F. Aguiar, M. A. Vieira, A. Staviack, C. Buarque, A. Marsico, L. Fonseca, R. Chaisson, A. Kristski, G. Werneck, F. Mello
S E T T I N G : A reference hospital for tuberculosis (TB) and human immunodeficiency virus/acquired immune- deficiency syndrome (HIV/AIDS) with a TB control pro- gramme in Rio de Janeiro, Brazil.
OBJECTIVE: To estimate the prevalence of resistance to anti-tuberculosis drugs and to identify associated factors.
D E S I G N : In a cross-sectional study, clinical and laboratory data were collected retrospectively from 2001 to 2005. Patients with isolation of Mycobacterium tuber- culosis and available drug susceptibility tests were considered eligible. Data on demographic characteristics, risk factors for resistance, HIV serology and past TB history were collected and analysed by χ2 Mann-Whitney test and Poisson regression.
R E S U LT S : We analysed 350 treatments, of which 62 were for patients with previous TB. HIV status was positive in 31.2% of cases. Resistance was found in 15.7% and multidrug resistance (MDR) in 4.3% of cases. Previous treatment (P < 0.001) and relapse within 2 years were associated with resistance (P < 0.03). Pulmonary cavities were associated with MDR (P < 0.001). Homelessness was associated with any resistance in newly diagnosed patients (P < 0.01). Working in a hospital was not associated with resistance.
C O N C L U S I O N : Suspicion of drug-resistant disease is necessary in patients with a history of previous TB in hospitals in Rio de Janeiro. The implementation of an effective hospital TB control programme can prevent transmission even in high TB prevalence settings.
KEY WORDS: drug resistance; tuberculosis chemotherapy; hospital infection control programme; multidrug- resistant TB; transmissible disease control