The effect of coffee on gastric emptying was addressed in a scintigraphic liquid-phase gastric emptying study in patients with non-ulcer dyspepsia. Ninety-three subjects (56 males, 37 females; mean age 40 years, range 17-77 years) diagnosed as having non-ulcer dyspepsia were enrolled in the study. The baseline study was to drink 500 ml of 5% glucose water and the coffee study was to drink 500 ml of 5% glucose water containing 4 g of regular instant coffee. The two studies were performed on separate days. Fifteen of the 93 subjects were chosen at random to undergo repeated coffee studies for evaluation of reproducibility. Overall the 93 subjects showed accelerated gastric emptying, as measured by half emptying time (T1/2) with coffee compared with baseline (35.7 +/- 10.5 vs 45.0 +/- 23.1 min, P < 0.001). However, 68 (73.2%) subjects showed accelerated emptying (-14.8 +/- 19.5 min), while 25 (26.8%) subjects showed delayed emptying (5.9 +/- 4.5 min) after ingestion of coffee. There was no significant difference in the change in gastric emptying with coffee in duplicate measurements from the 15 subjects who had two coffee studies (P = 0.082). We conclude that coffee accelerates liquid-phase gastric emptying in the majority of patients with non-ulcer dyspepsia.