Clostridium difficile is aslow-growing, gram positive, anaerobic bacillus. Clostridia species formspores which are resistant to ultra-violet radiation, temperature extremes andcommon hospital disinfectants. These spores are highly persistent andubiquitous in the environment. C. difficile produces two types of toxins; toxinA, an enterotoxin, and toxin B, a cytotoxin. Both toxins bind to specializedreceptors on intestinal mucosal cells and play a role in the organism’spathogenicity. These toxins can act synergistically. In addition, a thirdbinary CDT toxin exists in some strains. The pathenogenicity of this toxin iscurrently less clear.
Since its identification as thecausative organism in pseudomembranous colitis in 1978, infectious outbreakshave sporadically occurred in several hospitals throughout the US. This hasbrought about various institutional responses, including restricting certainantibiotics, changing housekeeping procedures, and establishing activemultidisciplinary antibiotic management programs. These measures in combinationhave met with some level of success.[i]Restricting the use of clindamycin has reduced Clostridium difficile-relateddiarrhea (CDAD) in some hospitals in the short term[ii],but increasingly other broad-spectrum antibiotics such as cephalosporins andquinolones cause the disease, so restricting clindamycin as a sole containmentpolicy is of very limited use.
New insight into thepathenogenesis of C. difficile infections has recently been gained. A highlytoxinogenic strain was identified in six US states, and in Alberta and Quebec,Canada between 2001 and 2004.[iii]The Society for Healthcare Epidemiology of America said at its 15thannual meeting in April 2005 that this strain, designated as ribotype 027,toxinotype III, has caused outbreaks in 7 hospitals in 6 US states. Thetoxinotype III strain produces 16 times more toxin A and 20 times more toxin Bin vitro than the more commonly isolated toxinotype 0 strain. This may be dueto the deletion of the tcdC gene, a negative regulator of toxin production.This gene limits toxin production in the log (exponential) phase of bacterialgrowth, delaying its production until the stationary phase of growth.[iv]This strain of C. difficile is responsible for the recent epidemic at StokeMandeville Hospital in the UK, announced on June 5th, 2005. Thisepidemic affected 300 persons at the small specialty hospital and killed 12patients, precipitating possible government action regarding hospital hygieneand the interaction between infected patients, staff, and visitors. Subsequentto the reports at Stoke-Mandeville, fourteen other UK hospitals were found tohave patients infected with the 027 strain.[v]Clostridium difficile has recently become a reportable disease in the UK. It’soccurrence is not yet a mandatory reporting issue in the US or Canada.
On July 6th,, 2005the 027 strain was identified at two hospitals in the Netherlands.[vi]Of particular concern was a lack of apparent clinical response to metronidazole.This phenomenon had been empirically observed previously in Canada. Curiously,the strain was susceptible to metronidazole in vitro by E-test.
A disturbance in the usualcontingent of human gut flora brings about C. difficile super infection andresultant diarrhea and colitis. This disruption of microbial flora is mostcommonly associated with orally administered antibiotics and/or broad- spectrumantibiotics. Canadian interventionists have reported a possible relationshipbetween the use of proton pump inhibitors and the disease, but in view of thevery rapid formation of large amounts of toxin, it is unclear how large a rolethe use of these medications play in the pathenogenesis of the toxinotype IIIstrain. The odds ratio of the instance of proton inhibitor use in a smaller(n=94) case-control study portion of their research was nearly equivalent tothat of other risk factors, such as hospital admission in the last 3 months(2.7 and 2.6, respectively). In addition, patients placed on proton pumpinhibitors are often more ill than other patients.[vii]
Since the year 2000, mortalityper case of C. difficile has been rising. Data presented at the 2004 InfectiousDiseases Society of America meeting has indicated that the C. difficile isresistant to fluoroquinolones, especially levofloxacin.[viii][ix] Previously, quinolones were thought to pose a low risk of C. difficileinfection.[x]Other evidence suggests that resistance to metronidazole may be occurring, aswell as documented resistance to macrolides and clindamycin. The risk ofinfection is now highest with cephalosporins and ampicillin/amoxicillin,presumably because they are the most widely used antibiotics. Indeed, almostall classes of antibiotics have been implicated, and even brief exposure to anysingle antibiotic can cause C. difficile colitis.
The C. difficile speciesinvolved in major hospital outbreaks appear to have both a dominant clonalstrain and resistance to a commonly used antibiotic. C. difficile selects forspecific antibiotic resistance and then can proliferate largely because it isresistant to that antibiotic. The antibiotic itself is responsible forobliterating much of the host’s protection by suppressing the bowel flora. C.difficile then produces copious amount of toxins. Hence, reduction of overallantibiotic use is necessary to reduce endemic CDAD rates in hospitals.
A recently reported randomizedcontrol study demonstrates the value of changing overall antibiotic usagepatterns. It involved 807 uncomplicated acute bronchitis patients treated inprimary care facilities that were assigned to one of three treatment arms,including: no antibiotics offered (n = 273), delayed antibiotics via a writtenprescription (n = 272), and immediate antibiotics ( n = 262). According toresponses from 78% of the patients, there was little difference in resolutionof symptoms. Prescribing practices that involved the delayed use or the non-useof antibiotics were effective in 70% to 90% of patients, resulted in adequatesymptom control, and were acceptable to patients.[xi]This study is important because evidence suggests that community-acquired C.difficile may be a problem. Since January 2005, this hospital has seen manycases of community-acquired C. difficile-associated diarrhea. Nearly all the patientswere previously treated with amoxicillin, Augmentin or a cephalosporin. Thistrend indicates that outpatient prescribing practices will play a role in thefuture of the disease.
Clostridium difficile colitisis no longer confined to elderly and very infirm nursing home or hospitalizedpatients. As with MRSA and VRE, utilizing strict enteric contact isolation isan essential means of controlling CDAD spread. Mandating hand-washing withchlorhexidine (not alcohol) before and after patient contact for all staff andvisitors, full personnel protective gowning and gloving, exclusive use ofdiagnostic equipment such as stethoscopes and blood-pressure cuffs on one andonly one patient, and cleaning rooms and equipment with a chlorine-baseddisinfectant must be initiated. Other approaches to combating the diseaseinclude improved surveillance and disease data collection, rapid same-day toxintesting and reporting, and developing new molecular typing systems using wholegenome sequences of C. difficile. Further understanding of the mode of actionof the pathogen, development and application of new treatment strategies,stressing the need for using non-antibiotic treatment modalities such as theuse of antitoxin, probiotic treatment and fecal donation, will also benecessary.
C. difficile is present in thegut of 50% of all neonates and in 7% of adults. It has only recently beenconsidered a pathogen instead of a commensurate organism. Its emergence isrelated to the widespread use of broad-spectrum antibiotics. Policies must beput in place that include the strict practice of enteric contact isolation andan active antibiotic management program, in order to reduce CDAD incidence andprotect patients and staff.
UPDATE: October 2006
A study from 6 North Carolina hospitals has indicated thatcommunity-acquired C. difficile-associated diarrhea (CA-CDAD) is indeed anincreasing problem. Moreover, the spread of the disease may not involve theprior use of antibiotics. C. difficile has been known to spread from patient topatient in hospitals. Now there is some basis to believe that it can spreadamong healthy individuals in the community as well. Of a total of 1137 cases of CDAD reviewed at 6 NorthCarolina hospitals between Jan and Dec 2005, nearly one in 5 (209, 18 percent)were acquired in the community, with 50 percent of those cases not originatingfrom prescription of an antimicrobial, stressed Dr. McDonald.” Cit. ProMedarchive number 20061018. Costridiumdifficle, community acquired – USA (NC).
[i]JohnsonStuart, Samore Matthew, et al. Epidemics of diarrhea caused by a clindamycinresitant strain of Clostridium difficile in four hospitals. NEJM Vol. 341, No.22: 1645-1651. Nov. 25th, 1999.
[ii]EggertsonLaura, Sibbald Barbara. Hospitals battling outbreaks of C. difficile. CMAJ171(1). July 6th, 2004.
[iii]ProMEDDigest. V2005, #155.
[iv]ProMEDDigest. V2005, #235.
[v]ProMEDDigest. V2005, #275.
[vi]ProMEDDigest. V2005, #286.
[vii]DialSandra, Alrasadi Khalid, et al. Risk of Clostridium difficile among hospitalinpatients prescribed proton pump inhibitors: cohort and case-control studies.Can. Med. Assoc. J. 2004 171: 33 – 38.
[viii]StephensonMichelle, correspondent. New epidemic strain of Clostridium difficilehas emerged. Infectious Disease News, February 2005.
[ix]MutoCarlene, Pokrywka Marian, Shutt Kathleen, et al. A large outbreak ofClostridium difficile-associated disease with an unexpected proportion ofdeaths and colectomies at a teaching hospital following increasedfluoroquinolone use. Infection Control and Hospital Epidemiology. Vol 26, No 3,273 – 279. March 2005.
[x]GorbachSherwood. Antibiotics and Clostridium difficile. NEJM Vol 341, No 22. Nov.25th,1999.
[xi]LittlePaul, Rumsby K, et al. Information Leaflet and Antibiotic PrescribingStrategies for Acute Lower Respiratory Tract Infection. JAMA June 22/29, 2005.Vol.293, No. 24, 3029-3035.