Aims/hypothesis: We evaluated in a double-blind study the effect of early treatment with the immunomodulatory drug fusidin in patients with newly diagnosed type 1 diabetes mellitus. Methods: Twenty-eight adults with newly diagnosed type 1 diabetes were included in the study. The patients were randomly assigned (computergenerated random number sequence) to two experimental groups. Patients allocated to the fusidin (FUS) group (n=15) received sodium fusidate (fusidin; 500 mg orally three times daily for 4 weeks). Subsequently the drug was given at the same dose and scheduled for two consecutive weeks a month followed by 2 weeks a month without the drug for 20 weeks. Subjects allocated to the placebo (PCB) group (n=13) received placebo according to the same schedule and conditions described for sodium fusidate in the FUS group. All patients received a diet adjusted to their age and BMI, and intensive insulin therapy. Results: There were no statistically significant differences between the FUS and PCB groups in beta cell function, evaluated by basal and glucagon-stimulated C-peptide values during the follow-up (24 and 48 weeks). There was also no difference between the two groups in insulin requirement after 48 weeks (0.4± 0.2 and 0.4±0.2 U/kg body weight for the FUS and PCB groups, respectively). Antibody titres, including insulin autoantibodies, were similar in the two groups during the follow-up. Conclusions/interpretation: Early treatment of newly diagnosed type 1 diabetes patients with intermittently administered fusidin failed to influence the natural course of the disease.