Helicobacter pylori infection and functional dyspepsia.: Meta-analysis of efficacy of eradication therapy

BACKGROUND: To establish a causal link between Helicobacter pylori infection and functional dyspepsia it is necessary to demonstrate that H. pylori eradication induces an improvement in dyspeptic symptoms. Our aim was to perform a meta-analysis of randomized studies comparing, in functional dyspepsia, the efficacy of H. pylori eradication treatment with that of treatments with no effect on H. pylori infection. PATIENTS AND METHODS: Data sources: PubMed database, Cochrane Controlled Trials Register and abstracts from congresses until 2001. Selection criteria: a) studies including patients with functional dyspepsia and H. pylori infection; b) randomized trials comparing H. pylori eradication treatment with treatment (control) with no effect on H. pylori infection, and c) follow-up of at least 6 months. The quality of studies was assessed by a validated score. Statistics:. the main outcome was the percentage o patients improving in each therapeutic group. A meta-analysis was performed combining the odds ratios (OR) of individual studies in a global OR. RESULTS: Nine studies fulfilled the inclusion criteria and were thus included in the meta-analysis. Overall, 953 patients received an eradication treatment and 958 received a control treatment. The overall percentage of patients with symptomatic improvement in the eradication group was 43% (95% CI, 40-46%), and it was 39% (95% CI, 36-42%) in the control group. The OR for the effect of the eradication treatment vs. the control treatment was 1.20 (95% CI, 0.91-1.58). The number needed to treat (NNT) with eradication therapy to achieve a symptomatic improvement, compared with the control group, was 25. Although heterogeneity was demonstrated, it disappeared when one study showing positive, clearly discordant results was excluded. Thus, the percentage of patients with symptomatic improvement in the eradication treatment group was 47% (95% CI, 43-50%) and it was 45% (95% CI, 41-48%) in the control group (OR: 1.06; 0.85-1.31; NNT: 50). CONCLUSIONS: H. pylori eradication treatment is not associated with a statistically significant improvement of symptoms in patients with functional dyspepsia.