Results from the ARTEMIS DISK Global Antifungal Surveillance Study: a 6.5-Year Analysis of Susceptibilities of Candida and Other Yeast Speci
http://www.100md.com
微生物临床杂志 2005年第12期
University of Iowa College of Medicine, Iowa City, Iowa
University of Texas Health Science Center, San Antonio, Texas
University of Wales College of Medicine, Cardiff, United Kingdom
Zhong Shan Hospital, Shanghai, China
Institute of Antimicrobial Chemotherapy, Smolensk, Russia
Hospital de Clinicas "Jose de San Martin," Buenos Aires, Argentina
Institute of Clinical Microbiology, Faculty of Medicine, University of Szeged, Szeged, Hungary
Giles Scientific, Inc., Santa Barbara, California
ABSTRACT
Fluconazole in vitro susceptibility test results for 140,767 yeasts were collected from 127 participating investigators in 39 countries from June 1997 through December 2003. Data were collected on 79,343 yeast isolates tested with voriconazole from 2001 through 2003. All investigators tested clinical yeast isolates by the CLSI (formerly NCCLS) M44-A disk diffusion method. Test plates were automatically read and results were recorded with the BIOMIC Vision Image Analysis System. Species, drug, zone diameter, susceptibility category, and quality control results were collected quarterly via e-mail for analysis. Duplicate (the same patient, same species, and same susceptible-resistant biotype profile during any 7-day period) and uncontrolled test results were not analyzed. The 10 most common species of yeasts all showed less resistance to voriconazole than to fluconazole. Candida krusei showed the largest difference, with over 70% resistance to fluconazole and less than 8% to voriconazole. All species of yeasts tested were more susceptible to voriconazole than to fluconazole, assuming proposed interpretive breakpoints of 17 mm (susceptible) and 13 mm (resistant) for voriconazole. MICs reported in this study were determined from the zone diameter in millimeters from the continuous agar gradient around each disk, which was calibrated with MICs determined from the standard CLSI M27-A2 broth dilution method by balanced-weight regression analysis. The results from this investigation demonstrate the broad spectrum of the azoles for most of the opportunistic yeast pathogens but also highlight several areas where resistance may be progressing and/or where previously rare species may be "emerging."
INTRODUCTION
Antifungal resistance surveillance with a focus on Candida is now widespread (5, 10, 17, 20, 29, 32). Most of these surveillance efforts are by necessity limited in terms of the numbers of participating sentinel sites and isolates tested. Furthermore, none of the programs is extensive enough to provide temporal and geographic data concerning the occurrence and resistance profiles of the less common Candida species and other, noncandidal opportunistic yeasts (21).
The ARTEMIS Global Antifungal Surveillance Program is among the most comprehensive and long-running fungal surveillance programs (6, 12, 17, 19, 22, 24, 25, 27). The ARTEMIS Program is made up of two components: (i) a broad international network of participating sites (127 sites in 39 countries), each of which performs Clinical and Laboratory Standards Institute (CLSI, formerly National Committee for Clinical and Laboratory Standards [NCCLS])-recommended disk diffusion testing (M44-A) (14) of fluconazole and voriconazole against consecutive yeast isolates from a variety of clinical sources (ARTEMIS DISK Surveillance Study) (6), and (ii) a central reference laboratory (University of Iowa, Iowa City), where CLSI-recommended broth microdilution (BMD) MIC and disk diffusion testing (M27-A2 and M44-A, respectively) (13, 14) is performed on blood and normally sterile-site isolates of Candida and other opportunistic yeasts and molds thatare referred according to protocol from the participating ARTEMIS study sites (19, 22, 24, 25, 27). As such, the ARTEMIS Program has been designed to address many of the potential limitations of resistance surveillance studies (7): (i) it is both longitudinal (1997 to present) and global (127 participating sites in 39 countries) in scope, (ii) it employs standardized antifungal susceptibility test methods (CLSI disk [M44-A] and BMD MIC [M27-A2]) (13, 14), (iii) both internal quality control (QC) performed in each participating laboratory and external quality assurance measures are used to validate test results (25, 27), (iv) results are recorded electronically using the BIOMIC image analysis plate reader system (Giles Scientific, Santa Barbara, Calif.) (6, 19, 25, 27) and are stored in a central database, and (v) both Candida and non-Candida yeast isolates obtained from consecutive clinical samples from all body sites are tested locally, thus avoiding misleading results based on biased selective testing. This so-called "routine" testing is augmented by testing of isolates from blood and normally sterile sites in the central reference laboratory (25, 27). Thus, the ARTEMIS Program generates massive amounts of data that have been externally validated and that can be used to identify temporal and geographic trends in the species distribution of Candida and other opportunistic yeasts, as well as the resistance profiles of these organisms to fluconazole and voriconazole as determined by standardized CLSI disk diffusion testing.
In the present study, we utilized the results from the ARTEMIS DISK Surveillance Program to evaluate global trends in the susceptibility of yeasts to fluconazole over a 6.5-year period (140,767 isolates from 127 study sites in 39 countries; June 1997 through December 2003). We also report results of voriconazole susceptibility testing performed on 79,343 isolates collected from 2001 to 2003. The scope of this study provides an unprecedented look at the occurrence and azole susceptibilities of several rare species of Candida, as well as several of the other opportunistic yeasts. The study is limited in that the numbers of isolates from certain regions are small and the time frame over which voriconazole data are available is relatively short.
MATERIALS AND METHODS
Organisms and test sites. A total of 134,715 isolates of Candida spp. and 6,052 isolates of noncandidal yeasts obtained from 127 different medical centers in Asia (23 sites), Latin America (16 sites), Europe (74 sites), the Middle East (2 sites), and North America (12 sites) were collected and tested against fluconazole between June 1997 and December 2003. In addition, a total of 79,343 isolates (75,810 isolates of Candida spp. and 3,533 other yeasts) from 115 study sites in 35 countries were tested against voriconazole between 2001 and 2003. All yeasts considered pathogens from all body sites (e.g., blood, normally sterile body fluids, deep tissue, genital tract, gastrointestinal tract, respiratory tract, skin, and soft tissue) and isolates from patients in all in-hospital locations during the study period were tested. Yeasts considered by the local site investigator to be colonizers, that is, not associated with an obvious pathology, were excluded, as were duplicate isolates from a given patient (the same species and the same susceptible-resistant biotype profile within any 7-day period). Identification of isolates was performed in accordance with each site's routine methods.
Susceptibility test method. Disk diffusion testing of fluconazole and voriconazole was performed as described by Hazen et al. (6) and in CLSI document M44-A (14). Agar plates (150-mm diameter) containing Mueller-Hinton agar (obtained locally at all sites) supplemented with 2% glucose and 0.5 μg of methylene blue per ml (MH-MB) at a depth of 4.0 mm were used. The agar surface was inoculated by using a swab dipped in a cell suspension adjusted to the turbidity of a 0.5 McFarland standard. Fluconazole (25-μg) and voriconazole (1-μg) disks (Becton Dickinson, Sparks, Md.) were placed onto the surfaces of the plates, and the plates were incubated in air at 35 to 37°C and read at 18 to 24 h. Slowly growing isolates, primarily members of the genus Cryptococcus, were read after 48 h of incubation. Zone diameter endpoints were read at 80% growth inhibition by using the BIOMIC image analysis plate reader system (version 5.9; Giles Scientific, Santa Barbara, Calif.) (6, 19).
The interpretive criteria for the fluconazole and voriconazole disk diffusion tests were those of the CLSI (1a,14): susceptible (S), zone diameters of 19 mm (fluconazole) and 17 mm (voriconazole); susceptible dose dependent (SDD), zone diameters of 15 to 18 mm (fluconazole) and 14 to 16 mm (voriconazole); and resistant (R), zone diameters of 14 mm (fluconazole) and 13 mm (voriconazole). The corresponding MIC breakpoints (13) are as follows: S, MIC of 8 μg/ml (fluconazole) and 1 μg/ml (voriconazole); SDD, MIC of 16 to 32 μg/ml (fluconazole) and 2 μg/ml (voriconazole); R, MIC of 64 μg/ml (fluconazole) and 4 μg/ml (voriconazole).
QC. QC was performed in accordance with CLSI document M44-A (14) by using Candida albicans ATCC 90029 and C. parapsilosis ATCC 22019. A total of 5,865 and 5,484 QC results were obtained for fluconazole and voriconazole, respectively, of which more than 99% were within the acceptable limits.
Analysis of results. All yeast disk test results were read by electronic image analysis and interpreted and recorded with a BIOMIC Plate Reader System (Giles Scientific Inc.). Test results were sent by e-mail to Giles Scientific for analysis. The zone diameter, susceptibility category (S, SDD, or R), and QC test results were all recorded electronically. In addition, MICs were calculated for each drug-organism pair by the BIOMIC System software. The MIC-versus-zone-diameter regression data used by the BIOMIC software were generated previously by ARTEMIS investigators (M.A.P. and M.G.R.) using CLSI BMD MIC and disk test methods (19, 25, 27). Patient and doctor names, duplicate test results (the same patient, the same species, and the same biotype results), and uncontrolled results were automatically eliminated by the BIOMIC system prior to analysis.
RESULTS
Isolation rates by species. A total of 140,767 yeast isolates were collected and tested at 127 study sites between June 1997 and December 2003 (Table 1). Candida species accounted for 95 to 97% of all isolates in each study year (overall, 95.7%). More than 16 different species of Candida were isolated, of which Candida albicans was the most common (overall, 66.2% of all Candida spp.). A decreasing trend in the rate of C. albicans isolation (overall decrease, 10 to 11%) was noted over the 6.5-year period. In contrast, increased rates of isolation of C. tropicalis (an increase of 2.9% from 1997 to 2003) and C. parapsilosis (an increase of 3.1% from 1997 to 2003) were noted. Neither C. glabrata nor C. krusei showed a consistent increase or decrease in isolation rate. Although isolates of more unusual Candida species, such as C. guilliermondii, C. kefyr, C. rugosa, and C. famata, constituted only a small percentage of the Candida isolates, the isolation rates of these four species increased from 2- to 10-fold over the course of the study. Likewise, although C. inconspicua, C. norvegensis, C. lipolytica, C. pelliculosa, and C. zeylanoides are rare species of Candida, the sheer size of the ARTEMIS database provides a significant number of each of these species for study.
Among the noncandidal yeasts, Cryptococcus neoformans (21% of 6,052 isolates), Saccharomyces spp. (6.8%), Trichosporon spp. (6.5%), and Rhodotorula spp. (2.3%) were the most commonly identified species (Table 1). Unidentified ("other") yeasts represented 0.46 to 3.05% of all isolates. As noted previously (6), this percentage decreased somewhat over the course of the study as more isolates were identified to the species level.
Fluconazole and voriconazole susceptibilities of Candida spp. Table 2 summarizes the in vitro susceptibilities of 78,463 and 75,787 isolates of Candida spp. to fluconazole and voriconazole, respectively, as determined by CLSI disk diffusion testing. These isolates were obtained from 115 institutions in 35 countries during the period 2001 through 2003. The distribution of zone diameters and their respective interpretive categories are shown in Fig. 1 for both agents. The percentages of isolates in each category (S, SDD, and R) were 89.6%, 4.0%, and 6.4% and 94.6%, 2.3%, and 3.1% for fluconazole and voriconazole, respectively. Fluconazole was most active against C. albicans (97.8% S), C. parapsilosis (93.2% S), C. lusitaniae (93.3% S), C. kefyr (95.3% S), C. dubliniensis (96.8% S), and C. pelliculosa (94.7% S). Decreased susceptibility to fluconazole was seen with C. glabrata (66.7% S; 16.6% R), C. krusei (9.4% S; 77.2% R), C. guilliermondii (73.3% S; 9.8% R), C. rugosa (39.3% S; 51.8% R), C. famata (79.8% S; 11.9% R), C. inconspicua (25.7% S; 49.2% R), C. norvegensis (50.0% S; 38.0% R), C. lipolytica (54.7% S; 39.6% R), and C. zeylanoides (54.1% S; 37.8% R). These findings confirm previously reported data for the more common species (e.g., C. albicans, C. glabrata, C. parapsilosis, and C. krusei) and markedly expand our understanding of the susceptibility, or lack thereof, of less common species, such as C. rugosa, C. inconspicua, and C. norvegensis, to fluconazole (5, 15, 16, 18, 21, 23).
Voriconazole was significantly more active than fluconazole against virtually every species, with the exception of C. tropicalis (89.1% S to fluconazole versus 87.1% S to voriconazole) (Table 2). Among the species with decreased susceptibility to fluconazole, more than 80% were susceptible to voriconazole, including C. glabrata (81.7% S), C. krusei (83.2% S), C. guilliermondii (91.2% S), C. famata (89.5% S), C. inconspicua (89.2% S), and C. norvegensis (92.3% S). Among the fluconazole-resistant (zone diameter, 14 mm) isolates of C. glabrata, 30% remained susceptible (zone diameter, 17 mm) to voriconazole; however, all voriconazole-resistant strains were also resistant to fluconazole (reference 22 and data not shown). Although voriconazole was more active than fluconazole against C. rugosa (61.4% S versus 39.3% S, respectively), C. lipolytica (67.3% S versus 54.7% S, respectively), and C. zeylanoides (74.3% S versus 54.1% S, respectively), these species were markedly less susceptible and more resistant (11.4% to 26.4%) to voriconazole than all other species of Candida. Again, these data confirm and extend previous observations, especially with the less common species of Candida (18, 20, 23, 24). Importantly, it is readily apparent from these data that although some degree of cross-resistance may be seen between fluconazole and voriconazole, it varies by species and should not be assumed in the absence of species identification and susceptibility testing results.
Trends in resistance to fluconazole among Candida spp. over a 6.5-year period. The longitudinal nature of the ARTEMISDISK Surveillance Program allows one to examine trends in fluconazole resistance among clinical isolates of Candida spp. with the important advantage of sufficient numbers of isolates of each species, all tested by a single standardized method (Table 3). Among the 10 species listed in Table 3, no consistent increase or decrease in fluconazole resistance was seen over time with C. albicans (range, 0.8% to 1.5%) or C. glabrata (range, 14.3% to 22.8%). Although resistance among C. tropicalis isolates appeared to decline from 1997-1998 (4.2%) thru 2001 (3.0%), increases were seen in 2002 (6.6%) and 2003 (5.0%). A slight increase in resistance was noted over time among C. parapsilosis and C. kefyr, whereas a major increase in resistance was detected among isolates of C. rugosa, where 61.2 to 66.0% resistance was observed in the last 2 years of data collection. In contrast, following a peak of 26.1% R in 2000, resistance among isolates of C. guilliermondii decreased steadily between 2001 (11.7% R) and 2003 (8.1% R). Although C. famata appeared to be quite resistant to fluconazole in 1997 and 1998 (47.4% of 19 isolates), this was likely due to the small number of isolates tested. As the numbers of C. famata isolates increased to >50 per year over the next 5 years, the level of resistance stabilized at 10 to 12%. Despite the increase in the overall percentage of isolates of C. krusei that tested as resistant to fluconazole, this is not an important finding, as the species must be considered to be clinically resistant to fluconazole. The CLSI recommends that C. krusei not be tested against fluconazole (13, 14). All such isolates should be reported as fluconazole resistant.
Trends in resistance to voriconazole among Candida spp., 2001 to 2003. Voriconazole has been used clinically since 2001 and since that time has been tested against Candida in the ARTEMIS Global Surveillance Program (Table 4). Overall, there has been a slight increase in the percentage of Candida isolates that appear to be resistant (zone diameter, 13 mm) to voriconazole, from 2.6% in 2001 to 3.5% in 2003. This may be accounted for by increases in resistance observed with C. glabrata (9.8% to 11.0%), C. tropicalis (4.7% to 7.0%), C. rugosa (3.1 to 38.0%), C. lipolytica (7.7% to 12.0%), and unidentified Candida species (4.2% to 7.0%). In contrast, no change or a decrease in resistance was seen with C. albicans, C. parapsilosis, C. krusei, C. lusitaniae, C. kefyr, C. famata, C. inconspicua, C. dubliniensis, and C. pelliculosa. Thus, the picture for voriconazole, in terms of spectrum and potency versus Candida spp., looks quite favorable. Emerging resistance, especially among C. glabrata, C. tropicalis, and C. rugosa, bears close monitoring.
Geographic variation in the susceptibilities of C. albicans and C. glabrata to fluconazole and voriconazole. Table 5presents the in vitro susceptibility results for fluconazole and voriconazole tested against the two most common species of Candida, C. albicans and C. glabrata, stratified by geographic region and country of origin for the time period 2001 to 2003. With the exception of those from India, isolates of C. albicans were highly susceptible to both fluconazole and voriconazole. The only other countries where the percentages of C. albicans susceptible to either agent dropped below 94% were Colombia (fluconazole, 91.2% S, 6.1% R) and Ecuador (fluconazole, 91.6% S, 4.9% R). Overall, there was no meaningful difference in the fluconazole or voriconazole susceptibility profile for C. albicans when stratified by specimen type (96.7% to 99.3% S to fluconazole; 97.9% to 99.3% S to voriconazole) or by hospital location (95.3 to 99.1% S to fluconazole; 97.2% to 99.4% S to voriconazole) (data not shown).
Fluconazole and voriconazole susceptibilities of C. glabrata isolates varied considerably among the various countries and geographic regions. Susceptibilities to fluconazole were lowest (<50%) in Venezuela (29.2% S), Malaysia (34.0% S), Belgium (39.7% S), the Czech Republic (44.8% S), and South Africa (49.6%) and highest (>80%) in India and the Middle East (100% S), Brazil (94.9% S), Greece (93.9% S), Canada (90.6% S), Portugal (87.1% S), Mexico (86.7% S), Poland (86.4% S), South Korea (83.7% S), Turkey (82.4% S), and Italy (81.3% S). Overall rates of resistance to fluconazole among C. glabrata isolates were 10.6% in the Asia-Pacific region, 13.2% in Latin America, 16.5% in Europe, and 18.0% in North America (data not shown). These rates of fluconazole resistance are considerably higher for each geographic region than those reported previously for blood and normally sterile-site infection isolates of C. glabrata (range, 2 to 9% R) tested by BMD between 1992 and 2000 (20).
In contrast to that seen with C. albicans, the susceptibility of C. glabrata isolates to fluconazole varied according to specimen type and hospital location. Isolates from blood and normally sterile sites were the most susceptible (71% S; 14.8% R) and genital tract isolates were the least susceptible (53.6% S; 21.2% R) to fluconazole (data not shown). The highest rates of resistance were seen in isolates of C. glabrata from the surgical intensive-care unit (21.3%), the obstetrics and gynecology service (21.5%), the hematology/oncology service (22.6%), and the neonatal intensive-care unit (35.0%) (data not shown).
Voriconazole was equally or more active than fluconazole against C. glabrata isolates from all countries and geographic regions (Table 5). Susceptibilities to voriconazole were lowest (<70%) in Venezuela (32.7% S), Belgium (53.2% S), Malaysia (59.5% S), the Czech Republic (65.3% S), and Ecuador (66.7% S) and highest (>90%) in India, Turkey and the Middle East (100% S), Brazil (96.8% S), Canada (95.7% S), Greece (95.5%), Thailand (92.5%), and Portugal (90.0%). Overall rates of resistance to voriconazole among C. glabrata isolates were 4.1% in the Asia-Pacific region, 5.4% in Latin America, 5.6% in Europe, and 9.0% in North America (data not shown). Our previous results using BMD MIC testing found resistance rates of 2.2 to 5.4% among blood and normally sterile-site isolates of C. glabrata tested in 2001 and 2002 (22). Similar to that seen with C. albicans, there was little variation in the susceptibility of C. glabrata to voriconazole when stratified by specimen type. Isolates from blood and normally sterile sites were the most susceptible (81%) and genital tract isolates were the least susceptible (70%) to voriconazole (data not shown). The rates of resistance to voriconazole ranged from 2.5% (neonatal intensive-care unit) to 8.2% (hematology/oncology service) across the different hospital locations.
Activities of fluconazole and voriconazole against other opportunistic yeasts and yeast-like fungi. Although they comprise only 3 to 5% of all of the isolates tested in this study, the number of noncandidal yeasts tested against fluconazole and voriconazole exceeds that published in the current literature (1, 3, 21, 26). Lack of standardized methods for testing most of these fungi may be considered problematic; however, the vast majority grew well on the MH-MB agar plates, and the zone diameters were easily determined. For the purposes of this study, we utilized the interpretive breakpoints for Candida, and we recognize that they may be adjusted for noncandidal yeasts in the future. Nevertheless, the data generated for these organisms are not dissimilar to those obtained using CLSI BMD MIC methods (1, 3, 21, 26). Using Cryptococcus neoformans as an example, the susceptibilities of the isolates shown in Table 6 indicated moderate susceptibility to fluconazole and a very high level of activity for voriconazole. Very similar findings for these two agents using BMD MIC methods were recently reported from our laboratory (26). As noted previously (21), most of these noncandidal yeasts were substantially less susceptible to both fluconazole and voriconazole than Candida species. Although voriconazole was more active than fluconazole for each of these different genera, it is notable that less than 80% of Trichosporon beigelii/Trichosporon cutaneum, Trichosporon asahi, and Rhodotorula spp. were susceptible to either of these agents. The diverse array of opportunistic yeasts and yeast-like fungi and their variable susceptibilities to these azole antifungals emphasize the need for prompt identification of noncandidal yeasts from clinical material. The flexibility of the CLSI disk diffusion method may well be an advantage in assessing the antifungal susceptibilities of these "emerging" pathogens.
Conversion of zone diameters to MICs. In addition to using image analysis technology to measure and record the zones of inhibition surrounding an antifungal disk, the BIOMIC system uses previously developed scatter plots and regression analysis to calculate MICs based on the relationship between the zone diameter and the MIC (Fig. 2). The data in Fig. 2 show the correlation between the MIC and the zone diameter for voriconazole with Candida spp. As seen previously with fluconazole (6, 19, 25), an excellent correlation was observed. Based on these data, the voriconazole MICs for Candida spp. were calculated and the data were compared to BMD MICs published previously (24) for the same species (Table 7). Although the numbers of isolates tested are considerably different in the two groups, it is readily apparent that the MIC50 and MIC90 values are very close for each species, as is the percent resistant. Thus, the large amount of qualitative disk diffusion data presented here can be converted to quantitative MIC data for purposes of comparing the activities of fluconazole and voriconazole for individual species (Fig. 3) or potentially for following trends across time. Additional work in this area is warranted.
DISCUSSION
The ARTEMIS Global Antifungal Surveillance Program is the largest and most comprehensive program of its kind and the only one to incorporate many of the features that arguably constitute an "ideal" resistance surveillance program (7-9, 11, 30). It is longitudinal and global, employs standardized methods used for "routine" testing in participating laboratories and for "reference" testing in a central reference laboratory, uses electronic data capture and storage in a central database, and conducts external validation of the data generated by participating laboratories. The current report from the ARTEMIS DISK Surveillance Study includes more than 140,000 opportunistic yeast isolates and is by far the largest and most geographically diverse study of antifungal susceptibility and resistance to date (5, 15, 16, 20, 28). Important findings regarding species distribution include a steady decrease in the isolation of C. albicans and an increase in the isolation of C. tropicalis and C. parapsilosis. Although they are still rare, it appears that C. rugosa, C. famata, C. inconspicua, and C. norvegensis may be "emerging" in recent years. Among the noncandidal yeasts, Cryptococcus neoformans, Saccharomyces, Trichosporon, and Rhodotorula species are prominent and may prove to be important due to their decreased susceptibilities to several antifungal agents (21).
Despite the use of a standard protocol, it is recognized that any surveillance program based on susceptibility tests performed by the participating laboratories needs to include some measure of quality assurance, beyond simple QC testing, in order to provide an independent assessment of laboratory performance and validation of the results generated by the various laboratories (7, 9, 31). One approach to cross-validation that has been suggested is to use centralized testing with high-quality microbiology to confirm the trends in routine data obtained from participating sentinel sites (7-9, 11, 30). Comparison of results obtained for isolates tested in participating laboratories with results obtained for the same organisms tested in a central reference laboratory would accomplish this goal (8, 9). This approach has been used to validate and support the epidemiologic relevance of findings from antibacterial surveillance programs (9, 30). Most recently, we have used the same approach to validate fluconazole and voriconazole disk test results generated by laboratories participating in the ARTEMIS Program (25, 27). More than 2,900 isolates of Candida obtained from blood and normally sterile-site infections were tested against fluconazole and voriconazole by ARTEMIS participating laboratories (CLSI disk test) and by the central reference laboratory (CLSI disk and BMD MIC tests) (25, 27). Categorical agreement between the reference MIC results and the disk diffusion test results performed in the participant laboratories was 87.4% and 94.1% for fluconazole and voriconazole, respectively (Table 8). A similar level of agreement was seen when the disk test results obtained in the reference laboratory were compared with those from the participant laboratories (references 25 and 27 and data not shown). It was noted that participating laboratories tended to err on the side of calling isolates more resistant than the reference laboratory did; however, the numbers of major and very major discrepancies were quite small (Table 8). This external quality assurance data, coupled with excellent QC performance, ensures the generation of accurate and useful surveillance data in the ARTEMIS DISK Surveillance Program.
The data reported here for the more common species of Candida (i.e., C. albicans, C. glabrata, C. parapsilosis, and C. tropicalis) confirm most of the previously published data regarding their susceptibilities to fluconazole and voriconazole (5, 16, 20, 24). The activity of fluconazole remains high against C. albicans, C. parapsilosis, and C. tropicalis, although resistance may be increasing among C. tropicalis isolates. Fluconazole resistance was considerable among isolates of C. glabrata, although the extent of resistance varied widely throughout the world. Fortunately, voriconazole remains quite active against this species. It is notable, however, that resistance to voriconazole has increased among C. glabrata isolates over the 3-year period of this study and was quite high in certain countries, such as Belgium (18.5%) and Venezuela (38.8%), where fluconazole resistance was also widespread. Again, our previous studies have shown that compared to reference laboratory testing of C. glabrata by MIC and disk methods, the fluconazole and voriconazole disk test results reported by ARTEMIS participating sites tended to overestimate resistance (25, 27). Thus, the rates of resistance to fluconazole and voriconazole reported in this study for C. glabrata may be somewhat higher than previously reported in the literature. Nevertheless, the geographical and temporal comparisons and differences remain important.
The ARTEMIS database is most valuable as it pertains to the less common species of Candida (Table 2). The excellent activity of voriconazole against C. krusei was confirmed by the results from almost 2,000 clinical isolates. Similarly, the high levels of activity of both azoles against C. lusitaniae, C. kefyr, C. dubliniensis, and C. pelliculosa were clearly demonstrated, confirming previous results based on comparatively few isolates (21, 23). Equally important was the demonstration of generally poor activities of fluconazole against C. guilliermondii, C. rugosa, C. famata, C. inconspicua, C. norvegensis, C. lipolytica, and C. zeylanoides. In most instances, these findings confirm what can only be called preliminary observations (21); however, for some of these species, these constitute new data and serve to underscore the imperative to identify Candida to the species level. Although voriconazole is active against the vast majority of these rare species, it is notable that decreased susceptibility to this agent, as well as to fluconazole, is seen with C. rugosa, C. lipolytica, and C. zeylanoides. These findings are especially important for C. rugosa, as the frequency of isolation of this species appears to be increasing over time (Table 1), it has been shown to cause clusters of nosocomial infection that are poorly responsive to amphotericin B (2, 4), and it was previously considered highly susceptible to voriconazole based on results for less than 20 clinical isolates (21).
As is the case for the less common Candida species, new information for noncandidal yeasts is provided by this data set. Although the antifungal susceptibility profile of Cryptococcus neoformans is well known (1, 26), much less is known of the susceptibilities of Saccharomyces, Trichosporon, Rhodotorula, and Blastoschizomyces species to fluconazole and voriconazole (3, 21, 33). The results presented in Table 6 indicate that most of these opportunistic yeasts have decreased susceptibility to fluconazole, and although voriconazole is clearly more active than fluconazole, decreased susceptibility to that agent is also seen with certain species of Trichosporon and with Rhodotorula spp. The fact that these yeast-like fungi are also nonsusceptible to the echinocandins (they lack -1,3-D-glucan) and respond variably to amphotericin B highlights the potential for their emergence as difficult-to-treat mycotic pathogens in the future (21, 33).
Finally, the ability of the BIOMIC software to convert disk diffusion zone diameters to MICs is an important feature of the ARTEMIS surveillance program, providing quantitative data that will be valuable in trend analysis. We have extended the previous work of Hazen et al. (6) and have shown that the voriconazole MICs calculated from the disk diffusion data for Candida spp. compare very favorably to those obtained by BMD MIC testing performed centrally (Table 7).
In summary, we present a tremendous volume of data describing temporal and geographic trends in the isolation and azole susceptibilities of opportunistic yeast pathogens. The data point to the strength of azole coverage for most of these organisms but also highlight several areas where resistance may be progressing and/or previously rare species may be "emerging." The strength of the ARTEMIS Global Surveillance Program is in the overall design, incorporating standardized test methods, "routine" and centralized testing of isolates, and a broad international network of study sites providing consistent data over time. The continued efforts of this surveillance program will provide data on pathogen frequency and antifungal susceptibility on a global scale.
ACKNOWLEDGMENTS
Linda Elliott provided excellent support in the preparation of the manuscript.
The ARTEMIS DISK Surveillance Program is supported by grants from Pfizer.
We express our appreciation to all ARTEMIS participants. Participants contributing to this study included Jorge Finquelievich, Buenos Aires University, and Nora Tiraboschi, Hospital Escuela Gral., Buenos Aires, Argentina; David Ellis, Women's and Children's Hospital, North Adelaide, Australia; Dominique Frameree, CHU de Jumet, Jumet, Annemarie van den Abeele, St Lucas Campus Heilige Familie, Ghent, and Jean-Marc Senterre, Hpital de la Citadelle, Liege, Belgium; Arnaldo Colombo, Escola Paulista de Medicina, Sao Paulo, Brazil; Robert Rennie, University of Alberta Hospital, Edmonton, and Steve Sanche, Royal University Hospital, Saskatoon, Canada; Bijie Hu, Zhong Shan Hospital, Shanghai, Yingchun Xu, Peking Union Medical College Hospital, Beijing, Yingyuan Zhang, Hua Shan Hospital, Shanghai, and Nan Shan Zhong, Guangzhou Institute of Respiratory Disease, Guangzhou, China; Pilar Rivas, Inst. Nacional de Cancerología, Bogota, Angela Restrepo and Catalina Bedout, CIB, Medellin, and Ricardo Vega and Matilde Mendez, Hospital Militar Central, Bogota, Colombia; Nada Mallatova, Hospital Ceske Budejovice, Ceske, and Eva Chmelarova, Krajska Hygienicka Stanice, Ostrava, Czech Republic; Julio Ayabaca, Hospital FF. AA HG1, Quito, and Jeannete Zurita, Hospital Vozandes, Quito, Ecuador; M. Mallie, Faculte de Pharmacie, Montpellier, and E. Candolfi, Institut de Parasitologie, Strasbourg, France; W. Fegeler, Universitaet Muenster, Münster, A. Haase, RWTH Aachen, Aachen, G. Rodloff, Inst. F. Med. Mikrobiologie, Leipzig, W. Bar, Carl-Thiem Klinikum, Cottbus, and V. Czaika, Humaine Kliniken, Bad Saarow, Germany; George Petrikos, Laikon General Hospital, Athens, Greece; Erzsebet Puskás, BAZ County Institute, Miskolc, Ilona Doczi, University of Szeged, Szeged, Mestyan Gyula, Medical University of Pecs, Pecs, and Radka Nikolova, Szt Laszlo Hospital, Budapest, Hungary; Uma Banerjee, All India Institute of Medical Sciences, New Delhi, India; Nathan Keller, Sheba Medical Center, Tel Hashomer, Israel; Vivian Tullio, Universita degli Studi di Torino, Turin, Gian Carlo Schito, University of Genoa, Genoa, Giacomo Fortina, Ospedale di Novara, Novara, Gian Piero Testore, Univerrsita di Roma Tor Vergata, Rome, Domenico D'Antonio, Pescara Civil Hospital, Pescara, Giorgio Scalise, Instituto di Malattie Infettive, Ancona, Pietro Martino, Dept. di Biotechnologie, Rome, and Graziana Manno, Universita di Genova, Genova, Italy; Kee Peng, University Malaya, Kuala Lumpur, Malaysia; Celia Alpuche and Jose Santos, Hospital General de Mexico, Mexico City, Eduardo Rodriguez Noriega, Universidad de Guadalajara, Guadalajara, andMussaret Zaidi, Hospital General O'Horan, Merida, Mexico; Jacques F. G. M. Meis, Canisius Wilhemina Hospital, Nijmegen, The Netherlands; Egil Lingaas, Rikshospitalet, Oslo, Norway; Danuta Dzierzanowska, Children's Memorial Health Institute, Warsaw, and Waclaw Pawliszyn, Pracownia Bakteriologii, Cracow, Poland; Mariada Luz Martins, Inst. de Higiene e Medicina Tropical, Lisbon, Luis Albuquerque, Centro Hospitalar de Coimbra, Coimbra, Laura Rosado, Instituto Nacional de Saude, Lisbon, Rosa Velho, Hosp. da Universidade de Coimbra, Coimbra, and Jose Amorim, Hospital de Santo Antonio, Porto, Portugal; Vera N. Ilina, Novosibirsk Regional Hospital, Novosibirsk, Olga I. Kretchikova, Institute of Antimicrobial Chemotherapy, Smolensk, Galina A. Klyasova, Hematology Research Center, Moscow, Sophia M. Rozanova, City Clinical Hospital No 40, Ekaterinburg, Irina G. Multykh, Territory Center of Laboratory Diagnostics, Krasnodar, Nikolay N. Klimko, Medical Mycology Research Institute, St. Petersburg, Elena D. Agapova, Irkutsk Regional Children's Hospital, Irkutsk, and Natalya V. Dmitrieva, Oncology Research Center, Moscow, Russia; Abdul Mohsen Al-Rasheed, Riyadh Armed Forces Hospital, Riyadh, Saudi Arabia; Jan Trupl, National Cancer Center, Leon Langsadl, NUTaRCH, Alena Vaculikova, Derer University Hospital, and Hupkova Helena, St. Cyril and Metod Hospital, Bratislava, Slovak Republic; Denise Roditi, Groote Schuur Hospital, Cape Town, Anwar Hoosen, GaRankuwa Hospital, Medunsa, H. H. Crewe-Brown, Baragwanath Hospital, Johannesburg, M. N. Janse van Rensburg, Pelanomi Hospital, UOFS, Bloemfontein, and Adriano Duse, Johannesburg General Hospital, Johannesburg, South Africa; Kyungwon Lee, Yonsei University College of Medicine, and Mi-Na Kim, Asan Medical Center, Seoul, South Korea; A. del Palacio, Hospital 12 De Octobre, and Aurora Sanchez-Sousa, Hospital Ramon y Cajal, Madrid, Spain; Jacques Bille, Institute of Microbiology CHUV, Lausanne, and K. Muhlethaler, Universitat Bern, Bern, Switzerland; Shan-Chwen Chang, National Taiwan University Hospital, Taipei, and Jen-Hsien Wang, China Medical College Hospital, Taichung, Taiwan; Malai Vorachit, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand; Deniz Gur, Hacettepe University Children's Hospital, Ankara, and Volkan Korten, Marmara Medical School Hospital, Istanbul, Turkey; John Paul, Royal Sussex County Hospital, Brighton, Brian Jones, Glasgow Royal Infirmary, Glasgow, F. Kate Gould, Freeman Hospital, Newcastle, Chris Kibbler, Royal Free Hospital, London, Nigel Weightman, Friarage Hospital, Northallerton, Ian M. Gould, Aberdeen Royal Hospital, Aberdeen, Ruth Ashbee, General Infirmary, P.H.L. S, Leeds, and Rosemarie Barnes, University of Wales College of Medicine, Cardiff, United Kingdom; Jose Vazquez, Harper Hospital, Wayne State University, Detroit, Michigan; Ed Chan, Mt. Sinai Medical Center, New York, and Davise Larone, Cornell Medical Center NYPH, Ithaca, N.Y.; Ellen Jo Baron, Stanford Hospital and Clinics, Stanford, Calif.; Mahmoud A. Ghannoum, University Hospitals of Cleveland, Cleveland, Ohio; Mike Rinaldi, University of Texas Health Science Center, San Antonio, Texas; Kevin Hazen, University of Virginia Health Systems, Charlottesville, Va.; Elyse Foraker, Christiana Care, Wilmington, Del.; and Heidi Reyes, Gen del Este Domingo Luciani, and Axel Santiago, Universitario de Caracas, Caracas, Venezuela.
REFERENCES
Brandt, M. E., M. A. Pfaller, R. A. Hajjeh, R. H. Hamill, P. G. Pappas, A. L. Reingold, D. Rimland, and D. W. Warnock for the Cryptococcal Disease Active Surveillance Group. 2001. Trends in antifungal drug susceptibility of Cryptococcus neoformans isolates in the United States: 1992 to 1994 and 1996 to 1998.Antimicrob. Agents Chemother. 45:3065-3069.
CLSI.2005 . Minutes of the CLSI Antifungal Subcommittee Meeting, 2005. CLSI, Wayne, Pa.
Colombo, A. L., S. A. Melo, R. F. C. Rosas, R. Salomao, M. Briones, R. J. Hollis, S. A. Messer, and M. A. Pfaller. 2003. Outbreak of Candida rugosa candidemia: an emerging pathogen that may be refractory to amphotericin B therapy. Diagn. Microbiol. Infect. Dis. 46:253-257.
Diekema, D. J., B. Petroelje, S. A. Messer, R. J. Hollis, and M. A. Pfaller. 2005. Activities of available and investigational antifungal agents against Rhodotorula species. J. Clin. Microbiol. 43:476-478.
Dube, M. P., P. N. R. Heseltine, M. G. Rinaldi, S. Evans, and B. Zawacki. 1994. Fungemia and colonization with nystatin-resistant Candida rugosa in a burn unit. Clin. Infect. Dis. 18:77-82.
Hajjeh, R. A., A. N. Sofair, L. H. Harrison, G. M. Lyon, B. A. Arthington-Skaggs, S. A. Mirza, M. Phelan, J. Morgan, W. Lee-Yang, M. A. Ciblak, L. E. Benjamin, L. Thompson Sanza, S. Huie, S. F. Yeo, M. E. Brandt, and D. W. Warnock.2004 . Incidence of bloodstream infections due to Candida species and in vitro susceptibilities of isolates collected from 1998 to 2000 in a population-based active surveillance program. J. Clin. Microbiol. 42:1519-1527.
Hazen, K. C., E. J. Baron, A. L. Colombo, C. Girmenia, A. Sanchez-Sousa, A. del Palacio, C. de Bedout, D. L. Gibbs, and The Global Antifungal Surveillance Group.2003 . Comparison of the susceptibilities of Candida spp. to fluconazole and voriconazole in a 4-year global evaluation using disk diffusion. J. Clin. Microbiol. 41:5623-5632.
Kahlmeter, G., and D. F. J. Brown. 2002. Resistance surveillance studies—comparability of results and quality assurance of methods. J. Antimicrob. Chemother. 50:775-777.
Livermore, D. M., E. J. Threlfall, M. H. Reacher, A. P. Johnson, D. James, T. Cheasty, A. Shah, F. Warburton, A. V. Swan, J. Skinner, A. Grahm, and D. C. E. Speller. 2000. Are routine test data suitable for the surveillance of resistance Resistance rates amongst Escherchia coli from blood and CSF from 1991-1997, as assessed by routine and centralized testing. J. Antimicrob. Chemother. 45:205-211.
Livermore, D. M., A. P. Macgowan, and M. C. J. Wale. 2005. Surveillance of antimicrobial resistance: centralized surveys to validate routine data offer a practical approach. Br. Med. J. 317:614-615.
Luzzati, R., G. Amalfitano, L. Lazzarini, F. Soldori, S. Bellino, M. Solbiatic, M. C. Danzi, S. Vento, G. Todeschini, C. Vivenza, and E. Concia. 2000. Nosocomial candidemia in non-neutrophenic patients at an Italian tertiary care hospital.Eur. J. Clin. Microbiol. Infect. Dis. 19:602-607.
Magee, J. T., M. L. Heginbothom, and B. W. Mason. 2005. Finding a strategy: the case for co-operative research on resistance epidemiology. J. Antimicrob. Chemother. 55:628-633.
Meis, J., M. Petrou, J. Bille, D. Ellis, D. Gibbs, and the Global Antifungal Surveillance Group. 2000. A global evaluation of the susceptibility of Candida species to fluconazole by disk diffusion. Diagn. Microbiol. Infect. Dis. 36:215-223.
National Committee for Clinical Laboratory Standards. 2002. Reference method for broth dilution testing of yeasts. Approved standard, 2nd ed. M27-A2. National Committee for Clinical Laboratory Standards, Wayne, Pa.
National Committee for Clinical Laboratory Standards. 2004. Method for antifungal disk diffusion susceptibility testing of yeasts: approved guidance M44-A. National Committee for Clinical Laboratory Standards, Wayne, Pa.
Nguyen, M. H., J. E. Peacock, Jr., A. J. Morris, D. C. Tanner, M. L. Nguyen, D. R. Snydman, M. M. Wagener, M. G. Rinaldi, and V. L. Yu. 1996. The changing face of candidemia: emergence of non-Candida albicans species and antifungal resistance. Am. J. Med. 100:617-623.
Ostrosky-Zeichner, L., J. H. Rex, P. G. Pappas, R. J. Hamill, R. A. Larsen, H. W. Horowitz, W. G. Powderly, N. Hyslop, C. A. Kauffman, J. Cleary, J. E. Mangino, and J. Lee. 2003. Antifungal susceptibility survey of 2,000 bloodstream Candida isolates in the United States. Antimicrob. Agents Chemother. 47:3149-3154.
Pfaller, M. A., and D. J. Diekema. 2002. Role of sentinel surveillance of candidemia: trends in species distribution and antifungal susceptibility. J. Clin. Microbiol. 40:3551-3557.
Pfaller, M. A., D. J. Diekema, S. A. Messer, L. Boyken, R. J. Hollis, R. N. Jones, and the International Fungal Surveillance Participant Group.2003 . In vitro activities of voriconazole, posaconazole, and four licensed systemic antifungal agents against Candida species infrequently isolated from blood. J. Clin. Microbiol. 41:28-83.
Pfaller, M. A., D. J. Diekema, S. A. Messer, L. Boyken, and R. J. Hollis. 2003. Activites of fluconazole and voriconazole against 1,586 recent clinical isolates of Candida species determined by broth microdilution, disk diffusion, and Etest methods: report from the ARTEMIS Global Antifungal Susceptibility Program, 2001. J. Clin. Microbiol. 41:1440-1446.
Pfaller, M. A., and D. J. Diekema. 2004. Twelve years of fluconazole in clinical practice: global trends in species distribution and fluconazole susceptibility of bloodstream isolates of Candida. Clin. Microbiol. Infect. 10(Suppl. 1):11-23.
Pfaller, M. A., and D. J. Diekema. 2004. Rare and emerging opportunistic fungal pathogens: concern for resistance beyond Candida albicans and Aspergillus fumigatus. J. Clin. Microbiol. 42:4419-4431.
Pfaller, M. A., S. A. Messer, L. Boyken, S. Tendolkar, R. J. Hollis, and D. J. Diekema.2004 . Geographic variation in the susceptibilities of invasive isolates of Candida glabrata to seven systemically active antifungal agents: a global assessment from the ARTEMIS Antifungal Surveillance Program conducted in 2001 and 2002.J. Clin. Microbiol. 42:3142-3146.
Pfaller, M. A., D. J. Diekema, S. A. Messer, L. Boyken, R. J. Hollis, and R. N. Jones.2004 . In vitro susceptibilities of rare Candida bloodstream isolates to ravuconazole and three comparative antifungal agents. Diagn. Microbiol. Infect. Dis. 48:101-105.
Pfaller, M. A., S. A. Messer, L. Boyken, R. J. Hollis, C. Rice, S. Tendolkar, and D. J. Diekema.2004 . In vitro activities of voriconazole, posaconazole, and fluconazole against 4,169 clinical isolates of Candida spp. and Cryptococcus neoformans collected during 2001 and 2002 in the ARTEMIS global antifungal surveillance program.Diagn. Microbiol. Infect. Dis. 48:201-205.
Pfaller, M. A., K. C. Hazen, S. A. Messer, L. Boyken, S. Tendolkar, R. J. Hollis, and D. J. Diekema. 2004. Comparison of results of fluconazole disk diffusion testing for Candida species with results from a central reference laboratory in the ARTEMIS Global Antifungal Surveillance Program. J. Clin. Microbiol. 42:3607-3612.
Pfaller, M. A., S. A. Messer, L. Boyken, C. Rice, S. Tendolkar, R. J. Hollis, G. V. Doern, and D. J. Diekema. 2005. Global trends in the antifungal susceptibility of Cryptococcus neoformans (1990-2004). J. Clin. Microbiol. 43:2163-2167.
Pfaller, M. A., L. Boyken, S. A. Messer, S. Tendolkar, R. J. Hollis, and D. J. Diekema.2005 . Comparison of results of voriconazole disk diffusion testing for Candida species with results from a central reference laboratory in the ARTEMIS Global Antifungal Surveillance Program. J. Clin. Microbiol. 43:5208-5213.
Rees, J. R., R. W. Pinner, R. A. Hajjeh, M. R. Brandt, and A. L. Reingold.1998 . The epidemiological features of invasive mycotic infection in the San Francisco Bay Area, 1992-1993: results of a population-based laboratory active surveillance. Clin. Infect. Dis. 27:1138-1147.
Richet, H., P. Roux, C. Des Champs, Y. Esnault, A. Andremont and the French Candidemia Study Group. 2002. Candidemia in French hospitals: incidence rates and characteristics. Clin. Microbiol. Infect. 8:405-412.
Stelling, J. M., K. Travers, R. N. Jones, P. J. Turner, T. F. O'Brien, and S. B. Levy.2005 . Integrating Escherichia coli antimicrobiol susceptibility data from multiple surveillance programs. Emerg. Infect. Dis. 11:873-882.
Tenover, F. C., M. J. Mohammed, J. Stelling, T. O'Brien, and R. Williams. 2001. Ability of laboratories to detect emerging antimicrobial resistance: proficiency testing and quality control results from the World Health Organizations' external quality assurance system for antimicrobial susceptibility testing.J. Clin. Microbiol. 39:241-250.
Tortorano, A. M., J. Peman, H. Bernhardt, L. Klingspor, C. C. Kibbler, O. Faure, E. Biraghi, E. Canton, K. Zimmerman, S. Seaton, R. Grillot, and the ECMM Working Group on Candidaemia.2004 . Epidemiology of candidaemia in Europe: results of 28-month European Confederation of Medical Mycology (ECMM) hospital-based surveillance study. Eur. J. Clin. Microbiol. Infect. Dis. 23:317-322.
Walsh, T. J., A. Groll, J. Hiemenz, R. Flemming, E. Roilides, and E. Anaissie. 2004. Infections due to emerging and uncommon medically important fungal pathogens. Clin. Microbiol. Infect. 10(Suppl. 1):48-66.(M. A. Pfaller, D. J. Diek)
University of Texas Health Science Center, San Antonio, Texas
University of Wales College of Medicine, Cardiff, United Kingdom
Zhong Shan Hospital, Shanghai, China
Institute of Antimicrobial Chemotherapy, Smolensk, Russia
Hospital de Clinicas "Jose de San Martin," Buenos Aires, Argentina
Institute of Clinical Microbiology, Faculty of Medicine, University of Szeged, Szeged, Hungary
Giles Scientific, Inc., Santa Barbara, California
ABSTRACT
Fluconazole in vitro susceptibility test results for 140,767 yeasts were collected from 127 participating investigators in 39 countries from June 1997 through December 2003. Data were collected on 79,343 yeast isolates tested with voriconazole from 2001 through 2003. All investigators tested clinical yeast isolates by the CLSI (formerly NCCLS) M44-A disk diffusion method. Test plates were automatically read and results were recorded with the BIOMIC Vision Image Analysis System. Species, drug, zone diameter, susceptibility category, and quality control results were collected quarterly via e-mail for analysis. Duplicate (the same patient, same species, and same susceptible-resistant biotype profile during any 7-day period) and uncontrolled test results were not analyzed. The 10 most common species of yeasts all showed less resistance to voriconazole than to fluconazole. Candida krusei showed the largest difference, with over 70% resistance to fluconazole and less than 8% to voriconazole. All species of yeasts tested were more susceptible to voriconazole than to fluconazole, assuming proposed interpretive breakpoints of 17 mm (susceptible) and 13 mm (resistant) for voriconazole. MICs reported in this study were determined from the zone diameter in millimeters from the continuous agar gradient around each disk, which was calibrated with MICs determined from the standard CLSI M27-A2 broth dilution method by balanced-weight regression analysis. The results from this investigation demonstrate the broad spectrum of the azoles for most of the opportunistic yeast pathogens but also highlight several areas where resistance may be progressing and/or where previously rare species may be "emerging."
INTRODUCTION
Antifungal resistance surveillance with a focus on Candida is now widespread (5, 10, 17, 20, 29, 32). Most of these surveillance efforts are by necessity limited in terms of the numbers of participating sentinel sites and isolates tested. Furthermore, none of the programs is extensive enough to provide temporal and geographic data concerning the occurrence and resistance profiles of the less common Candida species and other, noncandidal opportunistic yeasts (21).
The ARTEMIS Global Antifungal Surveillance Program is among the most comprehensive and long-running fungal surveillance programs (6, 12, 17, 19, 22, 24, 25, 27). The ARTEMIS Program is made up of two components: (i) a broad international network of participating sites (127 sites in 39 countries), each of which performs Clinical and Laboratory Standards Institute (CLSI, formerly National Committee for Clinical and Laboratory Standards [NCCLS])-recommended disk diffusion testing (M44-A) (14) of fluconazole and voriconazole against consecutive yeast isolates from a variety of clinical sources (ARTEMIS DISK Surveillance Study) (6), and (ii) a central reference laboratory (University of Iowa, Iowa City), where CLSI-recommended broth microdilution (BMD) MIC and disk diffusion testing (M27-A2 and M44-A, respectively) (13, 14) is performed on blood and normally sterile-site isolates of Candida and other opportunistic yeasts and molds thatare referred according to protocol from the participating ARTEMIS study sites (19, 22, 24, 25, 27). As such, the ARTEMIS Program has been designed to address many of the potential limitations of resistance surveillance studies (7): (i) it is both longitudinal (1997 to present) and global (127 participating sites in 39 countries) in scope, (ii) it employs standardized antifungal susceptibility test methods (CLSI disk [M44-A] and BMD MIC [M27-A2]) (13, 14), (iii) both internal quality control (QC) performed in each participating laboratory and external quality assurance measures are used to validate test results (25, 27), (iv) results are recorded electronically using the BIOMIC image analysis plate reader system (Giles Scientific, Santa Barbara, Calif.) (6, 19, 25, 27) and are stored in a central database, and (v) both Candida and non-Candida yeast isolates obtained from consecutive clinical samples from all body sites are tested locally, thus avoiding misleading results based on biased selective testing. This so-called "routine" testing is augmented by testing of isolates from blood and normally sterile sites in the central reference laboratory (25, 27). Thus, the ARTEMIS Program generates massive amounts of data that have been externally validated and that can be used to identify temporal and geographic trends in the species distribution of Candida and other opportunistic yeasts, as well as the resistance profiles of these organisms to fluconazole and voriconazole as determined by standardized CLSI disk diffusion testing.
In the present study, we utilized the results from the ARTEMIS DISK Surveillance Program to evaluate global trends in the susceptibility of yeasts to fluconazole over a 6.5-year period (140,767 isolates from 127 study sites in 39 countries; June 1997 through December 2003). We also report results of voriconazole susceptibility testing performed on 79,343 isolates collected from 2001 to 2003. The scope of this study provides an unprecedented look at the occurrence and azole susceptibilities of several rare species of Candida, as well as several of the other opportunistic yeasts. The study is limited in that the numbers of isolates from certain regions are small and the time frame over which voriconazole data are available is relatively short.
MATERIALS AND METHODS
Organisms and test sites. A total of 134,715 isolates of Candida spp. and 6,052 isolates of noncandidal yeasts obtained from 127 different medical centers in Asia (23 sites), Latin America (16 sites), Europe (74 sites), the Middle East (2 sites), and North America (12 sites) were collected and tested against fluconazole between June 1997 and December 2003. In addition, a total of 79,343 isolates (75,810 isolates of Candida spp. and 3,533 other yeasts) from 115 study sites in 35 countries were tested against voriconazole between 2001 and 2003. All yeasts considered pathogens from all body sites (e.g., blood, normally sterile body fluids, deep tissue, genital tract, gastrointestinal tract, respiratory tract, skin, and soft tissue) and isolates from patients in all in-hospital locations during the study period were tested. Yeasts considered by the local site investigator to be colonizers, that is, not associated with an obvious pathology, were excluded, as were duplicate isolates from a given patient (the same species and the same susceptible-resistant biotype profile within any 7-day period). Identification of isolates was performed in accordance with each site's routine methods.
Susceptibility test method. Disk diffusion testing of fluconazole and voriconazole was performed as described by Hazen et al. (6) and in CLSI document M44-A (14). Agar plates (150-mm diameter) containing Mueller-Hinton agar (obtained locally at all sites) supplemented with 2% glucose and 0.5 μg of methylene blue per ml (MH-MB) at a depth of 4.0 mm were used. The agar surface was inoculated by using a swab dipped in a cell suspension adjusted to the turbidity of a 0.5 McFarland standard. Fluconazole (25-μg) and voriconazole (1-μg) disks (Becton Dickinson, Sparks, Md.) were placed onto the surfaces of the plates, and the plates were incubated in air at 35 to 37°C and read at 18 to 24 h. Slowly growing isolates, primarily members of the genus Cryptococcus, were read after 48 h of incubation. Zone diameter endpoints were read at 80% growth inhibition by using the BIOMIC image analysis plate reader system (version 5.9; Giles Scientific, Santa Barbara, Calif.) (6, 19).
The interpretive criteria for the fluconazole and voriconazole disk diffusion tests were those of the CLSI (1a,14): susceptible (S), zone diameters of 19 mm (fluconazole) and 17 mm (voriconazole); susceptible dose dependent (SDD), zone diameters of 15 to 18 mm (fluconazole) and 14 to 16 mm (voriconazole); and resistant (R), zone diameters of 14 mm (fluconazole) and 13 mm (voriconazole). The corresponding MIC breakpoints (13) are as follows: S, MIC of 8 μg/ml (fluconazole) and 1 μg/ml (voriconazole); SDD, MIC of 16 to 32 μg/ml (fluconazole) and 2 μg/ml (voriconazole); R, MIC of 64 μg/ml (fluconazole) and 4 μg/ml (voriconazole).
QC. QC was performed in accordance with CLSI document M44-A (14) by using Candida albicans ATCC 90029 and C. parapsilosis ATCC 22019. A total of 5,865 and 5,484 QC results were obtained for fluconazole and voriconazole, respectively, of which more than 99% were within the acceptable limits.
Analysis of results. All yeast disk test results were read by electronic image analysis and interpreted and recorded with a BIOMIC Plate Reader System (Giles Scientific Inc.). Test results were sent by e-mail to Giles Scientific for analysis. The zone diameter, susceptibility category (S, SDD, or R), and QC test results were all recorded electronically. In addition, MICs were calculated for each drug-organism pair by the BIOMIC System software. The MIC-versus-zone-diameter regression data used by the BIOMIC software were generated previously by ARTEMIS investigators (M.A.P. and M.G.R.) using CLSI BMD MIC and disk test methods (19, 25, 27). Patient and doctor names, duplicate test results (the same patient, the same species, and the same biotype results), and uncontrolled results were automatically eliminated by the BIOMIC system prior to analysis.
RESULTS
Isolation rates by species. A total of 140,767 yeast isolates were collected and tested at 127 study sites between June 1997 and December 2003 (Table 1). Candida species accounted for 95 to 97% of all isolates in each study year (overall, 95.7%). More than 16 different species of Candida were isolated, of which Candida albicans was the most common (overall, 66.2% of all Candida spp.). A decreasing trend in the rate of C. albicans isolation (overall decrease, 10 to 11%) was noted over the 6.5-year period. In contrast, increased rates of isolation of C. tropicalis (an increase of 2.9% from 1997 to 2003) and C. parapsilosis (an increase of 3.1% from 1997 to 2003) were noted. Neither C. glabrata nor C. krusei showed a consistent increase or decrease in isolation rate. Although isolates of more unusual Candida species, such as C. guilliermondii, C. kefyr, C. rugosa, and C. famata, constituted only a small percentage of the Candida isolates, the isolation rates of these four species increased from 2- to 10-fold over the course of the study. Likewise, although C. inconspicua, C. norvegensis, C. lipolytica, C. pelliculosa, and C. zeylanoides are rare species of Candida, the sheer size of the ARTEMIS database provides a significant number of each of these species for study.
Among the noncandidal yeasts, Cryptococcus neoformans (21% of 6,052 isolates), Saccharomyces spp. (6.8%), Trichosporon spp. (6.5%), and Rhodotorula spp. (2.3%) were the most commonly identified species (Table 1). Unidentified ("other") yeasts represented 0.46 to 3.05% of all isolates. As noted previously (6), this percentage decreased somewhat over the course of the study as more isolates were identified to the species level.
Fluconazole and voriconazole susceptibilities of Candida spp. Table 2 summarizes the in vitro susceptibilities of 78,463 and 75,787 isolates of Candida spp. to fluconazole and voriconazole, respectively, as determined by CLSI disk diffusion testing. These isolates were obtained from 115 institutions in 35 countries during the period 2001 through 2003. The distribution of zone diameters and their respective interpretive categories are shown in Fig. 1 for both agents. The percentages of isolates in each category (S, SDD, and R) were 89.6%, 4.0%, and 6.4% and 94.6%, 2.3%, and 3.1% for fluconazole and voriconazole, respectively. Fluconazole was most active against C. albicans (97.8% S), C. parapsilosis (93.2% S), C. lusitaniae (93.3% S), C. kefyr (95.3% S), C. dubliniensis (96.8% S), and C. pelliculosa (94.7% S). Decreased susceptibility to fluconazole was seen with C. glabrata (66.7% S; 16.6% R), C. krusei (9.4% S; 77.2% R), C. guilliermondii (73.3% S; 9.8% R), C. rugosa (39.3% S; 51.8% R), C. famata (79.8% S; 11.9% R), C. inconspicua (25.7% S; 49.2% R), C. norvegensis (50.0% S; 38.0% R), C. lipolytica (54.7% S; 39.6% R), and C. zeylanoides (54.1% S; 37.8% R). These findings confirm previously reported data for the more common species (e.g., C. albicans, C. glabrata, C. parapsilosis, and C. krusei) and markedly expand our understanding of the susceptibility, or lack thereof, of less common species, such as C. rugosa, C. inconspicua, and C. norvegensis, to fluconazole (5, 15, 16, 18, 21, 23).
Voriconazole was significantly more active than fluconazole against virtually every species, with the exception of C. tropicalis (89.1% S to fluconazole versus 87.1% S to voriconazole) (Table 2). Among the species with decreased susceptibility to fluconazole, more than 80% were susceptible to voriconazole, including C. glabrata (81.7% S), C. krusei (83.2% S), C. guilliermondii (91.2% S), C. famata (89.5% S), C. inconspicua (89.2% S), and C. norvegensis (92.3% S). Among the fluconazole-resistant (zone diameter, 14 mm) isolates of C. glabrata, 30% remained susceptible (zone diameter, 17 mm) to voriconazole; however, all voriconazole-resistant strains were also resistant to fluconazole (reference 22 and data not shown). Although voriconazole was more active than fluconazole against C. rugosa (61.4% S versus 39.3% S, respectively), C. lipolytica (67.3% S versus 54.7% S, respectively), and C. zeylanoides (74.3% S versus 54.1% S, respectively), these species were markedly less susceptible and more resistant (11.4% to 26.4%) to voriconazole than all other species of Candida. Again, these data confirm and extend previous observations, especially with the less common species of Candida (18, 20, 23, 24). Importantly, it is readily apparent from these data that although some degree of cross-resistance may be seen between fluconazole and voriconazole, it varies by species and should not be assumed in the absence of species identification and susceptibility testing results.
Trends in resistance to fluconazole among Candida spp. over a 6.5-year period. The longitudinal nature of the ARTEMISDISK Surveillance Program allows one to examine trends in fluconazole resistance among clinical isolates of Candida spp. with the important advantage of sufficient numbers of isolates of each species, all tested by a single standardized method (Table 3). Among the 10 species listed in Table 3, no consistent increase or decrease in fluconazole resistance was seen over time with C. albicans (range, 0.8% to 1.5%) or C. glabrata (range, 14.3% to 22.8%). Although resistance among C. tropicalis isolates appeared to decline from 1997-1998 (4.2%) thru 2001 (3.0%), increases were seen in 2002 (6.6%) and 2003 (5.0%). A slight increase in resistance was noted over time among C. parapsilosis and C. kefyr, whereas a major increase in resistance was detected among isolates of C. rugosa, where 61.2 to 66.0% resistance was observed in the last 2 years of data collection. In contrast, following a peak of 26.1% R in 2000, resistance among isolates of C. guilliermondii decreased steadily between 2001 (11.7% R) and 2003 (8.1% R). Although C. famata appeared to be quite resistant to fluconazole in 1997 and 1998 (47.4% of 19 isolates), this was likely due to the small number of isolates tested. As the numbers of C. famata isolates increased to >50 per year over the next 5 years, the level of resistance stabilized at 10 to 12%. Despite the increase in the overall percentage of isolates of C. krusei that tested as resistant to fluconazole, this is not an important finding, as the species must be considered to be clinically resistant to fluconazole. The CLSI recommends that C. krusei not be tested against fluconazole (13, 14). All such isolates should be reported as fluconazole resistant.
Trends in resistance to voriconazole among Candida spp., 2001 to 2003. Voriconazole has been used clinically since 2001 and since that time has been tested against Candida in the ARTEMIS Global Surveillance Program (Table 4). Overall, there has been a slight increase in the percentage of Candida isolates that appear to be resistant (zone diameter, 13 mm) to voriconazole, from 2.6% in 2001 to 3.5% in 2003. This may be accounted for by increases in resistance observed with C. glabrata (9.8% to 11.0%), C. tropicalis (4.7% to 7.0%), C. rugosa (3.1 to 38.0%), C. lipolytica (7.7% to 12.0%), and unidentified Candida species (4.2% to 7.0%). In contrast, no change or a decrease in resistance was seen with C. albicans, C. parapsilosis, C. krusei, C. lusitaniae, C. kefyr, C. famata, C. inconspicua, C. dubliniensis, and C. pelliculosa. Thus, the picture for voriconazole, in terms of spectrum and potency versus Candida spp., looks quite favorable. Emerging resistance, especially among C. glabrata, C. tropicalis, and C. rugosa, bears close monitoring.
Geographic variation in the susceptibilities of C. albicans and C. glabrata to fluconazole and voriconazole. Table 5presents the in vitro susceptibility results for fluconazole and voriconazole tested against the two most common species of Candida, C. albicans and C. glabrata, stratified by geographic region and country of origin for the time period 2001 to 2003. With the exception of those from India, isolates of C. albicans were highly susceptible to both fluconazole and voriconazole. The only other countries where the percentages of C. albicans susceptible to either agent dropped below 94% were Colombia (fluconazole, 91.2% S, 6.1% R) and Ecuador (fluconazole, 91.6% S, 4.9% R). Overall, there was no meaningful difference in the fluconazole or voriconazole susceptibility profile for C. albicans when stratified by specimen type (96.7% to 99.3% S to fluconazole; 97.9% to 99.3% S to voriconazole) or by hospital location (95.3 to 99.1% S to fluconazole; 97.2% to 99.4% S to voriconazole) (data not shown).
Fluconazole and voriconazole susceptibilities of C. glabrata isolates varied considerably among the various countries and geographic regions. Susceptibilities to fluconazole were lowest (<50%) in Venezuela (29.2% S), Malaysia (34.0% S), Belgium (39.7% S), the Czech Republic (44.8% S), and South Africa (49.6%) and highest (>80%) in India and the Middle East (100% S), Brazil (94.9% S), Greece (93.9% S), Canada (90.6% S), Portugal (87.1% S), Mexico (86.7% S), Poland (86.4% S), South Korea (83.7% S), Turkey (82.4% S), and Italy (81.3% S). Overall rates of resistance to fluconazole among C. glabrata isolates were 10.6% in the Asia-Pacific region, 13.2% in Latin America, 16.5% in Europe, and 18.0% in North America (data not shown). These rates of fluconazole resistance are considerably higher for each geographic region than those reported previously for blood and normally sterile-site infection isolates of C. glabrata (range, 2 to 9% R) tested by BMD between 1992 and 2000 (20).
In contrast to that seen with C. albicans, the susceptibility of C. glabrata isolates to fluconazole varied according to specimen type and hospital location. Isolates from blood and normally sterile sites were the most susceptible (71% S; 14.8% R) and genital tract isolates were the least susceptible (53.6% S; 21.2% R) to fluconazole (data not shown). The highest rates of resistance were seen in isolates of C. glabrata from the surgical intensive-care unit (21.3%), the obstetrics and gynecology service (21.5%), the hematology/oncology service (22.6%), and the neonatal intensive-care unit (35.0%) (data not shown).
Voriconazole was equally or more active than fluconazole against C. glabrata isolates from all countries and geographic regions (Table 5). Susceptibilities to voriconazole were lowest (<70%) in Venezuela (32.7% S), Belgium (53.2% S), Malaysia (59.5% S), the Czech Republic (65.3% S), and Ecuador (66.7% S) and highest (>90%) in India, Turkey and the Middle East (100% S), Brazil (96.8% S), Canada (95.7% S), Greece (95.5%), Thailand (92.5%), and Portugal (90.0%). Overall rates of resistance to voriconazole among C. glabrata isolates were 4.1% in the Asia-Pacific region, 5.4% in Latin America, 5.6% in Europe, and 9.0% in North America (data not shown). Our previous results using BMD MIC testing found resistance rates of 2.2 to 5.4% among blood and normally sterile-site isolates of C. glabrata tested in 2001 and 2002 (22). Similar to that seen with C. albicans, there was little variation in the susceptibility of C. glabrata to voriconazole when stratified by specimen type. Isolates from blood and normally sterile sites were the most susceptible (81%) and genital tract isolates were the least susceptible (70%) to voriconazole (data not shown). The rates of resistance to voriconazole ranged from 2.5% (neonatal intensive-care unit) to 8.2% (hematology/oncology service) across the different hospital locations.
Activities of fluconazole and voriconazole against other opportunistic yeasts and yeast-like fungi. Although they comprise only 3 to 5% of all of the isolates tested in this study, the number of noncandidal yeasts tested against fluconazole and voriconazole exceeds that published in the current literature (1, 3, 21, 26). Lack of standardized methods for testing most of these fungi may be considered problematic; however, the vast majority grew well on the MH-MB agar plates, and the zone diameters were easily determined. For the purposes of this study, we utilized the interpretive breakpoints for Candida, and we recognize that they may be adjusted for noncandidal yeasts in the future. Nevertheless, the data generated for these organisms are not dissimilar to those obtained using CLSI BMD MIC methods (1, 3, 21, 26). Using Cryptococcus neoformans as an example, the susceptibilities of the isolates shown in Table 6 indicated moderate susceptibility to fluconazole and a very high level of activity for voriconazole. Very similar findings for these two agents using BMD MIC methods were recently reported from our laboratory (26). As noted previously (21), most of these noncandidal yeasts were substantially less susceptible to both fluconazole and voriconazole than Candida species. Although voriconazole was more active than fluconazole for each of these different genera, it is notable that less than 80% of Trichosporon beigelii/Trichosporon cutaneum, Trichosporon asahi, and Rhodotorula spp. were susceptible to either of these agents. The diverse array of opportunistic yeasts and yeast-like fungi and their variable susceptibilities to these azole antifungals emphasize the need for prompt identification of noncandidal yeasts from clinical material. The flexibility of the CLSI disk diffusion method may well be an advantage in assessing the antifungal susceptibilities of these "emerging" pathogens.
Conversion of zone diameters to MICs. In addition to using image analysis technology to measure and record the zones of inhibition surrounding an antifungal disk, the BIOMIC system uses previously developed scatter plots and regression analysis to calculate MICs based on the relationship between the zone diameter and the MIC (Fig. 2). The data in Fig. 2 show the correlation between the MIC and the zone diameter for voriconazole with Candida spp. As seen previously with fluconazole (6, 19, 25), an excellent correlation was observed. Based on these data, the voriconazole MICs for Candida spp. were calculated and the data were compared to BMD MICs published previously (24) for the same species (Table 7). Although the numbers of isolates tested are considerably different in the two groups, it is readily apparent that the MIC50 and MIC90 values are very close for each species, as is the percent resistant. Thus, the large amount of qualitative disk diffusion data presented here can be converted to quantitative MIC data for purposes of comparing the activities of fluconazole and voriconazole for individual species (Fig. 3) or potentially for following trends across time. Additional work in this area is warranted.
DISCUSSION
The ARTEMIS Global Antifungal Surveillance Program is the largest and most comprehensive program of its kind and the only one to incorporate many of the features that arguably constitute an "ideal" resistance surveillance program (7-9, 11, 30). It is longitudinal and global, employs standardized methods used for "routine" testing in participating laboratories and for "reference" testing in a central reference laboratory, uses electronic data capture and storage in a central database, and conducts external validation of the data generated by participating laboratories. The current report from the ARTEMIS DISK Surveillance Study includes more than 140,000 opportunistic yeast isolates and is by far the largest and most geographically diverse study of antifungal susceptibility and resistance to date (5, 15, 16, 20, 28). Important findings regarding species distribution include a steady decrease in the isolation of C. albicans and an increase in the isolation of C. tropicalis and C. parapsilosis. Although they are still rare, it appears that C. rugosa, C. famata, C. inconspicua, and C. norvegensis may be "emerging" in recent years. Among the noncandidal yeasts, Cryptococcus neoformans, Saccharomyces, Trichosporon, and Rhodotorula species are prominent and may prove to be important due to their decreased susceptibilities to several antifungal agents (21).
Despite the use of a standard protocol, it is recognized that any surveillance program based on susceptibility tests performed by the participating laboratories needs to include some measure of quality assurance, beyond simple QC testing, in order to provide an independent assessment of laboratory performance and validation of the results generated by the various laboratories (7, 9, 31). One approach to cross-validation that has been suggested is to use centralized testing with high-quality microbiology to confirm the trends in routine data obtained from participating sentinel sites (7-9, 11, 30). Comparison of results obtained for isolates tested in participating laboratories with results obtained for the same organisms tested in a central reference laboratory would accomplish this goal (8, 9). This approach has been used to validate and support the epidemiologic relevance of findings from antibacterial surveillance programs (9, 30). Most recently, we have used the same approach to validate fluconazole and voriconazole disk test results generated by laboratories participating in the ARTEMIS Program (25, 27). More than 2,900 isolates of Candida obtained from blood and normally sterile-site infections were tested against fluconazole and voriconazole by ARTEMIS participating laboratories (CLSI disk test) and by the central reference laboratory (CLSI disk and BMD MIC tests) (25, 27). Categorical agreement between the reference MIC results and the disk diffusion test results performed in the participant laboratories was 87.4% and 94.1% for fluconazole and voriconazole, respectively (Table 8). A similar level of agreement was seen when the disk test results obtained in the reference laboratory were compared with those from the participant laboratories (references 25 and 27 and data not shown). It was noted that participating laboratories tended to err on the side of calling isolates more resistant than the reference laboratory did; however, the numbers of major and very major discrepancies were quite small (Table 8). This external quality assurance data, coupled with excellent QC performance, ensures the generation of accurate and useful surveillance data in the ARTEMIS DISK Surveillance Program.
The data reported here for the more common species of Candida (i.e., C. albicans, C. glabrata, C. parapsilosis, and C. tropicalis) confirm most of the previously published data regarding their susceptibilities to fluconazole and voriconazole (5, 16, 20, 24). The activity of fluconazole remains high against C. albicans, C. parapsilosis, and C. tropicalis, although resistance may be increasing among C. tropicalis isolates. Fluconazole resistance was considerable among isolates of C. glabrata, although the extent of resistance varied widely throughout the world. Fortunately, voriconazole remains quite active against this species. It is notable, however, that resistance to voriconazole has increased among C. glabrata isolates over the 3-year period of this study and was quite high in certain countries, such as Belgium (18.5%) and Venezuela (38.8%), where fluconazole resistance was also widespread. Again, our previous studies have shown that compared to reference laboratory testing of C. glabrata by MIC and disk methods, the fluconazole and voriconazole disk test results reported by ARTEMIS participating sites tended to overestimate resistance (25, 27). Thus, the rates of resistance to fluconazole and voriconazole reported in this study for C. glabrata may be somewhat higher than previously reported in the literature. Nevertheless, the geographical and temporal comparisons and differences remain important.
The ARTEMIS database is most valuable as it pertains to the less common species of Candida (Table 2). The excellent activity of voriconazole against C. krusei was confirmed by the results from almost 2,000 clinical isolates. Similarly, the high levels of activity of both azoles against C. lusitaniae, C. kefyr, C. dubliniensis, and C. pelliculosa were clearly demonstrated, confirming previous results based on comparatively few isolates (21, 23). Equally important was the demonstration of generally poor activities of fluconazole against C. guilliermondii, C. rugosa, C. famata, C. inconspicua, C. norvegensis, C. lipolytica, and C. zeylanoides. In most instances, these findings confirm what can only be called preliminary observations (21); however, for some of these species, these constitute new data and serve to underscore the imperative to identify Candida to the species level. Although voriconazole is active against the vast majority of these rare species, it is notable that decreased susceptibility to this agent, as well as to fluconazole, is seen with C. rugosa, C. lipolytica, and C. zeylanoides. These findings are especially important for C. rugosa, as the frequency of isolation of this species appears to be increasing over time (Table 1), it has been shown to cause clusters of nosocomial infection that are poorly responsive to amphotericin B (2, 4), and it was previously considered highly susceptible to voriconazole based on results for less than 20 clinical isolates (21).
As is the case for the less common Candida species, new information for noncandidal yeasts is provided by this data set. Although the antifungal susceptibility profile of Cryptococcus neoformans is well known (1, 26), much less is known of the susceptibilities of Saccharomyces, Trichosporon, Rhodotorula, and Blastoschizomyces species to fluconazole and voriconazole (3, 21, 33). The results presented in Table 6 indicate that most of these opportunistic yeasts have decreased susceptibility to fluconazole, and although voriconazole is clearly more active than fluconazole, decreased susceptibility to that agent is also seen with certain species of Trichosporon and with Rhodotorula spp. The fact that these yeast-like fungi are also nonsusceptible to the echinocandins (they lack -1,3-D-glucan) and respond variably to amphotericin B highlights the potential for their emergence as difficult-to-treat mycotic pathogens in the future (21, 33).
Finally, the ability of the BIOMIC software to convert disk diffusion zone diameters to MICs is an important feature of the ARTEMIS surveillance program, providing quantitative data that will be valuable in trend analysis. We have extended the previous work of Hazen et al. (6) and have shown that the voriconazole MICs calculated from the disk diffusion data for Candida spp. compare very favorably to those obtained by BMD MIC testing performed centrally (Table 7).
In summary, we present a tremendous volume of data describing temporal and geographic trends in the isolation and azole susceptibilities of opportunistic yeast pathogens. The data point to the strength of azole coverage for most of these organisms but also highlight several areas where resistance may be progressing and/or previously rare species may be "emerging." The strength of the ARTEMIS Global Surveillance Program is in the overall design, incorporating standardized test methods, "routine" and centralized testing of isolates, and a broad international network of study sites providing consistent data over time. The continued efforts of this surveillance program will provide data on pathogen frequency and antifungal susceptibility on a global scale.
ACKNOWLEDGMENTS
Linda Elliott provided excellent support in the preparation of the manuscript.
The ARTEMIS DISK Surveillance Program is supported by grants from Pfizer.
We express our appreciation to all ARTEMIS participants. Participants contributing to this study included Jorge Finquelievich, Buenos Aires University, and Nora Tiraboschi, Hospital Escuela Gral., Buenos Aires, Argentina; David Ellis, Women's and Children's Hospital, North Adelaide, Australia; Dominique Frameree, CHU de Jumet, Jumet, Annemarie van den Abeele, St Lucas Campus Heilige Familie, Ghent, and Jean-Marc Senterre, Hpital de la Citadelle, Liege, Belgium; Arnaldo Colombo, Escola Paulista de Medicina, Sao Paulo, Brazil; Robert Rennie, University of Alberta Hospital, Edmonton, and Steve Sanche, Royal University Hospital, Saskatoon, Canada; Bijie Hu, Zhong Shan Hospital, Shanghai, Yingchun Xu, Peking Union Medical College Hospital, Beijing, Yingyuan Zhang, Hua Shan Hospital, Shanghai, and Nan Shan Zhong, Guangzhou Institute of Respiratory Disease, Guangzhou, China; Pilar Rivas, Inst. Nacional de Cancerología, Bogota, Angela Restrepo and Catalina Bedout, CIB, Medellin, and Ricardo Vega and Matilde Mendez, Hospital Militar Central, Bogota, Colombia; Nada Mallatova, Hospital Ceske Budejovice, Ceske, and Eva Chmelarova, Krajska Hygienicka Stanice, Ostrava, Czech Republic; Julio Ayabaca, Hospital FF. AA HG1, Quito, and Jeannete Zurita, Hospital Vozandes, Quito, Ecuador; M. Mallie, Faculte de Pharmacie, Montpellier, and E. Candolfi, Institut de Parasitologie, Strasbourg, France; W. Fegeler, Universitaet Muenster, Münster, A. Haase, RWTH Aachen, Aachen, G. Rodloff, Inst. F. Med. Mikrobiologie, Leipzig, W. Bar, Carl-Thiem Klinikum, Cottbus, and V. Czaika, Humaine Kliniken, Bad Saarow, Germany; George Petrikos, Laikon General Hospital, Athens, Greece; Erzsebet Puskás, BAZ County Institute, Miskolc, Ilona Doczi, University of Szeged, Szeged, Mestyan Gyula, Medical University of Pecs, Pecs, and Radka Nikolova, Szt Laszlo Hospital, Budapest, Hungary; Uma Banerjee, All India Institute of Medical Sciences, New Delhi, India; Nathan Keller, Sheba Medical Center, Tel Hashomer, Israel; Vivian Tullio, Universita degli Studi di Torino, Turin, Gian Carlo Schito, University of Genoa, Genoa, Giacomo Fortina, Ospedale di Novara, Novara, Gian Piero Testore, Univerrsita di Roma Tor Vergata, Rome, Domenico D'Antonio, Pescara Civil Hospital, Pescara, Giorgio Scalise, Instituto di Malattie Infettive, Ancona, Pietro Martino, Dept. di Biotechnologie, Rome, and Graziana Manno, Universita di Genova, Genova, Italy; Kee Peng, University Malaya, Kuala Lumpur, Malaysia; Celia Alpuche and Jose Santos, Hospital General de Mexico, Mexico City, Eduardo Rodriguez Noriega, Universidad de Guadalajara, Guadalajara, andMussaret Zaidi, Hospital General O'Horan, Merida, Mexico; Jacques F. G. M. Meis, Canisius Wilhemina Hospital, Nijmegen, The Netherlands; Egil Lingaas, Rikshospitalet, Oslo, Norway; Danuta Dzierzanowska, Children's Memorial Health Institute, Warsaw, and Waclaw Pawliszyn, Pracownia Bakteriologii, Cracow, Poland; Mariada Luz Martins, Inst. de Higiene e Medicina Tropical, Lisbon, Luis Albuquerque, Centro Hospitalar de Coimbra, Coimbra, Laura Rosado, Instituto Nacional de Saude, Lisbon, Rosa Velho, Hosp. da Universidade de Coimbra, Coimbra, and Jose Amorim, Hospital de Santo Antonio, Porto, Portugal; Vera N. Ilina, Novosibirsk Regional Hospital, Novosibirsk, Olga I. Kretchikova, Institute of Antimicrobial Chemotherapy, Smolensk, Galina A. Klyasova, Hematology Research Center, Moscow, Sophia M. Rozanova, City Clinical Hospital No 40, Ekaterinburg, Irina G. Multykh, Territory Center of Laboratory Diagnostics, Krasnodar, Nikolay N. Klimko, Medical Mycology Research Institute, St. Petersburg, Elena D. Agapova, Irkutsk Regional Children's Hospital, Irkutsk, and Natalya V. Dmitrieva, Oncology Research Center, Moscow, Russia; Abdul Mohsen Al-Rasheed, Riyadh Armed Forces Hospital, Riyadh, Saudi Arabia; Jan Trupl, National Cancer Center, Leon Langsadl, NUTaRCH, Alena Vaculikova, Derer University Hospital, and Hupkova Helena, St. Cyril and Metod Hospital, Bratislava, Slovak Republic; Denise Roditi, Groote Schuur Hospital, Cape Town, Anwar Hoosen, GaRankuwa Hospital, Medunsa, H. H. Crewe-Brown, Baragwanath Hospital, Johannesburg, M. N. Janse van Rensburg, Pelanomi Hospital, UOFS, Bloemfontein, and Adriano Duse, Johannesburg General Hospital, Johannesburg, South Africa; Kyungwon Lee, Yonsei University College of Medicine, and Mi-Na Kim, Asan Medical Center, Seoul, South Korea; A. del Palacio, Hospital 12 De Octobre, and Aurora Sanchez-Sousa, Hospital Ramon y Cajal, Madrid, Spain; Jacques Bille, Institute of Microbiology CHUV, Lausanne, and K. Muhlethaler, Universitat Bern, Bern, Switzerland; Shan-Chwen Chang, National Taiwan University Hospital, Taipei, and Jen-Hsien Wang, China Medical College Hospital, Taichung, Taiwan; Malai Vorachit, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand; Deniz Gur, Hacettepe University Children's Hospital, Ankara, and Volkan Korten, Marmara Medical School Hospital, Istanbul, Turkey; John Paul, Royal Sussex County Hospital, Brighton, Brian Jones, Glasgow Royal Infirmary, Glasgow, F. Kate Gould, Freeman Hospital, Newcastle, Chris Kibbler, Royal Free Hospital, London, Nigel Weightman, Friarage Hospital, Northallerton, Ian M. Gould, Aberdeen Royal Hospital, Aberdeen, Ruth Ashbee, General Infirmary, P.H.L. S, Leeds, and Rosemarie Barnes, University of Wales College of Medicine, Cardiff, United Kingdom; Jose Vazquez, Harper Hospital, Wayne State University, Detroit, Michigan; Ed Chan, Mt. Sinai Medical Center, New York, and Davise Larone, Cornell Medical Center NYPH, Ithaca, N.Y.; Ellen Jo Baron, Stanford Hospital and Clinics, Stanford, Calif.; Mahmoud A. Ghannoum, University Hospitals of Cleveland, Cleveland, Ohio; Mike Rinaldi, University of Texas Health Science Center, San Antonio, Texas; Kevin Hazen, University of Virginia Health Systems, Charlottesville, Va.; Elyse Foraker, Christiana Care, Wilmington, Del.; and Heidi Reyes, Gen del Este Domingo Luciani, and Axel Santiago, Universitario de Caracas, Caracas, Venezuela.
REFERENCES
Brandt, M. E., M. A. Pfaller, R. A. Hajjeh, R. H. Hamill, P. G. Pappas, A. L. Reingold, D. Rimland, and D. W. Warnock for the Cryptococcal Disease Active Surveillance Group. 2001. Trends in antifungal drug susceptibility of Cryptococcus neoformans isolates in the United States: 1992 to 1994 and 1996 to 1998.Antimicrob. Agents Chemother. 45:3065-3069.
CLSI.2005 . Minutes of the CLSI Antifungal Subcommittee Meeting, 2005. CLSI, Wayne, Pa.
Colombo, A. L., S. A. Melo, R. F. C. Rosas, R. Salomao, M. Briones, R. J. Hollis, S. A. Messer, and M. A. Pfaller. 2003. Outbreak of Candida rugosa candidemia: an emerging pathogen that may be refractory to amphotericin B therapy. Diagn. Microbiol. Infect. Dis. 46:253-257.
Diekema, D. J., B. Petroelje, S. A. Messer, R. J. Hollis, and M. A. Pfaller. 2005. Activities of available and investigational antifungal agents against Rhodotorula species. J. Clin. Microbiol. 43:476-478.
Dube, M. P., P. N. R. Heseltine, M. G. Rinaldi, S. Evans, and B. Zawacki. 1994. Fungemia and colonization with nystatin-resistant Candida rugosa in a burn unit. Clin. Infect. Dis. 18:77-82.
Hajjeh, R. A., A. N. Sofair, L. H. Harrison, G. M. Lyon, B. A. Arthington-Skaggs, S. A. Mirza, M. Phelan, J. Morgan, W. Lee-Yang, M. A. Ciblak, L. E. Benjamin, L. Thompson Sanza, S. Huie, S. F. Yeo, M. E. Brandt, and D. W. Warnock.2004 . Incidence of bloodstream infections due to Candida species and in vitro susceptibilities of isolates collected from 1998 to 2000 in a population-based active surveillance program. J. Clin. Microbiol. 42:1519-1527.
Hazen, K. C., E. J. Baron, A. L. Colombo, C. Girmenia, A. Sanchez-Sousa, A. del Palacio, C. de Bedout, D. L. Gibbs, and The Global Antifungal Surveillance Group.2003 . Comparison of the susceptibilities of Candida spp. to fluconazole and voriconazole in a 4-year global evaluation using disk diffusion. J. Clin. Microbiol. 41:5623-5632.
Kahlmeter, G., and D. F. J. Brown. 2002. Resistance surveillance studies—comparability of results and quality assurance of methods. J. Antimicrob. Chemother. 50:775-777.
Livermore, D. M., E. J. Threlfall, M. H. Reacher, A. P. Johnson, D. James, T. Cheasty, A. Shah, F. Warburton, A. V. Swan, J. Skinner, A. Grahm, and D. C. E. Speller. 2000. Are routine test data suitable for the surveillance of resistance Resistance rates amongst Escherchia coli from blood and CSF from 1991-1997, as assessed by routine and centralized testing. J. Antimicrob. Chemother. 45:205-211.
Livermore, D. M., A. P. Macgowan, and M. C. J. Wale. 2005. Surveillance of antimicrobial resistance: centralized surveys to validate routine data offer a practical approach. Br. Med. J. 317:614-615.
Luzzati, R., G. Amalfitano, L. Lazzarini, F. Soldori, S. Bellino, M. Solbiatic, M. C. Danzi, S. Vento, G. Todeschini, C. Vivenza, and E. Concia. 2000. Nosocomial candidemia in non-neutrophenic patients at an Italian tertiary care hospital.Eur. J. Clin. Microbiol. Infect. Dis. 19:602-607.
Magee, J. T., M. L. Heginbothom, and B. W. Mason. 2005. Finding a strategy: the case for co-operative research on resistance epidemiology. J. Antimicrob. Chemother. 55:628-633.
Meis, J., M. Petrou, J. Bille, D. Ellis, D. Gibbs, and the Global Antifungal Surveillance Group. 2000. A global evaluation of the susceptibility of Candida species to fluconazole by disk diffusion. Diagn. Microbiol. Infect. Dis. 36:215-223.
National Committee for Clinical Laboratory Standards. 2002. Reference method for broth dilution testing of yeasts. Approved standard, 2nd ed. M27-A2. National Committee for Clinical Laboratory Standards, Wayne, Pa.
National Committee for Clinical Laboratory Standards. 2004. Method for antifungal disk diffusion susceptibility testing of yeasts: approved guidance M44-A. National Committee for Clinical Laboratory Standards, Wayne, Pa.
Nguyen, M. H., J. E. Peacock, Jr., A. J. Morris, D. C. Tanner, M. L. Nguyen, D. R. Snydman, M. M. Wagener, M. G. Rinaldi, and V. L. Yu. 1996. The changing face of candidemia: emergence of non-Candida albicans species and antifungal resistance. Am. J. Med. 100:617-623.
Ostrosky-Zeichner, L., J. H. Rex, P. G. Pappas, R. J. Hamill, R. A. Larsen, H. W. Horowitz, W. G. Powderly, N. Hyslop, C. A. Kauffman, J. Cleary, J. E. Mangino, and J. Lee. 2003. Antifungal susceptibility survey of 2,000 bloodstream Candida isolates in the United States. Antimicrob. Agents Chemother. 47:3149-3154.
Pfaller, M. A., and D. J. Diekema. 2002. Role of sentinel surveillance of candidemia: trends in species distribution and antifungal susceptibility. J. Clin. Microbiol. 40:3551-3557.
Pfaller, M. A., D. J. Diekema, S. A. Messer, L. Boyken, R. J. Hollis, R. N. Jones, and the International Fungal Surveillance Participant Group.2003 . In vitro activities of voriconazole, posaconazole, and four licensed systemic antifungal agents against Candida species infrequently isolated from blood. J. Clin. Microbiol. 41:28-83.
Pfaller, M. A., D. J. Diekema, S. A. Messer, L. Boyken, and R. J. Hollis. 2003. Activites of fluconazole and voriconazole against 1,586 recent clinical isolates of Candida species determined by broth microdilution, disk diffusion, and Etest methods: report from the ARTEMIS Global Antifungal Susceptibility Program, 2001. J. Clin. Microbiol. 41:1440-1446.
Pfaller, M. A., and D. J. Diekema. 2004. Twelve years of fluconazole in clinical practice: global trends in species distribution and fluconazole susceptibility of bloodstream isolates of Candida. Clin. Microbiol. Infect. 10(Suppl. 1):11-23.
Pfaller, M. A., and D. J. Diekema. 2004. Rare and emerging opportunistic fungal pathogens: concern for resistance beyond Candida albicans and Aspergillus fumigatus. J. Clin. Microbiol. 42:4419-4431.
Pfaller, M. A., S. A. Messer, L. Boyken, S. Tendolkar, R. J. Hollis, and D. J. Diekema.2004 . Geographic variation in the susceptibilities of invasive isolates of Candida glabrata to seven systemically active antifungal agents: a global assessment from the ARTEMIS Antifungal Surveillance Program conducted in 2001 and 2002.J. Clin. Microbiol. 42:3142-3146.
Pfaller, M. A., D. J. Diekema, S. A. Messer, L. Boyken, R. J. Hollis, and R. N. Jones.2004 . In vitro susceptibilities of rare Candida bloodstream isolates to ravuconazole and three comparative antifungal agents. Diagn. Microbiol. Infect. Dis. 48:101-105.
Pfaller, M. A., S. A. Messer, L. Boyken, R. J. Hollis, C. Rice, S. Tendolkar, and D. J. Diekema.2004 . In vitro activities of voriconazole, posaconazole, and fluconazole against 4,169 clinical isolates of Candida spp. and Cryptococcus neoformans collected during 2001 and 2002 in the ARTEMIS global antifungal surveillance program.Diagn. Microbiol. Infect. Dis. 48:201-205.
Pfaller, M. A., K. C. Hazen, S. A. Messer, L. Boyken, S. Tendolkar, R. J. Hollis, and D. J. Diekema. 2004. Comparison of results of fluconazole disk diffusion testing for Candida species with results from a central reference laboratory in the ARTEMIS Global Antifungal Surveillance Program. J. Clin. Microbiol. 42:3607-3612.
Pfaller, M. A., S. A. Messer, L. Boyken, C. Rice, S. Tendolkar, R. J. Hollis, G. V. Doern, and D. J. Diekema. 2005. Global trends in the antifungal susceptibility of Cryptococcus neoformans (1990-2004). J. Clin. Microbiol. 43:2163-2167.
Pfaller, M. A., L. Boyken, S. A. Messer, S. Tendolkar, R. J. Hollis, and D. J. Diekema.2005 . Comparison of results of voriconazole disk diffusion testing for Candida species with results from a central reference laboratory in the ARTEMIS Global Antifungal Surveillance Program. J. Clin. Microbiol. 43:5208-5213.
Rees, J. R., R. W. Pinner, R. A. Hajjeh, M. R. Brandt, and A. L. Reingold.1998 . The epidemiological features of invasive mycotic infection in the San Francisco Bay Area, 1992-1993: results of a population-based laboratory active surveillance. Clin. Infect. Dis. 27:1138-1147.
Richet, H., P. Roux, C. Des Champs, Y. Esnault, A. Andremont and the French Candidemia Study Group. 2002. Candidemia in French hospitals: incidence rates and characteristics. Clin. Microbiol. Infect. 8:405-412.
Stelling, J. M., K. Travers, R. N. Jones, P. J. Turner, T. F. O'Brien, and S. B. Levy.2005 . Integrating Escherichia coli antimicrobiol susceptibility data from multiple surveillance programs. Emerg. Infect. Dis. 11:873-882.
Tenover, F. C., M. J. Mohammed, J. Stelling, T. O'Brien, and R. Williams. 2001. Ability of laboratories to detect emerging antimicrobial resistance: proficiency testing and quality control results from the World Health Organizations' external quality assurance system for antimicrobial susceptibility testing.J. Clin. Microbiol. 39:241-250.
Tortorano, A. M., J. Peman, H. Bernhardt, L. Klingspor, C. C. Kibbler, O. Faure, E. Biraghi, E. Canton, K. Zimmerman, S. Seaton, R. Grillot, and the ECMM Working Group on Candidaemia.2004 . Epidemiology of candidaemia in Europe: results of 28-month European Confederation of Medical Mycology (ECMM) hospital-based surveillance study. Eur. J. Clin. Microbiol. Infect. Dis. 23:317-322.
Walsh, T. J., A. Groll, J. Hiemenz, R. Flemming, E. Roilides, and E. Anaissie. 2004. Infections due to emerging and uncommon medically important fungal pathogens. Clin. Microbiol. Infect. 10(Suppl. 1):48-66.(M. A. Pfaller, D. J. Diek)