A

AeHfltaut©

K

1

gIhailttpotdAs

0d

International Journal of Antimicrobial Agents 29 (2007) 108–111

Short communication

Susceptibility of environmental versus clinical strainsof pathogenic Aspergillus

Ricardo Araujo a,∗, Cidalia Pina-Vaz a, Acacio Goncalves Rodrigues a,b

a Department of Microbiology, Faculty of Medicine, University of Porto, Alameda Prof. Hernani Monteiro, 4200-319 Porto, Portugalb Burn Unit, Department of Plastic and Reconstruction Surgery, Hospital S. Joao, Porto, Portugal

Received 14 August 2006; accepted 26 September 2006

bstract

The objective of this study was to compare the antifungal susceptibility profiles of 307 environmental strains and 139 clinical isolates ofspergillus belonging to six different species. Clinical and environmental strains with minimal inhibitory concentrations (MICs) or minimalffective concentrations ≥4 �g/mL to amphotericin B (AMB), itraconazole (ITC), voriconazole and caspofungin were seldom detected.owever, the susceptibility profile of environmental Aspergillus non-fumigatus strains suggests a native reduced susceptibility of Aspergillusavus and Aspergillus terreus to AMB. A single environmental strain of Aspergillus nidulans showed high in vitro resistance (MIC ≥ 16 �g/mL)o ITC. Aspergillus niger showed significantly higher MIC values to ITC compared with the other Aspergillus spp. Conversely, A. fumigatus

nd Aspergillus glaucus showed higher susceptible profiles to the antifungals. Definition of the breakpoints for the antifungal agents remainsrgent. The relationship between the susceptibility pattern and the pathogenic potential also deserves more attention, particularly with regardo non-fumigatus species.

2006 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.

onazol

AsnAvpHta

2

eywords: Aspergillus fumigatus; Aspergillus flavus; Amphotericin B; Itrac

. Introduction

Nowadays, invasive aspergillosis is an increasing fun-al infection in intensive care, transplant and burn units.mmunosuppression, particularly persistent neutropenia origh-dose corticosteroids, represents a serious risk factorssociated with invasive Aspergillus infections [1]. Minimalnhibitory concentrations (MICs) remain the most valuableaboratory tool available to predict resistance and to selecthe best alternative for antifungal treatment. The full rela-ionship between antifungal susceptibility and pathogenicotential remains unknown, however reduced susceptibilityf strains to antifungals has been shown to be closely relatedo higher mortality rates in murine models [2,3]. Many studies

etermining the susceptibility patterns of clinical strains ofspergillus spp. are available, but very few have evaluated theusceptibility patterns of environmental strains, especially of

∗ Corresponding author. Tel.: +351 91 603 5076; fax: +351 22 551 0119.E-mail address: ricjparaujo@yahoo.com (R. Araujo).

2

ic22

924-8579/$ – see front matter © 2006 Elsevier B.V. and the International Societyoi:10.1016/j.ijantimicag.2006.09.019

e; Voriconazole; Caspofungin; Susceptibility testing

spergillus non-fumigatus strains [4–7]. The objective of thistudy was to determine the susceptibility profiles of a largeumber of environmental strains of six different species ofspergillus to amphotericin B (AMB), itraconazole (ITC),oriconazole (VRC) and caspofungin (CPF) and to com-are them with the susceptibility profiles of clinical strains.ypothetical susceptibility differences among environmen-

al Aspergillus spp. might explain native resistance to certainntifungals.

. Material and methods

.1. Organisms and isolation sources

Four hundred and forty-six Aspergillus strains were stud-

ed, comprising A. fumigatus (105 environmental and 66linical strains), Aspergillus flavus (59 environmental and9 clinical strains), Aspergillus niger (46 environmental and0 clinical strains), Aspergillus terreus (27 environmental

of Chemotherapy. All rights reserved.

R. Araujo et al. / International Journal of Antimicrobial Agents 29 (2007) 108–111 109

Table 1Origin of clinical strains of Aspergillus spp.

Species S. Joao Hospital a CBS b IHEM c S. Antonio Hospital d Other sources e

A. fumigatus 58 0 0 7 1A. flavus 18 0 5 3 3A. niger 10 3 6 0 1A. terreus 2 0 4 1 6A. nidulans 1 2 5 0 2A. glaucus 1 0 0 0 0

a University hospital, north of Porto, Portugal.b Centraalbureau voor Schimmelcultures, Utrecht, The Netherlands.

nageme

ospital)

amewiAsrAfftoAlomafplrlcci

2

(SpImriCat[

2

SfgvSc0

3

a(agssractaipcMtMisltMo

c Scientific Institute of Public Health (L’Institut des Hautes Etudes en Mad University hospital, south of Porto, Portugal.e University hospital, Coimbra, Portugal; S. Maria Hospital (University h

nd 13 clinical strains), Aspergillus nidulans (20 environ-ental and 10 clinical strains) and Aspergillus glaucus (50

nvironmental and 1 clinical strain). Six additional strainsere also included for quality control of susceptibility test-

ng: reference strains A. fumigatus ATCC 46645 and A. nigerTCC 16404; and strains A1 and A3 (itraconazole-resistanttrains of A. fumigatus) and A7 and A8 (voriconazole-esistant strains of A. fumigatus), kindly provided by Drna Espinel-Ingroff. Environmental strains were collected

rom the air/water of 17 wards and intensive care units androm three different administrative departments of Hospi-al S. Joao, Porto, during the previous 4 years. The ratiof environmental strains used in this study was 4/5 of eachspergillus spp. collected from clinical areas and 1/5 col-

ected from non-clinical areas. Three environmental strainsf A. nidulans obtained from Centraalbureau voor Schim-elcultures were also included. Details of the clinical strains

re shown in Table 1, with the 139 clinical strains isolatedrom 130 patients (29 patients diagnosed with proved orrobable invasive aspergillosis). Clinical strains were iso-ated from the lower respiratory tract (106 isolates), upperespiratory tract (5 isolates), human biological fluids (7 iso-ates), eye (5 isolates), surgical wounds (6 isolates), earanal (4 isolates), central venous catheters (3 isolates), dis-overtebral puncture biopsies (2 isolates) and finger nail (1solate).

.2. Susceptibility testing

Determination of the MIC for the antifungal agents AMBBristol-Myers SP, Dublin, Ireland), ITC (Janssen-Cilag,aunderton, UK) and VRC (Pfizer Inc., New York, NY) waserformed according to Clinical and Laboratory Standardsnstitute (CLSI) M38-A protocol for susceptibility testing ofoulds [8]. Absence of visual growth defined the MIC. In

are cases of suspected trailing phenotype, 90% reductionn growth defined the MIC to avoid false MIC values [9].

PF (Merck Sharp & Dohme, Hoddesdon, UK) activity waslso determined according to CLSI M38-A protocol [8], buthe minimal effective concentration (MEC) was determined10,11].

aAci

nt), Brussels, Belgium.

, Lisbon, Portugal; Health Sciences Centre, Winnipeg, Manitoba, Canada.

.3. Statistical analysis

Excel 2000 (Microsoft Corp., New York, NY) andPSS 11.5 (SPSS Inc., Chicago, IL) software were usedor data elaboration, analysis and determination of theeometric mean for MIC and MEC values. Analysis ofariance (ANOVA), using the Bonferroni correction, andtudent’s t-test for paired samples were used for statisti-al analysis. Data were compared at a significance level of.05.

. Results and discussion

ATCC control strains of A. fumigatus and A. niger showedMIC < 1 �g/mL for all tested antifungals, except ITC

2 �g/mL for both strains). Aspergillus fumigatus strains A1nd A3 presented a MIC ≥ 16 �g/mL for ITC, as did A. fumi-atus strains A7 and A8 to VRC; however, these strains wereusceptible to the other antifungals (MIC < 1 �g/mL). Table 2hows the susceptibility patterns of all the clinical and envi-onmental strains of Aspergillus spp. to the tested antifungalgents. Based on their own experience, many authors usuallyonsider a strain with a MIC or MEC ≤ 1 �g/mL as suscep-ible [3,11]; the definitive breakpoints to antifungal agentsre not yet fully established. Resistance was rarely detectedn this study. A single strain of A. nidulans (environmental)resented high in vitro resistance to ITC (MIC ≥ 16 �g/mL,onfirmed in two different assays). The highest detectedIC values among environmental strains corresponded to

wo strains (one A. niger and one A. nidulans) showingICs ≥ 4 �g/mL to ITC, three strains of A. terreus show-

ng MICs ≥ 4 �g/mL to AMB and two strains of A. nidulanshowing a MEC of 4 �g/mL to CPF. Amongst the clinical iso-ates, two strains of A. flavus showed a MIC value of 4 �g/mLo AMB and ITC, and two strains of A. nidulans each showed

EC values of 4 �g/mL and 8 �g/mL to CPF. Comparisonf the percentage of strains with a MIC or MEC ≤ 1 �g/mL

mong all Aspergillus spp. revealed that A. fumigatus and. glaucus were three-fold more susceptible. No evidence ofross-resistance was found among the strains, and no signif-cant difference was found when comparing the MIC values

110 R. Araujo et al. / International Journal of Antimicrobial Agents 29 (2007) 108–111

Table 2Susceptibility patterns determined for 446 strains of Aspergillus to amphotericin B (AMB), itraconazole (ITC), voriconazole (VRC) and caspofungin (CPF)

Species Clinical strains (n = 139) Environmental strains (n = 307)

n MIC range(�g/mL)

Mean % of strains withMIC ≤ 1 �g/mL

n MIC range(�g/mL)

Meana % of strains withMIC ≤ 1 �g/mL

A. fumigatusAMB 66 0.25–2 0.79 97 105 0.25–2 0.77 94ITC 0.25–2 0.77 97 0.25–2 0.69 96VRC 0.25–2 0.52 98 0.125–2 0.54 99CPFb 0.125–2 0.62 92 0.06–2 0.36 98

A. flavusAMB 29 0.25–4 1.47 48 59 0.125–2 0.98 81ITC 0.125–4 0.88 90 0.06–1 0.44 100VRC 0.25–2 0.81 93 <0.03–2 0.66 98CPFb <0.03–2 0.41 97 0.06–2 0.33 98

A. nigerAMB 20 0.125–0.5 0.24 100 46 0.06–2 0.44 98ITC 0.06–2 1.04 70 0.06–4 1.01 78VRC 0.06–1 0.35 100 0.125–2 0.61 96CPFb 0.06–1 0.41 100 0.06–1 0.44 100

A. terreusAMB 13 0.25–2 0.98 77 27 0.5–4 1.54 56ITC 0.25–1 0.52 100 0.125–2 0.54 96VRC 0.25–1 0.52 100 0.25–2 0.90 93CPFb 0.06–0.5 0.30 100 0.06–1 0.50 100

A. nidulansAMB 10 0.125–2 0.96 70 20 0.125–2 0.79 90ITC 0.25–2 0.58 90 0.125 to ≥16 1.30 90VRC 0.125–0.25 0.19 100 0.06–2 0.37 95CPFb 0.125–8 1.68 70 0.125–4 0.81 90

A. glaucusAMB 1 2 – – 50 <0.03–2 0.17 98ITC 2 – – <0.03–1 0.15 100VRC 1 – – <0.03–2 0.26 98CPFb 0.25 – – <0.03–1 0.24 100

A. non-fumigatusAMB 73 0.125–4 0.98 70 202 <0.03–4 0.71 87ITC 0.06–4 0.83 85 0.06 to ≥16 0.60 94VRC 0.06–2 0.55 97 <0.03–2 0.55 97CPFa <0.03–8 0.56 95 <0.03–4 0.40 99

All Aspergillus spp.AMB 139 0.125–4 0.89 83 307 <0.03–4 0.73 89ITC 0.06–4 0.80 91 <0.03 to ≥16 0.63 94VRC 0.06–2 0.54 98 <0.03–2 0.55 97CPFb <0.03–8 0.51 94 <0.03–4 0.39 98

M

on

ssCrIDc

swhitt

IC, minimal inhibitory concentration.a Geometric mean.b CPF values refer to minimal effective concentration.

f environmental strains collected at different locations (dataot shown).

Our MIC range and geometric mean results for clinicaltrains did not differ from previous studies both when con-idering A. fumigatus or non-fumigatus species [4–7,10–12].omparison between environmental and clinical strains

evealed significantly higher MIC values (P < 0.05) to AMB,TC and CPF in the case of non-fumigatus clinical strains.etailed analysis of each species showed clearly distinct sus-

eptibility patterns; clinical isolates of A. flavus presented

utp[

ignificantly higher MIC values to AMB and ITC (P < 0.05),hilst clinical strains of A. nidulans presented significantlyigher MEC values to CPF (P = 0.022). Most clinical strainsncluded in this study were isolated from bronchial secretionshat had been collected before the prescription of antifungalreatment. Therefore, clinical strains with increased MIC val-

es did not result from exposure to the antifungals, especiallyo ITC as previously reported [4,13]. The assumption thatarticular strains may possess increased virulence potential7,14] can be additionally supported by these data. The rela-

l of Ant

tp

sApaaAtAboaptlTou

uArnsgpsocatp

A

fsDitpg

R

[

[

[

[

[

R. Araujo et al. / International Journa

ionship between reduced susceptibility and the infectiousotential of non-fumigatus strains deserves further attention.

Interestingly, the profile of non-fumigatus environmentaltrains suggested a native reduced susceptibility to AMB of. flavus and A. terreus, both species showing a significantercentage of strains with MIC values ≥2 �g/mL. The resultsre in accordance with recent reports describing A. terreus asn in vivo resistant species [4,12,15] or recalcitrant [15] toMB, although susceptible strains of A. terreus were sys-

ematically found in different in vitro testing studies [11,12].dditionally, environmental and clinical strains of A. nigeroth showed higher MIC values to ITC compared with thether Aspergillus spp., a fact also occurring with A. nidulans,lthough 90% of A. nidulans showed a MIC ≤ 1 �g/mL com-ared with 75% of A. niger strains. Aspergillus glaucus washe most susceptible Aspergillus spp., showing significantlyower geometric mean and MIC values to the antifungals.his species is rarely isolated from bronchial secretions orther clinical samples, and invasive infections are also veryncommon.

Resistant strains of Aspergillus thus appear to be veryncommon both in environment and clinical settings. Amongspergillus spp., A. fumigatus still remains the species

esponsible for most cases of invasive infections. In this study,either clinical nor environmental strains of A. fumigatushowed increased resistance to any of the tested antifun-als compared with the other Aspergillus spp., similar torevious results [5,6]. Although characterised as susceptible,ome non-fumigatus species showed a significant percentagef strains less susceptible to some of the antifungals, A. glau-us being the exception. Definition of the breakpoints for thentifungal agents remains urgent and more studies regardinghe susceptibility of non-fumigatus species in vivo should beerformed in the near future.

cknowledgements

The authors would like to thank Dr Ana Espinel-Ingroffor providing the itraconazole- and voriconazole-resistanttrains, and Dr Barb Weshnoweski, Dr Hortensia Sequeira,r Virginia Monteiro Lopes and Dr Rosa Velho for provid-

ng some of the strains included in the study. The authors alsohank Gloria Goncalves for the excellent technical assistancerovided throughout the study. This study was supported byrant no. 60 901 from Fundacao Calouste Gulbenkian.

[

imicrobial Agents 29 (2007) 108–111 111

eferences

[1] Marr KA, Carter RA, Boeckh M, Martin P, Corey L. Invasiveaspergillosis in allogeneic stem cell transplant recipients: changes inepidemiology and risk factors. Blood 2002;100:4358–66.

[2] Lass-Florl C, Kofler G, Kropshofer G, et al. In-vitro testing ofsusceptibility to amphotericin B is a reliable predictor of clinical out-come in invasive aspergillosis. J Antimicrob Chemother 1998;42:497–502.

[3] Warnock DW, Arthington-Skaggs BA, Li RK. Antifungal drug sus-ceptibility testing and resistance in Aspergillus. Drug Resist Update1999;2:326–34.

[4] Verweij PE, Mensink M, Rijs AJ, Donnelly JP, Meis JF, Denning DW.In-vitro activities of amphotericin B, itraconazole and voriconazoleagainst 150 clinical and environmental Aspergillus fumigatus isolates.J Antimicrob Chemother 1998;42:389–92.

[5] Guinea J, Pelaez T, Alcala L, Ruiz-Serrano MJ, Bouza E. Antifungalsusceptibility of 596 Aspergillus fumigatus strains isolated from out-door air, hospital air, and clinical samples: analysis by site of isolation.Antimicrob Agents Chemother 2005;49:3495–7.

[6] Meneau I, Sanglard D. Azole and fungicide resistance in clini-cal and environmental Aspergillus fumigatus isolates. Med Mycol2005;43(Suppl 1):S307–11.

[7] Teixeira AB, Silva M, Lyra L, et al. Antifungal susceptibility andpathogenic potential of environmental isolated filamentous fungicompared with colonizing agents in immunocompromised patients.Mycopathologia 2005;160:129–35.

[8] Clinical and Laboratory Standards Institute. Reference method for brothdilution antifungal susceptibility testing of conidium-forming filamen-tous fungi. Approved standard M38-A. Wayne, PA: CLSI; 2002.

[9] Mosquera J, Warn PA, Morrissey J, Moore CB, Gil-Lamaignere C,Denning DW. Susceptibility testing of Aspergillus flavus: inoculumdependence with itraconazole and lack of correlation between suscep-tibility to amphotericin B in vitro and outcome in vivo. AntimicrobAgents Chemother 2001;45:1456–62.

10] Arikan S, Lozano-Chiu M, Paetznick V, Rex JH. In vitro synergy ofcaspofungin and amphotericin B against Aspergillus and Fusarium spp.Antimicrob Agents Chemother 2002;46:245–7.

11] Espinel-Ingroff A. In vitro antifungal activities of anidulafunginand micafungin, licensed agents and the investigational triazoleposaconazole as determined by NCCLS methods for 12,052 fungalisolates: review of the literature. Rev Iberoam Micol 2003;20:121–36.

12] Lass-Florl C, Griff K, Mayr A, et al. Epidemiology and outcome ofinfections due to Aspergillus terreus: 10-year single centre experience.Br J Haematol 2005;131:201–7.

13] Dannaoui E, Borel E, Monier MF, Piens MA, Picot S, Persat F.Acquired itraconazole resistance in Aspergillus fumigatus. J Antimi-crob Chemother 2001;47:333–40.

14] Aufauvre-Brown A, Brown JS, Holden DW. Comparison of virulence

between clinical and environmental isolates of Aspergillus fumigatus.Eur J Clin Microbiol Infect Dis 1998;17:778–80.

15] Iwen PC, Rupp ME, Langnas AN, Reed EC, Hinrichs SH. Invasivepulmonary aspergillosis due to Aspergillus terreus: 12-year experienceand review of the literature. Clin Infect Dis 1998;26:1092–7.