Do hospital microbiology laboratories still need to distinguish Candida albicans from Candida dubliniensis?

As a clinical mycologist, one of the most frequent questions I get asked by clinical technicians is, “Do we still need to distinguish Candida dubliniensis from Candida albicans?” When I ask the questioner why they separate them, the usual response is that C. dubliniensis is often resistant to fluconazole. When C. dubliniensis was first described in 1995 (11), AIDS was epidemic and highly active antiretroviral therapy (HAART) was just becoming available. Because of these factors, mucocutaneous candidasis was commonly seen and fluconazole prophylaxis was often prescribed. Many of the first strains of C. dubliniensis from HIV-positive patients exhibited elevated fluconazole MIC values or the ability to develop fluconazole resistance under drug pressure in vitro (5, 6). Despite this, the original discoverers of C. dubliniensis reported that it remained fairly susceptible to fluconazole (9). To try to put this question to rest, we tested 42 isolates of C. dubliniensis for susceptibility to fluconazole. Isolates were collected as part of a population-based candidemia surveillance conducted by the CDC in Atlanta and Baltimore. Isolates were identified as C. dubliniensis by a Luminex-based assay (4) or by sequencing of the D1/D2 region of the ribosomal DNA (rDNA). Isolates were tested by broth microdilution using the methodology of Clinical and Laboratory Standards Institute standard M27-A3 (3). The MIC50 of these 42 isolates was 0.25 g/ml, and the MIC90 was 0.5 g/ml, with only two isolates (4.8%) having a MIC value above 0.5 g/ml, both of which had a MIC value of 16 g/ml. The two isolates with elevated MIC values would not recently have been considered resistant, but by applying the new C. albicans species-specific breakpoints (7) would now be considered resistant. Two other very recent studies had similar conclusions. In their 10-year global Candida surveillance, Pfaller and colleagues (8) found that only 2.6% (3.9% by new C. albicans species-specific breakpoints) of 310 C. dubliniensis isolates were resistant to fluconazole. In a 6-year national surveillance of fungemia in Denmark, Arendrup and colleagues (1) found only 3.1% fluconazole resistance among 65 clinical isolates of C. dubliniensis. This does not mean that there is never a reason to determine species. Certainly in an outbreak setting, it is important to know the species for epidemiological purposes, and it is important for surveillance trends. There are some commercially available systems that adequately identify C. dubliniensis, and these may be useful in centers with a high HIV-positive prevalence (2). However, for a recurrent isolate, it would be much more efficacious to perform susceptibility testing on the isolate than to identify it as C. dubliniensis, especially in light of the fact that C. albicans is considered to be the more virulent pathogen (10). The additional biochemical testing or molecular assay costs the clinical laboratory time and money and could slow the initiation of effective empirical therapy by the belief that fluconazole cannot be used for C. dubliniensis infections. Therefore, my answer to the question of whether it is necessary to distinguish C. dubliniensis from C. albicans in the clinical microbiology laboratory is usually, “No.”