Meta-analysis of randomized controlled trials comparing standard clinical doses of omeprazole and lansoprazole in erosive oesophagitis

Background: Omeprazole and lansoprazole are used to treat erosive oesophagi tis in the respective daily doses of 20 and 30 mg. Aim: To investigate, by meta-analysis, whether treatment with lansoprazole 30 mg increases erosive oesophagitis healing rates over omeprazole 20 mg. Methods: We searched for randomized, double-blind trials comparing omeprazo le 20 mg and lansoprazole 30 mg in endoscopically diagnosed erosive oesopha gitis. After assessing for homogeneity, non-heterogeneous trials were combi ned and pooled healing rates derived. We calculated the relative benefit in crease, absolute benefit increase and number needed to treat. Results: Six trials without significant heterogeneity met predetermined inc lusion criteria. By per protocol analysis, pooled healing rates for omepraz ole 20 mg and lansoprazole 30 mg were, respectively, 74.7% and 77.7% after 4 weeks and 87.0% and 88.7% after 8 weeks. The corresponding figures by int ention-to-treat analysis were 70.8% and 72.7% after 4 weeks and 81.8% and 8 3.3% after 8 weeks. In each analysis the absolute benefit increase for lans oprazole was small and its 95% confidence interval encompassed zero. Conclusion: Lansoprazole 30 mg produces healing rates in erosive oesophagit is that are not statistically significantly different to those of omeprazol e 20 mg.