Canda

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/7969636
Laboratory diagnosis of invasive candidiasis
Article in The Journal of Microbiology · March 2005

Source: PubMed
CITATIONS READS
177 2,025
2 authors, including:
Arjuna N B Ellepola
Kuwait University
80 PUBLICATIONS 1,951 CITATIONS
SEE PROFILE
All content following this page was uploaded by Arjuna N B Ellepola on 11 June 2014.
The user has requested enhancement of the downloaded file.

The Journal of Microbiology, February 2005, p.65-84 Vol. 43, special issue (No. S)
Copyright ⓒ 2005, The Microbiological Society of Korea
Laboratory Diagnosis of Invasive Candidiasis

Arjuna N.B. Ellepola1,2 and Christine J. Morrison1,*
1
Mycotic Diseases Branch, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia 30333, USA
2
Department of Oral Medicine and Periodontology, Faculty of Dental Sciences, University of Peradeniya, Peradeniya, Sri Lanka
(Accepted July 7, 2004)
Invasive candidiasis is associated with high morbidity and mortality. Clinical diagnosis is complicated by
a lack of specific clinical signs and symptoms of disease. Laboratory diagnosis is also complex because cir-
culating antibodies to Candida species may occur in normal individuals as the result of commensal col-
onization of mucosal surfaces thereby reducing the usefulness of antibody detection for the diagnosis of
this disease. In addition, Candida species antigens are often rapidly cleared from the circulation so that
antigen detection tests often lack the desired level of sensitivity. Microbiological confirmation is difficult
because blood cultures can be negative in up to 50% of autopsy-proven cases of deep-seated candidiasis
or may only become positive late in the infection. Positive cultures from urine or mucosal surfaces do not
necessarily indicate invasive disease although can occur during systemic infection. Furthermore, differ-
ences in the virulence and in the susceptibility of the various Candida species to antifungal drugs make
identification to the species level important for clinical management. Newer molecular biological tests
have generated interest but are not yet standardized or readily available in most clinical laboratory set-
tings nor have they been validated in large clinical trials. Laboratory surveillance of at-risk patients could
result in earlier initiation of antifungal therapy if sensitive and specific diagnostic tests, which are also cost
effective, become available. This review will compare diagnostic tests currently in use as well as those
under development by describing their assets and limitations for the diagnosis of invasive candidiasis.
Key words: candida, diagnosis, candidiasis, laboratory tests
The availability of modern and sophisticated medical care Candida BSIs, significantly decreased in this population
to prolong and improve the lives of the severely debili- in the 1990s (Trick et al., 2002). During the same period,
tated is becoming increasingly more common in today’s a significant increase in Candida glabrata BSIs occurred
health care system. Ironically, such medical advances (Trick et al., 2002). Changes in Candida spp. distributions
have resulted in a group of patients more vulnerable to in the bloodstream of patients outside of the ICU were
fungal infections. Such patients include those who receive also reported (Price et al., 1994; Nguyen et al., 1996; Abi-
therapy that produces prolonged neutropenia, such as Said et al., 1997; Hajjeh et al., 2004). Identification of the
some stem cell transplant recipients, or those undergoing infecting organism to the spp level has therefore become
solid organ transplantation, particularly liver transplant increasingly important for several reasons. First, not only
recipients. In these patient populations, Candida spp. are has the Candida spp. distribution changed in recent years,
a prominent cause of invasive disease (Rees et al., 1998; but Candida spp. differ in their susceptibility to antifungal
Pappas et al., 2004). In addition, Candida spp. are the agents. For instance, Candida krusei is often innately
fourth most common agent of all hospital-acquired blood- azole resistant, C. glabrata has been reported to acquire
stream infections (BSIs) in the United States; such infec- resistance in vitro and in vivo, and Candida dubliniensis
tions have an attributable mortality of up to 49% isolates have been observed to rapidly develop resistance
(Edmond et al., 1999; Centers for Disease Control and to fluconazole (Wingard, 1995; Moran et al., 1997, 1998;
Prevention, 2000; Gudlaugsson et al., 2003). Although Pfaller et al., 1999; Trick et al., 2002). Candida lusitaniae
sentinel surveillance studies in the 1980s indicated that can be resistant to amphotericin B, and Candida parap-
the overall incidence of Candida BSIs increased among silosis and Candida guilliermondii have higher MICs to
patients in hospital intensive care units (ICUs) (Banerjee the echinocandins than other Candida spp.; C. glabrata
et al., 1991), the incidence of infections caused by Can- and C. krusei require maximum doses of amphotericin B
dida albicans, historically the most common cause of to be effective, are resistant to itraconazole, and have high
MICs to voriconazole. Therefore, knowledge of the
* To whom correspondence should be addressed.
(Tel) 1-404-639-3098; (Fax) 1-404-639-3546 infecting species is highly predictive of likely drug sus-
(E-mail) [email protected] ceptibility and can be used as a guide to therapy (Pappas
66 Ellepola and Morrison J. Microbiol.
et al., 2004). Second, species-specific identification is rel- sive candidiasis.

evant for epidemiological purposes; for example, repeated Culture
identification of a particular species in a given hospital Empiric antifungal treatment is recommended following
ward or locale may indicate a point source outbreak, par- recovery of even one positive blood culture (Edwards
ticularly if a species occurs at a higher rate than that found 1992; Lecciones et al., 1992; Rex et al., 1994). Multiple
historically, or if the species is unusual (i.e., C. lusitaniae, blood cultures on successive days, especially after
Candida lipolytica) (Denning et al., 2003). Third, the risk removal of or in the absence of central lines, are compel-
of developing deep organ involvement, and the severity of ling evidence for invasive disease. Several advances in
clinical manifestations, differs depending on the infecting blood culturing techniques have been developed which
species. For example, Candida tropicalis is likely to appear to have improved the sensitivity and/or reduced
become invasive in neutropenic patients but not in liver the time required to obtain a positive blood culture. These
transplant patients (Rabkin et. al, 2000; Denning et al., include the development of lysis centrifugation tubes and
2003). Identification of relevant Candida spp., however, automated monitoring of blood culture bottles. The lysis
has traditionally required isolation of the infecting organ- centrifugation system (Wampole Laboratories, USA)
ism from a normally sterile site, such as blood, peritoneal increases the yield of Candida spp. recovered from blood
fluid, or intravenous lines. However, in cases of by using a detergent to release fungi trapped within host
hepatosplenic candidiasis, blood cultures are frequently phagocytic cells. The lytic mixture disrupts the host cells
negative; indeed, blood cultures were reported to be neg- and also inactivates both complement and some antimi-
ative in up to 50% of all autopsy-proven cases of invasive crobial agents which could be harmful to the viability of
candidiasis in such patients (Berenguer et al., 1993). the fungus. Tubes are centrifuged and the resultant pellet
Although newer blood culturing systems have been plated onto a solid medium, such as chocolate agar, for
reported to increase the recovery of Candida spp. (Fuller rapid recovery of most yeasts (Archibald et al., 2000; Pro-
et al., 2001), particularly if sufficient amounts of blood cop et al., 2000). The mean time to obtain a positive
(20 ml are recommended for aerobic culture) are inocu- blood culture is faster for patients with multiple organ
lated (Washington, 1994), recovery is not 100%. In addi- involvement than for those with single organ involve-
tion, blood cultures may only become positive late in ment, suggesting a correlation with severity of disease
infection. Therefore, in the absence of a positive blood (Berenguer et al., 1993). Others have not advocated the
culture, what other laboratory tools are available to aid in routine use of lysis centrifugation tubes because of a high
the early diagnosis of invasive candidiasis? Serological rate of contamination discovered during their use (Cregar
tests have been the subject of much study but can be dif- et al., 1998). However, the main disadvantages of this sys-
ficult to interpret. For example, circulating antibodies to tem are that it is expensive and labor-intensive, precluding
Candida spp. may occur in normal individuals as the routine use in many laboratories. In addition, it appears to
result of commensal colonization of mucosal surfaces and be more useful for the recovery of more fastidious fungi
antibody production in the immunocompromised patient such as Histoplasma capsulatum, rather than for Candida
population varies according to immune status. Candida species (Vetter et al., 2001).
spp. antigens and metabolites are often rapidly cleared Recent advances in the formulation of blood culture
from the circulation so that antigen detection tests may media, especially those containing resins to absorb out
lack the desired level of sensitivity required for an residual antimicrobial drugs found in the blood of patients
unequivocal diagnosis. Histopathological evidence of tis- as well as substances normally found in blood which are
sue invasion can confirm a diagnosis but commercial inhibitory to fungal growth, have significantly improved
reagents to identify and differentiate Candida spp. are not the recovery of Candida spp. from blood culture bottles
currently available. In addition, some subpopulations of (Petti et al., 1996; Doern et al., 1998; Archibald et al.,
patients at-risk for invasive candidiasis may be thromb- 2000; McDonald et al., 2001). Perhaps the single most
ocytopenic making tissue biopsy procedures hazardous. important improvement, however, was the institution of
Newer molecular biological methods, employing poly- automated blood culturing systems with continuous
merase chain reaction (PCR) technology, hold promise growth monitoring. Either colorimetric (BacT/ALERT
but are not yet standardized or readily available in most 3D, Organon Teknika Corp., Durham, NC) or fluorescent
clinical laboratories. Also, large clinical trials to deter- (BACTEC 9240, Becton Dickinson, USA) monitoring
mine the sensitivity and specificity of such molecular tests can now be conducted automatically and electronically at
are nonexistent. Ultimately, a combination of serological, approximately 10 min intervals. In a recent study, a direct
microbiological, histological, and molecular biological comparison of these two automated blood culture systems
tests may be required for the best possible diagnosis. This was conducted to compare detection of Candida growth
article will review currently available methods, as well as (Horvath et al., 2004). Aerobic, anaerobic, and mycology
newer methods not yet commercially available, and dis- medium, all of which are normally used in each system,
cuss their assets and limitations for the diagnosis of inva- were evaluated: Plus Aerobic/F, Plus Anaerobic/F, and
Vol. 43, special issue (No. S) Laboratory diagnosis of candidiasis 67
Myco/F lytic bottles for the BACTEC system and FA, accomplished by using carbohydrate assimilation and fer-
SN, and MB bottles for the BacT/ALERT system. Each mentation tests along with Dalmau plate (corn meal-
blood culture bottle was inoculated with fresh blood from Tween 80 agar) morphology. In addition, more rapid and
healthy donors and spiked with 103 yeast cells per bottle. less laborious phenotypic identification methods have been
Both systems gave comparable results for the growth of developed. These include tests such as the RapID Yeast
all Candida spp. tested when aerobic culture media were Plus System (Innovative Diagnostic Systems, Norcross,
used: the BACTEC and BacT/ALERT detected 90% and Ga) which contains conventional and chromogenic sub-
100% of Candida spp. isolates, respectively. In addition, strates and requires only 4 to 5 h to complete (Kitch et al.,
both systems detected 50 of 50 Candida spp. isolates using 1996; Heelan et al., 1998; Wadlin et al., 1999), the Fongis-
specialty mycology medium. The major difference noted creen test (Sanofi Diagnostics Pasteur, France) (Quindos et
between these two culturing systems was for growth in al., 1993; Hoppe and Frey, 1999), and the automated Rapid
anaerobic media: the BACTEC system only detected 10% Yeast Identification Panel (Dade Microscan, USA) (Land et
of the isolates inoculated compared to 70% detected by al., 1991; St-Germain and Beauchesne, 1991). Although
the BacT/ALERT system (Horvath et al., 2004). It must such tests can identify an isolate in as little as one day, most
be taken into consideration, however, that anaerobic cul- of these tests are more accurate for the identification of
tures are not generally inoculated without also inoculating common than uncommon yeast pathogens. For instance, in
aerobic cultures so the overall detection rate by combin- one study, the RapID Yeast Plus System correctly identified
ing aerobic and anaerobic culture results would not be 96% of common yeasts but only 79% of rarer Candida spp.
much different between these two systems. However, the and 75% of other uncommon yeasts (Espinel-Ingroff et al.,
mean time to growth detection in aerobic culture medium 1998).
was significantly faster for C. glabrata isolates in the Perhaps the most convenient and popular methods for
BacT/ALERT system than in the BACTEC system (Hor- Candida species identification consist of strips or plates
vath et al., 2004). Given the reduced susceptibility of C. for carbohydrate assimilation and/or enzyme detection
glabrata to several antifungal drugs (Pappas et al., 2004), which are commercially available in an assortment of dif-
a more rapid identification of this species could be advan- ferent formats. These systems are readily and easily inoc-
tageous for clinical management of disease. ulated with yeasts and include, but are not limited to, the
Other blood culture systems which use continuous man- API 20C AUX (bioMerieux-Vitek, Hazelwood, Mo.), the
ometric monitoring, instead of colorimetric or fluorescent API Candida (bioMerieux, France), the Auxacolor
monitoring, also exist (ESP, Difco, Detroit, Mich. and (Sanofi Diagnostics Pasteur, France), and the Uni-Yeast-
O.A.S.I.S., Unipath, Inc.). A comprehensive review of all Tek kit (Remel Laboratories, Lenexa, Kan.). These tests
these systems can be found in Reimer et al. (1997). use an increase in turbidity (API 20C AUX) or the pro-
Body fluids other than blood may also be culture pos- duction of color (API Candida, Auxacolor, Uni-Yeast-
itive for Candida spp.. Although the majority of patients Tek) in each of a series of wells containing different sub-
with candiduria are asymptomatic, positive urine cultures, strates to produce a particular biochemical profile. The
in the absence of indwelling urinary catheters, which yield profile produced is then translated into a numerical code
>1×104 cfu/ml should raise suspicion of infection. Posi- that is deciphered using the manufacturer’s reference man-
tive urine cultures from some patient populations (e.g., ual. These tests give good results for the more common
those in ICUs or neonatal ICUs, burn unit patients, and species of Candida and genera of yeasts (99.8% for Uni-
transplant patients) are particularly important and organ- Yeast-Tek, 98% for API 20C AUX, 81% to 91% for Aux-
isms recovered should be identified to the species level; acolor, and 79% to 92% for API Candida) (Bernal et al.,
these species are usually the same as those found in blood 1998; Campbell et al., 1999; Moghaddas et al., 1999; Ver-
or at other sites in multiply colonized patients. For exam- weij et al., 1999). The Auxacolor and API 20C AUX tests
ple, positive urine cultures from infants in neonatal ICUs are also relatively useful for identifying common germ
often represent invasive disease and may precede candi- tube negative Candida spp. (i.e., the accuracy of these
demia (Denning et al., 2003). However, Candida spp. tests ranged from 75% to 93%) (Davey et al., 1995; Ver-
may be absent from the urine even in disseminated infec- weij et al., 1999). However, identification of less common
tion and vice versa (Tiraboschi et al., 2000). Although Candida spp. and genera are not as accurate (i.e., the
candiduria may not be a specific marker for disseminated Auxacolor and API Candida tests failed to identify Can-
candidiasis in all cases, it has been proposed that it is an dida norvegensis, Candida catenulata, Candida haemu-
indicator of poor prognosis in patients with advancing age lonii, and C. dubliniensis and the API 20C AUX had only
because of the multiple serious underlying diseases found a 90% accuracy for isolates belonging to genera of Cryp-
in this population (Tiraboschi et al., 2000). tococcus, Trichosporon, and Geotrichum) (Campbell et
al., 1999; Verweij et al., 1999). Further, although the API
Phenotypic species identification Candida system correctly identified 92% of 146 clinical
Identification of the most common Candida spp. is often isolates, 23 required supplemental biochemical or mor-
phological tests for unequivocal confirmation (Campbell rate or colony size were observed for most species and no
et al., 1999). differences in the capacity to differentiate among colonies
A recent study to differentiate C. dubliniensis from C. of C. albicans, C. tropicalis, and C. krusei were reported
albicans revealed that the API 20C AUX carbohydrate for the new formulation compared to the original one.
assimilation system incorrectly identified C. dubliniensis However, all C. albicans isolates gave a lighter shade of
as C. albicans in all but 3 of 22 cases: remaining isolates green on this medium compared to the old formulation
were misidentified as C. albicans/C. tropicalis, C. tropi- whereas C. dubliniensis isolates gave the same typical
calis/C. albicans, and C. lusitaniae/C. albicans (Ellepola dark green color on agar made with both the old and new
et al., 2003). In addition, 82% of C. albicans isolates formulations (Jabra-Rizk et al., 2001). Hence, it was pro-
tested, and 100% of C. dubliniensis isolates, assimilated posed that the new medium could not only differentiate
trehalose; the latter finding was opposite to that expected between C. albicans, C. tropicalis, and C. krusei but could
for C. dubliniensis according to the API 20C AUX ref- also differentiate C. albicans from C. dubliniensis. In con-
erence manual. Xylose and α-methyl-D-glucoside assim- trast, in a recent study, it was demonstrated that C. albicans
ilation, respectively, were negative for 100% and 95% of rather than C. dubliniensis was more likely to demonstrate
C. dubliniensis isolates and positive for 100% and 91% of a dark green or blue green colony color on CHROMagar
C. albicans isolates, confirming earlier reports that assim- Candida medium made with the new formulation (i.e.
ilation results for xylose and α-methyl-D-glucoside may 100% of C. albicans isolates were dark green or blue green
be helpful in the discrimination of these two species (Elle- versus only 64% of C. dubliniensis isolates) (Ellepola et al.,
pola et al., 2003). 2003). Fluconazole has also been incorporated into CHRO-
It was suggested that a novel medium referred to as Magar Candida medium in some studies so as to not only
Pal’s medium (sunflower seed agar) could be used to dif- identify the Candida spp. present but to also identify anti-
ferentiate C. albicans from C. dubliniensis (Mosaid et al., fungal drug resistant isolates during initial isolation (Patter-
2003). Growth on this medium for 48 to 72 h resulted in son et al., 1996; Verghese et al., 2000).
smooth creamy-grey colonies for both species, but only C. Preliminary studies have demonstrated that a newly-for-
dubliniensis isolates exhibited a hyphal fringe surround- mulated chromogenic medium (Oxoid Chromogenic
ing the colonies. However, as some isolates belonging to Candida Agar, OCCA; Oxoid, Inc., USA) allows differen-
other Candida species (C. tropicalis, C. parapsilosis and tiation of C. tropicalis (dark blue colonies), C. albicans/C.
C. krusei) also produce a fringe on Pal’s agar, this dubliniensis (green colonies) and C. krusei (dry, irregular,
medium might best be used only for germ tube positive pink-brown colonies) from colonies of C. glabrata, C. kefyr,
and/or chlamydospore positive isolates. C. parapsilosis or C. lusitaniae (beige/yellow/brown colo-
The development of a chromogenic medium (CHRO- nies), and it has been suggested that experienced users may
Magar Candida, France) which incorporates substrates be able to differentiate among the latter four species. This
linked to chemical dyes in a solid medium to differentiate medium incorporates X-NAG (5-bromo-4-chloro-3-indolyl-
Candida spp. by the color and/or texture of the growth N-acetyl-β-D-glucosaminide) and BCIP (5-bromo-6-chloro-
produced (C.albicans = green/blue green; C. tropicalis 3-indolyl-phosphate-p-toluidine salt) as chromogenic sub-
= blue/purple with halo; C. krusei = pink/ruffled) has been strates, which detect the activity of hexosaminidase and
particularly helpful for the presumptive identification of alkaline phosphatase, respectively. Chloramphenicol is also
C. albicans, C. tropicalis, and C. krusei, especially from incorporated into the medium to inhibit bacterial growth (De
fresh cultures (Odds and Bernaerts, 1994; San-Millan et Caux et al., 2004). More extensive evaluations in the clinical
al., 1996). CHROMagar Candida medium is also valuable microbiology laboratory setting are needed to determine the
for the differentiation of mixed cultures which would full potential of this medium.
ordinarily be missed during conventional plating on solid Additional methods for Candida spp. identification
medium (Pfaller et al., 1996; Powell et al., 1998). Some include automated biochemical systems such as the ID
reports have suggested that this medium can also be used 32C strip system (bioMerieux), the Vitek Yeast Biochem-
to differentiate C. glabrata from other yeast species ical Card system (bioMerieux Vitek, Inc., Hazelwood,
(Pfaller et al., 1996; Willinger and Manafi, 1999). In con- Mo.), the Vitek 2 ID-YST card system (bioMerieux Vitek,
trast, others reported that it can not be used for this pur- Inc.), the Quantum II (Abbott Laboratories, USA), and the
pose because Candida kefyr, C. lusitaniae, Candida Biolog YT MicroPlate system (Biolog, USA), to mention
guilliermondii, Candida famata, Candida rugosa, Can- only a few. Details of a variety of other colorimetric and
dida utilis, Candida robusta, and Candida pelliculosa all enzymatic systems, as well as fatty acid analysis and
produce the same type of glossy pink colonies as C. gla- physicochemical spectroscopic identification methods can
brata, leading to misidentification (Baumgartner et al., be found in a review by Freydiere et al. (2001).
1996; Freydiere, 1996; Freydiere et al., 1997a).
CHROMagar Candida was recently reformulated (Bec- Molecular biological identification of Candida species iso-
ton Dickinson) but no significant differences in growth lated in culture
The limitations described above for phenotypic identifi- amplified ITS2 region by automated capillary electro-
cation techniques have led to the development of nucleic phoresis. Unique, species-specific PCR products were
acid-based identification systems. These systems offer the obtained from 92% of the clinical isolates tested; the
potential for more rapid and specific identification of remaining 8% contained DNA sequences that required
Candida spp. compared to traditional phenotypic meth- restriction enzyme analysis for differentiation. These data,
ods. Most nucleic acid-based systems use polymerase and the specificity of length polymorphisms for identify-
chain reaction (PCR) techniques to amplify fungal DNA ing yeasts, were confirmed by DNA sequence analysis of
as the first step in the identification process. Before PCR the ITS2 region for 93 isolates. Phenotypic and ITS2-
amplification can occur, appropriate DNA targets and based identifications were concordant for 427 of 434
PCR primers must be selected. Highly conserved regions yeast isolates examined. The remaining seven clinical iso-
of ribosomal DNA have been the most popular targets for lates contained ITS2 sequences that did not agree with
PCR amplification and have included the 5.8S, 18S, and their phenotypic identification, and ITS2-based phyloge-
28S rRNA genes (Sandhu et al., 1995; Kurtzman and netic analyses indicated the possibility that these isolates
Robnett, 1997; Mannarelli and Kurtzman 1998; Hui et al., represented new or clinically unusual species among the
2000; Jaeger et al., 2000; Loeffler et al., 2000a; Guiver et Rhodotorula and Candida genera. Whereas it was sug-
al., 2001). Other targets have included the more variable gested that size estimation and restriction enzyme analysis
internal transcribed spacer (ITS) regions between these of PCR-amplified ITS2 region DNAs were rapid and reli-
genes (Elie et al., 1998; Chen et al., 2000; Ellepola et al., able methods to identify clinically significant yeasts,
2003; Coignard et al., 2004; Massonet et al., 2004) or the including potentially new or emerging pathogenic species
intergenic spacer (IGS) region (Cirak et al., 2003). (Chen et al., 2000), few clinical laboratories have the
The main advantage of using amplification targets from capability or budget to conduct DNA sequence-based
regions of DNA which are conserved among all Candida methods and, in particular, restriction enzyme fragment
species is that a PCR product can be obtained from all length analyses.
species using a single set of PCR primers and conditions The application of an enzyme immunoassay (EIA) for-
optimal for that set of primers. Following amplification, mat, using a colorimetric substrate, is perhaps the easiest
species-specific probes can be designed from the more and least costly method available for amplicon detection
variable DNA regions, located between primer binding (Elie et al., 1998; Loeffler et al., 1998). It is also user-
sites, for the identification of specific organisms (Elie et friendly, does not require any dangerous radioactive
al., 1998; Ellepola et al., 2003; Coignard et al., 2004) or reagents, and can be automated. One example of such an
amplicons can be subjected to gel electrophoresis, with or assay uses universal fungal primers flanking the ITS2
without the use of restriction enzyme digestion, followed region of ribosomal DNA. Species-specific DNA probes,
by ethidium bromide staining to differentiate among the directed to the ITS2 region and labeled with digoxigenin,
species (Buchman et al., 1990; Crampin and Matthews are hybridized to the PCR amplicons along with a bioti-
1993; Hopfer et al., 1993; Rand et al., 1994; Wildfeuer et nylated all-Candida spp. probe which functions to capture
al., 1996; Morace et al., 1997; Irobi et al., 1999; Cirak et amplicons onto streptavidin-coated wells of a 96-well
al., 2003; Graf et al., 2004). Increased sensitivity can be microtiter plate. Amplicons are then detected in an EIA
achieved by Southern blotting of such gels onto nylon format (PCR-EIA) using horseradish peroxidase-labeled,
membranes and by detection with radiolabeled or colori- anti-digoxigenin antibodies, H2O2, and a colorimetric sub-
metric probes (Crampin and Matthews, 1993; Miyakawa strate. The assay can be completed in a single day and can
et al., 1993; Burgener-Kairuz et al., 1994; Holmes et al., identify up to 18 medically important Candida spp.,
1994; Jordan, 1994; van Deventer et al., 1995; Loeffler et including C. dubliniensis (Elie et al., 1998; Ellepola et al.,
al., 2000a). Others have used line-probe, reverse cross 2003; Coignard et al., 2004). This detection format is par-
blot or slot blot detection methods (Loeffler et al., 2000a). ticularly amenable to the identification of Candida spp. in
More sophisticated detection methods include analysis by a clinical laboratory.
direct fluorescent capillary automated DNA sequencing, Other PCR-based identification methods to specifically
amplified fragment length polymorphism (AFLP), ran- differentiate C. albicans from C. dubliniensis have
domly amplified polymorphic DNA (RAPD), or single included comparisons of ACT-1 intron amplicons, PCR
strand conformational polymorphisms (SSCP) (Sandhu et amplicons obtained using primers derived from the pH-
al., 1995; Walsh et al., 1995a; Hui et al., 2000; Loeffler regulated PHR1 gene of C. albicans, or from flanking
et al., 2000b; Martin et al., 2000; Posteraro et al., 2000; regions of the CaLSU intron (Haynes et al., 1995;
Bautista-Munoz et al., 2003; Borst et al., 2003). Boucher et al., 1996; Donnelly et al., 1999; Kurzai et al.,
Sequence polymorphisms in the ITS2 region of the 1999). Each of these non-EIA methods have limitations in
rRNA genes have also been used as a means to identify that they either lack a specificity control and, as such, can
yeasts. Chen et al. (2000) used 434 isolates, representing be prone to potential misidentifications (PHR1 gene
34 species of yeasts to determine the length of the PCR- method), require large amounts of DNA template (PHR1
gene method), have fluctuating reproducibility regarding their respective species; there was no cross-reaction
agarose gel detection of the larger of two required DNA between any set and human, fungal, bacterial or viral
bands (ACT-1 intron method), or require multiple internal DNA tested (Guiver et al., 2001). The TaqMan assay has
controls. These methods also require gel electrophoresis the advantage of reducing post-amplification manipula-
of the PCR amplicons for isolate identification. Molecular tion steps but the initial cost for the equipment and lab-
mass can be difficult to interpret in an agarose gel and can oratory set up to perform such assays is prohibitive in
vary depending upon the composition of the agarose used, many laboratory settings.
the length of run time, the size of the gel format, and the Another “real time” PCR method is the LightCycler
molecular markers employed. Few clinical laboratories system (Roche Molecular Systems, Indianapolis, Ind.)
are equipped to use electrophoresis gels on a regular basis which allows rapid amplification of DNA in glass capil-
whereas an EIA detection format is more readily adapt- laries and simultaneous fluorescent detection of ampli-
able. cons using fluorescence energy transfer or “FRET”
The PCR-EIA method has also been used to resolve dis- (Loeffler et al., 2000b; Schmidt et al., 2001). One DNA
crepant phenotype-based Candida spp. identifications probe is labeled at the 3' end with fluorescein and another
between the Centers for Disease Control and Prevention probe is labeled at the 5' end with Light Cycler Red 640
(CDC) laboratories and referring hospitals participating in fluorophore. The fluorescein is excited by the light source
an active, population-based surveillance for candidemia of the Light Cycler instrument and the energy emitted by
(Coignard et al., 2004). Species identifications were per- the fluorescein is transferred to the Light Cycler Red 640
formed at these institutions and then confirmed at the fluorophore. The light emitted by the fluorophore is then
CDC by phenotype-based methods using CHROMagar measured and is proportional to the amount of DNA
Candida medium and the API 20C AUX system. Discrep- amplified. Selvarangan et al. (2002) used this method in
ancies in species identification between the referring insti- an attempt to distinguish C. albicans from C. dubliniensis.
tution and the CDC were noted for 43 of 935 isolates Fluorescently-labeled nucleic acid probes, specific for C.
(4.6%). The PCR-EIA results were identical to the CDC albicans and C. dubliniensis, were bound to PCR ampl-
phenotypic identification method for 98% of the isolates icons. Subsequent denaturation produced characteristic
tested. Discrepancies between the CDC phenotypic peak melting temperatures (Tm) that identified each of the
method and the PCR-EIA results were resolved by DNA two species. No signal was generated when the C. albi-
sequence analysis of the ITS1 rRNA gene, which con- cans or C. dubliniensis probes were tested against DNA
firmed the PCR-EIA result. Among all misidentifications, from other Candida spp.. However, both probes also
the referring institutions most frequently misidentified C. reacted with C. tropicalis DNA, but the average Tm values
glabrata (37% of all discrepant identifications). Such mis- were sufficiently different to allow differentiation of C.
identifications could lead to the administration of inap- tropicalis from C. albicans and C. dubliniensis. Hsu et al.
propriate therapy given the propensity of C. glabrata to (2003) also used a “real-time” LightCycler assay to detect
develop resistance to azole antifungal drugs. Hence, use and identify Candida spp.. In this instance, template DNA
of the PCR-EIA for species identification might circum- from different Candida species was amplified and detected
vent such problems. in real time by employing SYBR Green fluorescent dye
Most recently, detection methods have been developed and specific target sequences from the 18S and 28S
which are referred to as “real-time” PCR. Detection of the regions of rDNA. C. albicans, C. glabrata, C. krusei, C.
amplicon occurs as the PCR product is produced and parapsilosis, C. guilliermondii, and C. tropicalis could be
quantitative results can be graphically displayed during discriminated by species-specific primers and confirmed
the process. Therefore, post-amplification manipulation of by melting-curve analyses (Hsu et al., 2003).
the product is not required. The TaqMan system (Perkin- A DNA sequence-based approach for the identification
Elmer; Applied Biosystems, USA) is a fluorogenic assay of medically important fungi, including Candida spp., by
which uses a reporter and a quencher dye in proximity to automated capillary electrophoresis of PCR products has
each other on the detector probe. As DNA amplification also been developed (Pryce et al., 2003). In this study, the
occurs, the 5'→3' endonuclease activity of the Taq DNA entire ITS1, 5.8S, and ITS2 ribosomal DNA region was
polymerase separates the quencher dye from the reporter sequenced using automated dye termination sequencing
dye allowing signal to be detected. Guiver et al. (2001) on 89 clinical isolates. Eighty eight of the 89 isolates had
used the TaqMan system and species-specific primer and DNA sequences that were highly homologous to those of
probe sets for the identification of C. albicans, C. gla- reference strains accessioned in GenBank and 87 of 89
brata, C. kefyr, C. krusei, and C. parapsilosis. The probes gave a sequence-based identification result that correlated
were labeled with three fluorescent dyes to enable differ- with the traditional phenotypic identification. Sequence-
entiation between species when three primer and probe based identification of Candida spp. in pure culture was
sets were combined in two multiplex reactions. The obtained within 24 h of DNA extraction. Another study
primer and probe sets were shown to be 100% specific for describing sequenced-based yeast identification used a
commercially available MicroSeq D2 large-subunit ribo- ferentiate infections by Candida spp. from those of other
somal DNA sequencing kit (Applied Biosystems, USA). fungi (Kaufman et al., 1997). A recent study described the
The use of this method, together with the API 20C AUX use of an IgG1 monoclonal antibody, 3H8, directed
system, revealed that of 100 isolates sequenced (repre- against C. albicans cell wall mannoprotein, to specifically
senting 19 species of Candida), 98% gave results concor- recognize C. albicans in culture and in paraffin-embedded
dant with identifications made by the API 20C AUX tissue sections using immunofluorescent and immunohis-
system. C. dubliniensis, however, was not included in the tochemical staining (Marcilla et al., 1999). This antibody
MicroSeq database and was identified as C. albicans preferentially detected mycelial forms and, to a lesser
(Hall et al., 2003). Several other less common yeasts were extent, blastospores of C. albicans and did not react with
also not included in the MicroSeq database. any other Candida spp. tested. This monoclonal antibody
In general, nucleic acid sequencing may provide the was originally produced for use in a latex agglutination kit
greatest benefit to the laboratory by identifying organisms (Bichro-latex albicans, Fomouze Diagnostics, Asnieres,
whose identities are questionable or cannot be determined France) for the rapid identification of C. albicans in cul-
by phenotypic methods. However, a cost analysis, com- ture (Freydiere et al., 1997b; Quindos et al., 1997; Mar-
paring phenotypic testing to nucleic acid sequencing, cilla et al., 1999). In addition, differentiation of C.
showed that the cost of sequencing was $29.50 higher albicans from C. dubliniensis has been reported by indi-
than the cost of using the API 20C AUX system (Hall et rect immunofluorescence (Bikandi et al., 1998). In this
al., 2003). The MicroSeq D2 library was also somewhat study, anti-C. dubliniensis serum was adsorbed with C.
limited and did not include sequences for all clinically albicans blastospores so that no reactivity was observed
important yeast species. However, the potential exists to against C. albicans or several other Candida species,
allow each laboratory to construct a custom database for except C. krusei.
the most common species found in a given laboratory, Fluorescent in-situ hybridization (FISH), using oligonu-
thereby making the system more useful. Constructing a cleotide probes directed against 18S rRNA has been used
database of species not already included in the MicroSeq to differentiate C. albicans from C. parapsilosis in tissues
D2 library and of isolates that show some genetic diversity of infected mice (Lischewski et al., 1997). The C. albi-
among their sequences would enhance the flexibility of cans probe detected fungal cells in tissue sections of the
the MicroSeq D2 sequencing kit. Other limitations of kidney, spleen, and brain of mice infected with C. albi-
direct sequencing systems are in the inaccuracies and cans, but not in tissues from mice infected with C. parap-
incompleteness of some sequence databases. For instance, silosis and vice versa for the C. parapsilosis probe. In
commercially available databases often contain sequences addition, the C. albicans probe could detect as few as
from cultures which may not have been accurately iden- three C. albicans cells per 500 µl of spiked human blood
tified initially (Hall et al., 2003). Public databases such as after a lysis-filtration assay and subsequent FISH (Lis-
GenBank are not refereed and contain sequencing errors chewski et al., 1997).
as well as nomenclature errors. Selection of gene sequenc- A new FISH method, using fluorescein-labeled peptide
ing targets can also greatly alter identification results (i.e., nucleic acid (PNA) probes targeting 26S rRNA, detects C.
use of conserved gene regions tend to “lump” some spe- albicans directly in smears taken from positive blood cul-
cies together whereas more variable gene targets, such as ture bottles (Rigby et al., 2002). The PNA probe is added
the ITS and IGS regions of rDNA, tend to “split” species and hybridized. Unhybridized probe is removed by wash-
apart). Direct sequencing also requires technical expertise ing, and the smears are examined by fluorescence micros-
and sophisticated equipment which is not available in copy. The performance of the C. albicans PNA FISH
many clinical laboratories. method as a diagnostic test was evaluated with 33 routine
and 25 simulated yeast-positive blood culture bottles and
Histopathology showed 100% sensitivity and 100% specificity (Rigby et
Histopathological examination of tissue sections is one of al., 2002). Oliveira et al. (2001) also used a PNA FISH
the most reliable methods of establishing a diagnosis of assay but to differentiate C. albicans from C. dubliniensis.
systemic fungal infection. However, the ease with which Samples from liquid cultures were smeared onto micro-
a fungal pathogen can be recognized in tissue depends on scope slides, heat fixed, and then hybridized with probes
its abundance and the distinctiveness of its appearance. for 30 min. Unhybridized probe was removed by wash-
The presence of typical blastospores and pseudohyphae of ing, and slides were examined by fluorescence micros-
Candida spp. in histochemically-stained tissue sections copy. Evaluation of the PNA FISH method using 79 C.
can be used as a diagnostic parameter for invasive can- dubliniensis and 70 C. albicans isolates showed 100%
didiasis. However, the production of fluorescent antibod- sensitivity and specificity for both probes.
ies specific for the identification of individual Candida
spp. has proved extremely difficult. Generic reagents Antibody detection
which react across Candida spp. have been used to dif- The clinical usefulness of antibody detection for the diag-
nosis of systemic candidiasis has been limited by false- albicans cell wall mannan (Platelia Candida Antibody
negative results in immunocompromised patients who test, Bio-Rad Laboratories, France) and the other to detect
produce low or undetectable levels of antibody and by mannan serum antigen using a monoclonal antibody
false-positive results in patients with superficial coloniza- (Platelia Candida Antigen test, Bio-Rad). The sensitivity
tion. A study was conducted to evaluate the usefulness of and specificity for the antibody test alone was reported to
antibody detection by double immunodiffusion (ID) in be 53 and 94%, respectively, whereas these values
214 patients admitted to the intensive care unit of a uni- changed to 80 and 93% when a combination of both anti-
versity hospital (Pallavicini et al., 1999). Patients were body and antigen detection tests were employed. It was
followed over a period of 42 months for the development suggested that a combination of both tests may be more
of invasive candidiasis. Seventeen percent of patients useful for the diagnosis of invasive candidiasis than either
developed invasive candidiasis but the sensitivity and test alone. In a later study, Yera et al. (2001) examined the
specificity of the ID test was only 29% and 67%, respec- contribution of these same two tests for the diagnosis of
tively, indicating that the ID test is not very useful for the invasive candidiasis in relation to the time of positive
diagnosis of invasive candidiasis in this patient popula- blood culture. In patients who presented with at least one
tion. positive blood culture (n=45), one or both serological tests
In an attempt to reduce false-positive results, several were positive in 73% of patients at least 2 days, and in
researchers have developed tests to detect antibodies some patients, up to 15 days before blood cultures became
directed against cytoplasmic antigens, assuming that the positive. It was suggested that serological surveillance of
host would not be exposed to intracellular antigens except patients at-risk for invasive candidiasis could therefore
during invasive disease. Unfortunately, in a study of result in earlier implementation of antifungal therapy than
patients undergoing induction chemotherapy for acute blood culture if both serological tests were employed
leukemia, antibody to a major 54 kDa cytoplasmic anti- simultaneously. It is of particular interest, however, that in
gen was detected in only 25% of patients with dissemi- this same study, patients were generally either positive for
nated candidiasis (Jones 1980a; 1980b; Greenfield and antibodies or for antigen but generally not for both simul-
Jones, 1981; Greenfield et al., 1983) and others found taneously. In addition, the kinetics of test positivity was
increases in antibody titers in 10% of patients without dependent upon the hospital ward in which the patient
candidiasis (Fujita et al., 1986). In contrast, El Moudni et was housed: i.e., hematology and ICU patients were more
al. (1998) described highly successful detection of anti- likely to have had positive antigen tests whereas surgical
bodies to a high-performance liquid chromatographically patients were more likely to have had positive antibody
purified 52 kDa metalloprotein of C. albicans in an tests. These results are not surprising given the more
enzyme-linked immunosorbent assay (ELISA). However, immunosuppressed state that might be associated with
they failed to specify whether the patients studied were patients from the former two hospital units.
immunocompromised. Nonetheless, at a cutoff absor- Indeed, the fact that antibody production can be variable
bance of 0.425, test sensitivity and specificity was or nonexistent in immunocompromised patients make
reported to be 83% and 97%, respectively. It was there- tests to detect antibodies for the diagnosis of systemic
fore suggested that this aminopeptidase may be a useful candidiasis less useful in this patient population. How-
antigen for the detection of antibodies formed during ever, immunosuppressed patients may be in antigen
invasive candidiasis. excess, making the detection of antigens a potentially
Na and Song (1999) described an ELISA assay for the more successful strategy for the diagnosis of invasive can-
detection of antibodies to the secreted aspartyl proteinase didiasis in these patients groups.
(Sap) of C. albicans, a putative virulence factor released
during tissue invasion (Ray et al., 1991; Hube, 1996). In Antigen detection
a retrospective evaluation of 33 patients with invasive Numerous circulating antigens have been used as poten-
candidiasis, the sensitivity and specificity, respectively, tial targets for the diagnosis of disseminated candidiasis.
for this test was only 70% and 76%, making it less desir- One such antigen is an inducible, extracellularly secreted
able for the diagnosis of invasive candidiasis than Sap aspartyl proteinase (Sap) produced by C. albicans and
antigen detection tests (for a description of the latter tests, some other Candida spp.. It has been studied extensively
please see the antigen detection section of this paper). as a virulence factor in the invasion and dissemination of
Some researchers have also examined the usefulness of C. albicans in animal models of infection (Staib, 1965;
combined antigen and antibody detection tests for the Kwon-Chung et al., 1985; Ruchel, 1992; Hube 1996;
diagnosis of invasive candidiasis. A retrospective study Morrison et al., 2003). The theoretical usefulness of Sap
(Sendid et al., 1999), using 162 serum samples from 43 as a diagnostic antigen stems from the hypothesis that
patients with mycologically and clinically proven candid- because Sap is an inducible enzyme, produced during
iasis, was conducted to evaluate the usefulness of two active tissue invasion (MacDonald and Odds, 1983; Ray
commercial ELISA tests: one to detect antibodies to C. and Payne, 1988), its production should correlate with
invasive disease and not simple colonization. seminated candidiasis depending on the frequency of
Ruchel et al. (1988) examined serum samples from sampling, the underlying disease, the degree of immuno-
patients for the utility of Sap detection as an aid to the suppression, the serotype of C. albicans, the Candida spe-
diagnosis of invasive disease. Using anti-Sap antibodies cies involved, the definition of disseminated candidiasis,
in an ELISA format, the sensitivity of detection was low the specificity and titer of the capture antibodies, and the
(positive in 50% of suspected plus confirmed cases) and immunoassay method used. Numerous laboratories have
may have been the result of complex formation between attempted to use radioimmunoassay (RIA), enzyme-linked
Sap and alpha-2-macroglobulins in the circulation immunosorbent assay (ELISA), latex agglutination (LA)
(Ruchel and Boning-Stutzer, 1983). or reverse passive latex agglutination (RPLA) to detect cir-
More recently, Na and Song (1999) compared two dif- culating mannan (Bougnoux et al., 1990; Lemieux et al.,
ferent ELISA assays for their efficacy to detect Sap anti- 1990; Reiss and Morrison, 1993).
gen in serum: an inhibition ELISA and an antigen capture Methods such as the sandwich ELISA (Bougnoux et al.,
ELISA. Both antigen detection tests used a monoclonal 1990; Lemieux et al., 1990; Fujita and Hashimoto, 1992)
antibody, CAP1, which was reported to be specific for C. and RPLA (Bailey et al., 1985; Kahn and Jones, 1986),
albicans Sap. The sensitivities and specificities of these have moderate sensitivity but good specificity for dissem-
tests, respectively, were 94% and 92% for the antigen cap- inated disease. In a retrospective study of patients with
ture ELISA and 94% and 96%, for the inhibition ELISA. malignancy, the sensitivity and specificity of the sandwich
These data suggest that the inhibition ELISA could be ELISA was 65% and 100%, respectively (de Repentigny
useful for the diagnosis of invasive C. albicans infections. et al., 1985). Fujita and Hashimoto (1992) compared the
However, further studies are warranted to validate these sensitivity of the sandwich ELISA format to that of the
findings in a well-controlled, prospective study. RPLA format using the same capture antibodies for both.
The potential use of proteinases as markers of invasive They found the sensitivity of the RPLA to be 38%
candidiasis has led to the development of a competitive whereas that of the sandwich ELISA was 74% (Fujita and
binding inhibition ELISA to detect Sap in serum and urine Hashimoto 1992). In other studies, the RPLA test detected
obtained from a rabbit model of disseminated candidiasis serum mannan in 78% of leukemic patients with dissem-
(Morrison et al., 2003). Although ELISA inhibition was inated candidiasis (Kahn and Jones, 1986) and in 72% of
observed when serum specimens were used, more signif- patients for whom disseminated candidiasis was con-
icant inhibition, which correlated with disease progres- firmed by biopsy, autopsy, or persistent candidemia dur-
sion, occurred when urine specimens were tested. Urine ing granulocytopenia (Bailey et al., 1985). In a more
collected as early as 1 day after infection resulted in sig- recent study, cerebrospinal fluid samples from five
nificant ELISA inhibition compared with preinfection patients from whom Candida cells were cultured were
control urine, and inhibition increased until experimental tested for the presence of mannan using an ELISA (Ver-
termination on day 5. The overall test sensitivity and spec- duyn Lunel et al., 2004). Samples from four of five
ificity was 83% and 92%, respectively. The specificity patients determined to have proven Candida meningitis
could be increased to 97% if at least two positive test reacted positively in the assay. Samples from the remain-
results were required to define a positive result. The use ing patient and from patients with other central nervous
of Sap detection in urine could therefore provide a non- system infections were negative. Although sample num-
invasive means to diagnose disseminated candidiasis bers in this study were small, test results are encouraging.
(Morrison et al., 2003). A sandwich ELISA, commercially available as the
Mannan, the major cell wall mannoprotein of C. albi- Platelia Candida antigen test (Bio-Rad), was developed
cans, is another diagnostic antigen. Dissociation of anti- for the diagnosis of systemic candidiasis based on the
gen-antibody complexes is necessary for the optimal detection of α-linked oligomannose residues (α-Man)
detection of mannan in the circulation. This antigen is released from Candida cells into the serum. This test has
heat stable and resists boiling, proteinase treatment, and been shown to have good specificity but must be repeated
acidic pH (Reiss and Morrison, 1993). Therefore, antigen- frequently because of the rapid clearance of this form of
antibody complexes are routinely dissociated by boiling in mannan from the circulation (Sendid et al., 2004). There-
the presence of EDTA or by enzymatic treatment. Bailey fore, a second ELISA, based on the detection of β-linked
et al. (1985) detected mannan in the serum of 17 of 21 oligomannoses (β-Man), was developed. It was suggested
patients with disseminated candidiasis when specimens that because β-Man are linked to different Candida mol-
were treated with pronase and heat, whereas only 3 of 21 ecules and interact differently with the host immune sys-
patients were positive if no dissociation step was tem and endogenous lectins than does α-Man, that β-Man
included. Mannan is cleared rapidly from the circulation, may therefore possess different serum clearance kinetics.
resulting in low serum concentrations, and multiple serum In a guinea pig model of systemic C. albicans infection,
sampling is required for optimal detection. Mannanemia the relative amounts of detectable α- and β-Man differed
occurs in approximately 31 to 90% of patients with dis- considerably according to the virulence of the strain, the
infecting dose, and the time after challenge that serum available commercially.
samples were drawn. Although detection of α-Man was In contrast to previously described antigens which had
more sensitive per serum sample than detection of β-Man, been identified and chemically purified, Gentry et al.
the sensitivity to detect the combination of both reached (1983) described detection of a structurally uncharacter-
90%. The same tests were then applied to 90 sera from 26 ized, 56oC-labile antigen, by RPLA. Latex particles were
patients selected retrospectively for having been infected sensitized with serum from rabbits immunized with
with C. albicans, C. tropicalis, and C. glabrata. A total of whole, heat-killed C. albicans blastoconidia. The test was
22 patients had positive antigenemia: 4 had α-mannane- commercialized as the Cand-Tec test (Ramco Laborato-
mia, 4 had β-mannanemia, and 14 showed the presence of ries, Houston, Tex.). The circulating antigen was heat-sen-
both types of mannan. For patients showing the presence sitive and susceptible to pronase, 2-mercaptoethanol, and
of both forms of mannan, the use of both tests enhanced sodium periodate treatment. The sensitized latex particles
the duration of mannanemia detection. A combined use of could not agglutinate mannan and it has been suggested
both tests had a cumulative specificity of 95%, and pos- that the assay may detect a neoantigen derived from C.
itive and negative predictive values of 79% and 97%, albicans after host processing or a host antigen which
respectively. These findings indicate that detection of both cross-reacts with those of C. albicans. The test was rel-
epitopes may contribute to increased test sensitivity (Sen- atively easy to perform. However, it lacks sensitivity
did et al., 2004). when an antigen titer of 1:8, which excludes most false-
Detection of cytoplasmic proteins of C. albicans has positive results, is used (Bougnoux et al., 1990; Lemieux
also been used diagnostically employing a variety of test et al., 1990; Phillips et al., 1990) and test specificity is
formats (Stevens et al., 1980; Araj et al., 1982; Strockbine also low (Lemieux et al., 1990). More recently, Pallavi-
et al., 1984a; Matthews et al., 1987; Matthews and cini et al. (1999) conducted a study to evaluate the use-
Burnie, 1988). The two predominant cytoplasmic proteins fulness of the Cand-Tec latex agglutination test. Over a
described to date include a 47-kDa protein which is period of 42 months, 214 patients admitted to the ICU of
thought to be a breakdown product of a 90-kDa heat- a university hospital were followed for the development
shock protein (HSP-90) and a 48 kDa protein, which was of invasive candidiasis. Although 17% of the patients
later found to be a C. albicans enolase (Strockbine et al., developed invasive candidiasis, the positive predictive
1984b; Matthews et al., 1987; Mason et al., 1988; Sund- value of the Cand-Tec test was low (13-17%). Most stud-
strom and Aliaga, 1992). Antigens in the 47 to 52 kDa ies to date suggest that the Cand-Tec test does not provide
range are not resolved by Western blot analysis unless adequate predictive value for a reliable diagnosis of dis-
monoclonal antibodies that specifically recognize the eno- seminated candidiasis.
lase antigen is applied (Strockbine et al., 1984b). The 47 Although the Cand-Tec test is not a reliable indicator of
kDa antigen could be detected in the sera of 77% of neu- invasive candidiasis, one study found this test to be a use-
tropenic patients with disseminated candidiasis using an ful predictor for timing the initiation of antifungal drug
enzyme-linked dot immunobinding assay (Matthews and therapy (Iwasaki et al., 2000). The Cand-Tec assay was
Burnie, 1988). This assay proved to be more sensitive performed serially on 10 patients with acute leukemia
than an RPLA test for the same antigen (Burnie, 1985). during 12 febrile episodes following post-remission che-
Early animal studies in mice and rabbits, using an motherapy. Febrile neutropenia after antileukemic chemo-
ELISA format, revealed that the presence of the 48 kDa therapy and an increased Cand-Tec titer relative to that
antigen in serum correlated with disseminated disease, measured before antileukemic chemotherapy were used as
was positive in the absence of candidemia, and declined indicators to administer intravenous azole antifungal drug
following antifungal therapy (Walsh and Chanock, 1998). therapy. In 9 of the 11 evaluated cases, antifungal therapy
The assay was commercialized as a double-sandwich was effective and the Cand-Tec titers declined to less than
liposomal assay using murine IgA monoclonal antibody or equal to baseline. In contrast, for the remaining two
adsorbed to a nitrocellulose membrane (Directigen1-2-3 cases, where antifungal drug therapy failed, the Cand-Tec
Disseminated Candidiasis Test, Becton Dickinson, USA). titers did not decline. The Cand-Tec test was therefore
Patient serum was added and polyclonal rabbit anti-C. suggested to provide a means to prevent excess use of
albicans enolase was applied and detected with a lipo- antifungal agents and to reduce the potential development
some coated with goat anti-rabbit IgG and containing a of azole-resistant Candida spp. infections. Recently, a
rhodamine dye (Walsh and Chanock, 1998). A multi- modification of the Cand-Tec test was described which
center study conducted at cancer centers over a two year uses a microtiter plate format (Cand-Tec MT, Ramco,
period revealed a sensitivity per sample of 54%; when Japan) and which expresses the Candida antigen level as
multiple samples were tested, detection of antigenemia a cutoff index (CI) by colorimetric analysis. The sensi-
was improved to 85% (Walsh et al., 1991). Hence, mul- tivity and specificity of the Cand-Tec MT assay to diag-
tiple sampling was recommended to optimize the detec- nose deep-seated Candida infection in 25 patients with
tion of antigenemia. Unfortunately this test is no longer hematologic diseases were 100% and 80%, respectively.
The CI value was followed during antifungal therapy and many samples to be processed rapidly. The same instru-
it decreased in 75% of cases that responded to antifungal ment could also be employed to detect serum creatinine
therapy, but did not show this tendency in non-responsive levels which could then be used to normalize readings to
cases. Therefore, this new format may improve the sen- compensate for the increased levels of serum arabinitol
sitivity of the Cand-Tec test but specificity issues remain. observed during renal dysfunction.
This study also supports the potential usefulness of this Walsh et al. (1995b) used the enzymatic-colorimetric
test to monitor response to therapy (Misaki et al., 2003). method in a study of 3223 serum samples from 274
In yet another study, the combined detection of Candida patients with cancer. They found that among patients with
antigen and antibody for the determination of systemic candidemia, the mean maximum serum D-arabinitol/cre-
Candida infections was investigated (Bar and Hecker, atinine ratio was 11 times greater than that of normal
2002). One hundred and four patients from an intensive blood bank donors and 4 times greater than that of all
care unit were analyzed in this study. Seventeen of the patient controls. Patients with persistent candidemia had
patients were suspected to have systemic Candida infec- the highest D-arabinitol/creatinine levels: 83% of 30
tions based on clinical and laboratory criteria. In these patients with persistent candidemia had elevated levels
patients, Candida antigens (Cand-Tec LA assay) and anti- compared to 74% of 42 cases of non-persistent candi-
bodies were analyzed. The sensitivity and specificity were demia. Elevated ratios preceded positive blood cultures in
58.8% and 97.6% for antigenemia, and 52.9% and 85.7% up to 50% of the cases. Moreover, serial D-arabinitol/cre-
for antibody detection. These values reached 100% and atinine ratios correlated with therapeutic response in 85%
83.3%, respectively, when the results of both tests were of 34 patients with evaluable cases of candidemia,
combined, indicating that the combined use of antigen decreasing in 89% of 9 patients with clearance of candi-
and antibody detection might improve the usefulness of demia and increasing in 84% of 25 patients with persistent
the Cand-Tec test. candidemia (Walsh et al., 1995b).
More recently, a recombinant C. albicans D-arabinitol
Detection of D-arabinitol dehydrogenase has been produced in E. coli and purified
Invasive candidiasis can also be diagnosed by detecting a by dye-ligand affinity chromatography. Whereas the pre-
metabolite, D-arabinitol, in serum or urine, produced by vious enzyme preparations metabolized host D-mannitol
most medically important Candida species except for C. as well as arabinitol, the purified C. albicans recombinant
krusei and perhaps C. glabrata (Christensson et al., enzyme only cross-reacts with xylitol (4.9%) among all
1999). It is uncertain whether the human host produces polyols tested (Yeo et al., 2000). The system using the
one or both enantimers of arabinitol or if baseline levels recombinant C. albicans enzyme could be automated to
in serum and urine are the result of dietary or microbial measure the initial production rate of NADH spectroflu-
arabinitol absorbed by the gut (Wong et al., 1990; Chris- orometrically in a COBAS FARA II centrifugal autoana-
tensson et al., 1999). Nonetheless, natural host serum ara- lyzer (Roche Diagnostic Systems) and creatinine levels
binitol accumulates during renal insufficiency so that D- can be measured simultaneously. For 11 patients with
arabinitol levels need to be reported as a D-arabinitol/cre- invasive candidiasis, the mean D-arabinitol/creatinine
atinine ratio to compensate for this occurrence (Wong et ratio was 2.74 µM/mg/dl whereas for healthy controls it
al., 1982). was only 1.14 µM/mg/dl (Yeo et al., 2000). Hence, the
Switchenko et al. (1994) used a semi-automatic enzy- higher throughput and ease of use of the newer methods
matic-colorimetric assay in which NADH-dependent end- to detect D-arabinitol in body fluids may allow for more
products were measured. In an effort to reduce cross- clinical utility of this test. Currently, however, D-arabini-
reactions with D-mannitol, these researchers employed a tol detection is only conducted in a limited number of
more highly purified arabinitol dehydrogenase enzyme research and specialty laboratories.
than that used previously. Whereas previous enzymes
were derived from bacteria, this enzyme was derived from Detection of (1→3)-β-D-Glucan
C. tropicalis and was cloned and expressed in Escherichia The cell walls of Candida spp contain (1→3)-β-D-glucan
coli (Walsh et al., 1994). In this assay, the D-arabinitol is (BDG) as a structural component. As this polysaccharide
converted to D-ribulose by the recombinant arabinitol is not found in bacteria, viruses, or mammals, its presence
dehydrogenase in the presence of NAD. The resultant in the circulation of patients has been used as an indicator
NADH reduces iodonitrotetrazolium in the presence of of invasive disease. An assay to detect BDG has been
diaphorase to form a blue black formazan dye. This dye developed which utilizes the activation of a clotting path-
is detected spectrophotometrically at 500 nm. The authors way found in amoebocyte lysates of the Japanese horse-
automated the detection of D-arabinitol by using a con- shoe crab, Tachypleus tridentalis. This clotting
ventional COBAS MIRA-S clinical chemistry analyzer mechanism can be activated by two pathways: one by
(Roche Diagnostic Systems, Inc., Montclair, NJ). This bacterial endotoxin and the other by fungal cell wall com-
system simplified the detection method and also allowed ponents. Different activating factors can be removed from
the system to make the reaction specific for either path- non-fungal fevers of unknown origin.
way. Removal of Factor G from the lysate permits acti- A recent study investigated plasma BDG levels as a
vation only by endotoxin whereas removal of factors B means for the selection of surgical patients with Candida
and C permit activation only by BDG. colonization who might benefit from empiric antifungal
A test to directly measure plasma BDG was developed therapy (Takesue et al., 2004). Fluconazole was adminis-
using isolated Factor G, the horseshoe crab coagulation tered to postoperative patients demonstrating Candida
factor that is highly sensitive to activation by BDG (Oba- colonization who had risk factors for candidemia and who
yashi et al., 1992). It is called the Fungitec-G test and is presented with persistent fever despite prolonged antibac-
commercially available (Seikagaku Corporation, Tokyo, terial therapy. Plasma BDG levels were analyzed to deter-
Japan). The BDG detection kit contains reagents consist- mine if a correlation existed between BDG levels and
ing of lyophilized horseshoe crab coagulation Factor G, clinical outcome. Of the 32 patients who were positive for
proclotting enzyme, and the chromogenic substrate, t- BDG, 47% responded to empiric therapy. In multiple
butyloxycarbonyl-Leu-Gly-Arg-p-nitroanilide. This sub- logistic regression analysis, being positive for BDG was a
strate is cleaved by the last step in the proteolytic cascade significant factor predicting response to empiric therapy
and can be detected colorimetrically (Obayashi et al., (Takesue et al., 2004). Therefore, in surgical patients with
1992; 1995). Patient plasma, derived from heparinized Candida colonization, assessment of BDG levels may be
blood, must initially be treated with perchloric acid to pre- useful as a guide for the initiation of empiric therapy for
cipitate interfering factors before application of the kit suspected candidiasis.
reagents (Obayashi et al., 1995). Although the BDG test can not identify which fungus is
Obayashi et al. (1995), in a multi-center study con- specifically causing an infection, results can be obtained
ducted in Japan, used this method to measure the plasma within 2 h. Such rapidity makes it very attractive as a
concentration of BDG at the time of routine blood culture screening test for invasive infection by common as well as
performed for 202 febrile episodes. Forty-one febrile epi- less common fungi, including those for which no other
sodes were attributed to infections by fungal species serological test is available (Yoshida et al., 1997). Auto-
including Candida spp.. An additional 59 episodes were mation of this assay (Tamura et al., 1994) will make it
attributed to Gram-negative or Gram-positive bacterial more attractive for evaluation in prospective studies and
infections or febrile responses due to drug therapy, and for use in the clinical laboratory.
102 were of unknown origin. Normal plasma concentra- Patients undergoing hemodialysis with cellulosic mem-
tions of BDG never exceeded 10 pg/ml and fungal febrile branes, such as cuprammonium rayon, which contain
episodes could be differentiated from non-fungal episodes polysaccharides that are shed into the bloodstream during
using a cut-off value of 20 pg/ml. Therefore, using a 20 dialysis and patients receiving parenteral infusions of
pg/ml cut-off value for positively, 90% of 41 episodes plasma components, such as γ-globulin, which is filtered
associated with culture or autopsy-confirmed fungal through cellulose membranes during manufacture may
infection were positive. One hundred percent of the epi- give false positive results by the BDG test (Obayashi et
sodes associated with non-fungal infection had BDG lev- al., 1992).
els below 20 pg/ml. Of 102 episodes of fever of unknown Another more newly created test, the Glucatell test, is
origin, 25.5% demonstrated elevated BDG levels. If these available for research purposes only in the U.S. (Associ-
episodes of fever of unknown origin were to be taken as ates of Cape Cod, Falmouth, Mass.) and also detects
non-fungal in etiology, the tests had a positive and neg- BDG. In this case, amebocytes collected from the
ative predictive value of 59% and 97%, respectively hemolymph of Limulus polyphemus are washed and
(Obayashi et al., 1995). The highest BDG levels were lysed. The lysate is processed to remove Factor C making
observed in patients with proven deep mycoses or the lysate relatively specific for activation by the fungal
fungemia. On the other hand, it was speculated that the glucan pathway. As in the Fungi-Tec test, BDG in the
patients with fever of unknown origin who had elevated sample activates Factor G, which then activates the pro-
BDG levels may, in fact, have had occult deep mycoses. clotting enzyme. The clotting enzyme cleaves p-nitro-
Therefore, 45 patients with neutropenia and fever who aniline (pNA) from the chromogenic peptide substrate
were unresponsive to antibacterial therapy were analyzed (Boc-Leu-Gly-Arg-pNA). The free pNA is measured at
to determine responsiveness to antifungal therapy. Plasma 405 nm (kinetic assay) or, alternatively, the pNA is dia-
BDG levels were >10 pg/ml in 49% and below 10 pg/ml zotized to form a compound that absorbs at 540-550 nm
in 51% of the 45 patients. The efficacy of intravenous flu- (end-point assay). The Glucatell test detected BDG in the
conazole or miconazole therapy was significantly greater sera of patients with positive blood cultures for Candida
in the high BDG group (81.8%) compared with the low species with a sensitivity and specificity of more than
BDG group (43.5%) as measured by resolution of fever 93% (Saeki et al., 2001). In a study by Odabasi et al.
after 2 weeks of antifungal drug therapy. Therefore BDG (2002), BDG was detected by the Glucatell test in the
levels may be helpful in the discrimination of fungal from serum of 100 leukemic patients undergoing chemotherapy
for invasive fungal infections, including candidiasis. Cir- chemicals such as phenol and/or chloroform before pre-
culating BDG levels were found to correlate with invasive cipitation of target DNA with ethanol (Buchman et al.,
fungal infection in these patients. Additional large scale 1990; Fujita et al., 1995). Such methods were quite effec-
studies are required to fully evaluate the utility of the Glu- tive in recovering sufficiently pure DNA for amplification
catell test for the diagnosis of invasive candidiasis and to and detection. For example, Fujita et al. (1995) could
determine its susceptibility to non-specific activation by detect as few as 2 C. albicans cells per 200 µl of spiked
environmental contaminants. whole blood. Einsele et al. (1997) added an incubation
with 50 mM NaOH at 95oC before zymolyase treatment
Molecular biological identification using clinical material and subsequently did not require the use of phenol-chlo-
Identification of Candida spp. by culture requires the roform extraction of DNA before PCR amplification.
presence of viable organisms in blood or body fluids. In Using this method, a detection limit of 1 to 10 fg of fungal
addition, several days may be required for blood cultures DNA (1 cfu per ml of blood) with a sensitivity of 100%
to become positive and, for non-albicans spp. of Candida, was reported for patients with invasive infections. Positive
additional subculturing is required to obtain pure cultures PCR preceded radiological signs by a median of 4 days
for use in subsequent phenotypic identification systems. for 12 of 17 patients with hepatosplenic candidiasis. Fla-
Molecular methods have been applied directly to clinical haut et al. (1998) further simplified the isolation and puri-
materials or the contents of blood culture bottles to cir- fication of C. albicans DNA from whole blood by
cumvent the time required for culture and subculture and modifying the use of a commercial kit (QIAmp Tissue
to increase the sensitivity of detection. Many of the meth- Kit, Qiagen AG, Basel, Switzerland) for purifying DNA
ods described for the molecular identification of Candida after the lysis of erythrocytes. Proteinase K and lysis buff-
species in pure culture have also been applied to the direct ers from the kit were applied and DNA was purified using
detection of organisms from clinical materials but with the silica-based spin columns. The limit of sensitivity from
required addition of more sophisticated DNA extraction clinical specimens was 20 cfu/ml after amplification using
methods to remove PCR inhibitors often present in blood a single copy SAP gene target. Colorimetric detection in
culture media or in host tissues or body fluids. Clinical an ELISA format was as sensitive as Southern blotting
materials such as whole blood, blood culture bottles, and detected 31 of 31 positive blood cultures (100% sen-
spiked whole blood, sera, urine, pus, and tissue samples, sitive compared to culture) after DNA purification and
to name a few, have been used for the molecular identi- amplification by this method.
fication of infecting Candida species. Going one step further, Shin et al. (1997) isolated Can-
In addition to reducing the time required for Candida dida species DNA directly from blood culture bottles that
species identification, the detection of fungal DNA had become positive for growth in a BacT/ALERT blood
directly from clinical materials offers several other advan- culturing system. A natural amplification of target DNA
tages compared to conventional methods. First, fungal occurred as yeasts were allowed to grow in blood culture
DNA can be amplified a million-fold or more using PCR bottles so that a very rapid mechanical disruption method,
methodology. In addition, DNA derived from dead as which did not require expensive enzymes or phenol-chlo-
well as from living fungal cells can be amplified in this roform extraction, could be used to obtain Candida spp.
system. Therefore, for both of these reasons, a PCR-based DNA. Universal fungal PCR primers were used to
assay should be more sensitive than culture. Not only amplify the DNA and species-specific oligonucleotide
should such amplification be more sensitive than culture, probes were then used to identify and differentiate C. albi-
it should be more rapid than culture or even antibody cans, C. krusei, C. parapsilosis, C. tropicalis and C. gla-
detection assays (i.e., requiring only a few hours com- brata in less than one day in a PCR-EIA format. The limit
pared to days for conventional culture or weeks for anti- of sensitivity for this system was 500 cfu per ml which
body production and detection). Use of specific was sufficient to detect 100% of the positive blood cul-
oligonucleotide probes can provide identification of an tures.
organism to the species level and can theoretically detect Iwen et al. (2004) used a similar approach to compare
mixed fungal infections accurately. a PCR assay with the capacity of the ESP blood culture
The earliest reports describing PCR amplification of bottle system (ESP system; TREK Diagnostic Systems,
Candida spp. DNA derived from blood or clinical spec- Inc., USA) to detect fungi including C. albicans, C.
imens used elaborate multi-step processes for DNA iso- parapsilosis, C. glabrata, C. krusei, C. tropicalis, and C.
lation and purification including lysis of host erythrocytes lusitaniae in positive blood culture bottles. The PCR
and leukocytes with salts, detergents and/or proteinase K, assay was performed using fungus-specific primers that
or pre-treatment with DNAses to remove host DNA, amplified the complete ITS1 and ITS2 regions of the
spheroplast production with zymolyase, RNase digestion, rDNA complex (ITS-PCR). Prior to amplification, the
precipitation of proteins with sodium or potassium ace- extracted DNA was diluted to prevent inhibitors in the
tate, and multiple purification steps using hazardous blood-broth samples from interfering with the PCR assay.
ITS-PCR was negative for all 183 samples that were neg- a second reaction, also using two probe sets. The assay
ative by culture while it was positive for all 27 samples was validated with DNA extracted from 62 BACTEC
from blood culture-positive bottles giving a sensitivity and blood culture bottles positive for yeasts and was 100%
specificity of 100% compared to culture. Although the concordant with culture identification results based on bio-
sensitivity limit was no higher than that for culture in chemical and morphological features.
blood bottles alone, the time to specific species identifi- Amplification of RNA instead of DNA has also been
cation was significantly reduced. evaluated for the identification of Candida spp. from clin-
Newer real time PCR assays such as TaqMan (Perkin ical specimens. Nucleic Acid Sequence Based Amplifica-
Elmer) and the Light Cycler (Roche Molecular Systems) tion (NASBA), a non-PCR-based amplification method,
require no post-amplification manipulations and can was evaluated for detection of RNA from 6 different Can-
potentially be automated for all steps from DNA extrac- dida spp. (C. krusei, C. glabrata, C. inconspicua, C.
tion to final PCR amplicon detection and quantitation. dubliniensis, C. norvegensis, and C. lusitaniae) and com-
The TaqMan system was used by Shin et al. (1999) to pared to a PCR assay which targeted the 18S rDNA
identify Candida species from positive blood culture bot- region (Loeffler et al., 2003). Eleven patients were
tles. Universal fungal primers were used to amplify the screened weekly after allogeneic stem cell transplantation
ITS2 rDNA from all Candida species and then species- for the presence of Candida spp. nucleic acid by both
specific probes to this region, labeled with one of three assays. All 5 patients culture positive for Candida species
fluorescent reporter dyes, were used to differentiate gave positive results by both methods. The NASBA assay
among Candida species. Each dye emitted a characteristic detected 1 CFU using a 100 µl sample compared to 5 mL
wavelength allowing up to three Candida species to be for PCR and, in contrast to PCR, NASBA is an isothermal
detected in a single reaction tube. Probes correctly amplification process which does not require thermal
detected and identified 95.1% of 61 Candida species cycling for nucleic acid amplification. Therefore, the use
recovered from blood culture bottles, including those cul- of NASBA may have some advantages compared to PCR.
ture positive for C. albicans, C. parapsilosis, C. glabrata, Given the limitations of some of the conventional diag-
C. krusei, and C. tropicalis. No false positive results were nostic methods for invasive candidiasis, DNA and RNA
obtained from bottles with no growth or from patients based methods hold promise for improved sensitivity and
with bacteremias. This assay detected and identified Can- specificity. However, most molecular tests to date have
dida to the species level in less than one day and did not not shown improved sensitivity compared to blood cul-
require post-PCR hybridization and incubation steps. It turing methods; the most valuable contribution made by
was specific for the detection and identification of Can- molecular methods has been the significantly more rapid
dida species from blood culture bottles, including those identification of Candida spp.. In addition, molecular
containing mixtures of Candida species (C. albicans and detection methods lack standardized and commercially
C. glabrata) that were overlooked by conventional culture available DNA processing systems. Many different DNA
methods due to overgrowth of culture plates with coagu- extraction kits such as the QIAmp Tissue kit, GeneRe-
lase-negative Staphylococcus. leaser, Puregene, Dynabeads DNA, DNAzol and Mag-
Another “real time” PCR system is the LightCycler NAPure are now commercially available which may help
(Roche Molecular Systems) which allows rapid amplifi- to resolve these issues. However, in all cases, a lysis step
cation of DNA in glass capillaries and simultaneous fluo- to release fungal DNA is required before DNA purifica-
rescent detection of amplicons using fluorescence energy tion and amplification can occur. Methods concerning the
transfer or “FRET” (Loeffler et al., 2000b; Schmidt et al., selection of the target region to be amplified, subsequent
2001). Addition of a commercial DNA extraction system primer design, post-PCR detection methods to identify the
to the Light Cycler detection format (MagNA Pure, Roche amplified product, as well as the PCR conditions and for-
Molecular Biochemicals, Indianapolis, Ind.), applied after mat used (single step, multiplex, or nested PCR), all need
lysis and removal of erythrocytes and disruption of C. albi- to be standardized. Contamination can occur during PCR
cans cells with glass beads, allowed a sensitivity limit of 1 assays and precautions to minimize false-positive results
CFU per ml of whole blood (Schmidt et al., 2001). A (Kwok and Higuichi, 1989) also need to be considered
recent study by Selvarangan et al. (2003) reported the use before any system can be commercialized and used in a
of the Light Cycler format to identify DNA from six Can- clinical laboratory.
dida species. Target sequences in the ITS2 regions of the
rRNA gene were simultaneously amplified and labeled References
with fluorescent dyes to identify C. albicans, C. glabrata,
C. parapsilosis, C. tropicalis, C. krusei, and C. lusitaniae. Abi-Said, D., E. Anaissie, O. Uzun, I. Raad, H. Pinzcowski, and S.
The first four species were identified in a single reaction in Vartivarian. 1997. The epidemiology of hematogenous candid-
glass capillaries using two fluorescent hybridization probe iasis caused by different Candida species. Clin Infect Dis. 24,
sets whereas C. krusei and C. lusitaniae were detected in 1122-1128. Erratum in: Clin. Infect. Dis. 25, 352.
Al Mosaid, A., J. Sullivan, and D.C. Coleman. 2003. Differentiation disseminated or peripheral candidiasis. J Clin. Microbiol. 28,
of Candida dubliniensis from Candida albicans on Pal’s agar. 905-909.
J Clin. Microbiol. 41, 4787-4789. Buchman, T.G., M. Rossier, W.G. Merz, and P. Charache. 1990.
Araj, G.F., R.L. Hopfer, S. Chesnut, V. Fainstein, and G.P. Bodey Detection of surgical pathogens by in vitro DNA amplification.
GP. 1982. Diagnostic value of the enzyme-linked immunosor- Part I. Rapid identification of Candida albicans by in vitro
bent assay for detection of Candida albicans cytoplasmic anti- amplification of a fungus-specific gene. Surgery. 108, 338-347.
gen in sera of cancer patients. J Clin. Microbiol. 16, 46-52. Burgener-Kairuz, P., J.P. Zuber, P. Jaunin, T.G. Buchman, J. Bille,
Archibald, L.K., L.C. McDonald, R.M. Addison, C. McKnight, T. and M. Rossier. 1994. Rapid detection and identification of
Byrne, H. Dobbie, O. Nwanyanwu, P. Kazembe, L.B. Reller, Candida albicans and Torulopsis (Candida) glabrata in clinical
and W.R. Jarvis. 2000. Comparison of BACTEC MYCO/F specimens by species-specific nested PCR amplification of a
Lytic and WAMPOLE ISOLATOR 10 (lysis-centrifugation) cytochrome P-450 lanosterol-a-demethylase (L1A1) gene frag-
systems for detection of bacteremia, mycobacteremia, and ment. J Clin. Microbiol. 32, 1902-1907.
fungemia in a developing country. J Clin. Microbiol. 38, 2994- Burnie, J.P. 1985. A reverse passive latex agglutination test for the
2997. diagnosis of systemic candidiasis. J Immunol. Methods. 82,
Bailey, J.W., E. Sada, C. Brass, and J.E. Bennett. 1985. Diagnosis 267-280.
of systemic candidiasis by latex agglutination for serum anti- Campbell, C.K., K.G. Davey, A.D. Holmes, A. Szekely, and D.W.
gen. J Clin. Microbiol. 21, 749-752. Warnock. 1999. Comparison of the API Candida system with
Banerjee, S.N., T.G. Emori, D.H. Culver, R.P. Gaynes, W.R. Jarvis, the AUXACOLOR system for identification of common yeast
T. Horan, J.R. Edwards, J. Tolson, T. Henderson, and W.J. Mar- pathogens. J Clin. Microbiol. 37, 821-823.
tone. 1991. Secular trends in nosocomial primary bloodstream Centers for Disease Control and Prevention. 2000. Monitoring hos-
infections in the United States, 1980-1989. National Nosoco- pital acquired infections to promote patient safety in the United
mial Infections Surveillance System. Am. J Med. 91, 86S-89S. States, 1990-1999. Morb. Mortal Wkly Rep. 49, 149-153.
Bar, W., and H. Hecker. 2002. Diagnosis of systemic Candida Chen, Y.C., J.D. Eisner, M.M. Kattar, S.L. Rassoulian-Barrett, K.
infections in patients of the intensive care unit. Significance of LaFe, S.L. Yarfitz, A.P. Limaye, and B.T. Cookson. 2000. Iden-
serum antigens and antibodies. Mycoses. 45, 22-28. tification of medically important yeasts using PCR-based
Baumgartner, C., A.M. Freydiere, and Y. Gille. 1996. Direct iden- detection of DNA sequence polymorphisms in the internal tran-
tification and recognition of yeast species from clinical material scribed spacer 2 region of the rRNA genes. J Clin. Microbiol.
by using Albicans ID and CHROMagar Candida plates. J Clin. 38, 2302-2310.
Microbiol. 34, 454-456. Christensson, B., G. Sigmundsdottir, and L. Larsson. 1999. D-
Bautista-Munoz, C., X.M. Boldo, L. Villa-Tanaca, and C. Hernan- arabinitol - a marker for invasive candidiasis. Med. Mycol. 37,
dez-Rodriguez. 2003. Identification of Candida spp. by ran- 391-396.
domly amplified polymorphic DNA analysis and Cirak, M.Y., A. Kalkanci, and S. Kustimur. 2003. Use of molecular
differentiation between Candida albicans and Candida dublin- methods in identification of Candida species and evaluation of
iensis by direct PCR methods. J Clin. Microbiol. 41, 414-420. fluconazole resistance. Mem. Inst. Oswaldo. Cruz. 98, 1027-
Berenguer, J., M. Buck, F. Witebsky, F. Stock, P.A. Pizzo, and T.J. 1032.
Walsh. 1993. Lysis-centrifugation blood cultures in the detec- Coignard, C., S.F. Hurst, L.E. Benjamin, M.E. Brandt, D.W. War-
tion of tissue-proven invasive candidiasis. Diagn. Microbiol. nock, and C.J. Morrison. 2004. Resolution of discrepant results
Infect. Dis. 17, 103-109. for Candida species identification by using DNA probes. J
Bernal, S., E. Martin Mazuelos, M. Chavez, J. Coronilla, and A. Clin. Microbiol. 42, 858-861.
Valverde. 1998. Evaluation of the new API Candida system for Crampin, A.C. and R.C. Matthews. 1993. Application of the poly-
identification of the most clinically important yeast species. merase chain reaction to the diagnosis of candidosis by ampli-
Diagn. Microbiol. Infect. Dis. 32, 217-221. fication of an HSP 90 gene fragment. J Med. Microbiol. 39,
Bikandi, J., R.S. Millan, M.D. Moragues, G. Cebas, M. Clarke, D.C. 233-238.
Coleman, D.J. Sullivan, G. Quindos, and J. Ponton. 1998. Creger, R.J., K.E. Weeman, M.R. Jacobs, A. Morrissey, P. Parker,
Rapid identification of Candida dubliniensis by indirect immu- R.M. Fox, and H.M. Lazarus. 1998. Lack of utility of the lysis-
nofluorescence based on differential localization of antigens on centrifugation blood culture method for detection of fungemia
C. dubliniensis blastospores and Candida albicans germ tubes. in immunocompromised cancer patients. J Clin. Microbiol. 36,
J Clin. Microbiol. 36, 2428-2433. 290-293.
Borst, A., B. Theelen, E. Reinders, T. Boekhout, A.C. Fluit, and Davey, K.G., P.M. Chant, C.S. Downer, C.K. Campbell, and D.W.
P.H. Savelkoul. 2003. Use of amplified fragment length poly- Warnock. 1995. Evaluation of the AUXACOLOR system, a
morphism analysis to identify medically important Candida new method of clinical yeast identification. J Clin. Pathol. 48,
spp., including C. dubliniensis. J Clin. Microbiol. 41, 1357- 807-809.
1362. De Caux, B.S., D. Thomas, P.J. Stephens, and C. Booth. 2004. Clin-
Boucher, H., S. Mercure, S. Montplaisir, and G. Lemay. 1996. A ical evaluation of a new selective and differential medium for
novel group I intron in Candida dubliniensis is homologous to the rapid isolation and presumptive identification of Candida
a Candida albicans intron. Gene. 180, 189-196. species. Abstr. C-104, p. 141. Abstr. 104th Annu. Meet. Am.
Bougnoux, M.E., C. Hill, D. Moissenet, M. Feuilhade de Chauvin, Soc. Microbiol.
M. Bonnay, I. Vicens-Sprauel, F. Pietri, M. McNeil, L. Kauf- Denning, D.W., C.C. Kibbler, and R.A. Barnes. 2003. British Soci-
man, and J. Dupouy-Camet. 1990. Comparison of antibody, ety for Medical Mycology proposed standards of care for
antigen, and metabolite assays for hospitalized patients with patients with invasive fungal infections. Lancet Infect. Dis. 3,
230-240. immunosorbent assay measurement of fluctuations in antibody

De Repentigny, L., L.D. Marr, J.W. Keller, A.W. Carter, R.J. Kuyk- titer and antigenemia in cancer patients with and without can-
endall, L. Kaufman, and E. Reiss. 1985. Comparison of didiasis. J Clin. Microbiol. 23, 568-575.
enzyme immunoassay and gas-liquid chromatography for the Fujita, S.I., B.A. Lasker, T.J. Lott, E. Reiss, and C.J. Morrison.
rapid diagnosis of invasive candidiasis in cancer patients. J 1995. Microtitration plate enzyme immunoassay to detect
Clin. Microbiol. 21, 972-979. PCR-amplified DNA from Candida species in blood. J Clin.
Doern, G.V., A. Barton, and S. Rao. 1998. Controlled comparative Microbiol. 33, 962-967.
evaluation of BacT/Alert FAN and ESP 80A aerobic media as Fujita, S.I. and T. Hashimoto. 1992. Detection of serum Candida
means for detecting bacteremia and fungemia. J Clin. Micro- antigens by enzyme-linked immunosorbent assay and a latex
bial. 36, 2686-2689. agglutination test with anti-Candida albicans and anti-Candida
Donnelly, S.M., D.J. Sullivan, D.B. Shanley, and D.C. Coleman. krusei antibodies. J Clin. Microbiol. 30, 3131-3137.
1999. Phylogenetic analysis and rapid identification of Can- Fuller, D.D., T.E. Davis Jr, G.A. Denys, and M.K. York. 2001. Eval-
dida dubliniensis based on analysis of ACT1 intron and exon uation of BACTEC MYCO/F Lytic medium for recovery of
sequences. Microbiol. 145, 1871-1882. mycobacteria, fungi, and bacteria from blood. J Clin. Micro-
Edmond, M.B., S.E. Wallace, D.K. McClish, M.A. Pfaller, R.N. biol. 39, 2933-2936.
Jones, and R.P. Wenzel. 1999. Nosocomial bloodstream infec- Gentry, L.O., I.D. Wilkinson, A.S. Lea, and M.F. Price. 1983. Latex
tions in United States hospitals: a three-year analysis. Clin. agglutination test for detection of Candida antigen in patients
Infect. Dis. 29, 239-244. with disseminated disease. Eur J Clin. Microbiol. Infect. Dis. 3,
Edwards Jr, J.E. 1992. Should all patients with candidemia be 122-128.
treated with antifungal agents? Clin. Infect. Dis. 15, 4222-4223. Graf, B., A. Trost, J. Eucker, U.B. Gobel, and T. Adam. 2004.
Einsele, H., H. Hebart, G. Roller, J. Loffler, I. Rothenhofer, C.A. Rapid and simple differentiation of C. dubliniensis from C.
Muller, R.A. Bowden, J. van Burik, D. Engelhard, L. Kanz, and albicans. Diagn. Microbiol. Infect. Dis. 48, 149-151.
U. Schumacher. 1997. Detection and identification of fungal Greenfield, R.A., M.J. Bussey, J.L Stephens, and J.M. Jones. 1983.
pathogens in blood by using molecular probes. J Clin. Micro- Serial enzyme-linked immunosorbent assay for antibody to
biol. 5, 1353-1360. Candida antigens during induction chemotherapy for acute leu-
Elie, C.M., T.J. Lott, E. Reiss, and C.J. Morrison. 1998. Rapid iden- kemia. J Infect. Dis. 148, 275-283.
tification of Candida species using species-specific DNA Greenfield, R.A. and J.M. Jones. 1981. Purification and character-
probes. J Clin. Microbiol. 36, 3260-3265. ization of a major cytoplasmic antigen of Candida albicans.
Ellepola, A.N., S.F. Hurst, C.M. Elie, and C.J. Morrison. 2003. Infect. Immun. 34, 469-477.
Rapid and unequivocal differentiation of Candida dubliniensis Gudlaugsson, O., S. Gillespie, K. Lee, J. Vande Berg, J. Hu, S.
from other Candida species using species-specific DNA Messer, L. Herwaldt, M. Pfaller, and D. Diekema. 2003. Attrib-
probes: comparison with phenotypic identification methods. utable mortality of nosocomial candidemia, revisited. Clin.
Oral. Microbiol. Immunol. 18, 379-388. Infect Dis. 37, 1172-1177.
El Moudni, B., M.H. Rodier, G. Daniault, and J.L. Jacquemin. 1998. Guiver, M., K. Levi, and B.A. Oppenheim. 2001. Rapid identifi-
Improved immunodiagnosis of human candidiasis by an cation of Candida species by TaqMan PCR. J Clin. Pathol. 54,
enzyme-linked immunosorbent assay using a Candida albicans 362-366.
52-kilodalton metallopeptidase. Clin. Diagn. Lab. Immunol. 5, Hajjeh, R.A., A.N. Sofair, L.H. Harrison, G.M. Lyon, B.A. Arthing-
823-825. ton-Skaggs, S.A. Mirza, M. Phelan, J. Morgan, W. Lee-Yang,
Espinel-Ingroff, A., L. Stockman, G. Roberts, D. Pincus, J. Pollack, M.A. Ciblak, L.E. Benjamin, L.T. Sanza, S. Huie, S.F. Yeo,
and J. Marler. 1998. Comparison of RapID Yeast Plus system M.E. Brandt, and D.W. Warnock. 2004. Incidence of blood-
with API 20C system for identification of common, new, and stream infections due to Candida species and in vitro suscep-
emerging yeast pathogens. J Clin. Microbiol. 36, 883-886. tibilities of isolates collected from 1998 to 2000 in a
Flahaut, M., D. Sanglard, M. Monod, J. Bille, and M. Rossier. 1998. population-based active surveillance program. J Clin. Micro-
Rapid detection of Candida albicans in clinical samples by biol. 42, 1519-1527.
DNA amplification of common regions from C. albicans- Hall, L., S. Wohlfiel, and G.D. Roberts. 2003. Experience with the
secreted aspartic proteinase genes. J Clin. Microbiol. 36, 395- MicroSeq D2 large-subunit ribosomal DNA sequencing kit for
401. identification of commonly encountered, clinically important
Freydiere, A.M. 1996. Evaluation of CHROMagar Candida plates. yeast species. J Clin. Microbiol. 41, 5099-5102.
J Clin. Microbiol. 34, 2048. Haynes, K.A., T.J. Westerneng, J.W. Fell, and W. Moens. 1995.
Freydiere, A.M., L. Buchaille, and Y. Gille. 1997a. Comparison of Rapid detection and identification of pathogenic fungi by poly-
three media for direct identification and discrimination of Can- merase chain reaction amplification of large subunit ribosomal
dida species in clinical specimens. Eur. J Clin. Microbiol. DNA. J Med. Vet. Mycol. 33, 319-325.
Infect. Dis. 16, 464-467. Heelan, J.S., E. Sotomayor, K. Coon, and J.B. D’Arezzo. 1998.
Freydiere, A.M., L. Buchaille, R. Guinet, and Y. Gille. 1997b. Eval- Comparison of the rapid yeast plus panel with the API20C
uation of latex reagents for rapid identification of Candida albi- yeast system for the identification of clinically significant iso-
cans and C. krusei colonies. J Clin. Microbiol. 35, 877-880. lates of Candida species. J Clin. Microbiol. 36, 1443-1445.
Freydiere, A.M., R. Guinet, and P. Boiron. 2001. Yeast identifica- Holmes, A.R., R.D. Cannon, M.G. Shepherd, and H.F. Jenkinson.
tion in the clinical microbiology laboratory: phenotypical meth- 1994. Detection of Candida albicans and other yeasts in blood
ods. Med. Mycol. 39, 9-33. by PCR. J Clin. Microbiol. 32, 228-231.
Fujita, S., F. Matsubara, and T. Matsuda. 1986. Enzyme-linked Hopfer, R.L., P. Walden, S. Setterquist, and W.E. Highsmith. 1993.
Detection and differentiation of fungi in clinical specimens 1069-1071.

using polymerase chain reaction (PCR) amplification and Kurtzman, C.P. and C.J. Robnett. 1997. Identification of clinically
restriction enzyme analysis. J Med. Vet. Mycol. 31, 65-75. important ascomycetous yeasts based on nucleotide divergence
Hoppe, J.E. and P. Frey. 1999. Evaluation of six commercial tests in the 5' end of the large-subunit (26S) ribosomal DNA gene.
and the germ tube test for presumptive identification of Can- J Clin. Microbiol. 35, 1216-1223.
dida albicans. Eur. J Clin. Microbiol. Infect. Dis. 18, 188-191. Kurzai, O., W.J. Heinz, D.J. Sullivan, D.C. Coleman, M. Frosch,
Horvath, L.L., B.J. George, C.K. Murray, L.S. Harrison, and D.R. and F.A. Muhlschlegel. 1999. Rapid PCR test for discriminat-
Hospenthal. 2004. Direct comparison of the BACTEC 9240 ing between Candida albicans and Candida dubliniensis iso-
and BacT/ALERT 3D automated blood culture systems for lates using primers derived from the pH-regulated PHR1 and
Candida growth detection. J Clin. Microbiol. 42, 115-118. PHR2 genes of C. albicans. J Clin. Microbiol. 37, 1587-1590.
Hsu, M.C., K.W. Chen, H.J. Lo, Y.C. Chen, M.H. Liao, Y.H. Lin, Kwok, S. and R. Higuchi. 1989. Avoiding false positives with PCR.
and S.Y. Li. 2003. Species identification of medically important Nature 339, 237-238.
fungi by use of real-time LightCycler PCR. J Med. Microbiol. Kwon-Chung, K.J., D. Lehman, C. Good, and P.T. McGee. 1985.
52, 1071-1076. Genetic evidence for role of extracellular proteinase in viru-
Hube, B. 1996. Candida albicans secreted aspartyl proteinases. lence of Candida albicans. Infect. Immun. 49, 571-575.
Curr. Top. Med. Mycol. 7, 55-69. Land, G.A., I.F. Salkin, M. El-Zaatari, M.R. McGinnis, and G.
Hui, M., M. Ip, P.K. Chan, M.L. Chin, and A.F. Cheng. 2000. Rapid Hashem. 1991. Evaluation of the Baxter MicroScan 4-hour
identification of medically important Candida to species level enzyme-based yeast identification system. J Clin. Microbiol.
by polymerase chain reaction and single-strand conformational 29, 718-722.
polymorphism. Diagn. Microbiol. Infect. Dis. 38, 95-99. Lecciones, J.A., J.W. Lee, E.E. Navarro, F.G. Witebsky, D. Mar-
Irobi, J., A. Schoofs, and H. Goossens. 1999. Genetic identification shall, S.M. Steinberg, P.A. Pizzo, and T.J. Walsh. 1992. Vas-
of Candida species in HIV-positive patients using the poly- cular catheter-associated fungemia in patients with cancer:
merase chain reaction and restriction fragment length polymor- analysis of 155 episodes. Clin. Infect. Dis. 14, 875-883.
phism analysis of its DNA. Mol. Cell. Probes. 13, 401-406. Lemieux, C., G. St-Germain, J. Vincelette, L. Kaufman, and L. de
Iwasaki, H., H. Misaki, T. Nakamura, and T. Ueda. 2000. Surveil- Repentigny. 1990. Collaborative evaluation of antigen detec-
lance of the serum Candida antigen titer for initiation of anti- tion by a commercial latex agglutination test and enzyme
fungal therapy after post-remission chemotherapy in patients immunoassay in the diagnosis of invasive candidiasis. J Clin.
with acute leukemia. Intl. J Hematol. 71, 266-272. Microbiol. 28, 249-253.
Iwen, P.C., A.G. Freifeld, T.A. Bruening, and S.H. Hinrichs. 2004. Lischewski, A., M. Kretschman, H. Hof, R. Amann, J. Hacker, and
Use of a panfungal PCR assay for detection of fungal patho- J. Morschhauser. 1997. Detection and identification of Candida
gens in a commercial blood culture system. J Clin. Microbiol. species in experimentally infected tissue and human blood by
42, 2292-2293. rRNA-specific fluorescent in situ hybridization. J Clin. Micro-
Jabra-Rizk, M.A., T.M. Brenner, M. Romagnoli, A.A. Baqui, W.G. biol. 35, 2943-2948.
Merz, W.A. Falkler Jr, and T.F. Meiller. 2001. Evaluation of a Loeffler, J., C. Dorn, H. Hebart, P. Cox, S. Magga, and H. Einsele.
reformulated CHROMagar Candida. J Clin. Microbiol. 39, 2003. Development and evaluation of the nuclisens basic kit
2015-2016. NASBA for the detection of RNA from Candida species fre-
Jaeger, E.E., N.M. Carroll, S. Choudhury, A.A. Dunlop, H.M. quently resistant to antifungal drugs. Diagn. Microbiol. Infect.
Towler, M.M. Matheson, P. Adamson, N. Okhravi, and S. Dis. 45, 217-220.
Lightman. 2000. Rapid detection and identification of Candida, Loeffler, J., H. Hebart, S. Magga, D. Schmidt, L. Klingspor, J.
Aspergillus, and Fusarium species in ocular samples using Tollemar, U. Schumacher, and H. Einsele. 2000a. Identification
nested PCR. J Clin. Microbiol. 38, 2902-2908. of rare Candida species and other yeasts by polymerase chain
Jones, J.M. 1980a. Kinetics of antibody responses to cell wall man- reaction and slot blot hybridization. Diagn. Microbiol. Infect.
nan and a major cytoplasmic antigen of Candida albicans in Dis. 38, 207-212.
rabbits and humans. J Lab. Clin. Med. 96, 845-860. Loeffler, J., H. Hebart, S. Sepe, U. Schumacher, T. Klingbiel, and
Jones, J.M. 1980b. Quantitation of antibody against cell wall man- J. Einsele. 1998. Detection of PCR-amplified fungal DNA by
nan and a major cytoplasmic antigen of Candida in rabbits, using a PCR-ELISA system. Med. Mycol. 36, 275-279.
mice, and humans. Infect. Immun. 30, 78-89. Loeffler, J., N. Henke, H. Hebart, D. Schmidt, L Hagmeyer, U.
Jordan, J. 1994. PCR identification of four medically important Schumacher, and H. Einsele H. 2000b. Quantification of fungal
Candida species by using a single primer pair. J Clin. Micro- DNA by using fluorescence resonance energy transfer and the
biol. 32, 2962-2967. light cycler system. J Clin. Microbiol. 38, 586-590.
Kahn, F.W. and J.M. Jones. 1986. Latex agglutination tests for MacDonald, F. and F.C. Odds. 1983. Virulence for mice of a pro-
detection of Candida antigens in sera of patients with invasive teinase-secreting strain of Candida albicans and a proteinase-
candidiasis. J Infect. Dis. 153, 579-585. deficient mutant. J Gen. Microbiol. 129, 431-438.
Kaufman, L., J.A. Kovacs, and E. Reiss. 1997. Clinical Immuno- Mannarelli, B.M. and C.P. Kurtzman. 1998. Rapid identification of
mycology, p. 585-604. In N.R. Rose, E.C. de Macario, J.D. Candida albicans and other human pathogenic yeasts by using
Folds, H.C. Lane, and R.M. Nakamura (eds.), Manual of Clin- short oligonucleotides in a PCR. J Clin. Microbiol. 36, 1634-
ical Laboratory Immunology. ASM Press, Washington, D.C. 1641.
Kitch, T.T., M.R. Jacobs, M.R. McGinnis, and P.C. Appelbaum. Marcilla, A., C. Monteagudo, S. Mormeneo, and R. Sentandreu.
1996. Ability of RapID Yeast Plus System to identify 304 clin- 1999. Monoclonal antibody 3H8: a useful tool in the diagnosis
ically significant yeasts within 5 hours. J Clin. Microbiol. 34, of candidiasis. Microbiol. 145, 695-701.
Martin. C., D. Roberts, M. van Der Weide, R. Rossau, G. Jannes, T. Nguyen, D.R. Snydman, M.M. Wagener, M.G. Rinaldi, and
Smith, and M. Maher. 2000. Development of a PCR-based line V.L. Yu. 1996. The changing face of candidemia: emergence of
probe assay for identification of fungal pathogens. J Clin. non-Candida albicans species and antifungal resistance. Am J
Microbiol. 38, 3735-3742. Med. 100, 617-623.
Mason, A.B., M.E. Brandt, and H.R. Buckley. 1989. Enolase activ- Obayashi, T., M. Yoshida, T. Mori, H. Goto, A. Yasuoka, H.
ity associated with a Candida albicans cytoplasmic antigen. Iwasaki, H. Teshima, S. Kohno, A. Horiuchi, and A. Ito. 1995.
Yeast 5, S231-239. Plasma (1→3)-beta-D-glucan measurement in diagnosis of
Massonet, C., J. Van Eldere, M. Vaneechoutte, T. De Baere, J. Ver- invasive deep mycosis and fungal febrile episodes. Lancet. 345,
haegen, and K. Lagrou. 2004. Comparison of VITEK 2 with 17-20.
ITS2-fragment length polymorphism analysis for identification Obayashi, T., M. Yoshida, H. Tamura, J. Aketagawa, S. Tanaka, and
of yeast species. J Clin. Microbiol. 42, 2209-2211. Kawai T. 1992. Determination of plasma (1→3)-β-D-glucan: a
Matthews, R.J. and J.P. Burnie. 1988. Diagnosis of systemic can- new diagnostic aid to deep mycosis. J Med. Vet. Mycol. 30,
didiasis by an enzyme-linked dot immunobinding assay for a 275-280.
circulating immunodominant 47-kilodalton antigen. J Clin. Odabasi, Z., G. Mattiuzzi, L. Ostrosky-Zeichner, E. Estey, H. Kan-
Microbiol. 26, 459-463. tarjian, and J.H. Rex. 2002. Detection of (1→3)-β-D-Glucan in
Matthews, R.J., J.P. Burnie, and S. Tabaqchali. 1987. Isolation of the serum of leukemia patients with invasive fungal infections.
immunodominant antigens from sera of patients with systemic Abstr. M-902, p. 396. Abstr. 42nd ICAAC.
candidiasis and characterization of serological response to Can- Odds, F. and R. Bernaerts. 1994. CHROMagar Candida, a new dif-
dida albicans. J Clin. Microbiol. 25, 230-237. ferential medium for presumptive identification of clinically
McDonald, L.C., M.P. Weinstein, J. Fune, S. Mirrett, L.G. Reimer, important Candida species. J Clin. Microbiol. 12, 1923-1929.
and L.B. Reller. 2001. Controlled comparison of BacT/ALERT Oliveira, K., G. Haase, C. Kurtzman, J.J. Hyldig-Nielsen, and H.
FAN aerobic medium and BACTEC fungal blood culture Stender. 2001. Differentiation of Candida albicans and Can-
medium for detection of fungemia. J Clin. Microbiol. 39, 622- dida dubliniensis by fluorescent in situ hybridization with pep-
624. tide nucleic acid probes. J Clin. Microbiol. 39, 4138-4141.
Misaki, H., H. Iwasaki, and T. Ueda. 2003. A comparison of the Pallavicini, F., I. Izzi, M.A. Pennisi, G. Morace, G.G. Portaccio, G.
specificity and sensitivity of two Candida antigen assay sys- Bello, F. Iodice, D. Godino, C. Del Borgo, and R. Proietti.
tems for the diagnosis of deep candidiasis in patients with 1999. Evaluation of the utility of serological tests in the diag-
hematologic diseases. Med. Sci. Monit. 9, MT1-7. nosis of candidemia. Minerva Anestesiologica. 65, 637-639.
Miyakawa, Y., T. Mabuchi, and Y. Fukazawa. 1993. New method Pappas, P.G., J.H. Rex, J.D. Sobel, S.G. Filler, W.E. Dismukes, T.J.
for detection of Candida albicans in human blood by poly- Walsh, and J.E. Edwards. 2004. Guidelines for treatment of
merase chain reaction. J Clin. Microbiol. 31, 3344-3347. candidiasis. Clin. Infect. Dis. 15, 161-189.
Moghaddas, J., A.L. Truant, C. Jordan, and H.R. Buckley. 1999. Patterson, T.F., S.G. Revankar, W.R. Kirkpatrick, R.K. McAtee,
Evaluation of the RapID Yeast Plus System for the identifica- A.W. Fothergill, D.I. McCarthy, and M.G. Rinaldi. 1996. Sim-
tion of yeast. Diagn. Microbiol. Infect. Dis. 35, 271-273. ple method for detecting fluconazole-resistant yeasts with chro-
Morace, G., M. Sanguinetti, B. Posteraro, G. LoCascio, and G. mogenic agar. J Clin. Microbiol. 34, 1794-1797.
Fadda. 1997. Identification of various medically important Petti, C.A., A.K. Zaidi, S. Mirrett, and L.B. Reller. 1996. Compar-
Candida species in clinical specimens by PCR-restriction ison of Isolator 1.5 and BACTEC NR660 aerobic 6A blood
enzyme analysis. J Clin. Microbiol. 35, 667-672. culture systems for the detection of fungemia in children. J
Moran, G.P., D. Sanglard, S.M. Donnelly, D.B. Shanley, D.J. Sul- Clin. Microbiol. 34, 1877-1879.
livan, and D.C. Coleman. 1998. Identification and expression Pfaller, M.A., A. Houston, and S. Coffmann. 1996. Application of
of multidrug transporters responsible for fluconazole resistance CHROMagar Candida for rapid screening of clinical specimens
in Candida dubliniensis. Antimicrob. Agents Chemother. 42, for Candia albicans, Candida tropicalis, Candida krusei, and
1819-1830. Candida (Torulopsis) glabrata. J Clin. Microbiol. 34, 58-61.
Moran, G.P., D.J. Sullivan, M.C. Henman, C.E. McCreary, B.J. Har- Pfaller, M.A., S.A. Messer, S. Gee, S. Joly, C. Pujol, D.J. Sullivan,
rington, D.B. Shanley, and D.C. Coleman. 1997. Antifungal D.C. Coleman, and D.R. Soll. 1999. In vitro susceptibilities of
drug susceptibilities of oral Candida dubliniensis isolates from Candida dubliniensis isolates tested against the new triazole
human immunodeficiency virus (HIV)-infected and non-HIV- and echinocandin antifungal agents. J Clin. Microbiol. 37, 870-
infected subjects and generation of stable fluconazole-resistant 872.
derivatives in vitro. Antimicrob. Agents Chemother. 41, 617- Phillips, P., A. Dowd, P. Jewesson, G. Radigan, M.G. Tweeddale, A.
623. Clarke, I. Geere, and M. Kelly. 1990. Nonvalue of antigen
Morrison, C.J., S.F. Hurst, and E. Reiss. 2003. Competitive binding detection immunoassays for diagnosis of candidemia. J Clin.
inhibition enzyme-linked immunosorbent assay that uses the Microbiol. 28, 2320-2326.
secreted aspartyl proteinase of Candida albicans as an anti- Posteraro, B., M. Sanguinetti, L. Masucci, L. Romano, G. Morace,
genic marker for diagnosis of disseminated candidiasis. Clin. and G. Fadda. 2000. Reverse cross blot hybridization assay for
Diagn. Lab. Immunol. 10, 835-848. rapid detection of PCR-amplified DNA from Candida speceis,
Na, B.K. and C.Y. Song. 1999. Use of monoclonal antibody in diag- Cryptococcus neoformans, and Saccharomyces cerevisiae in
nosis of candidiasis caused by Candida albicans: detection of clinical samples. J Clin. Microbiol. 38, 1609-1614.
circulating aspartyl proteinase antigen. Clin. Diagn. Lab. Powell, H.L., C.A. Sand, and R.P. Rennie. 1998. Evaluation of
Immunol. 6, 924-929. CHROMagar Candida for presumptive identification of clini-
Nguyen, M.H, J.E. Peacock Jr, A.J. Morris, D.C. Tanner, M.L. cally important Candida species. Diagn. Microbiol. Infect Dis.
32, 201-204. teinase by enzyme immunoassay and interaction of the enzyme

Price, M.F, M.T. LaRocco, and L.O. Gentry. 1994. Fluconazole sus- with alpha-2-macroglobulin. J Immunol. Methods. 61, 107-116.
ceptibilities of Candida species and distribution of species Ruchel, R., B. Boning-Stutzer, and A. Mari. 1988. A synoptical
recovered from blood cultures over a 5-year period. Antimi- approach to the diagnosis of candidosis, relying on serological
crob. Agents Chemother. 38, 1422-1424. antigen and antibody tests, on culture, and on evaluation of
Procop, G.W., F.R. Cockerill 3rd, E.A. Vetter, W.S. Harmsen, J.G. clinical data. Mycoses. 31, 87-106.
Hughes, and G.D. Roberts. 2000. Performance of five agar Saeki, F., R.J. Ridge, M.A. Finkelman, P.A. Ketchum, and J.H. Rex.
media for recovery of fungi from isolator blood cultures. J 2001. Application of a glucan-specific limulus amebocyte
Clin. Microbiol. 38, 3827-3829. lysate serum assay for diagnosis of invasive fungal infections.
Pryce, T.M., I.D. Kay, S. Palladino, and C.H. Heath. 2003. Real- Abstr. J-841, p. 386. Abstr. 41st ICAAC.
time automated polymerase chain reaction (PCR) to detect Sandhu, G., B.C. Kline, L. Stockman, and G.D. Roberts. 1995.
Candida albicans and Aspergillus fumigatus DNA in whole Molecular probes for diagnosis of fungal infections. J Clin.
blood from high-risk patients. Diagn. Microbiol. Infect Dis. 47, Microbiol. 33, 2913-2919.
487-496. San-Millan, R., L. Ribacoba, J. Ponton, and G. Quindos. 1996.
Quindos, G., R. San Millan, R. Robert, C. Bernard, and J. Poton. Evaluation of a commercial medium for identification of Can-
1997. Evaluation of Bichro-latex albicans, a new method for dida species. Eur. J Clin. Microbiol. Infect Dis. 15, 153-158.
rapid identification of Candida albicans. J Clin. Microbiol. 35, Schmidt, D., J. Loeffler, H. Hebart, U. Schumacher, and H. Einsele.
1263-1265. 2001. Rapid extraction and detection of DNA from Candida
Quindos, G., V. Lipperheide, and J. Ponton. 1993. Evaluation of two species in research samples by MagNA Pure LC and the light
commercialized systems for the rapid identification of medi- cycler. Biochemica. 3, 1-3.
cally important yeasts. Mycoses. 36, 299-303. Selvarangan, R., U. Bui, A.P. Limaye, and B.T. Cookson. 2003.
Rabkin, J.M., S.L. Oroloff, C.L. Corless, K.G. Benner, K.D. Flora, Rapid identification of commonly encountered Candida spe-
H.R. Rosen, and A.J. Olyaei. 2000. Association of fungal cies directly from blood culture bottles. J Clin. Microbiol. 41,
infection and increased mortality in liver transplant recipients. 5660-5664.
Am J Surg. 179, 426-430. Selvarangan, R., A.P. Limaye, and B.T. Cookson. 2002. Rapid iden-
Rand, K.H., H. Houck, and M. Wolff. 1994. Detection of candi- tification and differentiation of Candida albicans and Candida
demia by polymerase chain reaction. Mol. Cell. Probes. 8, 215- dubliniensis by capillary-based amplification and fluorescent
222. probe hybridization. J Clin. Microbiol. 40, 4308-4312.
Ray, T.L. and C.D. Payne. 1988. Scanning electron microscopy of Sendid, B., T. Jouault, R. Coudriau, D. Camus, F. Odds, M.
epidermal adherence and cavitation in murine candidiasis: a Tabouret, and D. Poulain. 2004. Increased sensitivity of man-
role for Candida acid proteinase. Infect. Immun. 56, 1942- nanemia detection tests by joint detection of α- and β-linked
1949. oligomannosides during experimental and human systemic
Ray, T.L., C.D. Payne, and B.J. Morrow. 1991. Candida acid pro- candidiasis. J Clin. Microbiol. 42, 164-171.
teinases: characterization and role in candidiasis. Adv. Exp. Sendid, B., M. Tabouret, J.L. Poirot, D. Mathieu, J. Fruit, and D.
Med. Biol. 306, 173-183. Poulain. 1999. New enzyme immunoassays for sensitive detec-
Rees, J.R., R.W. Pinner, R.A. Hajjeh, M.E. Brandt, and A.L. Rein- tion of circulating Candida albicans mannan and antimannan
gold. 1998. The epidemiological features of invasive mycotic antibodies: useful combined test for diagnosis of systemic can-
infections in the San Francisco Bay area, 1992-1993: results of didiasis. J Clin. Microbiol. 37, 1510-1517.
population-based laboratory active surveillance. Clin. Infect. Shin, J.H., F.S. Nolte, and C.J. Morrison. 1997. Rapid identification
Dis. 27, 1138-1147. of Candida species in blood cultures by a clinically useful PCR
Reimer, L.G., M.L. Wilson, and M.P. Weinstein. 1997. Update on method. J Clin. Microbiol. 35, 1454-1459.
detection of bacteremia and fungemia. Clin. Microbiol. Rev. 10, Shin, J.H., F.S. Nolte, and C.J. Morrison. 1999. Rapid identification
444-465. of up to three Candida species in a single reaction tube by a 5'
Reiss, E. and C.J. Morrison. 1993. Nonculture methods for the diag- exonuclease assay using fluorescent DNA probes. J Clin.
nosis of Candida infections. Clin. Microbiol. Rev. 6, 311-323. Microbiol. 37, 165-170.
Rex JH, Bennett JE, Sugar AM, P.G. Pappas, C.M. van der Horst, Staib, F. 1965. Serum-proteins as nitrogen source for yeast-like
J.E. Edwards, R.G. Washburn, W.M. Scheld, A.W. Karchmer, fungi. Sabouraudia. 4, 187-193.
and A.P. Dine. 1994. A randomized trial comparing flucona- Stevens, P., S. Huang, L.S. Young, and M. Berdischewsky. 1980.
zole with amphotericin B for the treatment of candidemia in Detection of Candida antigenemia in human invasive candid-
patients without neutropenia. N Eng. J Med. 331, 1325-1330. iasis by a new solid phase radioimmunoassay. Infection. 8,
Rigby, S., G.W. Procop, G. Haase, D. Wilson, G. Hall, C. Kurtzman, S334-338.
K. Oliveira, S. Von Oy, J.J. Hyldig-Nielsen, J. Coull, and H. St-Germain, G. and D. Beauchesne. 1991. Evaluation of the
Stender. 2002. Fluorescence in situ hybridization with peptide MicroScan rapid yeast identification panel. J Clin. Microbiol.
nucleic acid probes for rapid identification of Candida albicans 29, 2296-2299.
directly from blood culture bottles. J Clin. Microbiol. 40, 2182- Strockbine, N.A., M.T. Largen, and H.R. Buckley. 1984a. Produc-
2186. tion and characterization of three monoclonal antibodies to
Ruchel, R. 1992. Proteinase, p. 17-33. In J.E. Bennett, R.J. Hay, Candida albicans proteins. Infect. Immun. 43, 1012-1018.
and P.K. Peterson (eds.), New Strategies in Fungal Disease. Strockbine, N.A., M.T. Largen, S.M. Zweibel, and H.R. Buckley.
Churchill-Livingstone, Edinburgh. 1984b. Identification and molecular weight characterization of
Ruchel, R. and B. Boning-Stutzer. 1983. Detection of Candida pro- antigens from Candida albicans that are recognized by human
sera. Infect. Immun. 43, 715-721. infections: advances in nonculture systems. Curr. Clin. Top
Sundstrom, P. and G.R. Aliaga. 1992. Molecular cloning of cDNA Infect. Dis. 18, 101-153.
and analysis of protein secondary structure of Candida albi- Walsh, T.J., A. Francesconi, M. Kasai, and S.J. Chanock. 1995a.
cans enolase, an abundant, immunodominant glycolytic PCR and single-strand conformational polymorphism for rec-
enzyme. J Bacteriol. 174, 6789-6799. ognition of medically important opportunistic fungi. J Clin.
Switchenko, A.C., C.G. Miyada, T.C. Goodman, T.J. Walsh, B. Microbiol. 33, 3216-3220.
Wong, M.J. Becker, and E.F. Ullman. 1994. An automated Walsh, T.J., J.W. Hathorn, J.D. Sobel, W.G. Merz, V. Sanchez, S.M.
enzymatic method for measurement of D-arabinitol, a metab- Maret, H.R. Buckley, M.A. Pfaller, R. Schaufele, and C. Sliva.
olite of pathogenic Candida species. J Clin. Microbiol. 32, 92- 1991. Detection of circulating Candida enolase by immunoas-
97. say in patients with cancer and invasive candidiasis. N Eng J
Takesue, Y., M. Kakehashi, H. Ohge, Y. Imamura, Y. Murakami, M. Med. 324, 1026-1031.
Sasaki, M. Morifuji, Y. Yokoyama, M. Kouyama, T. Walsh, T.J., J.W. Lee, T. Sein, R. Schaufele, J. Bacher, A.C. Switch-
Yokoyama, and T. Sueda. 2004. Combined Assessment of β-d- enko, T.C. Goodman, and P.A. Pizzo. 1994. Serum D-arabinitol
Glucan and Degree of Candida Colonization before Starting measured by automated quantitative enzymatic assay for detec-
Empiric Therapy for Candidiasis in Surgical Patients. World J tion and therapeutic monitoring of experimental disseminated
Surg. In press. candidiasis: correlation with tissue concentrations of Candida
Tamura, H., Y. Arimoto, S. Tanaka, M. Yoshida, T. Obayashi, and albicans. J Med. Vet. Mycol. 32, 205-215.
T. Kawai. 1994. Automated kinetic assay for endotoxin and Walsh, T.J., W.G. Merz, J.W. Lee, R. Schaufele, T. Sein, P.O. Whit-
(1→3)-β-D-glucan in human blood. Clin Chim Acta. 226, 109- comb, M. Ruddel, W. Burns, J.R. Wingard, and A.C. Switch-
112. enko. 1995b. Diagnosis and therapeutic monitoring of invasive
Tiraboschi, I.N., J.E. Bennett, C.A. Kaufmann, J.H. Rex, C. Gir- candidiasis by rapid enzymatic detection of serum D-arabinitol.
menia, J.D. Sobel, and F. Menichetti. 2000. Deep Candida Am J Med. 99, 164-172.
infections in the neutropenic and non-neutropenic host: an Washington, J.A. 1994. Collection, transport, and processing of
ISHAM symposium. Med. Mycol. 38, 199-204. blood cultures. Clin Lab. Med. 14, 59-68.
Trick, W.E., S.K. Fridkin, J.R. Edwards, R.A. Hajjeh, and R.P. Wildfeuer, A., R. Schlenk, and W. Friedrich. 1996. Detection of
Gaynes. 2002. Secular trend of hospital-acquired candidemia Candida albicans DNA with a yeast-specific primer system by
among intensive care unit patients in the United States during polymerase chain reaction. Mycoses. 39, 341-346.
1989-1999. Clin. Infect. Dis. 35, 627-630. Willinger, B. and M. Manafi. 1999. Evaluation of CHROMagar
van Deventer, A.J., W.H. Goessens, A. van Belkum, H.J. van Vliet, Candida for rapid screening of clinical specimens for Candida
E.W. van Etten, and H.A. Verbrugh. 1995. Improved detection species. Mycoses. 42, 61-65.
of Candida albicans by PCR in blood of neutropenic mice with Wingard, J.R. 1995. Importance of Candida species other than C.
systemic candidiasis. J Clin. Microbiol. 33, 625-628. albicans as pathogens in oncology patients. Clin. Infect. Dis.
Verduyn Lunel, F.M., A. Voss, E.J. Kuijper, L.B. Gelinck, P.M. 20, 115-125.
Hoogerbrugge, K.L. Liem, B.J. Kullberg, and P.E. Verweij. Wong, B., E.M. Bernard, J.W. Gold, D. Fong, and D. Armstrong.
2004. Detection of the Candida antigen mannan in cerebrospi- 1982. The arabinitol appearance rate in laboratory animals and
nal fluid specimens from patients suspected of having Candida humans: estimation from the arabinitol/creatinine ratio and rel-
meningitis. J Clin. Microbiol. 42, 867-870. evance to the diagnosis of candidiasis. J Infect. Dis. 146, 353-
Verghese, S.L., P. Padmaja, P. Sutha, T. Mathew, and E.S. John. 359.
2000. Rapid identification of fluconazole resistance using Wong, B., K.L. Brauer, J.R. Clemons, and S. Beggs. 1990. Effects
CHROMagar Candida. Ind. J Pathol. Microbiol. 43, 343-346. of gastrointestinal candidiasis, antibiotics, dietary arabinitol,
Verweij, P.E., I.M. Breuker, A.J. Rijs, and J.F. Meis. 1999. Com- and cortisone acetate on levels of the Candida metabolite D-
parative study of seven commercial yeast identification sys- arabinitol in rat serum and urine. Infect. Immun. 58, 283-288.
tems. J Clin. Pathol. 52, 271-273. Yeo, S.F., Y. Zhang, D. Schafer, S. Campbell, and B. Wong. 2000.
Vetter, E., C. Torgerson, A. Feuker, J. Hughes, S. Harmsen, C. A rapid, automated enzymatic fluorometric assay for determi-
Schleck, C. Horstmeier, G. Roberts, and F. Cockerill 3rd. 2001. nation of D-arabinitol in serum. J Clin. Microbiol. 38, 1439-
Comparison of the BACTEC MYCO/F Lytic bottle to the iso- 1443.
lator tube, BACTEC Plus Aerobic F/bottle, and BACTEC Yera, H., B. Sendid, N. Francois, D. Camus, and D. Poulain. 2001.
Anaerobic Lytic/10 bottle and comparison of the BACTEC Contribution of serological tests and blood culture to the early
Plus Aerobic F/bottle to the Isolator tube for recovery of bac- diagnosis of systemic candidiasis. Eur. J Clin. Microbiol.
teria, mycobacteria, and fungi from blood. J Clin. Microbiol. Infect. Dis. 20, 864-870.
39, 4380-4386. Yoshida, M., T. Obayashi, A. Iwama, M. Ito, S. Tsunoda, T. Suzuki,
Wadlin, J.K., G. Hanko, R. Stewart, J. Pape, and I. Nachamkin. K. Muroi, M. Ohta, S. Sakamoto, and Y. Miura. 1997. Detec-
1999. Comparison of three commercial systems for identification of plasma (1→3)-β-D-glucan in patients with Fusarium,
tion of yeasts commonly isolated in the clinical microbiology Trichosporon, Saccharomyces and Acremonium fungaemias. J
laboratory. J Clin. Microbiol. 37, 1967-1970. Med. Vet. Mycol. 35, 371-374.
Walsh, T.J. and S.J. Chanock. 1998. Diagnosis of invasive fungal
View publication stats

Canda

Uploaded by

Copyright:

Available Formats

Canda

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Canda

Uploaded by

Copyright:

Available Formats

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

Laboratory diagnosis of invasive candidiasis

Article in The Journal of Microbiology · March 2005

The user has requested enhancement of the downloaded file.

Laboratory Diagnosis of Invasive Candidiasis

et al., 2004). Second, species-specific identification is rel- sive candidiasis.

230-240. immunosorbent assay measurement of fluctuations in antibody

Detection and differentiation of fungi in clinical specimens 1069-1071.

32, 201-204. teinase by enzyme immunoassay and interaction of the enzyme

View publication stats

You might also like