One-Year Prevalence of Candida dublinienis in a Dutch University Hospital

With reference to a recently published article by Polacheck et al. (9) reporting the isolation of Candida dubliniensis from non-human immunodeficiency virus (HIV)-infected patients in an Israeli hospital, we would like to add our data from a single center in The Netherlands. Since we described the occurrence of candidemia due to C. dubliniensis in non-HIV-infected patients (4), we prospectively studied the prevalence of C. dubliniensis during the year 1999 in our 1,000-bed tertiary-care university hospital. All germ-tube-positive yeasts isolated from various clinical specimens submitted to the diagnostic laboratory were further examined. Germ-tube-positive yeasts were suspected to be C. dubliniensis when there was no growth at 45°C, no assimilation of xylose, and no elaboration of β-glucosidase and when microscopic morphology showed abundant chlamydospore formation on rice-cream agar (8, 10, 12). Molecular typing with RP02 (4, 5) was used to confirm the phenotypic identification as C. dubliniensis. In 1999, a total of 3,848 yeast isolates were cultured from various clinical samples of which 2,605 (67.7%) were germ-tube positive. Of the germ-tube-positive yeasts, 21 (0.8%) isolated from 11 patients were C. dubliniensis as determined by phenotypic and molecular methods. Most isolates were recovered from oropharyngeal and fecal surveillance cultures from patients at the hematological ward (18 of 21 patients; 86%); one isolate each was cultured from a bronchial secretion, a urine specimen, and a skin biopsy specimen. Of the 11 patients, 6 (55%) were HIV negative, and of 5 patients, no antibodies against HIV were tested because there was no clinical suspicion for HIV-related disease. In vitro susceptibility testing was performed on one isolate per patient using the NCCLS broth microdilution method (6). All isolates were susceptible to amphotericin B (MIC, 0.25 mg/liter), fluconazole (MIC, 0.25 mg/liter), 5-flucytosine (MIC, <0.25 mg/liter), and itraconazole (MIC, 0.012 mg/liter). Only the patient with the positive skin biopsy specimen, without concomitant positive blood cultures, was treated with fluconazole (400 mg daily) and recovered. Yeast infections are becoming a world-wide problem, with C. albicans still the most frequently isolated species of nosocomial infections (3, 7, 11, 12). In The Netherlands, 60% of all candidemic episodes were caused by C. albicans (13). C. dubliniensis, which shares many phenotypic characteristics with C. albicans, was first described only 5 years ago for HIV-positive subjects in Ireland (12). Since the methods for correct identification of C. dubliniensis were established, the species has been reported world-wide (1, 2, 9, 10). In a recent prospective study (2) in the United States, 6 of 699 yeasts (0.9%) isolated were C. dubliniensis, a prevalence which is similar to ours. In contrast with our results, these isolates were mainly recovered from the oral cavity of HIV-positive patients. In 1999, three cases of C. dubliniensis candidemia in HIV-negative subjects were reported in Europe (4), and recently, another four cases, of which one was HIV positive, were reported in the United States (1). Although C. dubliniensis was first described in HIV-positive patients, recent data suggest that it is also common in HIV-negative patients as long as attempts are made to differentiate the isolates from C. albicans. Microbiologists should be alerted to consider C. dubliniensis when an atypical germ-tube-positive yeast is isolated because the most serious infections (six of seven) described until now have occurred in HIV-negative subjects with different forms of immunosuppression.