Seventeen studies were included in the review. Three of the studies were published in a full report, whilst the remainder were abstracts only. The median number of participants per treatment arm was 30 (range: 15-150).
Eradication rates in an ITT analysis ranged from 51 to 94%.
RBC-A-C versus PPI-A-C (11 studies):
Pooled ITT eradication rates were 78.5% (576 out of 734; range: 51- 94%) for RBC-based triple therapy, compared to 78.7% (605 out of 769; range: 60-88%) for PPI-based triple therapy. The difference between the pooled eradication rate was not statistically significant (weighted difference 0.8%, 95% CI: -3.1, 4.7).
Pooled PP eradication rates (8 out of 11 studies) were 85.8% (327 out of 381; range: 66-98%) for RBC-based triple therapy, compared to 82.9% (344 out of 415, range: 62-95%) for PPI-based triple therapy. The difference between the pooled eradication rate was not statistically significant (weighted difference 3.3%, 95% CI: -0.8, 7.4). Drop-out rates were 10.4% (44 out of 425) for RBC-A-C and 10.0% (46 out of 461) for PPI-A-C.
RBC-I-C versus PPI-I-C (7 studies):
Pooled ITT eradication rates were 85.6% (338 out of 395; range: 70-95%) for RBC-based triple therapy, compared to 78.9% (355 out of 442; range: 70-93%) for PPI-based triple therapy. The difference between the pooled eradication rate was statistically significant in favour of RBC-I-C (weighted difference 6.4%, 95% CI: 1.6, 11.1).
Pooled PP eradication rates (4 out of 7 studies) were 91.5% (161 out of 176; range: 91-93%) for RBC-based triple therapy, compared to 86.9% (152 out of 175; range: 77-92%) for PPI-based triple therapy. The difference between the pooled eradication rate was not statistically significant (weighted difference 2.0%, 95% CI: -4.1, 8.0). Drop-out rates were 12.9% (26 out of 202) for RBC-I-C and 12.9% (26 out of 201) for PPI-I-C.
No definite conclusions could be made about the impact of metronidazole or clarithromycin resistance since only three studies performed a formal resistance analysis.
No serious side-effects were reported , and drop-out rates were equal for the two regimens.