[Glibenclamide in type II diabetes as compared with placebo]

60 non-insulin-dependent diabetics (type II) were controlled in our outpatient clinic from the 1st February to the 31st May 1981 for investigation of the effect of glibenclamide on blood glucose. They had been treated with glibenclamide during the preceding 12 months at least. Satisfactory control of blood glucose [1, 9] was obtained in only 22 patients (37%) although primary failure of sulphonylurea therapy had been excluded in all patients. After being changed from glibenclamide to placebo, only 19 patients (32%) exhibited a significant increase in blood glucose. Altogether only six out of all sixty patients (10%) satisfied the criteria of adequate blood glucose control under glibenclamide with a significant increase in blood glucose on substituting placebo in place of glibenclamide. It is concluded that treatment with glibenclamide is indicated only if the fasting blood glucose is under satisfactory control (less than 120 mg%) and the efficacy of glibenclamide is confirmed at least once a year by a comparison with placebo.