Fecal Transplant vs Vancomycin for Recurrent C Difficile
Fecal Transplant vs Vancomycin for Recurrent C Difficile
In this open-label, randomised controlled clinical trial, we compared the two following treatments: (i) FMT via colonoscopy preceded by bowel cleaning and (ii) the standard vancomycin regimen.
The study was performed at the A. Gemelli University Hospital in Rome. Subjects referred to our hospital were examined by authors involved in the patients' recruitment (G.C., G.I., S.B., G.B. and A.G.) to determine their eligibility for the trial. All enrolled subjects provided their written informed consent. We planned an interim analysis 1 year after the start of the patients' enrolment.
The study protocol was approved by the local ethics committee (A.119/C.E./2013) and registered at ClinicalTrials.gov (NCT02148601). The study was conducted in accordance with the Consolidated Standards of Reporting Trials (CONSORT) Statement.
Patients who were at least 18 years of age, had a life expectancy equal to or longer than 3 months and a recurrence of C. difficile infection after one or more courses of specific antibiotic therapy (at least 10 days of vancomycin at a dosage of at least 125 mg four times daily or at least 10 days of metronidazole at a dosage of 500 mg three times a day) were considered for inclusion. At enrolment, recurrent C. difficile infection was defined as diarrhoea (at least three loose or watery stools per day for 2 or more consecutive days, or at least eight loose stools in 48 h) and positivity in the C. difficile toxin stool test within 10 weeks from the end of the previous antibiotic treatment. To obtain a homogeneous study population in both arms of treatment, only patients we believed able to undergo colonoscopy were enrolled.
The exclusion criteria included prolonged immunodeficiency due to recent chemotherapy; human immunodeficiency virus (HIV) infection; prolonged use of steroids; pregnancy; use of antibiotics other than metronidazole, vancomycin or fidaxomicin at baseline; admission to an intensive care unit; requirement for vasoactive drugs; and other infectious causes of diarrhoea.
Patients were randomly assigned to one of the following treatments: FMT – a short regimen of vancomycin (125 mg by mouth four times a day for 3 days), followed by bowel cleaning with 4 L of macrogol preparation (SELG ESSE) on the last 1 or 2 days (according to the clinical condition of the patients) of antibiotic treatment, followed by consequent faecal infusion from a healthy donor by colonoscopy the next day; or vancomycin) standard vancomycin treatment of 125 mg by mouth four times daily for 10 days, followed by a pulse regimen (125–500 mg/day every 2–3 days) for at least 3 weeks. Patients in whom recurrent C. difficile infection developed after the first faecal infusion were given a second infusion of faeces within 1 week. However, after the enrolment of the first two patients who underwent FMT, this part of the study protocol was amended; thereafter, all subsequent patients with PMC underwent repeated infusions every 3 days until the resolution of colitis. Patients who had to repeat faecal infusion after 3 days were restricted to a light diet and prepared for colonoscopy by taking only 2 L of bowel preparation before the colonoscopy. Patients in whom the two study treatments failed were re-evaluated to establish whether they were able to receive off-protocol treatment with donor faeces. The timeline of the scheduled treatments is shown in Figure 1.
(Enlarge Image)
Figure 1.
Timeline of scheduled treatments after patient randomisation.
Healthy volunteers less than 50 years of age (preferably the patient's relatives or intimates) were initially screened through a specific questionnaire about possible risk factors for potentially transmittable diseases due to their medical history and lifestyle habits. The donors could not have taken antibiotics in the previous 6 months or exhibited significant intestinal symptoms of other intestinal diseases. Other reasons for exclusion of potential donor candidates were: lifestyle associated with increased risk for contracting infections; recent (equal or less than 3 months) travels in tropical areas; new sexual relationship in the last 6 months; recent needle stick accident; previous reception of blood products; body tattoos; gastrointestinal diseases or complaints (abdomen discomfort, alvus disturbances); a family history of gastrointestinal cancer or inflammatory bowel disease; systemic diseases (i.e. diabetes or neurological disorders) or the use of drugs that could be excreted in faeces with potential risk for the recipients.
If no reasons of exclusion from faeces donation were identified through the questionnaire, the candidates underwent blood and stool exams to exclude potentially transmittable diseases. The blood samples were tested for hepatitis A, B and C, antibodies to HIV-1 and -2, Epstein–Barr virus, Treponema pallidum,Strongyloides stercoralis and Entamoeba histolytica. Blood cell counts and measurements of transaminase, C-reactive protein, albumin and creatinine analysis were also performed. The faeces were tested for C. difficile (culture and toxin), enteric bacteria, protozoa and helminths of the large and small bowel, VRE (vancomycin-resistant Enterococci), MRSA (methicillin-resistant Staphylococcus aureus), and Gram-negative MDR (multi-drug-resistant) bacteria.
Before donation, a further questionnaire was used to screen for any recent acute gastrointestinal illnesses, newly contracted infections or other situations that could represent a risk for the patients.
Faeces were collected by the donor on the day of infusion and rapidly transported to our hospital. In the Hospital's Microbiology Laboratory, the faeces were diluted with 500 mL of sterile saline (0.9%). The deriving solution was blended, and the supernatant strained and poured into a sterile container. Within 6 h after the supply of faeces by the donor, the solution was infused (using 50-mL syringes filled with the solution at the time of colonoscopy) through the operative channel of the scope within the proximal tract of the colon. At the time of infusion, the patients were placed in right lateral recumbency position and were asked to maintain this position for at least 1 h after the procedure to facilitate as much as possible the permanence of the material infused into the proximal portions of the colon. On average, the entire infusion procedure was performed within 10 min, and the tube was removed after the infusion. During the insertion and removal of the colonoscope, the endoscopist was able to make a judgment on the inflammatory conditions of the colonic mucosa and report any additional pathological conditions. Finally, the patients were monitored in the recovery room of the Endoscopy Center for 2 h after the procedures.
The primary end point was the resolution of diarrhoea associated with C. difficile infection 10 weeks after the end of the treatments. For patients in the FMT group who required more than one infusion of faeces, follow-up was extended to 10 weeks after the last infusion. For patients in the vancomycin group, follow-up lasted for 10 weeks after the end of the vancomycin course. The secondary end point was toxin negative without recurrent C. difficile infection 5 and 10 weeks after the end of the treatments.
We defined the cure of C. difficile infection as the disappearance of diarrhoea, or persistent diarrhoea explicable by other causes, with two negative stool tests for C. difficile toxin. Recurrence after treatment was defined as diarrhoea (at least three loose or watery stools per day for 2 or more consecutive days, or at least eight loose stools in 48 h) unexplainable by other causes, with or without positive stool toxin within 10 weeks from the end of the therapy.
Patients were closely followed up in the days after treatment, and a stool diary was kept by the patients themselves or by family members or the medical and nursing staff. Patients, family members and referral physicians were also questioned about stool frequency and consistency, medication use, and adverse events in the 7 days after the end of treatment and on weeks 2–10 after the end of treatment. Subjects in the vancomycin group were also questioned during the 14-day treatment period. Stool tests for C. difficile toxin were performed on weeks 5, 10, and whenever diarrhoea occurred using a Premier Toxins A&B (Liaison C. difficile GDH-Toxin A/B – DiaSorin Inc., Stillwater, MN, USA) kit in the central hospital laboratory.
Blocked randomisation of subjects was performed by an external person not involved in the study. An online random number generator software (https://www.sealedenvelope.com/simple-randomiser/v1/lists) was used to provide random permuted blocks with a block size of six and an equal allocation ratio; the sequence was concealed until the interventions were assigned. Because of the intrinsic difference between the two treatments, neither physicians nor patients were blinded to the randomisation groups.
Calculation of sample size was based on the superiority of FMT via colonoscopy over the vancomycin regimen. Respectively, a cure rate of 87% for FMT via colonoscopy and 55% for vancomycin were assumed. Considering a two-tailed α value of 0.05 and a power of 90% (β = 0.10), the enrolment of 41 patients per group was required. Sample size was calculated with an online software (http://www.stat.ubc.ca/~rollin/stats/ssize/b2.html). Considering the possibility of 20% possible dropouts, we planned to enrol 50 patients per group.
Analyses were performed both on an intention-to-treat (ITT) and per protocol (PP) basis. Differences among groups were assessed with Student's t-test for continuous data and with Fisher's exact probability test (using two-tailed P-values) for categorical data. Differences in cure percentages were determined with Fisher's exact probability test (using two-tailed P-values). Because the trial was stopped early after a 1-year interim analysis, the cure rates and odds ratio for the cure rates for the primary end points were calculated with their 99.9% confidence intervals according to the Haybittle–Peto boundary rule (i.e. P < 0.001 for the primary end point). Statistical analyses were performed with an online calculator (http://www.graphpad.com/quickcalcs/) and with Microsoft Excel for Mac (Microsoft Excel, Redmond, WA, USA; Microsoft, 2011).
Material and Methods
Study Design
In this open-label, randomised controlled clinical trial, we compared the two following treatments: (i) FMT via colonoscopy preceded by bowel cleaning and (ii) the standard vancomycin regimen.
The study was performed at the A. Gemelli University Hospital in Rome. Subjects referred to our hospital were examined by authors involved in the patients' recruitment (G.C., G.I., S.B., G.B. and A.G.) to determine their eligibility for the trial. All enrolled subjects provided their written informed consent. We planned an interim analysis 1 year after the start of the patients' enrolment.
The study protocol was approved by the local ethics committee (A.119/C.E./2013) and registered at ClinicalTrials.gov (NCT02148601). The study was conducted in accordance with the Consolidated Standards of Reporting Trials (CONSORT) Statement.
Study Population
Patients who were at least 18 years of age, had a life expectancy equal to or longer than 3 months and a recurrence of C. difficile infection after one or more courses of specific antibiotic therapy (at least 10 days of vancomycin at a dosage of at least 125 mg four times daily or at least 10 days of metronidazole at a dosage of 500 mg three times a day) were considered for inclusion. At enrolment, recurrent C. difficile infection was defined as diarrhoea (at least three loose or watery stools per day for 2 or more consecutive days, or at least eight loose stools in 48 h) and positivity in the C. difficile toxin stool test within 10 weeks from the end of the previous antibiotic treatment. To obtain a homogeneous study population in both arms of treatment, only patients we believed able to undergo colonoscopy were enrolled.
The exclusion criteria included prolonged immunodeficiency due to recent chemotherapy; human immunodeficiency virus (HIV) infection; prolonged use of steroids; pregnancy; use of antibiotics other than metronidazole, vancomycin or fidaxomicin at baseline; admission to an intensive care unit; requirement for vasoactive drugs; and other infectious causes of diarrhoea.
Treatments
Patients were randomly assigned to one of the following treatments: FMT – a short regimen of vancomycin (125 mg by mouth four times a day for 3 days), followed by bowel cleaning with 4 L of macrogol preparation (SELG ESSE) on the last 1 or 2 days (according to the clinical condition of the patients) of antibiotic treatment, followed by consequent faecal infusion from a healthy donor by colonoscopy the next day; or vancomycin) standard vancomycin treatment of 125 mg by mouth four times daily for 10 days, followed by a pulse regimen (125–500 mg/day every 2–3 days) for at least 3 weeks. Patients in whom recurrent C. difficile infection developed after the first faecal infusion were given a second infusion of faeces within 1 week. However, after the enrolment of the first two patients who underwent FMT, this part of the study protocol was amended; thereafter, all subsequent patients with PMC underwent repeated infusions every 3 days until the resolution of colitis. Patients who had to repeat faecal infusion after 3 days were restricted to a light diet and prepared for colonoscopy by taking only 2 L of bowel preparation before the colonoscopy. Patients in whom the two study treatments failed were re-evaluated to establish whether they were able to receive off-protocol treatment with donor faeces. The timeline of the scheduled treatments is shown in Figure 1.
(Enlarge Image)
Figure 1.
Timeline of scheduled treatments after patient randomisation.
Selection and Screening of Donors
Healthy volunteers less than 50 years of age (preferably the patient's relatives or intimates) were initially screened through a specific questionnaire about possible risk factors for potentially transmittable diseases due to their medical history and lifestyle habits. The donors could not have taken antibiotics in the previous 6 months or exhibited significant intestinal symptoms of other intestinal diseases. Other reasons for exclusion of potential donor candidates were: lifestyle associated with increased risk for contracting infections; recent (equal or less than 3 months) travels in tropical areas; new sexual relationship in the last 6 months; recent needle stick accident; previous reception of blood products; body tattoos; gastrointestinal diseases or complaints (abdomen discomfort, alvus disturbances); a family history of gastrointestinal cancer or inflammatory bowel disease; systemic diseases (i.e. diabetes or neurological disorders) or the use of drugs that could be excreted in faeces with potential risk for the recipients.
If no reasons of exclusion from faeces donation were identified through the questionnaire, the candidates underwent blood and stool exams to exclude potentially transmittable diseases. The blood samples were tested for hepatitis A, B and C, antibodies to HIV-1 and -2, Epstein–Barr virus, Treponema pallidum,Strongyloides stercoralis and Entamoeba histolytica. Blood cell counts and measurements of transaminase, C-reactive protein, albumin and creatinine analysis were also performed. The faeces were tested for C. difficile (culture and toxin), enteric bacteria, protozoa and helminths of the large and small bowel, VRE (vancomycin-resistant Enterococci), MRSA (methicillin-resistant Staphylococcus aureus), and Gram-negative MDR (multi-drug-resistant) bacteria.
Before donation, a further questionnaire was used to screen for any recent acute gastrointestinal illnesses, newly contracted infections or other situations that could represent a risk for the patients.
Faecal Infusion Procedure
Faeces were collected by the donor on the day of infusion and rapidly transported to our hospital. In the Hospital's Microbiology Laboratory, the faeces were diluted with 500 mL of sterile saline (0.9%). The deriving solution was blended, and the supernatant strained and poured into a sterile container. Within 6 h after the supply of faeces by the donor, the solution was infused (using 50-mL syringes filled with the solution at the time of colonoscopy) through the operative channel of the scope within the proximal tract of the colon. At the time of infusion, the patients were placed in right lateral recumbency position and were asked to maintain this position for at least 1 h after the procedure to facilitate as much as possible the permanence of the material infused into the proximal portions of the colon. On average, the entire infusion procedure was performed within 10 min, and the tube was removed after the infusion. During the insertion and removal of the colonoscope, the endoscopist was able to make a judgment on the inflammatory conditions of the colonic mucosa and report any additional pathological conditions. Finally, the patients were monitored in the recovery room of the Endoscopy Center for 2 h after the procedures.
Outcomes and Follow-up
The primary end point was the resolution of diarrhoea associated with C. difficile infection 10 weeks after the end of the treatments. For patients in the FMT group who required more than one infusion of faeces, follow-up was extended to 10 weeks after the last infusion. For patients in the vancomycin group, follow-up lasted for 10 weeks after the end of the vancomycin course. The secondary end point was toxin negative without recurrent C. difficile infection 5 and 10 weeks after the end of the treatments.
We defined the cure of C. difficile infection as the disappearance of diarrhoea, or persistent diarrhoea explicable by other causes, with two negative stool tests for C. difficile toxin. Recurrence after treatment was defined as diarrhoea (at least three loose or watery stools per day for 2 or more consecutive days, or at least eight loose stools in 48 h) unexplainable by other causes, with or without positive stool toxin within 10 weeks from the end of the therapy.
Patients were closely followed up in the days after treatment, and a stool diary was kept by the patients themselves or by family members or the medical and nursing staff. Patients, family members and referral physicians were also questioned about stool frequency and consistency, medication use, and adverse events in the 7 days after the end of treatment and on weeks 2–10 after the end of treatment. Subjects in the vancomycin group were also questioned during the 14-day treatment period. Stool tests for C. difficile toxin were performed on weeks 5, 10, and whenever diarrhoea occurred using a Premier Toxins A&B (Liaison C. difficile GDH-Toxin A/B – DiaSorin Inc., Stillwater, MN, USA) kit in the central hospital laboratory.
Randomisation
Blocked randomisation of subjects was performed by an external person not involved in the study. An online random number generator software (https://www.sealedenvelope.com/simple-randomiser/v1/lists) was used to provide random permuted blocks with a block size of six and an equal allocation ratio; the sequence was concealed until the interventions were assigned. Because of the intrinsic difference between the two treatments, neither physicians nor patients were blinded to the randomisation groups.
Statistical Analysis
Calculation of sample size was based on the superiority of FMT via colonoscopy over the vancomycin regimen. Respectively, a cure rate of 87% for FMT via colonoscopy and 55% for vancomycin were assumed. Considering a two-tailed α value of 0.05 and a power of 90% (β = 0.10), the enrolment of 41 patients per group was required. Sample size was calculated with an online software (http://www.stat.ubc.ca/~rollin/stats/ssize/b2.html). Considering the possibility of 20% possible dropouts, we planned to enrol 50 patients per group.
Analyses were performed both on an intention-to-treat (ITT) and per protocol (PP) basis. Differences among groups were assessed with Student's t-test for continuous data and with Fisher's exact probability test (using two-tailed P-values) for categorical data. Differences in cure percentages were determined with Fisher's exact probability test (using two-tailed P-values). Because the trial was stopped early after a 1-year interim analysis, the cure rates and odds ratio for the cure rates for the primary end points were calculated with their 99.9% confidence intervals according to the Haybittle–Peto boundary rule (i.e. P < 0.001 for the primary end point). Statistical analyses were performed with an online calculator (http://www.graphpad.com/quickcalcs/) and with Microsoft Excel for Mac (Microsoft Excel, Redmond, WA, USA; Microsoft, 2011).
