Background: In Tomsk Oblast, Russian Federation, during the period of 1996-2000, most previously untreated patients with tuberculosis received standardized short-course chemotherapy, irrespective of drug-susceptibility testing results. A retrospective analysis was done to determine the effect of initial drug resistance on treatment outcome and acquired drug resistance in new patients receiving standardized short-course chemotherapy.
Methods: During the period of 1 November 1996 through 31 December 2000, a total of 2194 patients received a category 1 treatment regimen. Drug susceptibility test results for 1681 patients were available for analysis. Drug resistance patterns before and during treatment were compared for 73 patients whose culture results were persistently positive during treatment. Acquired resistance was defined as new drug resistance (during or at the end of treatment) that was not present at the beginning of treatment.
Results: Pretreatment drug resistance was strongly associated with treatment failure. In patients who had strains with pretreatment resistance patterns that included isoniazid or rifampin resistance, but not resistance to both, 17 (70.8%) of 24 cases involving treatment failures acquired new multidrug resistance. In patients with pretreatment pan-susceptible or streptomycin-monoresistant strains, 13 (41.9%) of 31 cases involving treatment failures acquired new multidrug resistance.
Conclusions: Early diagnosis of drug-resistant tuberculosis and judicious use of second-line drugs is recommended to decrease transmission of drug-resistant strains and to prevent the creation of multidrug-resistant strains. Finally, if drug susceptibility tests are not available or results are delayed, physicians should recognize that patients who do not respond to directly observed empirical short-course chemotherapy are at high risk of having multidrug-resistant tuberculosis and should be treated accordingly.