[Clinical characteristics and predictors of mortality in patients with candidemia in intensive care unit].

OBJECTIVE: To investigate the clinical characteristics and predictors of mortality in patients with candidemia in intensive care unit (ICU).

METHODS: The patients with candidemia admitted to ICU of Peking University People's Hospital from January 2010 to December 2017 were enrolled. The general clinical data, indicators related to Candidia infection and prognosis were collected, and the clinical characteristics, infection characteristics and prognosis of patients with candidiasis were analyzed. Patients were divided into death group and survival group according to hospital survival status. The differences of each index were compared between two groups. The independent risk factors of mortality in patients with candidemia were analyzed by multivariate Logistic regression analysis.

RESULTS: A total of 95 patients (55 males) with candidemia were included, with an average age of (69.3±16.5) years, acute physiology and chronic health evaluation II (APACHE II) was 24.7±3.6, sequential organ failure assessment (SOFA) was 6.6±2.7. Candida albicans accounted for the largest proportion of Candida infections (n = 56, 58.9%). Thirty-two (33.7%) patients received inadequate antifungal therapy and 38 (40.0%) patients received inadequate source control. Fifty-five (57.9%) patients were died in hospital. Compared with the survival group, patients in the death group was older (years: 72.5±14.6 vs. 64.9±18.0, P < 0.05), had higher APACHE II and SOFA scores (26.6±2.2 vs. 22.1±3.6, 7.9±2.0 vs. 4.7±2.4, both P < 0.01), higher rate of glucocorticoid treatment (18.2% vs. 10.0%, P < 0.05), and higher proportion of Candida albicans and Candida glabrata (69.1% vs. 45.0%, 10.9% vs. 7.5%, both P < 0.05), the rate of multi-site Candida infection also significantly increased (47.3% vs. 17.5%, P < 0.05). Intra-abdominal infection was the primary infection site and more common in death group (49.1% vs. 35.0%, P < 0.05). The rates of sepsis (87.3% vs. 62.5%), inadequate antifungal therapy (49.1% vs. 10.0%), inadequate source control (60.0% vs. 12.5%) in death group were all higher than those in survival group (all P < 0.01). It was shown by multivariate Logistic regression analysis that APACHE II [odds ratio (OR) = 1.605, P = 0.002, β = 0.473], SOFA (OR = 1.501, P = 0.029, β = 0.406), inadequate antifungal therapy (OR = 12.084, P = 0.006, β = 2.492) and inadequate source control (OR = 7.332, P = 0.024, β = 1.992) were independent risk factors for mortality in ICU patients with candidemia.

CONCLUSIONS: Candidemia patients were severe and had poor prognosis. APACHE II, SOFA, inadequate antifungal therapy and inadequate source control were independent risk factors of mortality.