- Email: [email protected]
Clostridium difficile infection in patients with ulcerative colitis: Investigations of risk factors and efficacy of antibiotics for steroid refractory patients
Clinics and Research in Hepatology and Gastroenterology (2011) 35, 315—320
ORIGINAL ARTICLE
Clostridium difficile infection in patients with ulcerative colitis: Investigations of risk factors and efficacy of antibiotics for steroid refractory patients Takashi Kaneko a, Reikei Matsuda a,∗, Masataka Taguri b, Masahiko Inamori c, Asami Ogura d, Eiji Miyajima d, Katsuaki Tanaka e, Shin Maeda c, Hideaki Kimura a, Reiko Kunisaki a a
Inflammatory Bowel Disease Center, Yokohama City University Medical Center, Yokohama, Japan Department of Biostatistics and Epidemiology, Graduate School of Medicine, Yokohama City University Medical Center, Yokohama, Japan c Department of Gastroenterology, Yokohama City University, School of Medicine, Yokohama, Japan d Department of Laboratory Medicine and Clinical Investigation, Yokohama City University Medical Center, Yokohama, Japan e Gastroenterological Center, Yokohama City University Medical Center, Yokohama, Japan b
Available online 23 March 2011
Summary Background and objective: The incidence of Clostridium difficile infection (CDI) has increased throughout the world and patients with ulcerative colitis (UC) are at a high risk for CDI. Potentially, CDI can exacerbate UC. Therefore, knowledge on the prevalence of CDI should contribute to better management of UC patients. Methods: The presence of toxin A antigen was defined as CDI, and the outcome of the test in patients with active UC during 2006—2009 was reviewed for identifying patients with CDI. Demographic data (disease profile, clinical response to medications and the need for colectomy) in UC patients with CDI were compared with the data from CDI free UC patients. Results: Fifty-five of 137 patients (40.1%) were CDI positive. Univariate and multivariate analyses revealed that CDI was not associated with any demographic factor. Intensive antibiotic therapy spared five of 17 (29.4%) steroid refractory patients with CDI from steroids. CDI was not a predictor of colectomy although this could be an outcome of efficient eradication strategy. Conclusion: CDI was not associated with any demographic factor or colectomy rate. However, CDI eradication therapy allowed some refractory patients to withdraw from steroids. Patients with active UC benefit from regular CDI test and eradication treatment for CDI. © 2011 Elsevier Masson SAS. All rights reserved.
∗ Corresponding author. Inflammatory Bowel Disease Center,Yokohama City University Medical Center, 4-57 Urafune, Minami-ku, Yokohama, 232-0024, Japan. Tel.:+81 45 261 5656; fax: +81 45 253 9955. E-mail address: [email protected] (R. Matsuda).
2210-7401/$ – see front matter © 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.clinre.2011.02.004
316
Introduction Clostridium difficile infection (CDI) or Clostridium difficile associated diarrhea (CDAD) is alarmingly increasing and has become a frequently observed nosocomial infection in the US and many other countries [1—4]. Paralleling the rising incidence in general population, during 1998 to 2004 [5], the incidence of CDI in patients with inflammatory bowel diseases (IBD) has doubled in patients with Crohn’s disease and tripled in patients with ulcerative colitis. This finding has been confirmed in several other independent studies, which found IBD, in particular, UC to be a major independent risk factor for acquiring CDI [6—11]. At the beginning of the infection, CDI mimics a UC flare and responds to corticosteroid therapy. However, gradually, colitis exacerbates, and in some cases, it elicits a lifethreatening complication like perforation, toxic megacolon, sepsis, and even death [12]. Recently, the impact of CDI on unfavorable IBD treatment outcome has been highlighted. In several retrospective studies [9,13—15], IBD patients with concomitant CDI had a four times greater mortality rate, up to six times higher colectomy rate, and greater medication need for IBD than patients without CDI. With this background in mind, we undertook this investigation to assess the prevalence of CDI in Japan and identify factor(s), which predispose UC patients to CDI. In this endeavour, we compiled IBD and other relevant data on UC patients who received toxin A test (marker of CDI) and compared the data from patients with CDI with the data from patients without CDI. Factors included were clinical background, disease profile, treatment for UC, refractoriness of UC to corticosteroids [16], chronic active or frequently relapsed colitis. Additionally, in steroid refractory UC patients with CDI, we assessed the efficacy of antibiotics based on the cessation of steroid requirement, and the contribution of CDI to colectomy rate. To our knowledge, this is the first survey of CDI in UC patients in Japan.
Patients and methods Patients From January 2006 to April 2009, Clostridium difficile (C. difficile) toxin A test on the gut lavage which was aspirated during colonoscopy was carried out in patients with UC during active phase of IBD. A total of 149 patients were investigated and of these 137 were found to have active UC and were investigated further for CDI. Patients had been followed up in our hospital or were sent to us by other hospitals. The diagnosis of UC was based on colonoscopy and pathological findings prior to toxin A test. UC severity was assessed by Truelove and Witts index; mild, moderate, and severe [17]. Further, patients with active UC were treated according to the protocol with the primary intention of inducing clinical remission.
Ethical considerations The study protocol was reviewed and approved by the Ethics Committee of Yokohama City University Medical Center.
T. Kaneko et al. Informed consent was obtained from all patients before registering to this study, after being informed of the purpose and the nature of the investigations involved.
Methods CDI was diagnosed by the presence of toxin A in the gut lavage, which was determined by using an enzyme-linked immunosorbent assay (ELISA) technique (C. difficile Toxin A test, Oxoid Limited, Basingstoke, UK). To identify risk factors for acquiring CDI in UC patients, demographic information including age, gender, frequency of hospitalization, outpatient or inpatient, duration of UC, disease extent, disease severity, current use of immunosuppressive agents like azathioprine (AZA), 6-mercaptopurine (6-MP), or cyclosporin (CyA), prolonged use of steroid enema, systemic steroid, proton pump inhibitor (PPI), refractoriness of UC to corticosteroids, frequently flaring or chronic active UC, and the lengths of hospital stay until the toxin A test were factored. The compiled data on UC patients with CDI and UC patients without CDI were analysed as described in the statistics section. Further, we determined colectomy rates in UC patients who had been with CDI and UC patients who had been CDI free over an equal time period.
Antibiotic treatment Patients who tested positive for C. difficile in active colitis setting were treated with metronidazole (MNZ) 500mg1000 mg/day orally for two weeks or oral vancomycin (VCM) 2 g/day for two weeks. In patients with steroid refractory UC, we accessed the withdrawal rate from steroid after antibiotic therapy as a measure of antibiotic efficacy.
Statistics Continuous variables were analyzed by the Mann-Whitney Utest and categorical variables were analyzed by the Fisher’s exact test for detecting differences between groups. Univariate and multivariate logistic regression analyses were carried out by using the Stat View-J 5.0 (SAS Institute, Cary NC). P < 0.05 was considered statistically significant.
Results The prevalence of Clostridium difficile infection in the study population A total of 137 patients with an UC flare from a population of 149 UC patients were followed in this investigation. Of these 137 patients, 55 (40.1%) were diagnosed with CDI. As shown in Table 1, at the investigation time, 41 patients (29.9%) were hospitalized, and 96 (70.1%) visited an outpatient clinic. Further, 87 patients (63.5%) had pancolitis and 32 (23.3%) had severe colitis. Forty-four patients (32.1%) had steroid refractory UC and 79 (57.7%) were patients with frequently flared or chronic active UC. To exclude the possibility of nosocomial infection spread in our hospital; if CDI prevalence had increased 72 h after admission was determined. To investigate this, we divided the hospitalized
C. difficile infection in active UC Table 1
317
Disease characteristic in 137 study cohort.
Age, median; range, (years) Female gender, n (%) CDI, n (%) UC duration, median; range
36; 14—82 62 (45.2) 55 (40.1) 4; 0—34
Disease extent Proctosigmoidal, n (%) Left sided, n (%) Pan colitis, n (%)
11 (8.1) 38 (27.7) 88 (64.2)
Severity, n (%) Mild, n (%) Moderate, n (%) Severe, n (%)
24 (17.5) 81 (59.3) 32 (23.3)
Outpatient/hospitalized, n (%) Outpatient clinic Hospitalized
96 (70.1) 41 (29.9)
Episode of hospitalization median; range Prolonged use of steroid enema, n (%) Systemic steroid use, n (%) Immunomodulator use, n (%) Combination treatment of immunomodulator and steroid, n (%) Current PPI use, n (%) Refractory UC, n (%) Steroid refractory Frequently flared or chronic active
1; 0—10 18 (13.1) 79 (57.8) 21 (15.8) 16 (11.7)
Figure 1 This figure shows that the prevalence of CDI depends on time from admission. The prevalence of CDI among admitted patients, no significant differences between less than 72 h from admission group and longer than 72 h group or transferred from other hospitals were seen, suggesting no significant nosocomial infection had occurred in our hospital (P = 0.10).
51 (37.2) 44 (32.1) 79 (57.7)
patients into three groups depending on the time point of toxin A test (shorter than 72 h after admission, longer than 72 h after admission, and transferred from other hospitals) and compared the prevalence of CDI among these three subgroups (Fig. 1). Although an increased trend was seen after hospitalization, the difference was very weak and not statistically significant (P = 0.10). As antibiotics were given to most of the hospitalized patients with severe disease, and the number of patients with longer stay at hospital was limited, we thought that the possibility of nosocomial infection was too low to affect the result of this study.
Univariate and multivariate analyses The relationship between patients’ background and CDI are shown in Table 2. Age, gender, duration of IBD, extent of colitis (proctosigmoidal colitis, left-sided colitis, and pancolitis), clinical severity (mild, moderate, and severe), inpatient or outpatient, frequency of hospitalization, use of steroid enema more than 14 times within a month prior to the study, use of systemic steroid, AZA, 6-MP, CyA, combination therapy involving an immunomodulator and steroid, current PPI, refractoriness of UC to steroids, frequently flared or chronic active UC were compared between patients with CDI and patients without CDI. None of the included factors appeared to be significantly associated with CDI. Then, we carried out multivariate logistic regression analysis to identify independent influence factors of CDI. We included factors (steroid refractory condition, prolonged use of steroid enema, combination treatment of
immunomodulator and steroid, and frequently flared or chronic active UC) which would be potentially associated with CDI and UC refractoriness into the model (Table 3). No significant factor was detected except a weak association between steroid enema therapy and CDI (P = 0.09, OR = 2.35; 95% CI, 0.85—6.46). Further, 51 patients were receiving PPI and three of these were found to be CDI compared with 12 CDI in 70 PPI free UC patients (P = 0.16).
Assessment based on Immunosupressive therapy Table 1 shows that an immunosuppressive agent was given to an 84 of the 137 patients (61.3%), 79 were on systemic steroid, 18 were on AZA or 6-MP, and only three on CyA (16 patients receiving more than one immunosupressant). Analysis for each agent showed no association with CDI (data not shown). Subsequently, we analyzed the relationship between CDI and systemic steroid. Daily dose of steroid at the time of investigation or the cumulative amount of steroid used as an immunosuppressant also did not show association with CDI (P = 0.13 P = 0.64, respectively, data not shown).
Efficacy of antibiotics for withdrawal from steroid therapy MNZ 500—1000 mg/day (the dose was decided by consideration of historical side effects associated with this agent or based on body weight) or VCM 2 g/day orally for 2 weeks was given to 17 steroid refractory patients with CDI for the eradication of C. difficile. Two patients changed from MNZ to VCM because of side effects and inadequate efficacy of MNZ. Further, prior to this treatment, we increased the dose of mesalazine up to a maximum of 4 g/day or
318 Table 2
T. Kaneko et al. Demographic features of patients with UC showing disease characteristics depends on CDI.
Demography
CD (+) n = 55
CD (−) n = 82
P value
Age, median; range (years) Female gender, n (%)
36; 16—76 24 (43.6)
38; 14—82 38 (46.3)
0.49 0.86
Duration of UC UC duration, median; range (years)
4; 1—21
5; 0—34
0.13
Disease extent (n) Proctosigmoidal Left sided Pancolitis
8 16 31
3 22 57
0.55
Severity (n) Mild Moderate Severe
11 35 9
13 46 23
0.20
Outpatient/hospitalized, (n) Outpatient clinic Hospitalized
15 40
26 56
0.70
Episode of hospitalization median; range Prolonged use of steroid enema, n (%) Systemic steroid use, n (%) Immunomodulator use, n (%) Combination treatment of immunomodulator and steroid, n (%) Current PPI use, n (%)
1; 0—6 11 (20.0) 28 (50.9) 8 (14.5) 6 (10.9) 3 (5.5)
1; 0—10 8 (9.8) 51 (62.2) 13 (15.9) 10 (12.2) 12 (14.6)
0.18 0.13 0.22 0.99 0.99 0.16
Refractory UC, n (%) Steroid refractory Frequently flared or chronic active
19 (34.5) 34 (61.8)
25 (30.5) 45 (54.9)
0.71 0.48
Table 3
Multivariate logistic regression analyses for assessing the influences of independent factors on CDI.
Demography
Odds ratio
95% confidence interval
P-value
Steroid refractory Prolonged use of steroid enema Combination treatment of immunomodulator and steroid Frequently flared or chronic active UC
1.33 2.35 0.86 1.23
0.63—2.82 0.85—6.46 0.29—2.60 0.60—2.52
0.45 0.09 0.80 0.57
sulphasalazine up to a maximum of 4.5 g/day. Improved symptoms and subsequent withdrawal from steroid were observed in five of 17 patients (29.4%) without adding other medication.
Colectomy rates in patients with and without Clostridium difficile infection We were also interested to see the effect of CDI on colectomy rates in UC patients with active disease. CDI was included as an independent variable for colectomy, and univariate logistic regression was performed. At the toxin A test, 41 patients stayed in the hospital and 15 of these 41 (36.6%) were positive for CDI. Five of these 15 patients (33.3%) had colectomy versus six (23.1%) of 26 non-CDI patients (P = 0.94, OR = 1.03; 95% CI, 0.41—2.63). Therefore, CDI did not appear to impact colectomy rate in our patient population.
Discussion In patients with IBD, microbial infection by enteric organisms, an opportunistic agent like CDI or cytomegalovirus potentially can exacerbate the underlying IBD, cause refractoriness to the conventional steroid therapy, and affect the treatment outcome [6,7,18—22]. C. difficile is a grampositive spore-forming anaerobic bacterium associated with the development of diarrhoea, severe pseudomembranous colitis, sepsis, and sometime death in humans. In recent years, the reported incidence of CDI has increased, and the odds ratio of acquiring CDI in UC patients were found to be almost four times as many as in non-IBD patients [5,9]. Patients with IBD, in particular UC patients as a population susceptible to CDI has been suspected recently [6—11,23]. For example, in India, 32% of flared UC patients were reported to be with CDI [6]. The survey of children with UC in Italy reported 21% with CDI [23]. A national survey in the USA investigated the prevalence of CDI in UC
C. difficile infection in active UC patients and reported an incidence of 37.3/1000 [10]. Taken together, up to 24% of relapsed UC patients are reported to be with CDI [24]. With this background in mind, we became interested to identify possible risk factors for acquiring CDI in patients with active UC, assess the efficacy of antibiotics in steroid refractory UC patients with CDI, and explore any likely association between CDI and colectomy after intensive antibiotic treatment. Hitherto suspected risk factors for CDI in IBD patients have been reported to be IBD itself when it involves colonic lesions, antibiotic exposure, healthcare occupation, increased comorbidity, admission to teaching hospitals, nonsummer months, maintenance immunosuppressive or initiation of corticosteroid therapy, PPIs (for recurrent CDI), disease extent, and bacterial strain type, albeit some of these are still controversial [8—10,25—27]. To unequivocally determine predisposing factor(s) to CDI in UC patients, we applied univariate together with multivariate analyses to patients’ major demographic data, but none was associated with the likelihood of acquiring CDI. Then, we thought that refractory UC condition might bear an association with CDI, but the outcome was negative. Other conditions we considered were combined treatment with an immunomodulator plus systemic steroid, prolonged use of steroid enema, and frequently flared UC and chronic active UC. The outcome did not change. Broad-spectrum antibiotics, older age (> 65 yr), prolonged hospital stay, and presence of systemic comorbidities are recognized as the classical risks of acquiring CDI [28—31]. However, older age was not associated with CDI in our UC patients, while other three mentioned risk factors were not assessed in this study. Additionally, our toxin A test was performed in conjunction with colonoscopy, and was done immediately after UC exacerbation. Therefore, the effect of prolonged hospital stay was insignificant in this study. It might be necessary to mention here that Infliximab, which potentially can promote opportunistic infection, was not an approved medication for UC in Japan at the time of this investigation and therefore, none of the patients in our cohort was exposed to infliximab. Recently, CDI is observed in patients with IBD without recent exposure to antibiotics [10,23,32]. According to a study of general population, medication with PPI was associated with CDI through neutralization of gastric acid [33—36]. In our cohort, 51 patients were treated with PPI at the time of toxin A test, and three of these 51 patients were found to be CDI versus 12 CDI patients in 70 PPI free UC patients, suggesting no association between PPI therapy and CDI. We retrospectively investigated the prognosis in patients we surveyed, and then colectomy rate was recorded. Our logistic regression analyses indicated that CDI was not a predictor of colectomy. This observation is contrary to previous investigations [12,13,15,37] which have reported otherwise. However, considering that we diligently detected CD toxin A and immediately treated the infected patients with antibiotics, this difference could have impacted the outcome (no association between CDI and colectomy rate). This is to say that an increased risk of colectomy following a UC flare, which is complicated by concomitant CDI might have been compromised by an intensive antibiotic treatment [38]. In our patients, 40 of 96 (41.7%) with an active UC flare who visited our outpatient clinic had concomitant CDI,
319 suggesting that CDI in IBD patients was contracted in the community. It may also be true to say that IBD by itself causes the disruption of the normal flora in the colon and this may predispose IBD patients to CDI [24,39]. However, it is unclear how many patients are asymptomatic careers during their remission period and how many develop colitis later on. The rate of asymptomatic career in IBD patients is reported to be higher than general population [40]. Clayton et al. found that an 8.2% of IBD asymptomatic careers without prior exposure to antibiotics, corticosteroids, immunomodulators, or recent hospitalization were CDI positive [3,39]. Therefore, further prospective studies are warranted. In our cohort, CDI was detected in 40.1% of the 137 patients with active UC. This might be higher than the level in a general active UC population in Japan because our data included more than half of chronic refractory patients who had been introduced to us from other hospitals. However, considering that the average annual positive rate of CD toxin A test in our hospital laboratory was about 6%, the prevalence of CDI among our flared UC patients is in fact high. In conclusion, our results strongly indicate that patients with UC are a very high risk population for acquiring CDI. However, our univariate together with multivariate analyses of patients’ major demographic data did not identify any component of UC as a significant factor for predisposing UC patients to CDI. Likewise, CDI did not appear to be a predictor of colectomy. Finally, we believe that patients with active UC benefit from regular check ups for CDI and immediately receive medication to eradicate C. difficile.
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Responsibility disclaimer: None. Funding: None.
References [1] McDonald LC, Owings M, Jernigan DB. Clostridium difficile infection in patients discharged from US short-stay hospital, 1996—2003. Emerg Infect Dis 2006;12:409—15. [2] Pepin J, Valiquette L, Alary ME, Villemure P, Pelletier A, Forget K, et al. Clostridium difficile-associated diarrhea in a region of Quebec from 1991—2003: a changing pattern of disease severity. CMAJ 2004;171:466—72. [3] Paltansing S, van den Berg RJ, Guseinova RA, Visser CE, van der Vorm ER, Kuijper EJ. Characteristics and incidence Clostridium difficile-associated disease in The Netherlands, 2005. Clin Microbiol Infect 2007;13:1058—64. [4] McFarland LV, Mulligan ME, Kwok RY, Stann WE. Nosocomial acquisition of Clostridium difficile infection. N Engl J Med 1989;320:204—10. [5] Rodemann JF, Dubberke ER, Reske KA, Seo da H, Stone CD. Incidence of Clostridium difficile infection in inflammatory bowel disease. Clin Gastroenterol Hepatol 2007;5:339—44. [6] Kochhar R, Ayyagari A, Goenka MK, Dhali GK, Aggarwal R, Mehta SK. Role of infectious agents in exacerbations of ulcerative colitis in India. A study of Clostridium difficile. J Clin Gastroenterol 1993;16:26—30.
320 [7] Meyer AM, Ramzan NN, Loftus Jr EV, Heigh RI, Leighton JA. The diagnostic yield of stool pathogen studies during relapses of inflammatory bowel disease. J Clin Gastroenterol 2004;38:772—5. [8] Mylonaki M, Langmead L, Pantes A, Johnson F, Rampton DS. Enteric infection in relapse of inflammatory bowel disease: importance of microbiological examination of stool. Eur J Gastroenterol Hepatol 2004;16:775—8. [9] Issa M, Vijayapal A, Graham MB, Beaulieu DB, Otterson MF, Lundeen S, et al. Impact of Clostridium difficile on inflammatory bowel disease. Clin Gastroenterol Hepatol 2007;5:345—51. [10] Nguyen GC, Kaplan GG, Harris ML, Brant SR. A national survey of the prevalence and impact of Clostridium difficile infection among hospitalized inflammatory bowel disease patients. Am J Gastroenterol 2008;103:1443—50. [11] Ricciardi R, Ogilvie Jr JW, Roberts PL, Marcello PW, Concannon TW, Baxter NN. Epidemiology of Clostridium difficile colitis in hospitalized patients with inflammatory bowel disease. Dis Colon Rectum 2009;52:40—5. [12] Issa M, Ananthakrishnan AN, Binion DG. Clostridium difficile and inflammatory bowel disease. Inflamm Bowel Dis 2008;14:1432—42. [13] Ananthakrishnan AN, McGinley EL, Binion DG. Excess hospitalization burden associated with Clostridium difficile in patients with inflammatory bowel disease. Gut 2008;57:205—10. [14] Chiplunker A, Ananthakrishnan AN, Beaulieu DB, Naik AS, Zadvormova Y, Skaros S, et al. Long-term impact of Clostridium difficile on inflammatory bowel disease. Gastroenterology 2009;136(Suppl. 1):A-199. [15] Greenstein AJ, Byrn JC, Zhang LP, Swedish KA, Jahn AE, Divino CM. Risk factors for the development of fulminant Clostridium difficile colitis. Surgery 2008;143:623—9. [16] Matsumoto T, Andoh A, Okawa K, Ito H, Torii A, Yoshikawa S, et al. Multivariate analysis for factors predicting rapid response of leukocytapheresis in patients with steroid-resistant ulcerative colitis: a multicenter prospective open-label study. Ther Apher Dial 2008;12:484—90. [17] Truelove SC, Witts LJ. Cortisone in ulcerative colitis; final report on a therapeutic trial. Br Med J 1955;2:1041—8. [18] Irving PM, Gibson PR. Infections and IBD. Nat Clin Pract Gastroenterol Hepatol 2008;5:18—27. [19] Stallmach A, Carstens O. Role of infections in the manifestation or reactivation of inflammatory bowel disease. Inflamm Bowel Dis 2002;8:213—8. [20] Stone CD. Prognosis in Clostridium difficile infection complicating inflammatory bowel disease. Gut 2008;57:150—2. [21] Gryboski JD. Clostridium difficile in inflammatory bowel disease relapse. J Pediatr Gastroenterol Nutr 1991;13:39—41. [22] Viget N, Vernier-Massouille G, Salmon-Ceron D, Yazdanpanah Y, Colombel JF. Opportunistic infections in patients with inflammatory bowel disease: prevention and diagnosis. Gut 2008;57:549—58. [23] Pascarella F, Martinelli M, Miele E, Del Pezzo M, Roscetto E, Staiano A. Impact of Clostridium difficile infection on pediatric inflammatory bowel disease. J Pediatr 2009;154:854—8. [24] Musa S, Thomson S, Cowan M, Rahman T. Clostridium difficile infection and inflammatory bowel disease. Scand J Gastroenterol 2010;45:261—72.
T. Kaneko et al. [25] Powell N, Jung SE, Krishnan B. Clostridium difficile infection and inflammatory bowel disease: a marker for disease extent? Gut 2008;57:1183—4. [26] Ben-Horin S, Margalit M, Bossuyt P, Maul J, Shapira Y, Bojic D, et al. Combination immunomodulator and antibiotic treatment in patients with inflammatory bowel disease and Clostridium difficile infection. Clin Gastroenterol Hepatol 2009;7: 981—7. [27] Schneeweiss S, Korzenik J, Solomon DH, Canning C, Lee J, Bressler B. Infliximab and other immunomodulating drugs in patients with inflammatory bowel disease and the risk of serious bacterial infections. Aliment Pharmacol Ther 2009;30:253—64. [28] Bartlett JG, Clinical practice. Antibiotic-associated diarrhea. N Engl J Med 2002;346:334—9. [29] Barbut F, Petit JC. Epidemiology of Clostridium difficileassociated infection. Clin Microbiol Infect 2001;7: 405—10. [30] Redelings MD, Sorvillo F, Mascola L. Increase in Clostridium difficile-related mortality rates, United States, 1999—2004. Emerg Infect Dis 2007;13:1417—9. [31] Bignardi GE. Risk factors for Clostridium difficile infection. J Hosp Infect 1998;40:1—15. [32] Bossuyt P, Verhaegen J, Van Assche G, Rutgeerts P, Vermeire S. Increasing incidence of Clostridium difficile-associated diarrhea in inflammatory bowel disease. J Crohn’s Colitis 2009;3:4—7. [33] Dial S, Alrasadi K, Manoukian C, Huang A, Menzies D. Risk of Clostridium difficile diarrhea among hospital inpatients prescribed proton pump inhibitors: cohort and case-control studies. CMAJ 2004;171:33—8. [34] Dial S, Delaney JA, Barkun AN, Suissa S. Use of gastric acid-suppressive agents and the risk of communityacquired Clostridium difficile-associated disease. JAMA 2005;294:2989—95. [35] Hauben M, Horn S, Reich L, Younus M. Association between gastric acid suppressants and Clostridium difficile colitis and community-acquired pneumonia: analysis using pharmacovigilance tools. Int J Infect Dis 2007;11:417—22. [36] Aseeri M, Schroeder T, Kramer J, Zackula R. Gastric acid suppression by proton pump inhibitors as a risk factor for Clostridium difficile-associated diarrhea in hospitalized patients. Am J Gastroenterol 2008;103:2308—13. [37] Jodorkovsky D, Young Y, Abreu MT. Clinical outcomes of patients with ulcerative colitis and co-existing Clostridium difficile infection. Dig Dis Sci 2010;55:415—20. [38] Stange EF, Travis SP, Vermeire S, Reinisch W, Geboes K, Barakauskiene A, et al. European evidence-based consensus on the diagnosis and management of ulcerative colitis: definitions and diagnosis. J Crohn’s Colitis 2008;2:1—23. [39] Clayton EM, Rea MC, Shanahan F, Quigley EM, Kiely B, Hill C, et al. The vexed relationship between Clostridium difficile and inflammatory bowel disease: an assessment of carriage in an outpatient setting among patients in remission. Am J Gastroenterol 2009;104:1162—9. [40] Dorman SA, Liggoria E, Winn Jr WC, Beeken WL. Isolation of Clostridium difficile from patients with inactive Crohn’s disease. Gastroenterology 1982;82:1348—51.
ORIGINAL ARTICLE
Clostridium difficile infection in patients with ulcerative colitis: Investigations of risk factors and efficacy of antibiotics for steroid refractory patients Takashi Kaneko a, Reikei Matsuda a,∗, Masataka Taguri b, Masahiko Inamori c, Asami Ogura d, Eiji Miyajima d, Katsuaki Tanaka e, Shin Maeda c, Hideaki Kimura a, Reiko Kunisaki a a
Inflammatory Bowel Disease Center, Yokohama City University Medical Center, Yokohama, Japan Department of Biostatistics and Epidemiology, Graduate School of Medicine, Yokohama City University Medical Center, Yokohama, Japan c Department of Gastroenterology, Yokohama City University, School of Medicine, Yokohama, Japan d Department of Laboratory Medicine and Clinical Investigation, Yokohama City University Medical Center, Yokohama, Japan e Gastroenterological Center, Yokohama City University Medical Center, Yokohama, Japan b
Available online 23 March 2011
Summary Background and objective: The incidence of Clostridium difficile infection (CDI) has increased throughout the world and patients with ulcerative colitis (UC) are at a high risk for CDI. Potentially, CDI can exacerbate UC. Therefore, knowledge on the prevalence of CDI should contribute to better management of UC patients. Methods: The presence of toxin A antigen was defined as CDI, and the outcome of the test in patients with active UC during 2006—2009 was reviewed for identifying patients with CDI. Demographic data (disease profile, clinical response to medications and the need for colectomy) in UC patients with CDI were compared with the data from CDI free UC patients. Results: Fifty-five of 137 patients (40.1%) were CDI positive. Univariate and multivariate analyses revealed that CDI was not associated with any demographic factor. Intensive antibiotic therapy spared five of 17 (29.4%) steroid refractory patients with CDI from steroids. CDI was not a predictor of colectomy although this could be an outcome of efficient eradication strategy. Conclusion: CDI was not associated with any demographic factor or colectomy rate. However, CDI eradication therapy allowed some refractory patients to withdraw from steroids. Patients with active UC benefit from regular CDI test and eradication treatment for CDI. © 2011 Elsevier Masson SAS. All rights reserved.
∗ Corresponding author. Inflammatory Bowel Disease Center,Yokohama City University Medical Center, 4-57 Urafune, Minami-ku, Yokohama, 232-0024, Japan. Tel.:+81 45 261 5656; fax: +81 45 253 9955. E-mail address: [email protected] (R. Matsuda).
2210-7401/$ – see front matter © 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.clinre.2011.02.004
316
Introduction Clostridium difficile infection (CDI) or Clostridium difficile associated diarrhea (CDAD) is alarmingly increasing and has become a frequently observed nosocomial infection in the US and many other countries [1—4]. Paralleling the rising incidence in general population, during 1998 to 2004 [5], the incidence of CDI in patients with inflammatory bowel diseases (IBD) has doubled in patients with Crohn’s disease and tripled in patients with ulcerative colitis. This finding has been confirmed in several other independent studies, which found IBD, in particular, UC to be a major independent risk factor for acquiring CDI [6—11]. At the beginning of the infection, CDI mimics a UC flare and responds to corticosteroid therapy. However, gradually, colitis exacerbates, and in some cases, it elicits a lifethreatening complication like perforation, toxic megacolon, sepsis, and even death [12]. Recently, the impact of CDI on unfavorable IBD treatment outcome has been highlighted. In several retrospective studies [9,13—15], IBD patients with concomitant CDI had a four times greater mortality rate, up to six times higher colectomy rate, and greater medication need for IBD than patients without CDI. With this background in mind, we undertook this investigation to assess the prevalence of CDI in Japan and identify factor(s), which predispose UC patients to CDI. In this endeavour, we compiled IBD and other relevant data on UC patients who received toxin A test (marker of CDI) and compared the data from patients with CDI with the data from patients without CDI. Factors included were clinical background, disease profile, treatment for UC, refractoriness of UC to corticosteroids [16], chronic active or frequently relapsed colitis. Additionally, in steroid refractory UC patients with CDI, we assessed the efficacy of antibiotics based on the cessation of steroid requirement, and the contribution of CDI to colectomy rate. To our knowledge, this is the first survey of CDI in UC patients in Japan.
Patients and methods Patients From January 2006 to April 2009, Clostridium difficile (C. difficile) toxin A test on the gut lavage which was aspirated during colonoscopy was carried out in patients with UC during active phase of IBD. A total of 149 patients were investigated and of these 137 were found to have active UC and were investigated further for CDI. Patients had been followed up in our hospital or were sent to us by other hospitals. The diagnosis of UC was based on colonoscopy and pathological findings prior to toxin A test. UC severity was assessed by Truelove and Witts index; mild, moderate, and severe [17]. Further, patients with active UC were treated according to the protocol with the primary intention of inducing clinical remission.
Ethical considerations The study protocol was reviewed and approved by the Ethics Committee of Yokohama City University Medical Center.
T. Kaneko et al. Informed consent was obtained from all patients before registering to this study, after being informed of the purpose and the nature of the investigations involved.
Methods CDI was diagnosed by the presence of toxin A in the gut lavage, which was determined by using an enzyme-linked immunosorbent assay (ELISA) technique (C. difficile Toxin A test, Oxoid Limited, Basingstoke, UK). To identify risk factors for acquiring CDI in UC patients, demographic information including age, gender, frequency of hospitalization, outpatient or inpatient, duration of UC, disease extent, disease severity, current use of immunosuppressive agents like azathioprine (AZA), 6-mercaptopurine (6-MP), or cyclosporin (CyA), prolonged use of steroid enema, systemic steroid, proton pump inhibitor (PPI), refractoriness of UC to corticosteroids, frequently flaring or chronic active UC, and the lengths of hospital stay until the toxin A test were factored. The compiled data on UC patients with CDI and UC patients without CDI were analysed as described in the statistics section. Further, we determined colectomy rates in UC patients who had been with CDI and UC patients who had been CDI free over an equal time period.
Antibiotic treatment Patients who tested positive for C. difficile in active colitis setting were treated with metronidazole (MNZ) 500mg1000 mg/day orally for two weeks or oral vancomycin (VCM) 2 g/day for two weeks. In patients with steroid refractory UC, we accessed the withdrawal rate from steroid after antibiotic therapy as a measure of antibiotic efficacy.
Statistics Continuous variables were analyzed by the Mann-Whitney Utest and categorical variables were analyzed by the Fisher’s exact test for detecting differences between groups. Univariate and multivariate logistic regression analyses were carried out by using the Stat View-J 5.0 (SAS Institute, Cary NC). P < 0.05 was considered statistically significant.
Results The prevalence of Clostridium difficile infection in the study population A total of 137 patients with an UC flare from a population of 149 UC patients were followed in this investigation. Of these 137 patients, 55 (40.1%) were diagnosed with CDI. As shown in Table 1, at the investigation time, 41 patients (29.9%) were hospitalized, and 96 (70.1%) visited an outpatient clinic. Further, 87 patients (63.5%) had pancolitis and 32 (23.3%) had severe colitis. Forty-four patients (32.1%) had steroid refractory UC and 79 (57.7%) were patients with frequently flared or chronic active UC. To exclude the possibility of nosocomial infection spread in our hospital; if CDI prevalence had increased 72 h after admission was determined. To investigate this, we divided the hospitalized
C. difficile infection in active UC Table 1
317
Disease characteristic in 137 study cohort.
Age, median; range, (years) Female gender, n (%) CDI, n (%) UC duration, median; range
36; 14—82 62 (45.2) 55 (40.1) 4; 0—34
Disease extent Proctosigmoidal, n (%) Left sided, n (%) Pan colitis, n (%)
11 (8.1) 38 (27.7) 88 (64.2)
Severity, n (%) Mild, n (%) Moderate, n (%) Severe, n (%)
24 (17.5) 81 (59.3) 32 (23.3)
Outpatient/hospitalized, n (%) Outpatient clinic Hospitalized
96 (70.1) 41 (29.9)
Episode of hospitalization median; range Prolonged use of steroid enema, n (%) Systemic steroid use, n (%) Immunomodulator use, n (%) Combination treatment of immunomodulator and steroid, n (%) Current PPI use, n (%) Refractory UC, n (%) Steroid refractory Frequently flared or chronic active
1; 0—10 18 (13.1) 79 (57.8) 21 (15.8) 16 (11.7)
Figure 1 This figure shows that the prevalence of CDI depends on time from admission. The prevalence of CDI among admitted patients, no significant differences between less than 72 h from admission group and longer than 72 h group or transferred from other hospitals were seen, suggesting no significant nosocomial infection had occurred in our hospital (P = 0.10).
51 (37.2) 44 (32.1) 79 (57.7)
patients into three groups depending on the time point of toxin A test (shorter than 72 h after admission, longer than 72 h after admission, and transferred from other hospitals) and compared the prevalence of CDI among these three subgroups (Fig. 1). Although an increased trend was seen after hospitalization, the difference was very weak and not statistically significant (P = 0.10). As antibiotics were given to most of the hospitalized patients with severe disease, and the number of patients with longer stay at hospital was limited, we thought that the possibility of nosocomial infection was too low to affect the result of this study.
Univariate and multivariate analyses The relationship between patients’ background and CDI are shown in Table 2. Age, gender, duration of IBD, extent of colitis (proctosigmoidal colitis, left-sided colitis, and pancolitis), clinical severity (mild, moderate, and severe), inpatient or outpatient, frequency of hospitalization, use of steroid enema more than 14 times within a month prior to the study, use of systemic steroid, AZA, 6-MP, CyA, combination therapy involving an immunomodulator and steroid, current PPI, refractoriness of UC to steroids, frequently flared or chronic active UC were compared between patients with CDI and patients without CDI. None of the included factors appeared to be significantly associated with CDI. Then, we carried out multivariate logistic regression analysis to identify independent influence factors of CDI. We included factors (steroid refractory condition, prolonged use of steroid enema, combination treatment of
immunomodulator and steroid, and frequently flared or chronic active UC) which would be potentially associated with CDI and UC refractoriness into the model (Table 3). No significant factor was detected except a weak association between steroid enema therapy and CDI (P = 0.09, OR = 2.35; 95% CI, 0.85—6.46). Further, 51 patients were receiving PPI and three of these were found to be CDI compared with 12 CDI in 70 PPI free UC patients (P = 0.16).
Assessment based on Immunosupressive therapy Table 1 shows that an immunosuppressive agent was given to an 84 of the 137 patients (61.3%), 79 were on systemic steroid, 18 were on AZA or 6-MP, and only three on CyA (16 patients receiving more than one immunosupressant). Analysis for each agent showed no association with CDI (data not shown). Subsequently, we analyzed the relationship between CDI and systemic steroid. Daily dose of steroid at the time of investigation or the cumulative amount of steroid used as an immunosuppressant also did not show association with CDI (P = 0.13 P = 0.64, respectively, data not shown).
Efficacy of antibiotics for withdrawal from steroid therapy MNZ 500—1000 mg/day (the dose was decided by consideration of historical side effects associated with this agent or based on body weight) or VCM 2 g/day orally for 2 weeks was given to 17 steroid refractory patients with CDI for the eradication of C. difficile. Two patients changed from MNZ to VCM because of side effects and inadequate efficacy of MNZ. Further, prior to this treatment, we increased the dose of mesalazine up to a maximum of 4 g/day or
318 Table 2
T. Kaneko et al. Demographic features of patients with UC showing disease characteristics depends on CDI.
Demography
CD (+) n = 55
CD (−) n = 82
P value
Age, median; range (years) Female gender, n (%)
36; 16—76 24 (43.6)
38; 14—82 38 (46.3)
0.49 0.86
Duration of UC UC duration, median; range (years)
4; 1—21
5; 0—34
0.13
Disease extent (n) Proctosigmoidal Left sided Pancolitis
8 16 31
3 22 57
0.55
Severity (n) Mild Moderate Severe
11 35 9
13 46 23
0.20
Outpatient/hospitalized, (n) Outpatient clinic Hospitalized
15 40
26 56
0.70
Episode of hospitalization median; range Prolonged use of steroid enema, n (%) Systemic steroid use, n (%) Immunomodulator use, n (%) Combination treatment of immunomodulator and steroid, n (%) Current PPI use, n (%)
1; 0—6 11 (20.0) 28 (50.9) 8 (14.5) 6 (10.9) 3 (5.5)
1; 0—10 8 (9.8) 51 (62.2) 13 (15.9) 10 (12.2) 12 (14.6)
0.18 0.13 0.22 0.99 0.99 0.16
Refractory UC, n (%) Steroid refractory Frequently flared or chronic active
19 (34.5) 34 (61.8)
25 (30.5) 45 (54.9)
0.71 0.48
Table 3
Multivariate logistic regression analyses for assessing the influences of independent factors on CDI.
Demography
Odds ratio
95% confidence interval
P-value
Steroid refractory Prolonged use of steroid enema Combination treatment of immunomodulator and steroid Frequently flared or chronic active UC
1.33 2.35 0.86 1.23
0.63—2.82 0.85—6.46 0.29—2.60 0.60—2.52
0.45 0.09 0.80 0.57
sulphasalazine up to a maximum of 4.5 g/day. Improved symptoms and subsequent withdrawal from steroid were observed in five of 17 patients (29.4%) without adding other medication.
Colectomy rates in patients with and without Clostridium difficile infection We were also interested to see the effect of CDI on colectomy rates in UC patients with active disease. CDI was included as an independent variable for colectomy, and univariate logistic regression was performed. At the toxin A test, 41 patients stayed in the hospital and 15 of these 41 (36.6%) were positive for CDI. Five of these 15 patients (33.3%) had colectomy versus six (23.1%) of 26 non-CDI patients (P = 0.94, OR = 1.03; 95% CI, 0.41—2.63). Therefore, CDI did not appear to impact colectomy rate in our patient population.
Discussion In patients with IBD, microbial infection by enteric organisms, an opportunistic agent like CDI or cytomegalovirus potentially can exacerbate the underlying IBD, cause refractoriness to the conventional steroid therapy, and affect the treatment outcome [6,7,18—22]. C. difficile is a grampositive spore-forming anaerobic bacterium associated with the development of diarrhoea, severe pseudomembranous colitis, sepsis, and sometime death in humans. In recent years, the reported incidence of CDI has increased, and the odds ratio of acquiring CDI in UC patients were found to be almost four times as many as in non-IBD patients [5,9]. Patients with IBD, in particular UC patients as a population susceptible to CDI has been suspected recently [6—11,23]. For example, in India, 32% of flared UC patients were reported to be with CDI [6]. The survey of children with UC in Italy reported 21% with CDI [23]. A national survey in the USA investigated the prevalence of CDI in UC
C. difficile infection in active UC patients and reported an incidence of 37.3/1000 [10]. Taken together, up to 24% of relapsed UC patients are reported to be with CDI [24]. With this background in mind, we became interested to identify possible risk factors for acquiring CDI in patients with active UC, assess the efficacy of antibiotics in steroid refractory UC patients with CDI, and explore any likely association between CDI and colectomy after intensive antibiotic treatment. Hitherto suspected risk factors for CDI in IBD patients have been reported to be IBD itself when it involves colonic lesions, antibiotic exposure, healthcare occupation, increased comorbidity, admission to teaching hospitals, nonsummer months, maintenance immunosuppressive or initiation of corticosteroid therapy, PPIs (for recurrent CDI), disease extent, and bacterial strain type, albeit some of these are still controversial [8—10,25—27]. To unequivocally determine predisposing factor(s) to CDI in UC patients, we applied univariate together with multivariate analyses to patients’ major demographic data, but none was associated with the likelihood of acquiring CDI. Then, we thought that refractory UC condition might bear an association with CDI, but the outcome was negative. Other conditions we considered were combined treatment with an immunomodulator plus systemic steroid, prolonged use of steroid enema, and frequently flared UC and chronic active UC. The outcome did not change. Broad-spectrum antibiotics, older age (> 65 yr), prolonged hospital stay, and presence of systemic comorbidities are recognized as the classical risks of acquiring CDI [28—31]. However, older age was not associated with CDI in our UC patients, while other three mentioned risk factors were not assessed in this study. Additionally, our toxin A test was performed in conjunction with colonoscopy, and was done immediately after UC exacerbation. Therefore, the effect of prolonged hospital stay was insignificant in this study. It might be necessary to mention here that Infliximab, which potentially can promote opportunistic infection, was not an approved medication for UC in Japan at the time of this investigation and therefore, none of the patients in our cohort was exposed to infliximab. Recently, CDI is observed in patients with IBD without recent exposure to antibiotics [10,23,32]. According to a study of general population, medication with PPI was associated with CDI through neutralization of gastric acid [33—36]. In our cohort, 51 patients were treated with PPI at the time of toxin A test, and three of these 51 patients were found to be CDI versus 12 CDI patients in 70 PPI free UC patients, suggesting no association between PPI therapy and CDI. We retrospectively investigated the prognosis in patients we surveyed, and then colectomy rate was recorded. Our logistic regression analyses indicated that CDI was not a predictor of colectomy. This observation is contrary to previous investigations [12,13,15,37] which have reported otherwise. However, considering that we diligently detected CD toxin A and immediately treated the infected patients with antibiotics, this difference could have impacted the outcome (no association between CDI and colectomy rate). This is to say that an increased risk of colectomy following a UC flare, which is complicated by concomitant CDI might have been compromised by an intensive antibiotic treatment [38]. In our patients, 40 of 96 (41.7%) with an active UC flare who visited our outpatient clinic had concomitant CDI,
319 suggesting that CDI in IBD patients was contracted in the community. It may also be true to say that IBD by itself causes the disruption of the normal flora in the colon and this may predispose IBD patients to CDI [24,39]. However, it is unclear how many patients are asymptomatic careers during their remission period and how many develop colitis later on. The rate of asymptomatic career in IBD patients is reported to be higher than general population [40]. Clayton et al. found that an 8.2% of IBD asymptomatic careers without prior exposure to antibiotics, corticosteroids, immunomodulators, or recent hospitalization were CDI positive [3,39]. Therefore, further prospective studies are warranted. In our cohort, CDI was detected in 40.1% of the 137 patients with active UC. This might be higher than the level in a general active UC population in Japan because our data included more than half of chronic refractory patients who had been introduced to us from other hospitals. However, considering that the average annual positive rate of CD toxin A test in our hospital laboratory was about 6%, the prevalence of CDI among our flared UC patients is in fact high. In conclusion, our results strongly indicate that patients with UC are a very high risk population for acquiring CDI. However, our univariate together with multivariate analyses of patients’ major demographic data did not identify any component of UC as a significant factor for predisposing UC patients to CDI. Likewise, CDI did not appear to be a predictor of colectomy. Finally, we believe that patients with active UC benefit from regular check ups for CDI and immediately receive medication to eradicate C. difficile.
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Responsibility disclaimer: None. Funding: None.
References [1] McDonald LC, Owings M, Jernigan DB. Clostridium difficile infection in patients discharged from US short-stay hospital, 1996—2003. Emerg Infect Dis 2006;12:409—15. [2] Pepin J, Valiquette L, Alary ME, Villemure P, Pelletier A, Forget K, et al. Clostridium difficile-associated diarrhea in a region of Quebec from 1991—2003: a changing pattern of disease severity. CMAJ 2004;171:466—72. [3] Paltansing S, van den Berg RJ, Guseinova RA, Visser CE, van der Vorm ER, Kuijper EJ. Characteristics and incidence Clostridium difficile-associated disease in The Netherlands, 2005. Clin Microbiol Infect 2007;13:1058—64. [4] McFarland LV, Mulligan ME, Kwok RY, Stann WE. Nosocomial acquisition of Clostridium difficile infection. N Engl J Med 1989;320:204—10. [5] Rodemann JF, Dubberke ER, Reske KA, Seo da H, Stone CD. Incidence of Clostridium difficile infection in inflammatory bowel disease. Clin Gastroenterol Hepatol 2007;5:339—44. [6] Kochhar R, Ayyagari A, Goenka MK, Dhali GK, Aggarwal R, Mehta SK. Role of infectious agents in exacerbations of ulcerative colitis in India. A study of Clostridium difficile. J Clin Gastroenterol 1993;16:26—30.
320 [7] Meyer AM, Ramzan NN, Loftus Jr EV, Heigh RI, Leighton JA. The diagnostic yield of stool pathogen studies during relapses of inflammatory bowel disease. J Clin Gastroenterol 2004;38:772—5. [8] Mylonaki M, Langmead L, Pantes A, Johnson F, Rampton DS. Enteric infection in relapse of inflammatory bowel disease: importance of microbiological examination of stool. Eur J Gastroenterol Hepatol 2004;16:775—8. [9] Issa M, Vijayapal A, Graham MB, Beaulieu DB, Otterson MF, Lundeen S, et al. Impact of Clostridium difficile on inflammatory bowel disease. Clin Gastroenterol Hepatol 2007;5:345—51. [10] Nguyen GC, Kaplan GG, Harris ML, Brant SR. A national survey of the prevalence and impact of Clostridium difficile infection among hospitalized inflammatory bowel disease patients. Am J Gastroenterol 2008;103:1443—50. [11] Ricciardi R, Ogilvie Jr JW, Roberts PL, Marcello PW, Concannon TW, Baxter NN. Epidemiology of Clostridium difficile colitis in hospitalized patients with inflammatory bowel disease. Dis Colon Rectum 2009;52:40—5. [12] Issa M, Ananthakrishnan AN, Binion DG. Clostridium difficile and inflammatory bowel disease. Inflamm Bowel Dis 2008;14:1432—42. [13] Ananthakrishnan AN, McGinley EL, Binion DG. Excess hospitalization burden associated with Clostridium difficile in patients with inflammatory bowel disease. Gut 2008;57:205—10. [14] Chiplunker A, Ananthakrishnan AN, Beaulieu DB, Naik AS, Zadvormova Y, Skaros S, et al. Long-term impact of Clostridium difficile on inflammatory bowel disease. Gastroenterology 2009;136(Suppl. 1):A-199. [15] Greenstein AJ, Byrn JC, Zhang LP, Swedish KA, Jahn AE, Divino CM. Risk factors for the development of fulminant Clostridium difficile colitis. Surgery 2008;143:623—9. [16] Matsumoto T, Andoh A, Okawa K, Ito H, Torii A, Yoshikawa S, et al. Multivariate analysis for factors predicting rapid response of leukocytapheresis in patients with steroid-resistant ulcerative colitis: a multicenter prospective open-label study. Ther Apher Dial 2008;12:484—90. [17] Truelove SC, Witts LJ. Cortisone in ulcerative colitis; final report on a therapeutic trial. Br Med J 1955;2:1041—8. [18] Irving PM, Gibson PR. Infections and IBD. Nat Clin Pract Gastroenterol Hepatol 2008;5:18—27. [19] Stallmach A, Carstens O. Role of infections in the manifestation or reactivation of inflammatory bowel disease. Inflamm Bowel Dis 2002;8:213—8. [20] Stone CD. Prognosis in Clostridium difficile infection complicating inflammatory bowel disease. Gut 2008;57:150—2. [21] Gryboski JD. Clostridium difficile in inflammatory bowel disease relapse. J Pediatr Gastroenterol Nutr 1991;13:39—41. [22] Viget N, Vernier-Massouille G, Salmon-Ceron D, Yazdanpanah Y, Colombel JF. Opportunistic infections in patients with inflammatory bowel disease: prevention and diagnosis. Gut 2008;57:549—58. [23] Pascarella F, Martinelli M, Miele E, Del Pezzo M, Roscetto E, Staiano A. Impact of Clostridium difficile infection on pediatric inflammatory bowel disease. J Pediatr 2009;154:854—8. [24] Musa S, Thomson S, Cowan M, Rahman T. Clostridium difficile infection and inflammatory bowel disease. Scand J Gastroenterol 2010;45:261—72.
T. Kaneko et al. [25] Powell N, Jung SE, Krishnan B. Clostridium difficile infection and inflammatory bowel disease: a marker for disease extent? Gut 2008;57:1183—4. [26] Ben-Horin S, Margalit M, Bossuyt P, Maul J, Shapira Y, Bojic D, et al. Combination immunomodulator and antibiotic treatment in patients with inflammatory bowel disease and Clostridium difficile infection. Clin Gastroenterol Hepatol 2009;7: 981—7. [27] Schneeweiss S, Korzenik J, Solomon DH, Canning C, Lee J, Bressler B. Infliximab and other immunomodulating drugs in patients with inflammatory bowel disease and the risk of serious bacterial infections. Aliment Pharmacol Ther 2009;30:253—64. [28] Bartlett JG, Clinical practice. Antibiotic-associated diarrhea. N Engl J Med 2002;346:334—9. [29] Barbut F, Petit JC. Epidemiology of Clostridium difficileassociated infection. Clin Microbiol Infect 2001;7: 405—10. [30] Redelings MD, Sorvillo F, Mascola L. Increase in Clostridium difficile-related mortality rates, United States, 1999—2004. Emerg Infect Dis 2007;13:1417—9. [31] Bignardi GE. Risk factors for Clostridium difficile infection. J Hosp Infect 1998;40:1—15. [32] Bossuyt P, Verhaegen J, Van Assche G, Rutgeerts P, Vermeire S. Increasing incidence of Clostridium difficile-associated diarrhea in inflammatory bowel disease. J Crohn’s Colitis 2009;3:4—7. [33] Dial S, Alrasadi K, Manoukian C, Huang A, Menzies D. Risk of Clostridium difficile diarrhea among hospital inpatients prescribed proton pump inhibitors: cohort and case-control studies. CMAJ 2004;171:33—8. [34] Dial S, Delaney JA, Barkun AN, Suissa S. Use of gastric acid-suppressive agents and the risk of communityacquired Clostridium difficile-associated disease. JAMA 2005;294:2989—95. [35] Hauben M, Horn S, Reich L, Younus M. Association between gastric acid suppressants and Clostridium difficile colitis and community-acquired pneumonia: analysis using pharmacovigilance tools. Int J Infect Dis 2007;11:417—22. [36] Aseeri M, Schroeder T, Kramer J, Zackula R. Gastric acid suppression by proton pump inhibitors as a risk factor for Clostridium difficile-associated diarrhea in hospitalized patients. Am J Gastroenterol 2008;103:2308—13. [37] Jodorkovsky D, Young Y, Abreu MT. Clinical outcomes of patients with ulcerative colitis and co-existing Clostridium difficile infection. Dig Dis Sci 2010;55:415—20. [38] Stange EF, Travis SP, Vermeire S, Reinisch W, Geboes K, Barakauskiene A, et al. European evidence-based consensus on the diagnosis and management of ulcerative colitis: definitions and diagnosis. J Crohn’s Colitis 2008;2:1—23. [39] Clayton EM, Rea MC, Shanahan F, Quigley EM, Kiely B, Hill C, et al. The vexed relationship between Clostridium difficile and inflammatory bowel disease: an assessment of carriage in an outpatient setting among patients in remission. Am J Gastroenterol 2009;104:1162—9. [40] Dorman SA, Liggoria E, Winn Jr WC, Beeken WL. Isolation of Clostridium difficile from patients with inactive Crohn’s disease. Gastroenterology 1982;82:1348—51.