Tuberculosis burden on AIDS in Brazil: A study using linked databases.

OBJECTIVES:To estimate the burden of tuberculosis (TB) in reported AIDS cases, to compare the characteristics of TB/HIV subjects with those without TB and to evaluate survival with or without TB in Brazil. METHODS:The data source was the linked database between AIDS (2011-2014) and TB (2011-2014) databases from the Notifiable Diseases Information System (SINAN). The sociodemographic, clinical, laboratory results and use of antiretroviral therapy (ART) data were compared by TB occurrence or not. Survival probability was estimated using the Kaplan-Meier method and associated factors were sought using Cox regression. RESULTS:The proportion of TB diagnosed from 2011 to 2014 among AIDS cases reported between 2006 and 2014 was 6.3%. Subjects coinfected with TB were predominantly male, older, with lower schooling, with lower CD4 count, higher viral load, and higher proportion of ART initiation than those without TB. 57.5% were diagnosed with HIV before TB, 38.2% as concurrent TB/HIV and 4.3% with TB before HIV. 16,466 reported TB cases were not found in the AIDS database, although registered as HIV-infected in the SINAN TB database between 2011 and 2014. Median survival for PLHIV was 581 days, with 582 for those without TB, significantly higher than 547 for those with TB (log-rank teste, p = 0,001). In the Cox multivariate analysis, male gender [aHR = 1.27 (CI 95% 1.22-1.33)], older age [aHR = 1.020 (CI 95% 1.019-1.022)] and TB coinfection [aHR = 1.97 (CI 95% 1.88-2.07)] were positively associated with adjusted hazard of death, whereas CD4 count 200-499 cells [aHR = 0.21 (CI 95% 0.20-0.22)] and receiving ART [aHR = 0.2 2(CI 95% 0.21-0.23)] reduced the risk of death. CONCLUSIONS:HIV-infected subjects should be screened for TB at care entry, to minimize diagnosis and treatment delays when active TB is present or to increase the odds of being offered latent TB infection therapy to prevent TB. On the other hand, TB cases should be promptly tested for HIV. All those will contribute to reduce mortality among people living with AIDS.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.