Present status of the detection of antifungal resistance: the perspective

from both sides of the ocean

M. Cuenca-Estrella and J . L. Roddguez-Tudela

Servicio de Micologia, Centro Nacional de Microbiologia, Instituto de Salud Carlos 111, Majadahonda, Madrid, Spain

The NCCLS reference methodology for antifungal susceptibility testing is a new milestone of the evolution of
medical mycology. The use of t h s methodology however, is not problem-fi-ee. At present, major limitations
are a trailing phenomenon with azoles, unreliable detection of resistance to amphotericin B, poor growth of
some organisms and unpractical procedures for the clinical laboratory. Herein a overview of NCCLS
guidelines for yeasts and filamentous fungi is presented. Likewise, a review of studies conducted trying to
overcome the limitations of reference procedures is also included. Several alternative approaches are reviewed
as alternative meha, inoculum size and incubation time. Modifications of reading procedure and endpoint
determination are also evaluated. Agar di&sion methods and other methods for susceptibihty testing are cited.
Finally, we discuss the data on correlation of the in vitro results with the in vivo activity.
Clin Microbiol Znfect 2001: 7 (Supplement 2): 46-53

stuhes have been conducted to try to overcome the

INTRODUCTION
Over the years, many studies on in vitro susceptibility of fungi
have been reported [l]. Their findings presented basic
hmitations due to interlaboratory variations and the lack of
correlation between in vitro and in vivo activity [2,3]. However,
significant progress has been made in t h s field. Susceptibility
testing of fungi has recently been standardized by the National
Committee for Clinical Laboratory Standards (NCCLS),
Subcommittee on Antifungal Susceptibility Tests. At present,
these testing methods have only been standardized for Candida
spp. and C. neofoovmans (document M27-A) [4], and guidelines
have been proposed for filamentous fungi (document M38-P)
[5]. The broth dilution tests proposed by the NCCLS feature
good interlaboratory reproducibdity. This agreement is
essential to identify organisms unlikely to respond to certain
antifungal treatments.
The use of NCCLS methodology is not problem-free. At
present, the major limitation is the trading phenomenon that
makes visual determination of azole MICs more difficult,
because of partial idubition of fungal growth [6,7]. Some
limitations. Colorimetric, spectrophotometric and agar d i 6 -
sion methods have been proposed as modifications of reference
techniques [3]. The European Committee on Antibiotic
Susceptibdity Testing (EUCAST) Antifungal Susceptibility
Testing Subcommittee has performed several studies and is
conducting a cooperative study to elaborate a method for
testing the antifungal susceptibihty of fermentative species of
yeasts. Societies of Clinical Microbiology in several European
countries are also carrying out collective works [8,9]. The aim
of all the studies in progress is to develop a practical reference
methodology for clinical laboratories and a useful tool to
predct the outcome of therapy with an antifungal agent.
Here, we present an overview of the NCCLS guidelines for
yeasts and filamentous fungi, their limitations and some
modifications of these reference methods. We also review
the data on the correlation of the in vitro results with in
vivo activity.
NCCLS REFERENCE GUIDELINES
Antifungal susceptibility testing for yeasts

In 1985, the NCCLS published its first report, in which the

Corresponding author and reprint requests: M. Cuenca-Esaella,
results of a small collaborative study were presented [lo]. Since
Servicio de Micologia, Centro Nacional de Microbiologia, Instituto
de Salud Carlos 111, Ctra Majadahonda-Pozuelo Km 2, 28220 then, a reproducible reference procedure of antifungal
Majadahonda, Madrid, Spain susceptibility testing for yeasts has been developed. Several
Tel.: +34 91 509 7961
cooperative works were conducted to ascertain inoculum
Fax: +34 91 509 7966
E-mail: mcuenca-estrella@isciii.es preparation, inoculum size, choice among several synthetic

0 2001 Copyright by the European Society of Clinical Microbiology and infectious Diseases, CM/, 7 (Suppl. 2). 4653
 Cuenca-Estrella and Rodrigez-Tudela Antifungal susceptibility testing 47
media, temperature of incubation, duration of incubation,
end-point determination, reference MIC ranges for quality
control strains, and breakpoints for available antifungal agents
[2-4,11,12]. These studies established a broth macrodilution
reference procedure. A microdilution method as a modifica-
tion of the macrodilution technique was also developed in
order to be more practical for the clinical laboratory [13].
As a result, the NCCLS document M27-A was published
[4]. It describes both macrodilution and microdilution
methods for the antifungal susceptibility testing of Candida
spp. and C. neoformans. Drug stock preparation, storage and
dilution techniques are similar to antibacterial testing proce-
dures with minor modifications. A completely synthetic
medium is recommended for susceptibility testing. Undefined
media can give rise to widely varying results [6]. For this
reason, the synthetic medium RPMI-1640 with 1-glutamine
buffered to pH 7.0 with 3-[N-morpholino]-propanesulfonic
acid (MOPS) is advocated by the NCCLS. The pH must be
strictly controlled. Inoculum preparation follows a spectro-
photometric method at 530 nm, and tubes or plates are
inoculated with a final inoculum of 0.5-2.5 x lo3 CFU/mL.
It has been indicated that higher inocula could increase the
MICs for the drugs tested, and smaller inocula were associated
with improvements in interlaboratory reproducibihty [14,151.
Plates or tubes are incubated at 35 "C for 48 h for Candida spp.
and for 72 h for C. neofrmanr.
The NCCLS method recommends visual reading but this is
an important source of variability and inaccuracy [16]. For
amphotericin B, a fungicidal drug, endpoints are easily defined
and MIC is the lowest drug concentration that prevents any
discernible growth compared with the growth control (drug-free
tube or well). However, fungistatic agents (such as azoles
ketoconazole, fluconazole, itraconazole and flucytosine) show
endpoints less clearly defined and introduce sigyficant sub-
jectivity in reading results [15,17,18]. The document M27-A
proposes to quanti+ endpoint determination, so MIC is defined
for the macrodilution procedure as an 80% decrease in turbidity
as compared with growth control. For broth microdilution,
MIC is detem7ined according to the 0-4 scale, as follows:
0 0, optically clear;
0 1, slightly hazy;
0 2, prominent decrease in turbidity;
0 3, slight decrease in turbidity;
0 4, no reduction in turbimty.
MIC is defined as the lowest concentration of drugs with
w h c h the score is < 2. The quantification of endpoint
determination and agitation of solutions before reading led to
more reproducible endpoints determination [19]. Quality
control procedures and reference strains are also included in
the document [4].
0 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 7 (Suppl. 2). 46-53
Obviously, a great deal of effort has gone into the
development of the standardized method for antifungal
susceptibility testing and the reference techniques are now
more reliable and reproducible. However, the use of this
methodology still has some limitations. At present, the main
problem is the determination of MICs for azoles. This is subject
to variable interpretation, because of the trailing phenomenon
caused by partial inhibition offungal growth [18,20]. As already
stated, visual endpoint determination is a subjective operation,
particularly with fungistatic agents. Quantification of endpoint
determination and solution agitation do not settle the issue. For
some isolates, trailing growth is so significant that the MICs after
24 h are much lower than after 48 h. Furthermore, a worse
aspect is that some strains are susceptible to fluconazole and
itraconazole (MIC < 8 mg/L and MIC 6 0.12 mg/L
respectively) at 24 h, but resistant at 48 h (264 mg/L and > 1
mg/L) [18]. The relevance of this discordance is unknown but
evidence suggests that lower MICs correlate most closely with
the outcome in vioo [21]. MIC determination at 24 h only has
been proposed to solve the problem [14,16] but numerous
strains show little or no discernible growth after 24 h of
incubation [22]. Other authoe have addressed inoculum size,
reading method or medium p H [14,15,2>25]. Several studies
have used spectrophotometric determination of endpoints to
eliminate such subjective interpretation [19,20,24].Marr et al.
reported that adjustment in the pH ofmedia to acidic conditions
@H4.5) can reduce trailing in C. albicans without aEecting the
MICs of fluconazole [25]. More comprehensive studies are
needed to resolve the phenomenon of trailing and to ascertain
its effects on clinical response.
A second major limitation is poor growth of C. neoformans
and other nonfermentative yeasts with media recommended
by the NCCLS 126,271. Therefore, several authors have
advised a number of different ways of overcoming the
problem. Media such as buffered yeast nitrogen base (BYNB)
[26] or RPMI supplemented with 2% glucose have been
employed [28]. Better rates of growth were obtained with
BYNB, and MICs were not falsely elevated [29]. Odds et al.
demonstrated that oxygen is a limiting factor for C. neDfomans
growth in liquid media [30] and cultivation under constant
agitation has been recommended [28].
The other major limitation of the NCCLS methods
described for antifungal susceptibility testing is reliable
detection of resistance to amphotericin B [4]. RPMI medium
yields a range of MICs that span only three or four twofold
serial dutions [7]. This short range precludes reliable
discrimination between susceptible and resistant isolates.
Current reports suggest that testing with Antibiotic Medium
3 (AM3) instead of RPMI permits us to enhance the ability to
detect it [31]. However, the reproducibility of this method is
still under study, owing to lot-to-lot variability of AM3 [32].
48 Clinical Microbiology and Infection, Volume 7, Supplement 2. 2001
Despite its limitations, the NCCLS reference methodology
for antifungal susceptibility testing is a new d e s t o n e in the
evolution of medical mycology. It is possible, even likely, that
new antifungal agents require modified methods or that new
assays will show a better correlation between in vitro and in viuo
susceptibility results and patients' outcomes. However, at
present, the NCCLS procedure is the most reliable and
reproducible methodology, and any progress in this field must
take it into account.
Antifungal susceptibility testing for filamentous fungi
The document M38-P describes a method for testing the
susceptibility to antifungal agents of filamentous fungi
(moulds), including Aspergillus spp. Fusarium spp., Rhizopus
arrhizus, Pseudallescheria boydii and Sporothrix schenckii [5]. These
species are conidium-forming moulds. A working group was
formed and charged with the responsibility of carrying out
studles to collect data and to propose a way to perform
susceptibility testing of those fungal species. As a result of two
collaborative works, the document was published [33,34].
The M38-P methodology adopts some of the steps in yeast
testing [4]. The guidelines include both macro- and micro-
dilution methods [5]. Drug stock preparation and dilution
procedures are similar to the yeast techniques. The synthetic
medium RPMI-1640 is also advocated. Fungi are grown on
potato agar slants at 35 "C for 7 days. Inocula are prepared by a
spectrophotometric procedure at 530 nm, and tubes or plates
are inoculated with a final suspension of approximately 0.4-
5 x lo4 CFU/mL [7]. It has been indicated that inoculum
preparation as specified here provides the most reproducible
MIC data. Table 1 displays the optical density range at 530 nm
correlated with inoculum size for filamentous fungi included
in document M38-P. Plates or tubes are incubated at 35 " C for
72 h and examined daily. Readings are obtained with the aid
of a mirror and endpoint determination adopts criteria
developed for antifungal testing of yeasts [3]. Interpretative
G a b l
Species No. Optical density
observations range
Aspergillus flavus 105 0.09-0.1 1
Aspergillus fumigatus 104 0 . 0 9 4 . 11
Fusarium oxysporum 105 0.15-0.1 7
Fusarium soiani 103 0.15-0.1 7
Pseudallescherisboydii 99 0.15-0.1 7
Rhizopus arrhizus 99 0.1 5-0.1 7
Sporothrix schenckii NAb 0.09-0.1 1
Paecilomyces variotii NA 0.11-0.17
(ATCC 22315)
aData are from ref. [3].
bNA. not applicable.
Q 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, C M , 7 (Suppl. 2), 46-53
breakpoints are not avadable but some degree of correlation
has been demonstrated between in uitro test results and
response to treatment in animal models p,35].
The limitations in standardized susceptibility testing for
yeasts are considerable, and developing a reference methodol-
ogy for filamentous fungi represents a major challenge with
unsolved problems. First, the incubation period for obtaining
the inoculum lasts several days, which is a major limitation for
the clinical susceptibdity testing of moulds [2,3]. In addition,
the document M38-P describes guidelines for conidmm-
forming moulds, and conima are small spheres that are easily
quantifiable [5]. However, obtaining conidial suspensions is
not possible when the moulds present germinate to hyphal
forms. Preparation of hyphal or germinated sporangiospores
suspensions is difficult and limited data suggest that t h s is an
important source of discrepancies [36,37]. Collaborative
studies have not been conducted for nonconidium-forming
moulds and published results are less consistent in reliabihty
than data obtained with species included in the proposed
reference method [2,3,38].
Finally, as mentioned in relation to yeasts, visual reading
introduces subjective interpretations of susceptibility results.
Using the conventional criterion of complete growth inhlbi-
tion inaccuracy is minimized, but the NCCLS recommends
quantification of endpoint determination in order to avoid the
trailing phenomenon with fungistatic agents [5,33]. Data
recently reported have revealed that 100%inhibition instead of
50% can be the optimal testing condition to improve
reproducibility with azoles and even to detect potential azole
resistance in moulds. These refined gmdelines and results
obtained by the evaluation of other species of fun@ will be
included in a new NCCLS document (M38-T) [7].
MODIFICATIONS OF REFERENCE METHODS
The reference method already mentioned is not the best
technique for testing all organisms or the most convenient
e 1 Optical density ranges and inoculum
lnoculurn size
sizes for filamentous fungi included in M38-P
mean ( x 10' CFU/mL) dncumsnta
1.6
2.7
3.0
1.8
1 .o
1.3
2.4
1 .1
 Cuenca-Estrella and Rodrigez-Tudela Antifungal susceptibility testing 49

procedure for routine use in clinical laboratories. As a
consequence, a number of modifications of the NCCLS
methodology have been proposed. Several alternative ap-
proaches have been evaluated, including techniques used for
antibacterial susceptibility testing. In particular, microdilution
procedures are employed in these modifications.
Alternative media, inoculum size and incubation time
The document M27-A includes some alternative media for
special circumstances but indicates that the utility of these
modifications remains to be established. These alternative
medla are AM3, BYNB and RPMI supplemented with 2%
glucose [4]. Alternative media have been accepted, because of
limitations of the reference techmque [2,6].
The M27 methodology does not permit consistent detec-
tion of isolates resistant to amphotericin B. Several works
suggest that testing with AM3 supplemented with 2% glucose
permits more reliable detection of these isolates, but AM3 is
not standardized and substantial lot-to-lot variabibty has been
observed [31,32]. Variabihty is a basic limitation for
reproducibility. Thus, other defined synthetic medla could
represent alternatives to detect resistance to amphotericin B.
Preliminary data suggest that Iso-Sensitest could distinguish
susceptible and resistant isolates. Iso-Sensitest is a semidefined
medium for antimicrobial susceptibility testing in which
undefined components are kept to a minimal level [MI.
Ghannoum et ul. reported that the use of BYNB media may
enhance the growth of C. Neofomuns [26]. This yeast exhibits
poor growth with RPMI-1640. Thus, BYNB and an inoculum
of 10" CFU/mL have been defined as the optimal method for
determining the susceptibility of C. neoformans [29]. Inoculum
"C [28]. Some reports have indicated that the maximum rate of
growth of C. neojbmw is observed at 30 "C [40]. These reports
demonstrate the need for a more reliable and reproducible new
reference methodology of antihngal susceptibility testing of C.
neofotvnans and other nodermentative yeasts.
Another modification is the supplementation of RPMI with
2% glucose for all organisms [22]. It has been suggested that
glucose supplementation may simplifi endpoint determina-
tion, because of higher turbidity of the growth control [14,16].
In addition, numerous strains show no discernible growth after
24 h of incubation when the NCCLS reference method is
employed, and FWMI-2% glucose has the advantage of
shortening the incubation time [41]. Growth increases when
hgher inoculum sizes are employed (lo4 or lo5 CFU/mL)
and MICs are not falsely elevated [42,43]. Thus, the
Antihngal Susceptibility Testing Subcommittee of EUCAST
is c q n g out a collaborative study comparing M27-A
procedures with a new method for the determination of
MIC by broth dilution of fermentative species of yeasts. This
methodology includes RPMI-2% glucose and an inoculum
size of lo5 CFU/mL. Figure 1 and Table 2 display preliminary
data of the EUCAST cooperative work. The MICs for quality
control strains are similar to the NCCLS reference values, and
0 4 8 12.18 20 24 28 92 38 40 44 48

effects in determining MICs of antifungal agents were not IncubanonamelnhQur

observed and improvements in the clinical relevance of Fig 1 Growth kinetics of 24 clinical isolates of Candida spp. (six
antihngal MICs have been found. Other authors have different species). Comparison between NCCLS reference methodol-
recommended the cultivation of the microdilution trays under ogy (RPMI and inoculation of lo3 CFU/mL) and new method,
constant agitation and an inoculum size of lo5 CFU/mL at 35 including 2% glucose and lo5 CFU/mL. OD, optical density.

Table 2 In vitro data on qulaity

control strains Candida krusei Organism Antifungal NCCLS reference RPMI-2% glucose and
ATCC 6258 and Candida agent MIC range (mg/L) lo5 CFU/mL M I C range (rng/L)
parapsilosis ATTC 22019
Candida krusei Amphotericin 6 0.52.0 0.5-1.O
ATCC 6258 Fluconazole 16-64 16-32
ltraconazole 0.12-0.5 0.12-0.5
Ketoconazole 0.1245 0.0601500.25
Flucvtosine 4.0-16 4.0-8.0

Candida parapsilosis Amphotericin B 0.25-1 .O 0.25-1.O

ATCC 2201 9 Fluconazole 2.0-8.0 1.0-4.0
ltraconazole 0.06-0.25 0.12-0.25
Ketoconazole 0.06-0.25 0.06-0.25
Flucytosine 0.12-0.5 0.12-0.5

0 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 7 (Suppl. 2).46-53
50 Clinical Microbio/ogy and Infection, Volume 7. Supplement 2, 2001
the growth kinetics show that 24 h is an optimal incubation
period for MIC determination. These data could confirm that
the M27-A technique modified with 2% glucose and higher
inocula offers the advantages of reducing the incubation time
and simplifying endpoint determination, without sigmficant
inoculum effects in determining MICs.
Alternative mema are not described in the document M38-
P for moulds. However, it has been suggested that inoculum
preparation by a spectrophotometric procedure can be a
source of inaccuracy, owing to the color and size of spores,
which are species dependent [36-381. Therefore, individual
features of strains could have an influence on the optical
density of inoculum suspensions. Reports have indicated that
inoculum size should be adjusted by using a hemacytometer
cell-counting chamber [44,45]. Inoculum quantification has to
be performed by plating serial dilutions of suspensions. Other
authors have reported that the hemacytometer count is less
consistent in between-laboratory tests than the spectrophoto-
metric procedure [6]. RPMI-2% glucose and a higher
inoculum size have also been employed for antifungal
susceptibility testing of moulds, being an acceptable means
of evaluating the in vitro activity of antifungal agents against
these organisms [46,47].
Modifications of reading procedure and endpoint determination
An alternative to the reference method of visually gradmg
turbidity is the determination of endpoints with automated
spectrophotometric methods. Microtiter trays and automated
reading allow the determination of endpoints at hfferent levels
of growth inhibition [14]. Several studies have assessed various
levels of growth inhibition, in comparison with that of the
drug-free control well [15,19,20,24]. Levels of 50% with the
azoles and 80-90% with amphotericin B have provided the
most accurate values [3]. Excellent agreement has been
observed between the results of the spectrophotometric and
standard methods, but most studies tested exclusively C.
albicans. Therefore, the collaborative work of the Antifungal
Susceptibility Testing Subcommittee of EUCAST is assessing
both automated and visual methods with several yeasts species.
A novel modification is the use of a colorimetric oxidation-
reduction indicator [2]. This technique is used in bacterial
reference methods and allows the determination of viable cells.
Several indicators have been employed, but only the use of
Alamar Blue (Alamar Biosciences, Inc., Sacramento, USA) has
been analyzed in depth. Alamar Blue is an indicator that
changes color from blue to red when reduced in the presence
of microbial growth. It has been incorporated into the
Sensititre Yeast One Antifungal Panel (Accumed Intema-
tional, Ohio, USA), which is a disposable microdilution tray,
and numerous studies have been carried out [48,49].
0 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CM/, 7 (Suppl. 2). 4 6 5 3
The indicator aids in the determimoon of visual MIC
endpoints and could be a more convenient procedure than the
standard method for use in the clinical laboratory. The agreement
of colorimetricmicrodilution MICs with the NCCLS reference
procedures has been acceptable, with the exception of some
species (C. albicans for azoles and flucytosine or C.n e o f m n s for
amphotericin B) [49]. For filamentous fungi, colorimetricMICs
have also been determined and excellent agreement has been
observed. Moreover, colorimetric MICs afier 48-72 h of
incubation with an inoculum size of lo4 CFU/mL have been
defined as optimal testing conditions and proposed as guidelines
for a reference method for testing moulds [34].
Agar diffusion and other methods for susceptibility testing
Disk-diffusion tests and E-test strips (As Biodisk, Solna,
Sweden) have been employed for antifungal susceptibdity
testing. These procedures are straightforward, economical to
perform and offer a practical method for the clinical laboratory
[2,3]. However, agreement with reference methods is species
and medium dependent. Disk-diffusion tests, such as Neo-
sensitabs (Rosco Diagnostica, Taastrup, Denmark), have been
employed in several studies [43,50,51]. Most of these
exclusively evaluated disks of fluconazole and Candida spp.
Work has also been conducted to assess disk-ditrusion
susceptibility testing of dermatophytes [52]. E-test strips have
been more extensively analyzed [2,3,53,54].This appears to be
a suitable alternative procedure for testing the susceptibility of
yeasts and moulds to azoles and amphotericin B. However,
poor agreement with the reference methods has been reported
for Candida glabrata, C. tropicalis, C. neofomzans and some
filamentous fungi. At present, the role of agar diffusion
methods is not clearly defined, although it may have promise.
Other methodologes for testing the susceptibihty of yeasts
and moulds involve either direct determination of cell mass or
cell viability by fluorescence, or indirect determination of cell
mass by calorimetry, analysis of ATP production or radiometry
[55-571. Expensive equipment is required, some methods are
very laborious and some involve subjective interpretation.
Agreement with reference procedures is not evaluated in some
of published reports. More comprehensive studies are needed
to determine the usefulness of these methodologies.
Finally, several novel commercial systems for antifungal
susceptibility testing have been developed. They are based on
microtitration or colorimetric procedures. The usefulness of
these techniques remains undetermined [58,59].
CLINICAL CORRELATION AND RESISTANCE DETECTION
In vitro susceptibility testing should reliably predict the in uivo
response to therapy in human infections. However, numerous
 Cuenca-Estrella and Rodrigez-Tudela Antifungal susceptibility testing
factors af3ect the clinical response to antifungal therapy,
regardless of the MIC endpoint for the infecting organism.
Host factors may have more effect on the infection than
susceptibility to antifungal agents, and a low MIC does not
necessarily predict clinical success [7]. However, an in vitro
susceptibility test may be able to identify organisms unlikely to
respond to certain antimicrobial agents.
Evidence of a correlation between clinical outcome and
antifungal susceptibility testing has been limited untd recently.
Good interlaboratory reproducibility of the NCCLS reference
guidelines for yeasts has led to the establishment of
interpretative breakpoints for fluconazole, itraconazole and
flucytosine, but for Cundidu species only (Table 3) [4].
Breakpoints for fluconazole were established following
correlation between azole MICs and the clinical response of
patients with AIDS, and oropharyngeal candidosis and non-
neutropenic patients with candidemia or visceral infection [l-
3,7]. Itraconazole breakpoints were defined with clinical data
on oropharyngeal candidosis only [12]. Breakpoints for
flucytosine were established on the basis of historical data
and the pharmacokinetics of the drug, because of the shortage
of available in vitvo results obtained by the NCCLS methods
[4]. Breakpoints for amphotericin B and other antifungal
agents have not been proposed. Recent reports point out that
either E-test MICs or minimal hngicidal concentrations are
better predictors of clinical failure to amphotericin B therapy
than are the reference NCCLS procedures [7,60]. However,
these methodologes have not been standardized.
For the f3amentous fmgi, interpretative breakpoints are not
available. Historically, it has been suggested that an amphotericin
B MIC > 1 mg/L could correlate with a lack of response of
treatment in experimental mould infections [1,2,7]. However,
these data have been reviewed recently. An NCCLS study using
the M38-P methodology has indicated some degree of
correlation between in oitm tests and response to treatment in
animal models. Lack of response was observed with R. urrhizus
and Aspqillus spp. isolates for whch MICs of amphotericin B
were 2 2 mg/L and MICs of itraconazole >1 mg/L [35].
However, other authors report a lack of correlation of in oitro
amphotericin B susceptibihty testing with outcome in a murine
model of Aspergdlus infection [61]. Some data are also available
from clinical stumes. A certain correlation was observed for 29
Table 3 Interpretative guidelines for
susceptibility testing in vitro of Candida Antifungal Susceptible
agent
species
Fluconazole 68
ltraconazole G0.125
Flucytosine <4
Breakpoints in mg/rnL. Data are from ref. [4].
0 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI,7 (Suppl.
51
patients with hematological diseases and Aspgillus spp. infection.
Mortality was sigdcantly associated with MICs of amphotericin
B 2 2 mg/L. The relationship was more sigdcant for patients
infected by Asperrgillus terreus [62]. In addition, Denning et al.
have identified strains of A. &migatus resistant to itraconazole.
These strains showed MICs of itraconazole 2 16 mg/L and were
isolated from patients who had fsjled itraconazole treatment [63].
The results were confirmed in a neutropenic mouse model of
invasive aspergdlosis [64].
It must be emphasized that breakpoints are tentative and
more data are needed. The clinical usefulness of in vitro tests
must be established in other clinical settings and for other
organism-drug combinations. More reliable detection of
resistance to amphotericin B is clearly needed. However, the
currently available in oitro tests can detect both intrinsic and
primary resistance (moulds to fluconazole, Trichosporon spp. to
amphotericin B, Cundidu knrsei to fluconazole and others). The
reference methodology is also able to detect multiresistant
organisms (Scedospon'um prol$cam and other species of moulds)
and, moreover, breakpoints for fluconazole and itraconazole
have clinical relevance in oropharyngeal candidal infection in
HIV-infected patients [1,7,65,66].
CONCLUSIONS
Considerable progress has been achieved in antihngal
susceptibility testing. A reproducible reference methodology
has been developed by the NCCLS. These procedures are not
problem-free, but are essential to identify organisms unlikely
to respond to certain antifungal treatments. At present, major
limitations are the trailing phenomenon with azoles, unreliable
resistance to amphotericin B detection, poor growth of some
organisms and impractical procedures for the clinical labora-
tory. Several collaborative studies are in progress with the
intention to overcome the limitations and to develop a
practical reference methodology and a useful clinical tool.
Good interlaboratory reproducibility of the the NCCLS
reference guidelines for yeasts has led to the establishment of
interpretative breakpoints for fluconazole, itraconazole and
flucytosine. Resistant breakpoints for fluconazole and itraco-
nazole have been well documented, with treatment failure
d y in HIV patients with oropharyngeal candidosis.
Susceptible-dose Intermediate Resistant
dependent
16-32 - > 64
0.25-0.5 - 21
- 8-1 6 > 32
Z),46-53
sz Clinical Microbiology and Infection, Volume 7. Supplement 2, 2001
Correlations in other clinical settings are unknown. Further
studies are needed to evaluate the clinical usefulness of the in
vitro tests for other species of yeasts and filamentous fun@.
REFERENCES
2. Graybill J R , Revankar SG, Patterson TF. Antifungal agents and antifungal
susceptibility testing. In: Ajello L, Hay RJ, eds. Topley and Wilsons.
Microbiology and Microbial Infertiom, 9th edn. London: Arnold, 1998: 163-75.
3. Espinel-lngroff A, White T et al. Anhfungal agents and susceptibility tests.
In: Murray P R , Baron EJ, Pfaller MA, Tenover FC, Yolken R H , eds.
Manual .f Clinical Microbiolofl, 7th edn. Washington, DC: ASM Press,
1999: 1640-52.
4. National Committee for Clinical Laboratory Standards. Reference method
for broth dilution antifungal susceptibility testing of yeasts. Approved
standard M, 27 A. Wayne, PA. National Committee for Clinical
Laboratory Standards, 1997.
5 . National Committee for Clinical Laboratory Standards. Reference method
for broth dilution antifungal susceptibility testing of conidium-forming
filamentous fungi. Proposed standard M38-P. Wayne, P A National
Committee for Clinical Laboratory Standards, 1998.
6. Rex JH, Pfaller MA, Rinaldi MG, Polak A, Galgiani JN. Antifungal
susceptibility testing. Clin Microbiol Rev 1993; 6: 367-81.
7. Espinel-Ingroff A. Clinical utility of in vitro antifungal susceptibility testing.
Rev Esp Quimioter 2000; 13: 161-6.
8. Schmalreck AF. Susceptibility testing of fluconazole: evaluation of a
multicenter study of the working group Clinical Mycology of the German
Speaking Mycological Society. Mycoses 1996; 39 (Suppl. 2): Sl-11.
9. Rodriguez-Tudela JL, Cuenca-Estrella M. Grupo de trabajo para el estudio
de fungemias. Estudio multic&ntrico sobre fungemias por levaduras en
Espafia (abril-junio 1997). Reu Clin ESP 1999; 1 9 9 3 5 6 4 1 .
10. National Committee for Clinical Laboratory Standards. Antifungal
susceptibility testing; committee report. Document M2O-CR. Villanova,
P A National Committee for Clinical Laboratory Standards, 1985.
11. Pfaller MA, Bale M, Buscbelman B ef al. Quality control guidelines for
National Committee for Clinical Laboratory Standards recommended
macrodilution testing of amphotericin 8, fluconazole and flucytoine. J Clin
Microbiol 1995; 33: 1104-7.
12. Rex JH, PWer MA, Galgiani J N ef al. Development of interpretative
breakpoints for antifungal susceptibility testing: conceptual framework and
analysis of in-vitro-in vivo correlation data for fluconazole, itraconazole,
and Candida infections. Clin Ifect Dis 1997; 24: 23547.
13. Espinel-Ingroff A, Kish CW, Kerkering TM et al. Collaborative
comparison of broth macro-and microdilution antifungal tests. J Clin
Microbiol 1992; 30: 3138-45.
14. Odds FC, Messer SA, Woestenborghs F. Antifungal susceptibility testing of
yeasts: evaluation of technical variables for tests automation. Antimicrob
Agents Chemother 1995; 39: 2051-60.
15. Nguyen MH, Yu CY. Influence of incubation time, inoculum size, and
glucose concentration on spectrophotometric endpoint determinations
for amphotericin B, fluconazole, and Itraconazole. J Clin Microbiof 1999;
37: 141-5.
16. Lozano-Chiu M, Atikan S, Paetznick VL, Anaissie EJ, Rex JH. Optimizing
voriconazole susceptibility testing of Candida: effects of incubation time,
endpoint rule, species of Candida, and level of fluconazole susceptibility.J
Clin Microbiol 1999; 37: 2755-9.
17. Tomatore MA, Noskin GA, Hacek DM, Obias AA, Peterson LR. Effects of
incubation time and buffer concentration on in vitro activities of antifungal
agents against Candida albicans. J Clin Microbiol 1997; 35: 1473-6.
18. Revankar SG, Kirkpatnck WR, Mcatee RK et a/. Interpretation of trailing
0 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 7 (Suppl. 2). 46-53
endpoints in antifungal susceptibilitytesting by the National Committee for
Clinical Laboratory Standards meth0d.J Clin Minobiol 1998; 3 6 153-6.
19. Pfaller MA, Messer SA, Coffmann S. Comparison of visual and spectro-
photomemc methods of MIC endpoint determinations by using broth
microdilution methods to tests five antifungal agents, including the new
mazole D0870. J Clin Microbial 1995; 33: 1094-7.
20. Espinel-lngroff A, Rodriguez-Tudela JL, Martinez-Suarez JV. Comparison
of two alternative microdilution procedures with the National Committee
for Clinical Laboratory Standards reference macrodilution method M27-P
for in vitro testing of fluconazole-resistant and -susceptible isolates of
Candida albicms. J Clin Microbiol 1995; 33: 3154-8.
21. Rex JH, Nelson PW, Paetznick VL, Lozano-Chiu M, Espinel-lngroff A,
Anaissie EJ. Optimizing the correlation between results of testing in vitro
and therapeutic outcome in vivo for fluconazole by testing critical isolates
in a murine model of invasive candidiasis. Antimicrob Agents Chemother 1998;
42: 129-34.
22. Rodriguez-Tndela JL, Martinez-Suarez JV. Improved mPdium for
fluconazole susceptibility testing of Candida &cans. Antimicrob Agents
Chemother 1994; 3 8 45-8.
23. Barchiesi F, Colombo AL, McGough DA, Rinaldi MG. Comparative
study of broth macrodilution and microdilution techniques for in vitro
antifungal susceptibility testing of yeasts by using the National Committee
for Clinical Laboratory Standards proposed standard.J Clin Microbiol1994;
32: 2494-500.
24. Rodrigue-Tudela JL, Berenguer J, Martinez-Suarez JV, Sanchez R.
Comparison of a spectrophotomemc microdilution method with RPMI-
2% glucose with the National Committee for Clinical Laboratory Standards
reference macrodilution method M27-P for in vitro susceptibility testing of
amphotericin B, flucytosine, and fluconazole against Candida albicans.
Antimicrob Agents Chemother 1996; 40: 1998-2003.
25. Marr KA, Rustad T R , Rex JH, White TC. The trailing end point
phenotype in antifungal susceptibility testing is pH dependent. Anfiminob
Agents Chemother 1999; 43: 1383-6.
26. Ghannoum MA, Ibrahim AS, Fu Y, Shafiq MC, Edwards JE Jr. Cridle RS.
Susceptibility testing of Cryptococcus neofomans: a microdilution technique.J
Clin Minobiol 1992; 30: 2881-6.
27. St-Germain G , Dion C. Effect of media on growth rate and susceptibility
testing of C. neofmans. Mycoses 1996; 39: 2 0 1 4 .
28. Rodriguez-Tudela JL, Martin-Diez F, Cuenca-Estrella M, Rodero L,
Carpintero Y, Gorgojo B. Influence of shaking on antifungal susceptibility
testing of Cryptococms neofomans: a comparison of the NCCLS standard
M27A medium, buffered yeast nitrogen base, and RPMI-2% glucose.
Antimicrob Agents Chemother 20M); 44: 400-4.
29. Jessup CJ, Pfaller MA, Messer SA el al. FlUCOnazok susceptibility testing of
C. neof0nnans: comparison of two broth microdilution methods and clinical
correlates among isolates from Ugandan AIDS patients. J Clin Microbiol
1998; 3 6 2874-6.
30. Odds FC, De Backer T, Dams G, Vranckx L, Woestenborghs F. Oxygen is a
limiting nutrient for growth of C. neofotmuw.J Clin MimbiollW5; 33: 995-7.
31. Rex JH, Cooper C R Jr, Merz WG, Galgiani JN, Anaissie E. Detection of
amphotericin B-resistant Candida isolates in a broth-based system.
Antimicrob Agenfs Chemofher 1995; 3 9 906-9.
32. Lozano-Chiu M, Paetznick VL, Ghannoum MA, Rex JH. Detection to
resistance to amphotericin B among C. neoformans clinical isolates:
performances of three different media assessed by using E-test and National
Committee for Clinical Laboratory Standards M27-A methodo1ogies.J Clin
Minobiol 1998; 36: 2817-22.
33. Espinel-lngroff A, Dawson K, Pfaller MA et a/. Comparative and
collaborative evaluation of standardization of antifungal susceptibility
testing for filamentous fungi. Anfiminob Agmfs Chemother 1995; 3 9 314-9.
34. Espinel-Ingroff A, Bartlett M, Bowden R et al. Multicenter evaluation of
 Cuenca-Estrella and Rodrigez-Tudela Antifungal susceptibility testing 53
proposed standardized procedure for antifungal susceptibility testing of
filamentous fungi. J Cfin Microbiol 1997; 35: 139-43.
35. Odds F, Van Gerven F, Espinel-Ingroff A e f at. Evaluation of possible
correlations between antifungal susceptibilities of filamentous fungi in vitro
and antifungal treatment outcomes in animal infection models. Antimicrob
Agents Cliemotker 1998; 42: 282-8.
36. Gehrt A, Peter J, Pizzo PA, Walsh TJ. Effect of increasing inoculum sizes of
pathogenic filamentous f u n g on MICs of antifungal agents by broth
microdilution method. J Clin Mmobiof 1995; 33: 1302-7.
37. Llop C, Pujol I, Aguilar C, Sala J, Riba D, Guarro J. Companson of three
methods of determining MICs for filamentous fungi using different end
point criteria and incubation periods. Antimicrob Agents Chemother 2000; 44:
23942.
38. Cuarro J, Llop C, Aguilar C, Pujol I. Comparison of in vitro antifungal
susceptibilites of conidia and hyphae of filamentous fungi. Antimicrob ;Igents
Chemother 1997; 41: 2760-2.
39. Cuenca-EstreUa M, Rodriguez-Tudela JL, Diaz-Guerra TM, Mellado E.
Iso-censitest broth: a semi-dcfined medium with enhanced ability to detect
amphotencin b (amb) resistant Candida isolates [abstract]. In: Program and
abstracts of tke 40th Interscience Conference on Antimicrobial Agents and
Chemotherapy, Toronto, Ontario. Wacbington, DC: American Society for
Microbiology, 2000, in press.
40. Davey KG, Johnson EM, Holmes AD, Szekely A, Warnock DW. In-vitro
cusceptibility of Cryptococcus neofonnans isolates to fluconazole and
itraconazole. J Antimicrob Chemother 1998; 42: 217-20.
41. Rodriguez-Tudela JL, Martinez-Suarez JV. Defining conditions for
microbroth antifungal susceptibility testing: influence of RPMI and
RPMI-2,b glucose on the selection of endpoint criteria. J Antimicrob
Clwmotker 1995; 35: 739-49.
42. Cuenca-Estrella M, Diaz-Guerra TM, Mellado E, Monzon A, Rodriguez-
Tudela JL. Comparative in vitro activity of voriconazole and itraconazole
against fluconazoie-Fusceptibleand fluconazole-resistant clinical isolates of
Candida species from Spain. EurJ CIin Microbiof Infect Dis 1999; 18: 432-5.
43. Canton E, Peman J. Carrillo-MuBoz A et af. Fluconazole susceptibilities of
bloodstream Candida sp. isolates as determined by National Committee for
Clinical Laboratory Standards method M27-A and two other methods. J
Cfin Microbiof 1999; 37: 2197-200.
44. Denning DW, Hanson LH, Perlman AM, Stevens DA. In vitro
susceptibility and synergy studies of Aspergillus species to conventional
and new agents. Dingn Microbiol fnqicf Dis 1992; 15: 21-34.
45. Moore CB, Walls CM, Denning DW. In vitro activity of the new ttiazole
BMS-207147 against Aspergillus species in comparison with itraconazole
and amphotericin B. Antimicrob Agents Chemother 2000; 44: 441-3.
46 Cuenca-Estrella M, Rodriguez-Tudela JL, Mellado E, Martinez-Suirez JV,
Monz6n A. Comparison of the in-vitro activity of voriconazole (UK-
109,496), itraconazole and amphotericin B against clinical isolates of
Aspergillus fnmigafus.J Antimirrob Chemother 1998; 42: 531-3.
47. Cuenca-EstreUa M, Ruiz-Diez B, Martinez-Suirez JV, Monzdn A,
Rodriguez-Tudela JL. Comparative in-vitro activity of voriconazole (UK
109-496) and other six antifungal agents against clinical isolates of
Scedospon'um prolijcam and Scedospon'um apiosprnnum.J Antimimob Chemother
1999; 43: 149-51.
48. To WK, Fotherhill AW, Rinaldi MG. Comparative evaluation of macro-
dilution and Alamar colonmetric microdilution broth methods for antifungal
susceptibility testing of yeasts isolates.J Cfin Mimobiof 1995; 33: 2660-4.
49. Espinel-Ingroff A, Pfaller MA, Messer SA el al. Multicenter comparison of
the Sensititre YeastOne colorimetric antifungal panel with the National
Committee for Clinical Laboratory Standards M27-A reference method
0 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 7 (Suppl. 2). 48-53
for testing clinical isolates of common and emerging Candida spp.,
Crypfococnrs spp. & other yeasts and yeast-like organisms. J Clin Microbiof
1999; 37: 591-5.
50. Barry AL, Brown SD. Fluconazole disk diffusion procedure for determining
susceptibility of Candida species. J Cfi Microbiof 1996; 3 4 2154-7.
51. Meis J, Petron M, Bille J, Ellis D, Gibbs D. A global evaluation of the
susceptibility of Candida species to tluconazole by disk diffusion. Global
Antifungal Surveillance Group. Diagn Micmbiot kfcf Dis 2000; 36: 215-23.
52. Venugopal PV, Venugopal TV. Disk diffusion susceptibility testing
of dermatophytes with imidazoles. Indian J Pathof Microbiol 1995; 38:
369-74.
53. Espinel-Ingroff A, PMer MA, Erwin ME, Jones R N . An interlaboratory
evaluation of the E test method for antifungal susceptibility testing of
pathogenic yeasts for five antifungal agents using Casitone agar and
solidified RPMI 1640 with 2% glucose. J Cfin Microbiof 1996; 3 8 848-52.
54. Szekely A, Johnson EM, Wamock DW. Comparison of E-test and broth
microdilution methods for antifungal drug susceptibility testing of molds. J
Cfin Microbiol 1999; 37: 1480-3.
55. Riesselman MH, Hazen KC, Cutler JE. Determination of antifungal MICs
by a rapid susceptibility assay. J Clin Microbiof 2000; 38: 333-40.
56. Shimokawa 0, Nakayama H . Estimation of minimum sterol 14alpha-
demethylation-inhibitory concentration of azoles in Candido yeasts using
acetate-mediated growth inhibition: potential utility in susceptibility
testing. J Clin Minobio/2000; 3 8 2893-6.
57. Atthington-Skaggs BA, Wamock DW, Momson CJ. Quantitation of
Candida afbicans ergosterol content improves the correlation between In
vitro antifungal su5ceptibility test results and in vivo outcome after
fluconazole treatment in a murine model of invasive candidiasis. Antimicrob
Agents Chemother 2000; 44: 2081-55.
58. Druetta A, Freydiere A, Guinet R, Gille Y . Evaluation of five
commercial antifungal susceptibility testing systems. Etrr J Mirrobiol Znfrf
Dis 1993; 12: 3 3 6 4 2 .
59. Arthington-Skaggs BA, Motley M, Wamock DW, Momson CJ.
Comparative evaluation of PASCO and National Committee for Clinical
Laboratory Standards M27-A broth microdilution methods for antifungal
susceptibility testing of yeasts. J Cfin Microbiof 2000; 38: 2254-60.
60. Clancy CJ. Nguyen MH. Correlation between in vitro susceptibility
determined by Etest and response to therapy with amphotericin B. Results
&om a multicenter prospective study of candidemia. Antimictub Agents
Chernother 1999; 43: 1289-90.
61. Johnson EM, Oakley KL, Radford SA el af. Lack of correlation of in vitro
amphotericin B susceptibility testing with outcome in a murine model of
Aspergillus infection. J Antimictub Chemother 2000; 45: 85-93.
62. Lass-Florl C, Kofler G, Kropshofer G ei af. In-vitro testing of susceptibility
to amphotericin B is a reliable predictor of clinical outcome in invasive
aspergillosis. J Antimimob Chemother 1998; 42: 497-502.
63. Denning DW, Venkateswarlu K, Oakley KL et a/. Itraconazole resistance in
Aspe?giflusfumigatur. Anfiminob A p t Chemother 1997; 41: 1364-8.
64. Denning DW, Radford SA, Oakley KL, Hall L, Johnson EM, Wamock
DW. Correlation between in-vitro susceptibility testing to itraconazole and
in-vivo outcome of Aspergillus fumigafus infection. J Antimimob Chemother
1997; 4 0 401-14.
65. Anaissie EJ, Karyotakis NC, Hachem R , Dignani MC, Rez JH, Paetznick
V. Correlation between in vitro and in vivo activity of antifungal agents
against Candida species.J Jnfct Dis 1994; 170: 384-9.
66. White T C , Marr KA, Bowden RA. Clinical cellular, and molecular
factors that contribute to antifungal drug resistance. C h Microbiol Rev
1998; 11: 382-402.