Occult faecal blood loss determined by a 51Cr method and chemical tests in patients referred for upper gastrointestinal endoscopy.

In 119 patients referred for upper gastrointestinal endoscopy, the faecal blood loss was determined by a 51Cr method and 7 chemical tests. For patients with negative upper endoscopy (no. = 8), atrophic gastritis (no. = 30), gastric ulcer (no. = 31), or gastric cancer (no. = 23), the median 51Cr-determined faecal blood loss was 0.51, 0.61, 0.83, and 2.68 ml/24 h, respectively. For all chemical tests, the results were highly influenced by the upper time limit for positive reaction. Mixing of faecal specimens before testing did not prove essential. By repeated analyses of faecal samples stored for 3 days, the benzidine test showed a decreased sensitivity (p less than 0.01), whereas Fecatwin and Fecatwin sensitive showed an increased number of positive tests (p less than 0.01). Of cases of gastric cancer, tetramethylbenzidine tests including Hemo-Fec Test, benzidine test, Fecatwin sensitive, Hemoccult II, and Fecatwin could detect about 85%, 80%, 60%, 55%, and 30%, respectively.