[Mycobacterium tuberculosis drug resistance in patients with HIV and pulmonary tuberculosis infections in Rome: 1987-1996].

A retrospective chart review was performed on 118 HIV infected patients with pulmonary tuberculosis hospitalized between 1987 and 1996 in a tertiary care center for Infectious Diseases in Rome. The aims of this study were: a) to evaluate global prevalence of and risk factors for drug-resistant Mycobacterium tuberculosis and multidrug resistant tuberculosis; b) to assess trends in prevalence of drug-resistant tuberculosis over the 10-year study period. Prevalence of drug resistance of first Mycobacterium tuberculosis isolates was tested on Lowenstein-Jensen medium with the proportional method. Of the 118 patients studied, 83 had never been treated for tuberculosis and 35 had already been treated for at least 1 month. The overall prevalence of resistance to one or more drugs was 25% (17% in never treated patients vs 46% in already treated patients; p = 0.002). Five percent of isolates were resistant to both isoniazid and rifampin (1% in never treated patients vs 14% in already treated patients; p = 0.008). Resistance rates to individual drugs were: isoniazid 14%, rifampin 8%, ethambutol 0%, streptomycin 13%. During the study period no significant variations in prevalence of drug-resistant tuberculosis were found. In our area, empiric therapy should include 4 drugs: as well as isoniazid, rifampin and pyrazinamide, we recommend ethambutol. Surveillance of drug-resistant tuberculosis is needed. Directly observed therapy should be considered for HIV patients in order to prevent increases in drug resistance, relapses, and treatment failures.