Purpose: To determine whether nosocomial candidemia is associated with increased mortality in intensive care unit (ICU) patients.
Subjects and methods: We performed a retrospective (1992 to 2000) cohort study of 73 ICU patients with candidemia and 146 matched controls. Controls were matched based on disease severity as measured by the Acute Physiology and Chronic Health Evaluation (APACHE) II score (+/- 1 point), diagnostic category, and length of ICU stay before onset of candidemia.
Results: In comparison with the control group, patients with candidemia developed more acute respiratory failure (97% [n = 71] vs. 88% [n = 129], P = 0.03) during their ICU stay. They were mechanically ventilated for a longer period (29 +/- 26 days vs. 19 +/- 19 days, P<0.01) and had a longer stay in the ICU (36 +/- 33 days vs. 25 +/- 23 days, P = 0.02) as well as in the hospital (77 +/- 81 days vs. 64 +/- 69 days, P = 0.04). There was no difference in in-hospital mortality between the groups (48% [n = 35] vs. 43% [n = 62], P = 0.44), a difference of 5% (95% confidence interval [CI]: -8% to 19%). In a multivariate analysis, older age (hazard ratio [HR] = 1.13 per 10 years; 95% CI: 1.04 to 1.23; P = 0.004), acute renal failure (HR = 1.4; 95% CI: 1.1 to 2.0; P = 0.02), and unfavorable APACHE II scores (HR = 1.10 per 5 points; 95% CI: 1.00 to 1.20; P = 0.05) were independent predictors of mortality. Candidemia was not associated with mortality in a model that adjusted for these factors (HR = 0.9; 95% CI: 0.7 to 1.2; P = 0.53).
Conclusion: Nosocomial candidemia does not adversely affect the outcome in ICU patients in whom mortality is attributable to age, the severity of underlying disease, and acute illness.