What do the trials say — and why is every honest answer strain-shaped?
Probiotics research is a field of thousands of products and a short list of proven organisms — the evidence map is strain × indication, and nothing else generalizes. The strongest results: the De Simone formulation's pouchitis maintenance (probiotics' hardest clinical win), LGG and S. boulardii for antibiotic-associated diarrhea (the Cochrane-backed general use), 35624 for IBS (the ACG-conditional tier). Around them: the early-tier gut-brain signals, the provocative Weizmann personalization studies, the Stanford fermented-foods trial on the food side — and the generic-blend shelf, carrying none of it. This chapter reads the short list, the limits, and the harm card (PROPATRIA) with the strain-first discipline the whole page teaches.
The field where the unit of evidence is the strain teaches strain-first literacy:
The Nobel-bound immunologist pairs Bulgarian longevity with fermented milk and theorizes lactic-acid bacteria against 'autointoxication' — the wrong mechanism, the right instinct, the field's founding text (Chapter 2's story).
Minoru Shirota isolates L. casei Shirota and launches Yakult — the first named-strain commercial probiotic, still sold ninety years on: the strain-named product before the field had the science to deserve it.
Lilly and Stillwell define probiotics by opposition — the organisms that promote growth, against the antibiotics that suppress it. The vocabulary arrives decades before the evidence discipline.
Gorbach and Goldin isolate Lactobacillus rhamnosus GG — bile- and acid-resistant, adherent, and destined to become the most-published probiotic strain in history.
The high-dose 8-strain formulation's pouchitis-maintenance trials land among probiotics' hardest clinical results; Align's 35624 wins the IBS trials — the medical tier and the indication strain, both strain-named.
'Live microorganisms which, when administered in adequate amounts, confer a health benefit on the host' — every clause a future consumer skill: live, adequate, benefit, per strain.
The Human Microbiome Project maps the residents; the Weizmann personalization studies complicate the visitor story; the Stanford trial gives the food route its round; ACG conditionally recommends probiotics for IBS — and the billion-dollar blend shelf keeps selling the vocabulary the short list earned.
POPULATION · Patients after ileal pouch-anal anastomosis, high-dose 8-strain formulation vs placebo for pouchitis relapse prevention
FINDING · The formulation significantly reduced pouchitis relapse versus placebo — one of the strongest clinical results any probiotic product has produced, and the basis of its guideline use in IBD care.
LIMITATION · A specific surgical population, a specific high-dose product (the De Simone formulation at 112.5 billion CFU — not generalizable to the shelf), and gastroenterology-partnered use. The lesson in both directions: the strongest probiotic evidence is the LEAST generalizable to the general aisle.
POPULATION · Adults with IBS across randomized trials of the single strain, composite symptom and pain endpoints, 4–8 weeks
FINDING · Modest improvements in IBS symptom composite scores, abdominal pain, and bloating versus placebo — the strain-level evidence behind the ACG 2021 conditional recommendation.
LIMITATION · Modest effect sizes, industry lineage (the strain's licensee), and the guideline's own honesty: the ACG could not recommend specific strains — the trials are strain-by-strain, and this is the strain with the file. The studied dose is 1 billion CFU — the anti-arms-race lesson on one label.
POPULATION · 36 healthy adults, 10-week high-fermented-food vs high-fiber diets, microbiome diversity and inflammatory-marker endpoints
FINDING · The fermented arm increased microbiome diversity dose-dependently and reduced 19 inflammatory markers; the fiber arm showed minimal 10-week microbiome change — the food route's strongest modern mechanistic evidence.
LIMITATION · Small, healthy, short, no clinical endpoints — and the fiber arm's window was short for fiber's slower biology (fiber's benefits are established elsewhere on longer clocks). A proof-of-concept for the pattern, not a disease trial — graded exactly so.
POPULATION · Healthy adults given antibiotics then probiotics or none, with gut mucosal and stool microbiome tracking
FINDING · The standard multi-strain probiotic DELAYED the native microbiome's post-antibiotic reconstitution in a subset, and mucosal colonization by the supplement strains was limited and highly individual — the visitor model with personalization on top.
LIMITATION · Small, healthy volunteers, provocative design — not a clinical-outcome trial and not a basis for clinical recommendations, but a standing caution against 'one capsule for everyone' certainty and a foundation of the transient teaching this page carries. The field's most useful uncomfortable study.
POPULATION · Women with recurrent UTIs and bacterial vaginosis across randomized trials of the L. rhamnosus GR-1 + L. reuteri RC-14 pair
FINDING · The named pair restored protective vaginal flora and reduced recurrent UTI rates versus placebo — the urogenital chapter our UTI and bladder pages carry in full.
LIMITATION · Modest samples and moderate effect sizes; the evidence belongs to the named pair, not to lactobacilli generally — the strain rule stated from this page's side of the cross-reference. Full story: our UTI page's Chapter 5.