Received 6 May 2008; returned 15 June 2008; revised 7 August 2008; accepted 11 August 2008

Transcription

1 Journal of Antimicrobial Chemotherapy (2008) 62, doi: /jac/dkn381 Advance Access publication 8 September 2008 Fungaemia caused by Candida glabrata with reduced susceptibility to fluconazole due to altered gene expression: risk factors, antifungal treatment and outcome Mario Tumbarello 1, Maurizio Sanguinetti 2, Enrico Maria Trecarichi 1, Marilena La Sorda 2, Marianna Rossi 1, Elena de Carolis 2, Katleen de Gaetano Donati 1, Giovanni Fadda 2, Roberto Cauda 1 and Brunella Posteraro 2 * 1 Institute of Infectious Diseases, Università Cattolica del Sacro Cuore, Rome, Italy; 2 Institute of Microbiology, Università Cattolica del Sacro Cuore, Rome, Italy Received 6 May 2008; returned 15 June 2008; revised 7 August 2008; accepted 11 August 2008 Background: The role of Candida glabrata in fungaemia is attributed in part to its reduced susceptibility to azoles, usually due to altered expression of genes encoding drug efflux pumps. The aims of this study were to identify risk factors for fungaemia due to C. glabrata isolates with decreased susceptibility to fluconazole and to analyse the response to antifungal treatment and the clinical outcome of C. glabrata infections in hospitalized patients. Methods: A retrospective case case control study was conducted at a university hospital from 2000 to Three patient groups were studied: 14 patients infected by a fluconazole-less-susceptible isolate [susceptible-dose-dependent (SDD) or resistant]; 21 patients infected by a fluconazolesusceptible (FS) isolate; and 70 uninfected controls. We measured expression of the drug efflux pumpencoding CgCDR1 and CgCDR2 genes in isolates of the two infected groups using quantitative real-time PCR. Results: Multivariable analysis found that patients with prior fluconazole use [odds ratio (OR) 12.24, 95% confidence intervals (CIs) , P ], diabetes (OR 10.47, 95% CI , P ) and a central venous catheter (CVC) (OR 8.48, 95% CI , P ) were more likely to develop fungaemia due to a less-susceptible isolate. Previous surgery (OR 7.73, 95% CI , P ) was an independent risk factor for fungaemia due to a susceptible isolate, in addition to the presence of a CVC (OR 5.48, 95% CI , P ). The crude 30 day mortality rate was high for both case groups. Seven patients received inadequate antifungal treatment, including five infected by a fluconazole-resistant isolate but empirically treated with fluconazole; six of these seven patients died. Expression of the CgCDR genes was up-regulated in all fluconazole-resistant and, to a lesser extent, SDD isolates, but not in the FS isolates. Conclusions: Our data suggest that when candidaemia is suspected or detected, a more broadspectrum antifungal drug (i.e. echinocandins or amphotericin B) should be considered as initial treatment for patients with prior azole exposure. Keywords: candidaemia, antifungal resistance, mortality, inadequate therapy, drug efflux pumps Introduction Candida glabrata has emerged as a leading cause of fungal infection and is the second most common Candida organism responsible for fungaemia in the United States. 1 It is known to exhibit intrinsically low susceptibility to azoles and to develop resistance after exposure to these drugs, 2 9 through the CgPDR1-dependent 10 up-regulation of the genes encoding the ATP-binding cassette (ABC) transporters CgCDR1 and CgCDR2, 3,5,11 16 and also CgSNQ *Corresponding author. bposteraro@yahoo.it # The Author Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 Tumbarello et al. The aims of this study were to identify specific variables associated with fungaemia due to a C. glabrata isolate with reduced susceptibility to fluconazole (RFS) and to assess how the effectiveness of antifungal therapy for C. glabrata fungaemias caused by less-susceptible isolates compares with that for susceptible isolates. Molecular studies were also performed to investigate the mechanisms underlying the azole lesssusceptibility or resistance of the fungal isolates. Patients and methods Study site, patients and design This retrospective study was performed at the Catholic University Hospital, a 1700 bed university hospital located in Rome, Italy. It offers a full range of clinical services and admits patients per year. The microbiology laboratory historical database was used to identify patients, hospitalized from January 2000 through December 2006, who had at least one positive blood culture for C. glabrata. Patients could be entered into the study only once, at the time of the initial positive blood culture. A case case control study design was used. This approach involves two parallel case control studies with a common control group to give a more precise estimate of risk. Two case groups were studied: the first case group included patients with C. glabrata fungaemia due an isolate with RFS (MIC 16 mg/l) and the second case group consisted of patients with C. glabrata fungaemia due a fluconazole-susceptible (FS) isolate (MIC 8mg/L). The RFS isolates included either susceptible-dose-dependent (SDD; MICs mg/l) or resistant isolates (MICs 64 mg/l) (see below). A third group, the control group, consisted of randomly selected patients who had been hospitalized in our centre during the same periods of time and in the same wards as the case patients, but who did not have C. glabrata fungaemia. Patients were included as controls only if complete data series could be obtained from their medical records. The distribution of the control admissions throughout the study period was similar to that of the case admissions. 20 Variables and definitions Patient data were collected from computerized laboratory and medical records. The following variables were assessed as possible risk factors: age; sex; location of patient at the time of diagnosis in an intensive care unit (ICU) or a non-icu; underlying co-morbidities; invasive procedures [including insertion of a central venous catheter (CVC) or nasogastric tube, endoscopy, bronchoscopy and urinary catheterization] and total parenteral nutrition (TPN) administration within 72 h prior to the onset of candidaemia; and prior use of immunosuppressive agents, surgery, bacteraemia or exposure to antibiotics or antifungal agents within 30 days of the onset of candidaemia (or, for controls, at any point during hospitalization). 20 The Charlson score was used to provide a composite score of co-morbid conditions. 21 Risk days were determined as the number of hospital days from admission to the date of the first positive blood culture for case patients or total days in the hospital for control patients. 18 Antibiotic use was examined, both in terms of whether any antibiotics were used and by aggregate classes (glycopeptides, macrolides, fluoroquinolones, expanded-spectrum cephalosporins, aminoglycosides, carbapenems and b-lactam-b-lactamase inhibitors). 18 The type, dose and duration of antifungal therapy, as well as the 30 day mortality, were also recorded. If patients had positive blood cultures after 48 h of hospital admission, C. glabrata fungaemias were defined as hospital-acquired; also, C. glabrata fungaemias were defined as catheter-related if: (i) the semi-quantitative catheter tip culture yielded more than 15 cfu of the same Candida species; or (ii) simultaneous quantitative cultures of blood samples showed a ratio 5:1 in cfu of blood samples obtained through the catheter and a peripheral vein, as reported previously. 22 Inadequate C. glabrata fungaemia therapy was defined as a lack of antifungal treatment, administration of antifungal treatment 48 h after drawing the first positive blood sample for culture, administration of antifungal treatment with isolation of an organism found in vitro to be resistant to the antifungal agent or antifungal administration for,5 days. 22 The 30 day mortality was used as the outcome measure. Microbiological studies For each patient studied, blood samples were inoculated into aerobic media and processed using the BACTEC 9240 automated system (Becton Dickinson Microbiology Systems, Sparks, MD, USA). The first isolate from each patient was studied. Identification was performed using micromorphology analysis 23 and biochemical tests (Vitek 2 Yeast Identification, biomérieux, Marcy l Étoile, France). The in vitro antifungal activity of amphotericin B, fluconazole, itraconazole, voriconazole and caspofungin was determined, at least three times, using the broth microdilution procedure recommended by the CLSI (formerly the NCCLS). 24 Testing for posaconazole susceptibility was performed using the Sensititre YeastOne system according to the manufacturer s instructions (Trek Diagnostic Systems, Inc., Cleveland, OH, USA) 6 on three separate occasions. To classify the isolates, available interpretive CLSI MIC breakpoints for fluconazole, voriconazole and itraconazole were used: isolates were defined as susceptible (8, 1 and 0.12 mg/l, respectively), SDD (16 32, 2 and mg/l, respectively) or resistant (64, 4 and 1 mg/l, respectively). 24 Interpretive criteria are not available for amphotericin B, caspofungin and posaconazole; however, in agreement with previous studies, isolates for which the amphotericin B, caspofungin or posaconazole MIC was 1 mg/ L were defined as susceptible in this study. Isolates were maintained as frozen stocks and subcultured on Sabouraud dextrose agar plates at 378C as required. The clonality of the isolates was assessed using the repetitive probes Cg6 and Cg12 28 and by multilocus sequence typing, 29 as described elsewhere. 15,30 The biofilm-forming capability of the isolates was determined in microtitre plates as described previously 22 by measuring the amount of light blocked passing through the wells (%T bloc ) 31 and by using the XTT (2,3-bis(2-methoxy-4-nitro-5-sulfophenye)-5-[(phenylamino) carbonyl]-2h-tetrazodium hydroxide) reduction assay. 32 Quantitative expression of the azole resistance-related genes CgCDR1, CgCDR2 and CgSNQ2 was performed by real-time RT PCR, as described previously. 15 Genes with a DC T value, calculated as C T [test gene] C T [URA3 reference gene], that fell over the 3-SD range measured in the azole-susceptible isolates were considered overexpressed. 33 Statistical analysis Variables with normal or near normal distributions were described with means and SD, and compared using the Student t-test. Medians with interquartile ranges (IQRs) were used to describe non-normally distributed variables and compared using the Mann Whitney U-test. Categorical variables were compared using the two-tailed x 2 test or Fisher s exact test, as appropriate; crude odds ratios (ORs) 1380

3 Risk factors for C. glabrata fungaemia and 95% confidence intervals (CIs) were calculated. Multivariate analysis was performed by means of multiple logistic regression, including all variables with a P value,0.2 in the univariate analysis. Two-tailed tests of significance (P, 0.05) were used to determine statistical significance. All statistical analyses were performed with Intercooled Stata software (version 8) for Windows (Stata Corp., College Station, TX, USA). Results Patient characteristics Of 35 case patients studied, 14 (40%) were infected by an RFS isolate (i.e. fluconazole-sdd or -resistant) and the remaining 21 (60%) by an FS isolate (see below for isolate categorization). During the study period, the proportion of C. glabrata fungaemias due to RFS isolates increased from 25% in 2001 to 75% in 2006 (Figure 1). At the time of diagnosis, 17 patients (48.6%) were in a surgical ward, 14 patients (40%) in a medical ward and 4 patients (11.4%) in the ICU. Thirty-two patients (91.4%) had hospital-acquired fungaemia. While the primary source of infection was unknown for 22 patients (62.9%), it was identified as a CVC or surgical wound for 9 (25.7%) and 4 (11.4%) patients, respectively. The median length (IQR) of hospitalization in patients infected by an RFS isolate was 32.5 (25 55) days, and 32 (24 73) days in patients infected by an FS isolate. Compared with patients with fungaemia due to FS C. glabrata, patients in the RFS C. glabrata group were more likely to be male (P ¼ 0.02), older (P ¼ 0.06) and admitted to a medical ward (P ¼ 0.001). Risk factors for fungaemia due to RFS C. glabrata or FS C. glabrata The results of univariate and multivariable analyses for RFS or FS C. glabrata fungaemia are shown in Tables 1 and 2. Characterization of C. glabrata isolates As noted previously, the eight fluconazole-sdd and six fluconazole-resistant isolates were arbitrarily grouped into a unique fluconazole-reduced susceptibility category and designated as RFS isolates. All of the six fluconazole-resistant isolates were also resistant to itraconazole and voriconazole. Finally, by using an Percentage of isolates (n = 4) 2001 (n = 4) 2002 (n = 8) 2003 (n = 5) 2004 (n = 5) 2005 (n = 5) 2006 (n = 4) Figure 1. Annual distribution of RFS C. glabrata isolates causing fungaemias during a 6 year study period. The total number of isolates (RFS plus FS) by each year is indicated in parentheses. MIC of 1 mg/l to define resistance to amphotericin B, posaconazole and caspofungin (see the Patients and methods section), all 35 isolates were found to be susceptible to amphotericin B and caspofungin, although not to posaconazole. While 29 isolates had posaconazole MICs of,1 mg/l, the 6 azole-resistant isolates exhibited posaconazole MICs ranging from 4 to.8 mg/l and were thus considered resistant to this compound [Table S1, available as Supplementary data at JAC Online ( All isolates were examined for expression of CgCDR1, CgCDR2 and CgSNQ2 relative to that of the C. glabrata control isolate DSY562 (fluconazole MIC, 8 mg/l). 12 The six fluconazole-resistant isolates had up-regulated CgCDR1, CgCDR2 and/or CgSNQ2 expression. The eight fluconazole-sdd isolates also up-regulated CgCDR1, CgCDR2 or CgSNQ2, but the relative expression levels were much lower than those of the six fluconazole-resistant isolates. In contrast, no increase in the expression of CgCDR1, CgCDR2 and CgSNQ2 was observed for the 21 FS isolates studied (Table S1). Nine of the 35 isolates (25.7%) were able to form biofilms, as assessed by the %T method (Table S1). For all the isolates tested, %T determinations correlated well with XTT absorbance measurements (data not shown). When biofilm production of the isolates was analysed according to fluconazole susceptibility, the percentages of isolates that were able to form biofilms among the RFS and FS isolates were 35.7% (5/14 patients) and 19% (4/21 patients), respectively; there was no statistically significant difference between the two groups (P ¼ 0.26). Finally, when the 35 C. glabrata isolates were genotyped, 28 sequence types were found, of which 10 had not been published previously. 29 Isolates with identical sequence types were studied by the Cg6 and Cg12 hybridization method and yielded different patterns, indicating that these isolates are epidemiologically unrelated strains (data not shown). Except for 3 patients who were diagnosed with community-acquired fungaemia, we excluded any nosocomial transmission of C. glabrata isolate among the remaining 32 patients. Antifungal treatment and outcome Twenty-eight of the fungaemic patients (80%) received adequate systemic antifungal therapy for a median length (IQR) of 12 (6 20) days. Therapy consisted of fluconazole in 13 cases (46.4%), amphotericin B (conventional or lipidic formulations) in 8 cases (28.6%), caspofungin in 5 cases (17.9%) and voriconazole in 2 cases (7.1%). Therapy was considered inadequate for seven patients (20%). Of these seven patients, five who were treated with fluconazole were infected by a fluconazole-resistant isolate, one patient received antifungal treatment with caspofungin 4 days after a positive blood sample was drawn for culture and the remaining patient was never treated; except for two patients who were already dead at the time that the susceptibility testing results were known, for the remaining three of the five fluconazole-treated patients therapy was changed to amphotericin B (two patients) or caspofungin (one patient). Among the adequately treated patients, 8 patients who were infected by a fluconazole-sdd isolate received antifungal treatment with amphotericin B (3 patients) or caspofungin (5 patients), whereas 20 patients who were infected by a fluconazole-susceptible isolate were treated with fluconazole (13 patients), voriconazole (2 patients) or amphotericin B (5 patients). Among the last 1381

4 Tumbarello et al. Table 1. Demographic and clinical characteristics of patients with C. glabrata fungaemia due to an isolate with reduced fluconazole susceptibility (RFS group), patients with C. glabrata fungaemia due to a fluconazole-susceptible isolate (FS group) and uninfected control patients (control group) Variable Control group (n ¼ 70) RFS group (n ¼ 14) RFS group versus control group OR (95% CI) FS group (n ¼ 21) FS group versus control group OR (95% CI) Demographic characteristics male sex 42 (60%) 12 (85.7%) 4.00 ( ) 10 (47.6%) 0.60 ( ) age, median years (IQR) 66 (59 72) 75 (62 78) 68 (65 72) Length of hospital stay, median 16 (11 23) 15 (6 23) 17 (9 25) days (IQR) a ICU stay 4 (5.7%) 2 (14.3%) 2.75 ( ) 6 (28.6%) 6.60 ( ) Co-morbid condition solid-organ cancer 12 (17.1%) 6 (42.9%) 3.62 ( ) 8 (38.1%) 2.97 ( ) haematological malignancy 2 (2.9%) 1 (7.1%) 2.61 ( ) 2 (9.5%) 3.57 ( ) liver disease 10 (14.3%) 2 (14.3%) 1 ( ) 4 (19%) 1.41 ( ) diabetes mellitus 14 (20%) 6 (42.9%) 3.00 ( ) 5 (23.8%) 1.25 ( ) renal failure 14 (20%) 5 (35.7%) 2.22 ( ) 7 (33.3%) 2.00 ( ) Charlson index, median (IQR) 2 (1 4) 6 (2 7) 2 (1 6) Invasive procedures CVC 14 (20%) 8 (57.1%) 5.33 ( ) 14 (66.7%) 8.00 ( ) nasogastric tube 11 (15.7%) 1 (7.1%) 0.41 ( ) 6 (28.6%) 2.14 ( ) urinary catheterization 28 (40%) 8 (57.1%) 2.00 ( ) 19 (90.5%) ( ) TPN 15 (21.4%) 9 (64.3%) 6.60 ( ) 13 (61.9%) 5.95 ( ) Prior steroid use b 8 (11.4%) 6 (42.9%) 5.81 ( ) 2 (9.5%) 0.81 ( ) Previous surgery b 20 (28.6%) 6 (42.9%) 1.87 ( ) 17 (81%) ( ) Previous bacteraemia b 2 (2.9%) 4 (28.6%) ( ) 4 (19%) 8.00 ( ) Prior use of antibiotics by all or 33 (47.1%) 10 (71.4%) 2.80 ( ) 17 (81%) 4.76 ( ) aggregate classes b glycopeptides 2 (2.9%) 2 (14.3%) 5.66 ( ) 7 (33.3%) ( ) macrolides 2 (2.9%) 1 (7.1%) 2.61 ( ) 2 (9.5%) 3.57 ( ) fluoroquinolone 13 (18.6%) 5 (35.7%) 2.43 ( ) 7 (33.3%) 2.19 ( ) expanded-spectrum 10 (14.3%) 3 (21.4%) 1.63 ( ) 9 (42.9%) 4.5 ( ) cephalosporins aminoglycosides 2 (2.9%) 2 (14.3%) 5.66 ( ) 5 (23.8%) ( ) carbapenems 5 (7.1%) 4 (28.6%) 5.20 ( ) 4 (19%) 3.05 ( ) b-lactam-b-lactamase 9 (12.9%) 3 (21.4%) 1.84 ( ) 4 (19%) 1.59 ( ) inhibitors Prior fluconazole use b 4 (5.7%) 4 (28.6%) 6.60 ( ) 2 (9.5%) 1.73 ( ) Data are number (%) of patients, unless otherwise indicated. a Days of hospital stay prior to the onset of candidaemia (or, for controls, total days in the hospital). b Within 30 days prior to the onset of candidaemia (or, for controls, at any point during hospitalization). patients, amphotericin B was switched to fluconazole in three patients, based on the susceptibility testing results. Fifteen of the 35 fungaemic patients died (30 day crude mortality rate, 42.8%). The mortality rate was 38.1% for the 21 patients with fungaemia due to an FS isolate versus 50% among the 14 patients with fungaemia due to an RFS isolate (P ¼ 0.48). Among the seven patients who did not receive adequate therapy, the mortality rate was 85.7%. This rate differed significantly (P ¼ 0.01) from the rate (32.1%) among the 28 patients who received adequate therapy. Discussion We identified two unique risk factors for fungaemia due to RFS C. glabrata: prior fluconazole exposure and diabetes mellitus. The selective pressure of fluconazole has often been cited as the major driver of low-susceptibility Candida species (C. glabrata and Candida krusei), 34 raising concerns about the use of azolebased drugs for prophylaxis, pre-emptive therapy and empirical therapy in high-risk hospital settings. 35 In vitro work by Borst et al. 16 demonstrated that acquisition of azole resistance in 1382

5 Risk factors for C. glabrata fungaemia Table 2. Multivariable analysis of risk factors for C. glabrata fungaemia due to an isolate with reduced fluconazole susceptibility (RFS) and C. glabrata fungaemia due to a fluconazole-susceptible (FS) isolate Risk factor for fungaemia OR (95% CI) Due to RFS C. glabrata prior fluconazole use ( ) diabetes mellitus ( ) CVC 8.48 ( ) Due to FS C. glabrata previous surgery 7.73 ( ) CVC 5.48 ( ) C. glabrata can occur in the absence of prior exposure to these drugs and is associated with the increased expression of ABC transporter genes. However, another study of azole resistance mechanisms in C. glabrata isolates from transplant recipients receiving fluconazole prophylaxis showed that the MICs doubled, on average, every 31 days. 5 Expression of CgCDR1, CgCDR2 and CgSNQ2 was up-regulated to different extents in the fluconazole-resistant (6 patients) or fluconazole-sdd (8 patients) isolates, but not in the FS isolates (21 patients). Our findings seem to be in disaccord with those of Lin et al., 18 who reported the lack of an association between fluconazole use and the development of C. glabrata (or C. krusei) candidaemia. Nevertheless, in that study, C. glabrata isolates were assumed to be inherently more resistant than C. albicans to fluconazole, since no antifungal susceptibility testing was routinely performed for yeast isolates during the time period; this raises the possibility of a misclassification bias. For reasons we cannot explain, diabetes mellitus was an independent predictor of C. glabrata fungaemia due to RFS isolates but not of fungaemia due to FS isolates. Diabetes has previously been reported to be a risk factor for candiduria due to C. glabrata in renal transplant recipients, 36 but association of hyperglycaemia with overall C. glabrata fungaemia has never been reported. In addition, people with diabetes are at increased risk for oesophageal 37 and vulvovaginal candidiasis. 38 Prior surgery was a risk factor for fungaemia due to FS C. glabrata but not for fungaemia due to RFS C. glabrata. Interventions that affect the intestine are highly likely to create a conduit for yeast to enter the bloodstream. 39 The association between surgical procedures involving the bowel and fungaemia caused by FS strains can be explained either by greater fitness of the FS strains 40 or by local factors in our hospital. Treatment failure can be attributed to a combination of factors related to the host, the antifungal agent or the pathogen, 9 indicating that in vitro susceptibility testing alone is not sufficient for predicting clinical success. On the other hand, fluconazole susceptibility testing shows promise: isolates in which MICs are 64 mg/l are significantly less likely to respond clinically to fluconazole therapy than isolates with low MICs (i.e. susceptible). 41 Excluding the 15 case patients (42.9%) who died, 20 of 35 fungaemic patients had a favourable outcome, probably as a result of the receipt of appropriate antifungal therapy. In contrast, 6 of the 7 inadequately treated patients died (mortality rate of 85.7% versus 32.1% for the 28 adequately treated patients), including 5 who were empirically started with fluconazole before their isolates were known to have elevated MICs to fluconazole (and other azoles). The last finding supports the conclusion that reduced susceptibility of C. glabrata to fluconazole contributes to a poor outcome after invasive infection, especially in conjunction with inadequate empirical antifungal treatment. 42,43 Although we cannot exclude the possibility that modifying treatment based on the susceptibility testing results contributes to improved clinical outcome, it should be noted that even a minimal delay in implementing the appropriate antifungal therapy in patients hospitalized with candidaemia has a significant impact on mortality. 44,45 In this regard, new strategies such as early identification of resistance in clinical isolates (i.e. real-time detection of CgPDR1 mutations) or identification of unique risk factors are needed to reduce infection-related deaths. In conclusion, azole antifungal non-susceptibility in C. glabrata remains an especially acute problem in the management of diseases caused by this organism. Further studies evaluating the relationship between non-susceptibility and clinical outcome will be critical for determining the most effective treatment approach for patients with invasive candidiasis. For now, we advise caution when considering azole therapy for C. glabrata fungaemia in patients with prior azole drug exposure. Acknowledgements We would like to thank Paul Kretchmer at San Francisco Edit for revising our English text. Funding This work was supported by grants from the Istituto Superiore di Sanità (National Program of Research on AIDS, No. 50G.33), the Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Lazzaro Spallanzani (Strategic Research Program 2006) and the Università Cattolica del S. Cuore (Fondi Ateneo Linea D1-2007), Italy. Transparency declarations No financial conflicts of interest to declare. Paul Kretchmer at San Francisco Edit contributed to the preparation of this manuscript by revising the English language. Supplementary data Table S1 is available as Supplementary data at JAC Online ( References 1. Pfaller MA, Diekema DJ. Epidemiology of invasive candidiasis: a persistent public health problem. Clin Microbiol Rev 2007; 20: 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:

6 Tumbarello et al. 3. Redding S, Kirkpatrick WR, Saville S et al. Multiple patterns of resistance to fluconazole in Candida glabrata isolates from a patient with oropharyngeal candidiasis receiving head and neck radiation. J Clin Microbiol 2003; 41: Pfaller MA, Messer SA, Boyken L et al. Variation in susceptibility of bloodstream isolates of Candida glabrata to fluconazole according to patient age and geographic location. J Clin Microbiol 2003; 41: Bennett JE, Izumikawa K, Marr KA. Mechanism of increased fluconazole resistance in Candida glabrata during prophylaxis. Antimicrob Agents Chemother 2004; 48: Magill SS, Shields C, Sears CL et al. Triazole cross-resistance among Candida spp.: case report, occurrence among bloodstream isolates, and implications for antifungal therapy. J Clin Microbiol 2006; 44: Panackal AA, Gribskov JL, Staab JF et al. Clinical significance of azole antifungal drug cross-resistance in Candida glabrata. J Clin Microbiol 2006; 44: Posteraro B, Tumbarello M, La Sorda M et al. Azole resistance of Candida glabrata in a case of recurrent fungemia. J Clin Microbiol 2006; 44: Kanafani ZA, Perfect JR. Antimicrobial resistance to antifungal agents: mechanisms and clinical impact. Clin Infect Dis 2008; 46: Vermitsky JP, Earhart KD, Smith WL et al. Pdr1 regulates multidrug resistance in Candida glabrata: gene disruption and genome-wide expression studies. Mol Microbiol 2006; 61: Miyazaki H, Miyazaki Y, Geber A et al. Fluconazole resistance associated with drug efflux and increased transcription of a drug transporter gene, PDH1, incandida glabrata. Antimicrob Agents Chemother 1998; 42: Sanglard D, Ischer F, Calabrese D et al. The ATP binding cassette transporter gene CgCDR1 from Candida glabrata is involved in the resistance of clinical isolates to azole antifungal agents. Antimicrob Agents Chemother 1999; 43: Sanglard D, Ischer F, Bille J. Role of ATP-binding-cassette transporter genes in high-frequency acquisition of resistance to azole antifungals in Candida glabrata. Antimicrob Agents Chemother 2001; 45: Vermitsky JP, Edlind TD. Azole resistance in Candida glabrata: coordinate upregulation of multidrug transporters and evidence for a Pdr1-like transcription factor. Antimicrob Agents Chemother 2004; 48: Sanguinetti M, Posteraro B, Fiori B et al. Mechanisms of azole resistance in clinical isolates of Candida glabrata collected during a hospital survey of antifungal resistance. Antimicrob Agents Chemother 2005; 49: Borst A, Raimer MT, Warnock DW et al. Rapid acquisition of stable azole resistance by Candida glabrata isolates obtained before the clinical introduction of fluconazole. Antimicrob Agents Chemother 2005; 49: Torelli R, Posteraro B, Ferrari S et al. The ATP-binding cassette transporter-encoding gene CgSNQ2 is contributing to the CgPDR1-dependent azole resistance of Candida glabrata. Mol Microbiol 2008; 68: Lin MY, Carmeli Y, Zumsteg J 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: Harris AD, Karchmer TB, Carmeli Y et al. Methodological principles of case control studies that analyzed risk factors for antibiotic resistance: a systematic review. Clin Infect Dis 2001; 32: Tumbarello M, Spanu T, Sanguinetti M et al. Bloodstream infections caused by extended-spectrum-b-lactamase-producing Klebsiella pneumoniae: risk factors, molecular epidemiology, and clinical outcome. Antimicrob Agents Chemother 2006; 50: Charlson ME, Pompei P, Ales KL et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987; 40: Tumbarello M, Posteraro B, Trecarichi EM et al. Biofilm production by Candida species and inadequate antifungal therapy as predictors of mortality for patients with candidemia. J Clin Microbiol 2007; 45: Warren NG, Hazen KC. Candida, Cryptococcus, and other yeasts of medical importance. In: Murray PR, Baron EJ, Pfaller MA et al., eds. Manual of Clinical Microbiology, 7th edn. Washington, DC, USA: ASM Press, 1999; National Committee for Clinical Laboratory Standards. Reference Method for Broth Dilution Antifungal Susceptibility Testing of Yeasts: Approved Standard M27-A2. NCCLS, Wayne, PA, USA, Nguyen MH, Clancy CJ, Yu VL et al. Doin vitro susceptibility data predict the microbiologic response to amphotericin B? Results of prospective study of patients with Candida fungemia. J Infect Dis 1998; 177: Espinel-Ingroff A, Canton E, Gibbs D et al. Correlation of Neo-Sensitabs tablet diffusion assay results on three different agar media with CLSI broth microdilution M27-A2 and disk diffusion M44-A results for testing susceptibilities of Candida spp. and Cryptococcus neoformans to amphotericin B, caspofungin, fluconazole, itraconazole, and voriconazole. J Clin Microbiol 2007; 45: Pfaller MA, Messer SA, Boyken L et al. 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 J Clin Microbiol 2004; 42: Lockhart SR, Joly S, Pujol C et al. Development and verification of fingerprinting probes for Candida glabrata. Microbiology 1997; 143: Dodgson AR, Pujol C, Denning DW et al. Multilocus sequence typing of Candida glabrata reveals geographically enriched clades. J Clin Microbiol 2003; 41: Feil EJ, Li BC, Aanensen DM et al. eburst: inferring patterns of evolutionary descent among clusters of related bacterial genotypes from multilocus sequence typing data. J Bacteriol 2004; 186: Shin JH, Kee SJ, Shin MG et al. Biofilm production by isolates of Candida species recovered from nonneutropenic patients: comparison of bloodstream isolates with isolates from other sources. J Clin Microbiol 2002; 40: Kuhn DM, Chandra J, Mukherjee PK et al. Comparison of biofilms formed by Candida albicans and Candida parapsilosis on bioprosthetic surfaces. Infect Immun 2002; 70: Posteraro B, Sanguinetti M, Fiori B et al. Caspofungin activity against clinical isolates of azole cross-resistant Candida glabrata overexpressing efflux pump genes. J Antimicrob Chemother 2006; 58: Blot S, Janssens R, Claeys G et al. Effect of fluconazole consumption on long-term trends in candidal ecology. J Antimicrob Chemother 2006; 58: Pappas PG, Rex JH, Sobel JD et al. Guidelines for treatment of candidiasis. Clin Infect Dis 2004; 38: Safdar N, Slattery WR, Knasinski V et al. Predictors and outcomes of candiduria in renal transplant recipients. Clin Infect Dis 2005; 40:

7 Risk factors for C. glabrata fungaemia 37. Weerasuriya N, Snape J. Oesophageal candidiasis in elderly patients: risk factors, prevention and management. Drugs Aging 2008; 25: Goswami D, Goswami R, Banerjee U. Pattern of Candida species isolated from patients with diabetes mellitus and vulvovaginal candidiasis and their response to single dose oral fluconazole therapy. J Infect 2006; 52: Odds FC, Hanson MF, Davidson AD et al. One year prospective survey of Candida bloodstream infections in Scotland. J Med Microbiol 2007; 56: Anderson JB. Evolution of antifungal-drug resistance: mechanisms and pathogen fitness. Nat Rev Microbiol 2005; 3: Pfaller MA, Diekema DJ, Sheehan DJ. Interpretive breakpoints for fluconazole and Candida revisited: a blueprint for the future of antifungal susceptibility testing. Clin Microbiol Rev 2006; 19: Ibrahim EH, Sherman G, Ward S et al. The influence of inadequate antimicrobial treatment of bloodstream infections on patient outcomes in the ICU setting. Chest 2000; 118: Parkins MD, Sabuda DM, Elsayed S et al. Adequacy of empirical antifungal therapy and effect on outcome among patients with invasive Candida species infections. J Antimicrob Chemother 2007; 60: 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: Garey KW, Rege M, Pai MP et al. Time to initiation of fluconazole therapy impacts mortality in patients with candidemia: a multiinstitutional study. Clin Infect Dis 2006; 43: