Ethambutol plus isoniazid for the treatment of pulmonary tuberculosis-A controlled trial of four regimens

Abstract

A controlled clinical trial of 4 regimens was undertaken in patients with bacteriologically positive, newly-diagnosed pulmonary tuberculosis. The regimens were: ethambutol 15 mg/kg plus isoniazid 400 mg, daily (E 7H 7) ; ethambutol 45 mg/kg plus isoniazid 15 mg/kg, twice a week (E 2H 2) ; ethambutol 90 mg/kg plus isoniazid 15 mg/kg, once a week plus isoniazid 15 mg/kg, mid-way between the weekly doses (E 1 H 2) ; and ethambutol 90 mg/kg plus isoniazid 15 mg/kg, once a week (E 1H 1). All patients received streptomycin 1 g plus ethambutol 25 mg/kg body-weight plus isoniazid 400 mg daily for the first 2 weeks. The total duration of treatment was 12 months for all patients. There were 484 patients admitted to the study. After excluding 60 (41 with initial drug resistance to isoniazid), there remain 424 patients (107 E 7H 7, 101 E 2H 2, 107 E 1H 2, 109 E 1H 1) in the main analyses. The pretreatment characteristics of the 4 groups were broadly similar. A favourable response at 12 months was observed in 96%, 88%, 93% and 75% of patients respectively: the differences between the E,H, regimen and the other 3 regimens were all significant (P≤0.03) as was that between the E 7H 7 and E 2H 2 regimens (P=0.05). Among the slow inactivators of isoniazid, the proportions with a favourable response at one year were similar in the 4 groups (range 95-91 %). However, among the rapid inactivators, the proportion with a favourable response in the E 1H 1 group was only 57%. There was suggestive evidence that the E 1H 2 regimen was superior to the E 2H 2 regimen. Of the patients with bacteriologically quiescent disease at 1 year, approximately a half, at random, had no further chemotherapy and were followed up for a 4-year period. Bacteriological relapse requiring retreatment occurred in 15 % of 54 E 7H 7 patients, 26 % of 38 E 2H 2, 33 % of 43 E 1H 2 and 54 % of 37 E 1H 1 patients, a significant difference (P<0.001). A final evaluation of long-term (5-year) favourable response achieved by the 12-month regimens was 83 % for the E 7H 7, 63 % for the E 1H 2 and 33 % for the E 1H 1 regimens. In general, the regimens were well tolerated and the incidence of adverse reactions to the drugs was low. Of the 424 patients, 6 (1.4 %) developed a visual disturbance during the year of chemotherapy.