The oral antidiabetic agent 1-(hexahydro-1-H-azepin-1-yl)-3-(p-[2-(5-methyl-isoxazol-3-carboxamido)-ethyl]-phenylsulfonyl)-urea (glisoxepide, BS 4231, Pro-Diaban¿) was studied in a multi-centre open clinical trial comprising 4337 patients and was found to be an effective therapeutic agent for maturity-onset diabetics. About 66% of the patients had been pre-treated with oral antidiabetics. 54.7% of those 3572 patients treated with Pro-Diaban for at least 3 months achieved stabilization, which was classified according to strict criteria as "good", 73.4% achieved at least "satisfactory" stabilization. 12.5% of the cases dropped out after having received at least 12 mg/day because of inadequate efficacy of Pro-Diaban; 51.9% of these patients dropped out in the first 3 months (early failures 6.9% of the total case number). 61.9% received their daily dose in one portion, 29% in two and 10% in three portions. 98% received maximally 16 mg, 70% maximally 8 mg, and 42% 4 mg per day. Hypoglycaemic episodes led in 0.9% to a termination of the trial. In 1.22% of the cases the trial was terminated because of side effects, in 0.18% of cases they consisted of allergic or suspected allergic complaints and in 0.69% they consisted mainly of gastro-intestinal symptoms. Of the side-effects which did not lead to a termination of the trial, dizziness, headache and nausea were the relatively more frequent symptoms. An analysis of the laboratory data under consideration of pathological and normal initial and subsequent values showed that under the influence of Pro-Diaban pathological data reverted to normal more frequently than initially normal values changed to pathological ones. The analysis of laboratory and blood pressure data from 4 subgroups of patients, e.g. patients with diseases of the liver or the kidneys and hyperlipaemic or hypertensive patients, revealed that those data of special interest in each subgroup had mostly improved or remained unchanged under Pro-Diaban.