Background:  As a result of shared transmission routes, the hepatitis B virus (HBV) and hepatitis C virus (HCV) epidemics overlap with that of HIV. Understanding the epidemiology of HBV, HCV, and HIV infection in pregnant women is important, not only because of vertical transmission risk, but also to inform treatment decisions.

Methods:  In a sub-study nested in the European Collaborative Study, a cohort study, prevalence of HCV co-infection (HCV seropositive) and hepatitis B surface antigen (HBsAg) were estimated and risk factors investigated among HIV-infected pregnant women from 7 countries. Women with HCV co-infection or HBsAg carriage were compared with HIV-mono-infected women regarding immunological/virological characteristics and ART use. Information on HBsAg positivity and anti-HCV was collected retrospectively from antenatal records of women enrolled January 1999 to October 2005. These data were linked to prospectively collected data in the ECS database. Analyses included linear mixed effects and logistic regression models.

Results:  Of 1050 women (530 in Western Europe, 520 in Ukraine), 4.9% (95%CI 3.6 to 6.3) were HBsAg positive and 12.3% (95%CI 10.4 to 14.4) HCV-co-infected. Women who had ever injected drugs (IDU) had the highest HCV prevalence (28% [95%CI 22.8 to 35.7]). Risk factors for HCV co-infection included IDU history (AOR 2.92 [1.86 to 4.58]), older age (AOR 3.45 [1.66 to 7.20] for ≥35 years vs <25 years) and HBsAg carriage (AOR 5.80 [2.78 to 12.1]). HBsAg positivity was associated with African origin (AOR 2.74 [1.20 to 6.26]), and HCV infection (AOR 6.44 [3.08 to 13.5]). Antenatal HAART use was less likely in HIV/HCV-co-infected than in HIV-only-infected women, adjusting for timing of maternal HIV diagnosis and CD4 count (AOR 0.34 [95%CI 0.20 to 0.58]; p <0.0001). HIV/HCV-co-infected women had antenatal HIV RNA levels 0.36 log₁₀ higher [95%CI 0.12 to 0.60] than HIV-only-infected women, adjusting for baseline CD4 count, ART use and time of HIV RNA measurement (p = 0.004); they were also twice as likely to have detectable HIV RNA in the third trimester or at delivery than HIV-only infected women (AOR 1.95; p = 0.049).

Conclusions:  In 8 HIV-infected pregnant women, 1 was HCV co-infected and 1 in 20 was HsBAg positive. Although HCV co-infection affected use of HAART in pregnancy, the significant association found between HCV co-infection and uncontrolled HIV viremia in late pregnancy or at delivery was independent of HAART use. Research to guide evidence-based management of HIV/HCV-co-infected women is urgently needed.