- Email: [email protected]
Candida glabrata fungaemia in a tertiary centre in Taiwan: antifungal susceptibility and outcomes
International Journal of Antimicrobial Agents 34 (2009) 236–239
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
Candida glabrata fungaemia in a tertiary centre in Taiwan: antifungal susceptibility and outcomes Sheng-Yuan Ruan a , Yu-Tsung Huang b , Chen-Chen Chu c , Chong-Jen Yu d , Po-Ren Hsueh b,d,∗ a
Department of Internal Medicine, National Taiwan University Hospital Yun-Lin Branch, Yun-Lin, Taiwan Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, Taiwan Department of Pharmacology, National Taiwan University Hospital, Taipei, Taiwan d Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan b c
a r t i c l e
i n f o
Article history: Received 18 November 2008 Accepted 13 February 2009 Keywords: Antifungal susceptibility Candida glabrata Candidaemia Fungaemia Outcome
a b s t r a c t The proportion of non-albicans candidaemia has increased during recent decades, especially Candida glabrata. We evaluated the antifungal susceptibility, clinical features and outcome of C. glabrata fungaemia treated in a tertiary centre in Taiwan. All episodes of C. glabrata fungaemia during 1999–2005 were identified from microbiology laboratory records and all C. glabrata isolates were subjected to antifungal susceptibility testing by the broth microdilution method. A total of 177 episodes of C. glabrata fungaemia were documented, accounting for 30% of the 598 episodes of candidaemia. A dramatic decline of C. glabrata causing candidaemia from 2003 (46.8%) to 2005 (15.8%) was noted, accompanied by decreased fluconazole consumption. The most common underlying diseases in these patients were cancer (49%), diabetes (34%) and renal failure (25%). The most common risk factors were central venous catheter use (88%), antimicrobial treatment (87%) and parenteral nutrition (51%). The 30-day all-cause mortality was 48.6%, but only 31% of patients were eventually discharged from the hospital. There was no significant survival difference between patients with C. glabrata and Candida albicans fungaemia. Rates of antifungal susceptibility were 63% for fluconazole, 93% for voriconazole, 96% for caspofungin, 98% for amphotericin B and 99% for flucytosine. The different levels of susceptibility to fluconazole (susceptible, susceptible-dose dependent and resistant) were not significantly associated with 30-day mortality. © 2009 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction A global trend of increasing incidence of invasive candidiasis and candidaemia has occurred in the past two decades [1–3]. A trend of an increasing proportion of non-albicans species in Candida bloodstream infection has also been reported [4,5], although it remains controversial [6,7]. The most important non-albicans Candida species in North America and some other regions is Candida glabrata [8], accounting for 10–20% of candidaemias [8,9]. Candida glabrata is unique to other Candia species in biology and genetics, with a smaller morphology and having a haploid genome [10]. Despite these differences between C. glabrata and other Candida spp., the clinical features of C. glabrata fungaemia are often studied in the more general context of candidaemias and few studies have focused solely on C. glabrata fungaemia. In this study, clinical data and treatment outcomes of C. glabrata fungaemia at National Taiwan University Hospital over a 7-year period from 1999 to 2005 were retrospectively analysed. In addition, all C. glabrata isolates
during this period were analysed to determine their antifungal susceptibility. 2. Materials and methods 2.1. Patients and setting All episodes of C. glabrata fungaemia during 1999–2005 were identified from the records of the microbiology laboratory at National Taiwan University Hospital, a 2000-bed, universityaffiliated medical centre in Taiwan. This hospital provides both primary and tertiary medical care and has ca. 200 Intensive Care Unit (ICU) beds, including medical, surgical and paediatric ICUs. All C. glabrata isolates during this period had been stored in the mycology laboratory at −70 ◦ C in trypticase soy broth supplemented with 15% glycerol. Each isolate was retrieved from storage for antifungal susceptibility testing. The medical records of each episode of C. glabrata fungaemia were retrospectively reviewed. 2.2. Definitions
∗ Corresponding author. Tel.: +886 2 2312 3456x5355; fax: +886 2 2322 4263. E-mail address: [email protected] (P.-R. Hsueh).
Candida glabrata fungaemia was defined as at least one blood culture positive for C. glabrata. Fungaemia in the same patient
0924-8579/$ – see front matter © 2009 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. doi:10.1016/j.ijantimicag.2009.02.021
S.-Y. Ruan et al. / International Journal of Antimicrobial Agents 34 (2009) 236–239
occurring during the same hospital course was counted as the same episode. The probable portals of entry were determined by medical records, which mainly reflected the judgement of the primary care clinicians. ‘Renal failure’ in the analysis of risk factors was defined as end-stage renal disease or need for dialysis. Liver cirrhosis was defined as Child-Pugh class B or C. Neutropenia was defined as an absolute neutrophil count <500 cells/L for >48 h. Nil per os (NPO) was defined as withholding of feeding for >48 h for any reason. Patients who received chemotherapy, systemic steroids or fluconazole within 30 days before the onset of fungaemia were classified as positive for each corresponding factor. Thirty-day mortality in this study included mortality due to any cause. Data on annual consumption of fluconazole (intravenous and oral forms) from 1999 to 2005 were obtained from the pharmacy department of the hospital. Antibiotic consumption was expressed as the number of defined daily doses/1000 patient-days. 2.3. Antifungal susceptibility testing Standard powders of the following antifungal agents were obtained from their respective manufacturers: fluconazole and voriconazole (Pfizer Inc., New York, NY); flucytosine (Sigma Chemical Co., St Louis, MO); caspofungin (Merck Sharp & Dohme, West Point, PA); and amphotericin B (Bristol-Myers Squibb Laboratories, Princeton, NJ). Prior to testing, isolates were passaged twice on potato dextrose agar (Difco Laboratories, Detroit, MI) at 35 ◦ C. Antifungal susceptibility testing was performed by the reference broth microdilution method of the Clinical and Laboratory Standards Institute (CLSI) [11]. The tested concentrations of the agents ranged from 0.03 g/mL to 64 g/mL. Final dilutions were made in RPMI 1640 medium (Sigma) buffered to pH 7.0 with 0.165 M morpholinepropanesulfonic acid (MOPS) buffer (Sigma). Following incubation at 35 ◦ C for 48 h, minimum inhibitory concentrations (MICs) were determined. Two reference strains, Candida parapsilosis ATCC 22019 and Candida krusei ATCC 6258, were used as the control strains for susceptibility testing. Susceptibilities to fluconazole and flucytosine were determined according to CLSI guidelines. The proposed susceptibility breakpoint of 1 g/mL was used for voriconazole and amphotericin B and 2 g/mL was used for caspofungin [12–14]. 2.4. Statistical analyses Categorical variables were analysed using Fisher’s exact test or 2 test. The probability of survival was estimated using the Kaplan–Meier method. Pearson’s correlation coefficient was used to determine the relationship between annual fluconazole consumption and trends in the percentage of C. glabrata among all Candida spp. causing fungaemia each year. An R-value greater than 0.72 (or less than −0.72) and a P-value <0.05 were considered statistically significant. 3. Results 3.1. Epidemiology During the 7-year study period (1999–2005), there were 598 episodes of candidaemia. The species distribution of these 598 episodes was 41% Candida albicans (n = 246), 30% C. glabrata (n = 177), 14% C. tropicalis (n = 82), 11% C. parapsilosis (n = 65), 2% Candida guilliermondii (n = 11) and 2% other Candida spp. Candida glabrata was the second most common aetiology of candidaemia. The proportion of C. glabrata causing all candidaemia ranged from 15.4% (1999) to 46.8% (2003) and markedly declined in 2005 (15.8%) (Fig. 1). Although Fig. 1 might suggest a relationship between fluconazole consumption and the proportion of C. glabrata among all
237
Fig. 1. Percentage of Candida glabrata causing all candidaemias and fluconazole consumption. Table 1 Characteristics of 177 episodes of Candida glabrata fungaemia and association with 30-day mortality. No. of episodes (%)a
Association with 30-day mortality (P-value)
Gender (male:female) Age (mean ± S.D.) (years)
89:88 62.9 ± 20.0
Underlying factors Use of central catheters Use of antibacterial agent Receipt of parenteral nutrition Nothing per os Prior abdominal surgery Prior fluconazole use Anticancer chemotherapy Gastrointestinal bleeding Steroid Neutropenia
156 (88) 154 (87) 91 (51) 72 (41) 60 (34) 47 (27) 41 (23) 38 (21) 30 (17) 5 (3)
0.646 0.377 0.176 0.006 0.752 0.499 0.158 0.103 0.015 0.675
Underlying diseases Malignancy Diabetes Renal failure Liver cirrhosis Autoimmune diseases HIV
87 (49) 60 (34) 45 (25) 20 (11) 10 (6) 1 (1)
0.051 0.528 0.039 0.001 0.052 1.000
S.D., standard deviation; HIV, human immunodeficiency virus. a Data are n (%) unless otherwise stated.
candidaemia, Pearson’s correlation analysis revealed no significant correlation between these variables (R = 0.57; P = 0.24). The characteristics of the 175 patients with 177 episodes of C. glabrata fungaemia are summarised in Table 1. The mean age was 63 years and 70% of patients were >60 years of age. Onset of fungaemia occurred at a median of 27 days after admission. Most of the patients had several co-morbidities and risk factors (Table 1). The most common underlying diseases were cancer (49%), diabetes (34%) and renal failure (25%). The most common risk factors were central venous catheter use (88%), antibacterial treatment (87%) and parenteral nutrition (51%). The probable portals of entry for C. glabrata fungaemia are shown in Table 2. The probable portal of entry could be determined retrospectively from the medical records in only 130 (73%) of 177 Table 2 Probable portals of entry of Candida glabrata fungaemia. Probable portals of entry (N = 130)
No. of episodes (%)
Central venous catheter Urinary tract Intra-abdominal Airway or lung Surgical wound Other
79 (61) 20 (15) 18 (14) 3 (2) 4 (3) 6 (5)
238
S.-Y. Ruan et al. / International Journal of Antimicrobial Agents 34 (2009) 236–239
Table 3 In vitro susceptibilities of Candida glabrata isolates to five antifungal agents. Antifungal
Fluconazole Voriconazole Caspofungin Flucytosine Amphotericin B
Susceptible (%)a
MIC (g/mL) Range
MIC50
MIC90
2 to >64 0.06–64 0.25 to >64 0.06 to >64 0.25–2
8 0.25 0.5 0.06 1
32 1 1 0.12 1
63 93 96 99 98
MIC, minimum inhibitory concentration; MIC50/90 , MIC at which 50% and 90% of the isolates are inhibited, respectively. a Susceptibility breakpoints were defined as ≤8 g/mL for fluconazole and ≤4 g/mL for flucytosine, and presumptively identified as ≤1 g/mL for voriconazole and amphotericin B and ≤2 g/mL for caspofungin.
fungaemic episodes. The most commonly identified portal of entry was a central venous catheter (61%), followed by the urinary tract (15%) and intra-abdominal (14%).
Fig. 2. Survival curves for 177 episodes of Candida glabrata fungaemia and 246 episodes of Candida albicans fungaemia (P = 0.331, log-rank test).
C. glabrata fungaemia, the 30-day all-cause mortality rate was 48.6% (41.1% for C. albicans fungaemia), but only 31% of patients were eventually discharged from the hospital.
3.2. Antifungal susceptibility testing 4. Discussion The MICs of the C. glabrata isolates to five antifungal agents are shown in Table 3. The susceptibility rate was 63% for fluconazole, 93% for voriconazole, 96% for caspofungin, 99% for flucytosine and 98% for amphotericin B. The proportion of isolates with fluconazole resistance (MIC > 32 g/mL) was 7% and with susceptible-dose dependence was 30% (MIC = 16–32 g/mL). The different levels of susceptibility to fluconazole (susceptible, susceptible-dose dependent and resistant) were not significantly associated with 30-day mortality (P = 0.09). 3.3. Treatment and outcomes Of the 177 C. glabrata fungaemia episodes, 21 (11.9%) were not treated with antifungal agents. Lack of antifungal treatment was most often due to death of the patient before the time of diagnosis. Of the 156 treated episodes, the mean interval between culture and the start of empirical antifungal treatment was 2.3 days, and 78% of patients were treated within 3 days. Early antifungal treatment, starting <72 h after culture, was associated with worse outcome (52% vs. 23%; P = 0.03). The treatment regimens are summarised in Table 4. The most common treatment regimen, used in 48.1% of episodes, began with fluconazole and then switched to amphotericin B owing to poor treatment response or results of susceptibility testing. The mortality rate of the various regimens was not significantly different (P = 0.85). Amphotericin Bcontaining regimens had a lower 30-day mortality (42%) compared with non-amphotericin B regimens (50%) but this difference was not significant (P = 0.34). The mortality of C. albicans and C. glabrata fungaemia was compared to determine the relative impact of C. glabrata fungaemia on patients. Fig. 2 shows the survival curves for C. glabrata and C. albicans fungaemia over the 3 months. There was no significant survival difference between patients with C. glabrata and C. albicans fungaemia (P = 0.33, log-rank test). For patients with Table 4 Antifungal regimens and associated 30-day mortality. Regimen (N = 156)
No. of episodes (%)
No. of deaths
Mortality (%)
Fluconazole then amphotericin B Fluconazole alone Fluconazole then caspofungin Amphotericin B alone Amphotericin B then fluconazole Caspofungin then amphotericin B
75 (48.1) 63 (40.4) 5 (3.2) 10 (6.4) 2 (1.3) 1 (0.6)
32 31 3 4 1 0
43 49 –a 40 – –
a
Mortality rate not shown owing to small number of cases.
Several studies have reported an increasing trend of C. glabrata in candidaemia, especially in cancer patients [15–17]. This trend was proposed to be related to the use of fluconazole prophylaxis [15,18]. However, another study found that exposure to antibacterial agents, specifically vancomycin or piperacillin/tazobactam but not fluconazole, was associated with subsequent hospital-acquired C. glabrata or C. krusei candidaemia [19]. In the present study, C. glabrata accounted for 30% of candidaemias, which is higher than the average reported value (10–20%) [8]. The annual percentage of C. glabrata causing all candidaemia is shown in Fig. 1. The reason for the dramatic decline of C. glabrata causing candidaemia from 2003 (46.8%) to 2005 (15.8%) is not clear, although fluconazole consumption also decreased during these 3 years. During the study period, fluconazole or other azoles were not routinely used as prophylactic agents for patients with haematological cancer in our hospital. The predisposing factors for C. glabrata candidaemia identified in this study (Table 1) are similar to those in previous reports [20,21]. Our findings that 30-day mortality was associated with renal failure, liver cirrhosis, steroid use and NPO are also similar to results of previous studies which found that old age, poor performance status and extensive organ involvement were poor prognostic factors for candidaemia [22–24]. The gut and skin were thought to be the two major portals of entry of candidaemia and the latter has been implicated in the pathogenesis of catheter-related candidaemia [25]. Candida glabrata can be a non-pathogenic saprophyte in the gastrointestinal tract, on the skin and in the vagina of humans [10]. Therefore, the sources of C. glabrata fungaemia should reflect this biological habit. Although the gut has been considered as the most important source of candidaemia, the supporting data were not focused on C. glabrata [25]. In the present study, the most common sources of C. glabrata fungaemia were central venous catheters (61%). Despite several drawbacks with the determination of the source of candidaemia, our findings suggest that the skin is the predominant portal of entry of C. glabrata fungaemia. More prospective studies focusing on C. glabrata are needed to confirm these findings. Compared with C. albicans, C. glabrata is relatively resistant to fluconazole [8]. Fluconazole resistance is one of the greatest concerns when managing patients with C. glabrata fungaemia. In the present study, the susceptibility rate was only 63% for fluconazole. However, fluconazole susceptibility and the choice of antifungal regimen were not correlated with mortality. Previous studies also reported that patients with candidaemia were often seriously ill and outcome appeared to be associated with underlying conditions
S.-Y. Ruan et al. / International Journal of Antimicrobial Agents 34 (2009) 236–239
rather than fluconazole susceptibility or treatment agents [26–28]. Amphotericin B is an important therapeutic agent for infection caused by azole-resistant Candida spp. The guidelines of the Infectious Diseases Society of America (IDSA) suggest treating serious C. glabrata infections with amphotericin B or high-dose fluconazole [29]. In this study, patients treated with amphotericin B-containing regimens had a lower mortality rate than those treated with nonamphotericin B regimens, but this difference was not significant (42% vs. 50%; P = 0.34). However, because the majority of patients who received amphotericin B-containing regimens were treated with fluconazole prior to amphotericin B, the lack of difference in mortality between the two regimens might have been due to patients receiving inappropriate initial therapy. Candida glabrata fungaemia is associated with poor outcome and the 30-day mortality rate ranged from 25% to 50% in previous reports [20,21,30]. In this study, the 30-day all-cause mortality rate was 48.6%, and only 31% of patients could be discharged from hospital. Early antifungal treatment, starting <72 h after culture, was associated with worse outcome (52% vs. 23%; P = 0.03). This conflicting result might be due to the bias of greater severity of underlying diseases in these patients because primary care physicians tend to early use of antifungal agents in patients with more serious conditions. However, this assumption could not be tested because of the lack of Acute Physiology and Chronic Health Evaluation (APACHE) II score or other severity score data needed to adjust for disease severity in the analysis. Recent studies have documented that delayed antifungal therapy significantly impacted mortality in candidaemic patients [31,32]. In addition, the main antifungal agent used for early treatment in our study was fluconazole and it was possibly inappropriate. This may be another reason why early treatment does not result in better outcome. Our study has several limitations. First, because of the retrospective nature of this study, determination of the fungaemia source was arbitrary and mainly relied on the clinical judgement of primary care physicians. Second, the choice of antifungal agent and the switching of antifungal agents did not follow a consistent protocol. There might be biases in the selection of antifungal agents. Third, the sample size was small when various treatment subgroups were assessed for statistically meaningful associations with mortality. In conclusion, C. glabrata was the second most common pathogen causing candidaemia in this study from Taiwan, accounting for 30% of candidaemia in a tertiary care hospital. The outcome was poor, with a 30-day all-cause mortality rate of 48.6%, and only 31% of patients were eventually discharged from hospital. There was no significant survival difference between patients with C. glabrata and C. albicans fungaemia. Fluconazole susceptibility of the isolates was not significantly associated with mortality in these patients. Funding: No funding sources. Competing interests: None declared. Ethical approval: Not required. References [1] Asmundsdottir LR, Erlendsdottir H, Gottfredsson M. Increasing incidence of candidemia: results from a 20-year nationwide study in Iceland. J Clin Microbiol 2002;40:3489–92. [2] Beck-Sague C, Jarvis WR. Secular trends in the epidemiology of nosocomial fungal infections in the United States, 1980–1990. National Nosocomial Infections Surveillance System. J Infect Dis 1993;167:1247–51. [3] Hsueh PR, Teng LJ, Yang PC, Ho SW, Luh KT. Emergence of nosocomial candidemia at a teaching hospital in Taiwan from 1981 to 2000: increased susceptibility of Candida species to fluconazole. Microb Drug Resist 2002;8: 311–9. [4] Nguyen MH, Peacock JE, Morris AJ, Tanner DC, Nguyen ML, Snydman DR, et al. The changing face of candidemia: emergence of non-Candida albicans species and antifungal resistance. Am J Med 1996;100:617–23.
239
[5] Trick WE, Fridkin SK, Edwards JR, Hajjeh RA, Gaynes RP. Secular trend of hospital-acquired candidemia among intensive care unit patients in the United States during 1989–1999. Clin Infect Dis 2002;35:627–30. [6] Sandven P, Bevanger L, Digranes A, Haukland HH, Mannsaker T, Gaustad P. Candidemia in Norway (1991 to 2003): results from a nationwide study. J Clin Microbiol 2006;44:1977–81. [7] Pfaller MA, Diekema DJ. Twelve years of fluconazole in clinical practice: global trends in species distribution and fluconazole susceptibility of bloodstream isolates of Candida. Clin Microbiol Infect 2004;10(Suppl. 1):11–23. [8] Pfaller MA, Diekema DJ. Epidemiology of invasive candidiasis: a persistent public health problem. Clin Microbiol Rev 2007;20:133–63. [9] Eggimann P, Garbino J, Pittet D. Epidemiology of Candida species infections in critically ill non-immunosuppressed patients. Lancet Infect Dis 2003;3:685–702. [10] Fidel PL, Vazquez JA, Sobel JD. Candida glabrata: review of epidemiology, pathogenesis, and clinical disease with comparison to C. albicans. Clin Microbiol Rev 1999;12:80–96. [11] National Committee for Clinical Laboratory Standards. Reference method for broth microdilution antifungal susceptibility testing of yeasts; approved standard. Document M27-A2. Wayne, PA: NCCLS; 2002. [12] Nguyen MH, Clancy CJ, Yu VL, Yu YC, Morris AJ, Snydman DR, et al. Do in vitro susceptibility data predict the microbiologic response to amphotericin B? Results of a prospective study of patients with Candida fungemia. J Infect Dis 1998;177:425–30. [13] Pfaller MA, Boyken L, Hollis RJ, Kroeger J, Messer SA, Tendolkar S, et al. In vitro susceptibility of invasive isolates of Candida spp. to anidulafungin, caspofungin, and micafungin: six years of global surveillance. J Clin Microbiol 2008;46:150–6. [14] Pfaller MA, Diekema DJ, Rex JH, Espinel-Ingroff A, Johnson EM, Andes D, et al. Correlation of MIC with outcome for Candida species tested against voriconazole: analysis and proposal for interpretive breakpoints. J Clin Microbiol 2006;44:819–26. [15] AbiSaid D, Anaissie E, Uzun O, Raad I, Pinzcowski H, Vartivarian S. The epidemiology of hematogenous candidiasis caused by different Candida species. Clin Infect Dis 1997;24:1122–8. [16] Safdar A, Chaturvedi V, Cross EW, Park S, Bernard EM, Armstrong D, et al. Prospective study of Candida species in patients at a comprehensive cancer center. Antimicrob Agents Chemother 2001;45:2129–33. [17] Hachem R, Hanna H, Kontoyiannis D, Jiang Y, Raad I. The changing epidemiology of invasive candidiasis: Candida glabrata and Candida krusei as the leading causes of candidemia in hematologic malignancy. Cancer 2008;112:2493–9. [18] Marr KA, Seidel K, White TC, Bowden RA. Candidemia in allogeneic blood and marrow transplant recipients: evolution of risk factors after the adoption of prophylactic fluconazole. J Infect Dis 2000;181:309–16. [19] Lin MY, Carmeli Y, Zumsteg J, Flores EL, Tolentino J, Sreeramoju P, et al. Prior antimicrobial therapy and risk for hospital-acquired Candida glabrata and Candida krusei fungemia: a case–case–control study. Antimicrob Agents Chemother 2005;49:4555–60. [20] Gumbo T, Isada CM, Hall G, Karafa MT, Gordon SM. Candida glabrata fungemia—clinical features of 139 patients. Medicine 1999;78:220–7. [21] Malani A, Hmoud J, Chiu L, Carver PL, Bielaczyc A, Kauffman CA. Candida glabrata fungemia: experience in a tertiary care center. Clin Infect Dis 2005;41:975–81. [22] Boo TW, O’Reilly B, O’Leary J, Cryan B. Candidaemia in an Irish tertiary referral hospital: epidemiology and prognostic factors. Mycoses 2005;48:251–9. [23] Uzun O, Anaissie EJ. Predictors of outcome in cancer patients with candidemia. Ann Oncol 2000;11:1517–21. [24] Weinberger M, Leibovici L, Perez S, Samra Z, Ostfeld I, Levi I, et al. Characteristics of candidaemia with Candida albicans compared with non-albicans Candida species and predictors of mortality. J Hosp Infect 2005;61:146–54. [25] Nucci M, Anaissie E. Revisiting the source of candidemia: skin or gut? Clin Infect Dis 2001;33:1959–67. [26] Antoniadou A, Torres HA, Lewis RE, Thornby J, Bodey GP, Tarrand JJ, et al. Candidemia in a tertiary care cancer center—in vitro susceptibility and its association with outcome of initial antifungal therapy. Medicine 2003;82:309–21. [27] Pappas PG, Rex JH, Lee J, Hamill RJ, Larsen RA, Powderly W, et al. A prospective observational study of candidemia: epidemiology, therapy, and influences on mortality in hospitalized adult and pediatric patients. Clin Infect Dis 2003;37:634–43. [28] Viudes A, Peman J, Canton E, Ubeda P, Lopez-Ribot JL, Gobernado M. Candidemia at a tertiary-care hospital: epidemiology, treatment, clinical outcome and risk factors for death. Eur J Clin Microbiol Infect Dis 2002;21:767–74. [29] Pappas PG, Rex JH, Sobel JD, Filler SG, Dismukes WE, Walsh TJ, et al. Guidelines for treatment of candidiasis. Clin Infect Dis 2004;38:161–89. [30] Wilson AG, Micek ST, Ritchie DJ. A retrospective evaluation of fluconazole for the treatment of Candida glabrata fungemia. Clin Ther 2005;27:1228–37. [31] Garey KW, Rege M, Pai MP, Mingo DE, Suda KJ, Turpin RS, et al. Time to initiation of fluconazole therapy impacts mortality in patients with candidemia: a multiinstitutional study. Clin Infect Dis 2006;43:25–31. [32] Morrell M, Fraser VJ, Kollef MH. Delaying the empiric treatment of Candida bloodstream infection until positive blood culture results are obtained: a potential risk factor for hospital mortality. Antimicrob Agents Chemother 2005;49:3640–5.
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
Candida glabrata fungaemia in a tertiary centre in Taiwan: antifungal susceptibility and outcomes Sheng-Yuan Ruan a , Yu-Tsung Huang b , Chen-Chen Chu c , Chong-Jen Yu d , Po-Ren Hsueh b,d,∗ a
Department of Internal Medicine, National Taiwan University Hospital Yun-Lin Branch, Yun-Lin, Taiwan Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, Taiwan Department of Pharmacology, National Taiwan University Hospital, Taipei, Taiwan d Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan b c
a r t i c l e
i n f o
Article history: Received 18 November 2008 Accepted 13 February 2009 Keywords: Antifungal susceptibility Candida glabrata Candidaemia Fungaemia Outcome
a b s t r a c t The proportion of non-albicans candidaemia has increased during recent decades, especially Candida glabrata. We evaluated the antifungal susceptibility, clinical features and outcome of C. glabrata fungaemia treated in a tertiary centre in Taiwan. All episodes of C. glabrata fungaemia during 1999–2005 were identified from microbiology laboratory records and all C. glabrata isolates were subjected to antifungal susceptibility testing by the broth microdilution method. A total of 177 episodes of C. glabrata fungaemia were documented, accounting for 30% of the 598 episodes of candidaemia. A dramatic decline of C. glabrata causing candidaemia from 2003 (46.8%) to 2005 (15.8%) was noted, accompanied by decreased fluconazole consumption. The most common underlying diseases in these patients were cancer (49%), diabetes (34%) and renal failure (25%). The most common risk factors were central venous catheter use (88%), antimicrobial treatment (87%) and parenteral nutrition (51%). The 30-day all-cause mortality was 48.6%, but only 31% of patients were eventually discharged from the hospital. There was no significant survival difference between patients with C. glabrata and Candida albicans fungaemia. Rates of antifungal susceptibility were 63% for fluconazole, 93% for voriconazole, 96% for caspofungin, 98% for amphotericin B and 99% for flucytosine. The different levels of susceptibility to fluconazole (susceptible, susceptible-dose dependent and resistant) were not significantly associated with 30-day mortality. © 2009 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction A global trend of increasing incidence of invasive candidiasis and candidaemia has occurred in the past two decades [1–3]. A trend of an increasing proportion of non-albicans species in Candida bloodstream infection has also been reported [4,5], although it remains controversial [6,7]. The most important non-albicans Candida species in North America and some other regions is Candida glabrata [8], accounting for 10–20% of candidaemias [8,9]. Candida glabrata is unique to other Candia species in biology and genetics, with a smaller morphology and having a haploid genome [10]. Despite these differences between C. glabrata and other Candida spp., the clinical features of C. glabrata fungaemia are often studied in the more general context of candidaemias and few studies have focused solely on C. glabrata fungaemia. In this study, clinical data and treatment outcomes of C. glabrata fungaemia at National Taiwan University Hospital over a 7-year period from 1999 to 2005 were retrospectively analysed. In addition, all C. glabrata isolates
during this period were analysed to determine their antifungal susceptibility. 2. Materials and methods 2.1. Patients and setting All episodes of C. glabrata fungaemia during 1999–2005 were identified from the records of the microbiology laboratory at National Taiwan University Hospital, a 2000-bed, universityaffiliated medical centre in Taiwan. This hospital provides both primary and tertiary medical care and has ca. 200 Intensive Care Unit (ICU) beds, including medical, surgical and paediatric ICUs. All C. glabrata isolates during this period had been stored in the mycology laboratory at −70 ◦ C in trypticase soy broth supplemented with 15% glycerol. Each isolate was retrieved from storage for antifungal susceptibility testing. The medical records of each episode of C. glabrata fungaemia were retrospectively reviewed. 2.2. Definitions
∗ Corresponding author. Tel.: +886 2 2312 3456x5355; fax: +886 2 2322 4263. E-mail address: [email protected] (P.-R. Hsueh).
Candida glabrata fungaemia was defined as at least one blood culture positive for C. glabrata. Fungaemia in the same patient
0924-8579/$ – see front matter © 2009 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. doi:10.1016/j.ijantimicag.2009.02.021
S.-Y. Ruan et al. / International Journal of Antimicrobial Agents 34 (2009) 236–239
occurring during the same hospital course was counted as the same episode. The probable portals of entry were determined by medical records, which mainly reflected the judgement of the primary care clinicians. ‘Renal failure’ in the analysis of risk factors was defined as end-stage renal disease or need for dialysis. Liver cirrhosis was defined as Child-Pugh class B or C. Neutropenia was defined as an absolute neutrophil count <500 cells/L for >48 h. Nil per os (NPO) was defined as withholding of feeding for >48 h for any reason. Patients who received chemotherapy, systemic steroids or fluconazole within 30 days before the onset of fungaemia were classified as positive for each corresponding factor. Thirty-day mortality in this study included mortality due to any cause. Data on annual consumption of fluconazole (intravenous and oral forms) from 1999 to 2005 were obtained from the pharmacy department of the hospital. Antibiotic consumption was expressed as the number of defined daily doses/1000 patient-days. 2.3. Antifungal susceptibility testing Standard powders of the following antifungal agents were obtained from their respective manufacturers: fluconazole and voriconazole (Pfizer Inc., New York, NY); flucytosine (Sigma Chemical Co., St Louis, MO); caspofungin (Merck Sharp & Dohme, West Point, PA); and amphotericin B (Bristol-Myers Squibb Laboratories, Princeton, NJ). Prior to testing, isolates were passaged twice on potato dextrose agar (Difco Laboratories, Detroit, MI) at 35 ◦ C. Antifungal susceptibility testing was performed by the reference broth microdilution method of the Clinical and Laboratory Standards Institute (CLSI) [11]. The tested concentrations of the agents ranged from 0.03 g/mL to 64 g/mL. Final dilutions were made in RPMI 1640 medium (Sigma) buffered to pH 7.0 with 0.165 M morpholinepropanesulfonic acid (MOPS) buffer (Sigma). Following incubation at 35 ◦ C for 48 h, minimum inhibitory concentrations (MICs) were determined. Two reference strains, Candida parapsilosis ATCC 22019 and Candida krusei ATCC 6258, were used as the control strains for susceptibility testing. Susceptibilities to fluconazole and flucytosine were determined according to CLSI guidelines. The proposed susceptibility breakpoint of 1 g/mL was used for voriconazole and amphotericin B and 2 g/mL was used for caspofungin [12–14]. 2.4. Statistical analyses Categorical variables were analysed using Fisher’s exact test or 2 test. The probability of survival was estimated using the Kaplan–Meier method. Pearson’s correlation coefficient was used to determine the relationship between annual fluconazole consumption and trends in the percentage of C. glabrata among all Candida spp. causing fungaemia each year. An R-value greater than 0.72 (or less than −0.72) and a P-value <0.05 were considered statistically significant. 3. Results 3.1. Epidemiology During the 7-year study period (1999–2005), there were 598 episodes of candidaemia. The species distribution of these 598 episodes was 41% Candida albicans (n = 246), 30% C. glabrata (n = 177), 14% C. tropicalis (n = 82), 11% C. parapsilosis (n = 65), 2% Candida guilliermondii (n = 11) and 2% other Candida spp. Candida glabrata was the second most common aetiology of candidaemia. The proportion of C. glabrata causing all candidaemia ranged from 15.4% (1999) to 46.8% (2003) and markedly declined in 2005 (15.8%) (Fig. 1). Although Fig. 1 might suggest a relationship between fluconazole consumption and the proportion of C. glabrata among all
237
Fig. 1. Percentage of Candida glabrata causing all candidaemias and fluconazole consumption. Table 1 Characteristics of 177 episodes of Candida glabrata fungaemia and association with 30-day mortality. No. of episodes (%)a
Association with 30-day mortality (P-value)
Gender (male:female) Age (mean ± S.D.) (years)
89:88 62.9 ± 20.0
Underlying factors Use of central catheters Use of antibacterial agent Receipt of parenteral nutrition Nothing per os Prior abdominal surgery Prior fluconazole use Anticancer chemotherapy Gastrointestinal bleeding Steroid Neutropenia
156 (88) 154 (87) 91 (51) 72 (41) 60 (34) 47 (27) 41 (23) 38 (21) 30 (17) 5 (3)
0.646 0.377 0.176 0.006 0.752 0.499 0.158 0.103 0.015 0.675
Underlying diseases Malignancy Diabetes Renal failure Liver cirrhosis Autoimmune diseases HIV
87 (49) 60 (34) 45 (25) 20 (11) 10 (6) 1 (1)
0.051 0.528 0.039 0.001 0.052 1.000
S.D., standard deviation; HIV, human immunodeficiency virus. a Data are n (%) unless otherwise stated.
candidaemia, Pearson’s correlation analysis revealed no significant correlation between these variables (R = 0.57; P = 0.24). The characteristics of the 175 patients with 177 episodes of C. glabrata fungaemia are summarised in Table 1. The mean age was 63 years and 70% of patients were >60 years of age. Onset of fungaemia occurred at a median of 27 days after admission. Most of the patients had several co-morbidities and risk factors (Table 1). The most common underlying diseases were cancer (49%), diabetes (34%) and renal failure (25%). The most common risk factors were central venous catheter use (88%), antibacterial treatment (87%) and parenteral nutrition (51%). The probable portals of entry for C. glabrata fungaemia are shown in Table 2. The probable portal of entry could be determined retrospectively from the medical records in only 130 (73%) of 177 Table 2 Probable portals of entry of Candida glabrata fungaemia. Probable portals of entry (N = 130)
No. of episodes (%)
Central venous catheter Urinary tract Intra-abdominal Airway or lung Surgical wound Other
79 (61) 20 (15) 18 (14) 3 (2) 4 (3) 6 (5)
238
S.-Y. Ruan et al. / International Journal of Antimicrobial Agents 34 (2009) 236–239
Table 3 In vitro susceptibilities of Candida glabrata isolates to five antifungal agents. Antifungal
Fluconazole Voriconazole Caspofungin Flucytosine Amphotericin B
Susceptible (%)a
MIC (g/mL) Range
MIC50
MIC90
2 to >64 0.06–64 0.25 to >64 0.06 to >64 0.25–2
8 0.25 0.5 0.06 1
32 1 1 0.12 1
63 93 96 99 98
MIC, minimum inhibitory concentration; MIC50/90 , MIC at which 50% and 90% of the isolates are inhibited, respectively. a Susceptibility breakpoints were defined as ≤8 g/mL for fluconazole and ≤4 g/mL for flucytosine, and presumptively identified as ≤1 g/mL for voriconazole and amphotericin B and ≤2 g/mL for caspofungin.
fungaemic episodes. The most commonly identified portal of entry was a central venous catheter (61%), followed by the urinary tract (15%) and intra-abdominal (14%).
Fig. 2. Survival curves for 177 episodes of Candida glabrata fungaemia and 246 episodes of Candida albicans fungaemia (P = 0.331, log-rank test).
C. glabrata fungaemia, the 30-day all-cause mortality rate was 48.6% (41.1% for C. albicans fungaemia), but only 31% of patients were eventually discharged from the hospital.
3.2. Antifungal susceptibility testing 4. Discussion The MICs of the C. glabrata isolates to five antifungal agents are shown in Table 3. The susceptibility rate was 63% for fluconazole, 93% for voriconazole, 96% for caspofungin, 99% for flucytosine and 98% for amphotericin B. The proportion of isolates with fluconazole resistance (MIC > 32 g/mL) was 7% and with susceptible-dose dependence was 30% (MIC = 16–32 g/mL). The different levels of susceptibility to fluconazole (susceptible, susceptible-dose dependent and resistant) were not significantly associated with 30-day mortality (P = 0.09). 3.3. Treatment and outcomes Of the 177 C. glabrata fungaemia episodes, 21 (11.9%) were not treated with antifungal agents. Lack of antifungal treatment was most often due to death of the patient before the time of diagnosis. Of the 156 treated episodes, the mean interval between culture and the start of empirical antifungal treatment was 2.3 days, and 78% of patients were treated within 3 days. Early antifungal treatment, starting <72 h after culture, was associated with worse outcome (52% vs. 23%; P = 0.03). The treatment regimens are summarised in Table 4. The most common treatment regimen, used in 48.1% of episodes, began with fluconazole and then switched to amphotericin B owing to poor treatment response or results of susceptibility testing. The mortality rate of the various regimens was not significantly different (P = 0.85). Amphotericin Bcontaining regimens had a lower 30-day mortality (42%) compared with non-amphotericin B regimens (50%) but this difference was not significant (P = 0.34). The mortality of C. albicans and C. glabrata fungaemia was compared to determine the relative impact of C. glabrata fungaemia on patients. Fig. 2 shows the survival curves for C. glabrata and C. albicans fungaemia over the 3 months. There was no significant survival difference between patients with C. glabrata and C. albicans fungaemia (P = 0.33, log-rank test). For patients with Table 4 Antifungal regimens and associated 30-day mortality. Regimen (N = 156)
No. of episodes (%)
No. of deaths
Mortality (%)
Fluconazole then amphotericin B Fluconazole alone Fluconazole then caspofungin Amphotericin B alone Amphotericin B then fluconazole Caspofungin then amphotericin B
75 (48.1) 63 (40.4) 5 (3.2) 10 (6.4) 2 (1.3) 1 (0.6)
32 31 3 4 1 0
43 49 –a 40 – –
a
Mortality rate not shown owing to small number of cases.
Several studies have reported an increasing trend of C. glabrata in candidaemia, especially in cancer patients [15–17]. This trend was proposed to be related to the use of fluconazole prophylaxis [15,18]. However, another study found that exposure to antibacterial agents, specifically vancomycin or piperacillin/tazobactam but not fluconazole, was associated with subsequent hospital-acquired C. glabrata or C. krusei candidaemia [19]. In the present study, C. glabrata accounted for 30% of candidaemias, which is higher than the average reported value (10–20%) [8]. The annual percentage of C. glabrata causing all candidaemia is shown in Fig. 1. The reason for the dramatic decline of C. glabrata causing candidaemia from 2003 (46.8%) to 2005 (15.8%) is not clear, although fluconazole consumption also decreased during these 3 years. During the study period, fluconazole or other azoles were not routinely used as prophylactic agents for patients with haematological cancer in our hospital. The predisposing factors for C. glabrata candidaemia identified in this study (Table 1) are similar to those in previous reports [20,21]. Our findings that 30-day mortality was associated with renal failure, liver cirrhosis, steroid use and NPO are also similar to results of previous studies which found that old age, poor performance status and extensive organ involvement were poor prognostic factors for candidaemia [22–24]. The gut and skin were thought to be the two major portals of entry of candidaemia and the latter has been implicated in the pathogenesis of catheter-related candidaemia [25]. Candida glabrata can be a non-pathogenic saprophyte in the gastrointestinal tract, on the skin and in the vagina of humans [10]. Therefore, the sources of C. glabrata fungaemia should reflect this biological habit. Although the gut has been considered as the most important source of candidaemia, the supporting data were not focused on C. glabrata [25]. In the present study, the most common sources of C. glabrata fungaemia were central venous catheters (61%). Despite several drawbacks with the determination of the source of candidaemia, our findings suggest that the skin is the predominant portal of entry of C. glabrata fungaemia. More prospective studies focusing on C. glabrata are needed to confirm these findings. Compared with C. albicans, C. glabrata is relatively resistant to fluconazole [8]. Fluconazole resistance is one of the greatest concerns when managing patients with C. glabrata fungaemia. In the present study, the susceptibility rate was only 63% for fluconazole. However, fluconazole susceptibility and the choice of antifungal regimen were not correlated with mortality. Previous studies also reported that patients with candidaemia were often seriously ill and outcome appeared to be associated with underlying conditions
S.-Y. Ruan et al. / International Journal of Antimicrobial Agents 34 (2009) 236–239
rather than fluconazole susceptibility or treatment agents [26–28]. Amphotericin B is an important therapeutic agent for infection caused by azole-resistant Candida spp. The guidelines of the Infectious Diseases Society of America (IDSA) suggest treating serious C. glabrata infections with amphotericin B or high-dose fluconazole [29]. In this study, patients treated with amphotericin B-containing regimens had a lower mortality rate than those treated with nonamphotericin B regimens, but this difference was not significant (42% vs. 50%; P = 0.34). However, because the majority of patients who received amphotericin B-containing regimens were treated with fluconazole prior to amphotericin B, the lack of difference in mortality between the two regimens might have been due to patients receiving inappropriate initial therapy. Candida glabrata fungaemia is associated with poor outcome and the 30-day mortality rate ranged from 25% to 50% in previous reports [20,21,30]. In this study, the 30-day all-cause mortality rate was 48.6%, and only 31% of patients could be discharged from hospital. Early antifungal treatment, starting <72 h after culture, was associated with worse outcome (52% vs. 23%; P = 0.03). This conflicting result might be due to the bias of greater severity of underlying diseases in these patients because primary care physicians tend to early use of antifungal agents in patients with more serious conditions. However, this assumption could not be tested because of the lack of Acute Physiology and Chronic Health Evaluation (APACHE) II score or other severity score data needed to adjust for disease severity in the analysis. Recent studies have documented that delayed antifungal therapy significantly impacted mortality in candidaemic patients [31,32]. In addition, the main antifungal agent used for early treatment in our study was fluconazole and it was possibly inappropriate. This may be another reason why early treatment does not result in better outcome. Our study has several limitations. First, because of the retrospective nature of this study, determination of the fungaemia source was arbitrary and mainly relied on the clinical judgement of primary care physicians. Second, the choice of antifungal agent and the switching of antifungal agents did not follow a consistent protocol. There might be biases in the selection of antifungal agents. Third, the sample size was small when various treatment subgroups were assessed for statistically meaningful associations with mortality. In conclusion, C. glabrata was the second most common pathogen causing candidaemia in this study from Taiwan, accounting for 30% of candidaemia in a tertiary care hospital. The outcome was poor, with a 30-day all-cause mortality rate of 48.6%, and only 31% of patients were eventually discharged from hospital. There was no significant survival difference between patients with C. glabrata and C. albicans fungaemia. Fluconazole susceptibility of the isolates was not significantly associated with mortality in these patients. Funding: No funding sources. Competing interests: None declared. Ethical approval: Not required. References [1] Asmundsdottir LR, Erlendsdottir H, Gottfredsson M. Increasing incidence of candidemia: results from a 20-year nationwide study in Iceland. J Clin Microbiol 2002;40:3489–92. [2] Beck-Sague C, Jarvis WR. Secular trends in the epidemiology of nosocomial fungal infections in the United States, 1980–1990. National Nosocomial Infections Surveillance System. J Infect Dis 1993;167:1247–51. [3] Hsueh PR, Teng LJ, Yang PC, Ho SW, Luh KT. Emergence of nosocomial candidemia at a teaching hospital in Taiwan from 1981 to 2000: increased susceptibility of Candida species to fluconazole. Microb Drug Resist 2002;8: 311–9. [4] Nguyen MH, Peacock JE, Morris AJ, Tanner DC, Nguyen ML, Snydman DR, et al. The changing face of candidemia: emergence of non-Candida albicans species and antifungal resistance. Am J Med 1996;100:617–23.
239
[5] Trick WE, Fridkin SK, Edwards JR, Hajjeh RA, Gaynes RP. Secular trend of hospital-acquired candidemia among intensive care unit patients in the United States during 1989–1999. Clin Infect Dis 2002;35:627–30. [6] Sandven P, Bevanger L, Digranes A, Haukland HH, Mannsaker T, Gaustad P. Candidemia in Norway (1991 to 2003): results from a nationwide study. J Clin Microbiol 2006;44:1977–81. [7] Pfaller MA, Diekema DJ. Twelve years of fluconazole in clinical practice: global trends in species distribution and fluconazole susceptibility of bloodstream isolates of Candida. Clin Microbiol Infect 2004;10(Suppl. 1):11–23. [8] Pfaller MA, Diekema DJ. Epidemiology of invasive candidiasis: a persistent public health problem. Clin Microbiol Rev 2007;20:133–63. [9] Eggimann P, Garbino J, Pittet D. Epidemiology of Candida species infections in critically ill non-immunosuppressed patients. Lancet Infect Dis 2003;3:685–702. [10] Fidel PL, Vazquez JA, Sobel JD. Candida glabrata: review of epidemiology, pathogenesis, and clinical disease with comparison to C. albicans. Clin Microbiol Rev 1999;12:80–96. [11] National Committee for Clinical Laboratory Standards. Reference method for broth microdilution antifungal susceptibility testing of yeasts; approved standard. Document M27-A2. Wayne, PA: NCCLS; 2002. [12] Nguyen MH, Clancy CJ, Yu VL, Yu YC, Morris AJ, Snydman DR, et al. Do in vitro susceptibility data predict the microbiologic response to amphotericin B? Results of a prospective study of patients with Candida fungemia. J Infect Dis 1998;177:425–30. [13] Pfaller MA, Boyken L, Hollis RJ, Kroeger J, Messer SA, Tendolkar S, et al. In vitro susceptibility of invasive isolates of Candida spp. to anidulafungin, caspofungin, and micafungin: six years of global surveillance. J Clin Microbiol 2008;46:150–6. [14] Pfaller MA, Diekema DJ, Rex JH, Espinel-Ingroff A, Johnson EM, Andes D, et al. Correlation of MIC with outcome for Candida species tested against voriconazole: analysis and proposal for interpretive breakpoints. J Clin Microbiol 2006;44:819–26. [15] AbiSaid D, Anaissie E, Uzun O, Raad I, Pinzcowski H, Vartivarian S. The epidemiology of hematogenous candidiasis caused by different Candida species. Clin Infect Dis 1997;24:1122–8. [16] Safdar A, Chaturvedi V, Cross EW, Park S, Bernard EM, Armstrong D, et al. Prospective study of Candida species in patients at a comprehensive cancer center. Antimicrob Agents Chemother 2001;45:2129–33. [17] Hachem R, Hanna H, Kontoyiannis D, Jiang Y, Raad I. The changing epidemiology of invasive candidiasis: Candida glabrata and Candida krusei as the leading causes of candidemia in hematologic malignancy. Cancer 2008;112:2493–9. [18] Marr KA, Seidel K, White TC, Bowden RA. Candidemia in allogeneic blood and marrow transplant recipients: evolution of risk factors after the adoption of prophylactic fluconazole. J Infect Dis 2000;181:309–16. [19] Lin MY, Carmeli Y, Zumsteg J, Flores EL, Tolentino J, Sreeramoju P, et al. Prior antimicrobial therapy and risk for hospital-acquired Candida glabrata and Candida krusei fungemia: a case–case–control study. Antimicrob Agents Chemother 2005;49:4555–60. [20] Gumbo T, Isada CM, Hall G, Karafa MT, Gordon SM. Candida glabrata fungemia—clinical features of 139 patients. Medicine 1999;78:220–7. [21] Malani A, Hmoud J, Chiu L, Carver PL, Bielaczyc A, Kauffman CA. Candida glabrata fungemia: experience in a tertiary care center. Clin Infect Dis 2005;41:975–81. [22] Boo TW, O’Reilly B, O’Leary J, Cryan B. Candidaemia in an Irish tertiary referral hospital: epidemiology and prognostic factors. Mycoses 2005;48:251–9. [23] Uzun O, Anaissie EJ. Predictors of outcome in cancer patients with candidemia. Ann Oncol 2000;11:1517–21. [24] Weinberger M, Leibovici L, Perez S, Samra Z, Ostfeld I, Levi I, et al. Characteristics of candidaemia with Candida albicans compared with non-albicans Candida species and predictors of mortality. J Hosp Infect 2005;61:146–54. [25] Nucci M, Anaissie E. Revisiting the source of candidemia: skin or gut? Clin Infect Dis 2001;33:1959–67. [26] Antoniadou A, Torres HA, Lewis RE, Thornby J, Bodey GP, Tarrand JJ, et al. Candidemia in a tertiary care cancer center—in vitro susceptibility and its association with outcome of initial antifungal therapy. Medicine 2003;82:309–21. [27] Pappas PG, Rex JH, Lee J, Hamill RJ, Larsen RA, Powderly W, et al. A prospective observational study of candidemia: epidemiology, therapy, and influences on mortality in hospitalized adult and pediatric patients. Clin Infect Dis 2003;37:634–43. [28] Viudes A, Peman J, Canton E, Ubeda P, Lopez-Ribot JL, Gobernado M. Candidemia at a tertiary-care hospital: epidemiology, treatment, clinical outcome and risk factors for death. Eur J Clin Microbiol Infect Dis 2002;21:767–74. [29] Pappas PG, Rex JH, Sobel JD, Filler SG, Dismukes WE, Walsh TJ, et al. Guidelines for treatment of candidiasis. Clin Infect Dis 2004;38:161–89. [30] Wilson AG, Micek ST, Ritchie DJ. A retrospective evaluation of fluconazole for the treatment of Candida glabrata fungemia. Clin Ther 2005;27:1228–37. [31] Garey KW, Rege M, Pai MP, Mingo DE, Suda KJ, Turpin RS, et al. Time to initiation of fluconazole therapy impacts mortality in patients with candidemia: a multiinstitutional study. Clin Infect Dis 2006;43:25–31. [32] Morrell M, Fraser VJ, Kollef MH. Delaying the empiric treatment of Candida bloodstream infection until positive blood culture results are obtained: a potential risk factor for hospital mortality. Antimicrob Agents Chemother 2005;49:3640–5.