The objective of the present prospective, noncomparative, multicenter study was assess the safety and efficacy of gentamicin and doxycycline therapy for human brucellosis. In the first part of the study, a cohort of 17 patients received 100 mg of doxycycline (or 50 mg/kg of body weight per day if the body weight was < 40 kg) orally every 12 h for 45 days (cohort 1). In the second part of the study a subsequent cohort of 35 patients was treated with doxycycline at the same dosage for 30 days (cohort 2). All patients were treated intramuscularly with gentamicin at 240 mg (or 5 mg/kg per day if the body weight was < 50 kg) once daily for the first 7 days. Both cohorts showed a favorable response during therapy, and there were no therapeutic failures. Relapse was noted in 1 (5.9%; 95% confidence interval [95% CI], 0.15 to 28.7%) of the 17 patients in cohort 1 and in 8 (22.9%; 95% CI, 10.4 to 40.1%) of the 35 patients in cohort 2. Nineteen patients (36.5%; 95% CI, 23.6 to 51.0%) had adverse effects, with no differences between cohorts, and no patients had a treatment-limiting adverse effect. The study indicates that the combination of doxycycline for 45 days and gentamicin for 7 days is an effective and well-tolerated therapy for human brucellosis. The relapse rates obtained with doxycycline treatment for 30 days appear to be higher than those obtained with doxycycline treatment for 45 days.