Alterations in fecal microbiota composition by probiotic supplementation in healthy adults: a systematic review of randomized controlled trials
© Kristensen et al. 2016
Received: 15 January 2016
Accepted: 8 April 2016
Published: 10 May 2016
The effects of probiotic supplementation on fecal microbiota composition in healthy adults have not been well established. We aimed to provide a systematic review of the potential evidence for an effect of probiotic supplementation on the composition of human fecal microbiota as assessed by high-throughput molecular approaches in randomized controlled trials (RCTs) of healthy adults.
The survey of peer-reviewed papers was performed on 17 August 2015 by a literature search through PubMed, SCOPUS, and ISI Web of Science. Additional papers were identified by checking references of relevant papers. Search terms included healthy adult, probiotic, bifidobacterium, lactobacillus, gut microbiota, fecal microbiota, intestinal microbiota, intervention, and (clinical) trial. RCTs of solely probiotic supplementation and placebo in healthy adults that examined alteration in composition of overall fecal microbiota structure assessed by shotgun metagenomic sequencing, 16S ribosomal RNA sequencing, or phylogenetic microarray methods were included. Independent collection and quality assessment of studies were performed by two authors using predefined criteria including methodological quality assessment of reports of the clinical trials based on revised tools from PRISMA/Cochrane and by the Jadad score.
Seven RCTs investigating the effect of probiotic supplementation on fecal microbiota in healthy adults were identified and included in the present systematic review. The quality of the studies was assessed as medium to high. Still, no effects were observed on the fecal microbiota composition in terms of α-diversity, richness, or evenness in any of the included studies when compared to placebo. Only one study found that probiotic supplementation significantly modified the overall structure of the fecal bacterial community in terms of β-diversity when compared to placebo.
This systematic review of the pertinent literature demonstrates a lack of evidence for an impact of probiotics on fecal microbiota composition in healthy adults. Future studies would benefit from pre-specifying the primary outcome and transparently reporting the results including effect sizes, confidence intervals, and P values as well as providing a clear distinction of between-group and within-group comparisons.
The human gut microbiota refers to the microbes that reside inside the gut and partake in several functions beneficial to the host including fermentation of otherwise indigestible dietary fibers and other food items , synthesis of vitamins and amino acids , prevention of pathogen colonization , maturation and regulation of the immune system , modulation of gastrointestinal hormone release, and regulation of brain behavior through bidirectional neuronal signaling as part of the gut-brain axis . The development of culture-indpendent, high-throughput molecular techniques has enabled the identification of previously unknown bacterial species, thereby providing novel insights into the compositional diversity and functional capacity of fecal microbiota. As a result, studies have suggested that disorders such as colorectal cancer, rheumatoid arthritis, type 2 diabetes, and obesity are associated with disease-specific dysbiotic shifts in fecal microbiota [6–11]. Consequently, in recent years the gut microbiota as a potential modifiable risk factor for disease development has received massive attention. One common approach applied to convey health benefits by way of modifying the gut microbiota has been the use of probiotic supplementation. Probiotics are defined as live microorganisms that, when administered in adequate amounts, confer a health benefit on the host in a safe and efficacious manner . Suggested mechanisms by which probiotics may benefit the gut environment and the health of the host include improvement of the intestinal barrier function through effects on the epithelium and mucus lining, production of anti-microbial substances, competition with pathogenic bacteria, and regulation of luminal acidity (reviewed in [13, 14]).
The therapeutic effect of probiotic supplementation has been studied in a broad range of diseases, particularly in regard to gastrointestinal and metabolic disorders where results have supported the potential use of probiotics as therapeutic agents (reviewed in [15, 16]). Common to both sets of disorders is a multitude of readily available, clinically relevant outcome measures (e.g. body mass index, fat mass, insulin resistance, severity of gastrointestinal symptoms) by which to measure treatment effect. The effect of probiotics in disease-free individuals is, however, not as easily assessed. Interpretation of an effect on the composition of fecal microbiota in healthy individuals may be particularly complicated due to the lack of an internationally accepted consensus definition of a normal or a healthy fecal microbial community [17, 18].
Terms such as ecological stability, idealized composition, or favorable functional profile have been suggested as hallmarks of a healthy gut microbiota . These are all very unspecific concepts and the compositional and functional characteristics of a healthy gut microbial community remain to be defined. Furthermore, an effect of probiotics on the composition of the gut microbiota is an intermediate outcome only and should be interpreted with caution in regard to implications for the health of the host. Despite these limitations, several probiotic interventions aiming to observe alterations in fecal microbiota composition have been performed in healthy adults [19–28]. Results from these studies have the potential to provide insights into the underlying mechanisms of probiotics and fecal microbiota. Currently, no systematic review has addressed the effects of probiotics on fecal microbiota composition using high-throughput metagenomic methods (i.e. phylogenetic microarrays, 16S ribosomal RNA (rRNA) sequencing, or shotgun metagenomic sequencing) in healthy adults. In the context of a billion dollar market for probiotic supplements  with products being marketed, in part, toward healthy individuals by stating effects on gastrointestinal health, alluding to the fecal microbial community, an overview of the current evidence is warranted.
The objective of the present systematic review was to explore in healthy adults the current evidence for an effect of probiotic supplementation compared to placebo on the composition of human fecal microbiota as assessed by high-throughput molecular approaches in randomized controlled trials (RCTs).
We undertook a systematic review of the possible effects of probiotic intervention on the composition of fecal microbiota in healthy adults. The available literature was identified and examined as a systematic review and not a meta-analysis due to the heterogeneity of the study designs and methods. The results are reported in accordance with the PRISMA statement guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) . The study followed an a priori established protocol.
The criteria for eligibility were healthy adults as study population, probiotics and placebo as intervention, alteration in composition of fecal microbiota assessed by shotgun metagenomic sequencing, 16S rRNA sequencing or phylogenetic microarray methods as the primary outcome, and RCT as the study design with no criteria on study duration. No limits were applied for language or time. Studies not exploiting the randomized controlled design and providing only within-group results (i.e. results before and after the intervention in the probiotic group only) were not included. Moreover, only studies assessing the overall bacterial ecology were included. Accordingly, studies investigating survival of the probiotic strains only were considered ineligible. Studies with interventions combining probiotics with other supplements (e.g. prebiotics, antibiotics, medications) were excluded. If studies had more than two arms, only the comparison of probiotics to placebo was considered. Studies examining both healthy and unhealthy participants were excluded.
Information sources, search strategy, and study selection
The identification of papers involved four sequential processes performed by two independent reviewers (NBK and TB). On 17 August 2015, a literature search was conducted through multiple electronic databases (PubMed, SCOPUS, and ISI Web of Science) to capture as many relevant citations as possible. The search phrase used was:
Healthy adult AND (probiotic OR bifidobac* OR lactobac*) AND (gut microbio* OR f*cal microbio* OR intestinal microbio*) AND (clinical trial OR intervention OR trial).
In PubMed, “Species” was limited to include only humans and “Article types” was limited to cover “Clinical trial,” “Comparative study,” “Controlled clinical trial,” “Journal article,” and “Randomized controlled trial.”
In ISI Web of Science, “Document types” was limited to contain “Article,” “Clinical trial,” “Other,” and “Abstract.”
In Scopus, “Source type and document type” was limited to comprise “Journals and article,” “Short survey,” and “Erratum.” For “Subject area,” “Agricultural and biological sciences,” “Nursing,” “Pharmacology,” “Toxicology and pharmaceutics,” “Environmental science,” “Veterinary,” “Chemistry,” and “Neuroscience” were excluded.
Data collection process
Independent data collection was performed by two authors (NBK and TB). Corresponding authors of the following studies were contacted in order to acquire missing information on allocation concealment or other measures of risk of bias: Lahti et al., Rampelli et al., Ferrario et al., Bjerg et al., Hanifi et al., and Simon et al. [19, 21–24, 27]. Unpublished information about blinding was obtained from Lahti et al., Bjerg et al., and Hanifi et al. [19, 21, 22] and the reason for excluding data from three participants from the intervention group (missing fecal samples) was obtained from Hanifi et al. .
Information extracted from each included RCT was: (1) participant characteristics (including age and gender); (2) intervention (including probiotic strain and dosage as well as mode of administration); (3) design (including study design and duration); and (4) outcome measure (including the microbiomics and intervention effect on overall fecal microbiota structure).
The methodological quality assessment of reports of the clinical trials was performed independently by NBK and TB using a three-item instrument (the Jadad score) that evaluates likelihood of bias in research reports . The three items evaluated by a five-point scale are quality of randomization, quality of blinding, and reasons for withdrawal/drop-out (0 = worst, 5 = best). Risk of bias was further assessed regarding concealment of randomization, early termination of trial, blinding of patients, healthcare providers, data collectors and outcome assessors, reporting of drop-out or withdrawal, selective outcome reporting, and other potential biases [32, 33].
Intervention effects on the overall fecal microbiota structure, that is, richness, abundance, evenness, α-diversity or compositional dissimilarity (β-diversity), were the primary measures of treatment effects.
A total of 1368 citations were identified through the search in PubMed, SCOPUS, and ISI Web of Science and an additional five were identified through checking the references of relevant papers. After the removal of duplicates, 1287 citations were left. NBK and TB screened the initial search results using abstracts and 1256 citations were excluded as irrelevant for one or more of the following reasons: animal study, meta-analysis/review, non-healthy or non-adult participants, no probiotic intervention, or no assessment of fecal microbiota composition. The full papers of the remaining 31 citations and references therein were assessed to select studies for inclusion using the abovementioned criteria, resulting in the exclusion of 24 studies due to one or a combination of the following reasons: no assessment of fecal microbiota composition, assessment of single-strain survival only, inclusion of non-healthy participants, non-randomized controlled design, provision of only within-group results, and combined intervention of probiotic with prebiotics or other foods. Following the selection process (Fig. 1), seven studies [19–24, 27] remained (five of which were identified by checking the references of relevant papers) and were included in the present systematic review.
Characteristics of the studies reviewed
Sample size calculation
Probiotic effects on fecal microbiota compared to placebo
Lahti et al. 2013 
25/25 (72 %)
L. rhamnosus GG ATCC53103 (1.55 × 1010 CFU) in 250 mL milk-based fruit drink
Double-blinded, parallel, two-armed, placebo (drink) controlled (21 days)
16S rRNA (regions V1 and V6); HITchip based characterization of >1000 microbial species-like phylotypes
↔ Composition of the fecal microbiota
↔ Stability of the fecal microbiota quantified by inter-individual and intra-individual correlations within and between time points
Rampelli et al. 2013 
32/32 (59 %)
B. longum Bar33 and L. helveticus Bar13 (109 CFU) in biscuit
Double-blinded, parallel, two-armed, placebo (biscuit) controlled (30 days)
16S rRNA (region unknown);
HTF-Microbi.Array based characterization of 31 phylogenetically related groups
↔ Relative abundance of 31 phylogenetically related groups
Ferrario et al. 2014 
22/34* (56 %)
L. paracasei DG (>2.4 × 1010 CFU) in capsules
Double-blinded, cross-over, placebo (capsules) controlled (two 28-day intervention periods with a 28-day wash-out)
16S rRNA (region V3) sequencing on Ion Torrent platform
Modified β diversity (with absolute distances higher for the probiotic than for the placebo treatments)
Bjerg et al. 2015 
20/64* (50 %)
L. casei W8® (1010 CFU) in capsules
Double-blinded, parallel, two-armed, placebo (capsules) controlled (28 days)
16S rRNA (regions V3 and V4) sequencing on Roche 454 pyrosequencing platform
↔ α- and β-diversity
Brahe et al. 2015 
34/58* (100 %)
L. paracasei F19 (9.4 × 1010 CFU) or flaxseed mucilage (10 g) in sachets
Single-blinded, parallel, three-armed, placebo (sachets) controlled (42 days)
on a SOLiD 5500×l platform
↔ Bacterial gene count (richness) and abundance of specific bacterial genes compared to placebo
Hanifi et al. 2015 
37/81* (52 %)
Bacillus subtilis R0179
(0.1 × 109, 1.0 × 109, and 10 × 109 CFU, respectively) in capsules
Double-blinded, parallel, four-armed, placebo (capsules) controlled (28 days)
16S rRNA (regions V1 to V3) pyrosequencing
↔ β-diversity and OTU based richness
Simon et al. 2015 
21/21 (52 %)
L. reuteri SD5865 (2 × 1010 viable cells) in capsules
Double-blinded, parallel, two-armed, placebo (capsules) controlled (28 days)
16S rRNA (regions V5 and V6) sequencing on Illumina MiSeq platform
↔ α- and β-diversity
Risk of bias
Assessment of risk of bias in the studies reviewed
Concealment of randomization
RCT stopped early
Healthcare providers blinded
Data collectors blinded
Outcome assessors blinded
Reporting drop-out or withdrawal
Other potential bias
Selective outcome reporting
Lahti et al. 2013 
Rampelli et al. 2013 
Ferrario et al. 2014 
Bjerg et al. 2015 
Brahe et al. 2015 
Hanifi et al. 2015 
Simon et al. 2015 
Results of individual studies
In terms of richness, evenness, or α-diversity measures, no effects were observed on the fecal microbiota composition in any of the included studies when compared to placebo and only in the study by Ferrario et al.  was it found that probiotic treatment significantly modified the compositional dissimilarity (β-diversity).
In the study by Lahti et al. , the temporal stability of fecal microbiota, quantified by the correlation of the fecal microbiota profiles among three time points, did not differ between the probiotic (L. rhamnosus GG ATCC53103) and the placebo group.
In the study by Rampelli et al. , there was no effect of probiotic supplementation (B. longum Bar33 and L. helveticus Bar13) on the relative abundance of 31 phylogenetically related groups when compared to placebo. In the same study, the effect of probiotic supplementation on age-related dysbiosis was also evaluated. The probiotic intervention reverted an age-related increase of Clostridium cluster Xi, C. difficile, C. perfringens, Enterococcus Faecium, and Campylobacter when comparing the probiotic and the placebo group to a common reference of eight young, healthy adults; but no direct comparison was made between treatment groups.
In the study by Ferrario et al. , the α-diversity reported as Chao1 and Shannon coefficients and number of detected genera did not change as a result of the probiotic intervention (L. paracasei DG) when compared to placebo. Yet, the β-diversity between the probiotic and the placebo group was modified with absolute distances significantly higher for the probiotic than for the placebo treatments when assessed by principal coordinate analysis (PCoA) of weighted UniFrac distances. Accordingly, the relative abundance of Proteobacteria (P = 0.006) and Clostridiales genus Coprococcus (P = 0.009) were increased and the Clostridiales genus Blautia (P = 0.036) was decreased in the probiotic group when compared to placebo. Additionally, analyses of predicted functional profiles revealed changes in eight Kyoto Encyclopedia of Genes and Genomes modules related to bacterial pathways in membrane transport, amino acid metabolism, energy metabolism, and metabolism of cofactors and vitamins (P < 0.05).
In the study by Bjerg et al. , the β-diversity was not affected by the probiotic intervention (L. casei W8®) compared to placebo when assessed by PCoA of species and genus level Operational Taxonomic Unit (OTU) based UniFrac distances. Furthermore, no difference in α-diversity (Chao1 and Shannon index) or species richness was observed between the probiotic and the placebo group.
In the study by Brahe et al. , fecal microbiota was assessed by shotgun-sequencing-based metagenomics. The bacterial gene count (richness) did not change within the probiotic group (L. paracasei F19) compared to placebo. Alterations in the abundance of individual bacterial genes (2493 genes assigned to two metagenomic species) were observed after the intervention in the probiotic group. However, fewer alterations were observed in the intervention group compared to the placebo group (7436 genes assigned to six metagenomic species). Yet, no direct comparison between the groups is explicitly stated.
In the study by Hanifi et al. , no difference in compositional dissimilarity (β-diversity) between the treatment groups (Bacillus subtilis R0179 in different doses) and placebo was shown when analyzed using PCoA based on the UniFrac metric. Sequence reads binned to multiple OTUs assigned to the genus Ruminococcus increased in the probiotic group (with the highest dose (10 × 109 CFU, Table 2) compared to placebo (P < 0.01).
In the study by Simon et al. , the overall composition of fecal microbiota was unaffected by probiotic supplementation (L. reuteri SD5865) both in terms of α- (Chao1, Shannon, and Simpson indices) and β- (Bray-Curtis, Morisita-Horn, and weighted UniFrac) diversity.
Overall, this systematic review demonstrates that there is no convincing evidence for consistent effects of probiotics on fecal microbiota composition in healthy adults.
No effects were observed on the fecal microbiota composition in terms of α-diversity, richness, or evenness in any of the included studies when compared to placebo. Only in the paper by Ferrario et al.  was it reported that probiotic supplementation significantly modified the overall structure of the fecal bacterial community in terms of compositional dissimilarity (β-diversity) when compared to placebo.
Study design and reporting of results
Overall, the reporting of the analyses and results was non-transparent and difficult to assess with very few effect sizes, confidence intervals, and P values reported. This is possibly due to the fact that fecal microbiomics is a relatively new research area that currently relies heavily on non-parametric statistics and lacks an internationally accepted standard approach of reporting results. Unfortunately, this impedes the comparison of results in the present review. As the only study, Ferrario et al.  used a cross-over design, which may not be the ideal design to assess the effects of a probiotic intervention due to the risk of carry-over effects . In the study the probiotic cell count was decreased after the 4-week wash-out period compared with baseline count, suggesting that wash-out was effective. However, a carry-over effect at the outcome level cannot be excluded. Only two studies provided a priori sample size calculations [20, 24], of which two calculated statistical power based on alterations in fecal microbiota composition [20, 22]. Thus, several of the studies may well have been underpowered, with an inherent risk of unequal distribution of confounding factors. A potential confounder in the studies reviewed is habitual diet. Human studies have revealed that short-term and long-term changes in diet (such as plant-based vs. animal-based, amount of dietary fibers and fat) influence the fecal microbiota composition and function [36–38]. Hence, the enormous inter-individual variation in the dietary intake and its effect on the fecal microbiota may mask the true picture of the impact of a probiotic treatment. Only one of the included studies monitored habitual diet with the aim of accounting for differences in dietary habits, specifically considering prebiotic fibers, during the intervention period . A major limitation of most included studies is an unclear, inexplicit statement of the pre-specified primary outcome and delimitation of secondary outcomes. Only one study  is explicitly labeled as a pilot trial, reporting a multitude of outcomes, only in part addressing multiple testing. Two studies do not address the issue of multiple testing [23, 27], while others report multiple primary outcomes or make no distinction between primary and secondary outcomes [19–22]. Reporting of the results is generally unclear, with between-group comparisons on primary outcomes intermixed with results on secondary outcomes and within-group comparisons of differences between baseline and post-intervention measures.
Although study participants in the included trials were all healthy adults, the demographic makeup varied widely among studies. Rampelli et al.  included only elderly individuals, who may respond differently to probiotics than young individuals and Brahe et al.  included postmenopausal women only.
Considering that the impact on the fecal microbiota may differ among strains of the same bacterial species , despite close phylogenetic relationships, a potential source of heterogeneity is the use of various probiotic agents. Six studies used single-strain interventions with probiotic products belonging to the genera Lactobacillus [19, 20, 22, 24, 27] or Bacillus . One study used a double-strain probiotic mixture of bacteria belonging to the genera Lactobacillus and Bifidobacterium . Whereas the use of different probiotic agents makes it difficult to draw any meta-analytical conclusions, the choice between single-strain and multi-strain intervention is probably of less importance. In most cases, inert bacteria are administered and within a few hours are entering a diverse ecosystem where they are numerically a minority. So while additive or synergistic effects might be observed in vitro, the opportunity for metabolically active strains delivered in combination to result in similar effects in vivo may not present itself.
None of the studies included in the present review comment on the rationale behind their choice of dosage. The International Scientific Association for Probiotics and Prebiotics provides a list of dosages ranging from 1 × 108 to 1.8 × 1012 CFU twice daily depending on strain and disease, based on at least one well designed clinical trial showing a beneficial effect for a health promoting or therapeutic outcome . However, the list covers gastrointestinal disorders only and does not address fecal microbiota in healthy participants. In general, different dosages should be assessed to facilitate an interpretation of the dose-response relationship of probiotic consumption on relevant outcomes, rather than on safety and viability alone. The information provided by such studies would enable the identification of the dosage needed to observe an impact on the relevant outcome and add to the likelihood that an observed association is causal . Hanifi et al.  examined and detected oral dose-response relationships, but for tolerance and gastrointestinal viability only. As of now, it is impossible to draw any conclusions on the ideal dosage regarding effects on the fecal microbiota composition. Likewise, the optimal duration of intervention remains elusive.
Mode of administration may also contribute to the observed lack of impact on fecal microbiota. Ingested probiotics must survive hostile environments including acidic, protease and bile-salt rich conditions during their passage through the gastrointestinal tract [42, 43]. Currently, screening feces is the only way to assess whether the probiotics have survived through the gastrointestinal tract. Yet, the site of action may be proximal to the colon and it is not necessarily possible to conclude on the degree of colonization or even the amount of bacteria that produce the effect . In contrast to the findings in five of the included studies, the study by Rampelli et al.  showed only a trend towards enrichment of the probiotic strain, perhaps due to the use of biscuits as the mode of administration, yet, another reason may be low compliance. This may add to the explanation why little effect of probiotics was found in Rampelli et al. . Compliance was evaluated as sufficient in the remaining studies [19–22, 24, 27].
Further adding to the heterogeneity between studies is the application of different methods for assessing fecal microbiota. Even though all studies applied high-throughput metagenomics approaches, the resolution and specificity levels varied tremendously and no studies used the same methodological approach. Only one study examined the fecal microbiota by an untargeted metagenomic approach using shotgun sequencing and thereby providing information on microbial gene and derived functional levels, free of bias introduced by amplification of a specific genomic region as is the case in targeted 16S rRNA sequencing and array-based analyses . Compositional information can be achieved by mapping the reads to a microbial gene reference catalogue , as was done by Brahe et al. . Still, only a fraction of sequencing reads can be mapped to the existing reference catalogues. The targeted 16S rRNA approach provides information at the taxonomic level in the form of abundance and phylogenetic relationship, but the method has pitfalls in PCR amplification steps  and cross-platform comparison is not straightforward. Of the included studies, two use 454 pyrosequencing [19, 21], one uses Illumina MiSeq , another uses Ion Torrent sequencing , and two use phylogenetic microarrays [22, 23]. The sequencing platforms differ in costs, coverage, and length of reads with the Illumina platform becoming more widely used . Community profiles from HITChip correlate well with pyrosequencing-based compositions (Pearson correlations at phylum (average r = 0.94), class (0.93), order (0.94) and family levels (0.77)) and the HTF microbe.array has demonstrated good reproducibility by cluster analysis of the phylogenetic fingerprint in samples from the same participant [47, 48]. In general, using phylogenic microarray approaches have the advantages of being cost-efficient for compositional characterization; however cross-hybridization may occur and only taxa that are covered by the reference sequences can be detected . Another well-known source of bias in 16S rRNA-based studies is the targeted hypervariable region of the 16S rRNA gene. The region used for analysis in the included studies applying 16S rRNA-based methods varies with one study using V1 and V6 , one study using V3 , one study using V3 and V4 , one study using V5 and V6 , and one study does not specify . Several studies have examined the effects of region choice when evaluating fecal microbiota composition with no current international consensus [47, 49–51].
Probiotics in health and disease
In a recent systematic review including 29 trials studying healthy adults with undisturbed microbiota (using non-high-throughput molecular techniques) only ~20 % showed an effect of probiotic treatment on fecal microbiota. It is concluded that there is little, if any, evidence of an effect of probiotic treatment in circumstances where the microbiota is unperturbed by pathophysiological processes or pharmaceutical treatment (antibiotics or chemotherapy), either concurrent with or prior to intervention. However, where dysbiosis is present or where the microbiota is perturbed, there is some evidence for a restorative or protective effect of certain strains of probiotics, both on the fecal microbial community itself, but more importantly, also on host physiology, e.g. alleviation of gastrointestinal symptoms .
In the case of dysbiotic or perturbed microbial communities, any restorative or protective effect on the microbiota alone, without any measurable beneficial effect for the host, would predominantly be of academic interest by improving our understanding of the intestinal ecosystem. In the case of undisturbed microbiota, any inference of health benefit from changes to the microbiota alone would be highly speculative without a direct linkage to relevant host phenotypes. Ideally, hard endpoint data would determine the effects of probiotics in healthy individuals, but considering the time perspective of generating such data this may be long in coming. Until such studies are available, any statement on health benefits of probiotic supplementation in healthy participants would rely on observed effects on biomarkers or other intermediate outcomes.
Limitations of this review include the search terms used to identify relevant papers. In addition to probio*, we specifically searched on bifido* and lacto*, but other search terms such as Bacillus and Saccharomyces could have been relevant. Publication bias is a well-known challenge within the field of systematic reviews and meta-analyses; however, the majority of the studies included in the present review provide null findings, indicating that this concern may be settled to some extent. Language bias cannot be ruled out since our search was exclusively based on English language dominated sources.
Based on our review of the available RCTs, we find there is a lack of evidence to conclude whether or not there is an effect of probiotics on fecal microbiota composition in healthy adults, as assessed by high-throughput molecular techniques. A number of issues blur the conclusions that can be drawn from the studies, including small sample sizes with lack of statistical power, low resolution-methods of assessing fecal microbiota composition, inter-individual variation in susceptibility toward the probiotic, use of different probiotic strains either in isolation or in combination, variations in dosage and administration mode of probiotics, duration of intervention, or variation in the habitual diet of participants. Future research on the impact of probiotics on fecal microbiota configuration and function should involve statistically well-powered RCTs in well-phenotyped individuals. Importantly, future studies would also benefit from pre-specifying the primary outcome and transparently reporting the results including effect sizes, confidence intervals, and P values as well as providing a clear distinction of within-group and between-group comparisons. For the purpose of demonstrating health benefits of probiotic supplementation, effects should be demonstrated on relevant host phenotypes, which is non-trivial in healthy participants. Studies with microbiome features as the primary outcome should be reserved for improving our understanding of biology in general and our insight into microbial interactions in vivo in particular.
Colony forming units
Operational taxonomic unit
Principal coordinate analysis
Randomized controlled trial
The Novo Nordisk Foundation Center for Basic Metabolic Research is an independent research center at the Faculty of Health and Medical Sciences, University of Copenhagen and is partially funded by an unrestricted grant from the Novo Nordisk Foundation (http://www.metabol.ku.dk). The funders played no role in study design, collection, analysis, interpretation of data, writing of the report, or in the decision to submit the paper for publication.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Bäckhed F, Ding H, Wang T, Hooper LV, Koh GY, Nagy A, et al. The gut microbiota as an environmental factor that regulates fat storage. P Natl Acad Sci USA. 2004;101(44):15718–23.View ArticleGoogle Scholar
- LeBlanc JG, Milani C, de Giori GS, Sesma F, van Sinderen D, Ventura M. Bacteria as vitamin suppliers to their host: a gut microbiota perspective. Curr Opin Biotechnol. 2013;24(2):160–8.View ArticlePubMedGoogle Scholar
- Endt K, Stecher B, Chaffron S, Slack E, Tchitchek N, Benecke A, et al. The microbiota mediates pathogen clearance from the gut lumen after non-typhoidal Salmonella diarrhea. PLoS Pathog. 2010;6(9):e1001097.View ArticlePubMedPubMed CentralGoogle Scholar
- Round JL, Mazmanian SK. The gut microbiota shapes intestinal immune responses during health and disease. Nat Rev Immunol. 2009;9(5):313–23.View ArticlePubMedPubMed CentralGoogle Scholar
- Collins SM, Surette M, Bercik P. The interplay between the intestinal microbiota and the brain. Nat Rev Microbiol. 2012;10(11):735–42.View ArticlePubMedGoogle Scholar
- Feng Q, Liang S, Jia H, Stadlmayr A, Tang L, Lan Z, et al. Gut microbiome development along the colorectal adenoma-carcinoma sequence. Nat Commun. 2015;6:6528.View ArticlePubMedGoogle Scholar
- Yu J, Feng Q, Wong SH, Zhang D, Liang QY, Qin Y, et al. Metagenomic analysis of faecal microbiome as a tool towards targeted non-invasive biomarkers for colorectal cancer. Gut. 2015. doi:10.1136/gutjnl-2015-309800.Google Scholar
- Zhang X, Zhang D, Jia H, Feng Q, Wang D, Liang D, et al. The oral and gut microbiomes are perturbed in rheumatoid arthritis and partly normalized after treatment. Nat Med. 2015;21(8):895–905.View ArticlePubMedGoogle Scholar
- Le Chatelier E, Nielsen T, Qin J, Prifti E, Hildebrand F, Falony G, et al. Richness of human gut microbiome correlates with metabolic markers. Nature. 2013;500(7464):541–6.View ArticlePubMedGoogle Scholar
- Qin J, Li Y, Cai Z, Li S, Zhu J, Zhang F, et al. A metagenome-wide association study of gut microbiota in type 2 diabetes. Nature. 2012;490(7418):55–60.View ArticlePubMedGoogle Scholar
- Forslund K, Hildebrand F, Nielsen T, Falony G, Le Chatelier E, Sunagawa S, et al. Disentangling type 2 diabetes and metformin treatment signatures in the human gut microbiota. Nature. 2015;528(7581):262–6.View ArticlePubMedGoogle Scholar
- Hill C, Guarner F, Reid G, Gibson GR, Merenstein DJ, Pot B, et al. Expert consensus document: The International Scientific Association for Probiotics and Prebiotics consensus statement on the scope and appropriate use of the term probiotic. Nat Rev Gastroenterol Hepatol. 2014;11(8):506–14.View ArticlePubMedGoogle Scholar
- Gerritsen J, Smidt H, Rijkers GT, de Vos WM. Intestinal microbiota in human health and disease: the impact of probiotics. Genes Nutr. 2011;6(3):209–40.View ArticlePubMedPubMed CentralGoogle Scholar
- Rijkers GT, Bengmark S, Enck P, Haller D, Herz U, Kalliomaki M, et al. Guidance for substantiating the evidence for beneficial effects of probiotics: current status and recommendations for future research. J Nutr. 2010;140(3):671S–6S.View ArticlePubMedGoogle Scholar
- Ford AC, Quigley EM, Lacy BE, Lembo AJ, Saito YA, Schiller LR, et al. Efficacy of prebiotics, probiotics, and synbiotics in irritable bowel syndrome and chronic idiopathic constipation: Systematic review and meta-analysis. Am J Gastroenterol. 2014;109(10):1547–61.View ArticlePubMedGoogle Scholar
- Le Barz M, Anhê FF, Varin TV, Desjardins Y, Levy E, Roy D, et al. Probiotics as complementary treatment for metabolic disorders. Diabetes Metab J. 2015;39(4):291–303.View ArticlePubMedPubMed CentralGoogle Scholar
- Bäckhed F, Fraser CM, Ringel Y, Sanders ME, Sartor RB, Sherman PM, et al. Defining a healthy human gut microbiome: current concepts, future directions, and clinical applications. Cell Host Microbe. 2012;12(5):611–22.View ArticlePubMedGoogle Scholar
- McFarland LV. Use of probiotics to correct dysbiosis of normal microbiota following disease or disruptive events: a systematic review. BMJ Open. 2014;4(8):e005047.View ArticlePubMedPubMed CentralGoogle Scholar
- Bjerg AT, Sørensen MB, Krych L, Hansen LH, Astrup A, Kristensen M, et al. The effect of Lactobacillus paracasei subsp. paracasei L. casei W8® on blood levels of triacylglycerol is independent of colonisation. Benef Microbes. 2015;6(3):263–9.View ArticlePubMedGoogle Scholar
- Brahe LK, Le Chatelier E, Prifti E, Pons N, Kennedy S, Blædel T, et al. Dietary modulation of the gut microbiota–a randomised controlled trial in obese postmenopausal women. Br J Nutr. 2015;114(03):406–17.View ArticlePubMedPubMed CentralGoogle Scholar
- Hanifi A, Culpepper T, Mai V, Anand A, Ford A, Ukhanova M, et al. Evaluation of Bacillus subtilis R0179 on gastrointestinal viability and general wellness: a randomised, double-blind, placebo-controlled trial in healthy adults. Benef microbes. 2014;6(1):19–27.View ArticleGoogle Scholar
- Lahti L, Salonen A, Kekkonen RA, Salojärvi J, Jalanka-Tuovinen J, Palva A, et al. Associations between the human intestinal microbiota, Lactobacillus rhamnosus GG and serum lipids indicated by integrated analysis of high-throughput profiling data. PeerJ. 2013;1:e32.View ArticlePubMedPubMed CentralGoogle Scholar
- Rampelli S, Candela M, Severgnini M, Biagi E, Turroni S, Roselli M, et al. A probiotics-containing biscuit modulates the intestinal microbiota in the elderly. J Nutr Health Aging. 2013;17(2):166–72.View ArticlePubMedGoogle Scholar
- Simon MC, Strassburger K, Nowotny B, Kolb H, Nowotny P, Burkart V, et al. Intake of Lactobacillus reuteri improves incretin and insulin secretion in glucose-tolerant humans: a proof of concept. Diabetes Care. 2015;38(10):1827–34.View ArticlePubMedGoogle Scholar
- McNulty NP, Yatsunenko T, Hsiao A, Faith JJ, Muegge BD, Goodman AL, et al. The impact of a consortium of fermented milk strains on the gut microbiome of gnotobiotic mice and monozygotic twins. Sci Transl Med. 2011;3(106):106ra–106.View ArticleGoogle Scholar
- Zhang J, Wang L, Guo Z, Sun Z, Gesudu Q, Kwok L, et al. 454 pyrosequencing reveals changes in the faecal microbiota of adults consuming Lactobacillus casei Zhang. FEMS Microbiol Ecol. 2014;88(3):612–22.View ArticlePubMedGoogle Scholar
- Ferrario C, Taverniti V, Milani C, Fiore W, Laureati M, De Noni I, et al. Modulation of fecal Clostridiales bacteria and butyrate by probiotic intervention with Lactobacillus paracasei DG varies among healthy adults. J Nutr. 2014;144(11):1787–96.View ArticlePubMedGoogle Scholar
- Plaza-Diaz J, Fernandez-Caballero JA, Chueca N, Garcia F, Gomez-Llorente C, Saez-Lara MJ, et al. Pyrosequencing analysis reveals changes in intestinal microbiota of healthy adults who received a daily dose of immunomodulatory probiotic strains. Nutrients. 2015;7(6):3999–4015.View ArticlePubMedPubMed CentralGoogle Scholar
- marketsandmarkets.com. Probiotic Ingredients Market by Function (Regular, Preventative, Therapy), Application (Food & Beverage, Dietary Supplements & Animal Probiotics), Ingredient (Bacteria & Yeast), and by Region - Global Trends & Forecast to 2020. Available at: http://www.marketsandmarkets.com/Market-Reports/probiotic-market-advanced-technologies-and-global-market-69.html (accessed 14 Mar 2016).
- Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151(4):264–9.View ArticlePubMedGoogle Scholar
- Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJM, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996;17(1):1–12.View ArticlePubMedGoogle Scholar
- Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 2009;6(7):e1000100.View ArticlePubMedPubMed CentralGoogle Scholar
- Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration; 2011. Available at: http://handbook.cochrane.org/.
- Qin J, Li R, Raes J, Arumugam M, Burgdorf KS, Manichanh C, et al. A human gut microbial gene catalogue established by metagenomic sequencing. Nature. 2010;464(7285):59–65.View ArticlePubMedPubMed CentralGoogle Scholar
- Haller D, Antoine JM, Bengmark S, Enck P, Rijkers GT, Lenoir-Wijnkoop I. Guidance for substantiating the evidence for beneficial effects of probiotics: probiotics in chronic inflammatory bowel disease and the functional disorder irritable bowel syndrome. J Nutr. 2010;140(3):690s–7s.View ArticlePubMedGoogle Scholar
- David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE, Wolfe BE, et al. Diet rapidly and reproducibly alters the human gut microbiome. Nature. 2014;505(7484):559–63.View ArticlePubMedPubMed CentralGoogle Scholar
- Shoaie S, Ghaffari P, Kovatcheva-Datchary P, Mardinoglu A, Sen P, Pujos-Guillot E, et al. Quantifying diet-induced metabolic changes of the human gut microbiome. Cell Metab. 2015;22(2):320–31.View ArticlePubMedGoogle Scholar
- Zeevi D, Korem T, Zmora N, Israeli D, Rothschild D, Weinberger A, et al. Personalized nutrition by prediction of glycemic responses. Cell. 2015;163(5):1079–94.View ArticlePubMedGoogle Scholar
- Menard O, Butel MJ, Gaboriau-Routhiau V, Waligora-Dupriet AJ. Gnotobiotic mouse immune response induced by Bifidobacterium sp. strains isolated from infants. Appl Environ Microbiol. 2008;74(3):660–6.View ArticlePubMedPubMed CentralGoogle Scholar
- Guarner F, Khan AG, Garisch J, Eliakim R, Gangl A, Thomson A, et al. World Gastroenterology Organisation Global Guidelines: probiotics and prebiotics October 2011. J Clin Gastroenterol. 2012;46(6):468–81.View ArticlePubMedGoogle Scholar
- Hill AB. The Environment and disease: association or causation? Proc R Soc Med. 1965;58:295–300.PubMedPubMed CentralGoogle Scholar
- Cook MT, Tzortzis G, Charalampopoulos D, Khutoryanskiy VV. Microencapsulation of probiotics for gastrointestinal delivery. J Control Release. 2012;162(1):56–67.View ArticlePubMedGoogle Scholar
- Charteris WP, Kelly PM, Morelli L, Collins JK. Development and application of an in vitro methodology to determine the transit tolerance of potentially probiotic Lactobacillus and Bifidobacterium species in the upper human gastrointestinal tract. J Appl Microbiol. 1998;84(5):759–68.View ArticlePubMedGoogle Scholar
- Farnworth ER. The evidence to support health claims for probiotics. J Nutr. 2008;138(6):1250S–4S.PubMedGoogle Scholar
- Weinstock GM. Genomic approaches to studying the human microbiota. Nature. 2012;489(7415):250–6.View ArticlePubMedPubMed CentralGoogle Scholar
- Rastogi G, Sani RK. Molecular techniques to assess microbial community structure, function, and dynamics in the environment. In: Ahmad I, Ahmad F, Pichtel J, Rastogi G, Sani R, editors. Microbes and Microbial Technology. New York: Springer; 2011. p. 29–57.View ArticleGoogle Scholar
- Claesson MJ, O’Sullivan O, Wang Q, Nikkila J, Marchesi JR, Smidt H, et al. Comparative analysis of pyrosequencing and a phylogenetic microarray for exploring microbial community structures in the human distal intestine. PLoS One. 2009;4(8):e6669.View ArticlePubMedPubMed CentralGoogle Scholar
- Candela M, Consolandi C, Severgnini M, Biagi E, Castiglioni B, Vitali B, et al. High taxonomic level fingerprint of the human intestinal microbiota by ligase detection reaction--universal array approach. BMC Microbiol. 2010;10:116.View ArticlePubMedPubMed CentralGoogle Scholar
- Mizrahi-Man O, Davenport ER, Gilad Y. Taxonomic classification of bacterial 16S rRNA genes using short sequencing reads: evaluation of effective study designs. PLoS One. 2013;8(1):e53608.View ArticlePubMedPubMed CentralGoogle Scholar
- Sundquist A, Bigdeli S, Jalili R, Druzin ML, Waller S, Pullen KM, et al. Bacterial flora-typing with targeted, chip-based Pyrosequencing. BMC Microbiol. 2007;7:108.View ArticlePubMedPubMed CentralGoogle Scholar
- Wang Q, Garrity GM, Tiedje JM, Cole JR. Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl Environ Microbiol. 2007;73(16):5261–7.View ArticlePubMedPubMed CentralGoogle Scholar