Atlas chapter 9 · Gastrointestinal

Saccharomyces boulardii

Saccharomyces boulardii — probiotic lineage within Saccharomyces cerevisiae

E5++ — Supports the tested propositionR5

Proposition / claim: Clinically studied Saccharomyces boulardii preparations reduce the risk of antibiotic-associated diarrhoea.
Evidence owner: Clinically studied, properly characterized probiotic Saccharomyces boulardii preparations within the populations and settings represented by the evidence; not generic Saccharomyces cerevisiae, baker's yeast, brewer's yeast or arbitrary yeast products.
Dominant uncertainty: How strain- and preparation-specific the mature antibiotic-associated-diarrhoea evidence is, and how consistently the effect transports across populations and antibiotic settings. The evidence context also retains a large negative multicentre randomized, double-masked, placebo-controlled study in hospitalized adults that found no significant reduction and was stopped early for futility; the overall direction remains + while this null result stays visible rather than being smoothed away.

Organism: Saccharomyces boulardii
Tested intervention: Clinically studied probiotic Saccharomyces boulardii preparations
Evidence target: Standardized preparation
Intervention description: Live probiotic yeast preparations used during antibiotic treatment. CNCM I-745 is particularly important in the modern clinical literature, although historical trials and meta-analyses do not uniformly report strain identity.
Population: People receiving antibiotic treatment across multiple randomized clinical settings and populations represented in the Saccharomyces boulardii evidence programme.
Comparator: Placebo or control.
Outcome: Incidence / risk of antibiotic-associated diarrhoea.
Evidence maturity (E): E5+
Direction: + — Supports the tested proposition
Historical flag (H): Not specified at dossier level
Replication (R): R5
Claim Gap (CG): Claim ceiling: CG1 for broad ordinary-language prevention wording. CG0–1: Randomized trials found that clinically studied Saccharomyces boulardii preparations reduce antibiotic-associated diarrhoea risk in some populations. CG1: Saccharomyces boulardii helps prevent antibiotic-associated diarrhoea. CG2: Saccharomyces boulardii supports gut health during antibiotics. CG3: Saccharomyces boulardii restores the gut microbiome after antibiotics. CG4: This probiotic heals antibiotic damage and strengthens immunity.
Evidence Transfer Vector: Transferred claim: "Any Saccharomyces cerevisiae probiotic prevents antibiotic-associated diarrhoea." — T / I / R. T — Taxon: clinical Saccharomyces boulardii lineage → broader Saccharomyces cerevisiae identity. I — Intervention: clinically studied probiotic preparation → generic yeast product. R — Replication: specific evidence programme → arbitrary product assumed to inherit it.
Taxonomic Transfer Confidence: Clinical Saccharomyces boulardii lineage → generic Saccharomyces cerevisiae: LOW
Intervention Transfer Confidence: CNCM I-745 → arbitrary Saccharomyces boulardii product: NOT AUTOMATIC / REQUIRES VERIFICATION
Safety: S2
Conservation: C0
Resolution zone: RZ-5 — applies to the remaining strain/provenance and transport question rather than implying that another generic efficacy trial is automatically required.
Next best experiment: First perform strain-provenance reconstruction across the grade-determining randomized studies, identifying which strain/preparation each trial used and clustering outcomes by strain, preparation, population and setting. If a new trial is subsequently required, directly compare clinically important strains/preparations, or prospectively test a clearly identified alternative strain against an appropriate control in the same antibiotic-exposure setting, with the prevention endpoint, timing and case definition prespecified.
Primary references: Szajewska H, Kołodziej M. Systematic review with meta-analysis: Saccharomyces boulardii in the prevention of antibiotic-associated diarrhoea. Alimentary Pharmacology & Therapeutics. 2015;42(7):793-801. PMID 26216624. DOI 10.1111/apt.13344. Primary grade-owning efficacy synthesis. Twenty-one randomized controlled trials involving 4,780 participants. Antibiotic-associated diarrhoea occurred in approximately 8.5% of participants receiving S. boulardii compared with 18.7% of controls; pooled RR approximately 0.47 and NNT approximately 10. The synthesis supports the narrow AAD-prevention proposition while not establishing universal benefit across populations or indications. This is the principal efficacy reference for the dossier. The same synthesis also marks an indication boundary: the strong antibiotic-associated-diarrhoea result does not transfer automatically into established prevention efficacy for adult Clostridioides difficile-associated diarrhoea or infection.
Ehrhardt S, Guo N, Hinz R, Schoppen S, May J, Reiser M, et al.; SacBo Study Group. Saccharomyces boulardii to Prevent Antibiotic-Associated Diarrhea: A Randomized, Double-Masked, Placebo-Controlled Trial. Open Forum Infectious Diseases. 2016;3(1):ofw011. PMID 26973849. DOI 10.1093/ofid/ofw011. Major contradictory randomized evidence. The multicentre hospital trial randomized 477 adults receiving systemic antibiotics and was stopped early for futility. Twenty-one AAD events occurred in the S. boulardii group and 19 in placebo; HR approximately 1.02 (95% CI 0.55-1.90; P=.94). The trial administered a defined S. boulardii probiotic preparation rather than generic yeast. The result demonstrates that the favourable pooled AAD effect does not transport uniformly to every population and setting. It is retained to preserve genuine discordance and population/setting heterogeneity, not as cancellation of the wider positive evidence programme.
Rannikko J, Holmberg V, Karppelin M, et al. Fungemia and Other Fungal Infections Associated with Use of Saccharomyces boulardii Probiotic Supplements. Emerging Infectious Diseases. 2021;27(8):2090-2096. DOI 10.3201/eid2708.210018. Primary safety reference. Finnish multicentre retrospective registry study covering five university hospitals during 2009-2018. Forty-six patients with Saccharomyces fungemia were identified; at least 20 (43%) had documented S. boulardii probiotic exposure. Compared with controls with bacteremia or candidemia, the odds ratio for probiotic use was 14 (95% CI 4-44). This supports a genuine invasive-infection safety signal in vulnerable clinical populations but does not provide an incidence estimate for ordinary users. The distinction between an association or safety signal and population incidence or universal causal attribution is preserved; the source itself notes limitations in microbiological confirmation of the probiotic strain across cases.
Martin IW, Tonner R, Trivedi J, Miller H, Lee R, Liang X, Rotello L, Isenbergh E, Anderson J, Perl T, Zhang SX. Saccharomyces boulardii probiotic-associated fungemia: questioning the safety of this preventive probiotic's use. Diagnostic Microbiology and Infectious Disease. 2017;87(3):286-288. PMID 28024866. DOI 10.1016/j.diagmicrobio.2016.12.004. Supporting causal-attribution evidence. Genomic and proteomic typing demonstrated strain relatedness between the Saccharomyces organism recovered from the patient's bloodstream and the probiotic preparation being administered. This strengthens causal attribution in that individual case while not establishing that all reported Saccharomyces fungemia is probiotic-derived.
Source chapter: Chapter 9, The Fungal Evidence Atlas