Eleven studies met the inclusion criteria. Two studies were subdivided into two parts because they analysed four branches of treatment with different durations of eradication therapy. The total number of participants included was 2,391. Seven studies examined lansoprazole (n=1,414), 3 pantoprazole (n=555), 2 omeprazole (n=251) and 1 rabeprazole (n=240).
The quality of the included articles ranged from 0.24 ('acceptable') to 0.74 ('well designed').
The pooled analysis showed that double doses achieved higher H. pylori eradication rates than single doses in both the ITT (83.9% versus 77.7%; OR 1.51, 95% CI: 1.23, 1.85, P<0.01) and per protocol (88.8% versus 80.5%; OR 1.96, 95% CI: 1.51, 2.47, P<0.01) analyses. Heterogeneity was of borderline significance (P<0.10), but diminished with the exclusion of one small study (P=0.23); the main results remained unchanged.
A double dose of PPI was also found to be superior to a single dose in the subgroup analysis of lansoprazole (ITT analysis OR 1.52, 95% CI: 1.16, 2.00, P<0.01); pantoprazole (ITT analysis OR 1.76, 95% CI: 1.19, 2.58, P<0.01); 7-day therapy (ITT analysis OR 1.37, 95% CI: 1.09, 1.72, P not stated); and the combination of clarithromycin (500 mg) and amoxicillin (1 g) antibiotics (5 studies; ITT analysis OR 1.73, 95% CI: 1.38, 2.18, P not stated). There was no significant difference (2 studies) between single and double doses of PPIs for the combination of clarithromycin (250 mg) and metronidazole (500 mg) antibiotics.