Objective: To analyze and compare the mutations in two different regions of the katG gene, which is responsible for isoniazid (INH) resistance. Methods: We analyzed 97 multidrug-resistant Mycobacterium tuberculosis strains isolated in cultures of sputum samples obtained from the Professor Hélio Fraga Referral Center, in Brasília, Brazil. Another 6 INH-sensitive strains did not present mutations and were included as controls. We used PCR to amplify two regions of the katG gene (GenBank accession no. U06258)-region 1, (from codon 1 to codon 119) and region 2 (from codon 267 to codon 504)-which were then sequenced in order to identify mutations. Results: Seven strains were resistant to INH and did not contain mutations in either region. Thirty strains carried mutations in region 1, which was characterized by a high number of deletions, especially at codon 4 (24 strains). Region 2 carried 83 point mutations, especially at codon 315, and there was a serine-to-threonine (AGC-to-ACC) substitution in 73 of those cases. The analysis of region 2 allowed INH resistance to be diagnosed in 81.4% of the strains. Nine strains had mutations exclusively in region 1, which allowed the proportion of INH-resistant strains identified to be increased to 90.6%. Conclusions: The number of mutations at codon 315 was high, which is consistent with cases described in Brazil and in other countries, and the analysis of region 1 resulted in a 9.2% increase in the rate at which mutations were identified.

Keywords: Isoniazid; Mutation; Tuberculosis; Drug resistance, multiple.

Objective: To analyze the clinical characteristics and evolution of a group of patients with positive sputum cultures for multidrug-resistant (MDR) Mycobacterium tuberculosis and treated at a referral center in the city of Rio de Janeiro, Brazil. Methods: Based on the positive results in sputum cultures for MDR M. tuberculosis, 50 patients were selected, and their clinical data were obtained from the Brazilian Ministry of Health MDR-TB Database. The frequencies of noncompliance, relapses, failures and previous treatments for TB up to diagnosis of MDR-TB were compiled. The radiological patterns were classified as unilateral or bilateral, and with or without cavitation. Two years after the end of the standard treatment for MDR-TB, the outcome (cure, failure, noncompliance or death) for each patient was evaluated and reassessed every two years. The post-treatment follow-up period was eight years. Results: The mean number of previous treatments was 2.3 ± 0.9. The mean interval between the initial diagnosis and the development of MDR-TB was 2.0 ± 1.7 years. Two years after the initial treatment for MDR-TB, 2 patients had abandoned treatment, 8 had died, 18 had been cured, and 22 had presented treatment failure. The bivariate analysis showed that bilateral pulmonary involvement and cavitary pattern markedly reduced the chances for cure, with a relative risk of 1-0.6 (40%) and 1-0.7 (30%), respectively. At the end of the follow-up period, 2 patients had abandoned treatment, 9 had presented treatment failure, 17 had been cured, and 22 had died. Conclusions: Bilateral pulmonary involvement and cavity pattern greatly reduced the chances for cure of the patients with MDR-TB. Most patients who presented treatment failure died within the 8-year follow-up period.

Keywords: Tuberculosis, multidrug-resistant; Treatment outcome; Disease-free survival; Treatment refusal; Treatment failure.