My Healthcare News & Research — September 24, 2026 · Fecal Transplant Shows No Benefit in Phase 3 IBS Trial: One Donor-Stool Enema Did No Better Than Patients’ Own Stool in 448 Adults

On September 24, 2026, The Lancet published online the results of REFIT2, a phase 3 randomised trial run at five hospitals in Norway. It tested faecal microbiota transplantation (FMT) — moving processed stool, and the bacteria in it, from a healthy donor into a patient’s bowel — in 450 adults with moderate-to-severe irritable bowel syndrome (IBS). Each participant received a single rectal enema: two in every three got stool from a healthy donor, and the rest got their own stool back as the placebo. Ninety days later, 40% of the donor group and 38% of the own-stool group had improved by at least 75 points on the standard IBS symptom score. The difference was 1.9 percentage points (95% confidence interval −8.1 to 12.0; p=0.76), too small to tell apart from chance.

A large, double-blind, placebo-controlled trial like this is one of the strongest kinds of evidence for whether a treatment works. The authors’ own conclusion is direct: “The trial did not show a clinical benefit of faecal microbiota transplantation in patients with IBS.” The result matters partly because the same Norwegian research group had reported a positive, smaller pilot trial in 2018, and because earlier IBS trials of FMT had pointed in different directions. The journal ran an accompanying comment the same day under the title “Healthy donor faecal microbiota transplantation in irritable bowel syndrome: interpreting negative findings.”

This report is built from the paper’s PubMed abstract, its PubMed and Crossref records, and the trial’s public registry entry (ClinicalTrials.gov NCT04691544). The full text is behind the journal’s paywall and could not be opened for this report. That means the results of the secondary outcomes, the exact count of side effects and the authors’ own list of limitations are not reported here. Where a detail comes from the registry rather than the paper, the text says so. FMT itself, and its established use in C. difficile infection, is covered in the site’s March 2, 2026 overview of FMT.

Table of Contents

  1. At a Glance
  2. What Was Done
  3. What Was Found, Number by Number
  4. How FMT Is Meant to Work, and Where It Already Does
  5. How It Fits With Earlier Research
  6. Donors, Delivery Routes and the “Super-Donor” Idea
  7. What It Cannot Show and What Remains Unknown
  8. What the Authors, the Journal and the Guidelines Said
  9. Sources and Primary Documents
  10. Key Research Papers
  11. Connections

At a Glance

Back to Table of Contents

What Was Done

Who could join. Participants needed an IBS diagnosis that met the Rome IV criteria, the current international definition of IBS, which groups people by their main bowel pattern (constipation, diarrhoea, mixed or unclassified; see the site’s page on Rome IV subtypes). They also needed a score of at least 175 on the IBS-SSS, the threshold for moderate disease. Everyone aged 50 or older needed a colonoscopy within the previous five years to rule out bowel cancer, and people with diarrhoea-predominant IBS needed negative bowel biopsies to exclude microscopic colitis, a condition that can look like IBS. The registry entry lists further exclusions, among them diabetes, a body-mass index of 35 or more, inflammatory bowel disease, any earlier FMT, antibiotics in the previous 12 weeks, regular use of acid-suppressing drugs, and pregnancy.

How the groups were formed. Participants were randomly assigned in a 2:1 ratio to FMT with stool from healthy donors or to FMT with their own stool. Using a person’s own stool as the dummy treatment, called autologous FMT, means the placebo goes through the same collection, processing and delivery as the real thing. In the group’s 2018 pilot trial, staff not involved in the study made the two kinds of transplant “identical in appearance and temperature.” In REFIT2, according to the abstract, “All investigators, participants, and study personnel involved in treatment administration, patient care, and outcome assessment were masked to treatment allocation throughout the study.”

The treatment. “Treatments were delivered as a once-only rectal enema.” The registry entry lists three donor arms (donor A, donor B and donor C) alongside the own-stool arm. The abstract describes the donors only as “healthy donors.” How they were screened, how much stool each enema contained and whether the bowel was emptied beforehand are in the full text, which could not be read for this report.

The main measure. The primary endpoint was the share of participants whose IBS-SSS score fell by 75 points or more at 90 days after treatment, compared with their score before treatment. It was analysed in “all patients who received their allocated treatment.” Two randomised participants were left out of the analysis “because their allocated treatments were switched,” leaving 448.

The IBS-SSS. The IBS Severity Scoring System was published in 1997 by a British team. It combines pain, abdominal distension, bowel dysfunction and quality of life or general well-being into a score with a maximum of 500. Scores of 75 to 175 count as mild, 175 to 300 as moderate and above 300 as severe; healthy controls scored below 75. The original paper found that “a change of 50 reliably indicating improvement.” REFIT2 set a higher bar for success: a drop of 75 points.

Other registered measures. The registry lists secondary outcomes that include the same 75-point response at one year (day 365), a 14-point gain on the IBS quality-of-life questionnaire at 90 days, adequate relief of symptoms together with a 30% fall in abdominal pain, adverse and serious adverse events to day 90 and day 365, stool consistency in people with diarrhoea-predominant IBS, complete spontaneous bowel movements in people with constipation-predominant IBS, fatigue, anxiety and depression scores, heart-rate variability, and whether a ten-species bacterial profile from the pilot trial could predict who would respond. Exploratory measures cover changes in the microbiome, the immune system, metabolites and the gut barrier. The abstract reports none of these results.

Patients in the planning. The abstract notes that “A patient representative was involved in the planning of the trial.” Side effects were collected by asking participants to report “any adverse event during follow-up to their study contact, preferably by telephone.”

Back to Table of Contents

What Was Found, Number by Number

Reading the confidence interval. The 95% confidence interval is the range of true differences that fit the data reasonably well. Here it runs from 8.1 percentage points in favour of the placebo to 12.0 percentage points in favour of donor stool, and it includes zero. In plain terms, the trial found no detectable difference. Its data are compatible with donor stool being slightly worse than placebo, no better, or at most modestly better. A p-value of 0.76 means that a difference at least this large would be expected often if the two treatments truly had the same effect.

The placebo response. More than a third of the people who received their own stool improved by 75 points or more. Earlier FMT trials also saw sizeable placebo responses: in the group’s 2018 pilot, 43% of the own-stool group responded, and in a 2020 Norwegian trial led from Stord Hospital and the University of Bergen, which used a looser definition of response (a 50-point drop), 23.6% did. A 2019 Swedish trial found that the gut bacteria of people given their own stool also shifted substantially, and its authors suggested that the bowel cleansing before colonoscopy, the processing of the stool, or both, might contribute to changes in symptoms and microbes.

What is not reported here. Results for quality of life, fatigue, pain, bowel habits, the one-year response, subtypes of IBS and the microbiome analyses are not in the abstract, and the full text could not be read. This report therefore does not say whether any secondary outcome or subgroup differed between the groups.

Back to Table of Contents

How FMT Is Meant to Work, and Where It Already Does

The idea. The bowel holds a large community of bacteria and other microbes, together called the gut microbiome. A 2023 review of the FMT trials notes that “A growing number of trials have demonstrated an aberrant gut microbiota composition in IBS, also known as ‘gut dysbiosis’.” The REFIT2 authors set out the reasoning in their abstract: “Gut microbiota composition might have a role in the pathogenesis of irritable bowel syndrome (IBS), and faecal microbiota transplantation has been proposed as a means of restoring healthy gut microbiota for individuals with this condition.” FMT tries to do that in one step, by replacing or reseeding a patient’s gut community with a donor’s. How the gut and the nervous system talk to each other is covered on the site’s gut-brain axis page, and the bowel lining’s role on the gut barrier and microbiome animation.

Where it works: recurrent C. difficile. FMT’s best-established use is against repeated infections with Clostridioides difficile, a bacterium that can take over the bowel after antibiotics. In a Dutch trial published in the New England Journal of Medicine in 2013 and stopped early after an interim analysis, 13 of 16 patients (81%) who received donor stool through a tube into the small intestine were cured after one infusion, and 2 of the remaining 3 after a second infusion from a different donor. With the antibiotic vancomycin alone, 4 of 13 (31%) were cured, and with vancomycin plus bowel washing, 3 of 13 (23%) (P<0.001 for both comparisons). The 2024 guideline of the American Gastroenterological Association (AGA) states that “Fecal microbiota-based therapies are effective therapy to prevent recurrent C difficile in select patients.” The site’s FMT for recurrent C. diff page covers that use in detail.

The same research network, a different disease. Several REFIT2 authors also ran a 2025 Norwegian trial in first-episode C. difficile infection, published in the Annals of Internal Medicine. In that open-label trial of 100 analysed patients, 34 of 51 (66.7%) given FMT and 30 of 49 (61.2%) given vancomycin were cured with no recurrence within 60 days without further treatment, which met the trial’s test for FMT being no worse than the antibiotic. So the network’s FMT methods have produced results in an infection; REFIT2 found no such effect in IBS.

Why IBS may be different. C. difficile infection is caused by one identifiable bacterium. IBS, by contrast, is diagnosed from a pattern of symptoms under the Rome IV criteria rather than from a single identified cause. The REFIT2 authors wrote that their findings “suggest that microbiota modulation alone might be insufficient for symptom improvement in IBS.” A Danish team reached a similar view in 2018, after FMT capsules changed their patients’ gut bacteria but not their symptoms: “Altering the gut microbiota is not enough to obtain clinical improvement in IBS.” Other approaches to IBS that work through the gut, such as the low-FODMAP diet, probiotics and peppermint oil, are described on the site’s IBS pages.

Back to Table of Contents

How It Fits With Earlier Research

Before REFIT2, randomised trials of FMT in IBS had pointed in both directions, and most were small.

Meta-analyses. A 2023 meta-analysis of 8 randomised trials (484 participants) found “no significant benefit in IBS symptoms three months after treatment compared to placebo” (risk ratio 1.19, 95% CI 0.68–2.10) and rated the overall quality of evidence as low. In subgroups, delivery by endoscope improved symptoms while capsules did not. Its authors concluded that there was “insufficient evidence to support or refute the use of FMT for IBS.”

An updated meta-analysis in Gastroenterology, published online on May 19, 2026, pooled 13 randomised trials with 693 patients, searched up to February 20, 2026, so before REFIT2 appeared. In the main analysis the risk ratio for symptoms not improving with FMT was 0.72 (95% CI 0.50–1.03), which the authors described as “FMT may reduce IBS symptoms at 12 weeks compared with placebo, but the evidence is very uncertain.” Subgroups using a single dose (risk ratio 0.62, 95% CI 0.41–0.93) or the Rome IV criteria (0.38, 95% CI 0.17–0.86) favoured FMT. REFIT2 was a single-dose trial in Rome IV patients and found no difference. For scale: REFIT2 analysed 448 participants; the 2023 meta-analysis pooled 484 from 8 trials, and the 2026 meta-analysis 693 from 13.

Back to Table of Contents

Donors, Delivery Routes and the “Super-Donor” Idea

The earlier trials differed in several ways at once: who the donors were, how much stool was given, how it was prepared and which end of the gut it went into. That makes it hard to say which difference explains which result.

The donor. A 2019 review in Frontiers in Cellular and Infection Microbiology noted that “a number of studies suggest that FMT success is dependent on the microbial diversity and composition of the stool donor, leading to the proposition of the existence of FMT super-donors.” The 2020 Stord–Bergen trial used a single donor, chosen as “healthy, well-characterised,” and its authors concluded that “Utilising a well-defined donor with a normal DI and favourable specific microbial signature is essential for successful FMT,” where DI is a dysbiosis index. REFIT2’s registry entry lists three donors. Whether those donors were chosen for any particular microbial profile is not stated in the abstract.

The route. REFIT1 placed stool at the start of the large bowel by colonoscope; the Stord–Bergen trial delivered it from above, by gastroscope; the US and Danish trials used capsules; REFIT2 used a single rectal enema. In 2023, members of the REFIT team published a small proof-of-concept study in eight volunteers, using a contrast fluid with a thickness similar to an FMT. The fluid reached the caecum, the first part of the large bowel, in 100% of participants after colonoscopy, 50% after an enema followed by a positioning procedure and 38% after an enema without it. In the transverse colon the figures were 100%, 88% and 63%. The authors wrote that “colonoscopy is the only method that ensures delivery to the cecum” and that “Studies are needed to see if FMT colon distribution correlates with treatment effectiveness.”

The dose. REFIT1 used 50–80 g of stool once; the Stord–Bergen trial found more responders at 60 g than at 30 g. The amount in each REFIT2 enema is in the full text, which was not available for this report.

The definition of success. REFIT1 and REFIT2 counted a 75-point fall as a response; the Stord–Bergen trial counted a 50-point fall, the change the IBS-SSS developers described as reliable improvement. A lower bar counts more people as responders in every group, including the placebo group.

Back to Table of Contents

What It Cannot Show and What Remains Unknown

Back to Table of Contents

What the Authors, the Journal and the Guidelines Said

The authors. The abstract’s interpretation reads in full: “The trial did not show a clinical benefit of faecal microbiota transplantation in patients with IBS. These findings suggest that microbiota modulation alone might be insufficient for symptom improvement in IBS.” The 23 authors work at hospitals and universities in Tromsø, Harstad, Oslo, Bergen, Ålesund and Kristiansand, with two at a clinical-research organisation in Scotland. In their declarations, four authors report interests: one gave an unpaid talk at a scientific meeting organised by a drug company, one received lecture fees and conference support from drug companies, one sits on data-monitoring committees for two UK academic trials, and one received travel reimbursement from a clinical-nutrition society. The others declare no competing interests.

The journal. The Lancet published a comment alongside the trial by Jacqueline Liu and Andrea Shin of the Vatche and Tamar Manoukian Division of Digestive Diseases at the University of California, Los Angeles, titled “Healthy donor faecal microbiota transplantation in irritable bowel syndrome: interpreting negative findings.” Its text is behind the paywall and was not read for this report, so its arguments are not summarised here. One of the two commentators declares advisory, consulting, travel and speaking relationships with companies and professional societies; the other declares none.

The 2024 AGA guideline. The American Gastroenterological Association’s guideline on fecal microbiota-based therapies, published in Gastroenterology in March 2024, states: “The AGA suggests against the use of conventional fecal microbiota transplant as treatment for inflammatory bowel diseases or irritable bowel syndrome, except in the context of clinical trials.” It concludes that “Fecal microbiota transplant cannot yet be recommended in other gastrointestinal conditions” beyond C. difficile. REFIT2 is the kind of clinical trial that guideline pointed to.

The 2026 meta-analysis. The authors of the Gastroenterology meta-analysis published four months earlier concluded that there was “very low certainty of evidence that FMT improved IBS symptoms,” and that the positive subgroup and per-protocol findings needed “to be interpreted with caution.”

Press coverage. No press release from the journal or the hospitals was found. The only secondary coverage located was a physician’s blog post on September 26 that restated the abstract.

Back to Table of Contents

Sources and Primary Documents

Every figure attributed to REFIT2 above was taken from the paper’s PubMed abstract (the full text is paywalled and was not read) or, where marked, from the trial’s ClinicalTrials.gov record. Bibliographic details were checked against the PubMed and Crossref records.

  1. Johnsen PH, Juul FE, Hoff DAL, Hillestad ER, Holme Ø, Skjevling LK, et al. (2026). Faecal microbiota transplantation in irritable bowel syndrome (REFIT2): a randomised, double-blind, placebo-controlled, phase 3 trial. The Lancet, published online September 24, 2026, S0140-6736(26)01424-8. — doi:10.1016/S0140-6736(26)01424-8 · PubMed PMID: 42785338
  2. Liu J, Shin A (2026). Healthy donor faecal microbiota transplantation in irritable bowel syndrome: interpreting negative findings. The Lancet, published online September 24, 2026, S0140-6736(26)01815-5 (comment). — doi:10.1016/S0140-6736(26)01815-5 · PubMed PMID: 42785337
  3. University Hospital of North Norway (sponsor). Donor Versus Autologous Fecal Microbiota Transplantation for Irritable Bowel Syndrome: a Double Blind, Placebo-Controlled, Randomized Trial. ClinicalTrials.gov registration, first posted December 31, 2020. — ClinicalTrials.gov NCT04691544

Back to Table of Contents

Key Research Papers

Background papers cited above. Each entry was checked against its PubMed record and its Crossref DOI record before it was written down.

  1. Johnsen PH, Hilpüsch F, Cavanagh JP, Leikanger IS, Kolstad C, Valle PC, Goll R (2018). Faecal microbiota transplantation versus placebo for moderate-to-severe irritable bowel syndrome: a double-blind, randomised, placebo-controlled, parallel-group, single-centre trial. Lancet Gastroenterology & Hepatology 3(1):17-24. — PubMed PMID: 29100842
  2. Halkjær SI, Christensen AH, Lo BZS, Browne PD, Günther S, Hansen LH, Petersen AM (2018). Faecal microbiota transplantation alters gut microbiota in patients with irritable bowel syndrome: results from a randomised, double-blind placebo-controlled study. Gut 67(12):2107-2115. — PubMed PMID: 29980607
  3. Aroniadis OC, Brandt LJ, Oneto C, Feuerstadt P, Sherman A, Wolkoff AW, et al. (2019). Faecal microbiota transplantation for diarrhoea-predominant irritable bowel syndrome: a double-blind, randomised, placebo-controlled trial. Lancet Gastroenterology & Hepatology 4(9):675-685. — PubMed PMID: 31326345
  4. Holster S, Lindqvist CM, Repsilber D, Salonen A, de Vos WM, König J, Brummer RJ (2019). The Effect of Allogenic Versus Autologous Fecal Microbiota Transfer on Symptoms, Visceral Perception and Fecal and Mucosal Microbiota in Irritable Bowel Syndrome: A Randomized Controlled Study. Clinical and Translational Gastroenterology 10(4):e00034. — PubMed PMID: 31009405
  5. El-Salhy M, Hatlebakk JG, Gilja OH, Bråthen Kristoffersen A, Hausken T (2020). Efficacy of faecal microbiota transplantation for patients with irritable bowel syndrome in a randomised, double-blind, placebo-controlled study. Gut 69(5):859-867. — PubMed PMID: 31852769
  6. Halkjær SI, Lo B, Cold F, Højer Christensen A, Holster S, König J, et al. (2023). Fecal microbiota transplantation for the treatment of irritable bowel syndrome: A systematic review and meta-analysis. World Journal of Gastroenterology 29(20):3185-3202. — PubMed PMID: 37346153
  7. Aumpan N, Watanabe J, Yuan Y, Kanno T, Leontiadis GI, Chan FKL, Moayyedi P (2026). Fecal Microbiota Transplantation for Symptom Improvement in Patients With Irritable Bowel Syndrome: Systematic Review and Meta-Analysis of Randomized Controlled Trials. Gastroenterology 171(4):667-681. — PubMed PMID: 42162772
  8. Peery AF, Kelly CR, Kao D, Vaughn BP, Lebwohl B, Singh S, et al.; AGA Clinical Guidelines Committee (2024). AGA Clinical Practice Guideline on Fecal Microbiota-Based Therapies for Select Gastrointestinal Diseases. Gastroenterology 166(3):409-434. — PubMed PMID: 38395525
  9. van Nood E, Vrieze A, Nieuwdorp M, Fuentes S, Zoetendal EG, de Vos WM, et al. (2013). Duodenal infusion of donor feces for recurrent Clostridium difficile. New England Journal of Medicine 368(5):407-415. — PubMed PMID: 23323867
  10. Juul FE, Bretthauer M, Johnsen PH, Samy F, Tonby K, Berdal JE, et al. (2025). Fecal Microbiota Transplantation Versus Vancomycin for Primary Clostridioides difficile Infection: A Randomized Controlled Trial. Annals of Internal Medicine 178(7):940-947. — PubMed PMID: 40523286
  11. Wilson BC, Vatanen T, Cutfield WS, O’Sullivan JM (2019). The Super-Donor Phenomenon in Fecal Microbiota Transplantation. Frontiers in Cellular and Infection Microbiology 9:2. — PubMed PMID: 30719428
  12. Skjevling LK, Hanssen HM, Valle PC, Goll R, Juul FE, Arlov Ø, Johnsen PH (2023). Colonic distribution of FMT by different enema procedures compared to colonoscopy – proof of concept study using contrast fluid. BMC Gastroenterology 23(1):363. — PubMed PMID: 37872499
  13. Francis CY, Morris J, Whorwell PJ (1997). The irritable bowel severity scoring system: a simple method of monitoring irritable bowel syndrome and its progress. Alimentary Pharmacology & Therapeutics 11(2):395-402. — PubMed PMID: 9146781

PubMed Topic Searches

  1. PubMed: Randomised trials of FMT in irritable bowel syndrome
  2. PubMed: The FMT super-donor question
  3. PubMed: FMT delivery by enema, colonoscopy and capsules
  4. PubMed: FMT for recurrent Clostridioides difficile
  5. PubMed: Placebo response in IBS trials

Back to Table of Contents

Connections

Back to Table of Contents