1376. Risk factors for complications of Clostridium difficile Infection in a prospective multicentre cohort
Session: Poster Abstract Session: Clostridium difficile
Saturday, October 5, 2013
Room: The Moscone Center: Poster Hall C
Background:  Clostridium difficile infection (CDI) is the most common cause of nosocomial diarrhea and often results in severe complications including sepsis, colectomy and death. We conducted a prospective study to identify risk factors for complications or death due to CDI.

Methods:  Adult inpatients with confirmed CDI in 10 acute care hospitals across Quebec and Ontario, Canada, were enrolled in a prospective cohort between 2005 and 2008. A follow-up was performed on day 30 and 90 after enrollment. CDI was defined by diarrhea and a positive toxin (EIA or direct cytotoxin). Complicated CDI (cCDI) was confirmed if one or more of the following was observed: colonic perforation, toxic megacolon, septic shock, colectomy or death within 30 days. Data on potential factors predictive of cCDI were collected within 48 hours of the diagnosis. Isolates were typed by PCR-ribotyping. A multivariate logistic regression model was used to identify predictors of cCDI.

Results: A total of 1380 patients were enrolled (median age: 71 years; IQR= 58-80), among whom 86% were experiencing a first episode of CDI, 11% were long-term care residents and only 5% had community-acquired CDI. A stool specimen was collected in 1153 patients, a ribotype was obtained in 807 (70%) among them and 420 (52%) were ribotype 027. A cCDI was observed in 210 patients (16%), from which 169 (81%) died within 30 days and 20% had septic shock. Risk factors predictive of cCDI in the multivariate model were:  age (OR per year = 1.02; 95% CI = 1.002-1.03), tube feeding (OR = 2.7; 95% CI = 1.6- 4.5), white cell count >= 20 x 109C/mL (OR = 3.7; 95%CI = 2.1- 6.2), serum creatinine >=150 mmol/dL (OR = 3.02; 95% CI = 2.0- 4.6) and serum albumin ≤ 35 g/L (OR = 5.8; 95% CI = 2.5- 13.6). Surgery in the two months prior CDI was protective (OR = 0.52; 95% CI = 0.35-0.7). Comorbidities, immunosuppression or ribotype 027 were not significantly associated with cCDI.

Conclusion:  In a prospective multicenter cohort study we identified age, tube feeding, leukocytosis, increased serum creatinine, and hypoalbuminemia as independent predictors of cCDI. Ribotype 027 was not associated with cCDI and recent surgery appeared protective.

Claire Nour Abou Chakra, MSc1, Annie-Claude Labbé, MD, FRCPC2, Allison McGeer, MD, MSc, FRCPC, FSHEA3, Andrew E. Simor, MD, FRCPC, FACP4, Wayne Gold, MD5, Roslyn Devlin6, Kevin Katz, MD CM, MSc7, Jeff Powis, MD, MSc, FRCPC8, Louis-Charles Fortier, PhD9, Matthew P. Muller, MD, PhD, FRCPC10, Julian R. Garneau, BSc11, Jacques Pepin, MD1 and Louis Valiquette, MD, MSc1, (1)Department of Microbiology and Infectious Diseases, Université de Sherbrooke, Sherbrooke, QC, Canada, (2)Hôpital Maisonneuve-Rosemont, Montréal, QC, Canada, (3)Department of Microbiology, University of Toronto, Toronto, ON, Canada, (4)Department of Microbiology, Sunnybrook Health Sciences Centre, Toronto, ON, Canada, (5)Toronto General Hospital, Toronto, ON, Canada, (6)St. Michael's Hospital, Toronto, ON, Canada, (7)Infection Prevention & Control, North York General Hospital, Toronto, ON, Canada, (8)Infectious Disease, Toronto East General Hospital, Toronto, ON, Canada, (9)Microbiology and Infectious Diseases, University of Sherbrooke, Sherbrooke, QC, Canada, (10)St.Michael's Hospital, Toronto, ON, Canada, (11)Microbiology and Infectious Disease, University of Sherbrooke, Sherbrooke, QC, Canada


C. N. Abou Chakra, None

A. C. Labbé, None

A. McGeer, Pfizer Canada: Grant Investigator, Research grant

A. E. Simor, Pfizer canada Inc: Investigator and Scientific Advisor, Research grant and Speaker honorarium

W. Gold, None

R. Devlin, None

K. Katz, None

J. Powis, None

L. C. Fortier, None

M. P. Muller, None

J. R. Garneau, None

J. Pepin, None

L. Valiquette, Pfizer: Collaborator, Investigator and Speaker's Bureau, Research grant and Speaker honorarium

Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.