The study, which appears in the New England Journal of Medicine, involved patients who had repeated bouts of diarrhea caused by a bacterium known as Clostridium difficile. So-called C. diff can take over the intestines after antibiotic treatment has killed off the beneficial bacteria found in the gut.
One transplant of fecal material from a volunteer - with its mix of healthy bacteria - resolved severe diarrhea in 13 out of 16 volunteers. Standard treatment with an antibiotic, in comparison, worked in four of 13 patients.
"This is the first hard evidence that has been provided for the treatment," senior author Dr. Josbert Keller of the University of Amsterdam told Reuters Health.
"Those of us who do it know how effective it is," said Dr. Colleen Kelly of Brown University's Alpert Medical School in Providence, Rhode Island, who was not connected with the study but uses fecal transplant in her practice.
About 3 million people in the U.S. are infected annually with C. diff, which spreads mainly through hospitals, nursing homes and doctors' offices.
The bacterium underlying the problem is called "difficile" for a reason. When it controls the gut, it can be difficult to eradicate. Antibiotics typically only work in 15 to 26 percent of patients with C.
"I've done 90 of these now in the last four and a half years. In patient after patient who has failed multiple courses of antibiotic, if you give them a dose of stool, they get better," she told Reuters Health.
Stool transplants have been proposed as one alternative.
Keller and his colleagues compared three treatments in a small trial.
Thirteen volunteers with C. diff received a standard antibiotic, vancomycin, four times a day for 14 days. After 10 weeks, four were free of bacteria-related diarrhea.
Another 13 patients had the same drug therapy after drinking a solution to clean out the bowel, a process known as bowel lavage that is similar to what people go through if they are getting a colonoscopy. That worked in three cases.
The remaining 16 volunteers had a brief treatment with vancomycin, combined with bowel lavage, followed by the infusion of 500 milliliters of diluted donor feces through a tube that went into the nose, down the throat, past the stomach and into the small intestine.
In the three cases where that treatment failed, the doctors re-treated patients with fecal material from a different donor. That worked in all but one case.
Among the volunteers in the non-transplant groups who had a relapse of C. diff, 18 were later given a fecal transplant. It cured 15 of them, although four of the 15 needed two treatments.
All of the donors - people who worked in the hospital who were not involved with patient care - were periodically screened for a variety of diseases.
When side effects were tallied in the transplant group, 94 percent of patients reported diarrhea, 31 percent had cramping and 19 percent had belching, but all of those symptoms disappeared within three hours. Nineteen percent ultimately reported constipation after treatment.
Feces transplants may sound extremely distasteful, but "the patients were desperate because they had had several episodes. There was nothing else they could do. There was no psychological hurdle for them," said Keller.
"I think the 'yuck' thing is overplayed," Kelly said. "There's a desperation when you're this sick this long."
Only eight of the 43 patients in the study were willing to try the treatment after their first relapse of C. diff, Keller and his colleagues reported.
"The efficacy of antibiotic therapy decreases with subsequent recurrences, and it seems reasonable to initiate treatment with donor-feces infusion after the second or third relapse," they wrote.