Transplantation of Fecal Microbiota for C. Difficile
The group of patients with recurrent CDI described in this report clearly had refractory disease as evidenced by the average number of sequential relapses and duration of the condition (Table 1). Furthermore, many patients had multiple risk factors for high probability of recurrence, such as history of severe CDI as evidenced by hospitalization, frequent use of non-C. difficile intercurrent antibiotics, and advanced age. All patients failed a long taper or pulsed regimen of vancomycin, and 40% of the patients also failed an additional long course of vancomycin followed by a 2-week rifaximin "chaser" regimen. One of these patients also failed a 4-week course of rifaximin. Several patients (3/43) took 2–4 weeks course of nitazoxanide, which also failed to clear the infection. Patients with IBD were not excluded from the protocol. Thirty-five percent of our patients (14/40) had underlying IBD, including Crohn's disease (6/14), ulcerative colitis (4/14), and lymphocytic colitis (4/14). The patients with IBD were generally younger (Table 2), but did not differ in the refractory nature of CDI or severity of presentation than older patients. However, the majority of patients without underlying IBD had moderate-to-severe diverticulosis.
The overall rate of infection clearance was 86% in response to a single infusion of donor fecal material, as evidenced by symptom resolution and negative PCR testing for C. difficile toxin B after 2 months of follow-up (Table 1). Negative testing for C. difficile toxin B for 2 months was accepted as therapeutic success in patients with underlying IBD, even in the absence of complete symptom resolution. In all, 3/10 patients (30%) who received FMT using material from patient-identified individual donors had a recurrence of CDI. Two standard donors were employed for the remaining 33 cases in this series, but the majority (30/33) were done using material prepared from a single donor. In all, 3/33 patients who received FMT from a standard donor (fresh or frozen) had a recurrence of CDI. The difference in donor source, patient-identified vs. standard, was not significant (P=0.1270). There was no significant difference in clearing the infection with fresh (11/12) or frozen (19/21) donor material. All six patients who experienced recurrence of CDI after FMT were offered a repeat procedure. Two of these patients, both >80 years of age, had multiple other active medical problems and preferred to remain on indefinite treatment with vancomycin. Four other patients were treated with a second infusion, and all cleared the infection bringing the overall success rate to 95% (41/43 patients). All second infusions were performed using the standard donor-derived material. One of the recurrences of CDI occurred in a patient who received his first infusion from the second standard donor. The same donor source was used for his second FMT. Three of the four patients who received a second FMT had underlying IBD, two patients had Crohn's disease, and one had lymphocytic colitis. Finally, the fourth patient had a partial colon resection done for a stricture that developed following her initial CDI episode. She has a colostomy draining her proximal colon and a long segment of residual distal colon. After recurrence of CDI within 3 weeks following her first FMT, we thought it was likely that engraftment in this case was complicated by difficulty in retaining the donor material due to high flow of fecal contents and relatively small size of the infected colon. The second infusion in this case was done with two doses of frozen standard donor material: one via the colostomy into the colon and the other into the jejunum using upper push enteroscopy. C. difficile testing of her fecal material was done weekly in the first month and monthly thereafter. No C. difficile was found over 3 months of follow-up.
No serious adverse events were noted following FMT in any of the patients, with either fresh or frozen materials. A minority of patients (approximately a third) noted some irregularity of bowel movements and excessive flatulence during the first couple of weeks following the procedure, but these symptoms resolved by the time they were seen in clinic follow-up. Enhanced colitis activity in patients with underlying IBD was not observed and there was improvement in overall colitis activity in all patients with ulcerative colitis, although that is easily attributable to clearing the CDI. Interestingly, all diagnoses of lymphocytic colitis were made for the first time from biopsies taken during the colonoscopies performed at the time of FMT. These patients completely normalized their bowel function and had no diarrhea after FMT without any additional medical therapy for lymphocytic colitis. Follow-up biopsies were not performed in these patients when they became asymptomatic.
Results
Demographics
The group of patients with recurrent CDI described in this report clearly had refractory disease as evidenced by the average number of sequential relapses and duration of the condition (Table 1). Furthermore, many patients had multiple risk factors for high probability of recurrence, such as history of severe CDI as evidenced by hospitalization, frequent use of non-C. difficile intercurrent antibiotics, and advanced age. All patients failed a long taper or pulsed regimen of vancomycin, and 40% of the patients also failed an additional long course of vancomycin followed by a 2-week rifaximin "chaser" regimen. One of these patients also failed a 4-week course of rifaximin. Several patients (3/43) took 2–4 weeks course of nitazoxanide, which also failed to clear the infection. Patients with IBD were not excluded from the protocol. Thirty-five percent of our patients (14/40) had underlying IBD, including Crohn's disease (6/14), ulcerative colitis (4/14), and lymphocytic colitis (4/14). The patients with IBD were generally younger (Table 2), but did not differ in the refractory nature of CDI or severity of presentation than older patients. However, the majority of patients without underlying IBD had moderate-to-severe diverticulosis.
Response to Treatment
The overall rate of infection clearance was 86% in response to a single infusion of donor fecal material, as evidenced by symptom resolution and negative PCR testing for C. difficile toxin B after 2 months of follow-up (Table 1). Negative testing for C. difficile toxin B for 2 months was accepted as therapeutic success in patients with underlying IBD, even in the absence of complete symptom resolution. In all, 3/10 patients (30%) who received FMT using material from patient-identified individual donors had a recurrence of CDI. Two standard donors were employed for the remaining 33 cases in this series, but the majority (30/33) were done using material prepared from a single donor. In all, 3/33 patients who received FMT from a standard donor (fresh or frozen) had a recurrence of CDI. The difference in donor source, patient-identified vs. standard, was not significant (P=0.1270). There was no significant difference in clearing the infection with fresh (11/12) or frozen (19/21) donor material. All six patients who experienced recurrence of CDI after FMT were offered a repeat procedure. Two of these patients, both >80 years of age, had multiple other active medical problems and preferred to remain on indefinite treatment with vancomycin. Four other patients were treated with a second infusion, and all cleared the infection bringing the overall success rate to 95% (41/43 patients). All second infusions were performed using the standard donor-derived material. One of the recurrences of CDI occurred in a patient who received his first infusion from the second standard donor. The same donor source was used for his second FMT. Three of the four patients who received a second FMT had underlying IBD, two patients had Crohn's disease, and one had lymphocytic colitis. Finally, the fourth patient had a partial colon resection done for a stricture that developed following her initial CDI episode. She has a colostomy draining her proximal colon and a long segment of residual distal colon. After recurrence of CDI within 3 weeks following her first FMT, we thought it was likely that engraftment in this case was complicated by difficulty in retaining the donor material due to high flow of fecal contents and relatively small size of the infected colon. The second infusion in this case was done with two doses of frozen standard donor material: one via the colostomy into the colon and the other into the jejunum using upper push enteroscopy. C. difficile testing of her fecal material was done weekly in the first month and monthly thereafter. No C. difficile was found over 3 months of follow-up.
No serious adverse events were noted following FMT in any of the patients, with either fresh or frozen materials. A minority of patients (approximately a third) noted some irregularity of bowel movements and excessive flatulence during the first couple of weeks following the procedure, but these symptoms resolved by the time they were seen in clinic follow-up. Enhanced colitis activity in patients with underlying IBD was not observed and there was improvement in overall colitis activity in all patients with ulcerative colitis, although that is easily attributable to clearing the CDI. Interestingly, all diagnoses of lymphocytic colitis were made for the first time from biopsies taken during the colonoscopies performed at the time of FMT. These patients completely normalized their bowel function and had no diarrhea after FMT without any additional medical therapy for lymphocytic colitis. Follow-up biopsies were not performed in these patients when they became asymptomatic.
SHARE
