HIV, antiretroviral therapy, and tuberculosis: outcomes in Johannesburg, South Africa Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 21, 2019
  • Westreich, Daniel J.
    • Affiliation: Gillings School of Global Public Health, Department of Epidemiology
  • The rollout of highly active antiretroviral therapy (HAART) in sub-Saharan Africa poses many challenges. In one of the largest HAART clinics in Africa, we described overall outcomes of HAART, the impact of treated pulmonary tuberculosis (PTB) at the time of HAART initiation on all-cause mortality, and the impact of prevalent and incident tuberculosis treatment on the risk of the substitution of the antiretroviral drug stavudine. In these analyses, robust methods including inverse probability weighted marginal structural models were used where appropriate to control for confounding and bias due to losses to follow-up and competing risks. This analytic approach made these results more robust compared with more "traditional" analyses. Relatively few patients died after HAART initiation. Risk of death during follow-up increased in individuals who at baseline were older or had low CD4 cell counts, low body mass index, low hemoglobin, or advanced disease stage. The crude hazard ratio (HR) for the effect of treated PTB at time of HAART initiation on mortality was 1.71 (95% confidence interval [CI] 1.31-2.23) and the adjusted HR was 1.06 (95% CI 0.75-1.49), a difference due to the fact that individuals with PTB at time of HAART initiation had more severe immunodepression. Additionally, individuals who started HAART within 30 days of initiation of treatment for PTB were not at higher risk of death than other individuals, suggesting that early HAART initiation can be safely recommended. In analysis of the effect of treatment for pulmonary or extrapulmonary tuberculosis (TB) on risk of stavudine substitution, we found that if TB treatment is ongoing at the time of initiation of HAART, there was between a two- and sevenfold increase in risk of stavudine substitution during the early months of HAART. However, incident TB treatment after HAART initiation, did not raise the risk of stavudine substitution. These results were robust to sensitivity analysis, and suggest that the use of stavudine might be reconsidered in patients receiving treatment for TB. Overall, these results suggest that early initiation of HAART is warranted in patients with TB, but that clinicians need to monitor these patients closely for indications for stavudine substitution.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Van Rie, Annelies
  • Open access

This work has no parents.