Therapy for Helicobacter pylori in patients with nonulcer dyspepsia. A meta-analysis of randomized, controlled trials.

PURPOSE To assess the effect of eradication therapy for Helicobacter pylori on symptoms of nonulcer dyspepsia. DATA SOURCES Duplicate searches of bibliographic databases, reviews of proceedings of annual gastroenterology and H. pylori meetings from 1995 to 1999, reviews of reference lists, and contact with primary investigators and pharmaceutical manufacturers. STUDY SELECTION Included studies 1) examined patients with nonulcer dyspepsia and H. pylori infection; 2) used combination therapy for H. pylori and a control therapy without efficacy against H. pylori; 3) were randomized, controlled trials; 4) lasted for at least 1 month after the end of therapy; and 5) assessed symptoms of nonulcer dyspepsia. Ten studies were included. DATA EXTRACTION Independent, duplicate data extraction of the methodologic quality, population, intervention, study design, duration, and outcome of the trials. DATA SYNTHESIS The odds ratio (OR) for treatment success in nonulcer dyspepsia with H. pylori eradication therapy compared with control therapy was 1.29 (95% CI, 0.89 to 1.89; P = 0.18). However, significant heterogeneity (P = 0.04) calls the validity of aggregating the data into question. Heterogeneity resolved with the exclusion of one study (OR, 1.07 [CI, 0.83 to 1.37]; P > 0.2). For predefined analysis of trials that used a specifically stated definition of dyspepsia (that is, upper abdominal pain or discomfort), the OR was 1.04 (CI, 0.80 to 1.35) without heterogeneity. For treatment that resulted in cure rather than persistent infection, the OR was 1.17 (CI, 0.87 to 1.59) without heterogeneity. CONCLUSION This meta-analysis provides little support for the use of H. pylori eradication therapy in patients with nonulcer dyspepsia.