About The Study

This study meant to show the equivalence in efficacy between rifabutin (RBT) 300 mg/day for three months and isoniazid (INH) 300 mg/day for 12 months, in preventing tuberculosis in HIV-positive subjects with latent tuberculosis.

Study Approach

Volunteers with a positive tuberculin skin test and documented HIV infection were randomized to receive either isoniazid for 12 months, or rifabutin for three months. During months 1-3, group one took rifabutin and placebo, and group two took placebo and isoniazid. During months 4-12, group one took only the placebo, and group two took isoniazid. A follow-up period of 24 months was planned. The comparison between the two groups was to focus on the time to development of active tuberculosis. Clinical and laboratory examinations performed during drugs administration were used for the safety analysis.

Study population

From October 1995 to April 1997, 628 patients were enrolled in centres located in North America, Latin America, Africa, Far East, and Europe. Since this represented only 33% of the 1880 volunteers expected by that date and required to demonstrate the equivalent efficacy of the two drugs, the study enrollment was terminated in August 1997.


When the study enrollment was terminated, it was calculated that a follow-up period of five years would still be insufficient to provide enough data for the statistical analysis of efficacy. Therefore, the follow-up period was also terminated.

In August 1997, 107 (17%) of the patients had discontinued treatment. Five of them developed active tuberculosis. Analysis of reasons for discontinuation and adverse reactions is being completed.


Difficulties of enrollment, interaction with protease inhibitors, and declining interest in tuberculosis prevention in northern countries contributed to the early termination of this multinational study. Consequently, the equivalence between rifabutin and isoniazid for preventive therapy of latent tuberculosis in HIV-positive patients could not be demonstrated.

Note: These results were taken from abstract 60858 presented at the 12th World AIDS Conference in Geneva.