Antifungal susceptibility profiles of Candida isolates from a prospective survey of invasive fungal infections in Italian intensive care units

Transcription

1 Journal of Medical Microbiology (2012), 61, DOI /jmm Antifungal susceptibility profiles of Candida isolates from a prospective survey of invasive fungal infections in Italian intensive care units Anna Maria Tortorano, 1 Anna Prigitano, 1 Giovanna Dho, 1 Anna Grancini, 2 Marco Passera 3 for the ECMM-FIMUA Study Group3 Correspondence Anna Maria Tortorano annamaria.tortorano@unimi.it 1 Dipartimento di Sanità Pubblica-Microbiologia Virologia, Università degli Studi di Milano, via Pascal 36, Milano, Italy 2 Laboratorio di Microbiologia, Fondazione IRCCS Ospedale Maggiore Policlinico, Mangiagalli, Regina Elena, via F. Sforza 35, Milano, Italy 3 Laboratorio di Microbiologia, Ospedali Riuniti di Bergamo, largo Barozzi 1, Bergamo, Italy Received 6 September 2011 Accepted 14 November 2011 The antifungal susceptibility pattern of 302 Candida isolates collected during an Italian survey on invasive fungal infections in an intensive care setting was investigated. The results were correlated with some epidemiological data and compared with the antifungal profiles obtained in a previous survey. No resistance to echinocandins was detected. The overall resistance levels to fluconazole, posaconazole and voriconazole were 12.6, 6.0 and 7.1 %, respectively. Candida tropicalis and Candida parapsilosis accounted for more than half of all the fluconazole resistant isolates. Reduced susceptibility to fluconazole is not uncommon among isolates (12.3 %) and appears to be increasing, particularly among C. parapsilosis isolates, which showed an increase in resistant isolates from 2 % in the 1990s to 25.8 % in the present study. Routine antifungal susceptibility testing of this species is therefore recommended. INTRODUCTION Despite the increased awareness of Candida as an important pathogen for intensive care unit (ICU) patients, candidaemia and invasive candidiasis continue to be significant problems in this patient population (Kett et al., 2011; Pappas & Mycoses Study Group, 2011). A recently published 1-day multicentre point prevalence study, involving adult patients that were hospitalized in one of 1265 participating ICUs in 76 countries, reported 99 patients with Candida bloodstream infection (BSI) with a prevalence rate of 6.9 cases per 1000 patients and a crude mortality rate of 42.6 % (Kett et al., 2011). These data are consistent with the results of a recent Italian survey reporting 318 invasive Candida infections, 276 of which were BSIs, diagnosed in ICU patients (Tortorano et al., 2012). In this survey, fluconazole monotherapy was the dominant antifungal treatment, given to 63 % of the patients, followed by lipid-based amphotericin B (given to 18 % of patients), caspofungin (given to 7 % of patients) and voriconazole (given to 6 % of patients). In the present study we investigated the antifungal susceptibility pattern of 302 invasive Candida isolates collected during this Italian survey and compared the 3Members listed in Acknowledgements. Abbreviations: BSI bloodstream infection; ICU, intensive care unit. results with some epidemiological data and antifungal profiles that were obtained in a previous survey. METHODS Between May 2006 and April 2008, a prospective sequential survey of invasive fungal infections was conducted in Italy involving 38 ICUs of 27 Italian hospitals (Tortorano et al., 2012). All cases of invasive candidiasis with available isolates were selected for this analysis. Isolates. A total of 302 Candida isolates obtained from blood or other normally sterile body sites of 295 patients (the first isolate from each episode) were studied. A mixed infection was reported in seven cases. The strains, received by the coordinating centre of the survey, were subcultured on Chromagar Candida medium (CHROMagar) to ensure viability and purity. The identification of the yeast was checked by determining production of germ tubes in serum, examining morphology on potato carrot ox gall agar and analysing the biochemical patterns by using ID 32 test strips (biomérieux). Isolates were stored as suspensions in distilled water at room temperature until needed. Susceptibility testing. The isolates were tested for susceptibility to: the echinocandins anidulafungin (Pfizer), caspofungin (Merck) and micafungin (Astellas Pharma); the triazoles fluconazole (Pfizer), voriconazole (Pfizer), and posaconazole (Schering Plough Research Institute); and flucytosine (Sigma) by using the Clinical and Laboratory Standards Institute (CLSI) broth microdilution method G 2012 SGM Printed in Great Britain 389

2 A. M. Tortorano and others (CLSI, 2008). Stock solutions of the antifungals fluconazole, flucytosine, caspofungin and micafungin were prepared in water and stock solutions of voriconazole, posaconazole, anidulafungin were prepared in DMSO (Sigma). Testing was conducted in RPMI 1640 medium without sodium bicarbonate (Sigma), which was buffered to ph 7.0 with M MOPS (Sigma) and supplemented with 0.03 % L- glutamine (Sigma). All tests were performed in duplicate. The MICs of each antifungal for each isolate were determined visually as the lowest concentration of drug that caused a significant reduction ( 50 % inhibition) of growth compared to control growth. Susceptibility to amphotericin B was determined using Etest strips (biomérieux) on antibiotic medium 3 (Difco). Candida parapsilosis ATCC and Candida krusei ATCC 6258 were used as quality control strains. The species-specific breakpoints recently revised by CLSI were used to identify Candida strains that were resistant to anidulafungin, micafungin and azoles (Pfaller et al., 2010, 2011a, b). Candida albicans, Candida tropicalis and C. krusei isolates were considered resistant to anidulafungin and micafungin in cases of MICs 1 mg l 21 ; C. parapsilosis was considered resistant to these antifungals in cases of MIC 8 mgl 21. Candida glabrata isolates were considered resistant to anidulafungin and micafungin in cases of MIC 0.5 mg l 21 and MIC 0.25 mg l 21, respectively (Pfaller et al., 2011b). A breakpoint MIC of 2 mg l 21 was used to identify caspofungin susceptible isolates (Pfaller et al., 2008). C. albicans, C. tropicalis and C. parapsilosis were categorized as fluconazole resistant in cases of MIC 8 mgl 21 ; C. glabrata was categorized as fluconazole resistant in cases of MIC 64 mg l 21 ; and all C. krusei isolates were considered as fluconazole resistant. On the basis of the recently revised clinical breakpoints for voriconazole, C. albicans, C. tropicalis and C. parapsilosis were considered resistant in cases of MIC 1 mg l 21 and susceptible dose-dependent in cases of MIC mg l 21, while C. krusei was considered resistant to voriconazole in cases of MIC 2 mgl 21 (Pfaller et al., 2011a). According to this study, due to insufficient data to demonstrate a correlation between in vitro susceptibility testing and clinical outcome for C. glabrata, the epidemiological MIC cut-off value of 1 mgl 21 was used to detect resistance in this yeast (Pfaller et al., 2011a). Resistance breakpoints for voriconazole were used as an indicator of resistance for posaconazole. An MIC.16 mg l 21 was considered indicative of resistance to flucytosine (CLSI, 2008) and an MIC 1 mgl 21 was considered indicative of resistance to amphotericin B according to Etest manufacturer suggestions. MIC profiles were analysed and correlated with data collected during the epidemiological study (Tortorano et al., 2012). The antifungal susceptibility patterns were compared with resistance data from a previous survey on candidaemia conducted in Italy in the 1990s (Tortorano et al., 2003, 2005) after stratification according to the new breakpoints. The x 2 -test was used to compare proportions and a value of P,0.05 was considered statistically significant. RESULTS AND DISCUSSION The distribution of MIC values for triazoles and echinocandins is shown in Table 1. The overall proportions of isolates that were resistant to posaconazole, voriconazole and fluconazole were 6 % (18/ 294), 7.1 % (21/294) and 12.6 % (37/294), respectively; in addition to the six intrinsically fluconazole-resistant C. krusei isolates, nine C. albicans (5.3 %), three C. glabrata (9.7 %), 13 C. parapsilosis (22.4 %), and six C. tropicalis (22 %) isolates were also found to be resistant to this azole. Cross resistance to the other triazoles was detected in six out of the nine fluconazole-resistant C. albicans isolates. Six C. glabrata isolates, five C. tropicalis isolates and one C. krusei isolate were found to be resistant to voriconazole, and eight C. glabrata and four C. tropicalis isolates were resistant to posaconazole. No resistance to echinocandin drugs was detected among the 302 Candida isolates using the new species-specific breakpoints for anidulafungin and micafungin (Pfaller et al., 2011b) and the interpretative MIC breakpoints for caspofungin proposed by Pfaller et al., (2008). These caspofungin breakpoints were adopted for the elevated MIC values obtained with this drug in our study. The reasons for these elevated MICs could be attributable to the quality of the batch and/or to the choice of solvent used for stock solutions as already reported by another study (Arendrup et al., 2011b), in which the authors noticed that, despite water being recommended as a solvent by CLSI, from a chemical point of view this drug is more soluble in DMSO (Arendrup et al., 2011b). Resistance to flucytosine (MIC.16 mg l 21 ) was detected in 2 % of the isolates, including two isolates of C. albicans, two of C. tropicalis, one of C. krusei and one of C. lusitaniae, and all the tested isolates showed an MIC,1 mg l 21 of amphotericin B. On the contrary, as reported in other recent surveys (Pfaller et al., 2011b; Arendrup et al., 2011a; Lockhart et al., 2011), we demonstrated that cases of decreased susceptibility to fluconazole were frequent and increasing. Overall, the median rate of cases of fluconazole resistance in the 11 centres reporting at least ten episodes of deep-seated Candida infection was 17.8 %, with proportions of cases ranging from 4.3 to 28.6 %. The comparison of the rates of resistance found in the present study with those reported in a previous survey on candidaemia using Candida strains isolated in Italy in (Tortorano et al., 2003, 2005), after a stratification of resistance according to the recently adopted breakpoints, revealed a significant increase in the proportion of isolates with decreased susceptibility to fluconazole from an overall rate of 6.6 % in the 1990s to 12.6 % in (P50.004). This trend is not attributable to a change in species distribution as the proportion of nonalbicans aetiologies remained stable over the 10-year period of (41 % vs 44 %), but it seems to be due to an increase in resistant isolates of species previously considered to be highly susceptible, such as C. parapsilosis. While the fluconazole resistance rate among C. albicans isolates was similar between the two time periods (5.3 % vs 6.7 %), the fluconazole resistance rate among non-albicans Candida isolates was significantly higher in the more recent survey (22.9 % vs 6.5 %; P,0.001). C. tropicalis and C. parapsilosis accounted for more than half of all fluconazole resistant isolates in the present survey (19/37) and the proportion of the resistant isolates 390 Journal of Medical Microbiology 61

3 Antifungal susceptibility of invasive Candida isolates Table 1. In vitro susceptibilities of 302 Candida isolates from blood or other normally sterile body sites 2, Concentrations not investigated; NA, no breakpoints available. Species (no. isolates tested) Antifungal agent No. of isolates with MIC (mg l 1 ) of: % R C. albicans (169) Anidulafungin Caspofungin Micafungin Fluconazole Posaconazole Voriconazole C. parapsilosis (58) Anidulafungin Caspofungin Micafungin Fluconazole Posaconazole Voriconazole C. glabrata (31) Anidulafungin Caspofungin Micafungin Fluconazole Posaconazole Voriconazole C. tropicalis (27) Anidulafungin Caspofungin Micafungin Fluconazole Posaconazole Voriconazole C. krusei (6) Anidulafungin Caspofungin Micafungin Fluconazole Posaconazole Voriconazole Other spp.* (8) Anidulafungin NA Caspofungin NA Micafungin NA Fluconazole NA Posaconazole NA Voriconazole NA *Other species include C. lusitaniae (4 isolates), C. guilliermondii (2), C. kefyr (1) and C. dubliniensis (1). increased over the 10-year period of from 7 to 22 % (P50.127) and from 2 to 25.8 % (P,0.001) for C. tropicalis and C. parapsilosis, respectively. However, the data concerning C. parapsilosis may be biased by the elevated fluconazole MIC values (8 and 16 mg l 21 ) of the strains isolated during an outbreak that occurred in one centre. If the susceptibility data of the nine outbreak isolates are excluded from the analysis, 8.2 % of the isolates showed resistance and the increase in the proportion of resistance compared to the survey of the 1990s is not statistically significant. Fluconazole resistance among C. parapsilosis isolates is not a rare event and appears to be increasing worldwide, as noted by the SENTRY Antimicrobial Surveillance Program (Pfaller et al., 2011c). The resistance to fluconazole but not to the echinocandins that was observed in our survey was considered a consequence of increased drug use to combat this species, resulting from the recommendations of the Infectious Diseases Society of America; therefore, a close monitoring of this species in the future has been strongly recommended (Pfaller et al., 2011c). In 10 out of 23 patients receiving fluconazole prophylaxis, a resistant isolate caused a breakthrough infection, providing further evidence that 391

4 A. M. Tortorano and others prior exposure to the antifungal is associated with the risk of isolates developing a resistance. In our survey, no patients had been previously exposed to echinocandins and this may be the reason why resistance was not detected. Fluconazole resistance seems to be more frequently acquired outside the ICU setting, even if not statistically significant, as it occurred in 27 out of 257 (10.5 %) infections that were considered to be ICU-acquired and in 7 out of 45 (15.5 %) infections that were already present at admission. Resistance appeared to be associated with high mortality rates. The crude mortality rate in patients infected by nonsusceptible isolates was 60.5 % compared to an overall crude mortality rate detected during the survey of 46 %, however the difference was not statistically significant (P50.179). In conclusion, this 2-year survey has revealed that, while resistance to echinocandins and amphotericin B is irrelevant, fluconazole resistance seems to be an important problem in the Italian ICU setting, and the problem relates not only to intrinsically resistant species but also to isolates of species that were previously considered to be susceptible, such as C. parapsilosis. Routine antifungal susceptibility testing of this species is therefore recommended. Since MIC breakpoints are continuously evolving it would be useful for susceptibility results to be published as MIC distribution rather than as synthesized data in order to enable retrospective comparison or comparison with results of different studies. ACKNOWLEDGEMENTS Members of the ECMM-FIMUA Study Group: A. Astolfi, G. Breda, L. Gasparini, C. Brambilla, A. De Gasperi, G. Marino, S. Morero, C. Ossi, G. Giuliani, A. Corona, S. Cozzi, M. Tejada, M. Dei Poli, C. Bellucci, M. Favaro, C. Vismara, S. Perin, M. Cigada, M. Cainarca (Milano), L. Crema, L. Ferrari (Cremona), C. Cavanna, E. Carretto, V. Emmi (Pavia), G. Delvecchio (Bergamo), A. Grossi, S. Ferraris (Treviglio), C. Bezzi (Magenta), G. Pinsi, S. Magri, L. Signorini (Brescia), P. Casella, C. Fontana, S. Magni (Vimercate), A. Ceraminiello, S. Protti (Lodi), E. Costanzo, G. Monticone (Asti), M. Sanguinetti (Roma), G. Lo Cascio, L. Gottin (Verona), G. Bonaccorso, M. Scarin, A. Cavallaro (Padova), C. Venturelli, L. Donno, S. Assenza (Modena), E. Manso, C. Passatempi (Ancona), F. Di Bernardo, S. Giordano, R. Monastero, D. Palma (Palermo), V. Cusumano, A. David (Messina), M. G. Garau, S. Falchi, C. Cannas (Cagliari), S. Fracchiolla, P. Quaranta (Taranto), G. Defina, L. Pagani, A. Dietmar (Bolzano) and M. L. Deiana (Trieste). This study was supported, in part, by a grant from Merck Sharp & Dohme, Rome, Italy. REFERENCES Arendrup, M. C., Bruun, B., Christensen, J. J., Fuursted, K., Johansen, H. K., Kjaeldgaard, P., Knudsen, J. D., Kristensen, L., Møller, J. & other authors (2011a). National surveillance of fungemia in Denmark (2004 to 2009). J Clin Microbiol 49, Arendrup, M. C., Rodriguez-Tudela, J. L., Park, S., Garcia-Effron, G., Delmas, G., Cuenca-Estrella, M., Gomez-Lopez, A. & Perlin, D. S. (2011b). Echinocandin susceptibility testing of Candida spp. using EUCAST EDef 7.1 and CLSI M27-A3 standard procedures: analysis of the influence of bovine serum albumin supplementation, storage time, and drug lots. Antimicrob Agents Chemother 55, CLSI (2008). Reference method for broth dilution antifungal susceptibility testing of yeasts, 3rd edn. Document M27 A3. Wayne, PA: Clinical and Laboratory Standards Institute. Kett, D. H., Azoulay, E., Echeverria, P. M. & Vincent, J. L. for the Extended Prevalence of Infection in ICU Study (EPIC II) Group of Investigators (2011). Candida bloodstream infections in intensive care units: analysis of the extended prevalence of infection in intensive care unit study. Crit Care Med 39, Lockhart, S. R., Wagner, D., Iqbal, N., Pappas, P. G., Andes, D. R., Kauffman, C. A., Brumble, L. M., Hadley, S., Walker, R. & other authors (2011). Comparison of in vitro susceptibility characteristics of Candida species from cases of invasive candidiasis in solid organ and stem cell transplant recipients: Transplant-Associated Infections Surveillance Network (TRANSNET), 2001 to J Clin Microbiol 49, Pappas, P. G. & Mycoses Study Group (2011). Candidemia in the intensive care unit: miles to go before we sleep. Crit Care Med 39, Pfaller, M. A., Diekema, D. J., Ostrosky-Zeichner, L., Rex, J. H., Alexander, B. D., Andes, D., Brown, S. D., Chaturvedi, V., Ghannoum, M. A. & other authors (2008). Correlation of MIC with outcome for Candida species tested against caspofungin, anidulafungin, and micafungin: analysis and proposal for interpretive MIC breakpoints. J Clin Microbiol 46, Pfaller, M. A., Andes, D., Diekema, D. J., Espinel-Ingroff, A., Sheehan, D. & CLSI Subcommittee for Antifungal Susceptibility Testing (2010). Wild-type MIC distributions, epidemiological cutoff values and species-specific clinical breakpoints for fluconazole and Candida: time for harmonization of CLSI and EUCAST broth microdilution methods. Drug Resist Updat 13, Pfaller, M. A., Andes, D., Arendrup, M. C., Diekema, D. J., Espinel- Ingroff, A., Alexander, B. D., Brown, S. D., Chaturvedi, V., Fowler, C. L. & other authors (2011a). Clinical breakpoints for voriconazole and Candida spp. revisited: review of microbiologic, molecular, pharmacodynamic, and clinical data as they pertain to the development of species-specific interpretative criteria. Diagn Microbiol Infect Dis 70, Pfaller, M. A., Diekema, D. J., Andes, D., Arendrup, M. C., Brown, S. D., Lockhart, S. R., Motyl, M., Perlin, D. S. & CLSI Subcommittee for Antifungal Testing (2011b). Clinical breakpoints for the echinocandins and Candida revisited: integration of molecular, clinical, and microbiological data to arrive at species-specific interpretive criteria. Drug Resist Updat 14, Pfaller, M. A., Moet, G. J., Messer, S. A., Jones, R. N. & Castanheira, M. (2011c). Candida bloodstream infections: comparison of species distributions and antifungal resistance patterns in communityonset and nosocomial isolates in the SENTRY Antimicrobial Surveillance Program, Antimicrob Agents Chemother 55, Tortorano, A. M., Rigoni, A. L., Biraghi, E., Prigitano, A., Viviani, M. A. & FIMUA-ECMM Candidaemia Study Group (2003). The European Confederation of Medical Mycology (ECMM) survey of candidaemia in Italy: antifungal susceptibility patterns of 261 non-albicans Candida isolates from blood. J Antimicrob Chemother 52, Tortorano, A. M., Prigitano, A., Biraghi, E. & Viviani, M. A. on behalf of the FIMUA-ECMM Candidaemia Study Group (2005). The 392 Journal of Medical Microbiology 61

5 Antifungal susceptibility of invasive Candida isolates European Confederation of Medical Mycology (ECMM) survey of candidaemia in Italy: in vitro susceptibility of 375 Candida albicans isolates and biofilm production. J Antimicrob Chemother 56, Tortorano, A. M., Dho, G., Prigitano, A., Breda, G., Grancini, A., Emmi, V., Cavanna, C., Marino, G., Morero, S. & other authors (2012). Invasive fungal infections in the intensive care unit: a multicentre, prospective, observational study in Italy ( ). Mycoses 55,