Efficacy of twice weekly treatment for tuberculosis given under direct observation in Africa

The efficacy of a 6 months course of twice weekly therapy with 4 drugs for tuberculosis, preceded by a 2-3 weeks intensive daily phase, is unknown. Implementation of this regime as community-based directly observed therapy in Africa is highly effective (85% completion rate); it is important to estimate the efficacy of the regime before advocating its widespread use and before conducting prospective trials. We retrospectively evaluated 109 consecutive adults with culture-positive pulmonary tuberculosis who had documented completion of treatment; 84 (77%) were traced and in 15 (14%) a history was obtained from a close relative; 10 (9%) had left the area. Nineteen patients were producing sputum and 4 of these were culture-positive for Mycobacterium tuberculosis, giving an estimated cure rate of 95% (95% confidence interval, 89-98%). Follow-up specimens revealed no acquired drug resistance and restriction fragment length polymorphism analysis of patient-paired specimens showed them to be nearly identical, indicating that treatment had failed or there had been early relapse. This preliminary study suggested that generally twice weekly 4-drug treatment for tuberculosis, given under direct observation, is curative in an acceptable proportion of patients. Prospective trials are indicated.