 012, which compared regimens of different ARVs and duration of treatment, reported on neonatal mortality and found no significant difference in risk.

Table 3. Neonatal mortality
Author/YearIntervention and Comparison InterventionNumber Participants (# Trials)Control Group RiskRisk Difference(95% CI) (Random Effects Model)Relative Risk Quality of EvidenceComments
  1. IRA = inadequate reporting of allocation concealment, IRB = inadequate reporting of blinding, IRR = inadequate reporting of method of randomization, NVP = nevirapine, ZDV = zidovudine.

Brocklehurst 2002Hyperimmune immunoglobulin + ZDV vs. non-specific immunoglobulin + ZDV506 (1)0.4%01.91 [0.17, 20.90]Moderate— adequacy of allocation concealment unclear 
Wiysonge et al. 2005Vaginal disinfection vs. no disinfection111 (1)5%0.021.36 [0.32, 5.79]High 
Volmink et al. 2007Various ARV regimens     No meta-analysis was conducted as no 2 trials assessed identical drug regimes. Please see specific trials below.
ZDV vs. Placebo
RETRO-CIMaternal ZDV from 36 weeks gestation and in labour, no prophylaxis to the infant280 (1)6%− 0.050.13 [0.02, 0.99]Moderate—IRB 
DITRAMEMaternal ZDV from 36 to 38 weeks' gestation, during labour, and 7 days after delivery431 (1)1%0.012.03 [0.51, 8.00]Moderate—IRB 
Regimens using different ARVs and different duration
HIVNET 012Maternal NVP (single dose) at onset of labour + infant NVP (single dose immediately after birth) vs. maternal ZDV during labour and to infant for 1 week after birth626 (1)0.6%0.012.50 [0.49, 12.79]Moderate—IRR, IRB 

Postneonatal mortality (Table 4)

The majority of trials that reported on deaths after neonatal period did not find a significant reduction in deaths. The single trial that assessed hyperimmune HIV immunoglobulin plus ZDV versus IVIG plus ZDV, found no evidence that the addition of hyperimmune HIV immunoglobulin reduces the risk of death after neonatal period. Only one trial examining vaginal disinfection and assessing neonatal death and death after neonatal period was included. The trial, which included 108 women, found there was no evidence of effect on risk of death after the neonatal period. Four trials that compared ARVs to placebo and reported on death after the neonatal period (4 to 8 weeks) found no evidence of difference from placebo. Three trials reported mortality after 3 to 4 months with one (RETRO-CI) finding a significant reduction compared to placebo. This outcome was also assessed at 6 months, 12 months and 18 months by one trial, one trial, and three trials respectively, none finding a significant difference from placebo. For studies comparing longer versus shorter regimens using the same ARV, mortality was assessed at 4 to 8 weeks (1 trial), 3 to 4 months (1 trial), 6 months (3 trials), and 12 months (1 trial) with only one trial (PHPT-2) reporting a significant difference in risk at 6 months for those women on shorter regimens. Studies of regimens of different ARVs and duration of treatment reported neonatal mortality at 4 to 8 weeks (7 trials), 6 months (1 trial), and 18 months (1 trial), with no trials reporting a significant difference in risk.

Table 4. Postneonatal mortality
Author/YearIntervention and Comparison InterventionNumber Participants (# Trials)Control Group RiskRisk DifferenceRelative Risk (95% CI) (Random Effects Model)Quality of EvidenceComments
  1. IRA = inadequate reporting of allocation concealment, IRB = inadequate reporting of blinding, IRR = inadequate reporting of method of randomization, NVP = nevirapine, ZDV = zidovudine.

Brocklehurst 2002Hyperimmune immunoglobulin + ZDV vs. non-specific immunoglobulin + ZDV506 (1)0.8%00.48 [0.04, 5.23]Moderate—Adequacy of allocation concealment unclear 
Wiysonge et al. 2005Vaginal disinfection vs. no disinfection108 (1)11%0.041.39 [0.52, 3.71]High 
Volmink et al. 2007Various ARV regimens     No meta-analysis was conducted as no 2 trials assessed identical drug regimes. Please see specific trials below.
ZDV vs. placebo (deaths first 4 to 8 wks)
PETRAaMaternal ZDV + 3TC from 36 weeks' gestation through labour and ZDV + 3TC to mother and infant for first 7 days after delivery852 (1)2%− 0.020.35 [0.11, 1.14]Moderate—IRB, IRR 
PETRAbMaternal ZDV + 3TC from onset of labour, and ZDV + 3TC to mother and infant first 7 days after delivery848 (1)2%00.98 [0.41, 2.34]Moderate—IRB, IRR 
PETRAcMaternal ZDV + 3TC during labour only851 (1)2%− 0.010.71 [0.28, 1.81]Moderate—IRB, IRR 
Thai-CDCMaternal ZDV from 36 weeks' gestation and in labour with no prophylaxis to infant397 (1)1%− 0.010.50 [0.05, 5.50]High 
DITRAMEMaternal ZDV from 36 to 38 weeks' gestation, during labour, and 7 days after delivery431 (1)2%0.011.77 [0.53, 5.97]Moderate—IRB 
Longer vs. shorter regimens using same ARVs (death first 4 to 8 wks)
Thistle 2004Maternal ZDV in labour only and to infant for first 3 days after birth vs. ZDV to mothers from 36 weeks, through labour with no prophylaxis for infant222 (1)3%01.00 [0.21, 4.85]Moderate—IRB 
Regimens using different ARVs and different durations (death first 4 to 8 wks)
PACTG 316Maternal NVP (single dose) during labour + single dose NVP to infants immediately after birth vs. placebo (all women receiving one or more ARVs as part of clinical care)1506 (1)0.7%00.60 [0.14, 2.50]Moderate—IRA, IRB, IRR 
SAINTMaternal ZDV + 3TC during labour and maternal and infant ZDV + 3TC for 1 week after delivery vs. maternal NVP during labour and a single dose of NVP given to mothers and infants immediately after delivery/birth1319 (1)3%01.01 [0.54, 1.89]Moderate— IRA, IRB 
Kiarie 2003Thai-CDC vs. HIVNET 012139 (1)4%00.99 [0.21, 4.72]Moderate— IRA, IRB, IRR 
Gray 2005Single dose of NVP given to infants immediately after birth vs. ZDV given to infants for first 6 weeks after birth1051 (1)2%01.15 [0.52, 2.54]Moderate— IRA, IRB, IRR 
Taha 2003Infant NVP (single dose) immediately after birth and ZDV for first week vs. infant NVP (single dose) given immediately after birth1119 (1)2%0.011.26 [0.60, 2.67]Moderate— IRB 
Taha 2004Infant single dose of NVP + ZDV for 1 week vs. infants single dose NVP alone (all mothers received single dose of NVP at onset of labour)894 (1)0.9%0.011.74 [0.51, 5.91]Moderate— IRB 
HIVNET 012Maternal NVP (single dose) at onset of labour + infant NVP (single dose immediately after birth) vs. maternal ZDV during labour and to infant for 1 week after birth626 (1)1%0.022.50 [0.79, 7.89]Moderate— IRB, IRR 

Maternal deaths (Table 5)

Only two trials reported on maternal mortality. The single trial that assessed hyperimmune HIV immunoglobulin plus ZDV versus IVIG plus ZDV, found no evidence that the addition of hyperimmune HIV immunoglobulin reduces the risk maternal death. A single trial examining the effect of vitamin supplementation on the risk of maternal death found no evidence of effect.

Table 5. Maternal mortality
Author/YearIntervention and Comparison InterventionNumber Participants (# Trials)Control Group RiskRisk DifferenceRelative Risk (95% CI) (random effects model)Quality of EvidenceComments
Brocklehurst 2002Hyperimmune immunoglobulin + ZDV vs. non-specific immunoglobulin + ZDV459 (1)0.9%0.011.97 [0.37, 10.67]Moderate—adequacy of allocation concealment unclear 
Wiysonge et al. 2005Vitamin A supplementation vs. no supplementation728 (1)0.6%00.49 [0.04, 5.37]High 

Stillbirth rates (Table 6)

None of the interventions examined reported a significant difference in the risk of stillbirth. The single trial that assessed hyperimmune HIV immunoglobulin plus ZDV versus IVIG plus ZDV, found no evidence that the addition of hyperimmune HIV immunoglobulin reduces the risk of stillbirth. For vitamin A supplementation, the combined results of four trials found no evidence of effect on the risk of stillbirth. Six trials that compared ZDV to placebo reported on risk of stillbirth with no trial finding a significant reduction in risk compared with placebo. Three trials examining longer versus shorter regimens using the same ARV and reporting on risk of stillbirth found no evidence of difference. Three trials examining regimens of different ARVs and duration of treatment and reported on stillbirth rates found no evidence of difference.

Table 6. Stillbirth
Author/YearIntervention and Comparison InterventionNumber Participants (# Trials)Control Group RiskRisk DifferenceRelative Risk (95% CI) (Random Effects Model)Quality of EvidenceComments
  1. IRA = inadequate reporting of allocation concealment, IRB = inadequate reporting of blinding, IRR = inadequate reporting of method of randomization, NVP = nevirapine, ZDV = zidovudine.

Brocklehurst 2002Hyperimmune immunoglobulin + ZDV vs. non-specific immunoglobulin + ZDV459 (1)0.8%00.33 [0.01, 8.03]Moderate—adequacy of allocation concealment unclear 
Wiysonge et al. 2005Vitamin A supplementation vs. no supplementation2855 (4)0.8% to 8%00.99 [0.68, 1.43]High 
Volmink et al. 2007Various ARV regimens     No meta-analysis was conducted as no 2 trials assessed identical drug regimes. Please see specific trials below.
ZDV vs. placebo
DITRAMEMaternal ZDV from 36 to 38 weeks' gestation, during labour, and 7 days after delivery431 (1)3%− 0.030.14 [0.02, 1.17]Moderate—IRB 
PACTG 076Maternal ZDV from 14 to 34 weeks' gestation, through labour and to infant for first 6 weeks after birth477 (1)0.4%00.33 [0.01, 8.11]Moderate—IRA, IRB, IRR 
PETRAaMaternal ZDV + 3TC from 36 weeks' gestation through labour and ZDV + 3TC to mother and infant for first 7 days after delivery852 (1)1%− 0.010.40 [0.07, 2.15]Moderate—IRB, IRR 
PETRAbMaternal ZDV + 3TC from onset of labour, and ZDV + 3TC to mother and infant first 7 days after delivery851 (1)1%01.19 [0.34, 4.20]Moderate—IRB, IRR 
PETRAcMaternal ZDV + 3TC during labour only848 (1)1%00.80 [0.20, 3.18]Moderate—IRB, IRR 
Thai-CDCMaternal ZDV from 36 weeks' gestation and in labour with no prophylaxis to infant397 (1)0%0.013.02 [0.12, 73.57]High 
Limpongsanurak 2001Maternal ZDV from 38 weeks' gestation and in labour with no prophylaxis to infant182 (1)0%0.013.07 [0.13, 74.28]Moderate— IRA, IRR 
RETRO-CIMaternal ZDV from 36 weeks' gestation and in labour, no prophylaxis to the infant280 (1)1%0.043.50 [0.74, 16.55]Moderate— IRB 
Longer vs. shorter regimens using same ARVs
Thistle 2004Maternal ZDV in labour only and to infant for first 3 days after birth vs. ZDV to mothers from 36 weeks, through labour with no prophylaxis for infant222 (1)0%0not estimableModerate— IRB 
Bhoopat 2005Maternal ZDV given 62 to 92 days before and through labour vs. given 14 to 35 days before and through labour50 (1)0%− 0.02not estimableModerate— IRA, IRB, IRR 
PHPT-1aMaternal ZDV from 28 weeks' gestation, through labour and to infant for first 6 weeks vs. given from 35 weeks' gestation, through labour and to infant for first 6 weeks764 (1)4%− 0.020.55 [0.23, 1.33]Moderate— IRB, IRR 
PHPT-1bMaternal ZDV from 28 weeks' gestation, through labour and to infant for first 3 days vs. given from 35 weeks' gestation, through labour and to infant during first 6 weeks695 (1)4%− 0.020.33 [0.11, 1.01]Moderate— IRB, IRR 
PHPT-2Single dose of maternal NVP at labour and to infant shortly after birth vs. single dose of maternal NVP at labour (all mothers receive ZDV during third trimester and infants receive 1 week of ZDV after birth)1445 (1)1%− 0.010.25 [0.05, 1.17]Moderate— IRR 
Regimens using different ARVs and different durations
Kiarie 2003Thai-CDC vs. HIVNET 012139 (1)3%0.011.48 [0.25, 8.58]Moderate—IRA, IRB, IRR 
HIVNET 012Maternal NVP (single dose) at onset of labour + infant NVP (single dose immediately after birth) vs. maternal ZDV during labour and to infant for 1 week after birth626 (1)0.3%02.00 [0.18, 21.94]Moderate—IRB, IRR 
PACTG 316Maternal NVP (single dose) during labour + single dose NVP to infants immediately after birth vs. placebo (all women receiving one or more ARVs as part of clinical care)1506 (1)0%02.99 [0.12, 73.33]Moderate—IRA, IRB, IRR 

Low birth weight (Table 7)

The single trial that assessed hyperimmune HIV immunoglobulin plus ZDV versus IVIG plus ZDV, found no evidence that the addition of hyperimmune HIV immunoglobulin reduces the risk of low birth weight (<2500 g). The combined results of four trials found no evidence of effect in reducing risk of low birth weight for vitamin A supplementation. Three trials that examined the effect of ZDV on risk of low birth weight found no evidence of difference from placebo. Two trials comparing longer versus shorter regimens using the same ARV reported on risk of low birth weight with one arm of a trial (PHPT-1a) finding that women on longer regimens were one and a half times more likely to have a low birth weight infant than women on shorter regimens. Four trials compared regimens of different ARVs and duration of treatment and reported on risk of low birth weight, with none finding a significant difference.

Table 7. Low birth weight (<2500 g)
Author/YearIntervention and Comparison InterventionNumber Participants (# Trials)Control Group RiskRisk DifferenceRelative Risk (95% CI) (Random Effects Model)Quality of EvidenceComments
  1. IRA = inadequate reporting of allocation concealment, IRB = inadequate reporting of blinding, IRR = inadequate reporting of method of randomization, NVP = nevirapine, ZDV = zidovudine.

Brocklehurst 2002Hyperimmune immunoglobulin + ZDV vs. non-specific immunoglobulin + ZDV458 (1)14%0.021.14 [0.73, 1.76]Moderate—Adequacy of allocation concealment unclear 
Wiysonge et al. 2005Vitamin A supplementation vs. no supplementation2606 (4)14% to 21%− 0.020.83 [0.68, 1.01]High 
Volmink et al. 2007Various ARV regimens     No meta-analysis was conducted as no 2 trials assessed identical drug regimes. Please see specific trials below.
ZDV vs. placebo
Thai-CDCMaternal ZDV from 36 weeks' gestation and in labour with no prophylaxis to infant397 (1)9%− 0.040.61 [0.30, 1.27]High 
PACTG 076Maternal ZDV from 14 to 34 weeks' gestation, through labour and to infant for first 6 weeks after birth477 (1)16%− 0.030.75 [0.48, 1.19]Moderate—IRA, IRB, IRR 
DITRAMEMaternal ZDV from 36 to 38 weeks' gestation, during labour, and 7 days after delivery431 (1)14%− 0.010.92 [0.57, 1.47]Moderate—IRB 
Longer vs. shorter regimens using same ARVs
PHPT-1aMaternal ZDV from 28 weeks' gestation, through labour and to infant for first 6 weeks vs. given from 35 weeks' gestation, through labour and to infant for first 6 weeks764 (1)7%0.051.65 [1.04, 2.60]Moderate—IRB, IRR 
PHPT-1bMaternal ZDV from 28 weeks' gestation, through labour and to infant for first 3 days vs. given from 35 weeks' gestation, through labour and to infant during first 6 weeks695 (1)7%0.041.50 [0.92, 2.43]Moderate—IRB, IRR 
PHPT-2Single dose of maternal NVP at labour and to infant shortly after birth vs. single dose of maternal NVP at labour (all mothers receive ZDV during third trimester and infants receive 1 week of ZDV after birth)1445 (1)9%− 0.010.85 [0.60, 1.19]Moderate—IRR 
Regimens using different ARVs and different durations
HIVNET 012Maternal NVP (single dose) at onset of labour + infant NVP (single dose immediately after birth) vs. maternal ZDV during labour and to infant for 1 week after birth626 (1)7%− 0.020.67 [0.35, 1.29]Moderate—IRB, IRR 
PACTG 316Maternal NVP (single dose) during labour + single dose NVP to infants immediately after birth vs. placebo (all women receiving one or more ARVs as part of clinical care)1506 (1)10%0.011.14 [0.85, 1.53]Moderate—IRA, IRB, IRR 
SAINTMaternal ZDV + 3TC during labour and maternal and infant ZDV + 3TC for 1 week after delivery vs. maternal NVP during labour and a single dose of NVP given to mothers and infants immediately after delivery/birth1319 (1)0.8%01.41 [0.45, 4.42]Moderate—IRA, IRB 
Kiarie 2003Thai-CDC vs. HIVNET 012139 (1)1%0.011.97 [0.18, 21.24]Moderate—IRA, IRB, IRR 

Preterm birth (Table 8)

The single trial that assessed hyperimmune HIV immunoglobulin plus ZDV versus IVIG plus ZDV, found no evidence that the addition of hyperimmune HIV immunoglobulin reduces the risk of preterm birth. The combined results of three trials found no evidence that vitamin A supplementation has an effect on the risk of preterm births (<37 weeks). Two trials compared ZDV to placebo and reported on the risk of preterm birth, with one trial (DITRAME) finding a significant reduction in risk compared with placebo. Two trials, reporting preterm birth, comparing longer versus shorter ARV treatments found no evidence of difference in risk. Two trials compared regimens of different ARVs and duration of treatment and reported on risk of preterm birth; neither found evidence of difference in risk.

Table 8. Preterm birth (<37 completed weeks of gestation)
Author/YearIntervention and Comparison InterventionNumber Participants (# Trials)Control Group RiskRisk DifferenceRelative Risk (95% CI) (Random Effects Model)Quality of EvidenceComments
  1. IRA = inadequate reporting of allocation concealment, IRB = inadequate reporting of blinding, IRR = inadequate reporting of method of randomization, NVP = nevirapine, ZDV = zidovudine

Brocklehurst 2002Hyperimmune immunoglobulin + ZDV vs. non-specific immunoglobulin + ZDV458 (1)17%0.021.14 [0.77, 1.69]Moderate—adequacy of allocation concealment unclear 
Wiysonge et al. 2005Vitamin A supplementation vs. no supplementation (preterm = < 37 wk)2110 (3)15% to 22%− 0.020.91 [0.76, 1.09]High 
Volmink et al. 2007Various ARV regimens     No meta-analysis was conducted as no 2 trials assessed identical drug regimes. Please see specific trials below.
ZDV vs. placebo
DITRAMEMaternal ZDV from 36 to 38 weeks' gestation, during labour, and 7 days after delivery431 (1)7%− 0.060.14 [0.03, 0.58]Moderate—IRB 
PACTG 076Maternal ZDV from 14 to 34 weeks' gestation, through labour and to infant for first 6 weeks after birth477 (1)5%0.011.23 [0.60, 2.49]Moderate—IRA, IRB, IRR 
Longer vs. shorter regimens using same ARVs
Thistle 2004Maternal ZDV in labour only and to infant for first 3 days after birth vs. ZDV to mothers from 36 weeks, through labour with no prophylaxis for infant222 (1)4%0.031.75 [0.53, 5.81]Moderate—IRB 
PHPT-1aMaternal ZDV from 28 weeks' gestation, through labour and to infant for first 6 weeks vs. given from 35 weeks' gestation, through labour and to infant for first 6 weeks764 (1)3%0.032.01 [0.94, 4.31]Moderate—IRB, IRR 
PHPT-1bMaternal ZDV from 28 weeks' gestation, through labour and to infant for first 3 days vs. given from 35 weeks' gestation, through labour and to infant during first 6 weeks695 (1)3%0.021.86 [0.84, 4.12]Moderate—IRB, IRR 
Regimens using different ARVs and different durations
PACTG 316Maternal NVP (single dose) during labour + single dose NVP to infants immediately after birth vs. placebo (all women receiving one or more ARVs as part of clinical care)1506 (1)16%0.011.97 [0.37, 10.42]Moderate—IRA, IRB, IRR 
Kiarie 2003Thai-CDC vs. HIVNET 012139 (1)3%0.031.05 [0.83, 1.32]Moderate—IRA, IRB, IRR 

Discussion

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

This ‘umbrella review’ of five separate Cochrane reviews27, 28, 17, 31, 32 found that ARV prophylaxis (several regimens) and caesarean section before labour and before ruptured membranes were efficacious interventions to prevent MTCT of HIV. Other interventions evaluated (hyperimmune HIV immunoglobulin, vaginal disinfection, vitamin A supplementation) were not found to be efficacious for the prevention of MTCT.

There is not yet a Cochrane review on feeding interventions for prevention of MTCT, although one is in preparation. One randomized controlled trial, however, has demonstrated that complete avoidance of breastfeeding is efficacious in preventing MTCT6. MTCT of HIV through breast milk accounts for a large proportion of all HIV infections in children5, 2 and the presence or absence of breastfeeding is an important factor in MTCT. While breast milk is the recommended infant food for the first six months of life, current United Nations guidelines call for nursing mothers with HIV to give their infants replacement foods, and completely avoid giving breast milk, if it is acceptable, feasible, affordable, sustainable, and safe to do so11. In many parts of the world, however, this is not the case, particularly in resource-limited settings.

The authors of the Cochrane reviews examined here worked in accord with the Cochrane Handbook's current recommendations to search comprehensively and extensively through all relevant bibliographic databases, through conference abstracts, and other means, so that respectively they were able to find all eligible trials33. The authors in each case fully described the methods of included trials and endeavoured to contact investigators for clarifications or to obtain missing data.

There has been progress made in identifying efficacious interventions for the prevention of MTCT of HIV, but this has been primarily for prevention of in utero and intrapartum transmission. The majority of cases of MTCT of HIV occur in parts of the world where breastfeeding is the norm, and a large proportion of cases of MTCT are acquired through breast milk transmission. Thus, preventing transmission of HIV through breastfeeding remains a major hurdle.

Conclusions

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

ARV prophylaxis and caesarean section before labour and before ruptured membranes are efficacious interventions for the prevention of MTCT of HIV. Other interventions, including maternal/infant cleansing with chlorhexidine or other compounds, vitamin A supplementation, and hyperimmune HIV immunoglobulin have not been shown to be efficacious. A Cochrane review of infant feeding interventions for the prevention of MTCT of HIV is in progress.

Implications for practice

All HIV-infected pregnant women who meet criteria for ARV treatment should receive such treatment (and once started, it should be continued indefinitely). HIV-infected pregnant women who do not yet meet criteria for ARV treatment should receive ARV prophylaxis. Over the past several years, the complexity of ARV regimens recommended for prevention of MTCT by the World Health Organization and other groups has increased. An important implication for practice is that adherence to ARV regimens would be expected to decrease as the complexity (duration, number of doses per day) increases. Thus, not only must the efficacy of a given regimen be considered, but also its actual use effectiveness. The toxicity profile of a given ARV (or ARV regimen) must also be considered, as adherence diminishes with increasing toxicity of a given ARV. There is also room to ‘fine-tune’ ARV interventions, and to combine an ARV regimen with ECS, if it is safe and feasible to do so. In resource-limited places where HAART is not routinely available shorter, less expensive ARV regimens for reducing MTCT should be considered, as the evidence shows that the benefits associated with such an intervention outweigh the potential risks. While it isn't yet clear which regimen is best, a combination of ZDV and 3TC given to mothers in the antenatal, intrapartum and postpartum periods, and to infants for a week after delivery, or a regimen consisting of a single dose of NVP given to mothers in labour and infants immediately after birth, appear to be both effective and feasible. ZDV monotherapy is also useful, especially if it includes a long antenatal component. When HIV-infected pregnant women present only at the time of delivery (i.e. no antenatal care), post-exposure prophylaxis for the infant with a single dose of NVP immediately after birth, plus ZDV for the first 6 weeks of life, is efficacious.

With regard to caesarean section before labour and before ruptured membranes, the U.S. Public Health Service35 and other groups have recommended that caesarean section should be considered for women with peripheral blood viral loads greater than 1000 copies/mL, irrespective of receipt of ARVs, as an intervention to prevent transmission. Despite the conservative wording of these recommendations35, more recent data suggest a persistent effect of caesarean section delivery, even among those HIV-infected pregnant women with viral loads of less than 1000 copies/mL40, 41, 42. Caesarean delivery as an intervention to prevent MTCT of HIV is generally not feasible in resource-poor settings because of the lack of a skilled attendant during labour and other reasons. In general, the benefit of ECS outweighs the risk of PPM among HIV-infected women, but the risk/benefit ratio depends upon the underlying rate of MTCT. With very low rates of MTCT, the risks associated with ECS may outweigh the benefit. Studies of ECS among HIV-infected women have been conducted almost exclusively in North America and Europe. In developing countries, the risks and benefits associated with ECS have been largely unexplored.

Implications for research

There is an ongoing need for evaluation of ARV prophylaxis regimens, in order to improve the effectiveness and also to simplify regimens and to minimize toxicity for both the mother and the infant. Research on the long-term impact of drug resistance on future treatment of HIV-infected mothers and HIV-infected infants is urgently needed and ARV regimens that reduce the likelihood of resistance mutations emerging should be developed and evaluated. Future trials should, as far as possible, use standardized methods that will allow reliable comparisons of the efficacy of different prophylaxis regimens in different trials and that include adequate follow-up to fully assess long term safety of ARVs.

The risk of MTCT of HIV according to mode of delivery among HIV-infected women with low viral loads (low either because the woman's HIV disease is not advanced, or because her HIV disease is well-controlled with ARV therapy) is unclear. Preliminary data suggest continued benefit of ECS among HIV-infected women with plasma viral loads < 1000 copies/mL41 or women receiving HAART43. The benefit of ECS for prevention of MTCT of HIV may persist among women with low plasma viral loads because of compartmentalization of HIV reservoirs, i.e., a low plasma viral load does not necessarily indicate a low viral load among genital tract secretions. Therefore, an important issue to be addressed in one or more large studies (individual studies, or an individual patient data meta-analysis combining data from more than one study) is an assessment of the effectiveness of caesarean section for prevention of MTCT of HIV among HIV-infected women with undetectable viral loads (with or without receipt of HAART).

References

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References
  • 1
    UNAIDS/WHO. AIDS epidemic update: December 2006. [Cited 2007 Jan 10]. Available from: http://www.unaids.org/En/HIV_Data/Epi2006/Default.asp.
  • 2
    Kourtis AP, Bulterys M, Nesheim SR, Lee FK. Understanding the timing of HIV transmission from mother to infant. JAMA. 2001; 285(6): 70912.
  • 3
    Kuhn L, Stein ZA. Mother-to-infant HIV transmission: timing, risk factors and prevention. Paediatr Perinat Epidemiol. 1995; 9: 129.
  • 4
    Mofenson LM. Mother-child HIV-1 transmission: timing and determinants. Obstet Gynecol Clin North Am. 1997; 24: 75984.
  • 5
    BHITS. Breastfeeding and HIV International Transmission Study Group. Late postnatal transmission of HIV-1 in breast-fed children: an individual patient data meta-analysis. J Infect Dis. 2004; 189(12): 215466.
  • 6
    Nduati R, John G, Mbori Ngacha D, Richardson B, Overbaugh J, Mwatha A, et al. Effect of breastfeeding and formula feeding on transmission of HIV-1: a randomized clinical trial. JAMA. 2000; 283: 116774.
  • 7
    Rankin WW, Brennan S, Schell E, Laviwa J, Rankin SH. The stigma of being HIV-positive in Africa. PLoS Med. 2005; 2(8): e247.
  • 8
    Read JS, AIDS atCoP, American Academy of Pediatrics. Human milk, breastfeeding, and transmission of HIV infection in the U.S. Pediatrics. 2003; 112(5): 1196205.
  • 9
    Coutsoudis A, Pillay K, Spooner E, Kuhn L, Coovadia HM. Randomized trial testing the effect of vitamin A supplementation on pregnancy outcomes and early mother-to-child transmission of HIV-1 in Durban, South Africa. AIDS. 1999; 13: 151724.
  • 10
    Iliff PJ, Piwoz EG, Tavengwa NV, Zunguza CD, Marinda ET, Nathoo KJ. Early exclusive breastfeeding reduces the risk of postnatal HIV-1 transmission and increases HIV-free survival. AIDS. 2005; 19(7): 699708.
  • 11
    UNFPA/UNICEF/WHO/UNAIDS Inter-Agency Team on Mother-to-Child Transmission of HIV. New Data on the Prevention of Mother-to-Child Transmission of HIV and Their Policy Implications. Geneva: WHO; 2001; Conclusions and Recommendations. Ordering code: WHO/RHR/01.28. Available from: http://www.who.int/reproductive-health/publications/new_data_prevention_mtct_hiv/text.pdf.
  • 12
    Douglas GC. Maternal-fetal transmission of human immunodeficiency virus: a review of possible routes and cellular mechanisms of infection. Clin Infect Dis. 1992; 15: 67891.
  • 13
    Ebbesen P, Toth F, Aboagye-Mathiesen G, Zachar V, Hager H, Norskov-Lauritsenl N, et al. Vertical transmission of HIV: possible mechanisms and placental responses. Trophoblast Res. 1994; 8: 117.
  • 14
    Schwarts DA, Nahmias AJ. Human immunodeficiency virus and the placenta. Current concepts of vertical transmission in relation to other viral agents. Ann Clin Lab Sci. 1991; 21(4): 26474.
  • 15
    Reuben J, Lee B-N, Popek EJ. HIV and the Placenta. Immunol Allergy Clin North Am. 1998; 18: 371400.
  • 16
    Purswani MU, Johann Liang R, McNeeley M, Noel GJ. Effect of intravenous immune globulin (IVIG) on interleukin-6 (IL-6) production by whole blood [abstract 894]. Pediatr Res. 1998; 43(part 2): 154a.
  • 17
    Brocklehurst P. Interventions for reducing the risk of mother-to-child transmission of HIV infection. Cochrane Database of Systematic Reviews 2002, Issue 1. Art. No.: CD000102. DOI: 10.1002/14651858.CD000102.
  • 18
    Ross AC, Stephen CB. Vitamin A retinoids in antiviral responses. Federation of American Societies for Experimental Biology (FASEB). 1996; 10: 97985.
  • 19
    Semba RD. The role of vitamin A and related carotenoids in immune function. Nutr Rev. 1998; 56: S3848.
  • 20
    West KP Jr, Rice A, Sugimoto JD. Tables on the Global Burden of Vitamin A Deficiency among Preschool Children and Women of Reproductive Age. [Cited 2007 Mar 2]. Available from: http://www.jhsph.edu/CHN/GlobalVAD.html.
  • 21
    Christian P, West KP, Khatry SK. Night blindness in pregnancy in rural Nepal—nutritional and health risks. Int J Epidemiol. 1998; 27: 2317.
  • 22
    Christian P, West KP, Khatry SK. Night blindness during pregnancy and subsequent mortality among women in Nepal: effects of vitamin A and beta-carotene supplementation. Am J Epidemiol. 2000; 152: 5427.
  • 23
    Semba RD, Miotti PG, Chiphangwi JD, Saah AJ, Canner JK, Dallabetta GA, et al. Maternal vitamin A deficiency and mother-to-child transmission of HIV-1. Lancet. 1994; 343(8913): 15937.
  • 24
    Abrams B, Duncan D, Hertz-Picciotto I. A prospective study of dietary intake and acquired immune deficiency syndrome in HIV-seropositive homosexual men. J Acquir Immune Defic Syndr Hum Retrovirol. 1993; 6: 94958.
  • 25
    Tang AM, Graham NM, Kirby AJ, McCall LD, Willett WC, Saah AJ. Dietary micronutrient intake and risk of progression to acquired immunodeficiency syndrome (AIDS) in human immunodeficiency virus type 1 (HIV-1)-infected homosexual men. Am J Epidemiol. 1993; 138: 93751.
  • 26
    GRADE Working Group. Grading of Recommendations Assessment, Development and Evaluation. Available from: http://www.Gradeworkinggroup.org.
  • 27
    Volmink J, Siegfried NL, van der Merwe L, Brocklehurst P. Antiretrovirals for reducing the risk of mother-to-child transmission of HIV infection. Cochrane Database of Systematic Reviews 2007, Issue 1. Art. No.: CD003510. DOI: 10.1002/14651858.CD003510.pub2.
  • 28
    Read JS, Newell ML. Efficacy and safety of cesarean delivery for prevention of mother-to-child transmission of HIV-1. Cochrane Database of Systematic Reviews 2005, Issue 4. Art. No.: CD005479. DOI: 10.1002/14651858.CD005479.
  • 29
    ter Riet G. Validity checklist for clinical trials. Complement Ther Med. 1997; 5: 1168.
  • 30
    OHRI. Newcastle-Ottawa Quality Scale. Ottawa Health Research Institute. [Cited 2007 Mar 2]. Available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.htm.
  • 31
    Wiysonge CS, Shey MS, Shang JD, Sterne JAC, Brocklehurst P. Vaginal disinfection for preventing mother-to-child transmission of HIV infection. Cochrane Database of Systematic Reviews 2005, Issue 4. Art. No.: CD003651. DOI: 10.1002/14651858.CD003651.pub2.
  • 32
    Wiysonge CS, Shey MS, Sterne JAC, Brocklehurst P. Vitamin A supplementation for reducing the risk of mother-to-child transmission of HIV infection. Cochrane Database of Systematic Reviews 2005, Issue 4. Art. No.: CD003648. DOI: 10.1002/14651858.CD003648.pub2.
  • 33
    HigginsJPT, GreenS, editors. Cochrane Handbook for Systematic Reviews of Interventions 4.2.6 [updated September 2006]. In: The Cochrane Library, Issue 4, 2006; Chichester, UK: John Wiley & Sons, Ltd.
  • 34
    European Mode of Delivery Collaboration. Elective caesarean-section versus vaginal delivery in prevention of vertical transmission: a randomised clinical trial. Lancet. 1999; 353: 10359.
  • 35
    CDC. U.S. Public Health Service Task Force recommendations for the use of antiretroviral drugs in pregnant women infected with HIV-1 for maternal health and for reducing perinatal HIV-1 transmission in the United States. MMWR. 1998; 47(Rr-2): 130.
  • 36
    Wilson CM, Gray G, Read JS, Mwatha A, Lala S, Johnson S, et al. Tolerance and safety of different concentrations of chlorhexidine for peripartum vaginal and infant washes: HIVNET 025. J Acquir Immune Defic Syndr. 2004; 35(2): 13843.
  • 37
    Humphrey JH, Hargrove JW, Malaba LC, Iliff PJ, Moulton LH, Mutasa K, et al. HIV incidence among post-partum women in Zimbabwe: risk factors and the effect of vitamin A supplementation. AIDS. 2006; 20(10): 143746.
  • 38
    Humphrey JH, Iliff PJ, Marinda ET, Mutasa K, Moulton LH, Chidawanyika H, et al. Effects of a single large dose of vitamin A, given during the postpartum period to HIV-positive women and their infants, on child HIV infection, HIV-free survival, and mortality. J Infect Dis. 2006; 193(6): 86071.
  • 39
    Read JS. Commentary: vitamin A supplementation for reducing the risk of mother-to-child transmission of HIV infection. Int J Epidemiol. 2006; 35: 8334; discussion 835.
  • 40
    European Collaborative Study. Maternal viral load and vertical transmission of HIV-1: an important factor but not the only one. AIDS. 1999; 13: 137785.
  • 41
    Ioannidis JPA, Abrams EJ, Ammann A, Bulterys M, Goedert JJ, Gray L, et al. Perinatal transmission of human immunodeficiency virus type 1 by pregnancy women with RNA virus loads < 1000 copies/ml. J Infect Dis. 2001; 183: 53945.
  • 42
    Shaffer N, Roongpisuthipong A, Siriwasin W, Chotpitayasunondh T, Chearskul S, Young NL, et al. Maternal viral load and perinatal HIV-1 subtype E transmission, Thailand. J Infect Dis. 1999; 179: 5909.
  • 43
    European Collaborative Study. Mother-to-child transmission of HIV infection in the era of highly active antiretroviral therapy. Clin Infect Dis. 2005; 40: 45865.