Six relevant RCTs were identified (n=359, range nine to 98). All six RCTs were placebo-controlled. Five RCTs were double-blind. Two studies were in children (n=150) and four studies were in adults (n=209). LA1 was used in two studies and LGG in four studies; both the studies in children used LGG.
Relapse rates:
Lactobacilli administration compared to placebo did not significantly affect overall relapse rates. There was some heterogeneity among the studies (I2=43.6%). A sensitivity analysis found no difference in overall relapse rates after the omission of one unpublished study. Funnel plots showed evidence for publication bias.
Subanalyses that assessed studies in adults and children separately found no significant benefit for Lactobacilli. There was significant heterogeneity (I2=43.9%) for the meta-analysis in adults and no heterogeneity for the meta-analysis in children (I2=0%).
Subanalyses for the different Lactobacilli subtype interventions gave a significant benefit in relapse rate for placebo compared to LGG (RR 1.68, 95% CI 1.07 to 2.64, I2=0%), but no significant difference in relapse rate between LA1 and placebo.
Three RCTs (all in adults) reported endoscopic relapse rates. Meta-analysis showed no significant benefit for administration of Lactobacilli versus placebo and little heterogeneity (I2=0%).
Adverse events:
There was no significant difference in the number of adverse events for Lactobacilli administration versus placebo (five RCTs).