Skip Navigation

Bad Bug—The New Epidemic of Clostridium difficile

By John G. Bartlett, M.D., Johns Hopkins University School of Medicine, January 6, 2006


C. difficile was described as the agent of antibiotic-associated pseudomembranous colitis in 1978 and most of the relevant information regarding pathophysiology, putative toxins, clinical features, diagnostic testing, and treatment were described by 1980. This iatrogenic disease has been recognized and managed with substantial success for most of the past 25 years. However, it appears that during the past 2 to 3 years a new epidemic strain of C. difficile has been responsible for more common and serious disease. The renewed interest in C. difficile was apparent at the 45th ICAAC meeting in Washington, D.C. in December 2005, which included 3 symposia and 34 presentations on this topic. A brief review follows.

Historical Background

C. difficile is the major identifiable cause of antibiotic-associated diarrhea. Laboratories that process thousands of specimens consistently report that 15% to 25% of stool specimens submitted for diagnostic testing are positive. The cause of antibiotic-associated diarrhea in the remaining 75% to 85% of specimens is usually unknown.

The disease occurs almost exclusively in the presence of antibiotic exposure, and has been commonly associated with clindamycin and cephalosporins for the past 20 years. However, C. difficile can occur in the presence of just about any antibiotic with antibacterial spectrum. Other risk factors are advanced age and hospitalization. C. difficile-associated enteric disease is limited to the colon and, like most enteric pathogens, it causes a spectrum of disease from asymptomatic carriage, or “nuisance diarrhea,” to severe and sometimes life-threatening pseudomembranous colitis.

The best diagnostic test is the cytotoxin assay that was originally described in 1978, but it requires two days and is awkward for many laboratories to perform. The most commonly used diagnostic test in the U.S. is EIA, which has a sensitivity of only about 70%.

Standard treatment is with metronidazole or oral vancomycin. Metronidazole is preferred according to CDC, SHEA, and the IDSA because it is much less expensive and avoids the over utilization of vancomycin. Most patients respond to either drug, but both drugs are associated with a relapse rate of 15% to 25%.

The “New Epidemic”

  • Epidemiology: A review of ICD-9 codes indicates that the frequency of this diagnosis doubled from 1998 to 2003 in the 172-hospital VA system and in non-federal hospitals in the U.S. In Quebec, Canada, the increase has been substantially higher and, among persons over 65 years, the increase was approximately 10-fold.

  • Clinical features: The increase in C. difficile infection rates has been associated with an even greater increase in severe disease, especially in elderly patients. Severe disease has been defined as toxic megacolon, shock, requirement for colectomy, and death. In Sherbrooke, Quebec, the attributable mortality was most recently reported at 17% (Pepin J, et al. Mortality attributable to nosocomial Clostridium difficile-associated disease during an epidemic caused by a hypervirulent strain in Quebec. CMAJ. 2005;173:1037-42).

  • Epidemic strain: Studies of multiple strains from outbreaks in 6 states of the U.S. and Quebec show a new epidemic strain which was only occasionally encountered in prior studies. This is designated according to the method of study as NAP-1, REA-B1, Toxinotype-III and ribotype-027 (McDonald LC, et al. An epidemic, toxin gene-variant strain of Clostridium difficile. N Engl J Med. 2005;353:2433-41). Studies of this strain show several potential virulence factors: 1) Toxin A and B which are produced by most strains of C. difficile, but are produced in vitro in quantities 16- to 23-fold higher with this strain; 2) An 18 basepair deletion (TDC) which down regulates toxin production; and 3) A binary toxin similar to the iota toxin of C. perfringens type E, which is of uncertain significance (Warny M, et al. Toxin production by an emerging strain of Clostridium difficile associated with outbreaks of severe disease in North America and Europe. Lancet. 2005;366:1079-84). These strains are also universally resistant to fluoroquinolones, a resistance pattern that was infrequent in previous years. Extensive use of these agents may contribute substantially to this surge in cases.

  • Infection control: This is largely a nosocomial infection. Colonic colonization with C. difficile increases from about 3% in healthy adults to 20% to 40% in hospitalized patients. This emphasizes the need for aggressive infection control which includes: early detection of cases, barrier precautions, hand washing with soap and water (as alcohol-based sanitizers do not eradicate C. difficile), and cleaning of the environment with sporicidal agents active against C. difficile. With epidemics, it may be necessary to control implicated antibiotics, usually clindamycin, cephalosporins and/or fluoroquinolones.

  • Treatment: With regard to the question of treatment with metronidazole vs. oral vancomycin, the drugs appear equal in clinical trials, but most specialists prefer vancomycin for patients with serious disease and for those who have a delayed response to metronidazole.


This epidemic provides another example of an old pathogen emerging with greater virulence. Although perhaps driven in part by the increased use of fluoroquinolones, it is unrealistic to think that the restriction of this extremely useful class of antibiotics will successfully control this new form of disease. Early recognition of cases is important not only for optimal management of patient care, but also for the prompt implementation of infection control measures. Thus, the burden of disease control falls once again in the hands of infection control practitioners.