Aim: To study risk factors for failure of Helicobacter pylori eradication treatment.
Methods: Individual data from 2751 patients included in 11 multicentre clinical trials carried out in France and using a triple therapy, were gathered in a unique database. The 27 treatment regimens were regrouped into four categories.
Results: The global failure rate was 25.8% [95% CI: 24-27]. There was a difference in failure rate between duodenal ulcer patients and non-ulcer dyspeptic patients, 21.9% and 33.7%, respectively (P < 10(-6)). In a random-effect model, the risk factors identified for eradication failure in duodenal ulcer patients (n = 1400) were: to be a smoker, and to have received the group 4 treatment, while to receive a 10 day treatment vs. 7 days protected from failure. In non-ulcer dyspeptic patients (n = 913), the group 2 treatment was associated with failure. In both groups, age over 60 was associated with successful H. pylori eradication. There were less strains resistant to clarithromycin in duodenal ulcer patients than in non-ulcer dyspeptic patients. Clarithromycin resistance predicted failure almost perfectly.
Conclusion: Duodenal ulcer and non-ulcer dyspeptic patients should be managed differently in medical practice and considered independently in eradication trials.