Clinical, gut microbial and neural effects of a probiotic add-on therapy in depressed patients: a randomized controlled trial

Abstract

A promising new treatment approach for major depressive disorder (MDD) targets the microbiota-gut-brain (MGB) axis, which is linked to physiological and behavioral functions affected in MDD. This is the first randomized controlled trial to determine whether short-term, high-dose probiotic supplementation reduces depressive symptoms along with gut microbial and neural changes in depressed patients. Patients with current depressive episodes took either a multi-strain probiotic supplement or placebo over 31 days additionally to treatment-as-usual. Assessments took place before, immediately after and again four weeks after the intervention. The Hamilton Depression Rating Sale (HAM-D) was assessed as primary outcome. Quantitative microbiome profiling and neuroimaging was used to detect changes along the MGB axis. In the sample that completed the intervention (probiotics N = 21, placebo N = 26), HAM-D scores decreased over time and interactions between time and group indicated a stronger decrease in the probiotics relative to the placebo group. Probiotics maintained microbial diversity and increased the abundance of the genus Lactobacillus, indicating the effectivity of the probiotics to increase specific taxa. The increase of the Lactobacillus was associated with decreased depressive symptoms in the probiotics group. Finally, putamen activation in response to neutral faces was significantly decreased after the probiotic intervention. Our data imply that an add-on probiotic treatment ameliorates depressive symptoms (HAM-D) along with changes in the gut microbiota and brain, which highlights the role of the MGB axis in MDD and emphasizes the potential of microbiota-related treatment approaches as accessible, pragmatic, and non-stigmatizing therapies in MDD. Trial Registration: www.clinicaltrials.gov , identifier: NCT02957591.

Fig. 1. Trajectories and change scores of depressive symptoms in the probiotics and placebo group.

Mean trajectory of scores on the Hamilton Scale for Depression (HAM-D) from baseline to post-intervention (week 4) and follow-up assessment (week 8) in A the intention-to-treat sample (ITT) and B the modified intention-to-treat sample (mITT). Error bars indicate 95% confidence interval (based on a bootstrap). C Change scores from baseline to post-intervention in the mITT sample and D from baseline to follow-up.

Fig. 2. Enterotype distribution within time points.

Group differences at baseline (A), post-intervention (B), and follow-up (C) per study groups.

Fig. 3. Alpha-diversity comparisons between study groups at all three time points.

A Inversed Simpson index, B observed richness, C Pielou’s evenness, and D Shannon index.

Fig. 4. Significantly associated taxa to time and behavioral measures.

(A) Mean abundance of the taxa showing a significant increase over the study intervention in the placebo and probiotic group, respectively (MEM ANOVA p-FDR < 0.05 and Wald test p-FDR < 0.05). ^•<0.1, *<.05, **<0.01, *^**<0.001 p-FDR values after being adjusted for fecal moisture, sex, BMI, and age. Increases are indicated in relation to baseline values. (B) Line plots of the taxa-abundance of those taxa whose abundance differed significantly over time in the two study groups. The bottom of the panel shows ANOVA results for the time*group interaction, *p < .05 Wald test adjusted. (C) Mixed-effect β coefficients of the significantly associated taxa with the clinical measures (HAM-D, BDI, GSRS, STAI1) over time. The significance of the taxa was determined by using a negative binomial mixed-effect model; ^•<0.1, *<0.05, **<0.01 ANOVA adjusted p-values after being adjusted for fecal moisture, sex, BMI, and age. The gradient color indicates the negative binomial mixed-effect model coefficient. HAM-D = Hamilton Rating Scale for Depression; BDI = Beck Depression Inventory; STAI1 = State-Trait Anxiety Inventory 1; GSRS = Gastrointestinal Symptom Rating Scale.

Fig. 5. fMRI results: activation pattern during neutral face processing.

Blue: decreased activation after the 4-week intervention in the probiotics group during neutral face processing. Red: increased activation in depressive patients (i.e. probiotics and placebo groups combined at baseline) compared to healthy controls during neutral face processing.