SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

Aliment Pharmacol Ther 2011; 34: 409–415

Summary

Background  Evidence is emerging regarding the relationship between a dysbiosis of the human gut microbiota and a number of gastrointestinal diseases as well as diseases beyond the gut. Probiotics have been investigated in many gastrointestinal disease states, with variable and often modest outcomes. Faecal transplantation is an alternative approach to manipulate the gut microbiota.

Aim  To review the use of faecal transplantation therapy for the management of gastrointestinal disorders.

Methods  Available articles on faecal transplantation in the management of gastrointestinal disorders were identified using a Pubmed search and bibliographies of review articles on the subject were collated.

Results  A total of 239 patients who had undergone faecal transplantation were reported. Seventeen of 22 studies of faecal transplantation were in fulminant or refractory Clostridium difficile. Studies of faecal transplantation are heterogeneous regarding the patients, donors, screening, methods of administration and definition of response. Faecal transplantation for C. difficile has been demonstrated to be effective in 145/166 (87%) patients. Small numbers of patients are reported to have undergone successful faecal transplantation for irritable bowel syndrome and inflammatory bowel disease.

Conclusions  Faecal transplantation has been reported with good outcomes for fulminant and refractory C. difficile. No adverse effects of faecal transplantation have been reported. However, there are no level 1 data of faecal transplantation and reports to date may suffer from reporting bias of positive outcomes and under-reporting of adverse effects. This therapy holds great promise, where a dysbiosis of the gut microbiota is responsible for disease and further studies are necessary to explore this potential.


Background

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

The possibility of modifying the gut microbiota to replace harmful bacteria with more favourable microbes has been widely explored since Metchnikoff’s observations in 1907 of the potential health benefits of the ‘Bulgarian bacillus’.1 With the application of molecular techniques to the study of gut microbiology, mounting evidence is emerging regarding the relationship between a dysbiosis of the human gut microbiota and a number of gastrointestinal diseases as well as diseases beyond the gut including diabetes and metabolic syndrome.2–5

In vitro studies have demonstrated a positive effect of probiotic bacteria on gut inflammation by modulating gut immune cells.6, 7 Probiotics have been extensively investigated in many gastrointestinal disease states, where an abnormal microbiota is considered pathogenic.8–10 The outcomes of these studies have, however, been variable and modest.10 One confounding factor of the probiotic approach is the comparatively low number and diversity of bacterial species available in a typical commercial probiotic preparation in comparison with the gut microbiota. Furthermore, probiotic bacterial strains may not be able to compete effectively against the complex interactions of an established and adapted indigenous gut microbial community.

An alternative approach is transplantation of the gut microbiota. This is a concept that has been described in ruminants for some time.11 Its use as therapy in humans was first reported by Eiseman et al. in 1958 in the treatment of fulminant pseudomembranous enterocolitis.12 Over the subsequent decades, there have been a small number of case reports and case series of faecal transplantation for Clostridium difficile13–29 and also constipation,16, 30, 31 irritable bowel syndrome16, 30 and inflammatory bowel diseases.16, 30, 32, 33 In recent years, there has been a resurgence of interest in this procedure and its potential to modify the gut microbiota.

Reports of the procedure have originated from Canada and the United States, Australia and Northern Europe, but the methods of faecal transplantation, screening of donors and patient groups treated with this therapy have varied greatly. In this article, we review the use of faecal therapy since the 1958 report of Eiseman et al. Available articles on the use of faecal transplantation in the management of human gastrointestinal disorders, which were identified using a Pubmed search (15 January 2011) and bibliographies of review articles on the subject were collated. Articles including patients who were previously described or articles that were not available in the English language were not reviewed. The included publications encompassed different gastrointestinal pathologies, varying methods of treatment, screening and duration of follow-up. Twenty two reports of faecal transplantation meeting the inclusion criteria were identified. Ten of these were published since 2005, demonstrating the recently renewed interest in this area. In total, there are 239 patients who have undergone faecal transplantation reported.

Patient details

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

The majority of patients undergoing faecal transplantation were treated for C. difficile after standard treatments had failed. In 1989 Borody et al.16 reported 55 patients treated for constipation, diarrhoea, abdominal pain, ulcerative colitis or Crohn’s disease. This report did not specify the numbers of patients with each condition, although out of five cases described in more detail, two patients had irritable bowel syndrome, one ulcerative colitis, one Crohn’s disease and one had C. difficile diarrhoea. Andrews et al.31 described faecal enema treatment for two patients with constipation, and in the recent paper from Grehan et al.,30 nine patients had a diagnosis of constipation or diarrhoea predominant IBS and one patient had Crohn’s disease. One patient in the series from Aas et al.20 had C. difficile diarrhoea on a background of Crohn’s colitis. Seven other patients with ulcerative colitis are reported to have undergone faecal transplantation.32, 33

Faecal transplantation has been described in patients as young as 2 years24 to patients over 90 years of age.23 Several reports include patients with serious co-morbidities. Three of the four patients reported by Eiseman et al.12 were in a critical condition requiring the use of vasopressors. In the patients reported by Bowden et al.,13 eight had a previously treated carcinoma, two had chronic renal failure and two had an aortic aneurysm. In the study by Aas et al.,20 five patients undergoing faecal transplantation were hospitalised and of those treated as out-patients, three were nursing home residents. MacConnachie et al.22 described faecal transplantation in eighteen patients, eleven of whom were hospitalised with significant co-morbidity and a high proportion having hypoalbuminaemia, leucocytosis and renal dysfunction before faecal transplantation. The patient in the report of You and Franzos21 was treated in an intensive care unit with vasopressors and continuous veno-venous haemofiltration.

Donor screening

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

The potential risk of transmission of viral, bacterial or parasitic infection during the course of faecal transplantation is a concern. No guidelines currently exist regarding screening before faecal transplantation. A number of studies have proposed screening procedures.20, 24 In a recent review of faecal tranplantation for recurrent C. difficile,34 Bakken suggests a screening process based on previous studies. However, without established guidelines or data from randomised controlled trials, ethical approval for the procedure has to date depended on physician discretion with patient and donor consent, local hospitals’ or authorities’ approval or occurred within the framework of ethically approved research studies.

Screening methods of stool donors are not always detailed. In the majority of reports, a spouse or partner, close relative or household member of the patient is preferred as the stool donor. However, in a number of reports, donors who are unrelated healthy individuals have been used.13, 18, 33 Earlier cases did not employ rigorous screening protocols, whereas more recently, increased screening of donors’ medical histories, blood and stool tests have been implemented.

Donors have been screened for a history of gastrointestinal illness, cancer or polyps, hospitalisation within the three previous months25 and between 6 weeks33 to 6 months20 without the use of antibiotics. Screening blood tests have included full blood count and liver function tests31 as well as screening of viral pathogens including HIV 1+2,17–20, 22–27, 29, 32 HTLV I/II25 hepatitis A, B and C,18–20, 22–25, 29, 33 CMV, EBV18, 33 and also for Treponema pallidum20, 22–24, 32 and Helicobacter pylori antibody.25

Donor faecal specimens have been screened for Salmonella spp., Shigella spp., Campylobacter spp., Staphylococcus aureus, Aeromonas hydrophila, Yersinia spp., Vibrio parahaemolyticus, Vibrio cholerae, Candida albicans, Escherichia coli O157 and C. difficile toxins A and B.17, 18, 20, 22–29, 33 Stool microscopy has been screened for protozoa (trophozoites and cysts), helminths and ova including Entamoeba histolytica, Giardia lamblia, Microspora spp.,20, 22–25, 27, 33Cryptosporidium spp.,25Dientamoeba fragilis, Blastocystis hominis, Ascaris lumbricoides, trematodes and tape worms (Table 1).20, 22–25, 27, 33

Table 1.   Suggested screening investigations
SampleInvestigation
BloodFull blood count, Liver function tests
 Hepatitis A, B, C
 HIV 1+2, HTLV I/II
 CMV, EBV
 Treponema pallidum
StoolSelective stool culture
 C. difficile toxin A and B
 Microscopy for ova, cysts and parasites

Route of administration

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

The initial report of Eiseman et al. described administration of faecal enemas,12 which has been replicated in other studies.13–15, 17, 18, 21, 25, 30–33 Others have subsequently used instillation via a colonoscope to the right colon19, 26–30 or instillation of donor faeces via nasogastric tube20, 22–24 or duodenal29 or nasojejunal intubation.13, 30 The study of Grehan et al. employed a combination of colonoscopic instillation followed by enemas or nasojejunal tube.30 The majority of studies entailed a single administration of donor faeces. Some studies used repeated infusions over 2–15 days.12–15, 17, 31, 33 In the study by Garborg et al.,29 six patients underwent a second infusion of donor faeces having not responded to the initial transplantation.

Patient preparation

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

Preparation of the patient prior to faecal transplantation has varied depending on the method of administration of the donor stool. Studies in which donor stool is instilled at colonoscopy or via rectal enemas include patient preparation with bowel lavage treatments. Bennet and Brinkmann describe a bowel sterilisation procedure32 prior to transplantation of donor stool. Persky and Brandt described the use of prior bowel lavage with polyethylene glycol.19 The series of Borody et al. in six patients with refractory ulcerative colitis, gave 7–10 days of treatment with vancomycin, metronidazole and rifampicin prior to bowel lavage.33 This protocol was repeated in the study by Grehan et al.30 Two recent studies stopped treatment with metronidazole or vancomycin 24–48 h prior to faecal transplantation.27, 29 The study by Silverman et al., included prior treatment with Saccharomyces boulardii that was continued up to 60 days after the procedure.25 Patients treated at one centre in the study by Rholke et al.26 were treated with loperamide immediately following the procedure and again 6 h later to maximise contact time of the donor faeces with the colonic mucosa.

Studies of faecal transplantation administered into the upper gastrointestinal tract, do not report the use of prior bowel lavage. The method described by Aas et al. in 2003 and followed by those of MacConnachie, Rubin and Russell et al., includes pre-treatment with more than 4 days of vancomycin and 20 mg of omeprazole the evening before and the morning of the faecal transplantation procedure.20, 22–24

Preparation of donor stool

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

The interval between obtaining donor stool and its administration to the patient has varied between studies, from 24 h before, 6 h before20, 22–24 or immediately. One study homogenised donor stool in pasteurised cow’s milk and filtered the solution that was then stored at −20 °C and thawed in water at 37 °C 30–60 min prior to administration as an enema.18 Some studies have described the homogenisation of the stool and filtering to remove debris. The use of between 10 and 200 g of stool, diluted in 20–500 mL sterile saline, has been reported depending on the method of administration. Studies using an upper gastrointestinal protocol for faecal transplantation instilled between 30 and 50 g of stool homogenised with 50–250 mL sterile saline (Table 2).

Table 2.   Methods of faecal transplantation
 Upper gastrointestinal tractLower gastrointestinal tract
Donor stool collection prior to transplantation≤6 h13, 20, 22, 24, 30≤24 h14, 17, 25–27, 30, 33
Bowel cleansingNo13, 20, 22, 24, 30Yes (colonic instillation)20, 26–28, 32, 33
Donor stool volume30–50 g20, 22, 24, 2910–200 g15, 18, 19, 21, 25–30, 33
Volume of dilution in saline50–250 mL20, 22, 24, 2920–500 mL15, 17–19, 21, 25–30, 33
Volume instilled25–200 mL20, 22, 24, 2920–500 mL15, 17–19, 21, 25–30, 33
Repeated instillationNo13, 20, 22, 24 Yes29No18, 19, 26–28 Yes12–14, 17, 25, 29, 30, 33

Outcomes

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

In many reports of faecal transplantation, response is not clearly defined. Resolution of symptoms is most commonly stated. Some papers include absence of C. difficile toxin. In the 1989 paper by Borody et al., the indication for faecal transplantation in 50 of 55 patients treated was not stated. In this study, however, 20 patients were cured, 26 responded and nine patients did not respond to faecal transplantation.16 In the paper by Grehan et al., outcomes were not stated.30

Excluding these studies, faecal transplantation for C. difficile has been demonstrated to be effective in 145/166 (87%) patients. Time to response is often not stated, although ‘immediate’, ‘prompt’ or ‘rapid’ response is often reported. Where time to response is stated, this has been recorded to occur within 24 h to twelve days.13, 18, 18, 24, 29, 33 Response appears durable with follow-up of patients up to 8 years following faecal transplantation.27

In the initial report of Eiseman et al., three of the four patients were described as terminally or critically ill. All of these had cessation of diarrhoea and were completely asymptomatic between 24 h and 10 days following faecal transplantation. The report of Bowden et al. describes response as a reduction in frequency of bowel motions, absence of fever, normalisation of leucocyte counts and increased general well being. Tvede and Rask-Madsen describe normalisation of bowel function as well as reduction in inflammatory markers and increased albumin levels as response to faecal transplantation. In the report of You et al., the patient rapidly displayed normalisation of leucocytosis, stabilisation of blood pressure enabling cessation of vasopressors and improvement in renal function allowing cessation of continuous veno-venous haemofiltration as well as normalisation of bowel function. In the reports of Schwann et al., Gustaffson et al., Persky and Brandt, Aas et al., MacConnachie et al., Khoruts et al., Rholke et al. and Russell et al., cessation of diarrhoea is defined as response. Five of these studies also document a change from a positive to a negative C. difficile stool test.

For ulcerative colitis, of eight patients reported, all have responded and have remained in remission from 6 months to 13 years.32, 33 Patients with ulcerative colitis in the series of Borody et al. responded within 1–6 weeks, and were considered in remission by 4 months following faecal transplantation.33 Five of the six patients reported in this series had moderate to severe disease with moderate to severe endoscopic findings. All of the patients were asymptomatic with no endoscopic evidence of active inflammation following faecal transplantation. (Table 3).

Table 3.   Summary of the outcome studies of faecal transplantation
AuthorYearIndicationNumber of PatientsRoute of faecal instillationResponseStated time to responseDuration of follow-up
  1. PMC, pseudomembranous colitis; CDAD, C. difficile-associated diarrhoea; AAD, antibiotic associated diarrhoea; IBS, irritable bowel syndrome; IBD, inflammatory bowel disease; UC, ulcerative colitis.

Eiseman1958PMC 4Rectal4/42 days 
Bowden1981PMC16Rectal/Jejunal14/161–12 days5 days– 3 years
Schwan1984Relapsing CDAD 1Enema1/1 9 months
Tvede1989Relapsing CDAD 2Enema1/2 6 months
Bennet1989UC 1Enema1/1 6 months
Borody1989IBS, IBD, CDAD55Enema26 cure 20 response 9 no response 1–12 months
Andrews1992Constipation 1Enema1/1 18 months
Paterson1994Chronic CDAD 7Enema7/7 2 years
Gustaffson1998AAD/CDAD 9Enema9/96–10 days18 months
Persky2000Recurrent CDAD 1Colonic1/1 5 years
Aas2003Recurrent CDAD18Nasogastric15/18 90 days
Borody2003UC 6Enema6/61–6 weeks1–13 years
You2008Fulminant CDAD 1Enema1/136 h 
MacConnachie2009Recurrent CDAD15Nasogastric11/15 4–24 weeks
Rubin2009CDAD12Nasogastric10/12 90 days
Khoruts2010Chronic CDAD1Colonic1/12 days6 months
Rholke2010Relapsing CDAD19Colonic19/19 6 months– 5 years
Russell2010Relapsing CDAD 1Nasogastric1/136 h6 months
Yoon2010Refractory/Recurrent CDAD12Colonic12/12 3 weeks– 8 years
Garborg2010Recurrent CDAD40Duodenal/Colonic33/4024 h 
Silverman2010Chronic CDAD 7Enema7/7  

Adverse events

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

No studies of faecal transplantation report any adverse events related to the procedure. Some studies report patient deaths due to the underlying disease, where the patient has not responded to the faecal transplantation. In one study in which donor faeces were instilled via a nasogastric tube, the patient died of peritonitis. Although considered unlikely, the nasogastric tube insertion could not be discounted to have been contributory.17 One patient in the study by Silverman et al. developed irritable bowel symptoms following faecal transplantation.25

Analysis of effects of faecal transplantation on stool composition and faecal microbiota

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

Four studies have attempted to analyse stool before and after faecal transplantation. Using culture, Tvede and Rask-Madsen observed an absence of Bacteroides before bacteriotherapy and during vancomycin therapy whilst patients were symptomatic. During follow-up after bacteriotherapy (including faecal enemas in two patients), Bacteroides were regularly cultured.15 Gustafsson et al. studied stool short chain fatty acid concentrations before and after faecal transplantation in nine patients. All short chain fatty acids were found to be reduced in the patient group compared with healthy adults and following faecal enema therapy, the relative distribution and absolute amounts of short chain fatty acids returned to patterns similar to healthy adults.18 More recently, using modern molecular 16S rRNA gene sequencing techniques, two studies have shown a significant change in the microbiota following faecal transplantation. Khoruts et al. demonstrated a reduction in Bacteroidetes and Firmicutes in a patient with C. difficile diarrhoea. Following faecal transplantation, there was a rapid change in the patient’s microbiota to a composition that was highly similar to that of the healthy donor for at least 4 weeks (the duration of follow-up stool analysis).28 Grehan et al. undertook analysis on the stool of 10 patients who underwent faecal transplantation. A dramatic change was shown in the recipients’ microbiota to a composition similar to their donors’ microbiota. This study analysed stool from patients up to 24 weeks following faecal transplantation demonstrating a durable change in the recipients’ microbiota up to 24 weeks.30

Conclusions

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References

Evidence regarding the use of faecal transplantation as a means of modifying the gut microbiota and effecting cure of gastrointestinal illness is accumulating. To date the majority of studies of faecal transplantation have been in fulminant or refractory C. difficile. However, studies of faecal transplantation to date are heterogeneous regarding the patients treated, donors used, optimal screening protocols, methods and frequency of administration and definition of response. Furthermore, reports to date may suffer from reporting bias of positive outcomes and under-reporting of adverse effects.

Faecal transplantation, a therapy used for more than half a century, could hold great promise as a future treatment, where a dysbiosis of the gut microbiota is responsible for disease. This therapy is inexpensive as well as being effective in some cases. Standardised controlled studies are necessary to ascertain the most effective regimen as well as the most acceptable method of treatment. Two randomised controlled studies of faecal transplantation in C. difficile are on-going in North America and Europe and results from these are eagerly awaited as well as a study of faecal transplantation in metabolic syndrome. Studies of faecal transplantation for other gastrointestinal diseases, where a dysbiosis of the gut microbiota is evident, are necessary. Rigorous screening of potential donors is essential as is the use of partners or close relatives as donors to minimise the potential for transmitting disease. Close monitoring and long-term follow-up are necessary. Combining clinical studies with molecular analysis of the microbiota and the effects on the immune response may significantly enhance our understanding of the gut microbiota and its relationship with health and disease.

References

  1. Top of page
  2. Abstract
  3. Background
  4. Patient details
  5. Donor screening
  6. Route of administration
  7. Patient preparation
  8. Preparation of donor stool
  9. Outcomes
  10. Adverse events
  11. Analysis of effects of faecal transplantation on stool composition and faecal microbiota
  12. Conclusions
  13. Acknowledgement
  14. References