A retrospective survey to estimate the prevalence of hepatitis B (HBV) and C (HCV) infections was conducted on the samples of 247 African HIV-1 positive pregnant women who had participated to a mother-to-child prevention trial carried out in urban settings in Kampala, Uganda and Kigali, Rwanda. Hepatitis B markers studied were HBs antigen (HBsAg) and, if positive after confirmatory testing, HBe antigen/anti-HBe antibodies and HBV DNA. A fourth generation HCV enzyme immunoassay (EIA) was used for primary HCV screening. Positive samples were analyzed further with a second different EIA. Both for HBV and for HCV the use of confirmatory tests allowed the removal of frequent false-positive screening results. HBsAg was found in 10/246 women (seroprevalence 4.1%, 95% confidence interval (95%CI) 1.7-6.8): 8/164 (4.9%) in Uganda and 2/82 (2.4%) in Rwanda. HBe Ag was found in 33% of HBsAg-positive patients and HBV DNA was quantifiable in 71%. Anti-HCV antibodies were found in 5/247 women (seroprevalence 2.0% 95%CI 0.3-3.9): 1/165 (0.6%) in Uganda and 4/82 (4.9%) in Rwanda. There was no interrelation between HCV and HBV markers. There was no difference between patients with and without co-infection with HBV or HCV with regards to CD4+ cell count. Overall, hepatitis B and C co-infection was relatively infrequent in this group of pregnant women. However, since approximately 6% of HIV-positive women in these countries had a co-infection with one hepatitis virus, caution should be used in the monitoring of possible hepatotoxicity related to antiretroviral drugs in these populations.