Fourteen RCTs (n=1,671) were included.
The Jadad scores ranged from 2 to 5 out of a possible 5. Six trials scored 2 points and were considered to be of a low quality.
Eradication rates. For patients who had not been previously treated for H. pylori infection, probiotics were associated with significantly increased eradication rates compared with no probiotics: 83.6% versus 74.8% (OR 1.84, 95% CI: 1.34, 2.54), based on ITT analysis of 11 studies (n=1,074). No evidence of statistical heterogeneity was detected (p=0.65; I-squared 0). The results were similar for the per-protocol analysis and after the exclusion of low-quality studies. For patients who had previously failed treatment for H. pylori infection, probiotics were associated with significantly increased eradication rates compared with no probiotics: 88.5% versus 76.0% (OR 2.47, 95% CI: 1.16, 5.29), based on 2 studies (n=208). Probiotics were associated with a significant increase in eradication rates among patients who were symptomatic at baseline, but not in patients who were symptom-free at baseline. Probiotics were associated with a significant increase in eradication rates among adults and children. Probiotics containing Lactobacillus were also associated with a significant increase in eradication rates (based on 4 studies), but there was no significant difference in eradication rates for combinations of probiotics (5 studies), Bacillus clausii (1 study) or Clostridium butyrium (1 study) compared with no probiotics.
Side-effects.
Probiotics were associated with significantly decreased total side-effects compared with no probiotics: 24.7% versus 38.5% (OR 0.44, 95% CI: 0.30, 0.66). Probiotics were associated with significantly decreased diarrhoea (OR 0.34, 95% CI: 0.22, 0.52), epigastric pain (OR 0.62, 95% CI: 0.39, 0.97), nausea (OR 0.58, 95% CI: 0.38, 0.88) and taste disturbance (OR 0.38, 95% CI: 0.17, 0.85).
The funnel plot was slightly asymmetrical but Egger's test was not statistically significant, suggesting the absence of substantial publication bias.