Probiotics
Summary
Probiotic results depend on the exact strain. In pooled trials, Saccharomyces boulardii and Lactobacillus rhamnosus GG lowered the risk of diarrhea during antibiotic courses, but LGG and an R0011/R0052 combination did no better than placebo in two large trials in children with acute stomach illness. Bloodstream infections have occurred in severely ill people and preterm infants.
Key facts
- ODS: effects can be specific to a strain (genus, species, subspecies and strain designation, e.g. Lacticaseibacillus rhamnosus GG), so results for one strain do not transfer to another[9]
- Single-strain and multi-strain supplements (capsules, powders, liquids) and foods with added probiotics[9]
- S. boulardii: diarrhea risk during antibiotics ↓ (A); LGG: same outcome ↓ (B, significant in children only); LGG and L. rhamnosus R0011 + L. helveticus R0052 in children's acute stomach illness: no effect; mixed strains: LDL ↓ (C)[1][2][3][4][5]
- Measured in colony-forming units (CFU); many products contain 1–10 billion CFU per dose, and higher CFU counts are not necessarily more effective[9]
- Common strains are unlikely to harm healthy people; side effects are usually minor (gas). Bloodstream infections have been linked to probiotics, mostly in severely ill or immunocompromised people[9][7]
- FDA has not approved any probiotic product as a drug or biological product for infants of any age and warns of invasive, potentially fatal infection in hospitalized preterm infants[10]
- 5 graded human outcomes on this page
- October 9, 2026
How it works
EstablishedSome mechanisms, such as inhibiting the growth of pathogenic microorganisms in the gut and producing short-chain fatty acids, are widely shared among probiotic strains. Others, including vitamin synthesis and bile salt metabolism, are species-specific, and cytokine production and immune, endocrine and nervous-system effects are strain-specific. Probiotics may colonize the gut only transiently, in highly individual patterns (established).[9]
What the human research says
Human trials only, one row per outcome. The form column shows what each trial actually tested. Grades follow our evidence-grading policy.
At a glance: strongest evidence per outcome
| Outcome | Grade | Effect | Strongest evidence | Rows |
|---|---|---|---|---|
| Gut | Grade A | Worsened | Meta-analysis · n = 4,780 · 2015 | 4 rows |
| Heart | Grade C | Worsened | Meta-analysis · n = 1,624 · 2015 | 1 row |
Every graded row
| Outcome | Effect | Grade | Best evidence | PMID |
|---|---|---|---|---|
| Diarrhea risk during antibiotic treatment | Worsened | Grade A | Meta-analysis · n = 4,780 · 2015 Children and adults taking antibiotics in 21 RCTs Form: Saccharomyces boulardii (yeast) vs placebo or no treatment · Dose: Various doses (no clear dose-response per ODS) · Duration: Not stated in the abstract Antibiotic-associated diarrhea fell from 18.7% to 8.5% (RR 0.47; number needed to treat 10): in children from 20.9% to 8.8%, in adults from 17.4% to 8.2%. C. difficile-associated diarrhea fell significantly in children only. | 26216624 (opens PubMed)Source [1] |
| Diarrhea risk during antibiotic treatment | Worsened | Grade B | Meta-analysis · n = 1,308 · 2015 Children and adults taking antibiotics in 11 RCTs Form: Lactobacillus rhamnosus GG (LGG) vs placebo or no treatment · Dose: 4 x 10^8 to 12 x 10^10 CFU (per ODS) · Duration: 10 days to 3 months (per ODS) Antibiotic-associated diarrhea fell from 22.4% to 12.3% (RR 0.49; low quality of evidence). Separately, the reduction was significant in children (5 RCTs, n=445) but not in adults (6 RCTs, n=863), except in a subset on H. pylori eradication therapy. | 26365389 (opens PubMed)Source [2] |
| Total and LDL cholesterol | Worsened | Grade C | Meta-analysis · n = 1,624 · 2015 Participants in 30 placebo-controlled RCTs Form: Various live-bacteria probiotic products (no prebiotics); strains varied · Dose: Not stated in the abstract · Duration: Not stated in the abstract Total cholesterol fell by 7.8 mg/dL and LDL by 7.3 mg/dL versus placebo; HDL and triglycerides did not change. Effects were larger with higher starting cholesterol, longer treatment and certain strains, and seemed stronger in studies supported by probiotic companies. | 26512560 (opens PubMed)Source [5] |
| Moderate-to-severe illness and diarrhea duration in children with acute stomach illness | No effect | No effect | RCT · n = 971 · 2018 Children 3 months to 4 years with acute gastroenteritis at 10 US pediatric emergency departments Form: Lactobacillus rhamnosus GG vs placebo · Dose: 1 x 10^10 CFU twice daily · Duration: 5 days Moderate-to-severe gastroenteritis within 14 days: 11.8% with LGG vs 12.6% with placebo (RR 0.96). No differences in diarrhea duration (median 49.7 vs 50.9 hours), vomiting, day-care absence or household spread. | 30462938 (opens PubMed)Source [3] |
| Moderate-to-severe illness and diarrhea duration in children with acute stomach illness | No effect | No effect | RCT · n = 827 · 2018 Children 3 to 48 months with gastroenteritis at 6 Canadian pediatric emergency departments Form: Lactobacillus rhamnosus R0011 + L. helveticus R0052 vs placebo · Dose: 4.0 x 10^9 CFU twice daily · Duration: 5 days Moderate-to-severe gastroenteritis within 14 days: 26.1% with the probiotic vs 24.7% with placebo (odds ratio 1.06). No differences in diarrhea or vomiting duration, unscheduled visits or adverse events. | 30462939 (opens PubMed)Source [4] |
Forms
| Form | Elemental % | Absorption (human data) | GI tolerance | Studied for |
|---|---|---|---|---|
| Lacticaseibacillus (formerly Lactobacillus) rhamnosus GG (LGG) | Varies[9] | Not applicable; acts in the gut, colonization is transient and individual[9] | Linked to bloodstream infection in ICU patients given LGG capsules (genomic match to the product); no rise in Lactobacillus bloodstream infections in Finland's general population after LGG entered dairy products.[7][9] | Diarrhea risk during antibiotics (reduced, children); acute stomach illness in children (no effect) |
| Saccharomyces boulardii (probiotic yeast) | Varies[9] | Not applicable; acts in the gut[9] | ODS reports at least 60 published cases of fungemia with Saccharomyces cerevisiae probiotics, many in ICU patients or people with central catheters or broad-spectrum antimicrobials.[9] | Diarrhea risk during antibiotics (reduced in children and adults) |
| Lactobacillus rhamnosus R0011 + L. helveticus R0052 (two-strain combination) | Varies[4] | Not applicable; acts in the gut[9] | Adverse events were no more common than with placebo (34.8% vs 38.7%) over 5 days in young children.[4] | Acute stomach illness in children (no effect) |
| Multi-strain products (Lactobacillus, Bifidobacterium and others) | Varies[9] | No head-to-head human data between products[9] | Side effects are usually minor, self-limited digestive symptoms such as gas.[9] | Cholesterol (pooled, strain-dependent); gut-symptom scores (species-specific in a 10-trial analysis) |
Studied doses
Doses studied in human trials ranged from 4 x 10^8 to 12 x 10^10 CFU of LGG for 10 days to 3 months in antibiotic-diarrhea trials; the two large children's trials gave 8 x 10^9 to 2 x 10^10 CFU a day for 5 days. The effective amount depends on the strain and product.[9][3][4]
Interactions
Published interactions only: each row cites a source. No row means none was found in our sources, not that a combination is safe.
None found in our sources.
Who should be careful
- Preterm infants
- FDA warns that preterm infants given probiotic products in hospitals are at risk of invasive, potentially fatal disease or infection from the bacteria or yeast they contain, including one infant death in 2023 and more than two dozen other adverse events since 2018.[10]
- Severely ill or immunocompromised people
- Bloodstream bacterial and fungal infections linked to probiotics have occurred mostly in severely ill or immunocompromised people; the WGO advises restricting use in these groups to strains and indications with proven efficacy (ODS). In one ICU study, bloodstream Lactobacillus matched the probiotic given.[9][7]
- People with central lines or in intensive care
- Many reported Saccharomyces fungemia cases involved ICU patients, central venous catheters, tube or IV feeding, or broad-spectrum antimicrobials (ODS).[9]
- Before you start
- Talk to your pharmacist or clinician before adding this supplement, especially if you take prescription medicines, are pregnant or breastfeeding, or have a chronic health condition.[9]
Talk to your pharmacist or clinician before starting a supplement, especially if you take prescription medicines.
Articles about Probiotics
Plain-English answers to common questions, built on the evidence above.
Frequently asked questions
Why does the strain matter?
ODS says probiotics are identified by genus, species, subspecies and an alphanumeric strain designation, and that some effects are strain-specific, so recommendations need to be strain-specific. Two L. rhamnosus strains illustrate this: LGG was studied for antibiotic-related diarrhea, while R0011 was tested in a combination product. Pooling different strains can give misleading conclusions.[9][2][4]
Which strains have evidence for diarrhea during antibiotic courses?
Saccharomyces boulardii (21 trials, 4,780 people) and LGG (risk reduction significant in children). ODS says starting LGG or S. boulardii within 2 days of the first antibiotic dose reduced risk in children and adults aged 18 to 64, with no evidence that multiple strains work better.[1][2][9]
Do probiotics help children with acute stomach illness?
Two large emergency-department trials found no benefit: LGG (971 children, US) and an L. rhamnosus R0011 + L. helveticus R0052 combination (886 children, Canada) did no better than placebo.[3][4]
Do probiotics lower cholesterol?
A 30-trial meta-analysis found small drops in total and LDL cholesterol, but the effect depended on strain, was larger in industry-supported studies, and a review of healthy adults found insufficient evidence. ODS calls the results mixed.[5][8][9]
Do probiotics help gut symptoms such as pain and bloating?
In a meta-analysis of 10 trials, pain scores improved only with products containing certain species (B. breve, B. longum or L. acidophilus); bloating improved with others, and quality of life did not clearly improve. Effects appear species-specific.[6]
How do I read a probiotic label?
ODS says labels must list the total weight of microorganisms, which includes dead cells; CFU may be listed voluntarily. Look for the full strain name and the CFU count at the end of shelf life, not at manufacture; more CFU is not necessarily more effective.[9]
Sources
- [1]Systematic review with meta-analysis: Saccharomyces boulardii in the prevention of antibiotic-associated diarrhoea.. Aliment Pharmacol Ther, 2015. meta-analysis
- [2]Systematic review with meta-analysis: Lactobacillus rhamnosus GG in the prevention of antibiotic-associated diarrhoea in children and adults.. Aliment Pharmacol Ther, 2015. meta-analysis
- [3]Lactobacillus rhamnosus GG versus Placebo for Acute Gastroenteritis in Children.. N Engl J Med, 2018. RCT
- [4]Multicenter Trial of a Combination Probiotic for Children with Gastroenteritis.. N Engl J Med, 2018. RCT
- [5]Effect of Probiotics on Blood Lipid Concentrations: A Meta-Analysis of Randomized Controlled Trials.. Medicine (Baltimore), 2015. meta-analysis
- [6]Effect of probiotic species on irritable bowel syndrome symptoms: A bring up to date meta-analysis.. Rev Esp Enferm Dig, 2013. meta-analysis
- [7]Genomic and epidemiological evidence of bacterial transmission from probiotic capsule to blood in ICU patients.. Nat Med, 2019. observational
- [8]A review of probiotic supplementation in healthy adults: helpful or hype?. Eur J Clin Nutr, 2019. review
- [9]Probiotics: Fact Sheet for Health Professionals (Updated March 25, 2025). NIH Office of Dietary Supplements, 2025. fact-sheet
- [10]FDA Raises Concerns About Probiotic Products Sold for Use in Hospitalized Preterm Infants (October 26, 2023). U.S. Food and Drug Administration, 2023. safety-communication
What changed
- First draft. Every evidence-row PMID checked with NCBI E-utilities (esummary + efetch abstract) and re-read in an independent adversarial pass.