Setting: The Tuberculosis Research Centre, Chennai, and its unit at Madurai, South India.
Objective: To design oral short-course regimens for the treatment of sputum-positive pulmonary tuberculosis that could be more easily implemented under field conditions.
Design: A total of 1203 patients was randomly allocated to one of three regimens. I (2EHRZ7/6EH7): 8-month daily regimen of ethambutol (E), isoniazid (H), rifampicin (R) and pyrazinamide (Z) for 2 months, followed by E and H for 6 months. II (2EHRZ2/4EHR2): 6-month twice-weekly regimen with the same four drugs for 2 months, followed by EHR for 4 months. III (2HRZ2/4HR2): similar to Reg. II, but without ethambutol. In Reg. I, drugs were given completely unsupervised. Regs. II and III were either completely or partially supervised.
Results: Drug-susceptible group: At the end of treatment, 3.6% of 305 patients in Reg. I, 0.4% of 263 in Reg. II and 9.3% of 257 in Reg. III had an unfavourable bacteriological response. By 24 months after start of treatment, 5% of 290 in Reg. I, 11% of 258 in Reg. II and 10% of 229 in Reg. III had a bacteriological relapse requiring treatment. Giving the twice-weekly regimens partly unsupervised did not influence the response to treatment, emergence of drug resistance or relapse rates. Isoniazid resistant group: Unfavourable response and relapse with Reg. I (94 patients) was 17% and 8%, with Reg. II (59 patients) 20% and 25%, and with Reg. III (74 patients) 62% and 15%, respectively.
Conclusion: A fully unsupervised ethambutol-containing regimen given daily for 8 months (Reg. I) was found to be very effective even in the presence of isoniazid-resistant bacilli. With the ethambutol-containing twice-weekly regimen, the response at the end of treatment was near 100%, but the relapse rate was high (11%). The non-ethambutol twice-weekly regimen was not satisfactory. All three regimens failed in the presence of bacilli resistant to rifampicin and isoniazid.