In a prospective study we examined 20 Helicobacter pylori (HP)-positive duodenal ulcer patients (5 female, 15 male; age 26-70 [mean 43] years), 20 HP-positive patients with non-ulcer dyspepsia (10 female, 10 male; age 26-79 [mean 48] years) and 10 HP-negative patients with non-ulcer dyspepsia (5 female, 5 male; age 21-76 [mean 45] years) during upper GI-endoscopy. HP was detected by histology (H&E, Giemsa), rapid urease test (CLO) and serology (Cobas Core Anti-H. pylori EIA). IgA anti-HP in gastric juice was determined by ELISA. HP-positivity included positivity in all methods, and HP-negativity failure to detect HP-infection by all methods used. Of the 20 duodenal ulcer patients, 10 patients (2 female, 8 male; age 26-70 [mean 42] years) had an endoscopically documented duodenal ulcer at an earlier endoscopy with no current ulcer, 10 patients had florid duodenal ulcer disease at the time of examination. Duodenal ulcer patients compared with non-ulcer dyspepsia patients were tended to have higher serum IgG anti-HP (551 +/- 240 vs. 338 +/- 159 U/ml) and significantly higher gastric juice IgA anti-HP (50.0 +/- 7.3 vs. 26.5 +/- 4.3 relative units). Concentrations of both serum IgG anti-HP and gastric juice IgA anti-HP tended to be higher in patients with positive ulcer history but no present ulcer compared with patients with florid ulcer disease (934 +/- 456 vs. 170 +/- 63 U/ml and 60.0 +/- 8.6 vs. 40.8 +/- 10.4 relative units).(ABSTRACT TRUNCATED AT 250 WORDS)