Highly variable gastric emptying in patients with insulin dependent diabetes mellitus.

Some diabetic patients--particularly those with nausea and vomiting--frequently have evidence of delayed gastric emptying while other diabetic patients may in fact exhibit accelerated gastric emptying. Whether the presence or absence of symptoms of upper gastrointestinal dysfunction correlated with objective measures of gastric emptying in insulin dependent diabetic subjects was investigated. Twenty one insulin dependent diabetic patients underwent a solid phase gastric emptying scintiscan using in vivo labelled chicken liver. Thirteen patients had symptoms suggestive of gastrointestinal dysfunction (nausea, vomiting, early satiety, or constipation), while eight patients had no gastrointestinal symptoms. Eleven patients had orthostatic hypotension. All patients had been diabetic since childhood or adolescence. As a group, the diabetic patients showed a half time (T50) of gastric emptying (mean (SD) 150.0 min (163.7) that was not significantly different from that of 12 healthy control subjects (148.1 min (62.4)). Those diabetic patients without gastrointestinal symptoms and without orthostatic hypotension, however, showed a gastric emptying half time (70.1 min (41.6)) that was significantly faster than that of the control subjects. Conversely, those diabetic patients with nausea, vomiting, and early satiety (or early satiety alone) showed T50 values that were significantly greater than those of the diabetic patients without these symptoms. No correlation was found between the T50 value and the duration of diabetes, the fasting blood glucose at the time of study, or the respiratory variation in heart rate (E:I ratio). These observations indicate that highly variable rates of gastric emptying occur in insulin dependent diabetic patients, and that accelerated gastric emptying may occur in diabetic patients who have no symptoms of gastrointestinal dysfunction.