Treatment of shigellosis is currently limited by the high prevalence of multidrug-resistant strains of Shigella. Objective: To determine the efficacy of azithromycin in the treatment of shigellosis. Design; Randomized, double- blind clinical trial. Setting: Diarrhea treatment center in Dhaka, Bangladesh. Patients: 70 men with shigellosis that had lasted 72 hours or less. Interventions: Patients stayed in the hospital for 6 days. Thirty-four patients were randomly assigned to receive 500 mg of azithromycin on study day 1, followed by 250 mg once daily for 4 days; 36 patients were assigned to receive 500 mg of ciprofloxacin every 12 hours for 5 days. Measurements: Clinical treatment failure was considered to have occurred if frank dysentery persisted for 72 hours after therapy began or if on study day 5 a patient had more than six stools, had any bloody-mucoid stools, had more than one watery stool, or had an oral body temperature exceeding 37.8 °C. Bacteriologic treatment failure was considered to have occurred if Shigella strains could be isolated from a stool sample after study day 2. Therapy was considered either clinically or bacteriologically successful in patients who completed therapy and did not meet criteria for failure. Results: Therapy was clinically successful in 28 (82%) patients who received azithromycin and 32 (89%) patients who received ciprofloxacin (difference, 7% [95% CI, 23% to 10%]). Therapy was bacteriologically successful in 32 (94%) patients receiving azithromycin and 36 (100%) patients receiving ciprofloxacin (difference, 6% [CI, 14% to 2%]). Peak serum concentrations of azithromycin were equal to the minimum inhibitory concentration (MIC) of the infecting Shigella strains, whereas serum concentrations of ciprofloxacin were 28 times the MIC. Stool concentrations of both drugs were more than 200 times the MIC. Conclusion: Azithromycin is effective in the treatment of moderate to severe shigellosis caused by multidrug-resistant Shigella strains.