A Meta-analysis of Randomized Placebo-controlled Trials on the Effects of Probiotics for Autism Spectrum Disorders

Objective: The current meta-analysis aimed at updating evidence regarding the therapeutic potential of probiotics against autism spectrum disorder (ASD) symptoms and identifying outcome confounders. Methods: Following PRISMA guidelines, randomized placebo-controlled trials retrieved from PubMed, Embase, Cochrane CENTRAL, and ScienceDirect were analyzed for effect size of primary outcomes (i.e., overall behavioral symptom changes) expressed as standardized mean difference (SMD) and odds ratios (ORs) for continuous and categorical variables, respectively, with 95% confidence interval (CI). Results: Meta-analysis of eight studies (465 participants, mean age: 6.03, range: 1.5−45, female: 17.4%) revealed total ASD symptom improvements in the probiotic group compared to placebos (SMD: −0.19, 95% CI: −0.38 to −0.01, p = 0.04, I2 = 0%, eight studies, 465 participants, certainty of evidence [COE]: low). Subgroup analyses showed probiotics- associated improvements only in studies recruiting predominantly preschool children (SMD: −0.30, 95% CI: −0.59 to −0.01, p = 0.04, three studies, 191 participants) or those using probiotics for over three months (SMD: −0.39, 95% CI: −0.73 to −0.06, p = 0.04, three studies, 144 participants) without difference between multipleand single-strain probiotics. No difference was noted in overall dropouts between individuals treated with probiotics and those taking placebos (OR: 0.98, p = 0.94, eight studies, I2 = 0%, 464 participants, COE: low). Despite a low risk of bias in most studies, COE was deemed low from limited trials and inconsistencies on sensitivity analysis. Conclusion: The current study showed an association between probiotics use and an improvement in ASD symptoms, mainly in those aged below six or over three-month treatments. More large-scale investigations are warranted to support our findings.