Fecal microbiota transplantation (FMT or donor feces infusion) is effective against recurrent and antibiotic-refractory Clostridium difficile infections (CDI), with a success rate of approximately 90 percent. Restoration of bowel flora and thereby restoration of colonisation resistance is thought to be the mechanism responsible for cure. With the increasing interest for the role of the microbiome in many different disorders, there is also an increasing interest for the application of FMT for other diseases.
In 1958, Eiseman first described four patients with severe antibiotic-associated colitis that were cured by enemas with donor feces (Eiseman et al., 1958) Since then, it was reported as an effective treatment against antibiotic-refractory C. difficile infection in numerous publications including more than 500 patients (Gough et al., 2011; van Nood et al., 2009). After publication of a randomised controlled trial, showing that donor feces infusion (FMT) is superior compared to vancomycin for recurrent Clostridium difficile infection (van Nood et al., 2013), this unconventional treatment approach was also mentioned in guidelines for the treatment of recurrent Clostridium difficile infection (Surawicz et al., 2013).
However, wide application of FMT for patients that might benefit from this treatment is still hampered by a lack of experience in many centers, regulatory authorities struggling with the classification of this unstandardized and theoreticaly harmful medical application of stool, and by fear for infectious complications in (severely ill) patients. Interestingly, besides Clostridium difficile infection, there are more conditions for which FMT may be beneficial. In
57 Department of Gastroenterology, Haga Teaching Hospital, The Hague, the Netherlands.
58 Department of Internal Medicine, Academic Medical Center, University of Amsterdam, the Netherlands.
*Corresponding author email: email@example.com
particular, disorders that are associated with a disturbed bowel microbiota may be influenced by targetting the microbiome with FMT or powerful probiotics.
The use of donor feces infusion for Clostridium difficile infection and potential further applications of this approach will be discussed.
Clostridium difficile Infection
Clostridium difficile infections (CDI) are the most important cause of hospital-acquired infectious diarrhea in the Western world. There is an increasing incidence and severity in the last decade, which seems partly explained by the rise of more virulent strains of Clostridium difficile (Kuijper et al., 2006; Loo et al., 2005). Clostridium difficile is a spore-forming anaerobic gram-positive rod. Virulent Clostridium difficile strains produce the toxins A and B that are able to disrupt colonic epithelial cells, and initiate a local immune response thereby causing diarrhea. The spores formed by Clostridium difficile are able to survive outside an anaerobic environment, and are resistant to many disinfectants (such as hand alcohol) and antibiotics (Hall, 1935; Macleod-Glover and Sadowski, 2010; Viau and Peccia, 2009).
Clostridium difficile infections typically occur in (recently) hospitalised patients that are exposed to antibiotics. This reflects two important observations: (1) contraction of Clostridium difficile is strongly associated with hospitalization with an estimated (asymptomatic) carrier rate in the general population of about 3 percent, that gradually increases with (prolonged) hospitalization (Barbut and Petit, 2001); and (2) the outgrowth of Clostridium difficile strains to pathogenic levels requires the loss of colonization resistance of the healthy microbiome, which is importantly disturbed by antibiotic use (Dethlefsen et al., 2008; Walters et al., 1983). However, other conditions such as chemotherapy, GI surgery, or PPI use also predispose to Clostridium difficile infection, and community acquired Clostridium difficile is increasingly reported (Bauer et al., 2008; Kelly, 2012; Kwok et al., 2012; Rodrigues et al., 2010; Tleyjeh et al., 2012).
An initial episode of Clostridium difficile infection (although self-limiting in most cases) is usually treated with metronidazol or vancomycin orally. After initial treatment, a recurrence occurs in approximately 25 percent of patients, and those patients are at increased risk to develop subsequent recurrences. With subsequent recurrences, the failure rate of antibiotics gradually increases, and some patients develop chronic relapsing and antibiotic refractory Clostridium difficile infection (Maroo and LaMont, 2006). Recently, fidaxomicin was registered for treatment of CDI in the United States and Europe. Fidaxomicin, which relatively preserves the normal bowel flora compared to vancomycin, might by more effective against recurrent Clostridium difficile infection, although limited data are available (Cornely et al., 2012; Crook et al., 2012; Louie et al., 2011).
Donor Feces Infusion for Recurrent Clostridium difficile Infection
Because antibiotic treatment fails in a subset of patients with recurrent Clostridium difficile infection, there is an urgent need for more effective treatment strategies. To date, donor feces infusion (or fecal microbiota transplantation, FMT) seems the only effective alternative available, although unappealing. From 1958 there are more than 500 patients described in case series and case reports reporting successful treatment with FMTs for either antibiotic-associated diarrhea or Clostridium difficile infections (Bakken, 2009; van Nood et al., 2009). The majority of publications is published in the past 5 years, explained by the steep increase in the incidence of CDI and the concomitant increase of patients with recurrent disease. Historically, the majority of successfully treated patients received only one infusion with donor feces, but multiple infusions may be required in certain patients.
Recently, the first randomised trial of FMT for recurrent CDI was published. Patients were randomised to either a standard treatment with 14 days vancomycin orally (with or without a whole bowel lavage at day 5) or to a treatment with FMT (van Nood et al., 2013). Patients randomized to FMT were treated with vancomycin orally during 4 or 5 days followed by bowel lavage before infusion of the fresh donor feces (>50 gram) solution through a nasoduodenal tube. The study was stopped prematurely after an interim analysis, showing an overwhelming difference in outcome between treatment groups. Of 16 patients in the infusion group, 13 (81 percent) had resolution of C. difficile–associated diarrhea after the first infusion. The three remaining patients received a second infusion with feces from a different donor, with resolution in two patients. Overall, 15 of 16 patients were cured by FMT. In contrast, 4 of 13 patients (31 percent) receiving vancomycin alone and in 3 of 13 patients (23 percent) receiving vancomycin with bowel lavage were cured (P < 0.001 for both comparisons with the infusion group). No significant differences in adverse events among the study groups were observed except for short and mild diarrhea and abdominal cramping immediately after infusion of donor feces.
Taken together, FMT is a safe and effective treatment for recurrent Clostridium difficile infection. Patients with multiple relapses have a high failure rate (>50 percent) of current antibiotic therapy (Maroo and LaMont, 2006). After a third relapse, FMT should be considered as a serious treatment option (Surawicz et al., 2013). Whether FMT is safe in patients with severe colitis or immune-compromised patients is unknown, although anecdotal reports of the successful use of FMT as rescue treatment in severely ill patients exist (Brandt et al., 2011; Neemann et al., 2012).
Protocol Fecal Microbiota Transplantation (FMT)
Route of Infusion
A donor feces solution can be infused in the lower or upper gastrointestinal tract with fecal enemas, colonoscopy, or via a nasoduodenal or nasogastric tube. Currently, there is no consensus about the preferred route of infusion; both methods seem effective and safe. The overall reported success rate is >90 percent (Gough et al., 2011).
Pretreatment of Patients
Most patients are pretreated with antibiotics prior to FMT. Although this may not be necessary, for logistical reasons initiation of antibiotic therapy before performing FMT seems daily practice. Bowel lavage before infusion of donor feces has the theoretical advantage of clearing the bowel before infusion of healthy microbiota. We usually prescribe 2 liters of a cetomacrogol solution one day before infusion, regardless the route of administration of FMT. However, there is no study that compared donor feces with or without a whole bowel lavage, and FMT seems also effective without bowel lavage (Gough et al., 2011).
Preparation of Feces
We use freshly produced donor stool (>50 gram) that is dissolved in 500 cc of sterile saline and is infused preferably within 6 hours after collection by the donor. Water and other diluents (e.g., yogurt or milk) have also been described as vehiculum. Relatively more failures of FMT were described in patients receiving smaller amounts of feces (van Nood et al., 2009). Recently, standardized frozen stool samples were also reported to be effective (Hamilton et al., 2012). This could importantly simplify the implementation of FMT.
Protocol for Screening of Donors
Most patients receive donor feces from their spouse or relatives, but feces from healthy volunteers can also be used. To date, donor characteristics that predict the success of FMT are not known. Potential donors should not have an increased risk of (sexually) contractible infectious diseases. Even if the potential donor is a relative or spouse, extensive screening is required.
Initial selection of donors prior to screening of blood and feces is important for prevention of diseases in a window phase at the time of screening. We use a questionnaire, adapted from the questionnaire used for potential blood donors, addressing travel history, sexual behaviour, previous operations, blood transfusions, piercings, and all other interventions that might contribute to carriage of an infectious disease. Any risk of a recently contracted infectious disease that is still
in its window phase (HIV, hepatitis) warrants exclusion of the potential donor. Apart from the risk of transmitting an infectious disease, the risk of transmitting a noninfectious disease has to be taken into account. We also exclude individuals with ulcerative colitis, Crohn’s disease, irritable bowel syndrome, or a known increased colorectal cancer risk as donor.
In the future, thoroughly screened standardized frozen stool batches might overcome hurdles related to donor selection. Donor screening is summarized in Table A13-1.
The Mechanism of Donor Feces Infusions
The human bowel consists trillions of bacteria, with one gram of feces containing 10e11 bacteria (Gill et al., 2006). With 16sRNA sequencing, the composition of the microbiota can be assessed. There are four major bacterial phyla identified (Bacteroidetes, Firmicutes, Actinobacteria, and Proteobacteria). In early childhood, the gut microbiota finds an unique equilibrium and stays relatively stable throughout life (Adlerberth and Wold, 2009). The intestinal mucosa and thereby the microbiota is continuously confronted with passing antigens, food, possible toxic substances, and organisms. In a joint effort with the immune system the microbiota needs to balance between protective reactions against harmful pathogens and tolerance of commensal bacteria and dietary antigens to
|1. Initial screening:||Questionnaire addressing risk factors for potentially transmittable disorders|
|2. Donor screening||Blood tests: Cytomegalovirus (IgG and IgM)|
|Epstein-Barr Virus (VCA IgM, VCA IgG, VCA, antiEBNA)|
|Hepatitis A (total antibodies, and if positive also hepatitis A IgM)|
|Hepatitis B (HbsAg, antiHbsAg, antiHBcore)|
|Hepatitis C (anti HCV)|
|HIV type 1 and 2|
|Human T-lymphotropic virus type I and II (HTLV)|
|Treponema pallidum (TPHA)|
|Entamoeba histolytica (agglutination and dipstick test)|
|Strongyloides stercoralis (ELISA)|
|Bacteriological evaluation by local standards|
|Parasitological evaluation by local standards (triple feces test)|
|Test for Clostridium difficile (toxin ELISA and culture or PCR)|
|3. One day before
donation of feces:
|Questionnaire addressing stool frequency and pattern, general health, use of antibiotics, travel history, and (recent) sexual behaviour|
maintain intestinal homeostasis (Kootte et al., 2012). Colonisation resistance is the ability of the microbiota to prevent the outgrowth of unwanted pathogens. Antibiotic use is an important disruptive factor of normal homeostasis (Perez-Cobas et al., 2012). If colonisation resistance is disturbed, the outgrowth and differentiation of potential pathogenic bacteria such as Clostridium difficile is enabled. Restoration of microbiota seems likely to restore colonisation resistance.
Recent studies have addressed alterations in microbiota composition following FMT. Restoration of disturbed bowel flora is thought to be the principal mechanism responsible for success. The residual colonic microbiota of patients suffering from recurrent Clostridium difficile infection shows a reduced diversity and is deficient in members of the bacterial divisions of Firmicutes and Bacteroidetes (Chang et al., 2008; Khoruts et al., 2010; van Nood et al., 2013). FMT importantly alters the composition of the patient’s microbiota, which a change towards the fecal bacterial composition of the donor that persists over time, and a normalization of the diversity index (van Nood et al., 2013). In particular, there is a restoration of Bacteroidetes species and clostridium clusters IV and XIVa (Firmicutes), whereas Proteobacteria species decrease (Hamilton et al., 2013; Khoruts et al., 2010; van Nood et al., 2013).
Other Potential Applications for Fecal Microbiota Transplantation
Obesity is associated with changed bowel flora composition with a relative abundance of the two dominant bacterial divisions, the Bacteroidetes and the Firmicutes, and colonization of obese mice with “lean microbiota” results in a decrease in total body fat (Bäckhed et al., 2004; Turnbaugh et al., 2006). In addition, Bacteroidetes species are decreased and Firmicutes are increased in feces of obese people compared to lean people. Recently, the results of the FATLOSE trial were published, in which the effect of FMT on insulin resistance (hyperisulinemic clamp with stable isotopes) was assessed in male patients with metabolic syndrome (Vrieze et al., 2012). After 6 weeks, a significant reduction in both periferal and hepatic insulin resistance was observed, implicating substantial effects of whole body glucose metabolism. Moreover, a significant reduction in fasting lipid profiles after allogenic fecal therapy was noted, which is in line with previously published data suggesting a direct effect between duodenal lipid uptake and glucose production orchestrated by a gut microbiota-driven brain-gut axis. The efficacy of FMT may be explained by enhanced production of specific short-chain free fatty acid butyrate produced by the infused lean donor feces, which might restores normal fecal physiology by implantation of missing lean-figure flora components.
Further research is required, and FMT should not be offered to patients with metabolic syndrome outside a strict experimental setting.
FMT for Inflammatory Bowel Disease (IBD)
Several patients with IBD that were successfully treated with FMT are published in case reports (Bennet and Brinkman, 1989; Borody et al., 2003) Currently, there is increasing interest in the role of the microbiota in IBD, and several randomised controlled studies addressing the effect of FMT are recruiting patients. Pending the results of these studies, FMT should not be considered of proven benefit for patients with IBD.
FMT is a treatment modality that is effective in patients with recurrent Clostridium difficile infections. FMT seems safe, but careful donor screening is required. Although new drugs such as fidaxomicin demonstrate lower recurrence rates following initial infection, a subgroup of patients will remain dependent on FMT. The refinement of the protocol and selection of patients for FMT requires further investigation. Recently, the FDA has not approved FMTs as a regular treatment and an “investigational new drug application” (NDA) is required, which may result in reluctance of physicians to offer patients FMT.
In the future, further understanding of the optimal composition of the microbiota to prevent Clostridium difficile outgrowth or influence other disorders may result in the development of standardized mixtures of bacteria with a true therapeutic efficacy. This might finally eliminate the need for donor feces infusion.
Adlerberth, I., and A. E. Wold. 2009. Establishment of the gut microbiota in western infants. Acta Paediatrica 98(2):229-238.
Bäckhed, F., H. Ding, T. Wang, L. V. Hooper, G. Y. Koh, A. Nagy, C. F. Semenkovich, and J.I. Gordon. 2004. The gut microbiota as an environmental factor that regulates fat storage. Proceedings of the National Academy of Sciences USA 101(44):15718-15723.
Bakken, J. S. 2009. Fecal bacteriotherapy for recurrent Clostridium difficile infection. Anaerobe 15(6):285-289.
Barbut, F., and J. C. Petit. 2001. Epidemiology of Clostridium difficile-associated infections. Clinical Microbiology and Infection 7(8):405-410.
Bauer, M. P., A. Goorhuis, T. Koster, S. C. Numan-Ruberg, E. C. Hagen, S. B. Debast, E. J. Kuijper, and J. T. van Dissel. 2008. Community-onset Clostridium difficile-associated diarrhoea not associated with antibiotic usage—two case reports with review of the changing epidemiology of Clostridium difficile-associated diarrhoea. Netherlands Journal of Medicine 66(5):207-211.
Bennet, J. D., and M. Brinkman. 1989. Treatment of ulcerative colitis by implantation of normal colonic flora. Lancet 1(8630):164.
Borody, T. J., E. F. Warren, S. Leis, R. Surace, and O. Ashman. 2003. Treatment of ulcerative colitis using fecal bacteriotherapy. Journal of Clinical Gastroenterology 37(1):42-47.
Brandt, L. J., T. J. Borody, and J. Campbell. 2011. Endoscopic fecal microbiota transplantation: “Firstline” treatment for severe Clostridium difficile infection? Journal of Clinical Gastroenterology 45(8):655-657.
Chang, J. Y., D. A. Antonopoulos, A. Kalra, A. Tonelli, W. T. Khalife, T. M. Schmidt, and V. B. Young. 2008. Decreased diversity of the fecal microbiome in recurrent Clostridium difficile-associated diarrhea. Journal of Infectious Diseases 197(3):435-438.
Cornely, O. A., M. A. Miller, T. J. Louie, D. W. Crook, and S. L. Gorbach. 2012. Treatment of first recurrence of Clostridium difficile infection: Fidaxomicin versus vancomycin. Clinical Infectious Diseases 55 Suppl 2:S154-S161.
Crook, D. W., A. S. Walker, Y. Kean, K. Weiss, O. A. Cornely, M. A. Miller, R. Esposito, T. J. Louie, N. E. Stoesser, B. C. Young, B. J. Angus, S. L. Gorbach, and T. E. Peto. 2012. Fidaxomicin versus vancomycin for Clostridium difficile infection: Meta-analysis of pivotal randomized controlled trials. Clinical Infectious Diseases 55 Suppl 2:S93-103.
Dethlefsen, L., S. Huse, M. L. Sogin, and D. A. Relman. 2008. The pervasive effects of an antibiotic on the human gut microbiota, as revealed by deep 16s rRNA sequencing. PLoS Biology 6(11):e280.
Eiseman, B., W. Silen, G. S. Bascom, and A. J. Kauvar. 1958. Fecal enema as an adjunct in the treatment of pseudomembranous enterocolitis. Surgery 44(5):854-859.
Gill, S. R., M. Pop, R. T. Deboy, P. B. Eckburg, P. J. Turnbaugh, B. S. Samuel, J. I. Gordon, D. A. Relman, C. M. Fraser-Liggett, and K. E. Nelson. 2006. Metagenomic analysis of the human distal gut microbiome. Science 312(5778):1355-1359.
Gough, E., H. Shaikh, and A. R. Manges. 2011. Systematic review of intestinal microbiota transplantation (fecal bacteriotherapy) for recurrent Clostridium difficile infection. Clinical Infectious Diseases 53(10):994-1002.
Hall, I. C., E. O’Toole. 1935. Intestinal flora in new-born infants: With a description of a new pathogenic anaerobe, Bacillus difficilis. American Journal of Diseases in Children 49:390.
Hamilton, M. J., A. R. Weingarden, M. J. Sadowsky, and A. Khoruts. 2012. Standardized frozen preparation for transplantation of fecal microbiota for recurrent Clostridium difficile infection. American Journal of Gastroenterology 107(5):761-767.
Hamilton, M. J., A. R. Weingarden, T. Unno, A. Khoruts, and M. J. Sadowsky. 2013. High-throughput DNA sequence analysis reveals stable engraftment of gut microbiota following transplantation of previously frozen fecal bacteria. Gut Microbes 4(2):125-135.
Kelly, C. P. 2012. Can we identify patients at high risk of recurrent Clostridium difficile infection? Clinical Microbiology and Infection 18 Suppl 6:21-27.
Khoruts, A., J. Dicksved, J. K. Jansson, and M. J. Sadowsky. 2010. Changes in the composition of the human fecal microbiome after bacteriotherapy for recurrent Clostridium difficile-associated diarrhea. Journal of Clinical Gastroenterology 44(5):354-360.
Kootte, R. S., A. Vrieze, F. Holleman, G. M. Dallinga-Thie, E. G. Zoetendal, W. M. de Vos, A. K. Groen, J. B. Hoekstra, E. S. Stroes, and M. Nieuwdorp. 2012. The therapeutic potential of manipulating gut microbiota in obesity and type 2 diabetes mellitus. Diabetes Obesity and Metabolism 14(2):112-120.
Kuijper, E. J., B. Coignard, and P. Tull. 2006. Emergence of Clostridium difficile-associated disease in North America and Europe. Clinical Microbiology and Infection 12(Suppl 6):2-18.
Kwok, C. S., A. K. Arthur, C. I. Anibueze, S. Singh, R. Cavallazzi, and Y. K. Loke. 2012. Risk of Clostridium difficile infection with acid suppressing drugs and antibiotics: Meta-analysis. American Journal of Gastroenterology 107(7):1011-1019.
Loo, V. G., L. Poirier, M. A. Miller, M. Oughton, M. D. Libman, S. Michaud, A. M. Bourgault, T. Nguyen, C. Frenette, M. Kelly, A. Vibien, P. Brassard, S. Fenn, K. Dewar, T. J. Hudson, R. Horn, P. Rene, Y. Monczak, and A. Dascal. 2005. A predominantly clonal multi-institutional outbreak of Clostridium difficile-associated diarrhea with high morbidity and mortality. New England Journal of Medicine 353(23):2442-2449.
Louie, T. J., M. A. Miller, K. M. Mullane, K. Weiss, A. Lentnek, Y. Golan, S. Gorbach, P. Sears, and Y. K. Shue. 2011. Fidaxomicin versus vancomycin for Clostridium difficile infection. New England Journal of Medicine 364(5):422-431.
Macleod-Glover, N., and C. Sadowski. 2010. Efficacy of cleaning products for C. difficile: Environmental strategies to reduce the spread of Clostridium difficile-associated diarrhea in geriatric rehabilitation. Canadian Family Physician 56(5):417-423.
Maroo, S., and J. T. LaMont. 2006. Recurrent Clostridium difficile. Gastroenterology 130(4):1311-1316.
Neemann, K., D. D. Eichele, P. W. Smith, R. Bociek, M. Akhtari, and A. Freifeld. 2012. Fecal microbiota transplantation for fulminant Clostridium difficile infection in an allogeneic stem cell transplant patient. Transplant Infectious Disease 14(6):E161-E165.
Pérez-Cobas, A. E., M. J. Gosalbes, A. Friedrichs, H. Knecht, A. Artacho, K. Eismann, W. Otto, D. Rojo, R. Bargiela, M. von Berger, S. C. Neulinger, C. Däumer, F. A. Heinsen, A. Latorre, C. Barbas, J. Seifert, V. M. Dos Santos, S. J. Ott, M. Ferrer, and A. Moya. 2012. Gut microbiota disturbance during antibiotic therapy: A multi-omic approach. Gut doi:10.1136/gutjnl-2012-303184.
Rodrigues, M. A., R. R. Brady, J. Rodrigues, C. Graham, and A. P. Gibb. 2010. Clostridium difficile infection in general surgery patients; Identification of high-risk populations. International Journal of Surgery 8(5):368-372.
Surawicz, C. M., L. J. Brandt, D. G. Binion, A. N. Ananthakrishnan, S. R. Curry, P. H. Gilligan, L. V. McFarland, M. Mellow, and B. S. Zuckerbraun. 2013. Guidelines for diagnosis, treatment, and prevention of Clostridium difficile infections. American Journal of Gastroenterology 108(4):478-498.
Tleyjeh, I. M., A. A. Bin Abdulhak, M. Riaz, F. A. Alasmari, M. A. Garbati, M. AlGhamdi, A. R. Khan, T. M. Al, P. J. Erwin, T. Ibrahim, A. Allehibi, L. M. Baddour, and A. J. Sutton. 2012. Association between proton pump inhibitor therapy and Clostridium difficile infection: A contemporary systematic review and meta-analysis. PLoS One 7(12):e50836.
Turnbaugh, P. J., R. E. Ley, M. A. Mahowald, V. Magrini, E. R. Mardis, and J. I. Gordon. 2006. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature 444(7122):1027-1031.
van Nood, E., A. Vrieze, M. Nieuwdorp, S. Fuentes, E. G. Zoetendal, W. M. de Vos, C. E. Visser, E. J. Kuijper, J. F. Bartelsman, J. G. Tijssen, P. Speelman, M. G. Dijkgraaf, and J. J. Keller. 2013. Duodenal infusion of donor feces for recurrent Clostridium difficile. New England Journal of Medicine 368(5):407-415.
van Nood, N. E., P. Speelman, E. J. Kuijper, and J. J. Keller. 2009. Struggling with recurrent Clostridium difficile infections: Is donor faeces the solution? Eurosurveillance 14(34).
Viau, E., and J. Peccia. 2009. Survey of wastewater indicators and human pathogen genomes in biosolids produced by class a and class b stabilization treatments. Applied Environmental Microbiology 75(1):164-174.
Vrieze, A., E. van Nood, F. Holleman, J. Salojärvi, R. S. Kootte, J. F. Bartelsman, G. M. Dallinga-Thie, M. T. Ackermans, M. J. Serlie, R. Oozeer, M. Derrien, A. Druesne, J. E. Van Hylckama Vlieg, V. W. Bloks, A. K. Groen, H. G. Heilig, E. G. Zoetendal, E. S. Stroes, W. M. de Vos, J. B. Hoekstra, and M. Nieuwdorp. 2012. Transfer of intestinal microbiota from lean donors increases insulin sensitivity in individuals with metabolic syndrome. Gastroenterology 143(4):913-916.
Walters, B. A., R. Roberts, R. Stafford, and E. Seneviratne. 1983. Relapse of antibiotic associated colitis: Endogenous persistence of Clostridium difficile during vancomycin therapy. Gut 24(3):206-212.