Clostridium difficile in the intensive care unit: epidemiology, costs, and colonization pressure.

OBJECTIVE To evaluate the epidemiology, outcomes, and importance of Clostridium difficile colonization pressure (CCP) as a risk factor for C. difficile-associated disease (CDAD) acquisition in intensive care unit (ICU) patients. DESIGN Secondary analysis of data from a 30-month retrospective cohort study. SETTING A 19-bed medical ICU in a midwestern tertiary care referral center. PATIENTS Consecutive sample of adult patients with a length of stay of 24 hours or more between July 1, 1997, and December 31, 1999. RESULTS Seventy-six (4%) of 1,872 patients were identified with CDAD; 40 (53%) acquired CDAD in the ICU, for an incidence of 3.2 cases per 1,000 patient-days. Antimicrobial therapy, enteral feeding, mechanical ventilation, vancomycin-resistant enterococci (VRE) colonization or infection, and CCP (5.5 vs 2.0 CDAD case-days of exposure for patients with acquired CDAD vs no CDAD; P=.001) were associated with CDAD acquisition in the univariate analysis. Only VRE colonization or infection (45% of patients with acquired CDAD vs 16% of patients without CDAD; adjusted odds ratio, 2.76 [95% confidence interval, 1.36-5.59]) and a CCP of more than 30 case-days of exposure (20% with acquired CDAD vs 2% with no CDAD; adjusted odds ratio, 3.77 [95% confidence interval, 1.14-12.49]) remained statistically significant in the multivariable analysis. Lengths of stay (6.1 vs 3.0 days; P<.001 by univariate analysis) and ICU costs ($11,353 vs $6,028; P<.001 by univariate analysis) were higher for patients with any CDAD than for patients with no CDAD. CONCLUSIONS In this nonoutbreak setting, the CCP was an independent risk factor for acquisition of CDAD in the ICU at the upper range of exposure duration. Having CDAD in the ICU was a marker of excess healthcare use.