In patients with invasive candidiasis, early treatment with antifungals reduces the mortality associated with this pathology, which remains high. Given the lack of precision of early diagnostic tests, antifungal therapy is established empirically in patients with sepsis in serious condition, as long as they have risk factors for invasive fungal infections and multiple infections. Sites of fungal colonization. Although this strategy, although recommended by recent IDSA guidelines in their 2016 update, has never been proven effective in critically ill patients outside of any neutropenia or transplantation setting. member. A blood test measuring B-D-glucans, a cell wall component of many fungi, including Candida-type yeasts, has been approved by the FDA as a complementary diagnostic tool for cultures to detect fungal infections. This test, however, lacks specificity in resuscitation patients with multiple organ failures and frequently suffering from bacterial infections.
Finally, the empirical treatments represent two-thirds of the antifungals used in intensive care, and the lack of a demonstrated benefit must be weighed against the costs of the empirical treatment, its toxicity and the associated risks of emergence of resistant strains.

The added value of this study :
In this multicentre, double-blind randomized trial, 260 patients with sepsis, non-transplanted, non-neutropenic, mechanically ventilated with organ failure and multiple Candida colonizations were selected from 19 ICUs. We compared micafungin (100mg / day) for 14 days with placebo. Empiric micafungin did not influence survival without invasive fungal infection at D28. This result remains the same in the subgroup of patients with a positive B-D-glucan serum. However, micafungin significantly reduces the incidence of nosocomial fungal infection.