CHROMagar dilution screening vs CLSI to detect fluconazole resistant Candida

Objectives: The Clinical and Laboratory Standards Institute (CLSI) reference method for broth dilution antifungal susceptibility testing of yeasts serves as the reference standard in susceptibility testing, yet recently, CLSI has published changes to the method in document M27-A3. Previously, we showed excellent correlation comparing CHROMagar agar dilution susceptibility screening against CLSI M27-A2 MIC results, normally, detecting over 90% of CLSI resistant isolates at 48 hours. Herein, we compare CHROMagar agar dilution antifungal susceptibility screening to the revised CLSI methodology, now read at 24hours. Methods: Oral rinse samples from HIV/AIDS patients were screened for Candida albicans and Candida glabrata with reduced susceptibility to fluconazole using agar dilution employing CHROMagar infused with 8 and 16 micrograms/ml of fluconazole. We examined yeasts with expected reduced susceptibility from agar dilution screening including 27 isolates of C. albicans that grew normally on plates containing 8 or 16 micrograms of fluconazole per ml and 32 isolates of C. glabrata that grew normally on plates with 16 micrograms of fluconazole per ml. Yeast screening on agar dilution was performed at 48 hours. Antifungal susceptibility testing against fluconazole was done according to current CLSI (M27-A3) methodology. Results: Agar dilution screening revealed 23/27 (85.2%) of the C. albicans isolates determined to have reduced susceptibility were found to be within 1 tube dilution of the revised CLSI method. Additionally, screening of C. glabrata specimens revealed that 24/32 (75%) isolates determined to have reduced susceptibility were found to be within 1 tube dilution of the CLSI method. Conclusion: Agar dilution screening of oral samples remains a valuable tool for detecting fluconazole resistance. Recent changes in the CLSI methodology have impacted the correlation unfavorably. Additional studies, using expanded panels of isolates need to be performed using the same 24 hour time point assessment. This work was partially supported by NIDCR Grant DE-18096.