Objectives: There is limited data on the effect of antiviral therapies on clinical outcomes in HIV and hepatitis B virus (HBV)-infected individuals in sub-Saharan Africa.
Design: Single center, prospective longitudinal cohort study at Management and Development for Health supported HIV Care and Treatment clinics in Dar es Salaam, Tanzania.
Methods: Between April 2014 and December 2015, HIV-infected, HBV-infected and HIV/HBV-coinfected, treatment naïve, Tanzanian adults more than 18 years of age were eligible for enrollment and followed for 10-18 months after initiating antivirals. All HIV-infected and HIV/HBV-coinfected participants received tenofovir, lamivudine and efavirenz; HBV-infected participants received lamivudine. Multivariate regression models were constructed to identify factors associated with mortality in HIV-infected and HIV/HBV-coinfected participants.
Results: A total of 265 HIV-infected, 165 HBV-infected and 64 HIV/HBV-coinfected participants were analyzed. At baseline, HBV-infected participants were younger and had a higher BMI than HIV-infected and HIV/HBV-coinfected participants. After a median of 371 (interquartile range 50) days on treatment, there were 40 deaths. Mortality was significantly higher among HIV/HBV-coinfected participants compared with HIV and HBV-infected participants [HIV/HBV-coinfected 12 of 64 (19%) vs. HIV-infected 26 of 265 (10%) and HBV-infected two of 265 (1%), P < 0.01]. High baseline HIV RNA and low hemoglobin levels, but not HBV coinfection were independently associated with early mortality in multivariate analyses of HIV-infected participants.
Conclusion: High rates of early mortality were observed after treatment initiation in HIV/HBV-coinfected individuals compared with participants with HIV or HBV alone, despite robust aspartate aminotransferase to platelet ratio index declines and high rates of virologic suppression. HIV rather than HBV-related factors are more important contributors to mortality in these individuals.