Clostridium difficile and Inflammatory Bowel Disease: What are the risks?

David G. Binion, M.D. Co-Director, Inflammatory Bowel Disease Center Director, Translational Inflammatory Bowel Disease Research Division of Gastroenterology, Hepatology and Nutrition UPMC Presbyterian Hospital Visiting Professor of Medicine University of Pittsburgh School of Medicine Pittsburgh,PA I. Background – Gut microflora and intestinal immune health!

II. Impact of C. difficile on IBD

III. Diagnostic considerations C. difficile

IV. Treatment considerations

>100 trillion bacteria 1-2 kg of body weight Up to 1 trillion/gm 1/3 of the dry weight of stool 2-4 million genes 10-20% culturable Normal GI mucosa Surface area with external environment Largest immune organ in body Physiologic inflammation Oral tolerance “You are only 10% human”

Crohn’s disease

Destructive chronic inflammation

No “autoimmune target” identified in body

Impaired tolerance to enteric flora? Stomach 0-102 Lactobacillus Candida 2 Streptococcus Duodenum 10 Helicobacter pylori Streptococcus Peptostreptococcus Lactobacillus

7 8 Distal Ileum 10 -10 2 Clostridium Jejunum 10 Bacteroides sp Streptococcus Coliforms Lactobacillus

Colon 1011 Bacteroides Bifidobacterium Clostridium coccoides Proximal Ileum 103 Clostridium lepium/fusobacterium Streptococcus Lactobacillus I. Clostridium difficile – the “difficult” bacteria

. 1930’s - Bacillus difficillis first described as part of the normal flora of neonates. Difficult to culture. . 1974 - C. difficile recognized as complication of Clindamycin use.

. 1978 - C. difficile identified as the cause of antibiotic-associated pseudomembranous colitis in humans.

. Clinical syndrome may range from watery diarrhea, abdominal pain, pseudo-membranous colitis, toxic megacolon, sepsis, colonic perforation and death * Hall, I. and E. O'Toole. Am J Dis Child, 1935. 48: p. 390-402. ‡Tedesco, F.J et al. Ann Intern Med, 1974. 81(4): p. 429-33. C. difficile: Changing spectrum of clinical disease

. In the past: C. difficile linked to antibiotic use. Most cases treated successfully with metronidazole . Doubling of C. difficile associated disease between 1996 - 2003.

. Diminished therapeutic response to metronidazole (50% failure rate with initial course of treatment).

* McDonald LC et al. Emerg Infect Dis 2006;12:409-415. ‡ Musher DM et al. Clin Infect Dis 2005;40:1586-1590. 700 C. difficile related deaths in Quebec, Canada in one year (2003-4)

400 C. difficile related deaths annually in Quebec at the present time BI/NAP1 Epidemic strain C. difficile

. Regional outbreaks - Pittsburgh, PA, Quebec, Canada and the mid-Atlantic and southeastern U.S.

. C. difficile in low risk populations – young individuals, peripartum women, community dwelling and in individuals with no exposure to antibiotics.

Muto, C.A., et al., Infect Control Hosp Epidemiol, 2005. 26(3): p. 273-80. McDonald, L.C., et al.,N Engl J Med, 2005. 353(23): p. 2433-41. 2005. MMWR Morb Mortal Wkly Rep, 2005. 54(47): p. 1201-5. C. difficile Epidemic in U.S. BI/NAP1 C. difficile in U.S. Nov. 2007 (n = 38)

HI P AK R

BI/NAP1 C. difficile in U.S. DC Oct. 2008 IDSA Meetings

HI P AK R

Centers for Disease Control and Prevention. Data & Statistics about Clostridium difficile infections. www.cdc.gov/ncidod/dhqp/ id_cdiff_data.html Current burden of C. difficile in U.S.

. October 2008 – BI/NAP1 has been isolated in all 50 states (IDSA).

. Total number of C. difficile cases annually in U.S. is >500,000.

. Total number of C. difficile related deaths annually in the U.S. is >15,000.

. Epidemic is predicted to worsen.

. Cause? Where does the majority of antibiotic use occur in the U.S.? Antibiotic use in food animal industry

. Poultry industry – antibiotic use to prevent diarrheal illness

. Corn fed beef require antibiotics to prevent bacterial overgrowth BI/NAP1 Epidemic strain C. difficile and food animals

. Colonization and carriage with the epidemic strain C. difficile (B1 NAP1 strain) reported in cows.

. C. difficile has been isolated from retail ground meat purchased in Canada.

Songer, J.G. and M.A. Anderson, Clostridium difficile: an important pathogen of food animals. Anaerobe, 2006. 12(1): p. 1-4. Rodriguez-Palacios, A., et al., Clostridium difficile in retail ground meat, Canada. Emerg Infect Dis, 2007. 13(3): p. 485-7. C. difficile infectious inoculum is 10 spores

C. difficile spores may be resistant to cooking. Source of bacterial food poisoning?

Poutanen SM et al. CMAJ. July 6,2004;171(1). C. difficile: Pathogenic mechanisms

1) Antibiotic destroys 2) C. difficile grows 3) Toxins inflame 4) Damaged mucosa normal bacterial flora and secretes toxins and ulcerate mucosa secretes fluid causing diarrhea

Gut Lumen Fluid secretion

Colonic Mucosa Normal flora C. difficile Antibiotic Toxin PMN

Poutanen SM et al. CMAJ. July 6,2004;171(1). Clostridium difficile Clostridium difficile and IBD

. C. difficile and IBD present in identical fashion ranging from mild diarrhea to fulminant colitis.

. Early studies performed 2 decades ago indicated little overlap between C. difficile and IBD, concluding “No need for routine screening for C. difficile in IBD population”.

. Recent studies: Increasing incidence and severity of C. difficile in IBD population

. C. difficile recently identified to have a significant negative impact on IBD morbidity.

Kochlar R et al. J Clin Gastroenterol 1993;16:26-30. Issa, M., et al. Clin Gastroenterol Hepatol, 2007. 5(3): p. 345-51. Bolton RP et al. Lancet 1980;1:383-384 Rodemann, J.F., et al.Clin Gastroenterol Hepatol, 2007. 5(3): p. 339-44. Trnka Y et al. Gastroenterology 1981;80:693-696. Ananthakrishnan, et al. Gut, 2008. 57(2): p. 205-10. Increasing impact of C difficile on IBD

Rodemann JF et al.

Rodemann JF et al. Clin Gastroenterol Hepatol. 2007;5: 339-44. Meyer AM et al. J Clin Gastroenterol. 2004;38(9): 772-5 Increasing Impact of Clostridium difficile on IBD

50

40

30

Number of Patients Number of 20

10

0 1998 1999 2000 2001 2002 2003 2004 2005

P≤.01 Issa M et al. Clin Gastroenterol Hepatol. 2007;5: 345-51. Increasing Proportion of Clostridium difficile Patients With IBD

p=<0.01

350 16% Total C. diff patients 300

7% IBD patients 250 4% with C. diff 200

150 Number of Patients Number of

100

50

0 2000 2001 2002 2003 2004 2005

Issa M et al. Clin Gastroenterol Hepatol. 2007;5: 345-51. P≤.01 Complications: Clostridium difficile Infected Patients With IBD*

15%

36%

Issa M et al. Clin Gastroenterol Hepatol. 2007;5: 345-51. Endoscopic Appearance of C. difficile

Endoscopic appearance of C. diff in control patients

Endoscopic appearance of C. diff in patients with IBD

Ulcerative Colitis Crohn’s Disease

Issa M, et al. Clin Gastroenterol Hepatol. 2007;5: 345-51. Bossuyt P, et al. J Crohns Colitis 2009;3:4-7. 91% Colonic 61% Recent IBD antibiotic exposure

Issa M et al. Clin Gastroenterol Hepatol. 2007;5: 345-51. Clostridium difficile in IBD: Morbidity and Mortality IBD patients with C. difficile compared with IBD alone:

. Longer hospital stay

. Increased hospitalization costs . Higher colectomy rates

. Increased mortality rate –

. 118 IBD C.diff deaths in NIS 2004

. (>500 IBD C.diff deaths in U.S. 2004)

. UC C diff operative mortality 25%

4.2% 3.7% IBD pts with C. difficile C. difficile alone 0.5% IBD alone Ananthakrishnan AN, et al. Gut, 2008. 57(2): p. 205-10. Clostridium difficile in IBD: Increasing U.S. hospitalizations 2004 - 2007

60

50

(per 1000 40

IBD 30 Crohn's C difficile C difficile UC

hospitalizations) hospitalizations) 20 All patients

Proportion 10

0 1998 2004 2007 Year Ananthakrishnan AN et al. Med Clin N Am. 2009. C. difficile in IBD: Asymptomatic carriage

Clostridium difficile and host immune status . Increased asymptomatic carriage of C difficile in IBD – 8% . IgG immune response to toxin A seen in 50% of patients. . Immune response correlates with asymptomatic carriage. . Inability to mount immune response may result in infection and chronic, relapsing clinical course.

Clayton EM et al. Am J Gastroenterol. 2009;104: 1162-9. C. difficile in IBD: Impact of immunosuppression

Clostridium difficile and IBD treatment . Patients with most severe colitis with worse outcomes . Maintenance immunosuppression correlated with infection (purine analogs, methotrexate) . No association of anti-TNF therapy with C difficile. . Corticosteroids associated with 3 fold increase in developing C difficile – inhibition of humoral immune response to toxin A?

Issa M et al. Clin Gastroenterol Hepatol. 2007;5: 345-51. Ben-Horin S. et al. Clin Gastroenterol Hepatol. 2009; 9: 981-7. Schneeweis S, et al. Aliment Pharmacol Ther. 2010; 30: 253-64. Clostridium difficile and IBD . Patients with colitis are at increased risk . Maintenance immunosuppression correlated with infection (purine analogs, methotrexate) . 10% of cases were new IBD presentations . Contributes to flare in setting of new and longstanding disease in remission . Recommend multiple stool samples for ELISA toxin A, B analysis. 54% of patients detected on first stool sample. . No prompt response to metronidazole, consider vancomycin p.o.

Issa M et al. Clin Gastroenterol Hepatol. 2007;5: 345-51. III. Diagnostic considerations: C. difficile in IBD

. Laboratory - Leukocytosis - Hypoalbuminemia

. Radiographic

. Endoscopy - Pseudomembranes in 50% of patients with CDAD – rare in IBD patients.

. Leukocytosis, hypoalbuminemia, pseudomembranes are markers of severity Diagnosis of Clostridium difficile

. Cell culture toxin assay (cytotoxicity) is the gold standard. Excellent sensitivity. Requires 24 – 48 hrs; labor intensive and expensive. Performed in 15 hospitals in U.S. (total of 6,000 hospitals). . ELISA for toxin A and/or B. More rapid , less expensive and requires less expertise. Sensitivity varies from 79% to 97%. Performed in >90% of US hospitals at this time. . PCR analysis of stool samples. Sensitivity in IBD population? . Stool culture. Important for strain identification. Too slow for routine clinical use. (A) Negative cell culture cytotoxicity assay. Human fibroblasts remain spindle-shaped and in contact with each other

(B) Positive cell culture cytotoxicity assay revealing cytopathic effects of C difficile toxin B causing cell rounding and separation. (Courtesy of Ray Hariri, PhD) Stool ELISA testing in IBD patients for C. Difficile toxins A and B

100 4 stool samples to reach 90 90% detection with ELISA 80 70 60 Positive C. confirmed 50 difficile 40 ELISA 30

C. difficileC. 20 10 0 1st 2nd 3rd 4th

Stool ELISA sample

Issa M et al. Clin Gastro Hep. 2007 Mar;5(3):345-51. Special IBD scenarios with C. difficile

C. difficile in ileo-anal C. difficile in segments of Pouchitis diverted bowel

. Chronic refractory . One case report of C. pouchitis difficile in UC pt . Unresponsive to broad following subtotal spectrum antibiotics colectomy with end- . C. difficile developed ileostomy. while patients were on . Treated successfully metronidazole therapy with 10 day course of . 19% of pouch patients metronidazole with C. difficile suppositories.

Mann, S.D., et al. Dis Colon Rectum, 2003. 46(2): p. 267-70. Tsironi, E., et al. Dis Colon Rectum, 2006. 49(7): p. 1074-7. Shen, B., et al. Dig Dis Sci, 2006. 51(12): p. 2361-4. Shen B, et. al. Clin Gastroenterol Hepatol 2008;6:782-8 C. difficile enteritis: An early complication in IBD patients following colectomy

. Rare but associated with significant morbidity with mortality rates ranging from 60-83% . MCW institutional series of six patients (2004-2006). C. difficile manifested in high volume watery ileostomy output, ileus and fever with leukocytosis. No mortality with prompt diagnosis and therapy.

Miller, D.L et al. Arch Surg, 1989. 124(9): p. 1082. Jacobs, A., et al. review of the literature. Medicine (Baltimore), 2001. 80(2): p. 88-101. Hayetian, F.D., et al. Arch Surg, 2006. 141(1): p. 97-9. Lundeen et al. J Gastroentest Surg (2007) 11:138-142 IV. Therapeutic considerations: C. difficile in IBD Approach for hospitalized IBD patients with Suspected/confirmed C. difficile

. C. difficile isolation and contact precautions.

. Daily stool testing for C. difficile (until positive sample). Possibility for in-hospital acquisition.

. Empiric oral vancomycin from day 1, alone or in combination with metronidazole (IV or po). . Maintain oral diet!

. Decrease corticosteroid dosing – steroids blunt humoral immunity and IgG response to toxin A is necessary to resolve CDAD. Oral vancomycin vs metronidazole for C. difficile

FDA-approved Colonic levels Effectivity Mild Severe Promotion of VRE Failure rate Relapse rate Side effects

Response(median time) Zar, F.A., et al.. Clin Infect Dis, 2007. 45(3): p. 302-7. LucasCost GM et al. Clin Infect Dis. 1998 May ;26 (5):1127-339 Aslam et al. Lancet Infect Dis 2005;5:549-557. Wilcox, M.H. and R. Howe, J Antimicrob Chemother, 1995. 36(4): p. 673-9. Oral vancomycin for C. difficile

. Only FDA approved drug for the treatment of C. difficile . Tablets of vancomycin – shortages in past, high cost . Patent for vancomycin tablets will expire at the end of 2010 . Parenteral (intravenous formulation) vancomycin for oral use . Decreased cost – involves hospital pharmacy formulation . Palatability can be improved mouthwash “chaser” Apple juice “chaser” . Parenteral vancomycin for enema formulation Decreasing colectomy rate among hospitalized IBD patients with C. difficile

Number of infections and rate of hospitalization remained constant, but Significant decrease in colectomy rate. - High index of suspicion - Use of oral vancomycin - Decreased corticosteroid dosing

50

40

30

Number of Cases Number Cases of 20

10

0 1999 2000 2001 2002 2003 2004 2005 2006

Issa M et al. Clin Gastro Hep. 2007 Mar;5(3):345-51. Preventive strategies - C. difficile

. Prophylaxis - Limit exposure to antibiotics - MacFarland et al. Probiotics (Saccharomyces boulardii, Lactobacillus rhamnosus GG, and probiotic mixtures) effective for the prevention of CDAD (OR 0.59). Data was strongest with S. boulardii - Environmental decontamination requires 10% sodium hypochlorite solutions. - Alcohol based hand gels are in-effective against spore- forming organisms. Soap and water dislodges spores from skin.

McFarland, L.V. Am J Gastroenterol, 2006. 101(4): p. 812-22. Mayfield, J.L., et al. Clin Infect Dis, 2000. 31(4): p. 995-1000. Wilcox, M.H., et al.J Hosp Infect, 2003. 54(2): p. 109-14. Leischner, J., et al., American Society for Microbiology, 2005. ((abstract # LB 29)). Refractory and recurrent C. difficile - I

Refractory C. difficile: - Intravenous immunoglobulin was used in a series of 14 patients (200 mg/kg). 64% responded. One patient required 2nd dose. - Consideration for hypogammaglobulinemia associated IBD.

Recurrent C. difficile: - 59% of IBD patients (27 out of 46) had a recurrence. Of the recurring patients, one-quarter required colectomy. C. difficile treatment regimens used: 1- Prolonged courses of vancomycin with or without pulse dosing (2 months) 2- Initial course of vancomycin followed by rifaximin maintenance course.

McPherson, S., et al. Dis Colon Rectum, 2006. 49(5): p. 640-5. Issa, M., et al. Am J Gastro, 2006. 101(9): p. S469. Johnson, S., et al. Clin Infect Dis, 2007. 44(6): p. 846-8. C. difficile Experimental Therapies Stool transplantation: - Following vancomycin course, PEG colonic prep, blended and filtered donor stool is introduced via nasal-duodenal tube or via colonoscopy. Loop ileostomy – vancomycin flush (UPMC Protocol): - Loop ileostomy is created in severely ill patients facing colectomy and vancomycin is perfused into the efferent limb. Monoclonal antibodies against C. difficile toxin A and B:

- 200 non-IBD patients randomized to receive either monoclonal antibody against CDA1 and CDA2 or placebo. Recurrence rates of C. difficile were 7% vs 25% (p<0.001). C. difficile toxin vaccine: - Inactivated C difficile toxoids A and B were administered to 3 patients with recurrent disease, with no relapse.

Aas J., et al.Clin Infect Dis, 2003. 36: p.580-585. Lowy I., et al. N Engl J Med, 2010. 362: 197-205. Sougioultzis S, et al. Gastroenterology 2005. 128; 764-70. . C. difficile has doubled in North American Medical Centers in the past 5 years. . IBD colitis patients have been affected at highest rate. . C. difficile in IBD is associated with high rates of hospitalization and colectomy and increased mortality. . Antibiotic use may not be required to precipitate infection. . Endoscopic and Histologic appearance is frequently not classical – pseudomembranes not always present. . Multiple stool ELISA samples for toxin analysis are required to make a diagnosis. . Metronidazole failure rate is 50%; Oral vancomycin may be superior in hospitalized patients. . C. difficile enteritis may occur in post-colectomy patients and patients with ileoanal reconstruction.

. C. difficile recurrence rates are high. . Early surgical consultation for patients developing severe disease (>10 BM/day, WBC>20K, severe abdominal pain, ileus).

. Hand washing with soap and water is essential to prevent nosocomial transmission.