Postprandial hypotension is an important disorder for which current management is suboptimal. In healthy older subjects, oral and small-intestinal glucose administration decreases blood pressure (BP), and the magnitude of the reduction is dependent on the rate of glucose entry into the small intestine and, possibly, the release of glucagon-like peptide-1 (GLP-1). There is little information about the effects of other carbohydrates, particularly those poorly absorbed, on BP. The aim of the present study was to compare the effects of drinks containing xylose, glucose or water alone on BP, gastric emptying (GE), incretin hormone secretion, glycaemia and insulinaemia in healthy older subjects. A total of eight healthy older subjects (aged 65–75 years) had simultaneous measurements of BP (DINAMAP), GE (three-dimensional ultrasound), blood glucose, serum insulin, GLP-1 and glucose-dependent insulinotropic peptide (GIP), on three separate occasions, in a double-blind, randomised order. On each day, subjects consumed a 300 ml drink of water, glucose (50 g) or d-xylose (50 g). Glucose (P = 0·02), but not xylose (P = 0·63), was associated with a fall in BP. There was no difference in the GE of glucose and xylose (P = 0·47); both emptied slower than water (P < 0·001). Xylose had minimal effects on blood glucose, serum insulin or serum GIP, but was more potent than glucose in stimulating GLP-1 (P = 0·002). In conclusion, in healthy older subjects, xylose empties from the stomach at the same rate as glucose, but has no effect on BP, possibly because it is a potent stimulus for GLP-1 release. Xylose may be considered as an alternative sweetener to glucose in the management of postprandial hypotension.