Two new trials support rifampin as a better option than isoniazid for latent tuberculosis.
Isoniazid (INH) monotherapy for 6 to 9 months has long been the standard of care for latent tuberculosis treatment. However, numerous observational studies have found low rates of regimen completion with INH, and comparable efficacy with 3- or 4-month rifampin or 12-dose INH-rifapentine regimens. To further address this issue, an international team of investigators has now reported on two parallel, open-label, randomized trials that compared the outcomes obtained with either 4 months of rifampin or 9 months of INH for management of latent tuberculosis.
Menzies and colleagues randomized 6063 adult patients 18 years of age and older in 9 countries to the two regimens; 5744 participants completed 28 months of follow-up. They found that the rate of treatment completion was significantly higher with rifampin than with INH therapy (78.8% vs. 63.2%), that the rate of confirmed or clinically diagnosed cases of active tuberculosis during follow-up was comparable (0.10 cases per 100 person-years for rifampin vs. 0.11 cases per 100 person-years for INH), and that there were significantly lower rates of grade 3 to 5 adverse events causing discontinuation of rifampin than of INH (all events, 1.5% vs. 2.6%; hepatotoxic events, 0.3% vs. 1.5%).
Diallo and colleagues performed a similar study, randomizing 844 children younger than 18 years in 7 countries to the two regimens. The primary outcome was the safety of the two regimens; treatment adherence and efficacy were secondary outcomes. No grade 1 to 5 adverse events were related to either drug, and no significant differences in minor symptoms were due to the drugs. The rate of treatment completion was significantly higher with rifampin than with INH therapy (86.5% vs. 77.1%). There were no cases of active tuberculosis in the rifampin group versus two cases of active tuberculosis in the INH group, but the study was underpowered to assess efficacy.
ARYAN'S BLOG 
Comment
These two studies add to the body of evidence that indicates that the shorter-course rifampin or INH-rifapentine regimens have a clear benefit over INH monotherapy for latent tuberculosis in terms of treatment adherence (NEJM JW Infect Dis Sep 2018 and MMWR Morb Mortal Wkly Rep 2018; 67:723). The treatments have comparable efficacy in adults and likely have comparable efficacy in children. In addition, the safety profile may slightly favor rifampin. In my own practice I will only opt for INH monotherapy when rifampin or rifapentine are not options due to a history of past adverse reactions, potential significant drug-drug interactions, or known patient exposure to a rifampin-resistant Mycobacterium tuberculosis isolate.