Candida spp. bloodstream infection: influence of antifungal treatment on outcome.

OBJECTIVES: To assess the influence of new antifungal treatments on candidaemia outcome. METHODS: Candidaemia episodes prospectively collected through a blood culture surveillance programme in a single institution. The study was divided into two periods of time, 1994-2003 (A) and 2004-2008 (B), according to the introduction of echinocandin treatment. Non-conditional logistic regression methods with mortality as the dependent variable were used. RESULTS: Four hundred and thirty-three (3%) candidaemias out of 15 628 bloodstream infection episodes were analysed. Candida albicans was the most frequent species (211; 49%). Mortality was noted in 132 cases (30%). A total of 262 and 171 candidaemias were reported in period A and B, respectively. There were 94 deaths in period A (36%) and 38 in period B (22%, P = 0.03). Treatment in period A was amphotericin B in 89 patients (41 dead, 46%) and fluconazole in 151 (41 dead, 27%, P = 0.003). In period B, 113 patients received a triazole (26 dead, 23%), 30 an echinocandin (3 dead, 10%, P = 0.08) and 9 (0 dead) were treated with combined therapy (echinocandin and triazole). Mortality was higher in period A (94 dead, 36%) than in period B (38 dead, 27%), P = 0.03. Independent risk factors associated with mortality in period B were: age, chronic renal failure, ultimately or rapidly fatal prognosis of underlying disease and shock. Echinocandin alone or in combination therapy was associated with better outcome (odds ratio = 0.22, 95% confidence interval = 0.06-0.81, P = 0.02). CONCLUSIONS: In patients with candidaemia, echinocandin therapy results in a better outcome.