Effect of post-treatment isoniazid on prevention of recurrent tuberculosis in HIV-1-infected individuals: a randomised trial

Background Patients with HIV-1 infection respond well to treatment for active tuberculosis, but whether such patients are at increased risk of disease recurrence after complete cure is uncertain. We did a randomised trial in Port au Prince, Haiti, to determine whether recurrent tuberculosis after curative tuberculosis treatment is more common in HIV-1-infected individuals than HIV-1-uninfected individuals, and to determine whether post-treatment isoniazid prophylaxis decreases the risk of recurrent tuberculosis. Methods Patients older than 18 years who were diagnosed with a first episode of tuberculosis at the national HIV testing centre in Haiti, and who successfully completed a 6-month rifampicin-containing regimen for active pulmonary tuberculosis, were randomly assigned 1 year of post-treatment isoniazid prophylaxis or placebo. The primary outcome measure was rate of recurrent tuberculosis after at least 24 months. An intention-to-treat analysis was used. Findings Of 354 patients with active pulmonary tuberculosis, 274 successfully completed treatment, and 233 were randomised. Of 142 HIV-1-positive patients, 68 were assigned isoniazid and 74 placebo. Of 91 HIV-1-negative individuals, 51 were assigned isoniazid and 40 placebo. The rate of recurrent tuberculosis was 4·8 per 100 person-years in HIV-1-infected individuals and 0·4 per 100 person-years in uninfected individuals (relative risk 10·7 [95% CI 1·4–81·6]). Among HIV-1-positive patients receiving isoniazid, the tuberculosis recurrence rate was 1·4 per 100 person-years, and among HIV-1-positive patients receiving placebo, it was 7·8 per 100 person-years (0·18 [0·04–0·83]). Among HIV-1-positive individuals, all cases of recurrent tuberculosis occurred in individuals with a history of HIV-1-related symptoms before initial tuberculosis diagnosis. Interpretation The rate of recurrent tuberculosis is higher in HIV-1-positive individuals than in HIV-1-negative individuals, and is strongly associated with a history of symptomatic HIV-1 disease before initial tuberculosis diagnosis. Post-treatment isoniazid prophylaxis decreases the risk of recurrence in HIV-1-positive individuals, and should be considered for HIV-1-positive individuals with a history of HIV-1-related symptoms at the time of tuberculosis diagnosis.