This study evaluated the effect of cisapride on glycaemic control, well-being and treatment satisfaction in insulin-dependent diabetes mellitus (IDDM) patients with documented moderate to severe glycaemic instability. Thirty-seven patients with glycaemic instability were included in a randomized, double-blind, placebo-controlled, cross-over study. Patients were instructed to take cisapride 10 mg q.i.d. or placebo tablets q.i.d. (15 to 30 min before each of the three main meals and before bedtime) for two periods of four weeks. The first treatment period was followed by a wash-out period of four weeks, with placebo treatment. Patients measured blood glucose with a glucometer before breakfast, lunch, dinner, 90 min after dinner and before bedtime every two days. After each treatment period, glucose data were collected from the glucometer to calculate the standard deviation (SDBG) of self-monitored blood glucose (SMBG). Questionnaires designed to measure well-being and treatment satisfaction were also completed initially and after each of the two treatment periods. There were no treatment effects of cisapride compared to placebo with respect to SDBG, mean glucose, percentage of extremes, number of hypoglycaemic episodes, insulin requirement, insulin variability, well-being, or treatment satisfaction. Cisapride 10 mg q.i.d. given for four weeks did not affect diabetic control or well-being and treatment satisfaction in a group of IDDM patients with glycaemic instability, i.e. a standard deviation of blood-glucose > 3.9 mmol/l.