Trends in Antifungal Susceptibility of Candida spp. Isolated from Pediatric and Adult Patients with Bloodstream Infections: SENTRY Antimicrobial Surveillance Program, 1997 to 2000

Study design.The SENTRY Antifungal Surveillance Program was established in 1997 and has been described in detail in previous publications (3, 11, 13-15). The present report focuses on BSIs due to Candida spp. from U.S., Canadian, Latin American, and European sites. BSIs due to Candida spp. were reported from 32 medical centers in the United States, 23 medical centers in Europe, 9 medical centers in Latin America, and 7 medical centers in Canada over the 4-year period from January 1997 through December 2000. The institutions that contributed data or isolates to the study are listed in the Appendix.

Organism identification.All fungal isolates from blood cultures were identified at the participating institutions by the routine method in use at each laboratory. Upon receipt at the monitoring site, the isolates were subcultured onto potato dextrose agar (Remel, Lenexa, Kans.) and CHROMagar Candida medium (Hardy Laboratories, Santa Maria, Calif.) to ensure viability and purity. Confirmation of species identification was performed with Vitek and API products (bioMerieux, St. Louis, Mo.), as recommended by the manufacturer, or by conventional methods, as required (22). Isolates were stored as suspensions in water or on agar slants at ambient temperature until needed.

Susceptibility testing.Antifungal susceptibility testing of isolates of Candida spp. was performed by the reference broth microdilution method described by the National Committee for Clinical Laboratory Standards (NCCLS) (8). The susceptibilities of the isolates to amphotericin B were determined by use of the Etest (AB BIODISK, Solna, Sweden) and RPMI 1640 agar with 2% glucose (Remel), as described previously (12). Standard powders of fluconazole (Pfizer, Inc., New York, N.Y.), voriconazole (Pfizer), ravuconazole (Bristol-Myers Squibb, Wallingford, Conn.), posaconazole (Schering-Plough, Kenilworth, N.J.), itraconazole (Janssen, Beerse, Belgium), and amphotericin B and flucytosine (Sigma, St. Louis, Mo.) were obtained from the respective manufacturers. The MICs of fluconazole, voriconazole, ravuconazole, posaconazole, itraconazole, and flucytosine were the lowest concentrations at which a prominent decrease (approximately 50%) in turbidity relative to that in the growth control well was observed (8). Amphotericin B MICs determined by the Etest were read after 48 h of incubation at 35°C and were determined to be the concentration at which 100% inhibition of growth occurred, which on the Etest strip was the location where the border of the elliptical inhibition zone intercepted the scale on the strip edge (12). Quality control was ensured by testing the strains recommended by NCCLS (8): Candida krusei ATCC 6258 and C. parapsilosis ATCC 22019.

Interpretive criteria for susceptibility to fluconazole (MIC, ≤8 μg/ml), itraconazole (MIC, ≤0.12 μg/ml), and flucytosine (MIC, ≤4 μg/ml) were those published by Rex et al. (17) and NCCLS (8). The investigational triazoles posaconazole, voriconazole and ravuconazole have not been assigned interpretive breakpoints. For purposes of comparison and because preliminary pharmacokinetic data indicate that the levels of these agents achievable in serum may range from 2 to 6 μg/ml, depending on the dosing regimen (20), we have used a breakpoint for susceptibility of ≤1 μg/ml for all three agents. Interpretive criteria have not yet been defined for amphotericin B; however for comparison purposes in this surveillance study we have determined isolates that were inhibited by ≤1 μg/ml to be susceptible.

Statistical methods.Comparison of the species distributions or MIC distributions in terms of other factors were made by the chi-square test for categorical variables and the Wilcoxon rank sum text for ordinal variables (MICs). All reported P values are two-tailed.

Frequency of occurrence of bloodstream pathogens.During the 4-year study period (January 1997 to December 2000), a total of 2,047 BSIs due to Candida spp. were reported by SENTRY participants. Table 1 compares the frequencies of occurrence of the seven most commonly isolated pathogens during the time period of the study. These seven species accounted for 99% of all Candida BSIs reported from SENTRY study sites. Overall, the rank order of the top five species was unchanged from year to year. These five species (C. albicans, C. glabrata, C. parapsilosis, C. tropicalis, and C. krusei) accounted for 97% of all Candida BSIs. As described previously (15), the only notable geographic difference in the species distribution among the three continents was a higher frequency of C. glabrata as a cause of BSIs in the United States.

Comparing the frequency of isolation of different species by age group, we found identical rank orders for the groups 16 to 64 and ≥65 years of age: C. albicans > C. glabrata > C. parapsilosis > C. tropicalis > C. krusei. The rank order of the frequency of isolation of different species for the groups ≤1 and 2 to 15 years of age was considerably different from that for the two older age groups. The dominant causes of infections in individuals in the infant and pediatric age groups were C. albicans and C. parapsilosis, and very few infections were due to C. glabrata and C. krusei. Although C. albicans was the most common species in all age groups, the proportions of BSIs due to this species decreased from 60% in the group ≤1 year of age to 50% in the group ≥65 years of age (P = 0.007). C. glabrata was the second most common species overall, causing 17 to 23% of BSIs in the groups ages 16 to 64 and ≥65 years. In contrast, C. glabrata accounted for only 3% of BSIs in the groups ≤1 year and 2 to 15 years of age (P < 0.001 compared to the groups 16 to 64 and ≥65 years of age) and was surpassed by both C. parapsilosis (which was the cause of 21 to 24% of BSIs) and C. tropicalis (which was the cause of 7 to 10% of BSIs) as causes of infection.

Differences in susceptibilities among isolates from individuals in different age groups to both licensed and investigational antifungal agents.Among the licensed antifungal agents fluconazole, itraconazole, amphotericin B, and flucytosine, the susceptibilities of isolates of Candida spp. with increasing patient age (P ≤ 0.01 for trend). Although flucytosine was active overall against isolates from individuals in all age groups (with 95 to 98% of isolates being susceptible), isolates of C. tropicalis from patients ≤1 year of age were more susceptible to this agent (MIC at which 90% of isolates are inhibited [MIC₉₀], 0.5 μg/ml; 0% resistant isolates) than those isolates from patients ≥65 years of age (MIC₉₀, 32 μg/ml; 11% resistant isolates) (data not shown).

Among the licensed and investigational azoles, fluconazole, posaconazole, ravuconazole, and voriconazole were all highly active (96 to 100% of isolates were susceptible) against C. albicans, C. tropicalis, C. guilliermondii, and C. lusitaniae, regardless of the age group (Table 2). Although fluconazole appeared to be less active against isolates of C. glabrata in those ≤1 year of age (63% of isolates were susceptible) than against isolates from those 16 to 64 years of age (70% of isolates were susceptible) and ≥65 years of age (73% of isolates were susceptible), there were no highly resistant (MIC, ≥64 μg/ml) isolates in those ≤1 year of age, whereas 5 to 9% of isolates from the two older groups were highly resistant. Itraconazole was less active than all of the other azoles tested against all species with the exception of C. albicans. Consistent with previous reports (3, 10, 13, 15), all three investigational triazoles were highly active against all species from all age groups with the exception of C. glabrata isolates from those in the group 16 to 64 years of age, with 88 to 90% of C. glabrata isolates from individuals in that group being susceptible to the three investigational triazoles at ≤1 μg/ml but with 95 to 100% of C. glabrata isolates from all other age groups being susceptible to the three agents.

The activities of amphotericin B, as determined by the Etest, against the five species that were the most common causes of BSIs are shown in Table 3. As expected, amphotericin B was the most active agent against C. albicans (MIC₉₀, 1 μg/ml). The activities of amphotericin B against the other species were significantly less (P < 0.001) than that observed against C. albicans: MIC₉₀ for C. glabrata, 4 μg/ml; MIC₉₀ for C. parapsilosis, 4 μg/ml; MIC₉₀ for C. krusei, 8 μg/ml. Although in our hands the Etest tended to give higher amphotericin B MICs for the non-C. albicans species than for C. albicans, the same tendency was seen with the NCCLS microdilution method when it was applied to a subset of 1,077 isolates (data not shown). Notably, 17.4% of C. glabrata isolates and 27.8% of C. krusei isolates appeared to be resistant (MICs, ≥2 μg/ml) when they were tested by the NCCLS method. As reported previously (12), the level of agreement between the results of the Etest and those of the broth microdilution method was excellent (98.7% within 2 log₂ dilutions).

It appears that the differences in susceptibility to amphotericin B and the azoles among the isolates from individuals in four age groups may be largely due to differences in the species distributions among individuals in the younger age groups compared to those among individuals in the older age groups. The groups ≤1 year and 2 to 15 years of age were predominantly infected with C. albicans and C. parapsilosis, both of which were considerably more susceptible than C. glabrata to the antifungal agents tested. Individuals in the older age groups, and particularly those ≥65 years old, had relatively fewer infections due to C. albicans and significantly more infections due to C. glabrata than individuals in the younger age groups. The decreased susceptibilities of C. glabrata to both the azoles and amphotericin B are evident in Tables 2 and 3 and accounted for the differences in susceptibilities among the isolates from individuals in the different age groups.

The predominance of C. albicans and C. parapsilosis as the etiologic agents of fungemia in neonates has previously been noted by Kao et al. (5) and Saiman et al. (19), among others; however, the lack of infections with C. glabrata and C. krusei in this age group is less well appreciated. The increasing importance of C. glabrata in the adult population is well known, and the proportion of BSIs due to C. glabrata has been noted to be considerably higher in individuals >60 years of age in some institutions (6, 15). The reasons for these differences are speculative but may relate to the tendency for neonatologists and pediatricians to use amphotericin B preferentially over fluconazole in the treatment of documented or suspected candidemia (18) and the pervasive use of fluconazole in the adult hospital population (1). Furthermore, vertical transmission of C. albicans from mother to infant and horizontal transmission of C. parapsilosis from patient to patient in the neonatal intensive care unit environment are well documented and may also account for the predominance of these two species in infants and children (9, 21).

Summary and conclusions.Antifungal resistance surveillance programs provide important information both for the development of recommendations for empirical antifungal therapy and for the design of programs for the control of antifungal resistance. The present study demonstrates differences in the spectrum of pathogens and in antifungal susceptibilities among isolates from individuals in the pediatric and adult age groups in the SENTRY Program. The most notable trends were those of a decreasing frequency of occurrence of C. albicans and an increasing frequency of occurrence of C. glabrata with increasing patient age.

Decreased susceptibilities to fluconazole, itraconazole, and amphotericin B were most prominent among C. glabrata and C. krusei isolates from individuals in all age groups. Ongoing surveillance is essential and will enhance efforts to limit the extent of resistance among isolates from individuals in the various age groups.