Invasive candidiasis treated in the intensive care unit: Observations from a randomized clinical trial

Abstract

Objectives The objectives of this study were to contrast risk factors, microbiology, and outcomes in patients with invasive candidiasis treated in an intensive care unit (ICU) with those in patients with invasive candidiasis treated outside an ICU and to describe therapeutic results with caspofungin in ICU patients. Materials and Methods We retrospectively identified patients with documented invasive candidiasis who received their first dose of the study drug in the ICU as part of a double-blind randomized trial. Participants were not stratified at entry by their ICU status. Patients received caspofungin (50 mg/d after a 70-mg loading dose) or conventional amphotericin B (0.6-1.0 mg/kg per day) for 10 to 14 days. A favorable response required resolution of signs and symptoms as well as eradication of Candida pathogens. Results Of the 224 patients, 97 (43%) received their first dose of the study drug in the ICU. Most patients had well-recognized risk factors for invasive candidiasis, including broad-spectrum antibiotics, central venous catheters, and hyperalimentation. Recent surgery was more common whereas malignancy, neutropenia, and immunosuppression were less common among ICU patients than among non-ICU patients. Candidemia was demonstrated in 81% of ICU patients and in 84% of non-ICU patients. Favorable response rates in the ICU patients vs the non-ICU patients were 68% (95% confidence interval [CI] = 53%, 82%) vs 77% (95% CI = 67%, 87%) for caspofungin and 56% (95% CI = 43%, 69%) vs 67% (95% CI = 55%, 79%) for amphotericin B. After accounting for differences in APACHE (Acute Physiology and Chronic Health Evaluation) II score, neutropenia status, and geographic region, we found that patients initiating the study therapy in an ICU were still more likely to die than patients initiating study therapy outside an ICU. For ICU patients, all-cause mortality rates were 45% (95% CI = 30%, 60%) for caspofungin recipients and 40% (95% CI = 28%, 53%) for amphotericin B recipients, whereas candidiasis-attributable mortality rates were 5% (95% CI = 0%, 12%) for caspofungin recipients and 11% (95% CI = 3%, 19%) for amphotericin B recipients. Overall, drug-related adverse events were reported less often among the ICU patients than among the non-ICU patients. Conclusions In ICU patients treated with antifungal therapy, invasive candidiasis is associated with substantial mortality, but most deaths cannot be directly attributed to this infection.