Seven studies (n=1,444) and three abstracts (n=303) were included in the review.
People receiving therapy versus controls - 1 month after intervention.
Four full papers (quality: three scored 5/5, one 4/5) and three abstracts (quality: one scored 3/5, two 2/5) compared the success of therapy in the treatment and control groups. When pooled, there was no statistically significant difference in success between the control and therapy groups (OR 1.29, 95% CI: 0.89, 1.89). This was also the case when only studies of better quality were pooled (OR 1.41, 95% CI: 0.85, 2.33), and when one study was removed from the analysis because of a different definition of dyspepsia. Statistical heterogeneity was reported in three of the analyses; statistical homogeneity when the single study was removed from the analysis.
Patients cured of dyspepsia and those with persistent infection.
Three full papers (quality: all 5/5) and two abstracts (quality: one 3/5, one 2/5) compared the success of therapy in people cured of dyspepsia and those with persistent infection. When pooled, there was no statistically significant difference in success between the groups (OR 1.17, 95% CI: 0.87, 1.59) and no statistical heterogeneity.
Scale used to score dyspepsia.
Seven full papers (quality: three 5/5, one 4/5, two 3/5, one 1/5) compared the changes in the scale used to score dyspepsia between those receiving therapy and controls. Only the smallest study (quality score 1/5) reported a clinically significant benefit of H. pylori therapy.