Low incidence of colonization and no cases of disseminated Mycobacterium avium complex infection (DMAC) in Brazilian AIDS patients in the HAART era.

OBJECTIVE Evaluate the incidence of mycobacterial disease and the colonization of the respiratory and gastrointestinal tracts by Mycobacterium avium complex (MAC) bacteria in AIDS patients. METHODS INCLUSION CRITERIA HIV-positive individuals with at least one CD(4)(+) count < 100 cells/mm(3). EXCLUSION CRITERIA Mycobacterial disease and MAC prophylaxis. Stool, sputum, and blood cultures were prospectively obtained every month from September, 1997, to December, 1999. The incidence was calculated using Poisson regression. Survival was estimated by the Kaplan Meier method and the Cox proportional hazard model. RESULTS We followed-up 79 patients during a median period of 428 days. Blood cultures (n = 742) were negative for all mycobacteria. Positive cultures (25 samples) were obtained from non-sterile sites: Stools (19/703 specimens = 2.7%) and sputum (14/742 specimens = 1.9%). MAC was isolated in 7/703 stool samples (1%) and 1/32 sputum specimens (0.1%). The incidence of patient colonization with MAC was 0.09 /year (CI=0.05-0.18). CD4 counts in patients colonized with MAC were below 100 cells/mm(3) in only 2 out of 8 cases. Restoration of CD(4)(+) counts >100 cells/mm(3) (HR = 0.18; CI = 0.05-0.70) predicted a lower risk of death (P<0.05) but was not protective for MAC colonization (HR=0.52;CI =0.62-4.35, P=0.55). CONCLUSION The absence of DMAC infection in colonized individuals argues in favor of a HAART protective effect against; DMAC; however, restoration of CD(4) counts did not protect patients against MAC colonization.