Vancomycin for the Treatment of Clostridium difficile Infection: For Whom Is This Expensive Bullet Really Magic?

IDSA LECTURE Vancomycin for the Treatment of Clostridium difficile Infection: For Whom Is This Expensive Bullet Really Magic? Jacques Pepin Department of Microbiology and Infectious Diseases, University of Sherbrooke, Sherbrooke, Quebec, Canada (See the IDSA lecture by Bartlett on pages 1489 92) The epidemiology, clinical severity, and case-fatality ratio of Clostridium difficile infection (CDI) changed dramatically with the emergence of a toxin hyperproducing strain (BI/NAP1/027) in North America and Europe in 2000. For the treatment of CDI, metronidazole and vancomycin remain the 2 most commonly used drugs. The 3 randomized controlled trials published thus far, as well as the upcoming tolevamer trial, use intermediate outcomes, rather than the outcomes that now preoccupy clinicians: the frequency of complications or recurrence. The major advantage of metronidazole is its low price. The major advantage of orally administered vancomycin is its more favorable pharmacokinetics. Facilitating vancomycin-resistant enterococci colonization and/or infection is a potential drawback of both drugs. Pending the development of a prospectively validated scoring system, members of the Infectious Diseases Society of America/Society for Healthcare Epidemiology of America expert committee will define severe CDI as present in any patient with a leukocyte count 15,000 cells/mm 3 or a creatinine level increased by 50% from baseline. For patients with mild-to-moderate CDI (defined by a leukocyte count!15000 cells/mm 3 and a creatinine level!1.5 times the baseline value), there is no evidence that treatment with vancomycin is superior to treatment with metronidazole (even for intermediate outcomes), and metronidazole therapy should be preferred. For patients with severe CDI who are not infected with BI/NAP1/027, there is reasonable evidence that the better pharmacokinetics of vancomycin translate into a lower probability of complications. For those patients who are infected with BI/NAP1/027, the superiority of vancomycin therapy remains to be proven. In practice, because it is not yet possible to rapidly type the strains, all patients with severe CDI should be treated with vancomycin. Future trials should use complicated CDI and recurrences as their primary outcomes. HISTORICAL PERSPECTIVE In March 1978, John Bartlett and colleagues published the first article that established the etiological role of Clostridium difficile toxins in causing pseudomembran- Received 22 November 2007; accepted 15 January 2008; electronically published 4 April 2008. This is a modified version of a paper given at the 45th Annual Meeting of the Infectious Diseases Society of America, San Diego, California, October 2007. Reprints or correspondence: Dr. Jacques Pepin, CHUS, 3001, 12eme Ave. Nord, Sherbrooke, Quebec, Canada, J1 H 5N4 (jacques.pepin@usherbrooke.ca). Clinical Infectious Diseases 2008; 46:1493 8 2008 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2008/4610-0002$15.00 DOI: 10.1086/587656 ous colitis [1]. Four months later, they reported on the use of oral vancomycin in 9 patients, all of whom had a good clinical response, and documented high fecal levels of vancomycin [2]. Similar results were reported shortly thereafter by a British group [3]. During the same year, the first reports of treatment of C. difficile infection (CDI) with metronidazole appeared [4, 5]. At the end of 1983, the results of the first randomized trial comparing metronidazole (42 patients) and vancomycin (52 patients) were published [6]. The groups did not differ with respect to mean time to resolution of diarrhea, frequency of treatment failure (which occurred in 2 patients, both of whom had received metronidazole), or relapses. This was before power, sample Vancomyin for Treatment of C. difficile Infection CID 2008:46 (15 May) 1493

size estimates, and b-type error became a collective preoccupation, and it was concluded that both treatments were equivalent [6]. The field of CDI treatment then entered a hibernation phase that lasted 120 years. Additional case series were published, treatments other than metronidazole or vancomycin came and went (e.g., bacitracin, fusidic acid, teicoplanin, and colestipol), and a second small, 4-arm, randomized controlled trial was conducted (with 31 patients per arm) which, predictably, did not show differences between metronidazole and vancomycin therapy [7]. Use of vancomycin for the treatment of CDI decreased when, in 1995, the Hospital Infection Control Practices Advisory Committee recommended avoiding it as much as possible, in the hope of reducing the selection pressure for the emergence of vancomycin-resistant enterococci [8]. The Society for Healthcare Epidemiology of America position paper stated that both drugs were effective and that metronidazole is less expensive and may be preferable to avoid vancomycin resistance in other nosocomial bacterial species [9, p. 473]. Two years later, the American College of Gastroenterology also recommended metronidazole for most patients, with the exception of small subgroups, such as those critically ill with CDI [10]. Throughout that period, CDI was considered to be a generally trivial condition, and there was little interest in evaluating novel therapeutic approaches, despite the shortcomings associated with available drugs, especially the high frequency of recurrences. Around 2000, somewhere in the United States, 1 specific strain of C. difficile to be designated BI/NAP1/027 became resistant to all fluoroquinolones and, probably because of a 1 base pair deletion in its tcdc gene, became a hyperproducer of toxins A and B [11]. Compared with historical strains, these changes enhanced its transmission within and between hospitals, and BI/NAP1/027 spread very effectively. In 2002, it was introduced into Montreal, Canada, in the province of Quebec, and spread from there to dozens of hospitals, presumably via transfer of patients or via health care workers who provided care in 11 institution [12, 13]. Incidence of nosocomial CDI increased dramatically: at the worst of the epidemic, 4% of all patients who were admitted into several hospitals of Quebec acquired nosocomial CDI. Between 14% and 17% of patients with nosocomial CDI died as a direct or indirect consequence of infection [13, 14]. In Quebec, mortality in which CDI was the main cause of death increased from 85 cases in 2000 to 691 cases in 2004 [15]. BI/NAP1/027 spread throughout the United States, where the incidence of CDI more than doubled and CDI mortality quadrupled [16 18], but it also spread throughout Western Europe [19]. At last, this crisis gave a new impetus to the development of novel drugs for the treatment of CDI and prompted a reassessment of the comparative efficacy of metronidazole and vancomycin. IN VITRO SUSCEPTIBILITY Metronidazole resistance in C. difficile is rare, and the MICs of nearly all strains are 2.0 mg/l [20, 21]. In a report from Spain, the MIC 90 of 415 isolates was 4.0 mg/l, and 6% of isolates had MICs 32 mg/l [22], but this was not confirmed by further work. The MIC 90 of vancomycin against C. difficile is 1.0 2.0 mg/l, and the highest MIC ever reported is 16 mg/l [20, 21]. There is no evidence that the epidemic BI/NAP1/027 strain or other recent North American isolates are more resistant to vancomycin or metronidazole than are nonepidemic strains or historical isolates [13, 23 25]. PHARMACOKINETICS Orally administered metronidazole is absorbed rapidly and almost completely; only 6% 15% of its metabolites are excreted in stools. Fecal concentrations of metronidazole reflect its secretion in the colon, and levels decrease rapidly after treatment of CDI is initiated, from 9.3 mg/g in watery stools to 1.2 mg/g in formed stools [26]. When metronidazole is administered intravenously, fecal levels are at least as high as and perhaps even higher than when the drug is given orally [26, 27]. Metronidazole is undetectable in stools of asymptomatic C. difficile carriers [28], and administering courses longer than 14 days seems to be irrational, particularly if diarrhea has resolved. In contrast, vancomycin is poorly absorbed, so that fecal concentrations following oral administration (at a dosage of 125 mg administered 4 times daily) are generally 11000 mg/g [28]. Vancomycin fecal levels are maintained throughout treatment. Thus, the ratio of fecal levels divided by the MIC is 2 orders of magnitudes higher with vancomycin treatment than with metronidazole treatment. It normally requires!6 h for nutriments to travel from the mouth to the ileo-caecal valve. In patients with some degree of ileus, this interval will be longer, but whether this becomes clinically relevant with regard to the delivery of vancomycin to the colon is unknown. In comparison, orally administered metronidazole peaks 1 2 h later and will quickly reach the colonic wall. However, we do not know how long it takes for metronidazole to be secreted into the colonic lumen. ADVERSE EFFECTS Adverse effects are not a major issue when selecting between metronidazole or vancomycin for patients with CDI. The most common adverse effect of metronidazole therapy is gastrointestinal intolerance, but in patients with CDI, it is difficult to determine whether nausea is caused by the drug or by the underlying illness. If given for a long duration, metronidazole can cause polyneuropathy, which is another reason to avoid long courses of therapy. When administered orally, little van- 1494 CID 2008:46 (15 May) Pepin

comycin is absorbed, and thus, it causes no systemic toxicity. Allergic reactions to either drug are uncommon. VANCOMYCIN-RESISTANT ENTEROCOCCI CDI and colonization and/or infection with vancomycin-resistant enterococci (VRE; or methicillin-resistant Staphylococcus aureus) share many risk factors: increased age, extended hospital stay, presence of numerous comorbidities, tube feeding, and the administration of antibiotics, such as cephalosporins and fluoroquinolones [29]. It is thus not surprising that these pathogens often coexist in some patients, as has been demonstrated repeatedly [30, 31]. Whether intravenously or orally administered vancomyin promotes the emergence of vancomycin-resistant enterococci was reviewed 10 years ago by Dale Gerding [32]. An association between intravenously administered vancomycin and VRE colonization or infection was documented in 9 of the 10 studies then available [32]. Orally administered vancomycin was associated with VRE in 3 of the 4 small studies that stratified risk by route of administration. Metronidazole was also associated with VRE [32]. Although this could have been a consequence of the concomitant administration of other antibiotics that themselves select for VRE (e.g., third-generation cephalosporins and fluoroquinolones), treatment with anti-anaerobic agents promotes high-density colonization with VRE [33]. A meta-analysis found substantial heterogeneity between studies: those that used patients with vancomycin-susceptible enterococci as control subjects yielded much stronger associations between administration of vancomycin and VRE than did those that used patients without VRE (many without any enterococcal colonization or infection at all) as control subjects, and the funnel plot suggested publication bias [34]. More recently, in a large case-control study that adjusted for multiple confounding factors, including duration of hospital stay and administration of other antibiotics, VRE was associated with prior use of metronidazole, third-generation cephalosporins, and long courses of fluoroquinolones; surprisingly, VRE was not associated with the use of either intravenous or oral vancomycin [31]. Control subjects consisted of a mix of patients documented not to have VRE and patients not tested at all; therefore, there must have been at least some misclassification bias. The only study that specifically looked at the frequency of VRE colonization in patients who were given oral vancomycin for the treatment of CDI was performed in a hospital with a low overall incidence of VRE colonization [35]. At the population level, the use of oral vancomycin has increased dramatically in Quebec and in parts of the United States since 2002. In Quebec, a provincial surveillance system for VRE documented that the number of cases increased from 106 in 2002 to 275 in 2003 and 554 in 2004, in parallel with the increasing use of oral vancomycin for CDI [36]. However, the availability of novel antibiotics active against VRE (e.g., daptomycin and tigecycline) attenuated the potentially devastating clinical impacts of this pathogen. COSTS The major advantage of metronidazole over vancomycin is its much lower price. Cost of a 10-day course of metronidazole is $20, whereas the cost of a 10-day course of oral vancomycin (Vancocin; ) increased from $300 to $600 when the drug became more widely used in the United States. In hospitals, this issue can be avoided by administering the generic intravenous formulation of vancomycin, reducing the price of a 10-day course to $45. There is no reason to believe that such an approach would alter fecal levels of the drugs. However, community pharmacies do not stock intravenous vancomycin and would not be interested in manipulating the product so that it could be taken orally. It is possible that generic formulations of oral vancomycin will reach the US market within the next few years. This having been said, if vancomycin therapy were truly superior to metronidazole therapy in avoiding complications, it could be inferred, even without a formal cost-efficacy analysis, that the additional costs incurred by vancomycin therapy would be compensated by savings on the management of CDI complications. Obviously, the cost per patient with CDI requiring admission to the intensive care unit must be in the 5-digit range. MARKERS OF SEVERITY What is complicated CDI? Most clinicians would agree that this must include cases in any patient who either (1) dies as a direct or indirect consequence of CDI, be it during the initial episode or during a recurrence; (2) requires admission to the intensive care unit for the care of CDI; (3) develops hypotension or shock (hypovolemic or septic); or (4) develops megacolon, perforation, or some other indication for an emergency colectomy. These are the main outcomes that clinicians want to avoid by using a putatively better treatment, and this definition will be used herein. Other outcomes of relevance include the prolongation of hospital stay, costs of an episode of nosocomial CDI, and the frequency of recurrences. Unfortunately, many therapeutic trials have used softer, intermediate outcomes, such as presence of diarrhea or a positive toxin assay result on a specified therapy day. A number of markers have been identified as predictive of complicated CDI in retrospective studies [12, 37]. First, the incidence of CDI increases with age, as does the frequency of complications and the case-fatality ratio. Second, a high leukocyte count, which reflects the severity of colonic inflammation, is strongly associated with complications. Third, the Vancomyin for Treatment of C. difficile Infection CID 2008:46 (15 May) 1495

severity of diarrhea, measured either clinically (e.g., number of stools per 24 h) or biochemically (an increasing creatinine level) correlates with complicated CDI. It is thus surprising that no simple scoring system has yet been validated prospectively. In a recently published trial, researchers used a score combining age 160 years, temperature 138.3 C, albumin level!2.5 mg/dl, and leukocyte count 15,000 cells/mm 3, giving 1 point for each characteristic and giving 2 points if pseudomembranous colitis was seen on endoscopic examination or if the patient required admission to the intensive care unit [38]. Severe CDI was defined as a score 2. However, the authors provided no information on how these variables individually or collectively correlated with complicated CDI. Pending the development of a prospectively validated scoring system, members of the Infectious Diseases Society of America/Society for Healthcare Epidemiology of America expert committee are going to define severe CDI as occurring in any patient with a leukocyte count 15000 cells/ mm 3 or a creatinine level increased by 50% from baseline. This is meant to have high sensitivity for identifying patients at risk for complications, while accepting a lower specificity. COMPARATIVE EFFICACY OF METRONIDAZOLE AND VANCOMYCIN For patients with mild-to-moderate CDI, there is no evidence from randomized controlled trials that treatment with vancomycin is superior to treatment with metronidazole. In a recently published, small, randomized controlled trial that enrolled patients from 1994 through 2002 (i.e., before the emergence of BI/NAP1/027) and used a soft outcome (no diarrhea and negative toxin assay results on day 6 and day 10 of therapy), no difference was seen between vancomycin therapy and metronidazole therapy for patients with mild disease (i.e., those with a score of 0 1) [38]. Similarly, in the tolevamer versus vancomycin versus metronidazole trial, whose preliminary results were presented at the 2007 Interscience Conference on Antimicrobial Agents and Chemotherapy, no difference was seen between treatment with metronidazole and treatment with vancomycin for patients with mild-to-moderate CDI, with the outcome being defined as resolution of diarrhea and abdominal discomfort by day 10 of therapy [39]. Both studies showed a better outcome with vancomycin therapy in patients with severe CDI, defined as a score 2 [38] or as having either a leukocyte count 20,000 cells/mm 3, 10 bowel movements per day, or severe abdominal pain [39]. However, these studies did not prove that treatment with vancomyin reduced the frequency of complicated CDI or of recurrences. The first study was conducted before the emergence of BI/NAP1/027 [38]. In the tolevamer trial, isolates were typed, and it will be interesting to stratify analysis according to genotypes. Even if they are intrinsically inferior to randomized controlled trials, because of the eternal problem of residual confounding, observational studies can provide useful information pending the results of large randomized controlled trials that will use a hard outcome. We recently extended our retrospective review of all 1616 cases of CDI treated in a single academic center in Canada from 1991 through 2006 [37]. Adjusting for the same confounding variables (age, immune status, hospital acquisition, peak leukocyte count, and peak creatinine level), administration of vancomycin as the initial treatment was associated with a lower probability of developing complicated CDI, compared with initial treatment with metronidazole, for the 1991 2002 period, before BI/NAP1/027 was introduced in our hospital (adjusted OR [AOR], 0.21; 95% CI, 0.05 0.99). However, for the 2003 2006 period, when BI/NAP1/027 was the predominate strain, use of vancomycin rather than metronidazole as the initial treatment no longer impacted on the risk of complicated CDI (AOR, 0.90; 95% CI, 0.53 1.55), after adjustment for the very same confounding factors. Interaction (i.e., effect modification) was present (i.e., the AOR for 1991 2002 significantly differed from the AOR for 2003 2006). Restricting analysis to 2003 2006, adjusting for the same confounders and using the same outcome, the AOR of vancomycin versus metronidazole was the same whether the patient fulfilled the Infectious Diseases Society of America/Society for Healthcare Epidemiology of America criteria of severity (AOR, 0.90; 95% CI, 0.49 1.65) or not (AOR, 0.92; 95% CI, 0.28 3.02). The emergence of BI/NAP1/027 is the most obvious difference between the 2 periods, potentially accounting for this epidemiological interaction. What could have happened? It may be that, in patients infected with BI/NAP1/027, production of toxins A and B was so rapid and overwhelming [40] and the effect so immediate in patients with no preexisting immunity that, by the time orally administered vancomycin reached the colonic lumen in patients with some degree of ileus, it was too late to make any difference. Recent reports from Canada and the United States suggested that the risk of postmetronidazole recurrences was increasing [12, 23]. However, extending the observation through 2006, an interesting phenomenon was documented [37]. After metronidazole treatment, the frequency of recurrences was 20% in 1991 2002, increased to 45% in 2003 2004, and then decreased to 35% in 2005 2006. After vancomycin treatment, the frequency of recurrences was 20% in 1991 2002, 39% in 2003 2004, and 23% in 2005 2006. The fact that the same temporal trend was seen with 2 completely unrelated drugs, in the absence of any change in their in vitro activity, suggests that a substantial fraction of the frequent recurrences in 2003 2004 corresponded to reinfections during a period when in-hospital exposure to C. difficile spores was very high. 1496 CID 2008:46 (15 May) Pepin

CONCLUSIONS The major advantage of metronidazole is that it is cheap. The major advantage of orally administered vancomycin lies in its more favorable pharmacokinetics. Although it is perhaps counter intuitive, facilitating VRE colonization and/or infection is a potential drawback of both drugs and cannot be used as an argument in favor of one drug over the other. For patients with mild-to-moderate CDI, metronidazole therapy should be preferred, because it is 20 times cheaper than vancomycin therapy, and no study has documented any difference, even in intermediate outcomes. Pending the availability of a prospectively validated scoring system, mild-to-moderate CDI can be defined according to the Infectious Diseases Society of America/Society for Healthcare Epidemiology of America guidelines (i.e., a leukocyte count!15,000 cells/mm 3 and a creatinine level!1.5 times the baseline value). For patients with severe CDI who are not infected with BI/ NAP1/027, there is reasonable evidence that the better pharmacokinetics of vancomycin translate into a lower probability of complications. For those patients who are infected with BI/ NAP1/027, the superiority of vancomycin therapy remains to be proven. In practice, because it is not yet possible to rapidly type the strains, all patients with severe CDI should be treated with vancomycin. There is a need for a simple and large study in which intravenous metronidazole and oral vancomycin would be compared head-to-head in patients with severe CDI, using complicated CDI as the primary outcome and with stratification according to strain type. Until then, we may use the golden bullet, but we do not know for sure whether it is magic, and that should not deter us from trying to find better bullets. Acknowledgments Potential conflicts of interest. J.P. has been on the speakers bureau for Wyeth and Merck and has served on advisory boards for Wyeth, Bayer, Acambis, and Viropharma. References 1. Bartlett JG, Chang TW, Gurwith M, Gorbach SL, Onderdonk AB. Antibiotic-associated pseudomembranous colitis due to toxin-producing clostridia. N Engl J Med 1978; 298:531 4. 2. Tedesco F, Markham R, Gurwith M, Christie D, Bartlett JG. Oral vancomycin for antibiotic-associated pseudomembranous colitis. Lancet 1978; 2:226 8. 3. Keighley MR, Burdon DW, Arabi Y, et al. Randomised controlled trial of vancomycin for pseudomembranous colitis and postoperative diarrhoea. Br Med J 1978; 2:1667 9. 4. Dinh HT, Kernbaum S, Frottier J. Treatment of antibiotic-induced colitis by metronidazole. Lancet 1978; 1:338 9. 5. Matuchansky C, Aries J, Maire P. Metronidazole for antibiotic associated pseudomembranous colitis. Lancet 1978; 2:580 1. 6. Teasley DG, Gerding DN, Olson MM, et al. Prospective randomised trial of metronidazole versus vancomycin for Clostridium difficile associated diarrhoea and colitis. Lancet 1983; 2:1043 6. 7. Wenisch C, Parschalk B, Hasenhundl M, Hirschl AM, Graninger W. Comparison of vancomycin, teicoplanin, metronidazole, and fusidic acid for the treatment of Clostridium difficile associated diarrhea. Clin Infect Dis 1996; 22:813 8. 8. Centers for Disease Control and Prevention. Recommendations for preventing the spread of vancomycin resistance. Recommendations of the Hospital Infection Control Practices Committee. MMWR Morb Mortal Wkly Rep 1995; 44:(RR-12)1 13. 9. Gerding DN, Johnson S, Peterson LR, Mulligan ME, Silva J Jr. Society for Healthcare Epidemiology of America position paper on Clostridium difficile associated diarrhea and colitis. Infect Control Hosp Epidemiol 1995; 16:459 77. 10. Fekety R. Guidelines for the diagnosis and management of Clostridium difficile associated diarrhea and colitis. Am J Gastroenterol 1997; 92: 739 50. 11. MacCannell DR, Louie TJ, Gregson DB, et al. Molecular analysis of Clostridium difficile PCR ribotype 027 isolates from Eastern and Western Canada. J Clin Microbiol 2006; 44:2147 52. 12. Pepin J, Valiquette L, Alary ME, et al. Clostridium difficile associated diarrhea in a region of Québec from 1991 to 2003: a changing pattern of disease severity. CMAJ 2004; 171:466 72. 13. Loo VG, Poirier L, Miller MA, et al. A predominantly clonal multiinstitutional outbreak of Clostridium difficile associated diarrhea with high morbidity and mortality. N Engl J Med 2005; 353:2442 9. 14. Pépin J, Valiquette L, Cossette B. Mortality attributable to nosocomial Clostridium difficile associated disease during an epidemic cause by a hypervirulent strain in Quebec. CMAJ 2005; 173:1037 41. 15. Institut de Statistique du Québec. La situation démographique au Québec. Bilan 2006. Available at: http://www.stat.gouv.qc.ca/publications/ demograp/pdf2006/bilan2006c4.pdf. Accessed 20 March 2008. 16. McDonald LC, Owings M, Jernigan DB. Clostridium difficile infection in patients discharged from US short-stay hospitals, 1996 2003. Emerg Infect Dis 2006; 12:409 15. 17. Redelings MD, Sorvillo F, Mascola L. Increase in Clostridium difficile related mortality, United States, 1999 2004. Emerg Infect Dis 2007; 13:1417 9. 18. Ricciardi R, Rothenberger DA, Madoff RD, Baster NN. Increasing prevalence and severity of Clostridium difficile colitis in hospitalized patients in the United States. Arch Surg 2007; 142:624 31. 19. Kuijper EJ, Coignard B, Brazier J, et al. Update of Clostridium difficile associated disease due to PCR ribotype 027 in Europe. Eurosurveillance Web site. Available at: http://www.eurosurveillance.org/em/ v12n06/1206-221.asp. Accessed 20 march 2008. 20. Wong SS, Woo PC, Luk WK, Yuen KY. Susceptibility testing of Clostridium difficile against metronidazole and vancomycin by disk diffusion and Etest. Diagn Microbiol Infect Dis 1999; 34:1 6. 21. Aspevall O, Lundberg A, Burman LG, Akerlund T, Svenungsson B. Antimicrobial susceptibility pattern of Clostridium difficile and its relation to PCR ribotypes in a Swedish university hospital. Antimicrob Agents Chemother 2006; 50:1890 2. 22. Pelaez T, Alcala L, Alonso R, et al. In vitro activity of ramoplanin against Clostridium difficile, including strains with reduced susceptibility to vancomycin or with resistance to metronidazole. Antimicrob Agents Chemother 2005; 49:1157 9. 23. Musher DM, Aslam S, Logan N, et al. Relatively poor outcome after treatment of Clostridium difficile colitis with metronidazole. Clin Infect Dis 2005; 40:1586 90. 24. Hubert B, Loo VG, Bourgault AM, et al. A portrait of the geographic dissemination of the Clostridium difficile North American pulsed-field type 1 strain and the epidemiology of C. difficile associated disease in Québec. Clin Infect Dis 2007; 44:238 44. 25. Bourgault AM, Lamothe F, Loo VG, Poirier L, and the CDAD-CSI study group. In vitro susceptibility of Clostridium difficile clinical isolates from a multi-institutional outbreak in Quebec. Antimicrob Agents Chemother 2006; 50:3473 5. 26. Bolton RP, Culshaw MA. Faecal metronidazole concentrations during oral and intravenous therapy for antibiotic associated colitis due to Clostridium difficile. Gut 1986; 27:1169 72. Vancomyin for Treatment of C. difficile Infection CID 2008:46 (15 May) 1497

27. Dion YM, Richards GK, Prentis JJ, Hinchey EJ. The influence of oral versus parenteral preoperative metronidazole on sepsis following colon surgery. Ann Surg 1980; 192:221 6. 28. Johnson S, Homann SR, Bettin KM, et al. Treatment of asymptomatic Clostridium difficile carriers (fecal excretors) with vancomycin or metronidazole: a randomized, placebo-controlled trial. Ann Intern Med 1992; 117:297 302. 29. Safdar N, Maki DG. The commonality of risk factors for nosocomial colonization and infection with antimicrobial-resistant Staphylococcus aureus, Enterococcus, gram-negative bacilli, Clostridium difficile and Candida. Ann Intern Med 2002; 136:834 44. 30. Donskey C, Ray AJ, Hoyen CK, et al. Colonization and infection with multiple nosocomial pathogens among patients colonized with vancomycin-resistant Enterococcus. Infect Control Hosp Epidemiol 2003; 24:242 5. 31. Carmeli Y, Eliopoulos GM, Samore MH. Antecedent treatment with different antibiotic agents as a risk factor for vancomycin-resistant Enterococcus. Emerg Infect Dis 2002; 8:802 7. 32. Gerding D. Is there a relationship between vancomycin-resistant enterococcal infection and Clostridium difficile infection. Clin Infect Dis 1997; 25(Suppl 2):S206 10. 33. Donskey CJ, Chowdhry TK, Hecker MT, et al. Effect of antibiotic therapy on the density of vancomycin-resistant enterococci in the stool of colonized patients. N Engl J Med 2000; 343:1925 32. 34. Carmeli Y, Samore MH, Huskins C. The association between antecedent vancomycin treatment and hospital-acquired vancomycin-resistant enterococci. Arch Intern Med 1999; 159:2461 8. 35. Salgado C, Giannetta ET, Farr BM. Failure to develop vancomycinresistant Enterococcus with oral vancomycin treatment of Clostridium difficile. Infect Control Hosp Epidemiol 2004; 25:413 7. 36. Jetté L, Ismail J, Bekal S, Laurence RA. Surveillance passive des entérocoques résistants à la vancomycine, Rapport 2004. Institut National de Santé Publique du Québec. Available at: http://www.inspq.qc.ca/ pdf/publications/429-enterocoquesvancomycine2004.pdf. Accessed 20 March 2008. 37. Pepin J, Valiquette L, Gagnon S, Routhier S, Brazeau I. Outcomes of Clostridium difficile associated disease treated with metronidazole or vancomycin before and after the emergence of NAP 1/027. Am J Gastroenterol 2007; 102:2781 8. 38. Zar FA, Bakkanagari SR, Moorthi KM, Davis MB. A comparison of vancomycin and metronidazole for the treatment of Clostridium difficile associated diarrhea, stratified by disease severity. Clin Infect Dis 2007; 45:302 7. 39. Louie T, Gerson M, Grimard D, et al. Results of a phase III trial comparing tolevamer, vancomycin and metronidazole in patients with Clostridium difficile associated diarrhea [abstact K-425a]. In: Program and abstracts of the 47th Interscience Conference on Antimicrobial Agents and Chemotherapy (Chicago, IL). Washington, DC: American Society for Microbiology, 2007. 40. Warny M, Pepin J, Fang A, 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. 1498 CID 2008:46 (15 May) Pepin