The aim of our study was to test the hypothesis that the better absorption of sorbitol when ingested with glucose could be related to a delayed gastric emptying. We tested the effect of the ingestion of glucose and lipids on the gastric emptying and intestinal absorption of sorbitol in six healthy volunteers, using gastric scintigraphy and hydrogen breath test. After an overnight fast, subjects ingested in random order, on 48-h test periods separated by at least one week, the following solutions: (a) 20 g sorbitol alone; (b) 20 g sorbitol and 20 g glucose; (c) 20 g sorbitol and 9 g lipids. Isotopic acquisitions were taken for 3 h following the ingestion of sorbitol labelled with 111Indium. Hydrogen concentration was measured in end-expiratory samples during 5 h, and the areas under the breath hydrogen curve, reflecting the amounts of sorbitol unabsorbed in the small bowel, were compared between periods. Mean area under the curve was 397 +/- 159 when sorbitol was ingested alone, and this was significantly lower when ingested with glucose or lipids (313 +/- 181 and 337 +/- 135, respectively; P < 0.05). The three curves of sorbitol gastric emptying differed significantly from each other, the gastric emptying being the slowest for sorbitol plus lipids, and the fastest for sorbitol taken alone. We found a positive correlation between the half-emptying time and the hydrogen areas under the curve (r = 0.46, P = 0.05). In conclusion, our study demonstrates that adding glucose or lipids to a solution of sorbitol slows the gastric emptying of sorbitol, resulting in a better intestinal absorption of sorbitol.