Fecal Microbiota Transplant for Recurrent Clostridium difficile Infection

A Dutch study, the first randomized trial to assess the efficacy of fecal microbiota transplant for treatment of recurrent Clostridium difficile infection, has demonstrated the superiority and safety of this novel approach and suggests that it should become the standard of care.¹ The treatment was so strikingly effective that the study was stopped for ethical reasons after evaluating results from the first 16 patients in the transplant (infusion) group. 

The modality is intuitively satisfying-C difficile colitis is known to be an opportunistic infection occurring after antibiotic-induced decrease in the number and diversity of normal intestinal flora. Fecal transplant aims to replace normal flora, allowing them to out-compete pathogenic bacteria. Cures of recurrent C difficile colitis through fecal microbiota transplant have been reported since 2000, but the procedure has not become the standard of care because randomized trials have been lacking and both patients and clinicians have resisted the handling and administration of donor feces. This study ought to put those concerns to rest, because the current standard of care is so ineffective-the estimated efficacy of antibiotic therapy for a first-time recurrence is 60%; this worsens with each subsequent recurrence. The cure rate in this study of fecal microbiota transplant for recurrent C difficile infection approached 95%.

Adult patients were selected for study inclusion if they had experienced a relapse of C difficile infection after at least 1 course of adequate antibiotic therapy. They were randomized to 1 of 3 treatment groups: 
1. Initial short vancomycin regimen followed by bowel lavage and infusion of donor feces via nasoduodenal tube
2. Control group given the standard 14-day vancomycin regimen
3. Control group given standard 14-day vancomycin regimen with bowel lavage

Other studies have considered donor feces delivery through colonoscope or enema, but this study strictly considered nasoduodenal delivery of donor stools that were collected after administration of a questionnaire screening for transmissible diseases and parasite screening of stools. Donors’ blood samples were screened for antibodies to HIV; human T-lymphotropic virus types 1 and 2; hepatitis A, B, and C viruses; cytomegalovirus; Epstein-Barr virus; Treponema pallidum; Strongyloides stercoralis; and Entamoeba histolytica. A donor pool was created, and screening was repeated every 4 months. Stools were diluted with sterile saline, then stirred and strained before infusion through a nasoduodenal tube. 

The primary end point was cure without relapse within 10 weeks after initiation of therapy, defined by absence of symptoms or persistent diarrhea that could be explained by other causes with 3 consecutive negative stool tests for C difficile toxin. If patients in the infusion group experienced relapse, they were offered a second infusion and follow-up was extended to 10 weeks after the second infusion. Relapse was defined as diarrhea with a positive stool test for C difficile toxin. 

Initial study size had been planned for 40 patients in each of the 3 study groups. Most patients in both control groups experienced relapse of C difficile infection, prompting the institutional review board’s early termination of the trial. Of 16 patients in the infusion (transplant) group, 13 (81%) were cured after the first infusion. Of the 3 not initially cured, 2 were cured with a second infusion, for a total cure rate of 15 of 16 (94%). For the control groups, cure rates were 4 of 13 (31%) in the vancomycin-alone group, and 3 of 13 (23%) in the vancomycin plus bowel lavage. Infusion was statistically superior to both vancomycin control groups (P < .01 for both controls after the first infusion and P < .001 for overall cure rates). After fecal microbiota transplant, patients showed increased fecal bacterial diversity, similar to healthy donors. 

The study is limited-it did not compare fecal infusion by nasoduodenal tube with other known effective methods, including colonoscopic and enema delivery. Enema in particular is a much less expensive approach and could conceivably be performed in primary care offices or at home. And it would have been nice to see a treatment arm that eliminated vancomycin altogether, for a less expensive intervention and a potentially lower adverse-effect profile. For safety’s sake, I’d have appreciated mention of recipients’ non-seroconversion for various blood-borne diseases (although donor screening is as rigorous as that for blood donors). But for the first time, we have clear evidence that virtually all of our patients with recurrent C difficile infection can be cured-and the intervention is low-cost and low-tech.


Reference
1. van Nood E, Vrieze A, Nieuwdorp M, et al. Duodenal Infusion of donor feces for recurrent Clostridium difficile. N Engl J Med. 2013;368:407-415.