Invasive Candida infections are burdened with a high mortality rate and is very common in intensive care units. This study aims to evaluate the efficacy of empirical treatment with micafungin in adult patients with suspected invasive candidiasis.

Multicenter, randomized, double-blind parallel groups comparing adult patients with suspected invasive candidiasis input from a 14-day empirical treatment with micafungin (MYCAMINE 100 mg) with placebo on survival without invasive candidiasis in 28 days after initiation of study treatment.

Inclusion Criteria:

• Age ≥ 18 years
• Persistent sepsis without documented invasive candidiasis:

systemic inflammatory response syndrome (SIRS) presence of two signs on the 4 [temperature <36 ° C or> 38 ° C, heart rate> 90/min, respiratory rate> 20/min or PaCO2 <32 mmHg, leukocytosis> 12,000 / mm3, <4.000/mm3 or presence of circulating immature forms (> 10% of cells)] mechanical ventilation (intubation or tracheostomy) for over 4 days (96 hours) central line use of broad-spectrum antibacterial for more than 4 calendar days (96 hours) in the previous week presence of at least one extra-digestive site colonized by Candida sp. (Urine, mouth, throat, upper and lower respiratory tract, skin folds, and suction drains after surgery ...), not lower digestive tract, are not taken into account the positive samples of rectal swabs and / or stool cultures, absence of proven bacterial infections untreated no evidence of invasive fungal infections (positive blood culture, positive culture of a surgical site, deep biopsy with fungal) infection or mold according to the criteria of the group "fungal infection of the EORTC" organ failure

• Hospitalization in intensive care for over 5 days (120 hours)
• Giving a free, informed and in writing. In the absence of the person of trust or a family member (if present)consent to emergency possible.
• Receiving a social security system,
• Negative pregnancy test for patients of childbearing age

Exclusion Criteria:

• Proven invasive fungal infection (positive blood culture, positive culture of a surgical site, deep biopsy with fungal infection), including aspergillosis requiring antifungal therapy at the time of randomization
• Prognosis of less than 48 hours (for which the patient outcome will be fatal whatever treatment),
• Echinocandin antifungal treatment by more than one day or another antifungal for over 72 hours in the week before the inclusion visit,
• Allergy, hypersensitivity or known intolerance to echinocandins antifungal or any of the excipients of the drug
• Neutropenia (ANC <500/mm3)
• History of organ and bone marrow,
• Recent chemotherapy (less than 6 months)
• Systemic immunosuppressive therapy in progress, other than with corticosteroids at doses below 2 mg / kg / day of prednisolone or equivalent
• Participation in another interventional study in the same ICU stay or making treatment being evaluated within 28 days prior to randomization
• Any clinical investigator deems incompatible with the conduct of the study in acceptable security conditions
• Pregnant and lactating women,
• Adults subject to a legal protection measure
• Persons deprived of their liberty by a judicial or administrative decision, those hospitalized without consent, persons admitted to a health facility or social purposes other than research