WHO/HIV/2013.23
© World Health Organization 2013
Title: What ARV prophylaxis should be given to pregnant women living with HIV who do not receive lifelong ART? Contents 1. 2. 3. 4. 5. 6. PICO question ....................................................................................................................................... 1 Search strategy ...................................................................................................................................... 1 Flow diagram of screening process ....................................................................................................... 2 Evidence summaries ............................................................................................................................. 3 Bibliography of included studies .......................................................................................................... 9 Excluded studies with reasons ............................................................................................................ 10
1. PICO question What ARV prophylaxis should be given to pregnant women living with HIV who do not receive lifelong ART? Pregnant women living with HIV WHO option A from 2010 PMTCT guidelines WHO option B from 2010 PMTCT guidelines Vertical transmission, health of the mother, severe adverse effects, viral response, tolerability, TB incidence, maternal adherence, infant adherence
P I C O
2. Search strategy Search #5 #4 #3 Query Search (((#1) AND #2) AND #3) AND #4 Search “mother to child”[tiab] “mother-to-child” [tiab] OR MTCT[tiab] OR PMTCT[tiab] OR “Disease Transmission, Vertical”[MeSH] Search “triple ARV*”[tiab] OR (ARV*[tiab] AND prophylaxis[tiab]) OR (antiretroviral*[tiab] AND prophylaxis[tiab]) OR (anti-retroviral*[tiab] AND prophylaxis[tiab]) OR “triple antiretroviral” [tiab] OR ("TDF+3TC/FTC+EFV"[tiab] OR "TDF + FTC + EFV"[tiab] OR "TDF+FTC+EFV"[tiab] OR "TDF + 3TC + EFV"[tiab] OR "TDF+3TC+EFV"[tiab] OR (tenofovir[tiab] AND (lamivudine[tiab] OR emtricitabine[tiab]) AND efavirenz[tiab]) OR (TDF[tiab] AND (3TC[tiab] OR FTC[tiab]) AND EFV[tiab]) OR (TDF[tiab] AND FTC[tiab]) OR truvada[tiab] OR atripla[tiab] OR (epivir[tiab] AND emtriva[tiab] AND viread[tiab]) Search randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized controlled trials [MeSH] OR random allocation [MeSH] OR double-blind method [MeSH] OR single-blind method [MeSH] OR clinical trial [pt] OR clinical trials [MeSH] OR ("clinical trial" [tiab]) OR ((singl* [tiab] OR doubl* [tiab] OR trebl* [tiab] OR tripl* [tiab]) AND (mask* [tiab] OR blind* [tiab])) OR (placebos [MeSH] OR placebo* [tiab] OR random* [tiab] OR research design [mh:noexp] OR follow-up studies [MeSH] OR prospective studies [MeSH] OR control*[tiab] OR prospectiv* [tiab]) OR nonrandomi*[tiab] OR before after study[tiab] OR time series[tiab] OR case control[tiab] OR prospective cohort[tiab] OR cohort*[tiab] OR cross-section*[tiab] OR prospective[tiab]
#2
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 1
WHO/HIV/2013.23
© World Health Organization 2013
Search
#1
Query OR retrospective[tiab] OR research design[mh:noexp] OR follow-up studies[MeSH] OR prospective studies[MeSH] OR control*[tiab] OR prospectiv*[tiab]) NOT (animals [MeSH] NOT human [MeSH]) Search "HIV Infections"[MeSH] OR HIV[MeSH] OR HIV[tiab] OR hiv-1*[tiab] OR hiv2*[tiab] OR hiv1[tiab] OR hiv2[tiab] OR HIV infect*[tiab] OR human immunodeficiency virus[tiab] OR human immunedeficiency virus[tiab] OR human immuno-deficiency virus[tiab] OR human immunedeficiency virus[tiab] OR ((human immun*) AND (deficiency virus[tiab])) OR acquired immunodeficiency syndrome[tiab] OR acquired immunedeficiency syndrome[tiab] OR acquired immuno-deficiency syndrome[tiab] OR acquired immune-deficiency syndrome[tiab] OR ((acquired immun*) AND (deficiency syndrome[tiab]))
3. Flow diagram of screening process Results from all search results (n=308) Duplicates removed (n=116)
Initial screening by one author (n=192)
Clearly irrelevant articles removed (n=62)
Records screened by two authors working independently (n=130)
References excluded based on details in abstracts (n=110)
Full-text articles assessed for eligibility (n=20)
Full-text articles excluded (n=14)
Studies included (n=6) 2 RCTs reported in five manuscripts and one prospective observational trial
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 2
WHO/HIV/2013.23
© World Health Organization 2013
4. Evidence summaries There were no RCTs directly comparing PMTCT option A (AZT to mother, infant NVP to baby during breastfeeding) and option B regimens from 2010 (AZT/3TC/LPV/r, AZT/3TC/ABC, AZT/3TC/EFV). Two randomized controlled trials reported in five manuscripts (Jamieson 2012, Fao 2012) compared maternal triple ARV prophylaxis and dual ARV prophylaxis with zidovudine (AZT) and nevirapine (NVP). The Breastfeeding, Antiretrovirals and Nutrition (BAN) trial compared postnatal maternal triple antiretroviral medicines (ARV) and extended infant NVP with a control arm receiving single-dose (SD) NVP and one week of AZT/3TC in mothers and infants (Jamieson 2012, Chasela 2010). The Kesho Bora trial (Fao 2012, de Vincenzi 2011) compared AZT, lamivudine and lopinavir/ritonavir versus antepartum AZT and SD-NVP. The Kesho Bora study also included a subset analysis of maternal adherence (Mepham 2011). In addition, one prospective observational study examined HIV transmission and infant mortality outcomes with triple ARVs versus AZT prophylaxis among pregnant women living with HIV (Dryden-Peterson 2011). In the Kesho Bora randomized trial, pregnant women living with HIV with WHO stage 1, 2, or 3 HIV infection who had CD4 counts of 200–500 cells/mm3 were enrolled and initiated on ARVs at 28–36 weeks of gestation. Women were randomized to either triple ARV prophylaxis (zidovudine, lamivudine and lopinavir/ritonavir until breastfeeding ended to a maximum of 6.5 months postpartum) or AZT with SD-NVP and, after a protocol amendment, one week of AZT and lamivudine following delivery. All infants received NVP at birth and, after a protocol amendment, AZT for one week after birth. The primary endpoints were HIV-free infant survival at 6 weeks and 12 months; HIV-free survival at 12 months in infants who were ever breastfed; AIDS-free survival in mothers at 18 months; and serious adverse events in mothers and infants. Mepham et al. evaluated maternal adherence in a subpopulation in the Kesho Bora trial. Pill counts were performed, and the reasons for poor adherence were addressed. The rate of good adherence, as defined by adherence >95% over the entire prophylaxis period, was 60% in the triple ARV group and 61% in the AZT/NVP group (P = 0.893). The BAN trial enrolled women living with HIV with CD4 counts ≥200 cells/mm3 (≥250 cells/mm3 following a protocol amendment) into three arms following delivery. All women and infants received SDNVP with one week of AZT and lamivudine regardless of the randomization arm. At delivery, maternalinfant pairs were randomized to one of three 28-week regimens: maternal triple ARVs (AZT, lamivudine and NVP – NVP was replaced with nelfinavir and then lopinavir/ritonavir after protocol amendments – an infant NVP arm and a control arm. The interventions were continued through 28 weeks postpartum, and the women were counselled to wean between 24 and 28 weeks following delivery. The primary endpoint was HIV infection at 48 weeks among infants that were not infected at 2 weeks. Dryden-Peterson et al. conducted a prospective cohort study of women living with HIV and their babies in Botswana from 2009 to 2010. Women with CD4 counts ≤250 cells/mm3 or WHO clinical stage 3 or 4 were eligible for HAART (AZT, lamivudine and NVP). Women with CD4 cell counts >250 cells/mm3 were eligible for AZT (300 mg twice daily) from 28 weeks gestation through delivery. For women This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 3
WHO/HIV/2013.23
© World Health Organization 2013
receiving less than four weeks of AZT before delivery, SD-NVP was recommended during labour. All infants were to receive SD-NVP at birth and four weeks of AZT. Infants in both groups were evaluated for HIV infection and HIV-free survival. Of the two randomized controlled trials, neither compared option A or option B as described in the 2010 WHO guidelines, but the trials included in this analysis provided the closest approximation of the two PMTCT regimens. OUTCOME: VERTICAL TRANSMISSION RATE • In the Kesho Bora trial, the cumulative rate of HIV transmission at six weeks was 3.3% (95% CI: 1.9–5.6%) in the triple ARV group compared with 5.0% (3.3–7.7%) in the AZT/SD-NVP group. At 12 months, these rates were 5.4% (95% CI: 3.6–8.1%) in the triple ARV group compared with 9.5% (95% CI: 7.0–12.9%) in the AZT/SD-NVP group (P = 0.03). In the BAN trial, the cumulative risk of HIV-1 transmission by 48 weeks was significantly higher in the control group (7%, 95% CI 5–9%) than in the maternal-antiretroviral (4%, 95% CI 3–6%; P = 0.03) or the infant-nevirapine (4%, 95% CI 2–5; P = 0.003) groups. One third of HIV transmission occurred between weeks 29 and 48 following the intervention phase. One prospective cohort study evaluated infants born to women living with HIV in Botswana who used either triple ARV if their CD4 count was <250 cells/mm3 or AZT/NVP if their CD4 was >250 cells/mm3 during pregnancy. The transmission rates were 5.5% (95% CI: 2.6–10.2%) for maternal AZT prophylaxis and 0.4% (95% CI: 0.0-2.2%) for maternal triple ARV.
•
•
Findings Analysis of two randomized controlled trials did not reveal a significant difference in vertical transmission between triple ARV prophylaxis and AZT prophylaxis with extended NVP. The quality of evidence was low primarily due to indirectness and imprecision. The Kesho Bora trial compared extended maternal triple ARV prophylaxis through 6.5 months of breastfeeding with prenatal AZT/SD-NVP without a postpartum component. No significant difference was observed at six weeks. The difference observed at 12 months likely reflects the need for ongoing prophylaxis during breastfeeding rather than the superiority of maternal triple ARV prophylaxis. In addition, women with CD4 counts <350 cells/mm3 were included in the analysis. A subgroup analysis revealed no difference in vertical transmission at 12 months in women with CD4 counts >350 cells/mm3. The BAN trial was not powered adequately to compare extended infant NVP with maternal triple ARV therapy through six months after delivery. There was no prenatal component beyond SD-NVP with a AZT/3TC tail. Although consistent with real-world scenarios in which pregnant women are diagnosed HIV positive at delivery, the lack of a prenatal component limits the ability of BAN to be applied directly to the question of whether option A or option B is superior in preventing vertical transmission. In addition, women with CD4 counts of <350 cells/mm3 were included in all arms of the BAN trial. The quality of evidence was also downgraded due to imprecision because of a relatively low number of events per trial.
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 4
WHO/HIV/2013.23
© World Health Organization 2013
Analysis of one prospective trial revealed a significant decrease in HIV transmission among women receiving triple ARVs as compared to AZT prophylaxis. The quality of evidence was very low and downgraded due to indirectness and imprecision. Women with CD4 counts between 250 and 350 cells/mm3 were treated with prophylactic AZT. This may have enhanced the significant difference in HIV transmission. However, a subset analysis in women with CD4 counts greater than 350 cells/mm3 was performed without loss of effect. This population was provided with replacement feeding following birth, which also contributed to indirectness, as this option is not available in many areas. OUTCOME: INFANT MORTALITY • In the Kesho Bora trial, the rate of HIV infection or death at 12 months following delivery was 10.2% (95% CI 7.6–13.5%) in the triple ARV arm compared with 16.0% (95% CI 12.7–20.0%) in the AZT/NVP arm (P = 0.02). In the BAN trial, the incidence of HIV infection or death by 48 weeks was 15% (95% CI 12–18%) in the control group versus 12% (95% CI 10–15%) in the maternal-antiretroviral group (P = 0.13) and 11% (95% CI 8–13%) in the infant-nevirapine group (P = 0.006). There was no difference in mortality among HIV-uninfected infants. One prospective cohort study found that rates of HIV-free survival were 95.7% in the HAART arm and 90.4% in the AZT/NVP arm at 6 months (P = 0.04).
•
•
Findings Analysis of two randomized controlled trials did not reveal a significant difference in infant mortality. The quality of evidence was downgraded to very low, primarily due to indirectness and imprecision. Specifically, the lack of postpartum prophylaxis in the control group of the Kesho Bora study does not equate fairly with option A and may have affected mortality. In addition, women with CD4 counts of <350 cells/mm3 were included in the AZT/NVP group. This may have affected infant mortality if maternal health deteriorated while not receiving HAART. The BAN trial did not demonstrate significant differences in infant death between the maternal triple ARV and infant NVP arms. Analysis of one prospective trial did not reveal a significant difference in infant mortality. The quality of evidence resulting from analysis of the prospective trial was also limited by indirectness and imprecision for reasons similar to those described in the HIV transmission outcome. OUTCOME: MATERNAL MORTALITY: • In the Kesho Bora trial, 2 women died in the triple ARV arm as compared to 5 in the AZT/SDNVP arm. In the BAN trial, the risk of death between the maternal ARV (1/849) and infant NVP (2/852) arms was not compared directly.
•
Findings This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 5
WHO/HIV/2013.23
© World Health Organization 2013
Analysis of the two randomized controlled trials did not reveal a significant effect on maternal mortality due to prophylaxis with maternal triple ARVs as compared to AZT/NVP. The evidence quality was downgraded to very low due to both indirectness and imprecision. Women in both studies with CD4 counts of <350 cells/mm3 were included in the AZT/NVP group, leading to an indirect comparison with option A and enhancing the risk for increased mortality in this group. Women in the BAN trial did not receive prenatal prophylaxis, which does not allow for an evaluation of the impact of the prenatal component of option A on maternal mortality. Women in the BAN trial also received nutritional supplementation, which is not available in all locations and may have affected maternal mortality. OUTCOME: MATERNAL MORBIDITY AND INFANT MORBIDITY AND CLINICAL SERIOUS ADVERSE EVENTS • In the Kesho Bora trial, maternal HIV disease progression up to 24 months following delivery in the triple ARV arm was significantly lower than in the AZT/SD-NVP arm (15.7% versus 28.3%; P = 0.001). However, the risks of progression after cessation of ARV prophylaxis were not different (15.0% versus 13.8% by 18 months after ARV cessation). Relatively few women in either arm (<5%) with initial CD4 counts of ≥350 cells/mm3 progressed. Women in the triple ARV arm and AZT arm did not experience a significant difference in clinical serious adverse effects (Triple ARV 14% versus AZT/SD-NVP 12%, P = 0.35). There were no statistically significant differences between the two groups in the proportions of infants with common laboratory serious adverse effects (anaemia and neutropaenia) or clinical serious adverse events. In the BAN trial, serious adverse effects among mothers occurred among 48/849 women in the triple ARV arm and 32/852 women in the infant NVP arm. Combined serious adverse effects at 48 weeks in mothers and infants were 230/849 (27%) in the maternal ARV arm and 269/852 (32%) in the infant NVP arm. The rate of serious adverse effects in infants from 29 weeks to 48 weeks (1.1 per 100 person-weeks, 95% CI 1.0–1.2) was significantly higher than in the intervention phase (0.7 per 100 person-weeks, 95% CI 0.7–0.8; P < 0.0001).
•
Findings Analysis of two randomized controlled trials revealed a significant difference in maternal morbidity. Clinical serious adverse effects in mothers receiving prophylaxis with maternal triple ARVs were increased as compared to those receiving AZT/NVP. The evidence quality was assessed as low due to both indirectness and imprecision. Many of the serious adverse effects described in the BAN trial may have been attributable to the maternal triple ARV regimen, which included AZT, 3TC accompanied by NVP, nelfinavir and lopinavir/ritonavir at varying time points. Only lopinavir/ritonavir is currently included among the recommended option B regimens. The specific nature of serious adverse effects in the Kesho Bora cohort was not provided. Clinical disease progression was not evaluated in the GRADE analysis, but the Kesho Bora trial demonstrated an increased risk of clinical or immunological progression among women receiving This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 6
WHO/HIV/2013.23
© World Health Organization 2013
AZT/NVP up to 24 months from delivery. However, no effect was seen in a subset of women with CD4 counts above 350 cells/mm3 at enrolment or from the time of cessation of ARVs. Analysis of two randomized controlled trials did not reveal a significant difference in infant morbidity between groups as measured by serious adverse effects. The evidence quality was moderate due to indirectness. OUTCOME: VIRAL RESPONSE • In the Kesho Bora trial, the median viral load decreased by 4.2 logs in the triple ARV arm and 1.1 logs in the zidovudine/NVP arm from enrolment to delivery. At enrolment, 4.7% of the women in the triple ARV arm and 3.5% in the zidovudine/NVP arm had undetectable viral loads. At delivery, this increased to 63.3% in the triple ARV arm and 29.9% in the zidovudine/NVP arm (P < 0.0001). This difference had resolved by 18 months postpartum, when 5.4% of the triple ARV arm had undetectable viral loads as compared to 4.1% of the zidovudine/NVP arm. The median viral load was equal between arms 18 months following delivery.
Findings No GRADE analysis was performed for maternal viral response in maternal triple ARV versus AZT. The data are limited to a single trial and not amenable to GRADE analysis. OUTCOME: CD4 RESPONSE AT THE END OF FOLLOW-UP • In the Kesho Bora trial, the median CD4 count at enrolment was 334 cells/mm3 in the triple ARV arm and 339 cells/mm3 in the AZT/NVP arm (P = 0.55). At delivery, the median CD4 in the triple ARV arm was 465 cells/mm3 as compared to 415 cells/mm3 in the AZT/NVP arm (P < 0.0001). The median CD4 counts in both arms decreased following cessation of ARVs to 396 cells/mm3 in the triple ARV arm and 331 cells/mm3 in the AZT/NVP arm (P = 0.04). In the BAN trial, there was no significant difference in maternal CD4 counts 28 weeks following delivery (P = 0.33). The maternal ARV arm had a median CD4 count of 439 cells/mm3 (95% CI 331–591) and the women in the infant NVP arm had a median CD4 count of 451 cells/mm3 (95% CI 343–584).
•
Findings Analysis of the two randomized controlled trials did not reveal a significant effect on maternal CD4 count due to prophylaxis with maternal triple ARVs as compared to AZT/NVP at the end of follow-up. The quality of evidence was moderate and only downgraded for indirectness. In the Kesho Bora trial, a significant difference remained at 24 months following delivery favouring the maternal ARV group. Importantly, a rapid decrease in the CD4 counts of women treated with triple ARVs was not observed following discontinuation. There was no significant difference in the duration of prenatal ARVs used in the Kesho Bora trial; however, neither arm delivered prophylaxis in the second trimester, as is the current recommendation. Women in the infant NVP arm of the BAN trial did not have This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 7
WHO/HIV/2013.23
© World Health Organization 2013
significantly lower CD4 counts at the conclusion of therapy than women in the maternal triple ARV arm despite the absence of any prenatal AZT. OUTCOME: MATERNAL ADHERENCE AND INFANT ADHERENCE • In the Kesho Bora trial, 73% of women in both arms reported no missed doses prior to delivery. In the AZT/NVP arm, 87% of women received SD-NVP at delivery. Following delivery, 64% of women reported missing none or one dose of triple ARVs. Adherence was evaluated in more detail at a South African study site. Adherence was measured by pill counts, and there was no significant difference between the triple ARV and zidovudine/NVP arms (94.3% versus 94.2%, P = 0.41). In the Kesho Bora trial, infants received SD-NVP at birth followed by 1 week of AZT. 97% of infants in both the triple ARV and AZT/NVP arm received SD-NVP. 98% of infants in the triple ARV arm received 1 week of AZT, as did 95% of infants in the AZT/NVP group (P = 0.33). The BAN trial measured maternal adherence based on adherence reports from five follow-up visits. Mothers reported taking all of their antiretroviral doses a mean of 89% (SD 20%) of the time. In the BAN trial, mothers reported giving all infant antiretroviral doses 94% (SD 14%) of the time.
•
Findings No GRADE analysis was performed because the data were not appropriate. OUTCOME: MATERNAL AND INFANT TOLERABILITY • In the Kesho Bora trial, 0.8% of the women in the triple ARV had to discontinue or change their regimen due to WHO grade 3 or 4 anaemia. No women in the AZT/NVP arm had to discontinue the regimen, but the rates of anaemia in the groups were similar (3% versus 2%, P = 0.47). In the BAN trial, rash or NVP hypersensitivity occurred in 7/849 mothers and 3/849 infants in the maternal ARV group. In the infant NVP arm, 19/852 infants experienced NVP hypersensitivity or rash.
•
Findings Analysis of the one randomized controlled trial did not reveal a significant effect on maternal tolerability among women receiving maternal triple ARV as compared to AZT/NVP. The evidence quality was downgraded to very low due to both indirectness and imprecision. Women in both arms of the Kesho Bora trial received either triple ARV or AZT for an average of six weeks before delivery. Maternal prophylaxis beginning at 14 weeks gestation, as the WHO guidelines currently recommend, may result in decreased tolerability to both triple ARV and AZT regimens. The BAN trial was not included in the analysis because women in the infant NVP arm received only SD-NVP with a one-week tail of AZT and lamivudine, which was not thought to be an appropriate comparison group. This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 8
WHO/HIV/2013.23
© World Health Organization 2013
Analysis of one randomized controlled trial revealed a significant decrease in tolerability of infant NVP as compared to infant exposure to maternal triple ARVs through breastfeeding. The quality of evidence was downgraded to very low primarily due to indirectness and imprecision. A significantly increased number of infants developed NVP hypersensitivity and rash in the infant NVP arm of the BAN trial, but this did not lead to any infant deaths and resolved with a change of medication. Infant NVP and maternal ARVs were discontinued after six months instead of being continued for the duration of breastfeeding, potentially underestimating the issues with tolerability. The decreased tolerability in the infant NVP should be interpreted with caution, as the benefits of the intervention outweigh the risk in the appropriate clinical scenario. OUTCOME: TB INCIDENCE • In the Kesho Bora trial, 19 women were diagnosed with TB, without designation of the number of women in each group. In the BAN trial, maternal TB occurred in 4/849 in the maternal ARV arm and 4/852 in the infant NVP arm. TB in infants occurred in 2/849 in the maternal ARV arm and 1/852 in the infant NVP arm.
•
Findings Analysis of one randomized controlled trial did not reveal a significant difference in maternal or infant TB incidence due to prophylaxis with maternal triple ARVs as compared to AZT/NVP. This outcome was downgraded to very low due to indirectness and imprecision due to the low event numbers. OUTCOME: STUDY RETENTION • The Kesho Bora trial had 93% retention of infants in the triple antiretroviral group and 91% in the AZT and SD-NVP group (P = 0.25) through 52 weeks. 63% (245/384) of the women in the maternal triple ARV arm were retained through 18 months following delivery as compared to 84% (343/405) in the AZT/NVP arm. In the BAN trial, 79% (676/849) of maternal-infant pairs were retained in the triple ARV arm as compared to 79% (680/852) of maternal-infant pairs in the infant NVP arm.
•
Findings Analysis of two randomized controlled trials did not reveal a significant difference in study retention between maternal-infant pairs given triple ARVs as compared to AZT or extended infant NVP. The quality of evidence was moderate and downgraded only for indirectness.
5. Bibliography of included studies Randomized controlled trials
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 9
WHO/HIV/2013.23
© World Health Organization 2013
1. Jamieson DJ, Chasela CS, Hudgens MG, et al. Maternal and infant antiretroviral regimens to prevent postnatal HIV-1 transmission: 48-week follow-up of the BAN randomized clinical trial. Lancet 2012; 379: 2449-58. • Chasela CS, Hudgens MG, Jamieson DJ, et al. Maternal or infant antiretroviral drugs to reduce HIV-1 transmission. N Engl J Med 2010; 362: 2271-2281 (not graded, since it reports earlier outcomes). 2. Fao P, Ky-Zerbo O, Gouem C, Somda P, Hien H, Ouedraogo PE, Kania D, Sanou A, Kossiwavi IA, Sanogo B, Ouedraogo M, Siribie I, Valea D, Ouedraogo S, Some R, Rouet F, Rollins N, McFetridge L, Naidu K, Luchters S, Reyners M, Irungu E, Katingima C, Mwaura M, Ouattara G, Mandaliya K, Wambua S, Thiongo M, Nduati R, Kose J, Njagi E, Mwaura P, Newell ML, Mepham S, Viljoen J, Bland R, Mthethwa L, Bazin B, Rekacewicz C, Taylor A, Flowers N, Thigpen M, Fowler MG, Jamieson D, Mofenson LM, Read JS, Bork K, Cames C, Cournil A, Claeys P, Temmerman M, Van de Perre P, Becquart P, Foulongne V, Segondy M, de Vincenzi I, Gaillard P, Farley T, Habib N, Landoulsi S on behalf of the Kesho Bora Study Group. Maternal HIV-1 disease progression 18–24 months postdelivery according to antiretroviral prophylaxis regimen (triple-antiretroviral prophylaxis during pregnancy and breastfeeding vs zidovudine/single-dose nevirapine prophylaxis): the Kesho Bora randomized controlled trial. Clin Infect Dis 2012; 55: 449-460. • de Vincenzi I. Triple antiretroviral compared with zidovudine and single-dose nevirapine prophylaxis during pregnancy and breastfeeding for prevention of mother-to-child transmission of HIV-1 (Kesho Bora study): a randomised controlled trial. Lancet Infect Dis 2011; 11: 171-180. • Mepham S, Zondi Z, Mbuyazi A, Mkhwanazi N, Newell ML. Challenges in PMTCT antiretroviral adherence in northern KwaZulu-Natal, South Africa. AIDS Care 2011; 23: 741-747 (not graded since it examined adherence in a study subgroup). Observational studies 1. Dryden-Peterson S, Jayeoba O, Hughes MD, Jibril H, Keapoletswe K, Tlale J, Modise TA, Asmelash A, Moyo S, van Widenfelt E, Makhema J, Essex M, Shapiro RL, Lockman S. Highly active antiretroviral therapy versus zidovudine for prevention of mother-to-child transmission in a programmatic setting, Botswana. J Acquir Immune Defic Syndr 2011; 58(3):353-7.
6. Excluded studies with reasons Reference Ezechi OC, David AN, Gab-Okafor CV, Ohwodo H, Oladele DA, Kalejaiye OO, Ezeobi PM, Gbajabiamila TA, Adu RA, Oke B, Musa ZA, Ekama SO, Ilesanmi O, Odubela O, Somefun EO, Herbertson EC, Onwujekwe DI, Ujah IA. Incidence of and socio-biologic risk factors for spontaneous preterm birth in HIV positive Nigerian women. BMC Pregnancy Childbirth. 2012 12:93 Reason for Exclusion This study had a retrospective design. There was no information on subsets of women with CD4 > 350 cells/mm3 in the analysis and no comparison of Option A versus Option B in women with CD4 >350 cells/mm3.
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 10
WHO/HIV/2013.23
© World Health Organization 2013
Reference Coovadia HM, Brown ER, Fowler MG, Chipato T, Moodley D, Manji K, Musoke P, Stranix-Chibanda L, Chetty V, Fawzi W, Nakabiito C, Msweli L, Kisenge R, Guay L, Mwatha A, Lynn DJ, Eshleman SH, Richardson P, George K, Andrew P, Mofenson LM, Zwerski S, Maldonado Y. Efficacy and safety of an extended nevirapine regimen in infant children of breastfeeding mothers with HIV1 infection for prevention of postnatal HIV-1 transmission (HPTN 046): a randomised, doubleblind, placebo-controlled trial. Lancet 2012; 379:9812, 221-228 Hussain A, Moodley D, Naidoo S, Esterhuizen TM. Pregnant women's access to PMTCT and ART services in South Africa and implications for universal antiretroviral treatment. PLoS One 2011; 6:12, e27907
Reason for Exclusion This was a randomized controlled trial. There was no comparison between PMTCT options A and B.
Nlend AEN, Ekobo CS, Junior MB, Ekani BB, Tchokoteu P, Lyeb S, Chewa G, Moyo SNT, Takam PT. Early outcomes of HIV exposed children in the first district-wide programme using extended regimens for the prevention of mother-tochild transmission of HIV, in Yaounde, Cameroon. J Trop Pediatr 2012; 58: 297-302 Briand N, Mandelbrot L, Blanche S, Tubiana R, Faye A, Dollfus C, Le Chenadec J, Benhammou V, Rouzioux C, Warszawski J. Previous antiretroviral therapy for prevention of mother-to-child transmission of HIV does not hamper the initial response to PI-based multitherapy during subsequent pregnancy. J Acquir Immune Defic Syndr 2011; 57: 126-135 Mirkuzie AH, Hinderaker SG, Sisay MM, Moland KM, Mørkve O. Current status of medication adherence and infant follow up in the prevention of mother to child HIV transmission programme in Addis Ababa: a cohort study. J Int AIDS Soc 2011; 14:50-54 Namukwaya Z, Mudiope P, Kekitiinwa A, Musoke P, Matovu J, Kayma S, Salmond W, Bitarakwate E, Mubiru M, Maganda A, Galla M, Byamugisha J, Fowler MG. The impact of maternal highly active antiretroviral therapy and short-course combination antiretrovirals for prevention of mother-to-child
This study was excluded due to the difference in CD4 count required for HAART as compared to current recommendations. Women were given AZT from 28 weeks and SD-NVP if CD4 count >200 cells/mm3 versus HAART if CD4 count <200 cells/mm3. The study reported outcomes on prematurity: however, this is difficult to interpret given the low CD4 counts. In addition, a large percentage of participants qualified for HAART but did not start. This group had worse outcomes. The criteria for HAART included VL measurements in women with CD4 >350 cells/mm3. Dual therapy was only from 28 weeks, and infant prophylaxis was one week AZT plus SD-NVP.
The primary endpoint was VL and not transmission. All women in the study received PIbased HAART. Most women with prior exposure received triple ARVs. It did not address the outcomes of interest.
This study compares adherence of HAART for therapy with CD4 counts <200 cells/mm3 and AZT for prophylaxis for CD4 counts >200 cells/mm3. There was no direct comparison of HAART for prophylaxis with dual therapy prophylaxis. AZT/SD-NVP therapy was only given from 28 weeks and was compared with AZT/3TC from 33 weeks + SD-NVP in women with CD4 counts >350 cells/mm3. Triple ART was only given to women with CD4 counts <350 cells/mm3. The design was retrospective. There was no direct comparison of
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 11
WHO/HIV/2013.23
© World Health Organization 2013
Reference transmission on early infant infection rates at the Mulago national referral hospital in Kampala, Uganda, January 2007 to May 2009. J Acquir Immune Defic Syndr 2011; 56: 69-75 Thomas TK, Masaba R, Borkowf CB, Ndivo R, Zeh C, Misore A, Otieno J, Jamieson D, Thigpen MC, Bulterys M, Slutsker L, De Cock KM, Amornkul PN, Greenberg AE, Fowler MG. Tripleantiretroviral prophylaxis to prevent mother-tochild HIV transmission through breastfeeding--the Kisumu Breastfeeding Study, Kenya: a clinical trial. PLoS Med 2011; 8: e1001015 Marazzi MC, Liotta G, Nielsen-Saines K, Haswell J, Magid NA, Buonomo E, Scarcella P, Doro Altan AM, Mancinelli S, Palombi L. Extended antenatal antiretroviral use correlates with improved infant outcomes throughout the first year of life. AIDS 2010; 24: 2819-2826 Marazzi MC, Palombi L, Nielsen-Saines K, Haswell J, Zimba I, Magid NA, Buonomo E, Scarcella P, Ceffa S, Paturzo G, Narciso P, Liotta G. Extended antenatal use of triple antiretroviral therapy for prevention of mother-to-child transmission of HIV-1 correlates with favorable pregnancy outcomes. AIDS 2011; 25: 1611-1618
Reason for Exclusion option A with option B.
This study featured a 34- to 36-week initiation of AZT/3TC and NVP or nelfinavir (neither of which is included in the recommended option B regimens). Most importantly, there was no comparison group.
Kilewo C, Karlsson K, Ngarina M, Massawe A, Lyamuya E, Swai A, Lipyoga R, Mhalu F, Biberfeld G. Prevention of mother-to-child transmission of HIV-1 through breastfeeding by treating mothers with triple antiretroviral therapy in Dar es Salaam, Tanzania: the Mitra Plus study. J Acquir Immune Defic Syndr 2009; 52: 406-416 de Vincenzi I, Gaillard P, Farley T, Habib N, Landoulsi S. Eighteen-month follow-up of HIV-1infected mothers and their children enrolled in the Kesho Bora study observational cohorts. J Acquir Immune Defic Syndr 2010; 54: 533-541 Geddes R, Giddy J, Butler LM, Van Wyk E, Crankshaw T, Esterhuizen TM, Knight S. Dual and triple therapy to prevent mother-to-child transmission of HIV in a resource-limited setting lessons from a South African programme. S Afr Med J 2011; 101: 651-654 Lahoz R, Noguera A, Rovira N, Catala A, Sanchez E, Jimenez R, Fortuny C. Antiretroviral-related hematologic short-term toxicity in healthy infants implications of the new neonatal 4-week
This was a retrospective cohort trial. Women with CD4 <350 cells/mm3 received HAART for treatment and those with >350 cells/mm3 received HAART for prophylaxis through six months following delivery. There was no comparison group for prophylaxis. This was a retrospective cohort. Women with CD4 counts of <350 cells/mm3 received HAART for treatment and >350 cells/mm3 received HAART for prophylaxis through six months following delivery. Outcome was maternal mortality, prematurity and stillbirth. All were found to be significantly decreased in women receiving HAART for >30 days regardless of CD4 count. There was no nonHAART comparison group. This was a non-randomized cohort trial with HAART from 34 weeks through 6 months postpartum. They compare the cohort with historical regimens from the Petra and Mitra trials. The control regimens are not currently in use and not consistent with option A. This study was excluded because the regimens and the cohorts included do not match the comparisons of interest.
Multiple ARV regimens were used with varying criteria over the course of the study period. None match the current guidelines in terms of entry criteria. Infant prophylaxis only extended through one week, even among the infants of mothers receiving dual therapy. This trial was not powered to address the question of comparing four weeks to six weeks of AZT in terms of efficacy on reducing transmission. It was deemed too narrow to be relevant in terms of the
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 12
WHO/HIV/2013.23
© World Health Organization 2013
Reference zidovudine regimen. Pediatr Infect Dis J 2010; 29: 376-379
Reason for Exclusion question of haematological toxicity.
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 13