Levofloxacin for the Prevention of Multidrug-Resistant Tuberculosis in Vietnam

BACKGROUND Prevention of drug-resistant tuberculosis is a global health priority. However, trials evaluating the effectiveness of treating Mycobacterium tuberculosis infection among contacts of persons with drug-resistant tuberculosis are lacking. METHODS We conducted a double-blind, randomized, controlled trial comparing 6 months of daily levofloxacin (weight-based doses) with placebo to treat M. tuberculosis infection. The trial population comprised household contacts of persons with bacteriologically confirmed rifampicin-resistant or multidrug-resistant (MDR) tuberculosis in Vietnam. Contacts of any age with a positive tuberculin skin test or immunologic impairment were eligible. The primary end point was bacteriologically confirmed tuberculosis within 30 months. Secondary end points included grade 3 or 4 adverse events, death from any cause, and acquired drug resistance. RESULTS Of 3948 persons screened for eligibility, 61 (1.5%) had coprevalent tuberculosis (defined as active tuberculosis disease diagnosed before randomization) and 2041 underwent randomization. Of these 2041 participants, 1995 (97.7%) completed 30 months of follow-up, had a primary end-point event, or died. Confirmed tuberculosis occurred in 6 participants (0.6%) in the levofloxacin group and 11 (1.1%) in the placebo group (incidence rate ratio, 0.55; 95% confidence interval [CI], 0.19 to 1.62); this difference was not significant. There was little difference in grade 3 or 4 adverse events between the two groups (risk difference, 1.0 percentage point; 95% CI, -0.3 to 2.4). Adverse events of any grade were reported in 306 participants (31.9%) taking levofloxacin and 125 (13.0%) taking placebo (risk difference, 18.9 percentage points; 95% CI, 14.2 to 23.6). No acquired fluoroquinolone resistance was observed. CONCLUSIONS Although the incidence of tuberculosis was lower in the levofloxacin group than in the placebo group at 30 months, the difference was not significant.