The effects and kinetics of oral glibenclamide (Gb) and glipizide (Gz) were studied in Caucasian and Chinese patients (ten in each group) with type-2 diabetes. In randomised order, 2.5 mg Gb, 2.5 mg Gz or placebo was given orally before the administration of 75 g oral glucose. Concentrations of insulin and proinsulin were determined using radioimmunoassay (RIA) without cross-reactivities, and sulphonylurea concentrations were determined using high-performance liquid chromatography (HPLC). There were no significant interethnic differences in Gb or Gz effects whether on glucose, insulin or proinsulin/insulin ratio at any time point. Following Gz, however, Chinese patients had greater increments of serum proinsulin at 10-30 min compared with Caucasians. Apart from the maximum plasma concentration (Cmax) and area under the plasma concentration-time curve (AUC) of Gz being higher among the Chinese, no significant interethnic differences in pharmacokinetics were found. It appears that the same dosage principles could be used for Caucasian and Chinese patients with type-2 diabetes when Gb or Gz are prescribed.