AIDS-related tuberculosis in Rio de Janeiro, Brazil

Antonio G. Pacheco, Betina Durovni, Solange C. Cavalcante, L. M. Lauria, Richard D. Moore, Lawrence H. Moulton, Richard E. Chaisson, Jonathan E. Golub

Research output: Contribution to journalArticlepeer-review

22 Scopus citations


Background: We studied the incidence of tuberculosis, AIDS, AIDS deaths and AIDS-TB co-infection at the population level in Rio de Janeiro, Brazil where and free access to combination antiretroviral therapy has been available since 1997. Methodology/Principal Findings: This was a retrospective surveillance database match of Rio de Janeiro databases from 1995-2004. Proportions of tuberculosis occurring within 30 days and between 30 days and 1 year after AIDS diagnosis were determined. Generalized additive models fitted with cubic splines with appropriate estimating methods were used to describe rates and proportions over time. Overall, 90,806 tuberculosis cases and 16,891 AIDS cases were reported; 3,125 tuberculosis cases within 1 year of AIDS diagnosis were detected. Tuberculosis notification rates decreased after 1997 from a fitted rate (fR per 100,000) of 166.5 to 138.8 in 2004. AIDS incidence rates increased 26% between 1995 and 1998 (30.7 to 38.7) followed by a 33.3% decrease to 25.8 in 2004. AIDS mortality rates decreased dramatically after antiretroviral therapy was introduced between 1995 (27.5) and 1999 (13.4). The fitted proportion (fP) of patients with tuberculosis diagnosed within one year of AIDS decreased from 1995 (24.4%) to 1998 (15.2%), remaining stable since. Seventy-five percent of tuberculosis diagnoses after an AIDS diagnosis occurred within 30 days of AIDS diagnosis. Conclusions/Significance: Our results suggest that while combination ART should be considered an essential component of the response to the HIV and HIV/tuberculosis epidemics, it may not be sufficient alone to prevent progression from latent TB to active disease among HIV-infected populations. When tuberculosis is diagnosed prior to or at the same time as AIDS and ART has not yet been initiated, then ART is ineffective as a tuberculosis prevention strategy for these patients. Earlier HIV/AIDS diagnosis and ART initiation may reduce TB incidence in HIV/AIDS patients. More specific interventions will be required if HIV-related tuberculosis incidence is to continue to decline.

Original languageEnglish (US)
Article numbere3132
JournalPloS one
Issue number9
StatePublished - Sep 10 2008

ASJC Scopus subject areas

  • General


Dive into the research topics of 'AIDS-related tuberculosis in Rio de Janeiro, Brazil'. Together they form a unique fingerprint.

Cite this