OP0136 Should we eradicate helicobacter pylory before prescribing an nsaid ? result of a placebo controlled study

Background The influence of the infection by Helicobacter pylori (Hp) on the gastro-intestinal (GI) tolerability of NSAIDs remains unknown. Objectives 1) to determine the prevalence of GI symptoms in patients treated with NSAIDs as they are or not infected by Hp, 2) to estimate the impact of Hp eradication on the prevalence of these symptoms. Methods Multicentric, randomised, double blind, placebo-controlled study. Selection: patients presenting with a degenerative articular disease motivating the prescription of an NSAID for a duration > = 2 weeks. Distribution of the patients in two groups Hp-positive and Hp-negative by a serologic fast digital test. Randomisation of Hp-positive patients in two subgroups, receiving an eradication treatment: lansoprazole 60 mg/d, amoxicilline 2 g/d and clarithromycine 1 g/d during 7 days (group 1), or a placebo (group 2). Prevalence of GI symptoms estimated in the 3 groups (group 3 = not infected patients) at 2, 6 and 12 weeks after the beginning of the treatment. Analysis in intention to treat. Results 330 patients; group 1: n = 93; group 2: n = 92; group 3: n = 145. The highest prevalence of GI symptoms in the group 1 at S2 was related to diarrhoea, main side effect of antibiotics (21,8% vs 4,6% in the group 2 and 5,6% in the group 3). The prevalence of GI symptoms at S6 was not different between the groups 2 and 3 (p = 0,77). After adjustment of the NSAIDs dosage received between S6 and S12, the prevalence of GI symptoms was the same between group 1 and e group 3 (9.4% and 10.0%). This prevalence was lower than in group 2 (17.3%). This difference was not significant because of the samples size, but the most frequent GI symptom identified in this group was pyrosis. Conclusion The GI tolerance of conventional NSAIDs does not differ whether or not patients are infected by Hp. Systematic Hp eradication in infected patients does not seem to improve the digestive tolerance of NSAIDs for the short-term; medium-term observed tendency deserves to be confirmed.