Five hundred and forty patients with severe infection were enrolled in a multicentre, prospective, randomized, non-blinded study to compare the efficacy and safety of i.v. ciprofloxacin with i.v. standard therapy. Five hundred and thirty-one patients received at least one dose of study drug for pneumonia (310), septicaemia (112) or skin and skin structure infection (109). Intravenous ciprofloxacin (400 mg, every 8 h) or i.v. ciprofloxacin (400 mg, every 8 h) plus a beta-lactam were compared with a standard monotherapy (beta-lactam) or combination (aminoglycoside plus a beta-lactam) therapy. Patients were treated parenterally for a minimum of 2 or 3 days, then at the discretion of the investigator could be switched to oral therapy (ciprofloxacin 750 mg, every 12 h or a standard oral therapy). Patients were randomized in the ratio of 2:1 for the ciprofloxacin and standard therapy treatment groups and stratified to monotherapy if the APACHE II score was < or = 20 or to combination therapy if the APACHE II score was 21-29. Three hundred and ninety-five (74%) patients were valid for the efficacy analysis: these comprised 242 pneumonia (167 ciprofloxacin and 75 standard therapy), 70 septicaemia (47 ciprofloxacin and 23 standard therapy), and 83 skin infections (56 ciprofloxacin and 27 standard). The primary efficacy variable was clinical response and the secondary efficacy assessment was bacteriological response at the end of therapy (2 or 3 days after treatment). The mean duration of therapy for patients receiving only i.v. monotherapy or combination therapy was shorter (9-10 days) than for patients receiving sequential i.v./p.o. therapy (14-17 days). At the end of therapy, overall clinical resolution/improvement (success) for monotherapy was 138/166 (83%) for the ciprofloxacin group, compared with 74/87 (85%) for standard-treated patients (95% CI = -11.5% to 7.6%), and for combination therapy the response was 43/51 (84%) for the ciprofloxacin group and 14/20 (70%) for standard-treated patients (95% CI = -6.3% to 34.9%). For pneumonia, the most frequent infection treated, clinical success rates following monotherapy were 85% for ciprofloxacin and 83% for standard-treated patients and 83% for ciprofloxacin compared with 69% for standard-treated patients in the combination therapy group. Bacteriological eradication/presumed eradication following monotherapy was 85/102 (83%) for ciprofloxacin and 31/46 (67%) for standard-treated patients (95% CI = 1.6% to 30.3%), and that for combination therapy was 29/36 (81%) for ciprofloxacin and 7/10 (70%) for standard-treated patients (95% CI = -18.3% to 39.5%). Drug-related adverse events, primarily diarrhoea and nausea, were reported in 22% of ciprofloxacin-treated patients and 20% of standard-treated patients. In summary, ciprofloxacin administered alone or in combination was found to be effective in treating a wide range of severe infections.