Risk of acquired drug resistance during short-course directly observed treatment of tuberculosis in an area with high levels of drug resistance.

BACKGROUND

Data on the performance of standardized short-course directly observed treatment (DOTS) of tuberculosis (TB) in areas with high levels of drug resistance and on the potential impact of DOTS on amplification of resistance are limited. Therefore, we analyzed treatment results from a cross-sectional sample of patients with TB enrolled in a DOTS program in an area with high levels of drug resistance in Uzbekistan and Turkmenistan in Central Asia.

METHODS

Sputum samples for testing for susceptibility to 5 first-line drugs and for molecular typing were obtained from patients starting treatment in 8 districts. Patients with sputum smear results positive for TB at the end of the intensive phase of treatment and/or at 2 months into the continuation phase were tested again. RESULTS. Among 382 patients with diagnoses of TB, 62 did not respond well to treatment and were found to be infected with an identical Mycobacterium tuberculosis strain when tested again; 19 of these patients had strains that developed new or additional drug resistance. Amplification occurred in only 1.2% of patients with initially susceptible or monoresistant TB strains, but it occurred in 17% of those with polyresistant strains (but not multidrug-resistant strains, defined as strains with resistance to at least isoniazid and rifampicin) and in 7% of those with multidrug-resistant strains at diagnosis. Overall, 3.5% of the patients not initially infected with multidrug-resistant TB strains developed such strains during treatment. Amplification of resistance, however, was found only in polyresistant Beijing genotype strains.

CONCLUSIONS

High levels of amplification of drug resistance demonstrated under well-established DOTS program conditions reinforce the need for implementation of DOTS-Plus for multidrug-resistant TB in areas with high levels of drug resistance. The strong association of Beijing genotype and amplification in situations of preexisting resistance is striking and may underlie the strong association between this genotype and drug resistance.