Empirical fluconazole versus placebo for intensive care unit patients: a randomized trial.

BACKGROUND: Invasive infection with Candida species is an important cause of morbidity and mortality in intensive care unit (ICU) patients. Optimal preventive strategies have not been clearly defined.

OBJECTIVE: To see whether empirical fluconazole improves clinical outcomes more than placebo in adult ICU patients at high risk for invasive candidiasis.

DESIGN: Double-blind, placebo-controlled, randomized trial conducted from 1995 to 2000.

SETTING: 26 ICUs in the United States.

PATIENTS: 270 adult ICU patients with fever despite administration of broad-spectrum antibiotics. All had central venous catheters and an Acute Physiology and Chronic Health Evaluation II score greater than 16.

INTERVENTION: Patients were randomly assigned to either intravenous fluconazole, 800 mg daily, or placebo for 2 weeks and were followed for 4 weeks thereafter. Two hundred forty-nine participants were available for outcome assessment.

MEASUREMENTS: A composite primary outcome that defined success as all 4 of the following: resolution of fever; absence of invasive fungal infection; no discontinuation because of toxicity; and no need for a nonstudy, systemic antifungal medication (as assessed by a blinded oversight committee).

RESULTS: Only 44 of 122 (36%) fluconazole recipients and 48 of 127 (38%) placebo recipients had a successful outcome (relative risk, 0.95 [95% CI, 0.69 to 1.32; P = 0.78]). The main reason for failure was lack of resolution of fever (51% for fluconazole and 57% for placebo). Documented invasive candidiasis occurred in 5% of fluconazole recipients and 9% of placebo recipients (relative risk, 0.57 [CI, 0.22 to 1.49]). Seven (5%) fluconazole recipients and 10 (7%) placebo recipients had adverse events resulting in discontinuation of the study drug. Discontinuation because of abnormal liver test results occurred in 3 (2%) fluconazole recipients and 5 (4%) placebo recipients.

LIMITATIONS: Twenty-one randomly assigned patients were not included in the analysis because they either did not meet entry criteria or did not have postbaseline assessments. Fewer fungal infections than anticipated occurred in the control group. Confidence bounds were wide and did not exclude potentially important differences in outcomes between groups.

CONCLUSION: In critically ill adults with risk factors for invasive candidiasis, empirical fluconazole did not clearly improve a composite outcome more than placebo.