OBJECTIVE In the present study, a recently developed manometric technique was used to study antroduodenal motility in ambulant type I diabetic subjects. METHODS In 12 patients with type I diabetes, antroduodenal manometry was performed for 20 h during the fasting period and the postprandial period after a standardized dinner and breakfast. All patients had evidence of cardiac autonomic neuropathy and complained of dyspeptic symptoms. During the manometric study, the blood glucose levels were frequently monitored and kept close to euglycemia in the diabetic patients. The results were compared with 12 healthy control subjects. RESULTS The migrating motor complex cycles observed in the diabetic subjects were longer than in the control subjects, 118.9 +/- 46.0 vs. 87.0 +/- 21.6 min (P < 0.05). This increase was attributable to a prolonged phase II, 78.0 +/- 35.5 vs. 37.7 +/- 18.5 min (P < 0.05). In the diabetic subjects, antral phase III was seen significantly less than in the control subjects, 16.7 vs. 43.3% (P < 0.005). In 50% of the diabetic patients, total absence of antral phase III was observed-this phenomenon was not seen in the healthy control subjects. After dinner, the antral motility index was less in diabetic subjects compared with the healthy volunteers, indicating antral hypomotility (P < 0.01). Six diabetic patients showed abnormal duodenal activity such as early recurrence of phase III and bursts after dinner. No significant differences in antral motility index or in duodenal motility patterns were observed after breakfast. Six diabetic patients complained of dyspeptic symptoms after dinner, whereas none had dyspeptic symptoms after breakfast. In 67% of the patients, nausea was reported after an early phase III or a burst. CONCLUSIONS This study shows that prolonged ambulatory antroduodenal manometry is a feasible technique in patients. Recording multiple migrating motor complexes showed that interdigestive motor abnormalities of the stomach and duodenum are common in diabetic patients. Furthermore, it shows the occurrence of antral hypomotility and abnormal duodenal motility patterns after a high-calorie meal, with dyspeptic symptoms in diabetic patients being related to the composition of the meal.