Rationale: Cavitary disease and delayed culture conversion have been associated with relapse. Combining patient characteristics and measures of bacteriologic response might allow treatment shortening with current drugs in some patients.

Objectives: To assess whether treatment could be shortened from 6 to 4 months in patients with noncavitary tuberculosis whose sputum cultures converted to negative after 2 months.

Methods: This study was a randomized, open-label equivalence trial. HIV-uninfected adults with noncavitary tuberculosis were treated daily with isoniazid, rifampin, pyrazinamide, and ethambutol for 2 months, followed by 2 months of isoniazid and rifampin. After 4 months, patients with drug-susceptible TB whose sputum cultures on solid media were negative after 8 weeks of treatment were randomly assigned to continue treatment for 2 more months or to stop treatment. Patients were followed for relapse for 30 months after beginning treatment.

Measurements and Main Results: Enrollment was stopped by the safety monitoring committee after 394 patients were enrolled due to apparent increased risk for relapse in the 4-month arm. A total of 370 patients were eligible for per protocol analysis. Thirteen patients in the 4-month arm relapsed, compared with three subjects in the 6-month arm (7.0 vs. 1.6%; risk difference, 0.054; 95% confidence interval with Hauck-Anderson correction, 0.01–0.10).

Conclusion: Shortening treatment from 6 to 4 months in adults with noncavitary disease and culture conversion after 2 months using current drugs resulted in a greater relapse rate. The combination of noncavitary disease and 2-month culture conversion was insufficient to identify patients with decreased risk for relapse.