A yeast fermentate improves gastrointestinal discomfort and constipation by modulation of the gut microbiome: results from a randomized double-blind placebo-controlled pilot trial

Abstract

Background: Constipation and symptoms of gastrointestinal discomfort such as bloating are common among otherwise healthy individuals, but with significant impact on quality of life. Despite the recognized contribution of the gut microbiome to this pathology, little is known about which group(s) of microorganism(s) are playing a role. A previous study performed in vitro suggests that EpiCor® fermentate has prebiotic-like properties, being able to favorably modulate the composition of the gut microbiome. Therefore, the aim of this study was to investigate the effects of EpiCor fermentate in a population with symptoms of gastrointestinal discomfort and reduced bowel movements and to evaluate its effect at the level of the gut microbiome.

Methods: This pilot study was performed according to a randomized, double-blind, placebo-controlled parallel design. Eighty subjects with symptoms of gastrointestinal discomfort and constipation were allocated to one of two trial arms (placebo or EpiCor fermentate). Randomization was done in a stratified manner according to symptom severity, resulting in two subgroups of patients: severe and moderate. Daily records of gastrointestinal symptoms were assessed on a 5-point scale, and also stool frequency and consistency were documented during a 2-week run-in and a 6-week intervention phases. Averages over two-week intervals were calculated. Constipation-associated quality of life and general perceived stress were assessed at baseline and after 3 and 6 weeks of intervention. Fecal samples were also collected at these same time points.

Results: EpiCor fermentate led to a significant improvement of symptoms such as bloating/distension (p = 0.033 and p = 0.024 after 2 and 4 weeks of intervention, respectively), feeling of fullness (p = 0.004 and p = 0.023 after 2 and 4 weeks of intervention, respectively) and general daily scores (p = 0.046 after 2 weeks of intervention) in the moderate subgroup. A significant improvement in stool consistency was observed for the total population (p = 0.023 after 2 weeks of intervention) as well as for the severe subgroup (p = 0.046 after 2 weeks of intervention), and a nearly significant increase in stool frequency was detected for the total cohort (p = 0.083 and p = 0.090 after 2 and 4 weeks of intervention, respectively). These effects were accompanied by an improvement in constipation-associated quality of life and general perceived stress, particularly in the moderate subgroup. Members of the families Bacteroidaceae and Prevotellaceae, two groups of bacteria that have been previously reported to be deficient in constipated patients, were found to increase with EpiCor fermentate in the severe subgroup. In the moderate subgroup, a significant increase in Akkermansia muciniphila was observed.

Conclusions: Despite the relatively low dose administered (500 mg/day), particularly when comparing to the high recommended doses for prebiotic fibers, EpiCor fermentate was able to modulate the composition of the gut microbiome, resulting in improvement of constipation-associated symptoms. Conversely, the reported increase in bowel movements may have altered the gut microbial community by increasing those groups of bacteria that are better adapted to a faster gastrointestinal transit time.

Trial registration: NCT03051399 at ClinicalTrials.gov. Retrospectively registered. Registration date: 13 February 2017.

Keywords: Bacteroides; Bacteroidetes; Constipation; EpiCor fermentate; GI transit time; Gastrointestinal discomfort; Gut microbiome; Human study; Prevotella; Saccharomyces cerevisiae.

Conflict of interest statement

Ethics approval and consent to participate

Ethical approval was obtained from the Independent Ethics Committee (IEC) / Institutional Review Board (IRB) from the Ghent University Hospital, De Pintelaan 185, 9000 Ghent, Belgium. All participants signed an informed consent prior to inclusion in the study.

Consent for publication

Not applicable.

Competing interests

This study was commissioned by Embria Health Sciences (USA) to ProDigest BVBA (Belgium), which subcontracted the D.R.U.G. Unit at the Ghent University hospital (Belgium) to perform the study. The project was coordinated by ProDigest but all parties involved took part on the discussion about study design, inclusion and exclusion criteria and outcome measurements. However, during study realization and data analysis, both ProDigest and D.R.U.G. Unit staff members involved in the study were blinded. Reporting was performed by ProDigest. Embria Health Sciences (USA) owns the proprietary product used in this study. Massimo Marzorati is one of the co-founders of ProDigest and Björn Winkens was subcontracted by ProDigest.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures

Fig. 1

Schematic diagram of the study flow (based on CONSORT 2010 guidelines). Legend: GI, gastrointestinal; PAC-QOL, Patient Assessment of Constipation Quality of Life; PSS, Perceived Stress Scale. a Gastrointestinal symptoms were assessed by asking the volunteers to grade daily in the evening the average severity over the previous 24 h on a 5-point scale from 0 (not at all) to 4 (extremely) for the following 5 GI characteristics: bloating/distension, passage of gas, GI rumbling, feeling of fullness and abdominal discomfort

Fig. 2

Schematic diagram of primary and secondary objectives and instruments used for data/sample collection. Legend: GI, gastrointestinal; PAC-QOL, Patient Assessment of Constipation Quality of Life; PSS, Perceived Stress Scale. a Gastrointestinal symptoms were assessed by asking the volunteers to grade daily in the evening the average severity over the previous 24 h on a 5-point scale from 0 (not at all) to 4 (extremely) for the following 5 GI characteristics: bloating/distension, passage of gas, GI rumbling, feeling of fullness and abdominal discomfort. b Stool consistency was recorded according to the Bristol Stool Form Scale

Fig. 3

Mean differences between EpiCor and placebo on gastrointestinal symptoms on the three cohorts. Legend: The enlisted gastrointestinal symptoms are BD (bloating/distension), GAS (passage of gas), RUM (GI rumbling), FF (feeling of fullness) and AD (abdominal discomfort). A daily total score (DTS) calculated as the sum of all items recorded each day is also shown. A linear mixed model analysis that takes into account the differences between groups at baseline was used (p-values <0.05 are depicted in bold text; p-values <0.1 are depicted in regular text)

Fig. 4

Mean differences between EpiCor and placebo on stool frequency (a) and consistency (b) on the three cohorts. Legend: A linear mixed model analysis that takes into account the differences between groups at baseline was used (p-values <0.05 are depicted in bold text; p-values <0.1 are depicted in regular text)

Fig. 5

Mean differences between EpiCor and placebo on constipation-associated quality of life (a) and perceived stress (b) on the three cohorts. Legend: The enlisted Patient Assessment of Constipation Quality of Life (PAC-QOL) items are PhD (physical discomfort), PsD (psychosocial discomfort), W&C (worries and concerns) and SAT (satisfaction). An instrument total score (ITS) calculated as the average of all 28 items recorded at each visit is also shown. PSS relates to the Perceived Stress Scale instrument. A linear mixed model analysis that takes into account the differences between groups at baseline was used (no p-values <0.05 were found; p-values <0.1 are depicted in regular text)

Fig. 6

Bacteroidetes and Firmicutes relative abundances within the total cohort (a) and the two subgroups, severe (b) and moderate (c) that have been treated either with placebo or EpiCor. Each box represents median (50th percentile) and interquartile range (25th and 75th percentiles). The symbol (+) represents the mean. The outliers are indicated as dots (Tukey method). Significant (p < 0.05) and nearly significant (p < 0.1) p-values are also indicated within the boxplots (two-way repeated measures ANOVA with Dunnett’s multiple comparison’s test against V1). The inner plots (a*, b* and c*) correspond to the calculated Firmicutes/Bacteroidetes (F/B) ratio at each visit within the two treatment groups. Legend: V1, V2 and V3 correspond to visit 1 (baseline), visit 2 (3-weeks after treatment) and visit 3 (6-weeks after treatment), respectively

Fig. 7

Principal component analysis (47.4%) of the relative fold-changes calculated for all taxonomic groups (family and genus levels). Each dot represents a treated group (either EpiCor or placebo) for all three cohorts (total cohort, severe and moderate). The first component (PC1) accounts for nearly 28% of the variance, and the second component (PC2) for nearly 20%. legend: E, Epicor; P, placebo; V, visit

Fig. 8

Variables (taxa) correlation plot. Each vector represents a taxonomic group (family or genus levels). Shorter vectors only slightly contribute for differentiation between groups. Longer vectors have a bigger weight in groups’ differentiation. Those variables that mostly contribute the first component (PC1) are indicated in bold text

Fig. 9

Hierarchical clustering heatmap of the Log2 relative fold-changes calculated for all taxonomic groups (family and genus levels). (a) Both taxa and treatment groups were subject to HCL analysis. (b) Shows a detail of the results for the groups’ HCL. The KMC analysis roughly revealed the presence of 7 major clusters (C1-C7) based on similarly of taxa relative fold-changes. Legend: C, cluster; E, Epicor; M, moderate subgroup, P, placebo; S, severe subgroup; TC, total cohort; V, visit