Vol. 30, No. 12 JOURNALOFCLINICALMICROBIOLOGY, Dec. 1992,p. 3132-3137
0095-1137/92/123132-06$02.00/0
Copyright X 1992, AmericanSocietyforMicrobiology
Detection of Serum Candida Antigens by
Enzyme-Linked
Immunosorbent Assay and
a Latex
Agglutination
Test
with
Anti-Candida albicans and Anti-Candida krusei Antibodies
SHIN-ICHIFUJITA* ANDTAKUMA HASHIMOTO
Central ClinicalLaboratory, Kanazawa UniversityHospital,
13-1,
Takara-machi,
Kanazawa920,
Japan
Received 3 March 1992/Accepted 1 September 1992
Serum samples from 197 patients with and without candidiasis were assayed for Candida albicans mannan andCandida krusei mannan by an enzyme-linked immunosorbent assay (ELISA) and a latex agglutination test (LA) and for D-arabinitol by the enzymatic fluorometric method. Of the 43 patients positive for C. albicans mannan ('0.2 ng/ml), 34 were infected with C.albicans and 9 were infected with Candida tropicalis. C. krusei mannan (_0.3 ng/ml) was detected in 10 patients infected with Candidaparapsilosis,2patients infected with Candidaguilliermondii,and 1patient infected with C.krusei.With both anti-C. albicans antibodies and anti-C. kruseiantibodies, the sensitivities of ELISA and LA for detection of invasive candidiasis (58 patients) were 74 and38%,
respectively.
Nofalse-positive reactions were observed by the ELISA or the LA. Thesensitivityand specificity of the D-arabinitol/creatinine ratio (_1.5 ,umol/mg) to invasive candidiasis were 50 and 91%,respectively.
The ELISA with antibodies against both C. albicans and C. krusei may be useful in diagnosing invasive candidiasis caused bymedically
important Candida strains excludingCandidaglabrata.Invasivecandidiasis remains asignificant complication in
patients with hematologic malignancies, patients receiving total parenteral nutrition, andpatientswith solid tumors in whomabdominal surgery has been performed. The diagnosis ofinvasive candidiasis is made by the frequent isolation of Candida species from normally sterile sites; histological
evidence of Candida invasion; or positive results for cell
wall mannan (3, 9, 17), enolase (16), or D-arabinitol (3). Candida albicansmannan hasbeen detected in the sera of
patientswithC. albicansorCandida tropicalisinfections by
radioimmunoassay (17), enzyme-linked immunosorbent as-say
(ELISA)
(3, 9), and the latex agglutination test(LA) (1, 12). The specificity of the ELISA formannan was 95% ormore,butthe sensitivities variedbetweeninvestigators and
ranged between 22 and 100% (6). On the other hand, Candida
parapsilosis,
Candidaguilliermondii,
Candidakrusei,
andCandidaglabrata have also been isolatedfrom blood cultures(5). Therefore, sensitivity mightbe increased if mannan aswell as other polysaccharide antigens ofCan-didaspeciesweredetected. This report describesanELISA andanLAusedtodiagnose invasive candidiasiscausedby medically important Candida isolates.
Candida metabolites can be detected in the serum of affected patients by the enzymaticfluorometric assay (15), which is now commercially available as a D-arabinitol test kit,andby gas-liquidchromatography(3).Becausethe latter method is morecomplex, aD-arabinitol kitwasused in the present study. These resultswere compared with those of testsfor serumCandida antigens.
MATERIALSANDMETHODS
Yeast isolates. Thirty-six Candida strains isolated from clinical materials and maintained in our laboratory were
used. One reference strain ofC. albicans serotypeB (IFO 10108) was obtained from the Institute for Fermentation,
*
Corresponding
author.Osaka, Japan. Each isolatewasidentifiedby usingCandida Check (latron Laboratories, Tokyo, Japan) with an addi-tional germ tubetest andby examining morphological char-acteristics. Microbiological observations of pseudohyphae, hyphae, and chlamydospores were made on cornmeal-Tween80 agarincubated at 35°C for 3 days.
Culture conditions.Candida isolatesweregrownin Yeast
Nitrogen Base broth (Difco Laboratories, Detroit, Mich.)
with0.3% glucosefor 24 hat35°C. Thepelletswerewashed
withsteriledistilledwater,and about105washed cellswere
inoculated into 2 ml of Yeast Nitrogen Base broth with glucose andinoculatedon ashaker(150 rpm) at35°Cfor 22
h. Afterboilingina waterbath for5min,the cultureswere centrifugedat3,000 xgfor 10min.Then,the supernatants werediluted 1:2indistilledwaterforD-arabinitol
measure-ment, and serial 10-fold dilutions (1:10, 1:100, and 1:1,000) weremade for antigen detection.
Antiserum preparation.Six Candida isolates (C albicans serotypeA,C. tropicalis, C. parapsilosis, C. guilliennondii,
C.
knrsei,
and C.glabrata)
from blood cultureswere used.First, each of the isolateswasgrown on Candida GS agar
plates(Eiken ChemicalCo. Ltd., Tokyo,Japan),harvested
by centrifugation, and washed three times in sterile saline.
Next, Japanese White rabbits were inoculated subcutane-ouslywith a1:1 emulsion(1 ml)ofa10% salinesuspension
of heat-killed (80°C for 30
min)
cells and Freundcomplete
adjuvant. Two weeks later, the rabbits received another inoculation in the same manner. Twoweeks after the last
inoculation, intravenous injectionsof about 108 heat-killed cells were given weekly, and the rabbits were tested for
antibodyresponsesbythehomologousagglutinationtestand thehemagglutinationtest
(Hoffmann-La Roche, Basel,
Swit-zerland)4 daysafter each intravenous inoculation. Whena
homologous agglutination titer of 1:128 or more was
ob-tained, blood was collected, and the serum was stored at
-80°Cuntiluse.
Patients and controls. During a 5-year period from July
1987toJuly 1992,540serumsampleswereobtained from 197 3132
on April 12, 2020 by guest
http://jcm.asm.org/
hospitalized patients, and all sera were frozen at -80°C.
Seraweretakenat theacutestageofinfection, andatleast
twoserum samples perpatientwere tested. When positive resultsfor antigens or theD-arabinitol:creatinine ratio were
obtained, sera were taken twice a week thereafter. Sera drawn during the administration of D-mannitol were ex-cluded. Patients were divided intofivegroups on thebasis of clinical andmicrobiological information. Group 1 (invasive
candidiasis)consistedof16patients (72serumsamples)with
deep-seated candidiasis confirmed by histological evidence atautopsy or surgery and 42patients (178 serum samples)
who had two or more positive blood cultures drawn over a
period of days. Group 2consisted of 20 patients (52 serum
samples) with possibleinvasive candidiasis, defined by sus-tained feverdespite antibacterial therapy. Antifungalagents
such as amphotericin B, miconazole, or fluconazole were administered parenterally to all patients in this group for treatment of suspected candidiasis. Cultures of sputum(12 patients), urine(2patients), orcathetertip(4 patients)were
positive forC. albicans. C. parapsilosis was isolated from
theperitonealfluidofonepatientand from the cathetertipof onepatient. Group 3(transient candidemia) consisted of 10
patients who had only one positive blood culture for Can-dida species (6 patientswithC. albicans,2patientswithC. parapsilosis, and 1 patient each with C glabrata and C. tropicalis) and whose temperature had fallen to normal
within 2 days of the candidemia. Group 4 consisted of 54
patients with bacterial sepsis.Thesepatients madefavorable progress after parenteral administration of antibiotics. Group 5 consisted of 55 cancer patients without clinical
evidence of infection. Forty-two normalhealthy adults were also includedinthestudy.
ELISA for antigens. TheELISA technique for detecting Candida antigens is similartothatdescribedpreviously(4). Briefly, polystyrene microtitration plates (Immulon II;
Dynatech) were coatedwith0.1 ml of the diluted
immuno-globulin G (IgG)
(anti-C.
albicans IgG, 5 ,ug/ml; anti-C krusei IgG,10,ug/ml)
in0.1Mbicarbonatebuffer(pH 9.6)at 25°C for 2.5 h and then at4°C overnight. Theplates were washed withphosphate-buffered
saline (PBS) containing 0.05% Triton X-100 (PBST) by using a microplate washer.After 0.1 ml of heat-treated (9) or protease-heat treated serum samples was added to the wells, the plates were
incubatedat25°Cfor2.5h. Eachwell was washedfivetimes withPBST, after which 0.1 ml of diluted biotin-linked IgG(in PBSTcontaining 0.5%bovine serumalbuminand 1% poly-ethylene glycol 6000)was added, and theplateswere
incu-batedat25°Cfor 1.5 h. Afterfive morewashes,0.1 ml ofa 1:2,000 dilution ofstreptavidin-horseradish peroxidase
con-jugate(BethesdaResearchLaboratories)wasaddedand the
plateswere incubatedat25°C for 30min. Todetermine the
optimal dilutions of the biotin-linked IgG and
streptavidin-enzymeconjugate, checkerboard titrations were done. The unbound conjugate was removed by washing thoroughly,
and a substrate (0.1 M citrate buffer [pH 4.5]) containing 0.1%
o-phenylenediamine
and 0.012% hydrogenperoxide
wasadded at 0.1 mlper well. Theenzyme-substratereaction was allowed toproceed at25°C for 15 min, and the actual
A450 was measured on a spectrophotometer. Pooled sera
containingantigensatconcentrationsrangingfrom 0.125 to 4
ng/ml
were included on each plate. The concentrations ofantigensin the serum samples were extrapolated from the standard curve.
Candidapolysaccharide antigensofC. albicans serotype A,
C.
glabrata,C.
guilliermondii,
C.krusei,
Cparapsilosis,andC.tropicaliswereprepared bythe method of Meisteret
al. (10). Carbohydrate was measured by the anthrone
reac-tion by using D-mannose as a standard (7). For thin-layer chromatography, samples of thecarbohydrate preparations
(5 mg/ml in 2 N HCl)were hydrolyzed for 2 h in a boiling water bath. Thin-layer chromatography was carried out on
cellulose F 254 (Merck, Darmstadt, Federal Republic of
Germany) with pyridine-ethyl acetate-water (26:66:8; by volume) asthesolvent(10). Protease and heat treatmentof
serum samples was performed as follows: 0.3 ml ofeach samplewasmixedwith40,ulofprotease(20 mg/mlinPBST
[typeVIII;Sigma]), incubated for 30min at45°C, andplaced
in a boilingwater bath for 5 min. The mixture was then centrifuged, and thesupernatantwasused for antigen detec-tion.
LAfor antigens. Polystyrenebeads (0.8
pum
in diameter;Difco) were coated with antibodies against C. albicans (LA-CAreagent) orC. krusei (LA-CK reagent). Inbrief, 1
volumeof latex beads was added to the same volumeofIgG
solution diluted with glycine-buffered saline (0.85% NaCl
and0.75% glycine [pH8.4]),and themixturewasincubated
at roomtemperaturefor2h. Aftercentrifugationat3,000 x
g for 10 min, the pelleted latex beads were washed three
times with glycine-buffered saline and resuspended in 2 volumes of glycine-buffered saline containing 0.1% bovine
serumalbumin and0.05% sodium azide. The sensitivitiesof
the LA reagents were assessed for each run by testing pooled serum containing mannan antigens at final
concen-trations of 0.5, 1,2,or4ng/ml. To25 ,ul ofprotease-treated
serum, 25
RI
of latex reagent was added and mixed on aslidefor 10min at 160 rpm in a moist chamber.
D-Arabinitolmeasurement.D-Arabinitol was measuredby
using acommercialtestkit fromNacalaiTesqueCo., Ltd., Kyoto,Japan.Inbrief, 0.5ml ofsamplewasmixedwith1ml of 0.05 mol ofcitrate buffer(pH4.0)per liter. The mixture was placed in a boiling water bath for 5 min and was
centrifuged at 1,500 x gfor 10min. Then, 2 ml ofreaction
mixture containing resazurin, NAD, and diaphorase was
addedto0.3 ml of the
supernatant.
Afteradding 0.05 ml of D-arabinitol dehydrogenase solution to the mixture, the increase influorescencewasmeasured witha spectrofluoro-photometer (15).RESULTS
Homologousagglutination titers of 1:256 to1:2,048
(indi-recthemagglutinationtiters of 320 to20,480)wereobtained.
The ELISA withantiserumagainst
C.
albicansor C.tropi-calis detected both
C.
albicans antigens and C.tropicalis
antigenswith almost the same sensitivity
(Table
1). In thedetection of Candidaantigens
excluding C.
albicans andC.
tropicalis, ELISAwith anti-C krusei antibodies was some-what more sensitive than that with antibodies against
C.
parapsilosis,
C. guilliermondii,
or C.glabrata.
From the resultsgiveninTable 1,bothantibodiesagainst
C. albicansand antibodiesagainst
C.
kruseiwerechosen for the detec-tion of Candida antigens. Thepatterns
obtained bythin-layer chromatographywere identical for each antigen and showed one strong spot whichcorrespondedwithmannose.
Broth culturesof C albicans serotypeB,C. parapsilosis, C.
guilliermondii,
C.krusei,
and C.glabrata
revealed C.albicans antigen levels of 0.01 ,ug/ml or less (Table 2). C.
krusei
antigen was not detected from the broth culture ofC.
glabrata.
D-Arabinitol concentrationsranging
from 84to480
p.mol/liter
weredetected in broth cultures ofC. albicans serotypes A and B, Ctropicalis,
C.parapsilosis,
and C.on April 12, 2020 by guest
http://jcm.asm.org/
3134 FUJITA AND HASHIMOTO
TABLE 1. Absorbance of brothcultures of Candida isolates A450byELISAwith antibodies against: Broth culture (1:100)
Calbicans C.glabrata C guilliennondii C krusei C. parapsilosis C. tropicalis
C. albicans serotypeA 0.985 0.147 0.179 0.337 0.161 1.099
C. glabrata 0.044 0.289 0.052 0.099 0.106 0.033
C guilliermondii 0.055 0.162 0.344 0.515 0.059 0.055
C. krusei 0.064 0.097 0.191 >1.500 0.179 0.071
C.parapsilosis 0.065 0.162 0.143 0.286 0.298 0.081
C. tropicalis 1.488 0.575 0.291 0.297 0.099 1.433
YNWB 0.032 0.054 0.084 0.057 0.055 0.038
aYNB,YeastNitrogenBasebrothwith 0.3% glucose.
guilliennondii, whereas cultures of C krusei and C. glabrata showed D-arabinitolconcentrations of 10 ,umol/literorless.
Figure 1 shows the standardcurvesof C. albicans and C
krusei mannans according to serum treatment. The upper
limit of negativity was determined by adding 3 standard
deviations of about0.06 eachtothe absorbance of the pooled
serumsamples from normal healthy adults. With the prote-ase and heat treatment, this value was 0.2 ng/ml for C.
albicans mannanand was 0.3 ng/ml for C krusei mannan.
With heattreatmentof thesamples, the sensitivity limitwas
0.3ng/ml for C. albicansmannan,whereasitwas0.5 ng/ml
for C. kruseimannan.From theseresults,proteaseand heat treatmentwaschosenforsubsequent studies. The
sensitiv-ities of the LA-CA and LA-CKtests were 1 and 4 ng/ml, respectively.
The results oftestsfor antigens and D-arabinitol in various
groups are given in Table 3. C. albicans mannan was
detected in 31 patients (116 serum samples) with invasive
candidiasis and in 12 patients (29 serum samples) with
possible invasive candidiasis. C. albicansmannanlevels in
serumranged from 0.2to4.8 ng/ml. C. krusei mannan was
positive in 12 patients with invasive candidiasis (34 serum
samples) and 1 patient with possible invasive candidiasis (3
serumsamples). The concentrations of C kruseimannan in serum ranged from 0.3 to 2.3 ng/ml. Of 56 antigenemic patients, persistentorintermittentantigenemiawasshown in
48 patients with invasive candidiasis and 8 patients with possible invasive candidiasis. In accordance with the reduc-tion in fever, the antigenemia disappeared in six patients with invasive candidiasis and two patients with possible invasive candidiasis.
The LA-CA and LA-CKtests showedpositive results in 22patients (15 patients with C. albicans and 7 patients with C tropicalis) and 5 patients (4 patientswithCparapsilosis
and 1patient with C krusei), respectively.
The definition of a positive test result was a sample
concentration with a D-arabinitol/creatinine ratio greater
than themean + 3 standard deviations in patients without infections. This valuewas 1.5 p.mol/mg.Of 46patients with elevatedD-arabinitol/creatinine ratios, 29 patients had inva-sive candidiasis. Elevated D-arabinitol/creatinine ratios
re-turned tonormal in five patients with invasive candidiasis, three patients with possible invasive candidiasis, three
pa-tients with transient candidemia, and three patients with bacteremia. Although no false-positive reactionswere
ob-served in either theELISAorthe LA, 10 control patients in
groups 4 and 5 had elevated D-arabinitollcreatinine ratios. Among 58 patients with invasive candidiasis, 43 showed positive results for ELISA with antibodies against C.
albi-cansorC krusei, whereas only 22 had positive LA results.
The sensitivities of the ELISA, the LA, and the enzymatic D-arabinitol test for invasive candidiasis were 74, 38, and
50%, respectively. The specificities of the ELISA, the LA, and the D-arabinitolassay were100, 100, and 91%,
respec-tively.
DISCUSSION
Previous studies have shown that Candidamannan
anti-gen detection in the diagnosis of invasive candidiasis is inconsistent probably because of the low detection limits (generally above 1 ng/ml) of variousdetection systems(1, 4, 8). InourELISA, the detection limitwas0.2ng/ml, whereas
itwas0.8ng/ml inaprevious study (4). This improvementin
sensitivitymaybe duetoantisera withhigh titers against C albicans mannan and the use ofprotease treatment rather than the boiling method for the dissociation of antigen-antibody complexes. Antimannan serum was readily
ob-tainedby intravenous (9, 13)orsubcutaneous(17) injections ofheat-treated C. albicans. However, serological tests for Candida antibody such as agar gel diffusion, whole-cell agglutination, and LA have not been reliable, and the
necessity for standardizedreagentsforantibody titration has beenstressed (11). Weused acommerciallyavailable
indi-TABLE 2. Antigen and metabolite concentrationsofmedicallyimportantCandidaspeciesinbrothcultures
Meanantigenconcn(pg/ml[range]) detected by D-Arabinitol
Organisms No. of strains ELISA with antibodies against: concn(pmoV
tested ltr[ag]
C.
albicans Ckusei
liter [range])
C. albicans serotypeA 13 16(2.0-56) 0.70(0.68-0.75) 200(84-350)
C. albicans serotypeB 1 <0.01 0.39 260
C.glabrata 5 <0.01 <0.01 <0.05
C.guilliennondii 5 <0.01 0.38(0.31-0.49) 270(230-310)
C kusei 1 <0.01 28 9.2
C.parapsilosis 7 <0.01 1.5(0.5-2.5) 280(170-360)
C. tropicalis 5 29(7.0-42) 0.19(0.11-0.38) 250(160-360)
J. CLIN. MICROBIOL.
on April 12, 2020 by guest
http://jcm.asm.org/
B
Protease
Tr
T
A
1 / ; Boiling
A0~~~~~~~
JL ~ I0.125 0.25 05 2 4
C.albicans mannan, ng/ml C.krusei mannan, ng/nl
FIG. 1. Standard curvesof theenzymeimmunoassay analyzed according tothe mode oftreatment ofserum samples (0, boiling; 0,
protease), preparedinduplicatewithknownconcentrations of C. albicansmannan(A)and C. k,useimannan(B)inpooledserafrom normal healthyadults. Means standarddeviationsaregiven,and dotted lines indicatedetection limits.
rect hemagglutination test kit and obtained antiserum with antibody titers of 1:20,480, whereaswe obtained a titer of
1:2,560 inourpreviousstudy (4). Heattreatmentof diluted
serum containing EDTA has been reported togive assays greater sensitivity than the proteaseand heat treatment or
heat extraction methods(9). However, the detection limit of the EDTA and boiling method was about twofold greater
than that of theproteasemethod inourstudy. This reduction
insensitivitymayhave resulted from thetrapping of antigen within the precipitates, the incomplete dissociation of
se-rum-antigen complexes, orthe dilution (1:1) ofserum
sam-ples.
Because of thesimplicity of the LA, this method has been applied for the detection of C. albicansmannan(1, 12). The
LA formannan,which hadadetection limit of 1.25ng/mlin bufferedsaline, had been reportedtobepositive in 17 (81%) of21patientswith disseminatedcandidiasis(1). On the other hand, Phillipsetal.(12) reported thatnoneof theirpatients withcandidiasis showed positive results by LA formannan.
In our study, 22 (51%) of 43 C. albicans mannan-positive
patients showed positive results for LA-CA which could detect1 ngofmannanpermlinpooled serum.TheELISA was about three- to eightfold more sensitive than the LA, andserafrom48% ofantigen-positivepatientswerenegative
fortheantigen bythe LA.Therefore, the diagnosticvalue of
theLA is limited.
C. albicans and C. tropicalis are important pathogenic
yeaststrainsinimmunocompromised patients.Inourseries,
55 (71%) of78 patientswith invasive or possible invasive
candidiasis were affected by these two Candida species. FourCandida species, C. glabrata, C. krusei, C.
parapsilo-sis, and C. guilliermondii, have reportedly been isolated from 52 of 200 fungemic patients (5). Horn et al. (5) have
reported that most fungemic patients with C. albicans, C tropicalis,orC. krusei infections had disseminatedorlocal
diseaseatautopsy,whereasnoneof thosewith C. parapsi-losisorC. glabrata infections did (5). Inourpresentstudy,
nopatients with histologicallyprovenC.glabrata infections were found. Of12patientswith C. parapsilosis infections,
however, 1 had histologically proven endocarditis with in-termittentantigenemia, and 8 patients showed persistentor
intermittent antigenemia, suggesting invasion of the tissues by Candida strains.
Broth cultures of C. glabrata showed negative results in testsforantigens and D-arabinitol, suggesting the nonvalue of these tests in diagnosing C. glabrata infections. The ELISA with antibodies against C. knrsei detected C.
albi-cansandC. tropicalis antigens in broth cultures, but it failed
to detect C. krusei antigens in patients infected with C. albicans or C. tropicalis. One of the reasons for these
negative results may be antigen concentrations that were
below the detection limit.
Although D-arabinitol is a major metabolic product of
Candida species, a higher D-arabinitol/creatinine ratio in
serafromhigh-risk patientswithout candidiasis than that in
serafrom normalsubjectswasobserved(3).Inourstudy,13 of 109 patients without invasive candidiasis had elevated
D-arabinitol/creatinine ratios. Heavy colonization by Can-A
A450
1.0
O
.94
Protease
0.8-
0.7-
0.6-
0.5-
0.4-
0.3-
0.2- 0.1-Boiling
0.2
-0.1
on April 12, 2020 by guest
http://jcm.asm.org/
3136 FUJITA AND HASHIMOTO
TABLE 3. Results ofantigen and metaboliteassays inpatients with and without candidiasis and normalhealthyadults No.of casespositivefora:
Group andCandidaisolate No. of casesa Antigensdetectedby Antigens detected byLA Elevated ELISA with antibodies withantibodies against: D-arabinitolcreatendne
against: D-ara tio
ratio C.albicans C.krusei C.albicans Ckrusei
Invasive candidiasis
C. albicans 26(18) 22(19) 0 11(8) 0 12(9)
C. glabrata 6(1) 0 0 0 0 1
C. guilliernondii 4(2) 0 2(1) 0 0 2(1)
C. krusei 1(1) 0 1(1) 0 1(1) 0
C.parapsilosis 10(6) 0 9(3) 0 3(2) 6(4)
C. tropicalis 11(10) 9(7) 0 7(7) 0 8(8)
Possibleinvasive candidiasis
C. albicans 18(6) 12(6) 0 4(4) 0 4
C.parapsilosis 2 0 1 0 1 0
Transient candidemia 10 0 0 0 0 3
Bacteremia 54 0 0 0 0 5
Patients without infections 55 0 0 0 0 5
Normalhealthy adults 42 0 0 0 0 0
a Valuesinparenthesesindicate the number ofpatientswhodiedwith sustained fever.
dida species at mucosal sites may have resulted in the
production of sufficient D-arabinitol to cause a significant rise in concentrations inserum, because high-risk patients
were readily colonized with Candida species (14). More-over, the enzymatic fluorometric assay for D-arabinitol demonstrated positive results in cultures of C. krusei from which no D-arabinitol production was detected (2). These
positive resultsmight belinked to the assayformat;
D-ara-binitol dehydrogenase alsooxidizedserumD-mannitol, and
D-mannitol generated NADH as efficiently as D-arabinitol
did(15).
In conclusion, the ELISA with antibodies against C.
albicans in combination with the ELISA with antibodies
against C. krusei is a useful method for the diagnosis of
invasive candidiasis caused bymedically important Candida
strains excludingC.glabrata. Because of the time-consum-ing nature of the ELISA method, however, a simple and
rapid methodother than the ELISA is in use in our labora-tory.
ACKNOWLEDGMENT
This workwas supported in part by a grant from the Clinical Pathology Research FoundationofJapan.
REFERENCES
1. Bailey, J. W., E. Sada, C. Brass, and J. E. Bennet. 1985. Diagnosis of systemic candidiasis by latex agglutination for serumantigens.J.Clin. Microbiol. 21:749-752.
2. Bernard,E. M.,K.J. Christiansen, S. F. Tsang, T. E.Kiehn, and D. Armstrong. 1981. Rate of arabinitol production by pathogenicyeastspecies. J. Clin. Microbiol. 14:189-194. 3. deRepentigney, L., L. D. Marr, J. W. Keller, A. W. Carter,
R.J. Kuykendall, L.Kaufman,and E. Reiss.1985.Comparison ofenzymeimmunoassayandgas-liquidchromatography for the rapid diagnosis of invasive candidiasis in cancer patients. J.
Clin. Microbiol. 21:972-979.
4. Fujita, S., K. Maeno, and K. Soyama. 1989. Mannan and D-arabinitol concentration in serumfromapatientwith Candida albicans endocarditis.Mycopathologia 105:87-92.
5. Horn, R., B. Wong, T. E. Kiehn, and D. Armstrong. 1985. Fungemia in a cancer hospital: changing frequency, earlier onset,and results oftherapy.Rev. Infect. Dis.7:646-655. 6. Jones, J. M. 1990.Laboratory diagnosis of invasive candidiasis.
Clin. Microbiol. Rev. 3:32-45.
7. Kabat, E.A.,and M. M.Mayer. 1961.Carbohydrateestimation, p.527-528.InExperimental immunochemistry, 2nd ed. Charles CThomas,Publisher, Springfield, Ill.
8. Lemieux, C., G. St-Germain, J. Vincelette,L.Kaufman,andL. deRepentigny. 1990.Collaborative evaluation ofantigen detec-tion by a commercial latex agglutination test and enzyme immunoassayinthediagnosis of invasive candidiasis. J. Clin. Microbiol. 28:249-253.
9. Lew, M. A., G. R. Siber, D. M. Donahue, and F. Maiorca.1982. Enhanced detection with an enzyme-linked immunosorbent assayof Candida mannan inantibody-containing serum after heatextraction.J. Infect. Dis. 145:45-56.
10. Meister, H., B.Heymer, H. Schafer,and0.Haferkamp. 1977. Role ofCandida albicansingranulomatous tissue reactions.I. Invitro degradation ofC. albicans and immunospecificity of split products.J. Infect. Dis.135:224-234.
11. Merz, W. G., G. L. Evans, S. Shadomy, S. Anderson, L. Kaufman,P.J.Kozinn,D. W.Mackenzie,W. P.Protzman, and J. S. Remington. 1977. Laboratory evaluation of serological tests for systemic candidiasis: a cooperative study. J. Clin. Microbiol. 5:596-603.
12. Phillips,P., A.Dowd,P. Jewesson,G. Radigan,M.G. Tweed-dale,A. Clarke, I. Geere, and M. Kelly. 1990. Nonvalue of antigen detectionimmunoassay for diagnosis of candidemia.J. Clin. Microbiol. 28:2320-2326.
13. Reiss, E., L.Stockman,R.J. Kuykendall,andS. J.Smith.1982. Dissociation of mannan-serum complexes and detection of Candida albicansmannanby enzyme immunoassayvariations. Clin. Chem.28:306-310.
14. Sanfold,G.R., W. G.Merz, J.R.Wingard,P.Charache,and R. Saral.1980. The value offungal surveillance culturesas predic-J. CLIN. MICROBIOL.
on April 12, 2020 by guest
http://jcm.asm.org/
torsofsystemic fungal infections. J. Infect. Dis. 142:503-509. 15. Soyama, K., and E. Ono. 1985. Enzymatic fluorometricmethod
for the determination of D-arabinitol in serum by initial rate
assay.Clin. Chim. Acta 149:149-154.
16. Walsh, T. J., J. W. Hathorn, J. D. Sobel, W. G. Merz, V. Sanchez, S. M. Maret, H. R. Bucldey, M. A. Pfaller, R. Schaufele, C. Sliva, E. Navarro, J. Lecciones, P. Chandrasekar,
J. Lee, andP. A. Pizzo. 1991. Detection of circulatingCandida enolase by immunoassay in patients withcancer andinvasive
candidiasis. N. Engl. J. Med. 324:1026-1031.
17. Weiner, M. H., and M. Coats-Stephen. 1979. Immunodiagnosis of systemic candidiasis:mannanantigenemia detected by
radio-immunoassay in experimental and human infections. J. Infect. Dis. 140:989-993.