Six RCTs (n=13,572) were included. Data from 2 excluded RCTs (n=1,228) were also included in some analyses.
Esomeprazole (40 mg) showed a significant improvement in the healing rates of reflux oesophagitis at both 4 weeks (RR 0.92, 95% CI: 0.90, 0.94, p<0.00001) and 8 weeks (RR 0.95, 95% CI: 0.94, 0.97, p<0.00001) in comparison with combined standard-dose PPIs. However, significant heterogeneity was detected at both 4 and 8 weeks; further analysis suggested that the majority of this heterogeneity was due to the inclusion of the third largest trial (n=2,425). Subgroup analyses showed that 40 mg esomeprazole was associated with a significant improvement in healing rates in comparison with 30 mg lansoprazole at 4 weeks (RR 0.95, 95% CI: 0.95, 0.98, p<0.001) and 8 weeks (RR 0.96, 95% CI: 0.94, 0.99, p<0.0009), with 20 mg omeprazole at 4 weeks (RR 0.88, 95% CI: 0.85, 0.91, p<0.00001) and 8 weeks (RR 0.93, 95% CI: 0.91, 0.95, p<0.00001), and with 40 mg pantoprazole at 4 weeks (RR 0.93, 95% CI: 0.89, 0.96, p=0.0002) and 8 weeks (RR 0.97, 95% CI: 0.95, 1.00, p=0.03). No trials compared 20 mg rabeprazole with 40 mg esomeprazole.
Sensitivity analyses.
The results remained similar when 2 excluded RCTs with limited grades of oesophagitis were included, and when a random-effects model was used.
Baseline grade of oesophagitis.
At both 4 and 8 weeks, esomeprazole showed a significant additional benefit over standard-dose PPIs for the more severe grades of oesophagitis. This effect increased in size with increasing severity of the disease.
No significant publication bias was detected at either 4 or 8 weeks.