Is rapid antifungal susceptibility testing directly from blood culture bottles concordant for Candida isolates?

Blood culture remains the gold standard for candidemia diagnosis but suffers from low sensitivity and long turnaround times, with diagnostic delays linked to increased mortality. This study evaluated the applicability of direct antifungal susceptibility testing from blood cultures. Clinical blood culture bottles collected as part of routine patient care when clinically indicated at a training and research hospital in Istanbul were included in the study (01.06.2022-31.12.2023). After a positive signal, yeast-positive samples on Gram staining were subcultured onto CHROMagar Candida and Sabouraud Dextrose Agar (SDA), then incubated at 35 °C for 24 h. Identification was performed using CHROMagar Candida and direct and standard germ tube tests (GTT) with Matrix-Assisted Laser Desorption/Ionization-Time of Flight Mass Spectrometry (MALDI-TOF MS) used as the reference method. Antifungal susceptibility to fluconazole, voriconazole, and caspofungin was determined by direct, standard disk diffusion (DD) and broth microdilution (BMD). Direct DD used yeast-positive blood on Mueller-Hinton agar with methylene blue (MB-MHA); standard DD and BMD followed Clinical and Laboratory Standards Institute (CLSI) M44-S2 and M27-A3, respectively. All the results were interpreted per CLSI M27M44S ED3 [1]. Among 64 Candida isolates, CHROMagar Candida identification showed 100% concordance with MALDI-TOF MS. Direct and standard GTT achieved accuracies of 90.62% and 96.87% (p > 0.05). Direct DD demonstrated good agreement with BMD for fluconazole (one minor error) and voriconazole (no errors), while caspofungin showed two minor and one major error. Direct DD enables a highly cost-effective and rapid preliminary susceptibility reporting for azoles but requires confirmation by standard methods, particularly for caspofungin.