The Lancet September 24, 2026

Faecal microbiota transplantation in irritable bowel syndrome (REFIT2): a randomised, double-blind, placebo-controlled, phase 3 trial

Peter Holger Johnsen, Frederik Emil Juul, et al.

Bottom Line

A large phase 3 trial demonstrated that a single administration of faecal microbiota transplantation via rectal enema provided no clinical benefit over an autologous placebo in patients with moderate-to-severe irritable bowel syndrome.

Key Findings

1. At 90 days, 119 (40%) participants in the donor faecal microbiota transplantation group achieved an improvement of at least 75 points on the IBS-SSS.
2. Similarly, 57 (38%) participants in the placebo group (autologous FMT) achieved an improvement of at least 75 points on the IBS-SSS.
3. There was no statistically significant difference between the groups, with an absolute difference of 1.9 percentage points (95% CI -8.1 to 12.0; p=0.76).
4. The proportion of participants experiencing adverse events was similar between the intervention and placebo groups.

Study Design

Design
RCT
Double-Blind
Sample
448
Patients
Duration
90 days
Median
Setting
5 hospitals, Norway
Population Men and women aged 18-65 years with moderate-to-severe IBS defined by Rome IV criteria and an IBS-Severity Scoring System (IBS-SSS) score ≥175 points. A colonoscopy within 5 years was required for participants ≥50 years, and negative biopsies for microscopic colitis were required for diarrhoea-predominant IBS.
Intervention Faecal microbiota transplantation (FMT) using faeces samples from healthy donors, delivered as a once-only rectal enema.
Comparator Placebo group receiving autologous FMT (using the participants' own faeces), delivered as a once-only rectal enema.
Outcome Proportion of participants with a reduction of 75 points or more in the IBS-SSS score at 90 days after treatment compared with baseline.

Study Limitations

• The intervention consisted of a single rectal enema; it remains unknown whether repeated administrations or alternative delivery methods (such as oral capsules) might have yielded different results.
• The 90-day primary outcome assessment window might miss very early transient effects or long-term delayed benefits.
• The use of autologous faeces as a placebo might theoretically induce unintended microbiome shifts or contribute to a strong placebo effect.
• The rigid donor selection and single-donor-to-many-recipients approach may have missed necessary individualized microbial matching.

Clinical Significance

The REFIT2 trial provides robust, definitive evidence refuting the efficacy of single-dose faecal microbiota transplantation (FMT) via enema for irritable bowel syndrome (IBS). These negative findings suggest that single-intervention microbiota modulation alone is insufficient to manage the complex pathophysiology of IBS, heavily dampening clinical enthusiasm for using FMT as a standard therapeutic option for these patients.

Historical Context

Irritable bowel syndrome (IBS) is a highly prevalent functional gastrointestinal disorder, and accumulating data over the past decade heavily implicated the gut microbiome in its pathogenesis. Previous, smaller trials of FMT for IBS produced conflicting results—some suggested striking clinical remission (such as the smaller REFIT1 trial or trials using super-donors), while others showed no benefit over placebo. REFIT2 was designed as a definitive, multi-center, phase 3 trial to answer the question using a rigorous autologous placebo control. Its definitively negative results mark a significant pivot in the field, challenging the previously optimistic trajectory of microbiome-modulating therapies for functional bowel disorders.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

How does the use of an autologous placebo in the REFIT2 trial help researchers isolate the true therapeutic effect of a healthy donor's microbiome in IBS compared to using a standard saline enema?

Key Response

Using the patient's own stool (autologous FMT) controls for the procedural effects (bowel prep, enema administration) and the psychological placebo effect, ensuring any benefit seen in the treatment group is solely due to the introduction of a novel, healthy donor microbiome rather than the mechanical or psychological process itself.

Resident
Resident

A patient with severe IBS-D brings you an article about the microbiome and requests an FMT referral. Based on the REFIT2 trial and current standard of care, how should you counsel this patient regarding their request and alternative evidence-based therapies?

Key Response

Residents must practice translating evidence into patient counseling. The REFIT2 trial provides robust phase 3 evidence that FMT via enema offers no clinical benefit for IBS. Counseling should validate the patient's interest in the gut-brain axis while redirecting them toward proven treatments like dietary modifications (low FODMAP), gut-directed psychotherapies, or targeted pharmacotherapy (e.g., eluxadoline, rifaximin).

Fellow
Fellow

Previous smaller, positive FMT trials in IBS often used oral capsules, repeated dosing, or highly specific 'super-donors' implanted via colonoscopy, while REFIT2 used a single rectal enema. How might these differences in route of administration, frequency, and donor matching explain the discordant outcomes in FMT IBS literature?

Key Response

Fellows need to synthesize conflicting literature. A single enema only reaches the distal colon, whereas oral capsules or upper GI delivery impact the small intestine where much of IBS pathophysiology (like SIBO or altered motility) might originate. Furthermore, IBS may require continuous microbiome modification rather than a single engraftment attempt, highlighting the nuance in FMT delivery mechanisms.

Attending
Attending

The microbiome has been heavily marketed as a panacea for functional GI disorders. How does a definitive negative phase 3 trial like REFIT2 change the way we teach trainees about the limitations of translational microbiome research, and how does it reshape our approach to the massive 'placebo response' in functional disorders?

Key Response

Attendings must contextualize negative data. This trial highlights that despite strong preclinical rationale linking dysbiosis to IBS, macroscopic microbiome replacement fails. It emphasizes teaching trainees that functional GI disorders have a notoriously high placebo response rate, which necessitates rigorous trial designs (like autologous FMT) before adopting highly publicized, biologically plausible therapies.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

IBS is a highly heterogeneous disorder with distinct subtypes (IBS-C, IBS-D, IBS-M) and overlapping pathophysiologies (visceral hypersensitivity, altered motility, gut-brain dysregulation). From a methodological standpoint, how does enrolling a broad moderate-to-severe IBS population in a phase 3 trial like REFIT2 risk washing out a true, subtype-specific microbiome therapeutic signal?

Key Response

Researchers must consider phenotype-microbiome interactions. Grouping all IBS subtypes together assumes a uniform microbial deficit. If FMT only works for post-infectious IBS or specific dysbiosis-driven IBS-D, analyzing them as a single monolithic cohort increases type II error, potentially masking a significant effect in a specific endotype.

Journal Editor
Journal Editor

Publishing a negative trial in a premier journal like The Lancet is a deliberate editorial choice. What specific methodological strengths of the REFIT2 trial, such as its choice of placebo, blinding integrity, and choice of primary endpoint, elevate it above previous negative studies to warrant high-impact publication?

Key Response

Editors value rigorous methodology that definitively answers a controversial clinical question. REFIT2's use of an autologous placebo is a gold-standard control that effectively eliminates the massive placebo response seen in functional GI trials. A definitive, well-powered negative trial prevents futile interventions from entering practice, making it highly impactful despite the negative result.

Guideline Committee
Guideline Committee

Current ACG and AGA guidelines strongly recommend against the use of FMT for IBS outside of clinical trials, based primarily on low-to-moderate quality, conflicting phase 2 data. How does the addition of the REFIT2 phase 3 data impact the strength of this recommendation and the overall quality of evidence supporting it in future iterations?

Key Response

Guideline committees use GRADE methodology. REFIT2 provides high-quality, large-scale, phase 3 data confirming the lack of efficacy. This allows the committee to upgrade the level of evidence from 'low/moderate' to 'high', cementing a 'strong recommendation against' FMT for IBS, thereby protecting patients from unproven, costly, and potentially risky off-label procedures.

Clinical Landscape

Noteworthy Related Trials

2018

Johnsen FMT IBS Trial

n = 90 · Lancet Gastroenterol Hepatol

Tested

FMT via colonoscopy

Population

Patients with moderate-to-severe IBS

Comparator

Autologous FMT via colonoscopy

Endpoint

Reduction of greater than 75 points in IBS-SSS at 3 months

Key result: 65 percent of patients in the active FMT group had significant symptom improvement compared to 43 percent in the placebo group.
2019

El-Salhy FMT IBS Trial

n = 165 · Gut

Tested

FMT (30g or 60g) from a super-donor via gastroscope

Population

Patients with moderate-to-severe IBS

Comparator

Autologous FMT (placebo)

Endpoint

Reduction of at least 50 points in IBS-SSS at 3 months

Key result: Significant symptom improvement was seen in 76 percent (30g) and 89 percent (60g) of FMT patients versus 23.6 percent in the placebo group.
2019

Aroniadis FMT IBS-D Trial

n = 48 · Gastroenterology

Tested

FMT via oral capsules

Population

Patients with diarrhea-predominant IBS

Comparator

Placebo capsules

Endpoint

IBS-SSS symptom reduction at 12 weeks

Key result: There was no significant difference in symptom relief between the FMT capsule group and the placebo group.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis
↑