Investigation of critical care unit utilization and mortality in patients infected with Clostridium difficile

Gasperino, James; Garala, Maya; Cohen, Hillel W.; Kvetan, Vladimir; Currie, Brian

doi:10.1016/j.jcrc.2009.04.002

« Previous Next »Journal of Critical Care
Volume 25, Issue 2 , Pages 282-286, June 2010

Investigation of critical care unit utilization and mortality in patients infected with Clostridium difficile

James Gasperino, MD, PhD
- Division of Critical Care, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY 10467, USA
- Corresponding author.
Maya Garala, PharmD
- Department of Pharmacy, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY 10467, USA
Hillel W. Cohen, MD, MPH
- Department of Epidemiology and Population Health, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY 10467, USA
Vladimir Kvetan, MD
- Division of Critical Care, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY 10467, USA
Brian Currie, MD, MPH
- Division of Infectious Diseases, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY 10467, USA
- Foundation for Applied Epidemiology, Northvale, NJ 07647, USA

published online 10 July 2009.

Abstract

Background

A nationwide increase in the rate and severity of Clostridium difficile–associated disease may reflect infection with a virulent strain characterized by polymerase chain reaction as ribotype 027 (NAP1/B1).

Hypothesis

The high prevalence of ribotype 027 at our institution would allow investigation of the risk of mortality and admission to the intensive care unit (ICU) associated with C difficile infection.

Methods

In a retrospective cohort study, we identified 108 patients with positive enzyme-linked immunosorbant assay tests for C difficile toxins over a 6-month period and compared them to 108 patients who were suspected to have C difficile but with negative toxin assays. Proportions of all-cause mortality and ICU admission were compared using χ², and odds ratios (ORs) were estimated using logistic regression to adjust for potential confounders. Mean log lengths of stay were compared using t test.

Results

Comparing patients with C difficile to those without, mortality (20% vs 8%) and ICU admission (32% vs 17%) were significantly higher (P = .02 for both), whereas log length of stay was not (P = .29). Adjusting for potential confounders, the OR for mortality was 6.8 (95% confidence interval, 1.8-25.4; P = .01), whereas for ICU admission, the association was no longer observed (OR, 1.0; 95% confidence interval, 0.4-2.5; P = .97).

Conclusion

C difficile infection was associated with increased all-cause mortality. An observed association with ICU admission and C difficile infection was identified through univariate analysis but was not significant in multivariate analysis. Although we did not strain-type isolates for patients infected with C difficile, the institutional prevalence of ribotype 027 C difficile infection was known to be high. These results document a strong association between ribotype 027 C difficile infection and mortality and underscore the need to identify effective C difficile preventive strategies.

Keywords: C difficile, Mortality, Intensive care