University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

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
Global Health Sciences Literature Digest
Published January 31, 2011
Journal Article

The Kesho Bora Study Group. 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 Infectious Diseases, Early Online Publication, 14 January 2011.

In Context

Breastfeeding accounts for a substantial portion of perinatally transmitted HIV infection, especially from mothers in the later stages of HIV infection. New WHO recommendations call for HIV-infected breastfeeding women, who do not meet clinical or immunologic criteria for initiating antiretroviral therapy (ART), to either give infants daily nevirapine prophylaxis (option A) or to take triple-drug ART, both until one week after exposure to breast milk has ended (option B).(1) Much of the data underlying the option B recommendation came from a poster presentation of results from the Kesho Bora study; final results from this study have now been published and are reviewed here.


To compare two different antiretroviral prophylaxis regimens in women who do not otherwise meet either clinical (WHO stage 4) or immunological (CD4 count >200 cells/µL at the time of the study) for initiating ART.


Randomized controlled trial


HIV-infected antiretroviral-naïve pregnant women with gestations ≤32 weeks were enrolled at five study sites in Burkina Faso, Kenya and South Africa. Women had WHO stage 1, 2 or 3 clinical disease and CD4 counts between 200 and 500 cells/µL. The study was conducted from June 2005 to August 2008.


Women who intended to breastfeed their infants were randomized into two arms: triple antiretroviral prophylaxis (300 mg zidovudine + 150 mg lamivudine + 400 mg lopinavir/100 mg ritonavir twice daily until cessation of breastfeeding (to a maximum of 6.5 months) or a standard regimen (300 mg zidovudine twice daily in the antenatal period, 600 mg zidovudine + 200 mg nevirapine at onset of labor and 300 mg zidovudine + 150 mg lamivudine twice daily for 7 days post partum). All infants, regardless of their mothers' assignment received 0.6 mL of nevirapine at birth and 4 mg/kg of zidovudine twice daily through 7 days of age. Prophylaxis was started between 34 and 36 weeks gestational age. The principal outcome was HIV-free survival at 12 months. The trial was analyzed using the intention to treat method.


Eight hundred twenty-four women were randomized and gave birth to 805 liveborn singleton or, in the case of multiple births, first-born infants. The 412 women randomized to the triple antiretroviral arm were similar to women randomized to the standard prophylaxis arm. Adherence was similar in both groups in the antenatal period. Mothers in the triple antiretroviral group had significantly higher CD4 counts at delivery and at 6 and 12 months and were more than twice as likely to have plasma viral load <300 copies/mL at delivery. The cumulative rate of HIV transmission at six weeks was 3.3% in the triple antiretroviral group compared to 5.0% in the standard prophylaxis group (relative risk reduction 34%) and at 12 months was 5.4% compared to 9.5% (relative risk reduction 43%, p=0.029). The proportion of infants who either became HIV-infected or died by 12 months of age was significantly lower in the triple antiretroviral arm (10.2%) compared to the standard arm (16%, relative risk reduction 36%, p=0.017). This protective effect was most pronounced in infants born to mothers with <350 CD4 cells/µL at delivery. Triple antiretroviral therapy was not associated with significant increases in severe adverse events either in mothers or infants compared to the standard regimen.


The investigators concluded that giving mothers, who were not otherwise eligible for ART, triple antiretroviral prophylaxis from 34-36 weeks gestation through the end of breast feeding resulted in 43% less transmission of HIV compared to a standard zidovudine plus nevirapine regimen in which mothers and infants did not receive antiretrovirals beyond 7 days post-partum. Moreover, triple antiretroviral therapy was not associated with higher numbers of severe adverse events than standard therapy.

Study Quality

This was an excellent trial. Using Cochrane Collaboration criteria for rating randomized controlled trials, the Kesho Bora trial has a very low risk of bias.

Programmatic Implications

Breastfeeding is an exceptionally important factor in infant and child survival but is a source of additional HIV transmission after the immediate post-partum period. As a consequence, WHO now recommends antiretroviral prophylaxis during the entire breastfeeding period for HIV-infected women who elect to breast feed. Notably, the most profound protective effects of the triple antiretroviral regimen occurred in women with <350 CD4 cells/µL, a group for whom WHO now recommends initiating ART. Estimates of benefit from this study are better than those found an earlier triple-drug antiretroviral prophylaxis study conducted in Malawi.(2) This earlier study, however, did not commence triple-drug prophylaxis until after delivery, likely failing to prevent some intrapartum transmission. Data from the Kesho Bora strongly support WHO recommendations for prevention of mother-to-child transmission during the breastfeeding period (option B).


  1. World Health Organization. Antiretroviral drugs for treating pregnant women and preventing HIV infection in infants: towards universal access. Recommendations for a public health approach (2010 version). Geneva: World Health Organization, 2010.
  2. Chasela C, Hudgens M, Jamieson D, et al. Maternal or infant antiretroviral drugs to reduce HIV-1 transmission. N Engl J Med 2010; 362: 2271-81.