Background: Rifamycins are an essential component of modern short-course regimens for treating tuberculosis. Rifabutin has favourable pharmacokinetic and pharmacodynamic properties and is less prone to drug-drug interactions than rifampicin. It could contribute to shortening of therapy or simplify treatment in HIV-positive people who also need antiretroviral drugs.
Objectives: To compare combination drug regimens containing rifabutin with those containing rifampicin for treating pulmonary tuberculosis
Search strategy: We searched Cochrane Infectious Diseases Group Specialized Register (January 2007), CENTRAL (The Cochrane Library 2006, Issue 4), MEDLINE (1966 to January 2007), EMBASE (1974 to January 2007), and LILACS (1982 to January 2007). We also searched the Indian Journal of Tuberculosis (1983 to 2006), conference abstracts, reference lists, and unpublished data on file at Pfizer Inc.
Selection criteria: Randomized and quasi-randomized trials including participants with sputum smear and/or culture-confirmed tuberculosis that compared a rifabutin-containing with an otherwise identical rifampicin-containing regimen.
Data collection and analysis: Two authors independently assessed study eligibility and methodological quality, and extracted data. Dichotomous data were analysed and combined using relative risks (RR) with 95% confidence intervals (CI) using a fixed-effect model. Subgroup analyses were carried out according to rifabutin dose.
Main results: Five trials with a total of 924 participants met the inclusion criteria; 5% of participants were HIV positive. Only one small trial was methodologically adequate. The two largest trials (818 participants) had unclear allocation concealment and included < 90% of randomized participants in the analysis. There was no statistically significant difference in between the regimens for cure (RR 1.00, 95% CI 0.96 to 1.04; 553 participants, 2 trials) or relapse (RR 1.23, 95% CI 0.45 to 3.35; 448 participants, 2 trials). The number of adverse events was not significantly different (RR 1.42, 95% CI 0.88 to 2.31; 714 participants, 3 trials), though the RR increased with rifabutin dose: 150 mg (RR 0.98, 95% CI 0.45 to 2.12; 264 participants, 2 trials); and 300 mg (RR 1.78, 95% CI 0.94 to 3.34; 450 participants, 2 trials). However, lack of dose adjustment by weight in the relevant trials complicates interpretation of this relationship.
Authors' conclusions: The replacement of rifampicin by rifabutin for first-line treatment of tuberculosis is not supported by the current evidence. HIV-positive people with tuberculosis, the group most likely to benefit from the rifabutin use, are under-represented in trials to date, and further trials in this group would be useful.