Systematic review and meta-analysis of randomized clinical trials in the treatment of human brucellosis

Abstract

Background: Brucellosis is a persistent health problem in many developing countries throughout the world, and the search for simple and effective treatment continues to be of great importance. Methods and Findings: A search was conducted in MEDLINE and in the Cochrane Central Register of Controlled Trials (CENTRAL). Clinical trials published from 1985 to present that assess different antimicrobial regimens in cases of documented acute uncomplicated human brucellosis were included. The primary outcomes were relapse, therapeutic failure, combined variable of relapse and therapeutic failure, and adverse effect rates. A meta-analysis with a fixed effect model was performed and odds ratio with 95% confidence intervals were calculated. A random effect model was used when significant heterogeneity between studies was verified. Comparison of combined doxycycline and rifampicin with a combination of doxycycline and streptomycin favors the latter regimen (OR = 3.17; CI95% = 2.05-4.91). There were no significant differences between combined doxycycline-streptomycin and combined doxycycline-gentamicin (OR = 1.89; CI95% = 0.81-4.39). Treatment with rifampicin and quinolones was similar to combined doxycycline-rifampicin (OR = 1.23; CI95% = 0.63-2.40). Only one study assessed triple therapy with aminoglycoside-doxycycline-rifampicin and only included patients with uncomplicated brucellosis. Thus this approach cannot be considered the therapy of choice until further studies have been performed. Combined doxycycline/co-trimoxazole or doxycycline monotherapy could represent a cost-effective alternative in certain patient groups, and further studies are needed in the future. Conclusions: Although the preferred treatment in uncomplicated human brucellosis is doxycycline-aminoglycoside combination, other treatments based on oral regimens or monotherapy should not be rejected until they are better studied. Triple therapy should not be considered the current treatment of choice. © 2012 Solís García del Pozo, Solera.