The progressive B-cell destruction at the onset of type 1 diabetes and the relevance of residual B-cell function to metabolic control prompted us to try to prevent further B-cell destruction in new-onset type 1 diabetic patients. Having already shown positive effects of sulfonylreas in long-standing IDDM patients, we decided to try the association of gliclazide with insulin in newly diagnosed IDDM patients. Therefore, 22 patients were randomly assigned to either gliclazide, 160 mg/24 h, or placebo (in association with standard insulin treatment) from the third week since the onset of the disease. Patients were studied every 6 months for 18 months. The insulin need was similar in the two groups at time 0, but thereafter it steadily decreased in the GCZ group becoming at 6, 12 and 18 months, respectively 65%, 42% and 39% of that of the PL group. The metabolic control was very similar in the two groups throughout the study. The residual B-cell function (fasting and post-breakfast C peptide), which was similar in the two groups at time 0, remained unchanged or improved in the GCZ group, whereas it decreased in the PL group. In conclusion, the combined insulin-gliclazide treatment appeared to improve B-cell function during the first 18 months of type 1 diabetes, inducing a major reduction of the insulin need with respect to patients treated with insulin alone.