Institutional members access full text with Ovid®

Share this article on:

Candida bloodstream infections in intensive care units: Analysis of the extended prevalence of infection in intensive care unit study*

Kett, Daniel H. MD; Azoulay, Elie MD, PhD; Echeverria, Pablo M. MD; Vincent, Jean-Louis MD, PhD, FCCM

doi: 10.1097/CCM.0b013e318206c1ca
Clinical Investigations

Objectives: To provide a global, up-to-date picture of the prevalence, treatment, and outcomes of Candida bloodstream infections in intensive care unit patients and compare Candida with bacterial bloodstream infection.

Design: A retrospective analysis of the Extended Prevalence of Infection in the ICU Study (EPIC II). Demographic, physiological, infection-related and therapeutic data were collected. Patients were grouped as having Candida, Gram-positive, Gram-negative, and combined Candida/bacterial bloodstream infection. Outcome data were assessed at intensive care unit and hospital discharge.

Setting: EPIC II included 1265 intensive care units in 76 countries.

Patients: Patients in participating intensive care units on study day.

Interventions: None.

Measurement and Main Results: Of the 14,414 patients in EPIC II, 99 patients had Candida bloodstream infections for a prevalence of 6.9 per 1000 patients. Sixty-one patients had candidemia alone and 38 patients had combined bloodstream infections. Candida albicans (n = 70) was the predominant species. Primary therapy included monotherapy with fluconazole (n = 39), caspofungin (n = 16), and a polyene-based product (n = 12). Combination therapy was infrequently used (n = 10). Compared with patients with Gram-positive (n = 420) and Gram-negative (n = 264) bloodstream infections, patients with candidemia were more likely to have solid tumors (p < .05) and appeared to have been in an intensive care unit longer (14 days [range, 5–25 days], 8 days [range, 3–20 days], and 10 days [range, 2–23 days], respectively), but this difference was not statistically significant. Severity of illness and organ dysfunction scores were similar between groups. Patients with Candida bloodstream infections, compared with patients with Gram-positive and Gram-negative bloodstream infections, had the greatest crude intensive care unit mortality rates (42.6%, 25.3%, and 29.1%, respectively) and longer intensive care unit lengths of stay (median [interquartile range]) (33 days [18–44], 20 days [9–43], and 21 days [8–46], respectively); however, these differences were not statistically significant.

Conclusion: Candidemia remains a significant problem in intensive care units patients. In the EPIC II population, Candida albicans was the most common organism and fluconazole remained the predominant antifungal agent used. Candida bloodstream infections are associated with high intensive care unit and hospital mortality rates and resource use.

From the Department of Medicine (DHK, PME), Division of Pulmonary and Critical Care, The University of Miami Miller School of Medicine and Jackson Memorial Hospital, Miami, FL; Service de Réanimation Médicale (EA), Groupe de Recherche Respiratoire en Réanimation Onco-Hématologique, Groupe de Recherche Famiré a Hôpital Saint-Louis, Université Paris VII, Paris, France; and the Department of Intensive Care (J-LV), Erasme University Hospital, Brussels, Belgium.

Presented in part at the Society of Critical Care Medicine's 39th Critical Care Congress, Miami Beach, FL, January 2010.

Supplementary digital content is available for this article. A list of Extended Prevalence of Infection in the Intensive Care Unit Study (EPIC II) Investigators can be viewed at

Dr. Kett consulted for Pfizer and Astellas and received honoraria/speaking fees from Pfizer, Astellas, GlaxoSmithKline, and Covidien. Dr. Azoulay consulted for Pfizer and Gilead, and received honoraria/speaking fees and a grant from Pfizer. The remaining authors have not disclosed any potential conflicts of interest.

For information regarding this article, E-mail:

© 2011 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins