Fecal bacteriotherapy for recurrent Clostridium difficile infection
Clostridium difficile infection (CDI) is a nosocomial diarrhea that is increasing in incidence and severity. The spectrum of disease ranges from mild diarrhea to fulminant pseudomembranous colitis that is associated with significant morbidity and mortality. Most cases of CDI have been associated with exposure to antimicrobial agents, and fluoroquinolones, cephalosporins and clindamycin are the agents that are associated with the highest risk. When possible, the offending antibiotic drug should be discontinued, and specific antimicrobial treatment directed against C. difficile should be instituted with oral metronidazole or vancomycin. The frequency of recurrent diarrhea (infection relapse versus reinfection) ranges between 5 and 35%, and risk factors for recurrent infection (RCDI) include age older than 65 years, low serum albumin concentration, recent abdominal surgery, prolonged hospitalization and stay in the intensive care unit. RCDI is generally treated with an additional course of metronidazole or vancomycin, but some patients develop a chronic relapsing pattern of diarrhea after initial retreatment improvement. Alternative treatment strategies for RCDI have recently been described for several new drugs, including rifaximin, nitazoxanide, and tolevamer. In spite of these developments some patients continue to manifest relapsing diarrhea after completed treatment with novel drugs.
Limited experience in northern Europe and the United States with enteral instillation of stool bacterial flora has been successful at breaking the relapsing pattern of diarrhea. Recently, Aas et al. treated 18 patients with RCDI with stool collected from healthy donors, and reported treatment success for 15 of 16 (94%) of their evaluable patients. This article discusses the rationale behind fecal bacteriotherapy for the treatment of RCDI, and reviews the current cumulative experience of published clinical reports in the peer-reviewed literature.
Colonization resistance
The indigenous colon bacterial flora can best be understood as a complex interdependent ecosystem that performs many simultaneous biological tasks. Some of these tasks include the degradation and digestion of food substrates, stimulation of the immune system, and the production of vitamins. One particularly important task is to assist the host in the defense against invasion by exogenous bacterial species. Indigenous bacterial inhabitants of the healthy colon can impede exogenous bacteria from establishing themselves as part of the residential flora through the production of antimicrobial factors, by competing for available binding sites in the epithelial lining of the colon, or by being able to utilize available nutrients and food-sources more efficiently than the competition. This biological defense mechanism has been labeled "colonization resistance" and it represents a primary line of defense against colonization and proliferation of opportunistic pathogenic bacteria, including C. difficile. More than 90% of the fecal biomass consists of strict anaerobic organisms that comprise members of the genera Bacteroides, Bifidobacterium, Eubacterium, Lactobacillus, Peptostreptococcus, Prevotella, Ruminococcus and others. However, the numbers and species diversity of the indigenous bacterial flora varies with age. Comparative studies of feces collected from healthy subjects at different age-ranges showed that the numbers and species diversity of Bifidobacterium decreased with age, whereas the numbers and species diversity of Bacteroides increased in the feces of elderly individuals compared to young adults.
Colonization resistance can prevent colonization and infection by bacterial pathogens in the human colon, unless the integrity of the microbiological barrier becomes disrupted, as may happen with systemic antibiotic therapy. Examination of stool from elderly patients with CDI demonstrated markedly reduced numbers and species diversity of Bacteroides, Prevotella and Bifidobacterium in feces when compared to healthy age matched individuals. Furthermore, repeated episodes of CDI progressively reduced the phylotype richness in the stool of these patients. Thus, it has been postulated that strategies that result in the preservation and restoration of the microbial diversity could be used to resolve RCDI when existing antibiotic treatment modalities have failed. This principle of biotherapy has long been used in veterinary medicine for the treatment of various enteric diseases of ruminants. For example, cud transfers from normal cows can be used to resolve the indigestion of anorexic cows, and young foals can resolve infectious diarrhea by ingesting manure from healthy adult horses (transfaunation). Colonization resistance probably explains the lack of increased risk of clinical CDI in individuals who are symptom-free excretors of C. difficile and prevents colonization with pathogenic C. difficile strains, which is the basis for an oral living vaccine with non-toxigenic C. difficile strains, currently in early clinical trials.
Fecal bacteriotherapy
Fecal bacteriotherapy refers to the process of instilling a liquid suspension of stool from a healthy donor into the gastrointestinal tract. The stool sample may be collected from a patient household contact or an unrelated donor on the day of use. Prior to collection, the sample donor has customarily been screened for potentially contagious infectious agents. After collection, the stool sample is processed in the clinical laboratory into a liquid suspension, and is subsequently instilled into the upper GI tract through a nasoduodenal catheter or into the colon through a colonoscope or a retention enema catheter.
Fecal enemas have been used as treatment for a variety of gastrointestinal diseases, including inflammatory bowel disease, chronic constipation, and pouchitis. A Medline search using the search words pseudomembranous enterocolitis and C. difficile associated diarrhea resulted in 13 published reports about fecal instillation therapy for recurrent episodes of CDI. Eiseman and coworkers were the first investigators to report the successful use of fecal enemas in the management of four patients with pseudomembranous enterocolitis (PEC). Three of their patients had recently undergone major abdominal surgery and almost died before the fecal enemas were employed. More than 20 years later Bowden et al. successfully resolved PEC in 13 of 16 surgical patients (81%) by administering stool collected from in-house family members or healthy medical students or hospital residents. Two of the 16 patients received their fecal instillation via an infusion catheter with tip placement in the proximal jejunum; the rest of the patients were given fecal enemas. Three (19%) of the patients died, but two of these patients died of unrelated illnesses. Tvede and Rask-Madsen successfully treated 6 patients with RCDI in 1989. One patient received stool collected from two separate family members; the remaining 5 patients were treated by rectal instillation of a mixture of ten different facultative aerobic and anaerobic bacterial species diluted in sterile saline, and included Streptococcus (Enterococcus) faecalis, Clostridium innocuum, C. ramnosum, C. bifermentans, Bacteroides ovatus, B. vulgatus, B. thetaiotaomicrom, Escherichia coli (two separate strains), and Peptostreptococcus productus. The bacterial suspension employed by Tvede et al has sometimes been referred to as synthetic stool.
Instillation of donor stool to patients with RCDI.
Once the individual who will serve as stool donor has been adequately prescreened, the actual instillation procedure can be scheduled in an expeditious manner. The patient and the treatment team together must decide on the route of administering the stool sample and the physical location where the stool instillation procedure will take place, which frequently may be performed as an outpatient procedure. Furthermore, in the days leading up to the instillation procedure the recipient gastrointestinal tract should be optimally prepared for acceptance of the donated stool sample. The planning of the actual instillation procedure typically requires scheduling and coordination between the operator of the instillation procedure, the clinical laboratory (typically the microbiology laboratory) and the radiology department at the treating facility.
Route of administration of donor stool to recipient. Instillation of donor feces can be accomplished from either the proximal or the distal end of the gastrointestinal tract. Most reported cases (76% of the cases reported in Table 1) received their fecal infusion through the rectum in the form of an enema. Arguments for instilling the stool sample through the upper GI tract include ease of patient preparation (placement of a nasoduodenal or nasojejunal tube), ease of preparation of stool sample (only a single fecal specimen of limited volume is usually needed), a relatively low risk of complications associated with the upper instillation procedure (perforation of the upper GI tract by the NG tube, and potential aspiration of fecal material are both theoretical complications), and limited cost for the radiology services (to verify that the tip of the infusion catheter is in the correct position). In contrast, even though most reported patients have received their stool instillation through the rectum, this approach is inherently associated with increased risk of complications (colon perforation) as a result of instrumentation of an inflamed colon. Furthermore, lower tract instillation requires a far larger volume of prepared stool, is associated with significant spillage of stool leaking backwards through the rectum after instillation, and frequently requires multiple instillations distributed over multiple days to achieve success. Thus, it is likely that the total cost of the instillation procedure is higher when the instillation is delivered through the lower GI tract.
The finished stool slurry can then be used immediately, or frozen as multiple aliquots stock samples at -20&deg;C for later use when stool has been collected from a healthy single anonymous donor.
Instillation of donor stool. As shown in Table 3, the volume of stool to be instilled and the number of instillations needed vary with the route of administration. Stored samples of frozen stool aliquots should be thawed in their original storage vessel in a waterbath prior to instillation. A nasogastric or nasojejunal catheter should be placed into the upper GI tract just prior to the instillation procedure, and catheter tip position should be verified by X-ray radiography. A single instillation is typically all that is needed when stool is delivered through the upper GI tract, which makes this approach expedient and convenient for the patient. Thirteen of 18 patients (72%) described by Aas were treated in the ambulatory setting and discharged in the afternoon of their procedure. Typically these patients responded quickly to a single stool instillation and reported rapid resolution of abdominal pain, normalization of stool frequency and consistency, and an increased sense of well-being within 24-48 h. In contrast, multiple instillations are frequently needed before the patient develops a normal bowel pattern when the rectal route is used for stool instillation. Thus, repeated instillations are usually necessary and have often been conducted over a period of 3-5 days, which may result in increased cost and loss of patient work productivity. Typically there is a significant back leakage of fecal material around the enema catheter which creates a bit of a mess and may be unappealing to the patient.
Success of fecal bacteriotherapy for RCDI
The published experience with fecal bacteriotherapy is still limited. However, with the increasing frequency and severity of CDI observed in the western world combined with a lack of highly effective treatment options for RCDI, this alternative form of treatment will likely increase in acceptance among patient and their caregivers. The principal potential risk associated with fecal bacteriotherapy is transmission of contagious agents contained in the donor stool, which argues for careful screening of the donor for presence of occult infections prior to the stool transfer process. However, no infectious complications or physical adverse effects from fecal biotherapy have so far been reported in the literature. Furthermore, the published treatment success-rates have varied between 81 and 100%. In fact, 89% of the patients shown in Table 1 successfully resolved their RCDI with fecal biotherapy, all having failed repeated prior treatment attempts with standard antimicrobial therapy. There was no apparent difference in treatment success-rates between patients who received fecal biotherapy administered via the upper or lower gastrointestinal tract, respectively. Fecal biotherapy therefore appears to be both effective and safe.
There is currently no consensus on treatment recommendations for patients who fail to respond to fecal instillation therapy.
Remaining questions
Most researchers agree that the fundamental problem with RCDI results from the absence of healthy bacteria in the colon to keep growth of C. difficile suppressed, rather than the presence of the pathogen per se. However, the relative roles played and the identities of individual bacterial species (especially Bacteroides and Prevotella) that effectively maintain colonization resistance are currently unknown. Future identification of specific key bacteria that are most important for upholding balance in the colon may permit the development of commercial, storable "synthetic stool" products that will resolve many of the practical and perceived esthetic problems that are associated with fecal bacteriotherapy today.