Synonyms: Clostridium difficile-associated diarrhoea/disease (CDAD), CD-positive diarrhoea, antibiotic-associated colitis

Pseudomembranous colitis (PMC) is an acute, exudative colitis caused by Clostridium difficile, a Gram-positive toxin-releasing bacillus.
It commonly occurs as a side-effect of antibiotic therapy, due to clearing of normal gut commensals allowing superinfection. It can occur with cytotoxic drugs too (for example, methotrexate, cyclophosphamide and cisplatin).

It has emerged, particularly in recent years, as a major and very expensive healthcare problem. Spores formed by the organism are implicated in spread of infection and have implications for hygiene and prevention of infection.
C. difficile is only responsible for 10-25% of the total burden of antibiotic-associated diarrhoea, but causes virtually all cases of PMC.¹

Antibiotics and pseudomembranous colitis

The antibiotics most often associated with it have changed since C. difficile was recognised as the main cause in 1978:²

• In the first 5 years (1978 through 1983), the most common cause was clindamycin, the standard diagnostic test was the cytotoxin assay, and standard management was to withdraw the implicated antibiotic and treat with oral vancomycin. Most patients responded well, but 25% relapsed when vancomycin was stopped.²
• Between 1983 and 2003 the most commonly implicated antibiotics were the cephalosporins (reflecting use). The enzyme immunoassay became easy to use with improved speed of results (compared with the standard cytotoxin assay). Metronidazole replaced vancomycin as standard treatment, and the principles of containment in hospitals became infection control and antibiotic control.²
• In the last 4 years resistant strains have emerged creating more difficulties with treatment and relapse as well as higher mortality.^2,3 Fluoroquinolones are now more often implicated along with cephalosporins.² Such changes are seen in the USA, Canada, and Europe. This change is now attributed to a new strain of C. difficile designated BI, NAP1, or ribotype 027 (which are synonymous terms). Infection is now often refractory to treatment.

Pathophysiology

• C. difficile is specifically an anaerobic Gram-positive rod. It has two high molecular weight toxins which are unique among bacterial enterotoxins in causing disruption to the barrier function of the colonic mucosa:
  • In about a third of those colonised, C. difficile secretes a toxin. There are two types of toxin, A and B. They are cytotoxic to cells of the intestinal tract, B being about 1,000 times more potent than A.
  • Toxin A has an identified bowel wall receptor and acts as an enterotoxin by loosening junctions between intestinal cells, allowing toxin B to enter epithelial cells. Toxin A activates the inflammatory cascade and, ultimately, these processes lead to cell disruption, fluid secretion, mucosal injury, oedema and inflammation.
  • Toxin B has no identified receptor and also disrupts cells and activates the inflammatory cascade.
• Transmission is via an indirect faeco-oral route, due to the presence of spores on surfaces. In other words, C. difficile can be transmitted via personal contact or environmentally.⁴ The spores can survive for months and patients can become carriers. The risk of colonisation increases with length of hospital stay.
• There are over 150 polymerase chain reaction (PCR) ribotypes and 24 toxinotypes of C. difficile. These have a pathogenicity locus (PaLoc) with genes encoding enterotoxin A (tcdA) and cytotoxin B (tcdB).⁴
• Some strains of C. difficile secrete a binary toxin whose role in human disease is unclear.⁵
  Disease due to C. difficile is therefore due to a progression from an uncolonised state, to being colonised, to toxin production. The likelihood of this occurring depends on the particular strain involved:
  • One particular strain, toxigenic S-type 5236, was reported as causing about 70% of UK cases.⁵
  • However, the emergence of newer, resistant strains associated with higher mortality is widely reported.^2,4
• Infectivity depends not just on the characteristics of the specific strain, but on factors that impact on cross infection (hygiene, bed spacing, and shared toilet facilities).⁵
• The presence of particular host factors predisposing to the progression of infection to pseudomembranous colitis (PMC) is undoubtedly important. This can be appreciated from the risk factors below.

Epidemiology

Prevalence

• Asymptomatic colonisation in the community is between 3 and 5%, and in hospitals about 20% (highest in the elderly and/or immunocompromised).⁵
• Asymptomatic colonisation rates are higher (up to 70%) amongst infants.
• Rates of C. difficile-associated colitis (CDAC) have been increasing steadily in Europe and the USA. The increase is greatest in the elderly.

Incidence

• There were approximately 45,000 cases of C. difficile-associated diarrhoea (CDAD) reported by 166 NHS acute Trusts in England in 2004.⁶
• The collection and reporting of data on CDAD has been included in the mandatory healthcare-associated infection surveillance system for acute Trusts in England since January 2004.
• The vast majority of symptomatic CDAD is hospital-related but cases do occur in the community. One Swedish study found that the incidence in hospitals was 1,300-fold higher than in the community (33,700 compared with 25 primary episodes per 100,000 persons per year).⁷
• CDAD amongst infants is uncommon despite their much higher colonisation rate, probably reflecting enhanced immunity compared with the older population.

Risk factors⁵

• Prolonged courses of antibiotics
• Multiple antibiotic usage
• Increasing age⁸
• Patients in long-term care facilities³
• Severe comorbidity
• Nonsurgical invasive gastrointestinal procedures
• Presence of a nasogastric tube
• Gastric acid-suppressing medication⁹
• Inpatient residence on ITU
• Duration of hospital stay
• Immunocompromised patients¹⁰

Presentation

Colonisation with C. difficile can be associated with a range of possible clinical states:

• The asymptomatic carrier state
• Mild self-limited diarrhoea
• Pseudomembranous colitis (PMC)
• Fulminant colitis

Generally there is a history of antibiotic exposure together with risk factors for colonisation:

• Typically symptoms come on between 5 and 10 days after antibiotic therapy. Occasionally patients will not have had antibiotic exposure.
• Most patients become unwell during their course of antibiotics, but 25–40% may not do so for as many as 10 weeks afterwards.¹¹
• Most affected individuals experience watery diarrhoea (varies from self-limiting to severe and debilitating) ±blood-stained stools, abdominal cramps, fever (especially so in severe cases), rigors ±septicaemia.
• Severe abdominal pain is uncommon but may mimic an acute abdomen.
• Frank rectal bleeding suggests other causes (for example, inflammatory bowel disease).

Differential diagnosis

• Crohn's disease
• Ulcerative colitis
• Diverticular disease
• Other infections:
  • Gastroenteritis
  • Campylobacter
  • Salmonella
  • Shigella
  • Cholera
  • Amoebiasis
• Acute abdomen due to surgical pathology
• Ischaemic colitis

Investigations¹¹

• FBC (WCC elevated in 80%, often very high)
• Urea and electrolytes including creatinine
• Hypoalbuminaemia may be present (due to a protein-losing enteropathy).
• Diagnosis in C. difficile-associated diarrhoea (CDAD) and pseudomembranous colitis (PMC) focuses on detection either of C. difficile or of its toxins in stool samples. The particular method used will depend on the laboratory. Generally repeat testing on 3 stool samples is recommended. Samples can usually be frozen or refrigerated if more than a 4-hour delay in processing is expected. Liaison with the laboratory may be helpful to avoid delay. Methods of testing include:
  • The stool cytotoxin test, which has high sensitivity (94–100%) and specificity (99%) and has been the standard test (it relies on detection of cytotoxic effect on cultured fibroblasts, negated by a specific antibody).⁴
  • Enzyme-linked immunoassay techniques. Toxin can be demonstrated in the stool but these have varying sensitivity and specificity (69–87%).⁴
  • Culture of C. difficile directly from the stool is the most sensitive diagnostic test but does not differentiate between toxin producing and non-toxin producing bacteria.
  • PCR technology provides a new and expensive test, which is sensitive and specific. High cost generally precludes use.
• Sigmoidoscopy (or colonoscopy):
  • Can show the characteristic pseudomembrane plaque appearance in about half of affected patients (see images).¹²
  • May require biopsy to confirm diagnosis.
  • Is not used routinely but is usually performed if rapid diagnosis is needed or in a patient who has ileus (often as part of work-up for other colonic disease).
• Imaging studies:
  • Useful in severe disease but not likely to be helpful in early or mild colitis.
  • Can detect complications (perforation, toxic dilatation).
  • Barium enemas can be harmful and should be avoided.
  • Plain X-rays and CT scanning may be helpful.

Management

• Only treat patients with a positive stool culture but no toxins if you strongly suspect that they have C. difficile-associated diarrhoea (CDAD) and if they are systemically unwell. This is because with no toxins detectable these patients may be asymptomatic carriers of C. difficile with diarrhoea from another cause.⁵
• Correct fluid losses or electrolyte imbalance with oral or IV electrolyte solutions.
• Avoid antiperistaltic agents such as loperamide or opiates (codeine) because of the risk of retention of toxins in the lumen.¹³
• Ceasing the causative antibiotic (if possible) allows resolution in ~3 days in 22%. Consider changing to an antibiotic less likely to cause pseudomembranous colitis (PMC) – aminoglycosides, macrolides, vancomycin or tetracyclines.
• If diarrhoea is severe or persists despite stopping the initial antibiotic, give oral metronidazole 400 mg tds or vancomycin orally 125 mg qds, (preferred drug if pregnant) both usually for 10 days (liaise with a microbiologist). Teicoplanin is another option.
• Metronidazole and vancomycin have been the mainstays of therapy, with some recent data supporting the expanding role of vancomycin in the treatment of severe CDAD. Adjunctive therapy with probiotics, intravenous immunoglobulin, or rifampin has been used in refractory or recurrent CDAD. Adherence to the recommended infection control measures and the judicious use of antibiotics should also be part of the global management of CDAD in long-term care facilities.³
• Some studies favour metronidazole.¹⁴ A recent systematic review calls into question the evidence base for current management strategies. It suggests that mild cases may best be left untreated and that vancomycin is superior to metronidazole but associated with significant side-effects. Teicoplanin is superior at symptomatic and bacteriological cure and may therefore be preferred in terms of preventing spread.¹⁵ Concurrent treatment with vancomycin is sometimes recommended if the causative antibiotic cannot be stopped, but recovery may be slower.

Prognosis

• In healthy individuals a good response to treatment is usually expected, but the illness can cause severe debility and prolonged hospital stays.
• Mortality can be as high as 25% in susceptible critically-ill or frail, older patients.^5,16
• Prevention is vastly preferable to cure. Early identification of C. difficile infection and prompt initiation of therapy with the most appropriate agent are critical to minimise morbidity and mortality in this era of increasingly severe C. difficile-associated diarrhoea (CDAD).³

Complications

Include dehydration and shock; rarely, toxic megacolon, perforation and septicaemia. The mortality rate from toxic megacolon is as high as 30%. Recurrence may be a feature in as many as a third of cases.⁵

Prevention

• Some have suggested the use of probiotics to prevent (and/or treat) antibiotic-induced diarrhoea. More research is needed to identify which specific probiotics may be beneficial (and at what dose) before they can be recommended. The evidence is, at best, equivocal.^5,17
• Overall preventative measures, such as strict handwashing and patient isolation policies for patients with diarrhoea, seem to be effective. There is less evidence of benefit for environmental cleansing measures.
• Handwashing should be done correctly to be effective. Alcohol gels do not kill spores and are not recommended.¹⁸
• Antibiotic policies have an important role.⁵
• Vaccines and local colonic immunological therapies are potential areas of useful preventative intervention.⁵
• The effectiveness of any given set of strategies is probably more related to the rigorousness with which they are carried out.

Document references

1. Barbut F, Meynard JL; Managing antibiotic associated diarrhoea. BMJ. 2002 Jun 8;324(7350):1345-6.
2. Bartlett JG; Narrative review: the new epidemic of Clostridium difficile-associated enteric disease. Ann Intern Med. 2006 Nov 21;145(10):758-64. [abstract]
3. Makris AT, Gelone S; Clostridium difficile in the long-term care setting. J Am Med Dir Assoc. 2007 Jun;8(5):290-9. [abstract]
4. Kuijper EJ, Coignard B, Tull P; Emergence of Clostridium difficile-associated disease in North America and Europe. Clin Microbiol Infect. 2006 Oct;12 Suppl 6:2-18. [abstract]
5. Starr J; Clostridium difficile associated diarrhoea: diagnosis and treatment.; BMJ. 2005 Sep 3;331(7515):498-501.
6. Surveillance of Clostridium difficile associated disease (CDAD); Dept of Health, August 2005
7. Noren T, Akerlund T, Back E, et al; Molecular epidemiology of hospital-associated and community-acquired Clostridium difficile infection in a Swedish county. J Clin Microbiol. 2004 Aug;42(8):3635-43. [abstract]
8. Bartlett JG; Clostridium difficile-associated Enteric Disease. Curr Infect Dis Rep. 2002 Dec;4(6):477-483. [abstract]
9. Bandolier; Proton pump inhibitors and Clostridium difficile
10. Morris AM, Jobe BA, Stoney M, et al; Clostridium difficile colitis: an increasingly aggressive iatrogenic disease? Arch Surg. 2002 Oct;137(10):1096-100. [abstract]
11. Curry J; Pseudomembranous Colitis, eMedicine, Aug 2009
12. Images; Gastrolab Image Gallery; Pseudomemranous Colitis Images
13. Bartlett JG; Clinical practice. Antibiotic-associated diarrhea. N Engl J Med. 2002 Jan 31;346(5):334-9.
14. Moyenuddin M, Williamson JC, Ohl CA; Clostridium difficile-associated diarrhea: current strategies for diagnosis and therapy. Curr Gastroenterol Rep. 2002 Aug;4(4):279-86. [abstract]
15. Bricker E, Garg R, Nelson R, et al; Antibiotic treatment for Clostridium difficile-associated diarrhea in adults. Cochrane Database Syst Rev. 2005 Jan 25;(1):CD004610. [abstract]
16. Pepin J, Valiquette L, Alary ME, et al; Clostridium difficile-associated diarrhea in a region of Quebec from 1991 to 2003: a changing pattern of disease severity. CMAJ. 2004 Aug 31;171(5):466-72. [abstract]
17. D'Souza AL, Rajkumar C, Cooke J, et al; Probiotics in prevention of antibiotic associated diarrhoea: meta-analysis. BMJ. 2002 Jun 8;324(7350):1361. [abstract]
18. A simple guide to Clostridium difficile, Dept of Health, July 2007

Internet and further reading

• Bandolier; Now wash your hands; Evidence base for hand washing

Acknowledgements