The gut reset that failed its biggest test

What the researchers did

The idea behind fecal microbiota transplantation, or FMT, is simple: take stool from a healthy, rigorously screened donor and place it into the gut of someone who is unwell, hoping the transplanted microbes re-set the recipient's microbial community. Earlier, smaller trials of FMT for irritable bowel syndrome (IBS) had shown promise, and IBS is a condition where the gut microbiome is thought by many researchers to play a role. So a team in Norway ran REFIT2, the largest and most rigorous test of the question so far, published in The Lancet in 2026.

The trial enrolled 450 adults aged 18–65 with moderate-to-severe IBS, diagnosed by Rome IV criteria (the standard diagnostic rules for IBS) and a severity score of at least 175 on the IBS-Severity Scoring System, or IBS-SSS — a validated questionnaire where a 75-point drop counts as meaningful improvement. Participants were treated at five hospitals in Norway; 65% were women and the median age was 36. They were randomized 2:1 to a single enema of donor stool from one of three screened donors (299 people) or a placebo enema made from their own stool (151 people). This is a strong design: double-blind means neither participants nor the doctors assessing them knew who got what, and using the patient's own stool as placebo is a clever way to keep everyone's experience identical. The main question: what share of each group improved by at least 75 points after 90 days?

What they found

The headline result is a statistical dead heat. In the donor-stool group, 40% improved by at least 75 points on the IBS-SSS at 90 days, versus 38% in the placebo group — an absolute difference of 1.9 percentage points (95% CI −8.1 to 12.0; p=0.76). In plain terms, the difference was so small that a real effect of this size or smaller can't be distinguished from chance. And the tie held over time: at one year, the numbers were 44% versus 45%.

Nor was there any hidden benefit lurking in secondary measures. The trial found no differences between groups in quality of life, global relief, or abdominal pain. Adverse events were similar in both groups, so the treatment wasn't dangerous here — it just didn't help beyond placebo.

Here is the part that makes this study genuinely interesting rather than merely negative: the transplant biologically worked. Stool analysis showed participants' gut microbiota shifted directionally toward the microbiome of the donor they received — engraftment, meaning the donated microbes actually took up residence, was confirmed. The microbes moved in. The symptoms did not care.

The authors' conclusion is direct: the trial did not show a clinical benefit of FMT in IBS, and the findings suggest that changing the microbiota alone may be insufficient for symptom improvement. They also note that earlier positive trials of FMT for IBS 'might not be robust' when tested under stricter methods. A commentary accompanying the paper, from UCLA, pointed to something the trial deliberately left out: there were no standardized dietary co-interventions — no structured food component alongside the transplant.

“The microbes moved in. The symptoms did not care.”

Why this result matters

If you live with IBS, this result lands on an old wound: the fear that nothing will ever work. IBS is exhausting, and the microbiome industry has spent years selling the gut as a switch you can flip — a reset, a re-seed, a fresh start. This was the strongest possible test of that idea, with the best design and the largest sample to date, and the reset didn't happen. Not because the microbes failed to arrive, but because arriving wasn't enough.

There is something quietly reassuring in that. A 38–40% improvement rate in both groups tells us people with IBS do improve — but it also shows how much of that improvement doesn't come from the expensive, dramatic intervention. And the UCLA commentary's observation about the missing dietary piece points somewhere constructive: the food side of IBS care was never given its strongest test here, which means it remains the sensible, evidence-building place to invest effort.

What the study can't tell us

The trial covered one delivery method (a single enema) and one population — adults 18–65 in Norway with moderate-to-severe symptoms, two-thirds women. It did not test FMT alongside any structured dietary strategy, which the accompanying commentary flagged as an absence. And like all trials of this design, it answers the average question for the group, not whether some individual might respond differently. Funding came from KLINBEFORSK, a Norwegian public clinical research program; the declared conflicts were minor, such as lecture fees from pharmaceutical companies unrelated to this intervention.

What this means for your plate

The study's own conclusion is the guidepost: microbiota modulation alone was insufficient for symptom improvement in IBS — and its commentators pointed to the absence of standardized dietary co-interventions as the open question. The flip of this finding is not another supplement or a stool transplant; it is slower and cheaper. Food-first strategies for IBS, built step by step with a clinician who knows your history, are where the evidence still has room to grow and where this trial directs attention. For decisions about your own symptoms, a qualified clinician — gastroenterologist or dietitian — is the right person to sit down with.