Transplantation of Fecal Microbiota for C. Difficile
Transplantation of Fecal Microbiota for C. Difficile
This report includes the first 43 patients who received FMT for recurrent CDI at the University of Minnesota Fairview Medical Center. All patients were identified by direct referral from clinicians at infectious disease and gastroenterology practices in the Minneapolis and St Paul metropolitan area. Inclusion criteria for FMT included a history of symptomatic, toxin-positive, infection by C. difficile, and at least two documented subsequent recurrences despite use of standard antibiotic therapy. At least one failed antibiotic regimen had to include a minimum of a 6-week course of tapered or pulsed vancomycin dosage, or at least a 1-month vancomycin course followed by a minimum of 2-week rifaximin "chaser." The only exclusion criteria in the protocol were age <18 years and medical fragility from non-C. difficile problems, resulting in life expectancy of <1 year. In the latter situation, we advised patients that the best therapeutic option was an indefinite course of vancomycin. All patients gave informed consent for FMT via colonoscopy, recognizing relatively limited experience with this treatment approach and the intrinsic unknowns associated with its use. The Institutional Review Board at the University of Minnesota approved prospective collection of clinical outcome data (project approval date was 2 October 2009), while recognizing this experience does not constitute a clinical trial, and as such was not designed to test the efficacy of FMT in comparison with any other therapeutic options.
At the start of the program, patients were asked to self-identify potential donors. These included mothers (n=2), daughters (n=1), sons (n=3), wives (n=1), husbands (n=1), and friends (n=2). Before recruitment, the donors were required to submit available medical records and have a separate medical history interview away from the recipient patient. The history included assessment of infectious risk, including identification of known risk factors for HIV and Hepatitis, current communicable diseases, and recent travel to areas of the world with a higher prevalence of diarrheal illnesses. Additional absolute donor exclusion criteria included gastrointestinal co-morbidities and the use of antibiotics within the preceding 3 months. Since gut microbiota are likely involved in various aspects of energy metabolism and the functioning of the immune system, the presence of features of metabolic syndrome, autoimmunity, or allergic diseases were treated as relative exclusion criteria. Donors provided separate informed consent to participate in the protocol, which included risks associated with laboratory screening. The donors underwent serologic testing for HIV and Hepatitis B and C, and stool testing that included screening for routine enteric pathogens, C. difficile toxin B, and examination for ova and parasites, and Giardia and Cryptosporidium antigens.
Given varying logistic difficulties in recruiting individual patient-identified donors, the lack of availability of donor materials when needed, and no evidence to suggest a clear therapeutic advantage of using a related vs. unrelated donor (e.g., son or daughter vs. friend or domestic partner), volunteer donors were recruited into the FMT program. The advantages of this change included removing the burden of donor identification from the patient, improving the efficiency and costs related to donor screening, a more consistent supply of donor fecal microbiota, and the ability to impose extensive and stringent exclusion criteria on donor selection (Supplementary Appendix 1 online). Two unpaid volunteer donors were recruited during this period, and one of them provided the majority of donated fecal material. Donor medical history was reviewed before every donation and complete laboratory screening, as described above, was done every 6 months.
Individual patient-identified donors used in the early phase of the program came into the outpatient endoscopy center 1–2 h before the scheduled procedure. The fecal material was collected in a toilet hat and processed in a dedicated bathroom separate from the procedure room. Approximately 50 g of fecal material was placed into a standard commercial blender (Oster, Subeam, Rye, NY) and homogenized in 250 ml of sterile, non-bacteriostatic normal saline. The slurry was then passed through stainless steel tea strainers to remove larger particles that could interfere with loading the syringes.
The material obtained from volunteer "universal" donors was transported on ice into the laboratory, where it was processed within 2 h of collection. The material was weighed and homogenized in a commercial blender in a dedicated biological cabinet under N2 gas. The slurry was then passed through 2.0, 1.0, 0.5, and 0.25 mm stainless steel laboratory sieves (WS Tyler, Mentor, OH) to remove undigested food and smaller particulate material. The resulting material passing through the 0.25-mm sieve was centrifuged at 6,000 Ă— g for 15 min in a Sorvall SS-34 rotor and resuspended to one-half the original volume in non-bacteriostatic normal saline. The resulting concentrated fecal bacteria suspension was administered to the patient immediately or amended with sterile pharmaceutical grade glycerol (Sigma, St Louis, MO) to a final concentration of 10%, and stored frozen at −80 °C for 1–8 weeks until used. Thawing was done over 2–4 h in an ice bath before the FMT procedure. The frozen preparation was diluted to 250 ml with non-bacteriostatic normal saline before infusion in the donor. This fecal material extract, whether fresh or frozen, was nearly odorless and of reduced viscosity, color, and texture relative to earlier material prepared in the endoscopy center. Filtration of donor material allowed for effortless loading of large tip 60 ml syringes without risk of clogging. All containers, bottles, and sieves used in material preparation were sterilized before use. Fecal material from universal donors was treated in the same manner as that obtained from patient-identified donors.
Patients were maintained on full dose of vancomycin (125 mg, four times daily by mouth) until 2 days before the FMT procedure. The day before the procedure, the patients were prepped using a split dosage polyethylene glycol purge (GoLYTELY or MoviPrep), which is standard in our endoscopy unit, before colonoscopies to wash out residual antibiotic and fecal material. The patients underwent a full colonoscopy under conscious sedation. Mucosal biopsies were taken to rule out lymphocytic colitis in the absence of obvious IBD. The majority of the prepared donor material (220–240 ml) was administered via the colonoscope's biopsy channel into the patient's terminal ileum and cecum. In some cases, however, a small portion (50 ml) was also instilled into colonic areas containing maximal diverticulosis. Recovery procedure was identical to that routinely used for standard colonoscopy patients. All patients were instructed to contact the endoscopist in case of symptom recurrence, and were formally followed in clinic 1–2 months after the procedure. Clearance of CDI was defined by resolution of diarrhea and negative stool testing for C. difficile at 2 months following FMT. All patients in this protocol also participated in a study examining fecal bacterial community structure, which involved collection of fecal specimens on days 3, 7, and 14 and 1, 3, 6, and 12 months after the procedure. The research staff collected these specimens from the patient's places of residence, providing additional opportunities for symptom follow-up.
Non-categorical data were compared using unpaired Student's t-test. Categorical data were compared using Fisher's exact test. GraphPad Prism software (La Jolla, CA) was used to calculate two-tailed and two-sided P values that were calculated with each test, respectively.
Methods
Patients
This report includes the first 43 patients who received FMT for recurrent CDI at the University of Minnesota Fairview Medical Center. All patients were identified by direct referral from clinicians at infectious disease and gastroenterology practices in the Minneapolis and St Paul metropolitan area. Inclusion criteria for FMT included a history of symptomatic, toxin-positive, infection by C. difficile, and at least two documented subsequent recurrences despite use of standard antibiotic therapy. At least one failed antibiotic regimen had to include a minimum of a 6-week course of tapered or pulsed vancomycin dosage, or at least a 1-month vancomycin course followed by a minimum of 2-week rifaximin "chaser." The only exclusion criteria in the protocol were age <18 years and medical fragility from non-C. difficile problems, resulting in life expectancy of <1 year. In the latter situation, we advised patients that the best therapeutic option was an indefinite course of vancomycin. All patients gave informed consent for FMT via colonoscopy, recognizing relatively limited experience with this treatment approach and the intrinsic unknowns associated with its use. The Institutional Review Board at the University of Minnesota approved prospective collection of clinical outcome data (project approval date was 2 October 2009), while recognizing this experience does not constitute a clinical trial, and as such was not designed to test the efficacy of FMT in comparison with any other therapeutic options.
Donor Identification and Screening
At the start of the program, patients were asked to self-identify potential donors. These included mothers (n=2), daughters (n=1), sons (n=3), wives (n=1), husbands (n=1), and friends (n=2). Before recruitment, the donors were required to submit available medical records and have a separate medical history interview away from the recipient patient. The history included assessment of infectious risk, including identification of known risk factors for HIV and Hepatitis, current communicable diseases, and recent travel to areas of the world with a higher prevalence of diarrheal illnesses. Additional absolute donor exclusion criteria included gastrointestinal co-morbidities and the use of antibiotics within the preceding 3 months. Since gut microbiota are likely involved in various aspects of energy metabolism and the functioning of the immune system, the presence of features of metabolic syndrome, autoimmunity, or allergic diseases were treated as relative exclusion criteria. Donors provided separate informed consent to participate in the protocol, which included risks associated with laboratory screening. The donors underwent serologic testing for HIV and Hepatitis B and C, and stool testing that included screening for routine enteric pathogens, C. difficile toxin B, and examination for ova and parasites, and Giardia and Cryptosporidium antigens.
Given varying logistic difficulties in recruiting individual patient-identified donors, the lack of availability of donor materials when needed, and no evidence to suggest a clear therapeutic advantage of using a related vs. unrelated donor (e.g., son or daughter vs. friend or domestic partner), volunteer donors were recruited into the FMT program. The advantages of this change included removing the burden of donor identification from the patient, improving the efficiency and costs related to donor screening, a more consistent supply of donor fecal microbiota, and the ability to impose extensive and stringent exclusion criteria on donor selection (Supplementary Appendix 1 online). Two unpaid volunteer donors were recruited during this period, and one of them provided the majority of donated fecal material. Donor medical history was reviewed before every donation and complete laboratory screening, as described above, was done every 6 months.
Donor Material Preparation
Individual patient-identified donors used in the early phase of the program came into the outpatient endoscopy center 1–2 h before the scheduled procedure. The fecal material was collected in a toilet hat and processed in a dedicated bathroom separate from the procedure room. Approximately 50 g of fecal material was placed into a standard commercial blender (Oster, Subeam, Rye, NY) and homogenized in 250 ml of sterile, non-bacteriostatic normal saline. The slurry was then passed through stainless steel tea strainers to remove larger particles that could interfere with loading the syringes.
The material obtained from volunteer "universal" donors was transported on ice into the laboratory, where it was processed within 2 h of collection. The material was weighed and homogenized in a commercial blender in a dedicated biological cabinet under N2 gas. The slurry was then passed through 2.0, 1.0, 0.5, and 0.25 mm stainless steel laboratory sieves (WS Tyler, Mentor, OH) to remove undigested food and smaller particulate material. The resulting material passing through the 0.25-mm sieve was centrifuged at 6,000 Ă— g for 15 min in a Sorvall SS-34 rotor and resuspended to one-half the original volume in non-bacteriostatic normal saline. The resulting concentrated fecal bacteria suspension was administered to the patient immediately or amended with sterile pharmaceutical grade glycerol (Sigma, St Louis, MO) to a final concentration of 10%, and stored frozen at −80 °C for 1–8 weeks until used. Thawing was done over 2–4 h in an ice bath before the FMT procedure. The frozen preparation was diluted to 250 ml with non-bacteriostatic normal saline before infusion in the donor. This fecal material extract, whether fresh or frozen, was nearly odorless and of reduced viscosity, color, and texture relative to earlier material prepared in the endoscopy center. Filtration of donor material allowed for effortless loading of large tip 60 ml syringes without risk of clogging. All containers, bottles, and sieves used in material preparation were sterilized before use. Fecal material from universal donors was treated in the same manner as that obtained from patient-identified donors.
Transplantation Procedure
Patients were maintained on full dose of vancomycin (125 mg, four times daily by mouth) until 2 days before the FMT procedure. The day before the procedure, the patients were prepped using a split dosage polyethylene glycol purge (GoLYTELY or MoviPrep), which is standard in our endoscopy unit, before colonoscopies to wash out residual antibiotic and fecal material. The patients underwent a full colonoscopy under conscious sedation. Mucosal biopsies were taken to rule out lymphocytic colitis in the absence of obvious IBD. The majority of the prepared donor material (220–240 ml) was administered via the colonoscope's biopsy channel into the patient's terminal ileum and cecum. In some cases, however, a small portion (50 ml) was also instilled into colonic areas containing maximal diverticulosis. Recovery procedure was identical to that routinely used for standard colonoscopy patients. All patients were instructed to contact the endoscopist in case of symptom recurrence, and were formally followed in clinic 1–2 months after the procedure. Clearance of CDI was defined by resolution of diarrhea and negative stool testing for C. difficile at 2 months following FMT. All patients in this protocol also participated in a study examining fecal bacterial community structure, which involved collection of fecal specimens on days 3, 7, and 14 and 1, 3, 6, and 12 months after the procedure. The research staff collected these specimens from the patient's places of residence, providing additional opportunities for symptom follow-up.
Statistical Analysis
Non-categorical data were compared using unpaired Student's t-test. Categorical data were compared using Fisher's exact test. GraphPad Prism software (La Jolla, CA) was used to calculate two-tailed and two-sided P values that were calculated with each test, respectively.
Source...