Cover Image
close this bookSummary Booklet of Best Practices in Africa - Issue 2 (UNAIDS, 2000, 116 p.)
close this folderMother-to-child transmission
View the document(introduction...)
View the documentHIVNET 012 study on effectiveness of Nevirapine in prevention of mother-to-child transmission (Uganda)

HIVNET 012 study on effectiveness of Nevirapine in prevention of mother-to-child transmission (Uganda)

Starting Year: 1998
Main Topic Area: Mother-to-child transmission
Other Topic Areas: Antiretroviral therapy · Access to drugs

Contact Person: Care and Support Team
Phone: +42 22 791-4808
Fax: +42 22 791-4741
Web Site:

20 avenue Appia
CH-1211 Geneva 27

Contact Person: Department of Reproductive Health and Research
Web Site:

World Health Organization
20 avenue Appia
CH-1211 Geneva 27


HIVNET 012 was undertaken at Mulago Hospital by researchers from Makerere University in Kampala and Johns Hopkins University in Baltimore, Maryland, USA.


The study was sponsored by the HIV Prevention Trials Network of the US National Institute of Allergy and Infectious Diseases (NIAID). Glaxo-Wellcome and Boehringer-Ingelheim Pharmaceuticals provided the drugs used in the study.


A variety of trials of short-course zidovudine (ZDV) and/or nevirapine regimens are being conducted in Thailand and Africa. One, the PETRA study, is a randomized controlled trial that has recruited HIV-positive pregnant women from five sites: two in South Africa, two in Uganda, and one in Tanzania. There are four arms to the trial, three using various regimens of ZDV and 3TC (lamivudine), the fourth using a placebo. Another study, ACTG 316, is ongoing in the United States, Europe, and Brazil. It compares a combination of zidovudine and nevirapine with zidovudine monotherapy.

Finally, a trial called SAINT (South African Intrapartum Nevirapine Trial) in South Africa aims to recruit 1,200 HIV-infected mothers and compare the HIVNET 012 nevirapine regimen with arm B of PETRA. Results should be available by June 2000. (For more on ongoing research trials into MTCT, see

Main Activities

The study was a randomized controlled trial enrolling 645 mothers (average age 24) and their infants living in and around Kampala. The dosage regimen for nevirapine was a single dose of 200mg given to the mother to take at the onset of labour, followed by 2mg per kg given to the neonate within 72 hours of birth. This was compared with a ZDV regimen of 600mg started by the mother at the onset of labour, followed by 300mg of ZDV every three hours during labour and a further course of ZDV of 4mg per kg given to the infant for seven days after birth.


The primary measures were of HIV infection and HIV survival at 6-8 weeks and 14-16 weeks. Of 609 mother-infant pairs, data were available for 529 at 14-16 weeks. At 6-8 weeks, 11.9 per cent of infants in the nevirapine group were HIV-infected, compared to 21.3 per cent of those in the ZDV group. At 14-16 weeks, the figures were 13.1 and 25.1 per cent, respectively.

Lessons Learned

A technical working group composed of experts from WHO, UNICEF, UNFPA, and the UNAIDS Secretariat reports consensus on the following points:

1. Nevirapine is cheap and simple to administer. Uganda has made relatively simple calculations of the costs of buying drug to treat all HIV-positive pregnant women in the country. For nevirapine, this would amount to $640,000 per year. With zidovudine, the cost would be $21,450,000 per year, and if the PETRA trial arm B were adopted this would cost $29,250,000 per year. With this overwhelming cost difference, it is likely that nevirapine will be widely seen as the drug of choice for this indication.

2. The low price of nevirapine must not blind decision-makers to the fact that effective MTCT interventions are not simply a matter of finding the right drug at the right price. Although drugs are a substantial element of the pilot projects, many of the additional components of the projects are independent of which drug is used. This includes the infrastructure of voluntary and confidential testing, many aspects of antenatal and postnatal care of women, psychosocial support to HIV-infected women, and the follow-up of babies to determine their HIV infection status. These elements need to be initiated and maintained regardless of what drug is recommended, now or in the future. Other elements of the infrastructure, however, such as laboratory testing (mainly haemoglobin measurement), the frequency of antenatal clinical attendance, and the monitoring of women for drug toxicity would be more extensive with short-course zidovudine than with nevirapine.

3. Voluntary counselling and testing are necessary. Although there has been discussion about using nevirapine in settings with high HIV seroprevalence without voluntary and confidential testing for HIV, there is consensus that this was not a desirable strategy. Even if nevirapine is safe for use in women who are HIV uninfected, the importance of women being aware of their HIV status offers substantial additional benefits. These include improved decision-making and the ability to plan for future health and well-being of families, and prevention of subsequent sexual transmission.

4. Considerations of risk and of drug resistance must continue to be examined. The beneficial effect of antiretrovirals given for MTCT prevention (a significant reduction in the baby's risk of infection with a fatal virus) far outweighs any risk to the infant. However, these risks have been and must continue to be researched. One serious risk - that of developing drug-resistant strains of HIV - is considered to be minimal when antiretrovirals are used for a short period of time. Research and extensive experience with ZDV since 1987 indicate that the virus takes at least three months to develop resistance to the drug when used as monotherapy. On the other hand, HIV has been shown to develop resistance to nevirapine monotherapy very quickly. Since only a single dose of nevirapine has been given to mother and baby in the study of MTCT prevention, resistance is not expected to be a problem, but no data are yet available to confirm this. If antiretroviral drugs are used in combination, resistance is even slower to develop than with monotherapy, as they offer the virus a more complex challenge. At present, however, the risk of developing drug resistance if a woman is treated with the same drug over several pregnancies is not known.

Further Reading

For more on HIVNET 012, see

For more on ongoing research trials into MTCT, see the UNAIDS website at