Impact of Clostridium difficile on Inflammatory Bowel Disease
2007; Elsevier BV; Volume: 5; Issue: 3 Linguagem: Inglês
10.1016/j.cgh.2006.12.028
ISSN1542-7714
AutoresMazen Issa, Aravind Vijayapal, Mary Beth Graham, Dawn B. Beaulieu, Mary F. Otterson, Sarah Lundeen, Susan Skaros, Lydia R. Weber, Richard Komorowski, Josh F. Knox, Jeanne Emmons, Jasmohan S. Bajaj, David G. Binion,
Tópico(s)Gastrointestinal motility and disorders
ResumoBackground & Aims:Clostridium difficile–associated disease has increased significantly in North American medical centers. The impact of C difficile on patients with IBD (Crohn’s disease, ulcerative colitis) at the present time is unknown. Methods: A retrospective, observational study evaluating IBD patients followed in a referral center to evaluate the impact of C difficile was performed. Diagnosis was confirmed with stool toxin analysis. Demographic information, diagnosis, anatomic location, IBD therapy, antibiotic exposure, hospitalizations, and surgeries were recorded. Available endoscopic and histologic data were evaluated. Results: Rate of C difficile infection increased from 1.8% of IBD patients in 2004 to 4.6% in 2005 (P < .01). Proportion of IBD patients within the total number of C difficile infections at our institution increased from 7% in 2004 to 16% in 2005 (P < .01). IBD colonic involvement was found in the majority of C difficile–infected patients in 2005 (91%), and the majority contracted infection as an outpatient (76%). Antibiotic exposure was identified in 61% of IBD patients with C difficile infection in 2005. Pseudomembranes and fibrinopurulent eruptions were not seen endoscopically or histologically. During 2004–2005 more than half of the infected IBD patients required hospitalization, and 20% required colectomy. Univariate and multivariate analysis identified maintenance immunomodulator use and colonic involvement as independent risk factors for C difficile infection in IBD. Conclusions:C difficile infection has increased significantly in IBD patients and negatively impacts clinical outcome. Increased vigilance regarding this infection in IBD patients with colitis activity is warranted. Background & Aims:Clostridium difficile–associated disease has increased significantly in North American medical centers. The impact of C difficile on patients with IBD (Crohn’s disease, ulcerative colitis) at the present time is unknown. Methods: A retrospective, observational study evaluating IBD patients followed in a referral center to evaluate the impact of C difficile was performed. Diagnosis was confirmed with stool toxin analysis. Demographic information, diagnosis, anatomic location, IBD therapy, antibiotic exposure, hospitalizations, and surgeries were recorded. Available endoscopic and histologic data were evaluated. Results: Rate of C difficile infection increased from 1.8% of IBD patients in 2004 to 4.6% in 2005 (P < .01). Proportion of IBD patients within the total number of C difficile infections at our institution increased from 7% in 2004 to 16% in 2005 (P < .01). IBD colonic involvement was found in the majority of C difficile–infected patients in 2005 (91%), and the majority contracted infection as an outpatient (76%). Antibiotic exposure was identified in 61% of IBD patients with C difficile infection in 2005. Pseudomembranes and fibrinopurulent eruptions were not seen endoscopically or histologically. During 2004–2005 more than half of the infected IBD patients required hospitalization, and 20% required colectomy. Univariate and multivariate analysis identified maintenance immunomodulator use and colonic involvement as independent risk factors for C difficile infection in IBD. Conclusions:C difficile infection has increased significantly in IBD patients and negatively impacts clinical outcome. Increased vigilance regarding this infection in IBD patients with colitis activity is warranted. See CME exam on page 266. An increased incidence and virulence of Clostridium difficile, the anaerobic spore-forming bacillus associated with diarrheal illness and pseudomembranous colitis, has recently been identified in multiple North American medical centers.1Bartlett J.G. Perl T.M. The new Clostridium difficile: what does it mean?.N Engl J Med. 2005; 353: 2503-2505Crossref PubMed Scopus (144) Google Scholar, 2McDonald L.C. Killgore G.E. Thompson A. et al.An epidemic, toxin gene-variant strain of Clostridium difficile.N Engl J Med. 2005; 353: 2433-2441Crossref PubMed Scopus (1741) Google Scholar, 3Loo V.G. Poirier L. Miller M.A. et al.A predominantly clonal multi-institutional outbreak of Clostridium difficile-associated diarrhea with high morbidity and mortality.N Engl J Med. 2005; 353: 2442-2449Crossref PubMed Scopus (1675) Google Scholar, 4Dial S. Delaney J.A. Barkun A.N. et al.Use of gastric acid-suppressive agents and the risk of community-acquired Clostridium difficile-associated disease.JAMA. 2005; 294: 2989-2995Crossref PubMed Scopus (813) Google Scholar U.S. hospital discharges have documented a doubling of C difficile–associated disease between 1996–2003.5McDonald L.C. Owings M. Jernigan D.B. Clostridium difficile infection in patients discharged from US short-stay hospitals, 1996-2003.Emerg Infect Dis. 2006; 12: 409-415Crossref PubMed Google ScholarC difficile produces several exotoxins, including toxin A and toxin B, which lead to a clinical syndrome manifesting as diarrhea, pseudomembranous colitis, and in its most severe forms toxic megacolon, sepsis, perforation, and death. In addition to the serious morbidity and mortality, C difficile–associated disease significantly increases the costs of care among patients who contract this infection. Historically, subgroups of patients are known to be at increased risk for the acquisition of C difficile–associated disease.6Bignardi G.E. Risk factors for Clostridium difficile infection.J Hosp Infect. 1998; 40: 1-15Abstract Full Text PDF PubMed Scopus (526) Google Scholar Patients recently treated with broad-spectrum antibiotics,7Wistrom J. Norrby S.R. Myhre E.B. et al.Frequency of antibiotic-associated diarrhoea in 2462 antibiotic-treated hospitalized patients: a prospective study.J Antimicrob Chemother. 2001; 47: 43-50Crossref PubMed Scopus (327) Google Scholar hospitalized patients,8McFarland L.V. Mulligan M.E. Kwok R.Y. et al.Nosocomial acquisition of Clostridium difficile infection.N Engl J Med. 1989; 320: 204-210Crossref PubMed Scopus (1145) Google Scholar, 9Johnson S. Clabots C.R. Linn F.V. et al.Nosocomial Clostridium difficile colonisation and disease.Lancet. 1990; 336: 97-100Abstract PubMed Scopus (287) Google Scholar oncology patients, and immunocompromised individuals as well as the elderly are believed to be at increased risk for C difficile–associated disease. Patients with the 2 major forms of IBD, Crohn’s disease and ulcerative colitis, share many of these same clinical risk factors for the development of C difficile–associated disease. Many IBD patients are maintained on long-term immunosuppression, frequently require antibiotic use for their treatment, and are often hospitalized. Past reports published during the 1980s from both the U.S. and Europe did not demonstrate an increased association of C difficile infection in IBD patients who were experiencing disease flare. More recent reports have suggested that up to 20% of IBD flare was associated with a positive stool analysis for C difficile.10Meyer A.M. Ramzan N.N. Loftus Jr, E.V. et al.The diagnostic yield of stool pathogen studies during relapses of inflammatory bowel disease.J Clin Gastroenterol. 2004; 38: 772-775Crossref PubMed Scopus (104) Google Scholar Given that the ability to clear infection is dependent on the generation of an immune response against C difficile toxin A,11Kyne L. Warny M. Qamar A. et al.Association between antibody response to toxin A and protection against recurrent Clostridium difficile diarrhoea.Lancet. 2001; 357: 189-193Abstract Full Text Full Text PDF PubMed Scopus (662) Google Scholar IBD patients are a particularly high-risk subgroup. Because IBD exacerbation is routinely treated with high-dose corticosteroids, the ability to mount an effective antibody response against C difficile might be compromised, further worsening this infectious complication. The impact of C difficile on IBD patients in the present age of increased infection rates and more severe disease is not defined. We examined the impact of C difficile on a tertiary referral population of IBD patients during the past 5 years (2000–2005). We report the following findings: (1) a significant increase in the number and rate of IBD patients who contracted C difficile–associated disease; (2) a significant association between IBD diagnosis and contraction of C difficile infection compared with the general hospital and clinic population; (3) a significant association of IBD colitis and maintenance immunomodulator use in patients who contracted C difficile infection; (4) a majority of IBD patients acquired C difficile as outpatients; and (5) the severe deleterious effect of C difficile on the clinical course of IBD, which frequently leads to hospitalization and colectomy. This was a retrospective, observational study evaluating all patients who were followed in the Medical College of Wisconsin’s Inflammatory Bowel Disease Center between the years 2000–2005. Patients who tested positive for the C difficile toxin A and/or toxin B stool enzyme-linked immunosorbent assay (ELISA) were considered infected if they presented with concomitant symptoms of colitis (ie, diarrhea, increased stool frequency, rectal bleeding, cramping, and/or tenesmus). No patients with presumptive C difficile infection in the absence of a positive toxin assay were included in this analysis. The total number of IBD patients followed during the calendar years 2004 and 2005 were recorded and used to determine rates of C difficile infection. The total number of C difficile–infected patients who were followed at Froedtert Hospital, the Medical College of Wisconsin’s major adult teaching hospital, in Milwaukee, WI were recorded annually. Finally, the total number of outpatient visits and inpatient admissions for all of Froedtert Hospital and its associated clinics as well as the total number of IBD Center patient encounters (admissions and outpatient visits) were tabulated and used to compare overall rates of C difficile infection during 2004 and 2005. IBD patient demographics, including age, gender, type of IBD (ie, Crohn’s disease, ulcerative colitis), anatomic distribution of disease (isolated small bowel vs any colonic involvement), type of therapy (mesalamine, immunomodulators [azathioprine, 6-mercaptopurine, methotrexate]), anti–tumor necrosis factor–α therapy (ie, infliximab [Remicade; Centocor, Horsham, PA]), adalimumab (Humira; Abbott Laboratories, Abbott Park, IL), antibiotics, corticosteroids (systemic [ie, prednisone] vs topical [ie, budesonide]), and smoking history were recorded. Among the IBD patients who tested positive, known risk factors for the diagnosis of C difficile were recorded. This included recent hospitalization within 2 months of the time of C difficile detection and use of antibiotics within 2 months of the time of infection. Statistical analysis was performed with χ2 analysis, as well as univariate and multivariate logistic regression analysis, by using MINITAB (Minitab Inc, State College, PA). This study was approved by the Medical College of Wisconsin’s human research review committee. Demographic and disease characteristics for IBD patients from the year 2005 with and without C difficile are displayed in Table 1. Patients who were positive for C difficile in 2005 were statistically more likely to have colonic involvement, more likely to be on immunomodulator therapy, and had significantly shorter disease duration than those without C difficile (Table 1). Seven C difficile–infected patients presented at the time of IBD diagnosis.Table 1IBD Patient Demographic and Disease Characteristics in 2005C difficile positive (n = 46)C difficile negative (n = 953)P valueAge (y)38.8 ± 14.643.2 ± 14.6NSFemale25535NSMale21418Crohn’s disease30 (65%)707 (74%)NSUlcerative colitis16 (35%)246 (26%)Duration of IBD (mean ± standard deviation, y)7.5 ± 8.013.9 ± 11.8.004Colonic IBD involvement42 (91%)675 (71%).002Immunomodulator maintenance34 (74%)535 (56%).02Biologic therapy13 (28%)208 (22%)NSCurrent tobacco use10 (27%)153 (16%)NSNOTE. Immunomodulator maintenance included azathioprine, 6-mercaptopurine, and methotrexate. Biologic therapy included infliximab and adalimumab. Open table in a new tab NOTE. Immunomodulator maintenance included azathioprine, 6-mercaptopurine, and methotrexate. Biologic therapy included infliximab and adalimumab. C difficile was rarely detected in IBD patients before 2003 and increased to 14 cases in 2004 and 46 cases in 2005 (Figure 1). This corresponded to a significant increase in the rate of C difficile infection from 1.8% in 2004 to 4.6% of the IBD Center patient population in 2005 (P < .01, χ2 analysis). The total number of C difficile–infected patients followed in our hospital system increased between the years 2000–2005, with 207 cases in 2004 and 287 detected in 2005 (Figure 2). IBD patients accounted for 4% of the total C difficile–infected patient cohort in 2003, which increased to 7% in 2004 and 16% in 2005. The percentage of C difficile–infected patients who had concomitant IBD increased significantly between 2004–2005 (P < .01, χ2 analysis). We next compared the rates of C difficile infection between the total cohort of patients who were followed in our hospital system and the IBD patients followed in our center. Rates of C difficile infection in non-IBD patients were higher in 2005 compared with 2004 (Table 2), and patients with IBD were at increased risk of developing C difficile infection compared with non-IBD patients in 2004 and 2005 (Table 2).Table 2Rate of C difficile Infections in 2004 and 200520042005Total non-IBD patient encounters (admissions and outpatient visits)455,682475,715Non-IBD cases with C difficile193241aSignificantly higher proportion of IBD patients with C difficile compared with non-IBD patients in 2005, compared with 2004 (P = .002).Total IBD patient encounters (admissions and outpatient visits)2,5472,349IBD cases with C difficile1446bSignificantly higher proportion of IBD patients with C difficile in 2005 compared with 2004 (P = .0001).NOTE. Odds ratio for developing C difficile in IBD patients compared with non-IBD patients: 13.0 (95% confidence interval, 7.5–22.4) in 2004 and 38.6 (95% confidence interval, 28.1–53.1) in 2005. Non-IBD patient encounters include both hospital admissions to Froedtert Hospital and visits to its affiliated outpatient clinics during 2004 and 2005. IBD patient encounters include inpatient admissions to Froedtert Hospital as well as outpatient clinic visits to the IBD Center during both 2004 and 2005.a Significantly higher proportion of IBD patients with C difficile compared with non-IBD patients in 2005, compared with 2004 (P = .002).b Significantly higher proportion of IBD patients with C difficile in 2005 compared with 2004 (P = .0001). Open table in a new tab NOTE. Odds ratio for developing C difficile in IBD patients compared with non-IBD patients: 13.0 (95% confidence interval, 7.5–22.4) in 2004 and 38.6 (95% confidence interval, 28.1–53.1) in 2005. Non-IBD patient encounters include both hospital admissions to Froedtert Hospital and visits to its affiliated outpatient clinics during 2004 and 2005. IBD patient encounters include inpatient admissions to Froedtert Hospital as well as outpatient clinic visits to the IBD Center during both 2004 and 2005. In 2004 Crohn’s disease accounted for 64% of the IBD patients who contracted C difficile and 65% of the diagnoses in 2005, with the remaining patients having UC. This preponderance of patients with Crohn’s disease was reflective of our IBD Center patient population, in which approximately 74% of the patients carried a diagnosis of Crohn’s disease in 2005. We then evaluated anatomic involvement in IBD patients who contracted C difficile infection. Colonic involvement was found in 93% of the 2004 patients and was present in 91% of the 2005 patients. Isolated small bowel Crohn’s disease was found in the remaining C difficile–infected IBD patients. The C difficile–infected IBD patients required 11 hospitalizations in 2004, which increased to 27 in 2005 (Figure 3). These admissions corresponded with 5 colectomies in 2004 and 7 colectomies in 2005. The rate of colectomy in hospitalized patients with IBD colitis with C difficile infection decreased from 45% in 2004 to 25% in 2005. During this time period colectomies for C difficile infection in non-IBD patients were performed in 3 patients during 2004 and 2 patients in 2005 at our institution. This implied that the majority of C difficile–associated colectomies occurred in IBD patients. The mean number of hospital days required for the care of IBD patients admitted with C difficile infection was 13.5 ± 8.8 days (mean ± standard deviation). During 2005, there were 12 hospitalizations for ulcerative colitis patients who were not C difficile–infected, and their mean length of stay was 6.0 ± 4.0 days (mean ± standard deviation). All IBD patients who tested positive for C difficile in the setting of colitis symptoms were treated with antibiotic therapy directed at the infection. This included the oral antibiotics vancomycin (oral or enema administration) or rifaximin (oral) and metronidazole (oral or intravenous). Hospitalized patients who were unable to tolerate oral medication were treated with intravenous metronidazole. Because of the high colectomy rate associated with C difficile during 2004, the antibiotic approach at our institution shifted from using metronidazole to vancomycin as the primary antibiotic. In addition, hospitalized patients with C difficile infection during 2005 had their treatment regimen for IBD adjusted, with a rapid decrease in steroid dosing emphasized after detection of the infection. This alteration in the treatment regimen correlated with a lower rate of surgery compared with the 2004 experience in our IBD cohort. The diagnosis of C difficile is most commonly based on the detection of exotoxin A and B in stool samples by ELISA. We retrospectively reviewed the pattern of stool ELISA analysis in the IBD patients diagnosed with C difficile in 2005. Fifty-four percent of the infected IBD patients were diagnosed with the first stool sample, 75% were diagnosed by the second stool sample, 78% were diagnosed by the third stool sample, and 92% were diagnosed by the fourth stool analysis. One hospitalized IBD colitis patient was diagnosed with C difficile infection on the 8th stool sample after a prolonged hospitalization. Next we analyzed whether the IBD patients contracted C difficile as a result of hospitalization (nosocomial infection) versus community-acquired infection. Patients who tested positive within 48 hours of hospital admission were labeled community-acquired for this analysis. Community-acquired C difficile was detected in 79% of cases in 2004 and in 76% of cases in 2005. Immunosuppression is a known risk factor for the development of C difficile–associated disease.11Kyne L. Warny M. Qamar A. et al.Association between antibody response to toxin A and protection against recurrent Clostridium difficile diarrhoea.Lancet. 2001; 357: 189-193Abstract Full Text Full Text PDF PubMed Scopus (662) Google Scholar, 12Dallal R.M. Harbrecht B.G. Boujoukas A.J. et al.Fulminant Clostridium difficile: an underappreciated and increasing cause of death and complications.Ann Surg. 2002; 235: 363-372Crossref PubMed Scopus (517) Google Scholar, 13Johnson S. Gerding D.N. Clostridium difficile-associated diarrhea.Clin Infect Dis. 1998; 26: 1027-1036Crossref PubMed Scopus (343) Google Scholar We analyzed the overall use of immunosuppressive medications in IBD patients with C difficile infection. In 2004, 64% of the IBD patients were receiving immunosuppressive medications (ie, azathioprine, 6-mercaptopurine, methotrexate, infliximab, and/or adalimumab) at the time that they contracted C difficile. In 2005, the rate of immunosuppression among IBD patients who contracted C difficile was 78%. Infliximab use among the IBD patients who contracted C difficile was identified in 14% in 2004 and in 28% of patients in 2005. Among the IBD patients who contracted C difficile in 2005, 46% were receiving a purine analog (ie, azathioprine or 6-mercaptopurine) for maintenance therapy alone or in combination with infliximab, whereas 13% were on methotrexate maintenance alone or in combination with infliximab. Dosages of the immunomodulators were individualized on the basis of routine clinical practice, while taking into account patient ability to tolerate the drug and adverse reactions. The classic endoscopic finding in C difficile–associated disease is the presence of pseudomembranes on the colonic mucosal surfaces, which have been estimated to occur in approximately 50% of infected patients.14Kelly C.P. LaMont J.T. Clostridium difficile infection.Annu Rev Med. 1998; 49: 375-390Crossref PubMed Scopus (368) Google Scholar, 15Fekety R. Shah A.B. Diagnosis and treatment of Clostridium difficile colitis.JAMA. 1993; 269: 71-75Crossref PubMed Scopus (318) Google Scholar In the 60 IBD patients who exhibited C difficile infection during the years 2004–2005, there were 16 who underwent endoscopic assessment. No IBD patient with C difficile infection who underwent endoscopy exhibited the presence of discrete pseudomembranes. In contrast, the presence of a nonspecific mucopus was the most commonly encountered endoscopic finding (Figure 4). The classic histologic finding in C difficile–associated colitis is a volcanic eruption of pseudomembranes from the mucosa, composed of polymorphonuclear leukocytes, fibrin, chronic inflammatory cells, and epithelial debris. When colonic endoscopic pinch biopsies (n = 12 patient biopsy samples from a total of 16 endoscopies) and colectomy specimens (n = 12 specimens) from C difficile–infected IBD patients were evaluated histologically between 2004–2005, no classic fibrin eruptions or pseudomembranes were detected (Figure 5). We analyzed established risk factors for the acquisition of C difficile infection in the IBD patients who acquired disease in 2005. Sixty-one percent of the patients had a history of antibiotic use within 2 months of the detection of C difficile infection. The most commonly used antibiotic was the oral fluoroquinolone, ciprofloxacin. Thirty-nine percent of the patients did not have antibiotic exposure during the 2-month time period immediately before C difficile detection. An additional potential risk factor for the acquisition of C difficile infection in the IBD patients who tested positive included a healthcare occupation with patient contact in an institutional setting (7%). To determine associations between IBD clinical parameters and emergence of C difficile infection, univariate and multivariate logistic regression analysis was performed. Patients with C difficile infection during 2005 (n = 46) were analyzed compared with the entire cohort of IBD patients followed at the Medical College of Wisconsin’s IBD Center during 2005 (n = 999). Data on the following clinical parameters from 2005 were analyzed: Crohn’s disease versus ulcerative colitis; anatomic involvement (isolated small intestine vs any colonic involvement); tobacco history (any history of smoking); gender; hospitalization; surgery; use of isolated mesalamine treatment; use of immunomodulator maintenance therapy (azathioprine, 6-mercaptopurine, methotrexate); and use of biologic therapy (infliximab, adalimumab). Univariate analysis demonstrated that presence of IBD colonic involvement (P = .03; 95% confidence interval, 1.10–7.22; odds ratio, 2.82) and use of immunomodulators (P = .011; 95% confidence interval, 1.23–4.86; odds ratio, 2.44) were significantly associated with the development of C difficile infection. None of the other parameters tested, including use of biologic treatment, were significantly associated with C difficile infection in patients with IBD. Multivariate logistic regression analysis confirmed that both colonic involvement (P = .018; odds ratio, 3.12; 95% confidence interval, 1.22–8.02) and maintenance immunosuppression (P = .008; odds ratio, 2.56; 95% confidence interval, 1.28–5.12) were independently associated with the emergence of C difficile infection in patients with IBD. We report that C difficile–associated disease has significantly increased among patients with IBD who were followed in a tertiary referral setting during the past several years. IBD patients with colonic involvement exhibited a significant association with development of C difficile infection, with both Crohn’s colitis and ulcerative colitis patients comprising the majority of those infected. Immunosuppression was used in the majority of IBD patients who contracted C difficile infection and was significantly associated with the emergence of this infection by using univariate and multivariate regression analysis. The majority of IBD patients contracted C difficile infection as outpatients and did not manifest classic endoscopic or histologic features including pseudomembranes and fibrinous exudate. More than half of the C difficile–infected IBD patients required hospitalization, and the colectomy rate in this cohort during a 2-year period was 20%. Perhaps most important, IBD patients comprised a disproportionate percentage of all the patients who contracted C difficile infection in our institution during the past 2 years, which suggests that IBD colitis is an important and previously unrecognized risk factor for the development of this infection. C difficile is a gram-positive, spore-forming anaerobic bacteria that can colonize the large intestine, producing a range of clinical activity from asymptomatic carriage to severe fulminant colitis in humans.14Kelly C.P. LaMont J.T. Clostridium difficile infection.Annu Rev Med. 1998; 49: 375-390Crossref PubMed Scopus (368) Google Scholar, 16Hurley B.W. Nguyen C.C. The spectrum of pseudomembranous enterocolitis and antibiotic-associated diarrhea.Arch Intern Med. 2002; 162: 2177-2184Crossref PubMed Scopus (193) Google Scholar There are 2 prerequisites for developing C difficile–associated diarrhea: disruption of the normal gastrointestinal flora causing diminished colonization resistance favoring C difficile and acquisition of the organism from an exogenous source.13Johnson S. Gerding D.N. Clostridium difficile-associated diarrhea.Clin Infect Dis. 1998; 26: 1027-1036Crossref PubMed Scopus (343) Google ScholarC difficile–associated disease is frequently but not always an iatrogenic complication of broad-spectrum antibiotics and often complicates the clinical course of hospitalized patients. Other important clinical factors include host susceptibility, virulence of the C difficile strain concerned, and the nature and extent of antimicrobial exposure.17McFarland L.V. Alternative treatments for Clostridium difficile disease: what really works?.J Med Microbiol. 2005; 54: 101-111Crossref PubMed Scopus (155) Google Scholar We found that recent use of antibiotics, specifically oral fluoroquinolones, occurred in the majority of IBD patients who contracted C difficile. Indeed, many of the patients, particularly those with Crohn’s disease colitis, had received ciprofloxacin in an attempt to provide treatment for IBD flare. Our data suggest that the empiric use of broad-spectrum antibiotics, which are associated with the development of C difficile, as a routine treatment of IBD flare should be considered cautiously. Recent reports have described a significant increase in the incidence of C difficile–associated disease throughout North America. The National Nosocomial Infection Surveillance system has reported a significant increase in the C difficile rates between 1987–2001.5McDonald L.C. Owings M. Jernigan D.B. Clostridium difficile infection in patients discharged from US short-stay hospitals, 1996-2003.Emerg Infect Dis. 2006; 12: 409-415Crossref PubMed Google Scholar, 18National Nosocomial Infections Surveillance (NNIS) System Report, data summary from January 1992 through June 2003, issued August 2003.Am J Infect Control. 2003; 31: 481-498Abstract Full Text Full Text PDF PubMed Scopus (604) Google Scholar Individual medical centers, including the University of Pittsburgh, described a doubling in the incidence rate of C difficile–associated disease in 2000–2001 compared with 1990–1999.12Dallal R.M. Harbrecht B.G. Boujoukas A.J. et al.Fulminant Clostridium difficile: an underappreciated and increasing cause of death and complications.Ann Surg. 2002; 235: 363-372Crossref PubMed Scopus (517) Google Scholar Case series from a Canadian hospital in Sherbrooke, Quebec, have described a 10-fold increase in C difficile incidence.19Pepin J. Valiquette L. Alary M.E. et al.Clostridium difficile-associated diarrhea in a region of Quebec from 1991 to 2003: a changing pattern of disease severity.CMAJ. 2004; 171: 466-472Crossref PubMed Scopus (949) Google Scholar Our data suggest that this increased detection of C difficile in hospitalized patients has been paralleled by increased infection in patients with IBD. C difficile–associated disease accounts for considerable increases in the length of hospital stay and more than $1.1 billion in healthcare costs each year in the United States.20Kyne L. Hamel M.B. Polavaram R. et al.Health care costs and mortality associated with nosocomial diarrhea due to Clostridium difficile.Clin Infect Dis. 2002; 34: 346-353Crossref PubMed Scopus (613) Google Scholar Each case of C difficile–associated disease has been estimated to result in more than $3600 in excess healthcare costs per patient at the present time. Our experience with C difficile infection in IBD was similar, with hospitalization and surgery emerging as frequent complications in patients who contracted this illness, in which the mean duration of admission was close to 2 weeks. The severity of illness and resultant economic burden from C difficile might be linked to increased virulence of isolates identified at the present time.2McDonald L.C. Killgore G.E. Thompson A. et al.An epidemic, toxin gene-variant strain of Clostridium difficile.N Engl J Med. 2005; 353: 2433-2441Crossref PubMed Scopus (1741) Google Scholar In a recent report, 50% of C difficile isolates from 8 healthcare facilities in 6 states belonged to BI/NAP1 that were positive for the binary toxin CDT. This C difficile isotype carries a higher rate of resistance to gatifloxacin and moxifloxacin and has been blamed for epidemics both inside and outside the U.S. Increased virulence of C difficile has also translated into an increased case-fatality rate. Published data from Quebec, Canada suggest that the proportion of cases that were complicated increased from 7.1% (12/169) in 1991–1992 to 18.2% (71/390) in 2003 (P < .001), and the proportion of patients who died within 30 days after diagnosis increased from 4.7% (8/169) in 1991–1992 to 13.8% (54/390) in 2003 (P < .001). Fortunately, no deaths were seen in the IBD patients during the time of C difficile infection at our center. None of our C difficile–infected IBD patients underwent specific microbiologic isolate identification. The epidemiology of C difficile–associated disease has changed during the past decade. Traditional risk factors, including advanced age (ie, age older than 65 years) and use of broad-spectrum antibiotics, must now include use of fluoroquinolone as a major risk factor.21Pepin J. Saheb N. Coulombe M.A. et al.Emergence of fluoroquinolones as the predominant risk factor for Clostridium difficile-associated diarrhea: a cohort study during an epidemic in Quebec.Clin Infect Dis. 2005; 41: 1254-1260Crossref PubMed Scopus (820) Google Scholar Our multivariate analysis demonstrated that immunomodulator use and presence of IBD colitis (either ulcerative colitis or Crohn’s colitis) are also significant risk factors for the development of C difficile infection. The larger question of why C difficile–associated disease has increased over recent years remains unanswered. This question is highlighted by the fact that 39% of the IBD patients in our cohort who contracted C difficile in 2005 did not have identifiable antibiotic exposure during the recent time period before manifesting infection, suggesting that other mechanisms for the acquisition of the pathogen might be involved. C difficile infection in IBD patients followed in our referral cohort had significant clinical ramifications and posed additional challenges in management. A majority of the C difficile–infected patients required hospitalization during the 2004–2005 time period. Detection of C difficile was often difficult, because classic endoscopic and histologic features were not readily identified in our IBD patients. Stool ELISA analysis for the presence of toxin A and B also demonstrated low sensitivity, mandating that multiple samples be routinely processed to accurately diagnose patients. Therefore, heightened vigilance regarding C difficile infection in IBD patients experiencing disease flare is warranted, because it occurs more commonly, fails to demonstrate classic morphologic features, and will frequently require multiple stool assays to detect. Patients with immunosuppression from chemotherapy or as a desired therapeutic goal in the setting of organ transplantation are known to be at increased risk for C difficile–associated disease. Our multivariate analysis confirmed that the use of maintenance immunomodulator therapy in IBD patients was significantly associated with C difficile infection. Interestingly, use of anti–tumor necrosis factor–α antibody therapy did not correlate with increased risk for acquisition of C difficile–associated disease in patients with IBD. These findings suggest that specific regimens of immunosuppression might carry differential risk for the acquisition of C difficile infection and associated disease. Colonic involvement with IBD was also significantly associated with C difficile infection, and isolated small bowel Crohn’s disease was found in a disproportionately lower number of patients. The reason for IBD colitis predisposing to C difficile infection is not clear. We hypothesize that the previously damaged and chronically inflamed colonic mucosa that was subject to chronic inflammation becomes more susceptible to infection with C difficile. We also suspect that the lack of pseudomembranes and fibrinous exudate encountered during endoscopic and histologic evaluation in IBD patients was related to preexisting mucosal changes associated with chronic inflammation as well. Although our article did not evaluate specific treatment regimens, 3 different antibiotics were used against C difficile among the 46 IBD patients who tested positive during 2005. Antibiotics included metronidazole, oral vancomycin, and rifaximin.22Mignini F. Falcioni E. Prenna M. et al.Antibacterial activity of rifaximin against Clostridium difficile, Campylobacter jejunii and Yersinia spp.J Chemother. 1989; 1: 220-222PubMed Google Scholar Initial courses of antibiotic were unsuccessful in 58% of the patients (27/46), who required additional courses of treatment for recurrent C difficile infection. A diminished therapeutic response to metronidazole was reported by Musher et al23Musher D.M. Aslam S. Logan N. et al.Relatively poor outcome after treatment of Clostridium difficile colitis with metronidazole.Clin Infect Dis. 2005; 40: 1586-1590Crossref PubMed Scopus (427) Google Scholar in 2005, who described a failure rate of 50% in C difficile–infected patients, which is in contrast to response rates of 90% described a decade prior. We found similar problems with metronidazole in our cohort of IBD patients infected with C difficile, and we changed our practice pattern in 2005 to use oral vancomycin as the primary antibiotic against this infection. Use of oral vancomycin in C difficile–infected IBD patients in 2005 corresponded with a diminished colectomy rate compared with our experience in 2004. However, vancomycin treatment of C difficile is also challenged by high rates of recurrence, and precise algorithms for treatment, including optimal dosages, combinations of agents, durations of therapy, and use of antibiotic tapering regimens, have yet to be defined. The reasons underlying the loss of response to metronidazole in C difficile infections at the present time have not been defined. Past published information regarding C difficile infection in IBD patients from 2 decades ago suggested that this was a minor clinical issue.24Rolny P. Jarnerot G. Mollby R. et al.Occurrence of Clostridium difficile toxin in inflammatory bowel disease.Scand J Gastroenterol. 1983; 18: 61-64Crossref PubMed Scopus (58) Google Scholar A Scandinavian study from 1983 stated “In our region there is no need for routine screening for this in symptomatic patients with IBD.” A U.S. study from pediatric patients in 1985 concurred,25Hyams J.S. McLaughlin J.C. Lack of relationship between Clostridium difficile toxin and inflammatory bowel disease in children.J Clin Gastroenterol. 1985; 7: 387-390Crossref PubMed Scopus (17) Google Scholar stating “Routine screening for C difficile toxin in children with inflammatory bowel disease appears unwarranted.” Our findings from the present age directly challenge these conclusions. C difficile has significantly increased during the past several years, and this trend is especially evident in patients with IBD, who appear to be at marked risk of infection. Our findings suggest that IBD patients with colonic involvement and those using maintenance immunomodulator therapy appear to be at highest risk for contracting C difficile. Detection of this infection in IBD patients is challenging, because classic pseudomembranes are not commonly encountered, and multiple stool samples are necessary to detect C difficile toxin in half of patients. Clinically, C difficile negatively impacts IBD patients, and a majority require hospitalization for management. In summary, clinicians must be vigilant regarding the potential for C difficile to contribute to colitis flare to rapidly identify and optimally treat IBD patients. We would like to thank H. Brandenburg for expert assistance with preparation of the manuscript. Exams 1 and 2Clinical Gastroenterology and HepatologyVol. 5Issue 3Preview Full-Text PDF
Referência(s)