See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/308121346

Antifungal susceptibility testing of Malassezia yeast: Comparison of two

different methodologies

Article  in  Mycoses · September 2016

DOI: 10.1111/myc.12556

CITATIONS READS

19 780

9 authors, including:

Florencia Rojas Susana Córdoba

Instituto de Medicina Regional - Universidad Nacional del Nordeste 1-Administración Nacional de Laboratorios e Institutos de Salud. Instituto Nacion…
39 PUBLICATIONS   462 CITATIONS    78 PUBLICATIONS   1,229 CITATIONS   

SEE PROFILE SEE PROFILE

María de los Ángeles Sosa Maria E Cattana

National University of the Northeast "Dr Julio C. Perrando" Hospital
32 PUBLICATIONS   407 CITATIONS    19 PUBLICATIONS   193 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

LEVADURAS AISLADAS de PACIENTES PEDIATRICOS HOSPITALIZADOS. FRECUENCIA Y SENSIBILIDAD ANTIFÚNGICA View project

Cryptococcus and Cryptocococis. View project

All content following this page was uploaded by Gustavo Giusiano on 19 September 2016.

The user has requested enhancement of the downloaded file.

 | |
Received: 31 March 2016    Revised: 11 August 2016    Accepted: 12 August 2016

DOI: 10.1111/myc.12556

ORIGINAL ARTICLE

Antifungal susceptibility testing of Malassezia yeast:

comparison of two different methodologies

Florencia D. Rojas1 | Susana B. Córdoba2 | María de los Ángeles Sosa1 | Laura C. Zalazar3 | 

Mariana S. Fernández1 | María E. Cattana1 | Liliana R. Alegre1 | 
Alfonso J. Carrillo-Muñoz4 | Gustavo E. Giusiano1

1
Departamento de Micología, Instituto de
Medicina Regional, Universidad Nacional del Summary
Nordeste, Resistencia, Chaco, Argentina All Malassezia species are lipophilic; thus, modifications are required in susceptibility
2
Departamento de Micología, Instituto
testing methods to ensure their growth. Antifungal susceptibility of Malassezia spe-
Nacional de Enfermedades Infecciosas,
ANLIS “Dr. Carlos G. Malbrán”, Buenos Aires, cies using agar and broth dilution methods has been studied. Currently, few tests
Chaco, Argentina
using disc diffusion methods are being performed. The aim was to evaluate the in
3
Cátedra de Matemática Aplicada, Facultad
vitro susceptibility of Malassezia yeast against antifungal agents using broth microdi-
de Humanidades, Universidad Nacional del
Nordeste, Resistencia, Chaco, Argentina lution and disc diffusion methods, then to compare both methodologies. Fifty
4
Departamento de Microbiología, ACIAM, Malassezia isolates were studied. Microdilution method was performed as described
Barcelona, Spain
in reference document and agar diffusion test was performed using antifungal tablets
Correspondence and discs. To support growth, culture media were supplemented. To correlate meth-
Florencia D. Rojas, Departamento de
Micología, Instituto de Medicina Regional, ods, linear regression analysis and categorical agreement was determined. The
Universidad Nacional del Nordeste, strongest linear association was observed for fluconazole and miconazole. The high-
Resistencia, Chaco, Argentina.
Email: florenciarojas@hotmail.com est agreement between both methods was observed for itraconazole and voricona-
zole and the lowest for amphotericin B and fluconazole. Although modifications made
to disc diffusion method allowed to obtain susceptibility data for Malassezia yeast,
variables cannot be associated through a linear correlation model, indicating that
inhibition zone values cannot predict MIC value. According to the results, disc diffu-
sion assay may not represent an alternative to determine antifungal susceptibility of
Malassezia yeast.

KEYWORDS
disc diffusion, lipid-dependent yeast, minimal inhibitory concentration

1 | INTRODUCTION The Clinical and Laboratory Standards Institute reference document,

Yeast of Malassezia genus are normal inhabitants of the human skin
microbiota and other warm-­blooded vertebrates, especially found in
1–3
seborrhoeic areas. These lipophilic yeast are associated with der-
matological diseases such as pityriasis versicolor, seborrhoeic dermati-
tis, dandruff, atopic eczema/dermatitis and folliculitis.3,4 Additionally,
some Malassezia species are recognised as emerging pathogens that,
in some cases, may cause invasive fungal infections in susceptible
patient groups.5–8.
CLSI M27-­A3, describes a standard methodology to determine in vitro
susceptibility of Candida species and Cryptococcus neoformans.9 However,
this method cannot be used to determine the susceptibility of lipophilic
yeast like Malassezia spp. All Malassezia species are lipophilic and most of
them, lipid-­dependent; thus, modifications are required in susceptibility
testing methods to ensure their growth such as the use of supplemented
culture medium, different inoculum size, temperature and reading time.
Antifungal susceptibility of Malassezia species using agar and
broth dilution alternative methods has been studied.5,10–21 Currently,

Mycoses 2016; 1–8 wileyonlinelibrary.com/journal/myc © 2016 Blackwell Verlag GmbH  |  1

 |
2       Rojas et al.
few antifungal susceptibility tests are performed using disc diffusion
methods, but only when studying the non-­lipid-­dependent species,
M. pachydermatis.22,23
Determination of minimal inhibitory concentration (MIC) by broth
microdilution is not an applicable method in a routine clinical labora-
tory. Therefore, other methods such as the agar diffusion test have
been developed as an alternative to simplify antifungal susceptibility
testing.
The aims of this study were to evaluate the in vitro antifun-
gal susceptibility of three Malassezia species against fluconazole,
voriconazole, ketoconazole, itraconazole, miconazole, amphotericin
B and terbinafine using both broth microdilution and agar diffusion
methods, and to compare both methodologies using linear regression
analysis and categorical agreement.
2 |  MATERIALS AND METHODS
2.1 | Malassezia strains
A total of 50 Malassezia isolates were studied including 30 Malassezia
furfur (M. furfur), 10 M. globosa and 10 M. sympodialis. All strains were
isolated from clinical samples obtained from human patients with
diagnosis of pityriasis versicolor, seborrhoeic dermatitis, folliculitis and
atopic dermatitis. The yeast were deposited in the culture collection
of Mycology Department, Instituto de Medicina Regional, Universidad
Nacional del Nordeste (UNNE) (Argentina). Identification was per-
formed by polymerase chain reaction -­ restriction fragment length
polymorphism (PCR-­RFLP).24 Isolates were sub-­cultured for 72 h onto
modified Dixon Agar at 32°C before antifungal susceptibility testing.
Quality control strains, Candida parapsilosis ATCC 22019 and
C. krusei ATCC 6258 were included in each assay.
Written informed consent from the patients was obtained accord-
ing to the current Ethics Committee of Instituto de Medicina Regional.
2.2 | Susceptibility testing
Broth microdilution method was performed as described in the CLSI
document M27-­A3.9 Stock solutions of fluconazole (FLC), voricona-
zole (VCZ), ketoconazole (KTZ), miconazole (MCZ), itraconazole (ITZ),
amphotericin B (AMB) and terbinafine (TRB) (Sigma-­Aldrich, Buenos
Aires, Argentina) were prepared at 100 times the highest concentration
to be tested in dimethyl sulfoxide (Anedra, San Fernando, Argentina)
(10 times for FLC) and stored at −70°C until use. The final concen-
−1
trations for all drugs were 0.03–16 μg mL except for FLC (0.125–
64 μg mL−1). To support growth of lipid-­dependent yeast, RPMI 1640
medium (Gibco, Buenos Aires, Argentina) was supplemented with glu-
cose (1.8%) (Cicarelli Reagents, San Lorenzo, Argentina), peptone (1%),
ox bile (0.5%), malt extract (0.5%) (Laboratorios Britania, Autónoma de
Buenos Aires, Argentina), Tween 40 (0.5%) (Sigma-­Aldrich), Tween 80
(0.05%) (Anedra), glycerol (1%) (Biopack, Autónoma de Buenos Aires,
Argentina) and chloramphenicol (250 mg L−1) (Sigma-­Aldrich).
All inoculum suspensions were prepared in sterile saline from
colonies grown on a 5-­day culture on modified Dixon Agar at 32°C.
After vortexing to disperse Malassezia clumps, the inoculum was
standardised spectrophotometrically at 530 nm to 1 McFarland stan-
dard and validated by quantitative plate count on Dixon Agar (106
UFC mL−1). Inocula were diluted 1:100 in supplemented RPMI 1640
medium providing a final inoculum size of 0.5–2.5 × 103 CFU mL−1.20
The plates were incubated at 32°C and visual reading was performed
every 24 h for about 3 days.
For azole drugs, MIC endpoints were defined as the lowest con-
centration that produces a prominent reduction in growth (≥50%) rel-
ative to the drug-­free growth control, slight reduction for TRB (80%)
and complete inhibition for AMB (100%). Data obtained were reported
in absolute frequencies displaying the number of isolates according to
MIC values, for each species tested.
At present, no breakpoints are established for antifungal agents
against Malassezia species. The interpretative criterion was assessed
according to breakpoints established in CLSI reference documents
for Candida albicans. For FLC, isolates with MIC ≤2 μg mL−1 were
categorised as susceptible, 4 μg mL−1 as sensible-­dose dependent
(S-­DD) and ≥8 μg mL−1 as resistant. For ITZ and VCZ, isolates with MIC
≤0.125 1 μg mL−1 were categorised as susceptible, 0.25–0.5 μg mL−1
as sensible-­dose dependent and ≥1 μg mL−1 as resistant. Isolates with
AMB MIC >1 μg mL−1 were categorised as resistant.
For disc diffusion method, Mueller-­Hinton agar with glucose (2%)
and methylene blue (5 mg L−1) was used. This medium was supple-
mented with peptone (1%), ox bile (0.5%), malt extract (0.5%), Tween
40 (0.5%), Tween 80 (0.05%), glycerol (1%) (Biopack) and chloram-
phenicol (250 mg L−1).
Plates were inoculated with 0.5 mL of the inoculum suspensions
adjusted to 1 McFarland poured onto the agar surface (flooding) and
the excess liquid was immediately removed.
AMB (10 μg), FLC (25 μg), KTZ (15 μg), ITZ (10 μg), MCZ (10 μg),
TRB (30 μg) and VCZ (1 μg) Neo-­Sensitabs tablets (Medica-­Tec,
Autónoma de Buenos Aires, Argentina) and paper discs of FLC (25 μg)
provided by the Department of Mycology (ANLIS ‘Dr. Carlos Malbrán’,
Argentina) were used. Plates were incubated at 32°C for 72 h, with
daily reading. The inhibition zone was considered up to the limit of col-
onies of confluent growth. For AMB, the complete clear zone with no
visible growth was measured. For both tablets and discs, the interpre-
tative criterion was according to the manufacturer’s guidelines.25,26
2.3 | Statistical analysis
The degree of linear relationship between the MICs and the inhibi-
tion zone diameters was analysed using Pearson’s correlation coeffi-
cient (ρ). To quantify the relationship between these two variables,
linear regression analysis was performed. As an adjustment measure,
coefficient of determination R2 was calculated. infostat software
2014 (Cátedra de Estadística y Biometría de la Facultad de Ciencias
Agropecuarias de la Universidad Nacional de Córdoba, Córdoba,
Argentina) was used for the statistical analysis.
The categorical agreements and percentages of discrepancies were
evaluated. Discrepancies between methods were considered very major
errors (VME) when the microdilution method categorised the organism
Rojas et al. |
      3

as resistant but the diffusion method categorised the organism as sus-
ceptible. Major errors (ME) occurred when the microdilution method
categorised the isolate as susceptible but the diffusion method catego-
rised it as resistant. Minor errors (mE) occurred when the microdilution
method categorised an organism as susceptible or resistant and the
diffusion method categorised it as susceptible dose dependent or the
microdilution method categorised it as susceptible dose dependent and
the diffusion method categorised it as susceptible or resistant.
3 | RESULTS
Both, supplemented RPMI 1640 medium and Mueller-­Hinton agar,
allowed growth of Malassezia species studied. Reading times for
microdilution and disc diffusion test were not species dependent and
results could be recovered at 72 h for both methods.
For disc diffusion, inhibition zone could be read at 48 h for some of
the isolates, but the diameters measured at 72 h were longer. This is due
to the trailing phenomenon, that it is quite evident 3 days after incubation.
MIC and zone diameter values obtained with QC strains were
within the range established in reference documents.25,27
Table 1 shows in absolute frequency the number of isolated strains
according to MIC value obtained.
FLC, AMB, MCZ and TRB displayed higher MIC values and were
more variable than KTZ, ITZ and VCZ, for the three species tested.
FLC showed the highest MIC values. For ITZ, 90% of isolates were
inhibited by the lowest antifungal drug dilution (MIC90: 0.03 μg mL−1).
Results of agar diffusion test for the 50 isolates studied are sum-
marised in Table 2. Both FLC disc and tablet showed the widest range
of inhibition zone diameters, with a broad dispersion of the values
around the mean value, especially for M. furfur. MCZ also showed a
wide range of results, but the dispersion was lower than FLC.
In general, there was not a significant difference between results
for the three species evaluated, except for M. sympodialis against FLC,
which resulted in higher inhibition zones and less dispersion of the
values compared with the other species.
Figure 1 shows scatter plots of inhibition zone diameters against
respective log2 MIC for the antifungal drugs tested. The relationship
between variables was not linear for KTZ and ITZ, as shown in the
figures.
Results of the Pearson’s correlation coefficient are presented
in Table 3. In all cases, negative relationship between variables was

T A B L E   1   Absolute frequencies of strains according to the MIC values obtained

MIC μg mL−1 (number of strains)

FLC KTZ AMB MCZ ITZ VCZ TBN

M. furfur (n = 30) 0.5 (2) 0.03 (27) 0.25 (1) 0.125 (1) 0.03 (27) 0.03 (8) 0.125 (5)
1 (3) 0.06 (2) 0.5 (1) 0.25 (3) 0.06 (2) 0.06 (10) 0.25 (5)
2 (2) 0.125 (1) 1 (5) 0.5 (2) 0.125 (1) 0.125 (7) 0.5 (2)
4 (7) 2 (11) 1 (3) 0.25 (2) 1 (5)
8 (5) 4 (7) 2 (6) 0.5 (3) 2 (5)
16 (3) 8 (4) 4 (6) 4 (3)
32 (5) 16 (1) 8 (3) 8 (1)
64 (2) 16 (4) 16 (2)
128 (1) 32 (2) 32 (2)
M. sympodialis (n = 10) 0.5 (3) 0.03 (9) 0.5 (2) 0.5 (2) 0.03 (8) 0.03 (7) 0.125 (4)
1 (3) 0.06 (1) 1 (4) 1 (3) 0.06 (2) 0.06 (3) 0.25 (3)
2 (2) 2 (4) 2 (2) 0.5 (3)
4 (2) 4 (2)
8 (1)
M. globosa (n = 10) 1 (1) 0.03 (8) 0.5 (2) 0.25 (1) 0.03 (10) 0.03 (3) 0.25 (5)
4 (3) 0.06 (2) 1 (2) 0.5 (1) 0.06 (3) 0.5 (1)
8 (4) 2 (1) 1 (1) 0.125 (3) 1 (2)
16 (1) 4 (3) 2 (1) 0.5 (1) 16 (1)
64 (1) 8 (1) 4 (3) 32 (2)
8 (1)
16 (1)
32 (1)

M., Malassezia; MIC, minimal inhibitory concentration; FLC, fluconazole; VCZ, voriconazole; KTZ, ketoconazole; MCZ, miconazole; ITZ, itraconazole; AMB,
amphotericin B; TRB, terbinafine.
 |
4       Rojas et al.

T A B L E   2   Inhibition zone diameters (mm) for the 50 Malassezia yeast

Disc diffusion inhibition zones (mm)

FLC FLC Malbrán KTZ AMB MCZ VCZ ITZ TRB

M. furfur (n = 30)
Range 9–52 7–48 25–40 15–21 9–32 31–55 22–35 14–28
AM 30.6 28.1 32.8 18.2 16.1 43.2 29.2 20.5
SD 10.6 12.2 4.02 1.36 4.41 5.2 3.44 3.21
M. globosa (n = 10)
Range 12–46 10–40 24–38 16–26 11–20 36–42 21–30 14–24
AM 28.1 25.3 30.7 20.2 14.8 38.8 25.4 19.7
SD 8.34 7.93 2.96 2.96 2.86 3.21 2.62 3.26
M. sympodialis (n = 10)
Range 19–35 17–43 26–38 16–25 9–21 30–48 23–32 16–29
AM 37.4 32.6 31.4 19.8 15.0 40.8 28.8 23.0
SD 8.32 7.52 3.92 2.48 4.8 5.44 2.64 2.8

Mm, millimetres; AM, arithmetic mean; SD, standard deviation.

observed. The strongest linear association was observed for FLC, both
disc (ρ = −0.71) and tablet (ρ = −0.70).
2
Table 3 shows the coefficient of determination, R , obtained by
linear regression analysis comparing disc diffusion and broth microdi-
lution results. The analysis was not performed for KTZ, ITZ and TRB
because variables were not linearly associated.
The categorical interpretation based on the in vitro results of the
50 Malassezia yeast against antifungal drugs and agreements between
the two methods are presented in Table 4. Clinical breakpoints have
not been established for MCZ and TRB in the reference documents;
so, the agreement could not be performed. Considering disc diffusion
method, the arithmetic mean was a value within the susceptible cate-
gory for all drugs, except for MCZ.
The highest numbers of isolates categorised as resistant were
recovered against FLC and AMB, regardless of the methodology used.
The highest agreement between both methods was observed for
ITZ and VCZ, and the lowest for AMB and FLC.
FCZ showed a significant difference in the categorisation of iso-
lates between both methods. For microdilution method, 68% of the
isolates showed a decreased susceptibility against this drug (SDD:
24% and R: 44%). However, with the diffusion method, both tablet
and disc, strains were mostly categorised as susceptible (78% and 72%
respectively).
Against VCZ and AMB, all isolates were categorised as susceptible
with disc diffusion method. However, with the microdilution method,
more than 60% of the strains were categorised as resistant according
to the CLSI document for Candida sp.
4 |  DISCUSSION
Data obtained with broth microdilution methodology indicated
that ITZ and KTZ were highly effective drugs against the Malassezia
species tested, obtaining a restricted range of MIC values, similar
to that observed in previous investigations.12,13,17–19,21 These azole
agents are widely used for treatment of skin diseases associated with
Malassezia yeast; ITZ is considered a good option for the management
of Malassezia superficial infections which require oral therapy and
KTZ is widely used for topical treatment.28–30 Although VCZ showed
itself as an active drug for most of the isolates tested, MIC values
were higher than those for ITZ and KTZ. These data are in concord-
ance with other studies.10,12,17,20
High MIC values and a significant dispersion were observed for
FLC, similar to those obtained in previous investigations.10,14,16,17,19–21
Using interpretative criteria established for Candida species and FLC,27
we found that more than 60% of isolates were categorised as SDD
or resistant to this drug. These results are important especially con-
sidering that FLC is the most widespread drug used for yeast fungal
­infection, including those caused by Malassezia species.
TRB shows low or no in vitro activity against common yeast
genera.31,32 High MIC values reported in this study suggest that
this drug has also low in vitro activity against Malassezia genus. The
results obtained in this study are even higher than those in previous
investigations.12,16,33
Wide MIC range and high geometric mean were obtained for MCZ,
similar to those reported in previous investigations.13,19,20 Due to the
high MIC values against the three species, it might be thought that
in some cases treatment with this azole drug could not be effective.
However, to consider that is debatable, because concentrations of this
antifungal used in formulated products are significantly higher than
those used in vitro testing.34
The M27-­A3 document determined that Candida species with MIC
higher than 1 μg mL−1 are likely to be resistant to AMB.9 Using this
breakpoint for Malassezia yeast, more than 50% of the isolates were
categorised as resistant. Velegraki et al. [16] reported high MIC for
AMB to M. furfur and M. globosa employing broth microdilution and
Rojas et al.       5 |
(a) 55 (b) 50
50 45
45

InhibitionzoneFLC Malbrán (mm)

40
40

Inhibition zone FLC (mm)

35
35
30
30
25
25
20
20
15
15
10
10
5 5
–2 –1 0 1 2 3 4 5 6 7 0
log 2 MIC FLC (µg mL–1) –2 –1 0 1 2 3 4 5 6 7
log 2 MIC FLC (µg mL–1)

(c) 45 (d) 45

40 40

Inhibition zone ITZ (mm)

35 35
Inhibition zone KTZ (mm)

30 30

25 25

20 20

15 15
–6 –5 –4 –3 –2 –1 0 –6 –5 –4 –3 –2 –1 0
–1
log 2 MIC KTZ (µg mL ) log 2 MIC ITZ (µg mL–1)

(e) 60 (f) 35

55 30

50
Inhibition zone VCZ (mm)

Inhibition zone MCZ (mm)

25

45
20
40
15
35
10
30
5
25
0
20 –4 –3 –2 –1 0 1 2 3 4 5 6
–6 –5 –4 –3 –2 –1 0
log 2 MIC MCZ (µg mL–1)
–1
log 2 MIC VCZ (µg mL )

(g) 30 (h) 35

30
25
Inhibition zone TRB (mm)
Inhibition zone AMB (mm)

25

20 20
F I G U R E   1   Scatter plots of MIC
(μg ml−1) vs. inhibition zone diameter 15

(mm) determined using FCZ tablet (a), 15

10
Malbrán paper disk (b), ketoconazole
(c), itraconazole (b), voriconazole (d), 5
10
–4 –3 –2 –1 0 1 2 3 4 5
amphotericin B (e), miconazole (f) and –3 –2 –1 0 1 2 3 4 5
log 2 MIC TRB (µg mL–1)
log 2 MIC AMB (µg mL–1)
terbinafine (g) tablets.

E-­test methodologies. Recently, Iatta et al. studied a large number of the Neo-­Sensitabs® tablets assay for testing FLC, VCZ, ITZ, MCZ, TRB
M. furfur from bloodstream infections and results showed high AMB and AMB against lipid-­dependent Malassezia species, compared with
MIC values regardless of the different media used.16,21 These high a broth microdilution method using supplemented medium. A similar
MIC against AMB are significant, especially because this drug is one study was realised by Yurayart et al., [41] but only for M. pachydermatis
of the main treatment options for systemic infection, particularly in and with no modification of the methodologies used.
paediatric patients.8,35,36 Although there is no available standardised methodology for
Most of the correlation studies of Neo-­Sensitabs® tablets and Malassezia sp., the comparison between both methods allowed deter-
CLSI broth microdilution were conducted for Candida species and fila- mining the extent to which inhibition zone variation represents the MIC
mentous fungi.26,37–40 We assessed for the first time, the suitability of variation. Scatter plot and determination of Pearson coefficients were
 |
6       Rojas et al.

T A B L E   3   Pearson’s linear correlation coefficients and coefficients Pearson coefficients for these drugs indicated a probable linear rela-
of determination (R2) obtained by correlating inhibition zone tionship, when linear regression analysis was applied, the determina-
diameter and corresponding MIC value tion coefficients (R2) obtained were very low, indicating that the linear
Pearson correlation model does not explain the relationship between the two variables.
Antifungal drug coefficient (ρ) p-value R² The R2 obtained for FLC was even lower than that obtained in the
Fluconazole −0.71 <.0001 0.680 study realised by Maldonado et al. for Candida species.39
Malbrán For all the other drugs, coefficients were not enough to estab-
Fluconazole −0.70 <.0001 0.565 lish that a linear relationship exists between MIC and inhibition zone
Ketoconazolea −0.2 .1627 – diameter variables.
Itraconazole a
−0.02 .8785 – The agreement between the disc diffusion and microdilution

Miconazole −0.52 .0001 0.231

methods has been extensively studied in Candida species and Cr. neo-
formans. Previous investigations conducted in Candida species against
Voriconazole −0.47 .0016 0.290
FLC, reported higher agreement (85.9–95.5%) and lower discrepan-
Amphotericin B −0.45 .0012 0.469
cies (0–4.8% for VME and 2.4–8.84% for mE) than those obtained in
Terbinafinea −0.34 .0220 –
this study.37–40 Regarding FLC Malbrán discs, Rodero et al. reported
a 2
R could not be calculated because variables were not linearly a concordance of 94.7% with 4.8% of mE, 0.3% ME and 0.2% VME
associated.
for Candida species, displaying a better correlation between the two
methods than the one observed in this study for Malassezia strains.26
useful tools, which allowed establishing whether these variables were Disc diffusion method was not able to detect those Malassezia
linearly related and, given the case, the magnitude of that relationship. strains categorised as resistant by the microdilution method for
Scatter plots and Pearson coefficients showed that the variables AMB, resulting in the highest VME compared to other drugs (64%).
were not linearly associated for KTZ and ITZ. The few publications Contrary to these results, Espinel-­Ingroff et al. obtained high agree-
regarding the correlation results for these drugs make the compari- ment between AMB Neo-­Sensitabs tablets and the reference method
son difficult. Espinel Ingroff et al. studied the correlation for Candida (98.2%) for Candida spp.37
species and Cr. neoformans against ITZ, obtaining Pearson coefficient It is proposed for a susceptibility test to be considered specific that
values of 0.85 and 0.79 respectively. These values are much higher occurrence of VME and ME should be <5%. Only for ITZ and VCZ, this
than those obtained for Malassezia yeast in this study.37 value was obtained, although in both cases resulting in a higher mE. This
The Pearson’s correlation coefficient was significant only for MCZ value represents isolates categorised as SDD with the microdilution
and FLC (P < .0001). The values obtained for FLC were similar to those method that were not detected by the diffusion method. It is important
reported in other studies for Candida species.26,37–40 But even though to highlight that as there were no isolates categorised as resistant to

T A B L E   4   Categorical interpretation, discrepancies and total agreement between broth microdilution and disc diffusion tests for 50
Malassezia yeast

Number of isolates Errors (%)

Agreement
Antifungal Method S SDD (I) R Very major Major Minor (%)

Fluconazole BMD 16 12 22
Tablet DD 39 3 8 25.0 0 28.0 47.0
Disc DD 36 2 12 18.0 0 24.0 58.0
Itraconazole BMD 50 0 0
DD 45 5 0 0 0 10.4 89.6
Voriconazole BMD 44 6 0 0 0 12.5 87.5
DD 50 0 0
Amphotericin B BMD 18 32
DD 50 0 0 64.0 0 0 36.0
Ketoconazole (a,b) DD 44 6 0 N/A
(a,b)
Miconazole DD 12 29 9 N/A
Terbinafine (a,b) DD 31 19 0 N/A

S, susceptible; SDD, susceptible dose dependent; R, resistant; BMD, Broth microdilution; DD, disc diffusion; I, intermediate; N/A, not applicable.
a
Interpretive breakpoints for ketoconazole, miconazole and terbinafine have not been established by reference method, so categorical agreements and
discrepancies were not evaluated.
b
For ketoconazole, miconazole and terbinafine there is no SDD category. Intermediate is used in these cases.
Rojas et al.
these drugs by the microdilution testing, so it was impossible to deter-
mine whether they would be detected by the diffusion method.
The disc diffusion method is a low-­cost, fast and simple technique
that is easily applicable in routine clinical laboratories. However, not
a good correlation between this technique and the broth microdilu-
tion method was obtained when evaluating antifungal susceptibility
of Malassezia yeasts. Collaborative studies to establish interpretative
MIC breakpoints for this genus are required in order to perform ­further
assessment of suitability of alternative methods.
5 | CONCLUSIONS
Although modifications made to disc diffusion and microdilution
methods allowed obtaining susceptibility data for Malassezia yeasts,
variables cannot be associated through a linear correlation model.
Additional studies are needed to find out if another kind of correlation
exists between both methodologies.
According to the data obtained in this study, disc diffusion assay
may not represent an alternative to determine antifungal susceptibility
of Malassezia yeast.
ACKNOWLE DG ME NTS
We gratefully acknowledge Prof. Mariana Climent for revising the
English of the manuscript.
CO NFLI CT OF I NTE RE S T
The authors have no conflict of interest to declare.
REFERENCES
1. Giusiano G. Malassezia: estado del conocimiento y perspectivas en su
estudio. Rev Argent Microbiol. 2006;38:41–48.
2. Cabañes FJ. Malassezia yeasts: how many species infect humans and
animals? PLoS Pathog. 2014;10:e1003892.
3. Saunders CW, Scheynius A, Heitman J. Malassezia fungi are special-
ized to live on skin and associated with dandruff, eczema, and other
skin diseases. PLoS Pathog. 2012;8:e1002701.
4. Crespo EV, Delgado FV. Malassezia species in skin diseases. Curr Opin
Infect Dis. 2002;15:133–142.
5. Giusiano G, Mangiaterra M, Saito VG, Rojas F, Gómez V, Díaz MC.
Etiology of fungaemia and catheter colonisation in Argentinean pae-
diatric patients. Mycoses. 2006;49:49–54.
6. Rosales CM, Jackson MA, Zwick D. Malassezia furfur meningitis asso-
ciated with total parenteral nutrition subdural effusion. Pediatr Dev
Pathol. 2004;7:86–90.
7. Dokos C, Pana ZD, Tragiannidis A. Malassezia species: a rare cause
of invasive fungal infections in immunocompromised patients. Curr
Fungal Infect Rep. 2010;5:18–22.
8. Arendrup MC, Boekhout T, Akova M, et al. ESCMID and ECMM joint
clinical guidelines for the diagnosis and management of rare invasive
yeast infections. Clin Microbiol Infect. 2014;20:76–98.
9. CLSI document M27-A3. Reference Method for Broth Dilution Antifungal
Susceptibility Testing of Yeasts; Approved Standard - Third Edition. Wayne
PA, USA: Clinical and Laboratory Standards Institute; 2008.
|
      7
10. Miranda KC, de Araujo CR, Costa CR, Passos XS, de Fátima Lisboa
Fernandes O, do Rosário Rodrigues Silva M. Antifungal activities of
azole agents against the Malassezia species. Int J Antimicrob Agents.
2007;29:281–284.
11. Shokohi T, Afshar P, Barzgar A. Distribution of Malassezia species
in patients with pityriasis versicolor in Northern Iran. Indian J Med
Microbiol. 2009;27:321–324.
12. Gupta A, Kohli Y, Li A, Faergemann J, Summerbell RC. In vitro suscep-
tibility of the seven Malassezia species to ketoconazole, voriconazole,
itraconazole and terbinafine. Br J Dermatol. 2000;142:758–765.
13. Hammer KA, Carson CF, Riley TV. In vitro activities of ketoconazole,
econazole, miconazole, and melaleuca alternifolia (tea tree) oil against
Malassezia species. Antimicrob Agents Chemother. 2000;44:467–469.
14. Van GF, Odds FC. The anti Malassezia furfur activity in vitro and in ex-
perimental dermatitits of sixs imidazole antifungal agents: bifonazole
clotrimazole, flutrimazole, ketoconazole, miconazole and sertacon-
azole. Mycoses. 1995;38:389–393.
15. Nakamura Y, Kano R. Susceptibility testing of Malassezia species using
the urea broth microdilution method. Antimicrob Agents Chemother.
2000;44:2185–2186.
16. Velegraki A, Alexopoulos EC, Kritikou S, Gaitanis G. Use of fatty acid
RPMI 1640 media for testing susceptibilities of eight Malassezia spe-
cies to the new triazole posaconazole and to six established antifun-
gal agents by a modified NCCLS M27-­A2 microdilution method and
Etest. J Clin Microbiol. 2004;42:3589–3593.
17. Garau M, Pereiro M, Palacio A, Del Palacio A. In vitro susceptibili-
ties of Malassezia species to a new triazole, albaconazole (UR-­9825),
and other antifungal compounds. Antimicrob Agents Chemother.
2003;47:2342–2344.
18. Rincón S, Cepero de García MC, Espinel-Ingroff A. A modified
Christensen’s urea and CLSI broth microdilution method for testing
susceptibilities of six Malassezia species to voriconazole, itraconazole,
and ketoconazole. J Clin Microbiol. 2006;44:3429–3431.
19. Carrillo-Muñoz AJ, Rojas F, Tur-Tur C, et al. In vitro antifungal activity
of topical and systemic antifungal drugs against Malassezia species.
Mycoses. 2013;56:571–575.
20. Rojas FD, Sosa MA, Fernández MS, Cattana ME, Córdoba SB,
Giusiano GE. Antifungal susceptibility of Malassezia furfur, Malassezia
sympodialis, and Malassezia globosa to azole drugs and amphoter-
icin B evaluated using a broth microdilution method. Med Mycol.
2014;52:641–646.
21. Iatta R, Figueredo LA, Montagna MT, Otranto D, Cafarchia C. In vitro
antifungal susceptibility of Malassezia furfur from bloodstream infec-
tions. J Med Microbiol. 2014;63:1467–1473.
22. Al-Sweih N, Ahmad S, Joseph L, Khan S, Khan Z. Malassezia pachy-
dermatis fungemia in a preterm neonate resistant to fluconazole and
flucytosine. Med Mycol Case Rep. 2014;5:9–11.
23. Bernardo FM, Martins HM, Martins ML. A survey of mycotic otitis
externa of dogs in Lisbon. Rev Iberoam Micol. 1998;15:163–165.
24. Mirhendi H, Makimura K, Zomorodian K, Yamada T, Sugita T, Yamaguchi
H. A simple PCR-­RFLP method for identification and differentiation of
11 Malassezia species. J Microbiol Methods. 2005;61:281–284.
25. Rosco Diagnostica. Document: 3.15.0. Interpretation zones and mic
breakpoints according to clsi. yeasts (CLSI M44-A2 method of diffusion
on agar). In: Rosco Diagnostica user guide. EUCAST-and CLSI Potency
NEO-SensitabsTM. Taastrup, Denmark; 2013:1–4.
26. Rodero L, Córdoba S, Vivot W, et al. Método de difusión con discos
para la determinación de sensibilidad a fluconazol en aislamientos de
Candida spp. Rev Argent Microbiol. 2006;38:155–163.
27. CLSI Document M27-S4. Reference Method for Broth Dilution Antifungal
Susceptibility Testing of Yeasts; Fourth Informational Supplement. Wayne
PA, USA: Clinical and Laboratory Standards Institute; 2012.
28. Nath J, Majumdar M, Majumdar V, Das J, Chakraborty U, Mazumdar
G. Oral itraconazole for the treatment of severe seborrhoeic dermati-
tis. Indian J Dermatol. 2011;56:515.
 |
8       Rojas et al.
29. Gaitanis G, Velegraki A. Malassezia in skin and systemic diseases. Arch
Hell Med. 2004;21:232–244.
30. Khondker L, Wahab A, Khann Islam S, Shirin S. Safety of oral itracon-
azole in the treatment of seborrheic dermatitis. J Pak Assoc Dermatol.
2011;21:102–105.
31. Jessup CJ, Ryder NS, Ghannoum MA. An evaluation of the in vitro
activity of terbinafine. Med Mycol. 2000;38:155–159.
32. Ryder NS, Wagner S, Leitner I. In vitro activities of terbinafine against
cutaneous isolates of Candida albicans and other pathogenic yeasts.
Antimicrob Agents Chemother. 1998;42:1057–1061.
33. Eichenberg ML, Appelt CE, Berg V, et al. Susceptibility of Malassezia
pachydermatis to azole antifungal agents evaluated by a new broth
microdilution method. Acta Sci Vet. 2003;31:75–80.
34. Peano A, Beccati M, Chiavassa E, Pasquetti M. Evaluation of the
­antifungal susceptibility of Malassezia pachydermatis to clotrimazole,
miconazole and thiabendazole using a modified CLSI M27-­A3 micro-
dilution method. Vet Dermatol. 2012;23:131–e29.
35. Iatta R, Immediato D, Montagna MT, Otranto D, Cafarchia C. In vitro
activity of two amphotericin B formulations against Malassezia fur-
fur strains recovered from patients with bloodstream infections. Med
Mycol. 2015;53:269–274.
36. Iatta R, Cafarchia C, Cuna T, et  al. Bloodstream infections by
Malassezia and Candida species in critical care patients. Med Mycol.
2013;52:264–269.
View publication stats
37. Espinel-Ingroff A, Canton E, Gibbs D, Wang A. 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;45:858–864.
38. Cantón E, Pemán J, Sastre M, Valentín A, Bosch M, Espinel-Ingroff
A. Evaluación y utilidad de los métodos E-­test® y Neo-­Sensitabs®
para estudiar la sensibilidad de las levaduras al fluconazol. Rev Esp
Quimioter. 2006;19:267–274.
39. Maldonado I, Fernández Canigia L, Vivot W, Domecq P, Davel G,
Cordoba S. Evaluación de tres métodos para la detección de la sensi-
bilidad in vitro de especies de Candida a los antifúngicos. Rev Argent
Microbiol. 2011;43:120–126.
40. Espinel-Ingroff A. Comparison of three commercial assays and a
modified disk diffusion assay with two broth microdilution reference
assays for testing zygomycetes, Aspergillus spp., Candida spp., and
Cryptococcus neoformans with posaconazole and amphotericin B. J
Clin Microbiol. 2006;44:3616–3622.
41. Yurayart C, Nuchnoul N, Moolkum P, et al. Antifungal agent suscep-
tibilities and interpretation of Malassezia pachydermatis and Candida
parapsilosis isolated from dogs with and without seborrheic dermatitis
skin. Med Mycol. 2013;51:721–730.