Background: Although a number of tests have been described to detect the presence of Helicobacter pylori in biopsy specimens, studies of positive and negative value have largely been performed on untreated patients; testing the reliability of posttherapy has not been done.
Methods: We examined the value of the number and site of biopsies performed and the method used for specimen evaluation posttherapy. For postantimicrobial therapy of 141 patients with previously confirmed H. pylori infection, three biopsies were taken, two from the antrum and one from the corpus. Individual slides were coded, randomized, and interpreted blindly by two pathologists. Furthermore, in 143 patients, a biopsy specimen was taken from the antrum and was immediately inserted into the gel of the rapid urease test, and the results were compared with those obtained from histopathology obtained at the same time.
Results: In 71 patients, H. pylori therapy was unsuccessful; in 61 (86%), all three sites were positive. The highest yield with a single large cup biopsy specimen was 94%; the lowest was 91%. Two antral biopsies were negative in 4% [95% confidence interval (CI) = 1-12%]. The combination of a biopsy from the angulus incisura and one from the greater curvature of the corpus correctly identified all treatment failures (95% CI = 95-100%). The rapid urease test was false-negative in 5% (95% CI = 1-13%); there were no false-positives.
Conclusion: Use of either the rapid urease test or two antral biopsies for evaluation of success of antimicrobial therapy for H. pylori infection will result in a false declaration of cure in at least 5% of cases. Three large cup gastric mucosal biopsies for histology are recommended for evaluation of the success of anti-H. pylori therapy.