Evaluation of advanced age as a risk factor for severe Clostridium difficile infection

Abstract

Background/purpose

Although advanced age has been associated with the incidence of Clostridium difficile infection (CDI), its relationship with disease severity remains inconclusive. The objective of this study was to evaluate risk factors, specifically advanced age, which may be associated with the acquisition of severe CDI.

Methods

A retrospective chart review at a Veterans Affairs Hospital was conducted on hospitalized veterans aged ≥ 18 years with a positive stool toxin assay for Clostridium difficile between May 2008 and September 2012 (n = 224). One hundred and sixty-one (72%) patients were in the mild–moderate infection group and 63 (28%) patients in the severe infection group. The primary outcome was to determine the effect of advanced age (≥70 years old) on the acquisition of severe CDI. The secondary outcome was to identify other potential risk factors for severe CDI. Disease severity was classified according to the criteria established in the 2010 Society for Healthcare Epidemiology of America/Infectious Disease Society of America practice guidelines for CDI. Demographic and disease-specific data were collected. A logistic regression model was used to identify characteristics predictive of disease severity.

Results

Our regression model found advanced age to be significantly associated with severe CDI (odds ratio 2.43, p ≤ 0.005, 95% confidence interval 1.31–4.50). A larger proportion of veterans were diagnosed with severe CDI in the intensive care unit (p = 0.004). In addition, multiple antibiotic use (≥3) and association with severe CDI was statistically significant (34% mild–moderate vs. 48% severe, p = 0.041). The univariate analyses did not reveal any other characteristics predictive of disease severity.

Conclusion

Advanced age was associated with severe CDI. A prospective evaluation is warranted to validate this finding. Efforts to identify patients at risk for severe CDI will be important as it may direct treatment and positively affect outcomes.