Saccharomyces boulardii (probiotic strain)Probiotic

• Antibiotic-associated diarrheaLikely helps

  A 2015 updated systematic review of 21 RCTs (4,780 participants) found the incidence fell from 18.7% to 8.5% [2].
  
  That is a risk ratio of 0.47 (95% CI 0.38–0.57), with an NNT of 10 [2].
  
  It held in both age groups: children 20.9% down to 8.8% (RR 0.43), adults 17.4% down to 8.2% (RR 0.49) [2].
  
  A separate adult-focused meta-analysis of 31 placebo-controlled arms (5,029 patients) landed on the same RR of 0.47 (95% CI 0.35–0.63, P<0.001) [1].
  
  84% of those 31 arms were rated efficacious and safe [1]. Natural Medicines-derived ratings put this at "likely effective" [6].
  
  This is the one use where independent reviews genuinely agree. ^[2]

• Diarrhea in tube-fed patientsLikely helps

  Rated "likely effective" for diarrhea in patients on enteral (tube) feeding, in Natural Medicines-derived ratings [6].
  
  The adult meta-analysis separately reported significant efficacy for enteral-nutrition-related diarrhea [1]. ^[6]

• C. difficile — recurrence versus a first episodePossibly helps

  Two reviews reach different answers, and the difference is which patients they looked at.
  
  The 2015 meta-analysis found a significant reduction in C. difficile-associated diarrhea in children: RR 0.25 (95% CI 0.08–0.73) [2].
  
  In adults, the same review found nothing significant: RR 0.80 (95% CI 0.47–1.34) [2].
  
  An earlier review of 4 adult RCTs found a benefit against recurrence in some subgroups [3]. In patients with a history of recurrent infection it was significant, OR 0.30 (95% CI 0.10–0.84) [3].
  
  The high-dose vancomycin subgroup showed a trend only, OR 0.33 (95% CI 0.04–1.01) [3].
  
  For a first episode, that review found the role poorly defined and not significant [3]. Natural Medicines rates recurrence use "possibly effective" and first-episode use unsupported [6].
  
  The two reviews used different trial sets and outcome definitions, so they are not strictly contradictory. Read together, they say this benefit is context-dependent rather than general. ^[2]

• Traveler's diarrheaPossibly helps

  The adult meta-analysis reported significant efficacy against traveler's diarrhea [1].
  
  Natural Medicines-derived ratings are more cautious about the same body of work, at "possibly effective" [6]. ^[1]

• Side effects of H. pylori eradication therapyPossibly helps

  Adding S. boulardii to standard eradication therapy reduced symptoms during it, per the adult meta-analysis [1].
  
  The rating is "possibly effective", at 5 billion CFU/day alongside the usual regimen [6]. ^[1]

• HIV-related diarrhea and acnePossibly helps

  Both sit at "possibly effective" in Natural Medicines-derived ratings [6].
  
  The adult meta-analysis describes HIV-related diarrhea as promising but less well supported than the antibiotic-diarrhea finding [1]. ^[6]

• Irritable bowel syndromePossibly helps

  Graded insufficient to rate — only early research existed when these reviews were written [6].
  
  A large phase 3 randomized, placebo-controlled multicentre trial of strain CNCM I-745 at 500 mg/day in adults with non-constipated IBS was still running as of late 2025 [15].
  
  No results were posted. Nothing is settled here in either direction [15]. ^[6]

• Crohn's disease and ulcerative colitisPossibly helps

  Both are graded insufficient to rate, even though specific research doses exist for each [6].
  
  The adult meta-analysis lists Crohn's among its promising-but-less-supported indications, alongside giardiasis [1]. ^[6]

• Lyophilized (freeze-dried) capsulesMost situations. The format most likely to still hold living yeast when you take it

  Elemental content

  Dried live yeast; the reference strain CBS 5926 is also listed as ATCC 74012 and CNCM I-745 [7]

  Absorption

  Stable at room temperature for over a year if kept dry, usually in blister packs [1].
  
  A German comparison of four products found the lyophilized one beat three heat-dried ones on viable-cell count and reactivation speed [1].

  Best for

  Most situations. The format most likely to still hold living yeast when you take it [1]

• Heat-dried preparationsNothing in particular. Worth checking the label before buying one

  Elemental content

  Same yeast, different drying process

  Absorption

  Not stable at room temperature.
  
  Potency drops quickly and the product has to be refrigerated [1].

  Best for

  Nothing in particular. Worth checking the label before buying one [1]

• Powder packets and sachetsChildren and anyone who can't swallow capsules

  Elemental content

  Single-dose dried yeast, same strains as the capsules

  Absorption

  Same oral route as a capsule.
  
  The contents can be swallowed directly or mixed into food or a drink [13].

  Best for

  Children and anyone who can't swallow capsules [13]

• Strain identity — CNCM I-745 versus everything elseMatching the trials. The strain code on the label is what actually matters

  Elemental content

  Not a separate species: genomics places it as a lineage of Saccharomyces cerevisiae [7]

  Absorption

  CBS 5926 / CNCM I-745 has been evaluated in more than 90 randomized clinical trials [7].
  
  Other proprietary strains — CNCM I-1079, CNCM I-3799, DBVPG 6763 — carry no equivalent evidence base [7].

  Best for

  Matching the trials. The strain code on the label is what actually matters [7]

• Gas and bloating

  The most commonly reported effect, and a mild one [6].
  
  Beyond the antifungal class, no clinically significant interactions with food, drink, or other drugs are well established — though the interaction research overall is incomplete [6]. ^[6]

• Hypersensitivity reactions

  Rare hypersensitivity reactions are documented [6].
  
  One published case describes an itchy rash appearing around both ankles about 1.5 hours after ingestion [16].
  
  People with a known yeast allergy are advised to avoid these products, and product hypersensitivity is listed as a contraindication [14]. ^[16]

• Fungaemia and sepsis in fragile patients

  This is the serious one, and it is the price of the organism being alive. The yeast can translocate into the bloodstream [9].
  
  It happens essentially only in one group: people with central venous catheters, the critically ill, and the immunocompromised [9].
  
  A Finnish five-hospital registry covering 2009–2018 looked at patients with Saccharomyces fungaemia. S. boulardii probiotic use carried an odds ratio of 14 (95% CI 4–44) versus bacteraemia and candidaemia controls [4].
  
  Other, non-blood Saccharomyces culture findings gave an odds ratio of 10 (95% CI 3–32) [4].
  
  The European Medicines Agency reviewed the spontaneous-report data in 2020. It counted 12 cases rated at least a reasonable causal possibility and 3 fatal cases with positive blood cultures [10].
  
  EMA concluded that warranted a formal sepsis warning on the product labelling for fragile patients [10].
  
  Malaysia's NPRA, citing the same review, reported 61 cumulative fungaemia cases and 10 fatal cases in EudraVigilance [9]. It advises avoiding the product in immunocompromised patients and anyone with a central line [9]. ^[9]

• Why the trial data and the real-world reports look so different

  A large adult systematic review found zero fungaemia cases among nearly 3,000 patients enrolled in controlled trials [1].
  
  That sits against 61 real-world cases and an odds ratio of 14. It is not actually a contradiction [4].
  
  Trials excluded the critically ill, central-line, and severely immunocompromised. The real-world cases sit almost entirely inside exactly those excluded groups [1].
  
  So the risk is real but concentrated, not spread evenly across everyone taking it. ^[1]

• Handling the capsules near a central line

  Some documented cases came not from swallowing the product at all, but from environmental contamination [1].
  
  Healthcare workers transferred yeast from an opened capsule or packet onto a patient's line [1].
  
  Drug references advise not opening these products near central-line patients, and changing gloves after handling them [14]. ^[1]

• Children and premature infants

  Rated possibly safe for children used appropriately, and the antibiotic-diarrhea evidence includes children specifically [6].
  
  Any childhood diarrhea should be medically assessed first. Premature infants are flagged as being at greater risk of probiotic-related side effects [6]. ^[6]

• Pregnancy and breastfeeding

  Reliable safety data is missing for both [6].
  
  Sources advise erring toward caution — avoid it, or check with a clinician first [14]. ^[6]

• What is actually in the capsule

  Independent testing of six S. boulardii products found only half matched their labelled concentration [1].
  
  One product held roughly 20,000-fold fewer live cells than its label claimed [1].
  
  Strain, manufacturing process, and labelled potency all vary between products. The dose you actually get cannot be assumed from the number on the bottle [6]. ^[1]

• Antifungal drugs

  S. boulardii is itself a fungus, so antifungals act on it [6].
  
  Fluconazole, itraconazole, ketoconazole, voriconazole, posaconazole, terbinafine, griseofulvin, nystatin, and amphotericin B formulations can reduce or wipe out its probiotic effect [6].
  
  The echinocandins do the same — caspofungin, micafungin, anidulafungin [14]. This combination is flagged for caution or avoidance. ^[6]

Not a dietary item. S. boulardii comes from a manufactured capsule or powder packet, not from a meal [5].

Its origin is food, though. French microbiologist Henri Boulard isolated the yeast in 1923 from the skins of lychee (Litchi chinensis) and mangosteen (Garcinia mangostana) in Southeast Asia [7].

He had watched local communities drink a preparation made from those fruit skins to settle diarrhea. The yeast behind it took his name [7].

Because it is a lineage of Saccharomyces cerevisiae, it is related to baker's and brewer's yeast — but it is a distinct strain, more tolerant of acid and heat [7].

Ordinary bread and beer do not reliably contain it. Manufacturer sources say some fermented foods and drinks carry it [5].

Meal timing barely matters here, which is unusual for a probiotic. Bacterial probiotics like Lactobacillus and Bifidobacterium are acid-sensitive, so guidance for them is usually before or with food [11].

S. boulardii is a yeast, and comparatively resistant to acid, bile, and antibiotics [1].

An in vitro upper-GI digestion model tested a multi-strain capsule at three timings relative to a meal [12].

The bacterial strains survived significantly better with a meal, or 30 minutes before it, than 30 minutes after [12]. S. boulardii's survival was unaffected by timing or by the meal's buffering capacity [12].

A leading manufacturer's FAQ says its capsules can be taken with or without food, and the drug-reference dosing entry sets no meal restriction [13].

No source found a morning-versus-evening difference. General probiotic guidance favours a consistent daily time instead [11].

When co-dosing with antibiotics, spacing the two by 2–3 hours is common probiotic practice — though this particular organism is largely unaffected by antibacterial drugs anyway [11].