928. Relationship Between Continued Antibiotic Use and All-cause Mortality in Hospital-onset Clostridium difficile Associated Diarrhea
Session: Poster Abstract Session: Clostridium difficile Infections: Epidemiology and Diagnostics
Friday, October 9, 2015
Room: Poster Hall
  • C diff all cause mortality.pdf (586.4 kB)
  • Background: Hypervirulent Clostridium difficile (CD) strains are associated with increased disease severity and death. Other mortality indicators include advanced age, underlying comorbidities, leukocytosis, hypoalbuminemia and a rising creatinine. The objective of our study was to determine predictors of death in hospital acquired CD associated diarrhea (CDAD).

    Methods: We conducted a retrospective review of hospital acquired, hospital onset (HAHO) CDAD over a 10-year period (2003-2012) in an integrated healthcare network. HAHO was defined according to 2010 national guidelines. Severe CDAD was defined using the modified University of Illinois criteria. Clinical and administrative data were collected from electronic sources.  Logistic regression was used to measure associations between clinical parameters and all-cause mortality during hospitalization. Variables with P≤0.1 in univariate analysis were included in the multivariable model.

    Results: We identified 2,712 HAHO CDAD encounters in 18 hospitals (5.2/10,000 patient days); 223 (8.2%) of these encounters ended with death. Severe CDAD was identified in 639 (24%) encounters.  Nearly all patients (98%) received antibiotics prior to CDAD; however, patients who died had more antibiotic days of therapy prior to diagnosis of CDAD (median [interquartile range]; 11 [7-19] vs. 9.0 [5-16]). Death was more frequent in patients who continued to receive antibiotics other than CD treatment after diagnosis of CDAD (9.7% vs. 2.1%; p<0.001). Factors independently associated with mortality in the multivariable model include extreme APRDRG severity of illness (OR [95% CI]; 13.4 [7.6-23.4]), severe CDAD (1.6 [1.1-2.2]) and continued antibiotic use after diagnosis of CDAD (3.0 [1.6-5.8]). Treatment for CD was associated with reduced mortality (0.4 [0.3-0.6]). Antibiotic exposure prior to CDAD, age and leukocytosis were not associated with death in the multivariable model.

    Conclusion: Antibiotic use before and after diagnosis of hospital onset CDAD is common. In addition to extreme severity illness and severe classification of CD, continued use of non-CD antibiotics after CDAD diagnosis was associated with mortality. Limiting antibiotic exposure after diagnosis of CDAD is essential for managing patients infected with CD.

    Bert K. Lopansri, MD1,2, Rajesh Mehta, R.Ph, MS1, Edward Stenehjem, MD MSc1, Kristin Dascomb, MD, PhD1 and John Burke, MD, FIDSA, FSHEA1,3, (1)Clinical Epidemiology and Infectious Diseases, Intermountain Medical Center, Murray, UT, (2)Division of Infectious Diseases, University of Utah School of Medicine, Salt Lake City, UT, (3)Internal Medicine, University of Utah School of Medicine, Salt Lake CIty, UT


    B. K. Lopansri, Cubist, Inc: Investigator , Research support

    R. Mehta, None

    E. Stenehjem, None

    K. Dascomb, None

    J. Burke, None

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.