Eighteen RCTs (total number of participants not given) met the inclusion criteria. However, a further 6 RCTs were excluded from the primary analysis due to a quality score of 2 or lower. Twelve RCTs were included in the primary analysis (number of participants not reported).
Jadad quality scores: six RCTs scored 2 points or less, five scored 3 points, three scored 4 points, and four scored 5 points.
Endoscopic healing rates of esomeprazole (40 mg) compared with omeprazole (20 mg): the RR was 1.14 (95% CI: 1.10, 1.18) at 4 weeks (n=3) and 1.08 (95%CI: 1.05, 1.10) at 8 weeks (n=3).
The endoscopic healing rates of other PPIs compared with omeprazole (20 mg) were as follows:
for lansoprazole (30 mg) compared with omeprazole, the RR was 1.02 (95% CI: 0.97, 1.08) at 4 weeks (n=5) and 1.01 (95% CI: 0.97, 1.06) at 8 weeks (n=4);
for pantoprazole (40 mg) compared with omeprazole, the RR was 0.99 (95% CI: 0.91, 1.07) at 4 weeks (n=3) and 0.98 (95% CI: 0.93, 1.04) at 8 weeks (n=3);
for rabeprazole (20 mg) compared with omeprazole, the RR was 1.00 (95% CI: 0.87, 1.14) at 4 weeks (n=1) and 0.98 (95% CI: 0.91, 1.05) at 8 weeks (n=1).
Pooling the data for lansoprazole, pantoprazole and rabeprazole compared with omeprazole showed no statistically-significant difference in treatment at 4 weeks (RR 1.01, 95% CI: 0.97, 1.06) and at 8 weeks (RR 1.07, 95% CI: 1.01, 1.13).
For the comparison of esomeprazole with omeprazole, the chi-squared test indicated significant heterogeneity; there was no evidence of heterogeneity in any of the other comparisons. The random-effects calculations made no significant difference to the results stated above for the fixed-effect calculations.
The sensitivity analysis, which was performed to include all 18 trials regardless of quality score, did not make a significant difference to any of the comparisons carried out in the primary analysis.