The aim of this survey was to examine the incidence of duodenogastric reflux in patients with abdominal complaints and the relations between the nature and extent of reflux abdominal complaints, the use of drugs, smoking, the drinking of coffee and alcohol and histological changes in the gastric mucosa. A comparison was also made between gastric ulcer patients and patients with upper abdominal complaints with respect to the nature and extent of reflux. The patients examined included 107 with abdominal complaints and 33 with a gastric ulcer. Gastroscopy was performed, followed by determination of intragastric bile acids and lysolecithin and a duodenogastric isotope reflux examination using technetium-99m-diethyliminodiacetic acid (Tc-99m HIDA). Intragastric bile acid concentrations in the patients with upper abdominal complaints were in the range 7-21,458 mumol/l (mean 964 +/- 2342 mumol/l) and lysolecithin concentrations in the range 0-1992 mumol/l (mean 70 +/- 273 mumol/l). Isotope reflux was observed in 48% of the patients, the reflux index varying in the range 0-70% (mean 4 +/- 9%). The patients suffered more frequently from nausea, epigastric fullness and flatulence with increasing reflux, as assessed by the various methods used here, but only the increase in epigastric fullness symptoms with rising intragastric bile acid concentrations was statistically significant (p less than 0.05). Similarly the various measures of reflux were higher in those patients taking anticholinergic, psychotherapeutic or cardiovascular drugs, antacids or metoclopramide than in the patients not taking the respective drugs, although the only statistically significant increases were in intragastric bile acids among the users of antacids and metoclopramide (p less than 0.01 and p less than 0.05, respectively) and the increase in lysolecithin concentrations among those taking metoclopramide (p less than 0.05). Those abstaining from alcohol had an intragastric bile acid concentration over 1000 mumol/l significantly more often than those who drank alcohol (p less than 0.05), but smoking and the drinking of coffee showed no significant correlation with duodenogastric reflux. The body gastritis score increased significantly with the extent of isotope reflux and the concentrations of intragastric bile acids (p less than 0.05 and p less than 0.01, respectively), and the latter also showed a significant correlation with serum gastrin (p less than 0.05). No significant relationship could be detected between intragastric lysolecithin concentrations and the gastritis score.(ABSTRACT TRUNCATED AT 400 WORDS)