Search

Summary Fungaemia is an increasing nosocomial pathology. The ‘gold standard’ for detection of fungaemia is blood culture, but it is time-consuming and its sensitivity for early detection is low. On the other hand, yeasts present different antifungal sensitivity patterns to be quickly detected to allow an effective treatment. The aim of this study was to evaluate the diagnostic performances of PNA-FISH to directly identify yeasts from blood cultures and to compare results with those obtained by culture. A total of 176 blood cultures positive for yeasts at direct Gram stain and 24 negative blood cultures as control collected from 15 Italian hospitals, included in a network coordinated by the Medical Mycology Committee, Italian Society of Clinical Microbiology (AMCLI), were examined both by culture and PNA-FISH technology. Sensitivity of the PNA-FISH technique evaluated for five Candida species was 99.3% and specificity, 100%. Distinguishing which yeast is implicated in fungaemia and whether the infection is caused by multiple species are important for the selection of antifungal therapy. The PNA-FISH technique is a very useful approach because the test discriminates between groups of Candida species with different susceptibility pattern, particularly against azoles and echinocandins, with only a 90-minute turn-around time after the Gram-stain reading.

A recently developed microdilution method (Sensititre) YeastOne) may represent a valid alternative to the National Committee for Clinical Laboratory Standards (NCCLS) method for routine testing. The Medical Mycology Committee of the Associazione Microbiologi Clinici Italiani (AMCLI) decided to evaluate its reproducibility and reliability compared with the NCCLS M27A protocol and the E-test. Nineteen strains each of Candida albicans and Ca. parapsilosis, isolated from systemic infections, were tested against amphotericin B, flucytosine, ketoconazole, itraconazole, and fluconazole. All the participating laboratories tested the YeastOne panels, while the E-test and the NCCLS method were performed by two laboratories each. Interlaboratory reproducibility showed a good correlation (from 95% for amphotericin B to 92.5% for flucytosine). The agreement between NCCLS and YeastOne ranged from 95 (ketoconazole and itraconazole) to 100% (amphotericin B and flucytosine), whereas the agreement between E-test and YeastOne ranged from 72.5 (fluconazole) to 100% (amphotericin B and flucytosine). The Sensititre YeastOne panels appear to be an excellent alternative to both the E-test and the NCCLS protocol for antifungal susceptibility testing.