Ulcerative colitis (UC) is difficult to treat and standard therapy does not always induce remission. Fecal microbial transplantation (FMT) is an alternative approach that induced remission in in small series of patients with active UC. We investigated its safety and efficacy in a placebo-controlled, randomized trial.
We performed a parallel study of patients with active UC without infectious diarrhea. Participants were examined by flexible sigmoidoscopy when the study began and then were randomly assigned to groups that received FMT (50 ml, via enema, from healthy anonymous donors; n=38) or placebo (50 ml water enema; n=37) once weekly for 6 weeks. Patients, clinicians, and investigators were blinded to the groups. The primary outcome was remission of UC, defined as a Mayo score ≤ 2 with an endoscopic Mayo score of 0, at week 7. Patients provided stool samples when the study began and during each week of FMT for microbiome analysis. The trial was stopped early for futility by the data monitoring and safety committee, but all patients already enrolled in the trial were allowed to complete the study.
Seventy patients completed the trial (3 dropped out from the placebo group and 2 from the FMT group). Nine patients who received FMT (24%) and 2 who received placebo (5%) were in remission at 7 weeks (a statistically significant difference in risk of 17%; 95% confidence interval, 2%−33%). There was no significant difference in adverse events between groups. Seven of the 9 patients in remission after FMT received fecal material from a single donor. Three of the 4 patients with UC ≤1 year entered remission, compared to 6/34 of those with UC > 1 year (P=.04 Fisher’s exact test). Stool from patients receiving FMT had greater microbial diversity, compared with baseline, than that of patients given the placebo (P=.02, Mann Whitney U test).
FMT induces remission in a significantly greater percentage of patients with active UC than placebo, with no difference in adverse events. Fecal donor and time of UC appear to affect outcomes.