The aim of this study is to evaluate the implementation of a new technique for invasive candidiasis diagnosis by PCR. The main objective will be to evaluate the effect of this strategy on the time required to obtain results in patients from intensive care units with suspected systemic invasive candidiasis. This PCR method could allow an antifungal switch to be made earlier, limiting the appearance of resistance and reducing the cost of antifungals for healthcare facilities.

Invasive candidiasis is a serious infection caused by yeasts of the genus Candida sp. Their incidence is increasing, particularly in intensive care. The prognosis is severe and depends on the early initiation of adequate antifungal treatment. At present, the diagnosis of invasive candidiasis is difficult since blood culture, the reference technique, has major limitations due to its low sensitivity, as well as very long delays in obtaining results. As a consequence, suspected invasive candidiasis is often treated empirically with echinocandins, resulting in the development of resistance and extremely high costs for health care facilities. The switch to treatments such as fluconazole is often delayed, although this molecule is effective on most Candida species except C. krusei and glabrata and is much less expensive than echinocandins.

For these reasons, the present study developed faster and more efficient methods to diagnose Candida krusei and glabrata species, based on the detection of DNA in the blood by PCR, allowing an earlier switch from echinocandins to Fluconazole.

The patients included will be divided into 2 groups: one group of patients will benefit from a diagnostic strategy based on blood culture (reference technique for the detection of Candida) and another group of patients will benefit from a diagnostic strategy based on this new PCR technique. The main objective will be to evaluate the effect of the PCR strategy on the time to obtain results and on the switch of antifungal agents. After inclusion, patients will be followed up at day 14, day 28 until discharge (or 3 months if the patient is still hospitalised).

Inclusion Criteria:

• Patient affiliated or benefiting from a social protection system
• Patient (or his/her trusted person) having given free and informed written consent to the investigating physician
• Non-neutropenic patient hospitalised in the Toulouse University Hospital intensive care unit.
• Patient with suspected invasive candidiasis for whom a blood culture for fungal infection diagnosis was taken and for whom an empirical treatment with echinocandin at an adapted dose was implemented.

Exclusion Criteria:

• Patients under guardianship or trusteeship, persons under court protection
• Pregnant or breastfeeding patients.