Add-On Bifidobacterium Bifidum Supplement in Children with Attention-Deficit/Hyperactivity Disorder: A 12-Week Randomized Double-Blind Placebo-Controlled Clinical Trial

Abstract

We conducted a 12-week randomized double-blind placebo-controlled clinical trial to investigate the potential impact of Bifidobacterium bifidum (Bf-688) supplementation on attention-deficit/hyperactivity disorder (ADHD). Children with ADHD who were already receiving a stable dose of methylphenidate (MPH) treatment were enrolled and were randomly assigned to two groups: one receiving add-on Bf-688 (daily bacterial count of 5 × 10⁹ CFUs) (n = 51) and the other receiving a placebo (n = 51). All participants underwent assessments using Conners' Continuous Performance Test (CPT) and Conners' Continuous Auditory Test of Attention (CATA). Additionally, fecal samples were collected at the beginning of the trial (week 0) and at the endpoint (week 12). Remarkably, the group receiving Bf-688 supplementation, but not the placebo group, exhibited significant improvements in omission errors in CPT as well as Hit reaction time in both CPT and CATA. Gut microbiome analysis revealed a significant increase in the Firmicutes to Bacteroidetes ratio (F/B ratio) only in the Bf-688 group. Furthermore, we identified significant negative correlations between N-Glycan biosynthesis and Hit reaction time in both CPT and CATA. Our results demonstrate that the probiotic Bf-688 supplement can enhance neuropsychological performance in children with ADHD, possibly by altering the composition of the gut microbiota, ultimately leading to reduced N-Glycan biosynthesis.

Keywords: ADHD; Bifidobacterium; gut–brain axis; microbiome; probiotic; psychobiotics.

Figure 1

CONSORT diagram for study procedure and flowchart. During the second visit at week 4, three individuals from the Bf-688 group discontinued their participation in the trial. This included two participants who found the taste of the probiotics unacceptable and one who ceased taking ADHD medication. In the placebo group, one participant withdrew from the trial due to an aversion to the taste of the placebo, and another participant withdrew their consent.

Figure 2

ADHD symptoms, body weight, BMI, and adverse effects of children with ADHD pharmacotherapy over the course of a 12-week double-blind placebo-controlled trial. (A) SNAP-IV Inattention symptoms; (B) SNAP-IV Hyperactivity symptoms; (C) ADHD-RS Inattention symptoms; (D) ADHD-RS Hyperactivity symptoms; (E) Body weight; (F) Body mass index; (G) SERS total scores; and (H) SERS GI symptoms. Bf-688, Bifidobacterium Bifidum plus ADHD medication for 12 weeks; Placebo, placebo plus ADHD medication for 12 weeks; SNAP-IV, the Swanson, Nolan, and Pelham Rating Scale; ADHD-RS, ADHD rating scale; SERS, Barkley’s Side Effects Rating Scale; SERS GI (gastrointestinal) symptoms, severity of stomachache, and loss of appetite. * p < 0.05, ** p < 0.01, *** p < 0.001 compared with baseline data.

Figure 3

Patients’ visual and auditory attention of children with ADHD pharmacotherapy at the baseline (week 0) and the endpoint (week 12). CPT, Conners’ Continuous Performance Test; CATA, Conners’ Continuous Auditory Test of Attention; Hit reaction time (RT). * p < 0.05, ** p < 0.01, *** p < 0.001 compared with baseline data.

Figure 4

Analysis of abundant bacteria (A) Shannon and Chao-1 analyses revealed significant differences in both the placebo and Bf-688 groups before and after probiotic supplementation. (B) PCoA plots of beta diversity for weighted and unweighted UniFrac distances. Ellipses represent the 95% confidence intervals for each group. Colors are assigned based on group allocation: red for the Bf-688 group after supplementation, green for the Bf-688 group before supplementation, blue for the placebo group after supplementation, and purple for the placebo group before supplementation. (C) Bar charts depict the average abundances of phylum, genus, and species-level ASVs across the four different samples studied: Bf-688 group after supplementation, Bf-688 group before supplementation, placebo group after supplementation, and placebo group before supplementation. “Other” refers to all species representing less than 10% abundance after the 10th most abundant species.

Figure 5

Correlations between CPT Hit reaction time (CPT Hit RT), CATA Hit reaction time (CATA Hit RT), and N-Glycan Biosynthesis In this extended error bar chart, we compare the predicted KEGG functional data based on four different samples: the Bf-688 group before and after supplementation, and the placebo group before and after supplementation. Welch’s t-tests were applied, and only predicted functions with p < 0.05 between the two groups are displayed. (A) The left bar chart represents the mean proportions of each KEGG pathway, while the right dot plot illustrates the differences in mean proportions between the two compared sample groups, along with their respective p-values. We employed the PICRUSt to predict the potential mechanisms through which gut bacteria might be involved in these comparisons. (B) We present the correlations between microbial communities, pathways, and clinical parameters, including the F/B ratio, SERS GI symptoms, CPT Hit reaction time, and CATA Hit reaction time. * p-values (C) Significant correlations were observed between CPT Hit RT, CATA Hit RT, and N-Glycan Biosynthesis.