-
Review
IntroductionAspergillus spp are ubiquitous opportunistic moulds
thatcause both allergic and invasive syndromes. The genuscomprises
approximately 180 species, of which 33 havebeen associated with
human disease. Most infections arecaused by Aspergillus fumigatus,
Aspergillus flavus,Aspergillus terreus, and Aspergillus niger;1
less commonly,Aspergillus nidulans can be implicated as the
causativepathogen, especially in the setting of
chronicgranulomatous disease.2
An accurate diagnosis of invasive aspergillosis isimportant for
clinical reasons; an earlier diagnosis isassociated with improved
patient survival3 and tests witha high negative predictive value
may allow expensive andpotentially toxic antifungal drugs to be
withheld. Newdrugseg, voriconazoleexhibit differential
mouldactivity; the ability to specifically exploit their
anti-aspergillus properties requires a rapid and accuratelaboratory
diagnosis. The epidemiology of invasiveaspergillosis is changing;
invasive disease is increasinglyobserved in the non-neutropenic
phase ofhaematopoietic stem cell transplantation46 and in
non-classic settings such as critically ill patients in
intensivecare units.7 Aspergillus spp other than A fumigatussomeof
which demonstrate inherent resistance to antifungaldrugsare
increasingly recognised.810 An internationalcollaborative effort
recently produced standardiseddefinitions for invasive fungal
infections.11 Thus, a reviewof the diagnostic modalities and their
use in establishinga diagnosis of invasive aspergillosis is
timely.
Diagnostic toolsDirect techniquesThe advantages of direct
techniques over culture includesuperior sensitivity and a
relatively rapid turn aroundtime. The principal disadvantage is the
inability todefinitively distinguish other filamentous fungi
(eg,
Penicillium spp and Scedosporium spp) or implicateAspergillus
spp as the causative pathogen in circumstancesin which there are
atypical or non-specific morphologicalfeatures. This disadvantage
may compromise diagnosticaccuracy and hence estimates of
therapeutic efficacy ifpatients are recruited to clinical trials
solely on the basis ofhyphae that resemble Aspergillus spp. Within
tissuesections, Aspergillus spp typically appear as slender
septatehyphae that exhibit angular dichotomous branching(figure
1).
Lancet Infect Dis 2005; 5: 60922
DWD and WWH are at the Schoolof Medicine, University
ofManchester and WythenshaweHospital, Manchester, UK; TJW isat the
Pediatric Oncology Branch,National Cancer Institute,National
Institutes of Health,Bethesda, MD, USA. WWH is alsoat the Pediatric
Oncology Branch,National Cancer Institute.
Correspondence to: Professor David W Denning,Education and
Research Centre,Wythenshawe Hospital,Southmoor Road, ManchesterM23
9LT, UK. Tel +44 (0)161 291 5811; fax +44 (0)161 291
5806;[email protected]
http://infection.thelancet.com Vol 5 October 2005 609
Invasive aspergillosis occurs in a wide range of clinical
scenarios, is protean in its manifestations, and is still
associatedwith an unacceptably high mortality rate. Early diagnosis
is critical to a favourable outcome, but is difficult to
achievewith current methods. Deep tissue diagnostic specimens are
often difficult to obtain from critically ill patients.
Newerantifungal agents exhibit differential mould activity, thus
increasing the importance of establishing a specific diagnosisof
invasive aspergillosis. For these reasons, a range of alternate
diagnostic strategies have been investigated. Mostinvestigative
efforts have focused on molecular and serological diagnostic
techniques. The detection of metabolitesproduced by Aspergillus spp
and a range of aspergillus-specific antibodies represent
additional, but relatively underused,diagnostic avenues. The
detection of galactomannan has been incorporated into diagnostic
criteria for invasiveaspergillosis, reflecting an increased
understanding of the performance, utility, and limitations of this
technique.Measurement of (1,3)--D glucan in blood may be useful as
a preliminary screening tool for invasive aspergillosis,despite the
fact that this antigen can be detected in a number of other fungi.
There have been extensive efforts directedtoward the detection of
Aspergillus spp DNA, but a lack of technical standardisation and
relatively poor understanding ofDNA release and kinetics continues
to hamper the broad applicability of this technique. This review
considers theapplication, utility, and limitations of the currently
available and investigational diagnostic modalities for
invasiveaspergillosis.
Laboratory diagnosis of invasive aspergillosisW W Hope, T J
Walsh, D W Denning
Figure 1: The appearance of Aspergillus spp in histological
sections(A) Gomori methanamine silver (GMS) stain of rabbit lung in
experimental invasive pulmonary aspergillosis(magnification x400).
(B) A similar section stained with periodic acid-Schiff (PAS)
(magnification x400). (C) and (D)show acute angle dichotomous
branching, which is typical of Aspergillus spp (magnification
x630).The GMS sectionsdemonstrate the prominent staining and stark
appearance of hyphae. By contrast, with PAS there is preservation
ofbackground histological detail and hyphal morphology, but hyphae
are less conspicuous against the background.
-
Review
Wet mounts, potassium hydroxide preparations, and use ofroutine
stainsAll specimens obtained in scenarios in which fungi
arepossible aetiological agents should be subject to a series
ofroutine direct procedures; these procedures may varyaccording to
the specimen, degree of clinical urgency, andthe individual
laboratory. Specimens may be examined asa wet mount preparation
with or without the addition of10% potassium hydroxide, which aids
in the visualisationof hyphal elements through the partial
digestion andclearing of proteinaceous material while leaving
thefungal cell wall intact.12 Subsequently, a smear is made ona
slide, fixed and subjected to a variety of stainingprocedures. A
Gram stain should be done as a matter ofroutine, but cytological
stains (eg, Papanicolaou stain),fungal stains, and fluorescent
stains may improvesensitivity.
Fungal stainsFungal-specific stains should be applied in all
cases inwhich invasive aspergillosis is considered a
diagnosticpossibility.13 Hyphal elements are stained
withhaematoxylin and eosin, although they may be difficultto
visualise if sparse, fragmented, or present in thesetting of
substantial tissue necrosis. Fungal-specificstainseg, Gomoris
methenamine silver stain (GMS)and periodic acid-Schiff (PAS)can be
applied tohistological sections and smears (figure 1). On
occasion,GMS is referred to as Grocotts stain or the Grocott-Gomori
silver stainRobert Grocott demonstrated thatGMS, which was
initially designed as a stain forglycogen and mucin, also readily
stained fungalelements.14 PAS has the advantage of providing
acounter stain that reveals the background host cellulardetail,
tissue architecture, and inflammatory response.By contrast, the GMS
counter stain removes the finedetails of background host cells and
tissues, butprovides a more sensitive stain for detecting
smallfragments of cell wall that may be otherwise obscuredby
surrounding tissue elements. Thus, for detection ofhyphal elements,
the use of the GMS stain may be moresensitive; whereas PAS provides
more of the cellulardetail and architecture that may be of help
inestablishing relations between the fungus and otherelements of
tissue. This may be important in definingthe individual
aspergillus-related syndromes that varyaccording to the
immunological status of the host. Inthis regard, GMS and PAS are
complementary.
Fluorescent techniquesFluorescent dyeseg, Calcofluor white,
Uvitex 2B, andBlankophorare water-soluble colourless dyes
thatselectively bind to beta-glycosidically linked poly-saccharides
within fungal cell walls. They are not specificfor Aspergillus spp,
but have the advantages of relativelyhigh sensitivity, rapid
turnaround time, and broadapplicability. They may be applied to
frozen sections,
paraffin-embedded tissue, and other fresh clinicalspecimenseg,
bronchoalveolar lavage fluid (BAL) orcorneal scrapings.15,16
Immunohistochemistry, immunofluorescence, and
in-situhybridisationImmunohistochemistry (using the monoclonal
antibodyWF-AF-117 or EB-A118,19), immunofluorescence,20 and in-situ
hybridisation21,22 have been studied as diagnosticmodalities.
Collectively, these techniques have thepotential to provide genus
and species specific data, whichmay be important to improve
diagnostic certainty whenhyphae are seen invading tissue, but
cultures or otheradjunctive diagnostic data are negative. The
availability ofthese modalities in routine clinical
microbiologylaboratories is variable.
CultureA culture yielding Aspergillus spp, in addition to
enabling adiagnosis of invasive aspergillosis, may further
definetherapeutic options via susceptibility testing or
theisolation of a species possessing inherent antifungalresistance;
examples of the latter include A terreus andA nidulans, which are
both resistant to amphotericin B.10,23
The main disadvantage of culture is that it is relativelyslow
(the process takes days), is relatively insensitive,24 andrequires
specialised expertise for species determination.
In common with other pathogenic fungi, the ability togrow at 37C
distinguishes Aspergillus spp from other non-pathogenic
environmental moulds. Aspergillus spp can berecovered on most
routine solid and liquidmicrobiological media (eg, blood agar,
chocolate agar,brain heart infusion broth). A fungal-specific
mediumeg, Sabouraud dextrose agarshould be included at thetime of
initial specimen set-up in clinical scenarios inwhich Aspergillus
spp (or other moulds) are consideredpossible pathogens, because of
superior yield.25 Theaddition of antibioticseg, chloramphenicol
andgentamicinto the medium is required for the recoveryof
Aspergillus spp from specimens obtained from non-sterile sites,
since they prevent bacterial overgrowth.Cycloheximide, a eukaryotic
protein synthesis inhibitor, isfrequently added to fungal media to
inhibit theovergrowth of cultures by non-pathogenic
environmentalmoulds; however, on occasion, cycloheximide may
inhibitthe growth of Aspergillus spp.26
The identity of a laboratory isolate can often be inferredon the
basis of colonial morphology and colour. Definitiveidentification,
however, is dependent on a detailedinspection of conidial
morphology and ontogeny andrequires a microscopic examination of a
simple teasedpreparation or a slide culture (a procedure in
whichsporulation is induced and the relevant diagnostic featuresare
visualised on the under-surface of a cover-slip). Theappearance and
diagnostic features of individual species isbeyond the scope of
this review and readers are referred todefinitive texts,27 useful
guides,28 and excellent websites.
Seehttp://www.aspergillus.man.ac.uk
http://www.mycology.adelaide.edu.au and
http://www.doctorfungus.org
610 http://infection.thelancet.com Vol 5 October 2005
-
Review
Several additional issues pertaining to culture requireemphasis.
First, the growth characteristics andmorphological appearances of
Aspergillus spp areprotean and in some circumstances quite
atypical; inthis regard, Aspergillus spp are great mimics and
shouldalways be included in the list of diagnostic possibilitiesfor
an unidentified mould. Second, at least on occasion,sporulation may
be difficult or impossible to induce,29
and other modalities must be used for the purposes
ofidentification. In this circumstance, moleculartechniques are
perhaps best placed to enable rapid andaccurate identification.
Serological techniquesGalactomannan Galactomannan is a
heat-stable heteropolysaccharidepresent in the cell wall of most
Aspergillus and Penicilliumspecies.30 The molecule is comprised of
a non-immunogenic mannan core with immunoreactive side-chains of
varying lengths containing galactofuranosylunits.30 The composition
of galactomannan varies betweengenera and strains, as well as the
strain and conditionsused for its production, extraction, and
purification.30
There are two commercial assays for the detection
ofgalactomannanthe Pastorex kit (Sanofi
DiagnosticsPasteur,Marnes-La-Coquette, France) and Platelia
ELISA(BioRad, Marnes-La-Coquette, France). Pastorex is nowrarely
used, while Platelia has been available in Europe forapproximately
10 years and has recently been licensed inthe USA. There has been a
progressive increase in theunderstanding of the diagnostic utility
of galactomannanto a point that has enabled its incorporation
intodiagnostic criteria.11 However, galactomannan testing isnot
universally available to clinicians; the decision to
offergalactomannan testing within a hospital microbiologylaboratory
depends on resources, the institutionalincidence of invasive
aspergillosis, and the hospital case-mix.
Details surrounding the release and kinetics ofcirculating
galactomannan remain largely undefined. Thegrowth phase,
microenvironment, host immune status,and pathology may all
influence galactomannan release.31
An abundance of data supports the notion thatgalactomannan
production is proportional to fungal loadin tissue;3234
furthermore, galactomannan levels appear tohave prognostic
significance, with high unremitting levelsin the face of antifungal
therapy associated with anunfavourable outcome.10,3337
Assays to detect galactomannan have mostly usedserum and BAL
fluid. Galactomannan can also bedetected in tissue and a number of
bodily fluids includingCSF, peritoneal fluid, urine, and
pericardial fluid,although data to support its use at these sites
is relativelyscant, and is likely to remain that way.38
Galactomannan assays use EB-A2, a monoclonalantibody derived
from rats, which is directed towards the (1,5)-linked
galactofuranoside side-chain residues of the
galactomannan molecule.39 Four or more epitopes arerequired for
antibody binding.31,39 Detection is achievedusing a sandwich ELISA
format, which is made possibleby multiple immunoreactive epitopes
on a singlegalactomannan molecule.39
There are a number of important determinants ofanalytical
sensitivity of galactomannan assays. First, thebinding of EB-A2
requires four or moregalactofuranoside epitopessensitivity may
becompromised by the inability to detect secreted antigensthat bear
fewer residues.31 Second, the Platelia assay isdependent on a
pretreatment step, the goal of which is toremove complexing
antibody that may block EB-A2binding. However, the acid-sensitive
galactofuranosideresidues may be degraded by the edetic acid used
in thisstep.31 Finally, the limit of detection using the
sandwichELISA format is lower (1 ng/L) than that achievableusing
latex agglutination (15 ng/L).40 In terms of theanalytical
specificity, cross reactivity with otherfilamentous fungi,
bacteria, drugs, and cotton swabshave been documented,4145 but
whether this is due to(exogenous) galactomannan or unrelated
cross-reactivemolecules is unclear.
There have been considerable efforts in establishing
theappropriate galactomannan ELISA cut-off to maximiseclinical
sensitivity and specificity. The ELISA endpoint is acontinuous
variable and the optimal cut-off should bedetermined after defining
the receiveroperator curverelation (ie, the relation between
sensitivity and1specificity).46 The cut-off level of 15 ng/L
initiallyrecommended by BioRad and used in many early studieshas
been progressively revised downwards; a cut-off of05 ng/mL is now
currently accepted by the US Food andDrug Administration (FDA),
while a level of 07 ng/L iscommonly used in Europe.47
The clinical sensitivity of galactomannan ELISA issomewhat
variable, with a range of 29100%.31 There are anumber of potential
reasons for these disparate results.First, the performance of the
assay may differ according tothe host group and therefore the
underlying pathologicalprocess. In studies of profoundly
immunocompromisedpatients, sensitivity has been generally reported
to be inexcess of 90%,48,49 while in other settingseg,
chronicgranulomatous disease50 and solid organ
transplan-tationsensitivity appears to be somewhat lower.5153
Second, accumulating evidence suggests that
concomitantantifungal therapy leads to a decrease in the
sensitivity ofgalactomannan.32,36,54 Finally, inadequate
samplingstrategies could conceivably compromise
clinicalsensitivity; the optimal sampling strategy for screeninghas
not been rigorously defined, but the twice weeklydetermination of
antigen levels has been generally used inpatients deemed to be at
risk of invasive aspergillosis. Bycontrast, galactomannan levels
should be determinedimmediately in a host with a constellation of
clinicalfeatures indicative of invasive aspergillosis to facilitate
adefinitive diagnosis.
http://infection.thelancet.com Vol 5 October 2005 611
PersonalHighlight
-
Review
The clinical specificity of galactomannan is generallyestimated
to be greater than 90%.32,36,4749,55 The specificity
ofgalactomannan in neonates and children appears to belower, which
is possibly due to the ingestion of extraneousgalactomannan (in
food and water) and translocationacross a damaged or immature gut
wall.45,47,56 Antibioticsrepresent an additional source of
extraneousgalactomannan that may compromise clinical
specificity.The in-vitro reactivity of a range of antibiotics
ingalactomannan assays was originally reported in 1997.57
More recently, positive galactomannan results in
patientsreceiving piperacillin-tazobactam have beendocumented.58,59
This phenomenon has been furtherexplored in vitro and in vivo and
probably relates to thepresence of galactomannan within the drug
itself.60,61 Thisfinding has forced some institutions to change
theirantibacterial protocols and the FDA to issue a warning.62
(1,3)--D glucanThere has been an emergence of clinical data
pertaining tothe diagnostic utility of the cell wall component,
(1,3)--Dglucan.6367 (1,3)--D glucan assays have been developed
byWako Pure Chemical Industries (Tokyo, Japan),Seikagaku Kogyo
Corporation (Tokyo, Japan), MaruhaCorporation (Tokyo, Japan) and
Associates of Cape Code(Falmouth, USA); the assay developed by
Associates ofCape CodeFungitellhas been approved by the FDA inthe
USA for the diagnosis of invasive fungal infections. -D glucan is
present in the cell wall of most fungi; thenotable exceptions are
Cryptococcus spp and thezygomycetes.67 The molecule is ubiquitous
in theenvironment and has been used as a marker of fungalbiomass.68
The presence of (1,3)--D glucan in fungalspecies other than
Aspergillus spp (eg, Candida spp,Fusarium spp, Acremonium spp, and
Pneumocystis jiroveci)means that its role in establishing a
specific diagnosis ofinvasive aspergillosis is not
straightforward.
Assays to detect (1,3)--D glucan typically use serum.The common
feature of all of the glucan assays is theability of (1,3)--D
glucan to activate a coagulation cascadewithin amoebocytes derived
from the haemolymph ofhorseshoe crabs. Horseshoe crab lysate
preparations werefirst used to detect endotoxin using the limulus
test orlimulus reaction (named after one type of horseshoe
crab,Limulus polyphemus). Endotoxin induces clot formation viaa
serine protease zymogen named factor C (figure 2).Subsequently,
evidence emerged that (1,3)--D glucan-induced clot formation
independently of factor C, via asecond serine protease zymogen,
factor G, thus providingthe impetus for the development of the
current assays.
The analytical sensitivity of the Fungitell assay is in theorder
of 1 pg/mL, which is less than the cut-off of60 pg/mL used in a
recent clinical study.67 A technicalconsideration pertinent to the
analytical sensitivity of (1,3)--D glucan assays is that human
plasma contains anumber of inhibitors of serine proteases that need
to beremoved in a pretreatment step; this removal can be
achieved by an alkali reagent method (Fungitell), or by
theaddition of Triton X-100 and heating to 70C for10 minutes (Wako
assay). The alkali pretreatment step inthe Fungitell assay also
converts triple-helix glucans intosingle-helix structures, which
appear to be more reactive.Since both endotoxin and (1,3)--D glucan
activate thehorseshoe crab coagulation pathway, an assay
thatspecifically detects (1,3)--D glucan requires removal
ofendotoxin from the specimen or the endotoxin-specificpathway from
the lysate; correspondingly, endotoxin isinactivated by the
addition of polymyxin in thepretreatment step in the Wako assay,
while the Fungitellassay uses factor C to deplete limulus lysate.
Thepretreatment step also enhances analytical specificity viathe
removal of non-specific activators of serine proteasespresent in
human serum.
There are no data that address the clinical sensitivity ofthe
(1,3)--D glucan assays specifically for Aspergillus spp.The
positive cut-off of 60 pg/mL was defined in a non-neutropenic group
of patients with candidaemia.69 Theperformance of (1,3)--D glucan
in the context ofantifungal therapy has not been rigorously
studied. False-positive (1,3)--D glucan results have been
documented inhaemodialysis, cardiopulmonary bypass, treatment
withimmunoglobulin products, and exposure to glucan-containing
gauze (eg, following major surgery).69
Environmental (1,3)--D glucan contamination may alsocompromise
specificity.
Antibodies directed toward Aspergillus sppThe demonstration of
specific antibody is required toestablish the diagnosis of chronic
pulmonaryaspergillosis.69 Traditionally, antibody detection has
notbeen considered useful for the diagnosis of acute
invasiveaspergillosis, following an early study that failed
todocument antibody formation in 15 patients with
invasiveaspergillosis.70 Subsequently, antibody has beendocumented
in approximately one-third of patients withinvasive
aspergillosis.47,71 The detection of antibody may
612 http://infection.thelancet.com Vol 5 October 2005
Endotoxin (1,3)--D glucan
Factor C Activated factor C Activated factor G Factor G
Factor B Activated factor B
Clotting enzymeProclotting enzyme
Clotting enzyme activity detected via cleaving of synthetic
chromagenicsubstrate or turbidimetric assay
Figure 2: The pathways for the activation of the amoebocyte
lysate byendotoxin and (1,3)--D glucan and the use of this pathway
for thedetection of (1,3)--D glucan
PersonalHighlight
-
Review
prove to be the best non-invasive means of establishingthe
diagnosis of subacute invasive aspergillosis in non-neutropenic
patients with invasive aspergillosis, asillustrated by a recent
case report describing invasivepulmonary aspergillosis in an
individual with chronicgranulomatous disease.72 Furthermore,
antibody detectioncould be useful as a means of establishing a
retrospectivediagnosis of invasive aspergillosis in
profoundlyimmunocompromised hosts who have undergoneimmunological
reconstitution, although more work isrequired in this regard.
The detection of antibodyMany assay formats have been used to
detect antibodies toAspergillus spp, including immunodiffusion,
counterimmunoelectrophoresis, complement fixation,
particle-haemagglutination, indirect-immunofluoresence,
radio-immunoassay, and ELISA.73,74 The large number ofepitopes in
crude extracts may compromise specificity.The use of recombinant
antigenseg, dipeptidyl-peptidases,75 superoxide dismutase,75,76
catalase,75 metallo-protease,75 mitogillin,77 and
galactomannoprotein71,78mayrectify this situation. One potential
advantage of usingassays with a single antigen is the prospect of
studyingprotective epitopes and thereby facilitating the
generationof assays that may also confer prognostic
information.
MetabolitesAspergillus spp produce a range of extracellular
enzymes(eg, metalloproteases, phospholipases) as well asprimary
(eg, mannitol)33 and secondary metabolites (eg,gliotoxin),79 all of
which at least have the potential toserve as diagnostic markers for
invasive aspergillosis.The ability of Aspergillus spp to produce
D-mannitol hasbeen known for many years80 and its diagnostic
potentialexamined in several experimental models of
invasiveaspergillosis,33,81 although it is limited in terms of
itsbroad applicability as a diagnostic tool because of
thecomplexity of measurements, which are done by gasliquid
chromatography and mass spectroscopy. Recentwork suggests that
gliotoxin is produced by most Afumigatus strains and the
possibility of using it as adiagnostic marker has been
entertained.82 Acomprehensive summary of the various
secondarymetabolites (mycotoxins) produced by Aspergillus sppcan be
found at http://www.aspergillus.man.ac.uk. Thedetection of
metabolites represents an under-researchedarea in terms of their
possible application as diagnosticmodalities for invasive
aspergillosis.
Nucleic acid testsAs far as the amplification of nucleic acid
and diagnosis ofinvasive aspergillosis is concerned, PCR technology
hasdominated. A limited number of publications have usedthe
isothermal technique nucleic acid sequence-basedamplification.83,84
Only PCR will be discussed here. Thelack of standardisation of
technical issues has and
continues to represent a considerable barrier for thewidespread
application of PCR as a diagnostic modalityfor invasive
aspergillosis and this is the focus of thefollowing
discussion.85
Clinical specimensMany studies have addressed the detection of
nucleicacid from various fractions of blood (serum, plasma,whole
blood) to establish a diagnosis of invasiveaspergillosis, but PCR
may also be applied to BALspecimens86,87 and tissue,88 including
paraffin-embeddedsections.89,90 The optimal blood fraction for the
detectionof aspergillus DNA remains unknown. One study,
usingquantitative PCR (qPCR), suggested that the yield ofDNA from
serum, plasma, and white cell pellet wassimilar,91 while another
demonstrated that the(qualitative) PCR signal from whole blood was
superiorto plasma.92 Serum has the advantage that it
enablesconcomitant antigen testing91 and does not require
theaddition of anticoagulants (eg, sodium citrate, edetic acid,or
heparin) that may inhibit PCR.93
DNA extractionThere are a multitude of extraction techniques;
theprincipal technical issues are summarised in table 1.The chosen
extraction method represents acompromise between efficiency,
freedom fromexogenous contamination, and applicability to
routinehigh-throughput laboratories. The fungal cell wallclearly
represents the major hurdle to high-efficiencyextraction of fungal
DNA. DNA may be extracted usingin-house methods, commercial kits
(eg, Qiagen QIAmpTissue Kit [Hilden, Germany]), and
automatedcommercial techniques (eg, MagNA Pure LC
[RocheDiagnostics, Basel, Switzerland). Automatedcommercial
techniques are probably required to makefungal DNA detection a
viable option for routine clinicallaboratories. The efficiency of
extraction of fungal DNAmay vary considerably between commercial
kits.94 Highspeed cell disruption incorporating chaotropic
reagentsand lysing matrices provide efficient and high yields ofDNA
from Aspergillus spp and other filamentous fungi.95
Fungal contamination of extraction systems andreagents has been
documented.96 Considerabledifferences in DNA extraction protocols
andperformance is one aspect of molecular assays thathinders the
comparison of studies.
Primer targetFor clinical diagnostic purposes, the detection of
a broadrange of fungi is important, as is the ability to
ultimatelyidentify the specific pathogen(s). The optimal approach,
inthis regard, involves the application of broad-rangingpanfungal
primers with post-amplification analysis forspecies determination.
Panfungal primers are directedtoward conserved regions, usually
within multicopygenes, which flank sequences containing species
specific
http://infection.thelancet.com Vol 5 October 2005 613
-
Review
polymorphisms that can be exploited in
post-amplificationanalysis.
The ribosomal DNA (rDNA) complex is the mostcommon target. This
complex contains both conservedand variable sequences and there is
a large volume of datadeposited in public databases for a wide
range of generaand species. The recent genome sequencing ofA
fumigatus, using strain Af293, revealed 35 repeatingunits;97 the
structure of the gene complex is illustrated infigure 3.98 The
mitochondrial genes encoding some of thetRNA genes91 and
(apo)cytochrome b99 have also been usedas primer targets.
Mitochondrial targets can be consideredmulticopy because of a
multiple number ofmitochondria per cell nucleus; in Af293, there
were12 copies of the mitochondrial genome present for everycopy of
the nuclear genome.97
Amplification formatNested PCR formats have been widely used for
Aspergillusspp in an attempt to optimise analytical sensitivity,
but the
requirement to open reaction tubes means that there
isconsiderable risk of contamination and the subsequentgeneration
of false-positive results. Real-time formats havebeen increasingly
used and are likely to dominate in thenear future.
Post-amplification analysisPost-amplification detection
techniques provide genusor species specific data but may also
increasesensitivity and specificity.100,101 Real-time
detectiontechniques (eg, TaqMan, LightCycler, molecularbeacons) are
automated, rapid, and reproducible, thusfacilitating comparisons
between studies. Southernblotting has had a valuable role in the
evolution of PCRas a diagnostic modality, but is unlikely to have
anysubstantial future role in routine clinical assays.Single-strand
conformational polymorphism,102,103
restriction fragment length polymorphism digestpattern,104 Line
Probes,105 fragment size deter-mination,106 and PCR-ELISA107 may
have a limited role
614 http://infection.thelancet.com Vol 5 October 2005
Feature Definitions and goals Specific considerations with
respect to invasive aspergillosis
Sample Sample type, volume, transport, and handling Serum and
white cell pellet equivalent and possibly superior to plasma as
sampleshould be defined Heparin and citrate inhibitory to PCR
Sample handling varies between studies (some have demonstrated
stability at room temperature for 48 h, others recommend immediate
freezing)
DNA extraction Target is of an adequate concentration and
quality Extraction efficiency for fungal DNA is low due to the
requirement to break the cell wallfor amplification. PCR inhibitors
and DNA nucleases Contamination rate 33% in one series, commercial
reagents may be contaminated with removed fungal DNA
Ideally negative and positive extraction controls should be
usedRemoval of red and white cells Red cell lysis buffer and white
cell lysis bufferDisruption of cell wall Enzymatic methods (eg,
lyticase, zymolase)
Chemical (eg, boiling in dilute alkali) Physical (eg, glass bead
milling, freeze-thawing, sonication, grinding in liquid
nitrogen)
Disruption of cell membrane Usually achieved with lysis buffer
(sodium dodecyl sulphate, beta-mercaptoethanol, EDTA)Precipitation
of protein and purification of DNA Phenol-chloroform
Silica fibres (eg, Qiagen Tissue Kit) Capture of DNA Alcohol
precipitation
Magnetic beads (eg, MagNA Pure)Silica fibres (eg, Qiagen Tissue
Kit)
Amplification Nested PCR, real-time formats, PCR-ELISA represent
Nested formats potentially allow for optimal analytical sensitivity
but are associated withthe commonest formats contamination and are
difficult to compare
Real-time formats will probably dominate in the
futureAmplification controls Negative and positive controls are
required
Analytical The smallest number of target organisms reliably and
Multicopy target preferablesensitivity reproducibly detected by the
assay Assessed by serial dilution of Aspergillus spp (conidia or
purified DNA) using the appropriate
clinical specimen as the diluent Circulating DNA in invasive
aspergillosis is typically less than 10 colony forming units per mL
or less than 30 fgExtraction method, primer target, and detection
method all influence analytical sensitivity
Analytical Does the test detect only what it purports to?
Specific primer and probe sequences initially identified from
public databasesspecificity Amplicon ideally should be sequenced
and a BLAST search done
Cross reactivity studies with a range of fungal and bacterial
pathogens as well as human DNA are requiredPost-amplification
detection probe(s) may enhance specificity
Inhibition Inhibitors of DNA polymerase (eg, heparin) Spiking
with purified aspergillus DNA and analysing in a separate reaction
controls Spiking with a plasmid construct containing different size
and sequence or label to the target
Amplifying a human housekeeping gene (eg, betaglobin, HLA2),
which also allows some determination of specimen adequacy, although
the relative dominance of human DNA in clinical samples may mask
low levels of inhibitors which could interfere with target
amplification
Contamination Uracil-D-glycolase, appropriate number of control
negative controls
Table 1: Technical variables required for a robust and
reproducible PCR assay
-
Review
in specific instances, such as the identification oflaboratory
isolates.
Analytical sensitivity and specificityThe analytical sensitivity
of a molecular assay is usuallydetermined by serial dilution of the
infectious agent inpooled non-infectious clinical material as the
diluent.108
Such a paradigm immediately presents a problem forAspergillus
spp or any other mould, since accurate andindeed meaningful
dilution of hyphae is not possible.Two commonly used approaches
include serial dilutionof conidia or DNA (either purified genomic
DNA or aplasmid construct), although neither are ideal; theformer
does not mimic a biologically valid scenario,since hyphae rather
than conidia are the invasive form,while the latter does not
control for issues in extractionefficiency. If it is intended that
more than one species isdetectable then DNA from those species
should beincluded in the assessment of analytical
sensitivity.109
The analytical sensitivity of published assays varies byseveral
orders of magnitude; however, most studiesreport detection limits
in the order of 110 fg DNA;variability in the detection limit is
yet another issue thatcompromises study comparability.
Studies differ considerably in terms of the methods andextent to
which analytical specificity is determined; there
are no standard techniques or criteria (table 2). Primertargets
are generally identified by aligning sequencesretrieved from public
databases. This practice should beviewed as a necessary but
insufficient step in establishingthe analytical specificity of an
assay and further validationprocedures are required. Ideally,
relatively early in assay
http://infection.thelancet.com Vol 5 October 2005 615
IGS
5'
5'
IGS IGSITS1 ITS2
58S18SSSU
28SLSU
3'
3'
IGS
Figure 3: The structure of the ribosomal DNA complexShaded areas
denote areas of variability that are present throughout the complex
that can be exploited to designassays of varying levels of
specificity. IGS=intergenic spacer; ITS=internal transcribed
spacer; LSU=long subunit;SSU=short subunit. Adapted from reference
98.
Primer target Assay format Intended BLAST search of primer and
probe sequences; isolates Method by which analytical specificity
determined and result Referencespecificity with same probability
match as intended target
18S rRNA PCR-ELISA Aspergillus spp Aspergillus spp, Penicillium
italicum, Penicillium commune, Penicillium Cross-reactivity
studies: Einsele chryosogenum, Penicillium brevicompactum,
Penicillium phialosporum, Amplification of Aspergillus fumigatus,
Aspergillus flavus, Aspergillus et al110
Penicillium tardum, Penicillium allii, Penicillium expansum,
Ajellomyces terreus, Aspergillus niger, Aspergillus nidulans,
Aspergillus versicolor, capsulatus (telemorph of Histoplasma
capsulatum), Paracoccidiodes Histoplasma capsulatumbrasiliensis,
Eupenicillium spp, Penicilliopsis spp No amplification of
Malassezia furfur (3 strains), Fusarium spp (3 strains),
Trichosporon cutaneum (2 strains), Mucor spp (3 strains),
Penicillium spp (2 strains), Pseudallescheria boydii (1 strain),
Paecilomyces spp (2 strains), Saccharomyces cerevisiae (2
strains)
18S rRNA TaqMan Aspergillus spp Aspergillus spp, P italicum,
Penicillium glabrum, P commune, Cross-reactivity studies: Kami P
chryosogenum, P brevicompactum, P phialosporum, Penicillium
Ampification of A fumigatus, A niger, A terreus, A flavus,
Aspergillus oryzae et al111
purpurogenum, P tardum, Penicillium verruculosum, Penicillium
hirsutum, No amplification of Candida albicans, Candida tropicalis,
Candida krusei, Penicillium radicum, Penicillium funiculosum,
Penicillium siamense, Candida parapsilosis, Candida glabrata,
Candida guilliermondiiPenicillium pittii, Penicillium minioluteum,
Penicillium pinophilum, Penicillium variabile, Penicillium
rugulosum, Penicillium crateriforme, Penicillium variotii,
Eupenicillium spp, and others
Mitochondrial Competitive PCR Aspergillus spp A fumigatus
Cross-reactivity studies: BretagneDNA (tRNA) with PCR-ELISA
Ampification of 30 isolates of A fumigatus, A niger, A terreus, A
flavus et al112
No amplification of A nidulans, C albicans, C tropicalis, C
krusei, C parapsilosis, C glabrata, Cryptococcus neoformans
Mitochondrial Competitive PCR A fumigatus, No database matches
Amplicon sequenced: revealing A fumigatus and A flavus Bretagne DNA
(tRNA) with PCR-ELISA A flavus Cross-reactivity studies: et
al113
No ampification of A niger, A terreus, A nidulans, Aspergillus
ustus, Penicillium purporogenum, Scopulariopsis brevicaulis
Mitochondrial LightCycler A fumigatus A fumigatus None, although
clinical specificity assessed using 20 serum samples from Costa DNA
(tRNA) healthy individuals et al91
Mitochondrial LightCycler A fumigatus Eupenicillium shearii,
Neosartorya fischerii, A fumigatus Cross-reactivity studies: Spiess
DNA (cyto- Amplification of A fumigatus, Aspergillus clavatus et
al99
chrome b) No amplification of Candida spp, other Aspergillus
spp, P chryosegenum, P expansum, P funiculosum, P variotii,
Rhizopus oryzae, Fusarium proliferatum
The BLAST searches were done at http://www.ncbi.nlm.nih.gov,
using search for short nearly exact matches. The primer and probe
sequences were searched simultaneously and were separated by a
string of at least ten nucleotides toensure only the specified
sequences were matched in the search algorithm. Only matches
identical to those of the intended target are displayed.
Table 2: Selected examples of issues in establishing the
analytical specificity of PCR assays to detect Aspergillus spp
-
Review
development, the amplicon should be sequenced and aBLAST search
done to verify that the intended target hasbeen amplified.109
Subsequently, the assay should bechallenged with organisms that
have a high likelihood ofcross-reacting with the target; in the
case of Aspergillusspp, genera that are close phylogenetic
relationseg,Penicillium spp and Paecilomyces spp114 are
especiallyimportant to consider (figure 4). A further
considerationis that sequences are being continuously deposited
inpublic databases; a unique sequence at the time of primerdesign
may subsequently align with a sequence from anunrelated species or
genus deposited at a later date. Somehave suggested that BLAST
searches are done on anannual basis to ensure there is no cross
reactivity withrecently submitted sequence.109 A final
consideration isthat false-positive reactions due to
carry-overcontamination of amplicon from previous reactions maybe
prevented with the addition of uracil-D-glycolase.85
Clinical sensitivity and specificityThere are a number of
factors that potentially have animpact upon the clinical
sensitivity of PCR. Themagnitude of the quantitative PCR signal
falls withantifungal therapy in both experimental models and
inclinical contextsthis may account for false-negative
PCRresults.110,111,115,116 Patients at risk for invasive
aspergillosisare also often prescribed a multitude of drugs and
fluids,all of which may act as non-specific inhibitors of PCR; as
a
result, inhibition controls are mandatory and may take theform
of spiking the sample with aspergillus DNA, aplasmid construct, or
amplification of a human gene suchas betaglobin (table 1).
The application of diagnostic modalitiesLaboratory isolatesGiven
the distinct differences in disease manifestations,prognosis, and
antifungal susceptibility betweendifferent fungal genera and
species, a rapid diagnosiswill assume increasing importance. The
inherentproblems with identification using culture methods havebeen
outlined. An increasing number of studies haveexamined the use of
PCR to enable the accurate andrapid detection of laboratory
isolates (table 3). The rapididentification of laboratory isolates
using microarraytechnology with a panfungal chip is possible and
nodoubt the relevant studies will emerge in the near future.
Clinical specimensThe application of diagnostic modalities to
tissue,respiratory tract secretions, and blood in the context of
thepathophysiology of invasive pulmonary aspergillosis
isillustrated in figure 5.
Tissue and sterile fluidsHistological and culture techniques
applied to tissue formthe reference diagnostic standard for
invasive
616 http://infection.thelancet.com Vol 5 October 2005
005
100
100
100
100
100
58
100
100
Candida krusei
Candida lusitaniaeCandida albicans
Taphrina deformans
Aureobasidium pullulans
Histoplasma capsulatum
Blastomyces dermatitidis
Coccidioides immitis
Trichophyton rubrum
Eremascus albusAscosphaera apis
Penicilllum marneffei
Paecilomyces variotii
Peniclllium chrysogenum
Aspergillus flavusAspergillus fumigatus
Aspergillus terreus
Aspergillus nigerAspergillus nidulans
Eurotium fubrumMonascus purpureus
Neurospora crassa
Colletotrichum gloeosporiodesOphiostoma uimi
Ophiostoma stenoceras
Sporothrix schenckii
Candida viswanathiiCandida tropicalis
Schizosaccharomyces pombe
Endomyces geotrichum
Figure 4: The phylogenetic relations between Aspergillus spp and
other fungi based on the 18S rRNA complexAdapted from reference
114.
-
Review
aspergillosis11 and have been, and continue to be, thestandard
tools by which tissue invasion and destruction byhyphae is
documented. Within this context, the followingpoints specifically
deserve emphasis. First, difficulties inobtaining deep tissue
specimens in patients who are leastable to tolerate invasive
procedures have been exhaustivelydocumented and remain one of the
principal factorsdriving the development of new diagnostic
techniques.Second, the analytical sensitivity of both histology
andculture is relatively low, meaning that invasive disease iswell
established by the time that culture and histology arepositive.
Third, the specificity of the reference standard forAspergillus spp
is optimised with the combination ofhistological and culture data
and this rigorous standardhas been used in some recent clinical
trials.48,120 Theproblem, however, is that Aspergillus spp can only
berecovered from tissue in the context of positive histologyin
3050% of cases.24 Finally, the possibility of accepting apositive
PCR result in tissue as the reference standard forinvasive
aspergillosis deserves increasing attention.Certainly, data from
experimental models suggests thatvalidated PCR is more sensitive
than culture for thedetection of Aspergillus spp in tissue,
especially in thesetting of substantial tissue necrosis;115,121 the
key in thisregard is assay validation.
Non-sterile sitesIn the absence of tissue specimens, samples
obtainedfrom contiguous non-sterile siteseg, the upper andlower
respiratory tractserve as a surrogate with whichto establish the
diagnosis of invasive aspergillosis. Inthe case of invasive
pulmonary aspergillosis, viablehyphal elements or related
serological or molecularmarkers are shed into the respiratory tract
from infectedparenchyma (figure 5). A body of data suggests
thisshedding occurs relatively late in the natural history,thus
compromising attempts to establish an earlydiagnosis using this
approach.25,122,123 The isolation ofAspergillus spp (or related
serological, molecular, orbiochemical markers) in the respiratory
tract may
represent one of three scenarios: (1) evidence of
currentdisease, (2) true colonisation, or (3) a marker for
thefuture development of invasive disease. An example ofthe latter
is provided by a study that demonstrated that apositive PCR result
from BAL at the time of bonemarrow transplant conditioning was
predictive of thesubsequent development of invasive
pulmonaryaspergillosis.124
There are a number of points to make about usingBAL specimens to
secure a diagnosis of invasivepulmonary aspergillosis. First,
although BAL is a safeprocedure, even in patients with
substantialimmunological impairment, it is not a trivialundertaking
and requires a dedicated and competentbronchoscopist and an
adequate commitment ofresources. Second, the overall sensitivity
(using cultureand microscopy) is relatively low and generally
estimatedto be in the order of 50%.122,123,125,126 Variations in
BALtechnique,127 the location, size, and type of
pulmonarylesions,128130 and the timing of bronchoscopy122 are
allimportant determinants of the overall estimate. Theimpact of
antifungal therapy in terms of the recovery ofaspergillus and
related markers in the respiratory tractremains poorly defined.
Third, the specificity of theisolation of Aspergillus spp from the
respiratory tract inpatients with substantial immunological
impairmenteg, those with allogeneic haematopoietic stem
celltransplantation or neutropeniais very high,131 but forother
patient groups, the likelihood of underlyinginvasive pulmonary
aspergillosis varies enormously.25,131
Fourth, qPCR may prove to be especially useful indetermining the
relation between the fungal burden inthe respiratory tract and the
probability of underlyinginvasive disease;87,99,115 however, at the
current time, thebenefit of PCR over conventional culture remains
to befurther defined. Finally, the diagnostic yield from BALfluid
is potentially optimised with the application ofmore than one test;
a recent study demonstratedsensitivity was improved with the
concomitantapplication of galactomannan and PCR.130
http://infection.thelancet.com Vol 5 October 2005 617
Specimen Target Demonstrated specificity PCR format Detection
method Reference
Cultures ITS1-58S rRNA-ITS2 Aspergillus fumigatus, Aspergillus
flavus, Aspergillus Conventional Sequencing of amplicon Henry et
al117
terreus, Aspergillus niger, Aspergillus ustusCultures ITS1-58S
rRNA-ITS2 A fumigatus, A flavus, Aspergillus nidulans,
Aspergillus
versicolor Conventional Line probe Martin et al105
Cultures ITS1-58S rRNA-ITS2 A fumigatus, A flavus, A terreus, A
niger, A nidulans Conventional SSCP Rath et al103
Cultures ITS1-58S rRNA-ITS2 A fumigatus Multiplex PCR Ethidium
bromide Luo et al118
Cultures ITS1-58S rRNA-ITS2 A fumigatus Nested Ethidium bromide
Zhao et al119
Cultures 58S rRNA-ITS2 region A fumigatus, A flavus, A terreus,
A niger Conventional Automated fluorescent capillary
electrophoresis (detection of different length of amplicon) Turenne
et al106
Cultures 18S rRNA A fumigatus, A terreus Conventional SSCP Walsh
et al102
Cultures and ITS2 Aspergillus spp and Penicillium spp, A
fumigatus, A flavus, Conventional PCR-ELISA De Aguirre et al107
tissue A terreus, A niger, A nidulans, A ustus, A versicolor
SSCP=single-strand conformational polymorphism
Table 3: The use of PCR in the identification of Aspergillus
spp
-
Review
BloodBlood sampling represents the optimal
non-invasivediagnostic approach for invasive aspergillosis.
Despitetheir propensity for vascular invasion, Aspergillus spp
areonly very infrequently isolated from blood usingconventional
culture techniques, hence the traditionaldependence on tissue
specimens to secure a definitivediagnosis of invasive
aspergillosis. There is an extensivebody of literature examining
the diagnostic utility ofmolecular and serological techniques in
blood.Galactomannan has been incorporated into diagnosticcriteria
for invasive aspergillosis and the technical issuesrequired for PCR
to be applied in the same manner have
been discussed. However, there remain some additionalpertinent
issues. First, specific sampling strategies are yetto be
systematically studiedyield is almost certainly afunction of the
volume and frequency of sampling, as isthe case with blood
cultures. Second, the appropriateinterpretation of a positive
galactomannan or validatedPCR result in a patient at risk of
invasive aspergillosis, butwithout subsequent evidence of invasive
disease, remainsunclear and difficult to resolve; the most
conservativeinterpretation in this context is that all single
positiveresults are false-positive, but at least on occasion,
suchresults may reflect true invasive disease that has abortedor is
non-progressive. Third, a body of evidence suggests
618 http://infection.thelancet.com Vol 5 October 2005
Appearance of viable or non-viablehyphal elements or surrogate
markers(galactomannan, DNA) in the contiguousrespiratory tract
Pathogenesis of IPA
Respiratory tract
Inhalation of conidia,germination, mucosalsurfaces breached
Blood
Dissemination tonon-contiguous sites
A B
Lung andsinus tissue Tissue
invasionanddamage
Haematogenousdissemination
Compartmental characteristicsand inter-compartmental
relations
Appearance of Aspergillus sppand associated markers occurslate
in the natural history of IA
Quantitative relation betweenfungal load in tissue and
respiratorytract difficult to determine with culture;quantitative
PCR may be useful
Positive predictive value dependson underlying disease
Destruction and invasion of tissue byhyphal elements represents
the conceptualunderpinning for IA; the demonstration ofAspergillus
spp in tissue is the referencestandard for IA
Sensitivity of tissue sampling may be low due to sampling error,
prior antifungal therapy, fungal tissue load beneath the analytical
sensitivity of histology and culture
Highest specificity achieved with combination of histology and
culture
Blood cultures typically negative
Risk factors for haematogenousdissemination remain poorly
defined
Clinical samples andsampling strategies
Bronchoalveolar lavageSensitivity ~50%
Sputum examinationPositive sputum culturesoccur relatively late
in thenatural history
Fine needle aspiration May be more sensitive than bronchoscopy
depending on the radiological pattern of disease
Complications include pneumothorax,haemoptysis,
haemorrhagiccomplications, seeding of needle tract
Open biopsySensitivity compromised by infarctionand
necrosisIncreased diagnostic certainty may not translate to
improved patient outcome
Blood samplingBlood cultures generally not helpfulPCR and
galactomannan from blood potentially usefulOptimal sampling
strategies yet to berigorously defined
Figure 5: Compartmental characteristics, inter-compartmental
relations, and sampling strategies as they relate to the
pathogenesis of invasive pulmonaryaspergillosis(A) Hepatosplenic
aspergillosis (courtesy of Damon Eisen). (B) Cerebral abscess due
to Aspergillus fumigatus. IA=invasive aspergillosis.
-
Review
that both PCR and galactomannan may enable a specificdiagnosis
to be established earlier than is possible using aconventional
approach.32,49,132 Fourth, the combination ofdifferent diagnostic
modalitieseg, concomitant meas-urement of galactomannan and
(1,3)--D glucanis astrategy that may optimise diagnostic
accuracy.63 Finally, itseems likely that both PCR and galactomannan
engenderimportant prognostic information; a falling galacto-mannan
titre or a positive-turning-negative PCR signal inthe context of
antifungal therapy is usually associated witha successful outcome.
However, at the current time,galactomannan and PCR have not been
systematicallyused to guide antifungal therapy.
The incorporation of diagnostic data into
managementstrategiesGalactomannan (and validated PCR) applied to
blood canbe used as screening tools to further improve
theidentification of patients at high risk of developinginvasive
aspergillosis.133 A positive result may enable thestart of early
targeted antifungal chemotherapy, whileexpensive and potentially
toxic antifungal drugs can bewithheld with persistently negative
results. Testing for(1,3)--D-glucan could be also be useful in this
regard.When the assays are used in this manner, a positive
resultshould also serve as a trigger for additional
diagnosticevaluationeg, a high-resolution computed tomographyscan
of the thoraxto investigate the possibility of asubclinical focus
of infection. The success of galac-tomannan (and validated PCR) as
a screening tool islargely dependent on the underlying prevalence
ofinvasive aspergillosis, which varies according to thespecific
host group and institution; thus, the requirementand extent of
galactomannan screening may varyaccordingly.
An alternate diagnostic strategy is to reservegalactomannan and
validated PCR for situations inwhich clinical and radiological data
are suggestive ofinvasive aspergillosis; in this scenario,
galactomannanand validated PCR applied to serum, and other
tissuesand fluids, may enable a definitive diagnosis of
invasiveaspergillosis to be secured. Although this approach doesnot
facilitate early antifungal therapy, it may minimisethe use of
invasive diagnostic modalities. Furthermore, amore definitive
diagnosis enables the administration ofspecific anti-aspergillus
therapy and would be of
considerable benefit for future diagnostic andtherapeutic
research.
Future challengesInvasive aspergillosis continues to pose many
challenges.From a diagnostic point of view, improving the
testaccuracy remains a priority for patient care,
therapeuticresearch, and future diagnostic research. The question,
ofcourse, is the manner in which these improvements canbe achieved.
The progressive refinement of existingtechniques and development of
new diagnostic technolo-gies is clearly a priority. Substantial
work remains in areasrelated to cost-effectiveness and whether
patients whoundergo intensive diagnostic testing have
improvedoutcome. Just as importantly, however, is the generationof
a clinical environment and culture that is amenable tohigh quality
diagnostic research, the provision of adequatefunding, multicentre
participation, international collabo-ration, and rigorous study
design.
Conflicts of interestWWH is supported by an unrestricted
educational grant from Merck & Coand the Fungal Research Trust.
TJW and DWD have no conflicts ofinterest to declare.
AcknowledgmentsWe thank Ruta Petraitiene for the
photomicrographs in figure 1.
References1 Perfect JR, Cox GM, Lee JY, et al. The impact of
culture isolation of
Aspergillus species: a hospital-based survey of aspergillosis.
Clin InfectDis 2001; 33: 182433.
2 Segal BH, DeCarlo ES, Kwon-Chung KJ, Malech HL, Gallin
JI,Holland SM. Aspergillus nidulans infection in chronic
granulomatousdisease. Medicine (Baltimore) 1998; 77: 34554.
3 von Eiff M, Roos N, Schulten R, Hesse M, Zuhlsdorf M, van de
Loo J.Pulmonary aspergillosis: early diagnosis improves
survival.Respiration 1995; 62: 34147.
4 Wald A, Leisenring W, van Burik JA, Bowden RA. Epidemiology
ofAspergillus infections in a large cohort of patients undergoing
bonemarrow transplantation. J Infect Dis 1997; 175: 145966.
5 Grow WB, Moreb JS, Roque D, et al. Late onset of invasive
Aspergillusinfection in bone marrow transplant patients at a
university hospital.Bone Marrow Transplant 2002; 29: 1519.
6 Alangaden GJ, Wahiduzzaman M, Chandrasekar PH.
Aspergillosis:the most common community-acquired pneumonia with
gram-negative bacilli as copathogens in stem cell transplant
recipients withgraft-versus-host disease. Clin Infect Dis 2002; 35:
65964.
7 Meersseman W, Vandecasteele SJ, Wilmer A, Verbeken
E,Peetermans WE, Van Wijngaerden E. Invasive aspergillosis
incritically ill patients without malignancy. Am J Respir Crit Care
Med2004; 170: 62125.
8 Hachem RY, Kontoyiannis DP, Boktour MR, et al. Aspergillus
terreus:an emerging amphotericin B-resistant opportunistic mold in
patientswith hematologic malignancies. Cancer 2004; 101:
1594600.
9 Steinbach WJ, Benjamin DK Jr, Kontoyiannis DP, et al.
Infectionsdue to Aspergillus terreus: a multicenter retrospective
analysis of 83cases. Clin Infect Dis 2004; 39: 19298.
10 Walsh TJ, Petraitis V, Petraitiene R, et al. Experimental
pulmonaryaspergillosis due to Aspergillus terreus: pathogenesis and
treatment ofan emerging fungal pathogen resistant to amphotericin
B. J Infect Dis2003; 188: 30519.
11 Ascioglu S, Rex JH, de Pauw B, et al. Defining opportunistic
invasivefungal infections in immunocompromised patients with cancer
andhematopoietic stem cell transplants: an international consensus.
ClinInfect Dis 2002; 34: 714.
12 Merz WG, Roberts GD. Algorithims for detection and
identificationof fungi. In: Murray PR, Baron EJ, Pfaller MA,
Tenover FC, YolkenRH, eds. Manual of clinical microbiology, 7th
edn. Washington DC:ASM Press, 1999: 116783.
http://infection.thelancet.com Vol 5 October 2005 619
Search strategy and selection criteria
PubMed was searched for English-language literature usingthe
following terms: Aspergillus, aspergillosis, diagnosis,fungus,
fungal, culture, histology, galactomannan,glucan, serology,
antibody, PCR, molecular,metabolite, mannitol, and gliotoxin.
Further relevantreferences, not identified by this strategy, were
retrievedfrom the primary publications.
-
Review
13 Denning DW, Kibbler CC, Barnes RA. British Society for
MedicalMycology proposed standards of care for patients with
invasivefungal infections. Lancet Infect Dis 2003; 3: 23040.
14 Grocott RG. A stain for fungi in tissue sections and smears
usingGomoris methenamine-silver nitrate technic. Am J Clin Pathol
1955;25: 97579.
15 Ruchel R, Schaffrinski M. Versatile fluorescent staining of
fungi inclinical specimens by using the optical brightener
Blankophor. J ClinMicrobiol 1999; 37: 269496.
16 Chander J, Chakrabarti A, Sharma A, Saini JS, Panigrahi
D.Evaluation of Calcofluor staining in the diagnosis of fungal
cornealulcer. Mycoses 1993; 36: 24345.
17 Choi JK, Mauger J, McGowan KL. Immunohistochemical
detectionof Aspergillus species in pediatric tissue samples. Am J
Clin Pathol2004; 121: 1825.
18 Kaufman L, Standard PG, Jalbert M, Kraft DE.
Immunohistologicidentification of Aspergillus spp. and other
hyaline fungi by usingpolyclonal fluorescent antibodies. J Clin
Microbiol 1997; 35:220609.
19 Verweij PE, Smedts F, Poot T, Bult P, Hoogkamp-Korstanje JA,
Meis JF. Immunoperoxidase staining for identification of
Aspergillusspecies in routinely processed tissue sections. J Clin
Pathol 1996; 49:798801.
20 De Vos MM, Nelis HJ. Detection of Aspergillus fumigatus
hyphae bysolid phase cytometry. J Microbiol Methods 2003; 55:
55764.
21 Zimmerman RL, Montone KT, Fogt F, Norris AH. Ultra
fastidentification of Aspergillus species in pulmonary cytology
specimensby in situ hybridization. Int J Mol Med 2000; 5:
42729.
22 Hayden RT, Isotalo PA, Parrett T, et al. In situ
hybridization for thedifferentiation of Aspergillus, Fusarium, and
Pseudallescheria species intissue section. Diagn Mol Pathol 2003;
12: 2126.
23 Kontoyiannis DP, Lewis RE, May GS, Osherov N, Rinaldi
MG.Aspergillus nidulans is frequently resistant to amphotericin B.
Mycoses2002; 45: 40607.
24 Tarrand JJ, Lichterfeld M, Warraich I, et al. Diagnosis of
invasiveseptate mold infections. Am J Clin Pathol 2003; 119:
85458.
25 Horvath JA, Dummer S. The use of respiratory-tract cultures
in thediagnosis of invasive pulmonary aspergillosis. Am J Med 1996;
100:17178.
26 Sutton DA. Specimen collection, transport, and
processing:mycology. In: Murray PR, Baron EJ, Jorgensen JH, Pfaller
MA,Yolken RH, eds. Manual of clinical microbiology, 8th
edn.Washington DC: ASM Press, 2003: 165967.
27 de Hoog GS, Guarro J, Gene J, Figueras MJ. Atlas of
clinicalfungi, 2nd edn. Utrecht: Centraalbureau voor
Schimmelcultures,2000.
28 Klich MA. Identification of common Aspergillus species.
Utrecht:Centraalbureau voor Schimmelcultures, 2002.
29 Callister ME, Griffiths MJ, Nicholson AG, et al. A fatal case
ofdisseminated aspergillosis caused by a non-sporulating strain
ofAspergillus fumigatus. J Clin Pathol 2004; 57: 99192.
30 Latge JP, Kobayashi H, Debeaupuis JP, et al. Chemical
andimmunological characterization of the extracellulargalactomannan
of Aspergillus fumigatus. Infect Immun 1994; 62:542433.
31 Mennink-Kersten MA, Donnelly JP, Verweij PE. Detection
ofcirculating galactomannan for the diagnosis and management
ofinvasive aspergillosis. Lancet Infect Dis 2004; 4: 34957.
32 Marr KA, Balajee SA, McLaughlin L, Tabouret M, Bentsen C,
Walsh TJ. Detection of galactomannan antigenemia by
enzymeimmunoassay for the diagnosis of invasive aspergillosis:
variablesthat affect performance. J Infect Dis 2004; 190:
64149.
33 Francis P, Lee JW, Hoffman A, et al. Efficacy of
unilamellarliposomal amphotericin B in treatment of pulmonary
aspergillosis inpersistently granulocytopenic rabbits: the
potential role ofbronchoalveolar D-mannitol and serum galactomannan
as markersof infection. J Infect Dis 1994; 169: 35668.
34 Bretagne S, Marmorat-Khuong A, Kuentz M, Latge JP,
Bart-Delabesse E, Cordonnier C. Serum Aspergillus
galactomannanantigen testing by sandwich ELISA: practical use in
neutropenicpatients. J Infect 1997; 35: 715.
35 Boutboul F, Alberti C, Leblanc T, et al. Invasive
aspergillosis inallogeneic stem cell transplant recipients:
increasing antigenemia isassociated with progressive disease. Clin
Infect Dis 2002; 34: 93943.
36 Becker MJ, Lugtenburg EJ, Cornelissen JJ, Van Der Schee
C,Hoogsteden HC, De Marie S. Galactomannan detection incomputerized
tomography-based broncho-alveolar lavage fluid andserum in
haematological patients at risk for invasive
pulmonaryaspergillosis. Br J Haematol 2003; 121: 44857.
37 Maertens J, Verhaegen J, Lagrou K, Van Eldere J, Boogaerts
M.Screening for circulating galactomannan as a noninvasive
diagnostictool for invasive aspergillosis in prolonged neutropenic
patients andstem cell transplantation recipients: a prospective
validation. Blood2001; 97: 160410.
38 Klont RR, Mennink-Kersten MA, Verweij PE. Utility of
Aspergillusantigen detection in specimens other than serum
specimens. ClinInfect Dis 2004; 39: 146774.
39 Stynen D, Goris A, Sarfati J, Latge JP. A new sensitive
sandwichenzyme-linked immunosorbent assay to detect galactofuran
inpatients with invasive aspergillosis. J Clin Microbiol 1995;
33:497500.
40 Verweij PE, Stynen D, Rijs AJ, de Pauw BE, Hoogkamp-Korstanje
JA,Meis JF. Sandwich enzyme-linked immunosorbent assay comparedwith
Pastorex latex agglutination test for diagnosing
invasiveaspergillosis in immunocompromised patients. J Clin
Microbiol 1995;33: 191214.
41 Dalle F, Lopez J, Caillot D, et al. False-positive results
caused by cotton swabs in commercial Aspergillus antigen latex
agglutination test. Eur J Clin Microbiol Infect Dis 2002; 21:
13032.
42 Swanink CM, Meis JF, Rijs AJ, Donnelly JP, Verweij
PE.Specificity of a sandwich enzyme-linked immunosorbent assay
fordetecting Aspergillus galactomannan. J Clin Microbiol 1997;
35:25760.
43 Kappe R, Schulze-Berge A. New cause for false-positive
results withthe Pastorex Aspergillus antigen latex agglutination
test. J ClinMicrobiol 1993; 31: 248990.
44 Hashiguchi K, Niki Y, Soejima R. Cyclophosphamide induces
false-positive results in detection of aspergillus antigen in
urine. Chest1994; 105: 97576.
45 Mennink-Kersten MA, Klont RR, Warris A, Op den Camp
HJ,Verweij PE. Bifidobacterium lipoteichoic acid and false
ELISAreactivity in aspergillus antigen detection. Lancet 2004;
363:32527.
46 Bennett JE, Kauffman C, Walsh T, et al. Forum report: issues
in theevaluation of diagnostic tests, use of historical controls,
and merits ofthe current multicenter collaborative groups. Clin
Infect Dis 2003; 36(suppl 3): S12327.
47 Herbrecht R, Letscher-Bru V, Oprea C, et al.
Aspergillusgalactomannan detection in the diagnosis of invasive
aspergillosis incancer patients. J Clin Oncol 2002; 20:
1898906.
48 Maertens J, Verhaegen J, Demuynck H, et al.
Autopsy-controlledprospective evaluation of serial screening for
circulatinggalactomannan by a sandwich enzyme-linked immunosorbent
assayfor hematological patients at risk for invasive aspergillosis.
J ClinMicrobiol 1999; 37: 322328.
49 Maertens J, Van Eldere J, Verhaegen J, Verbeken E,
Verschakelen J,Boogaerts M. Use of circulating galactomannan
screening for earlydiagnosis of invasive aspergillosis in
allogeneic stem cell transplantrecipients. J Infect Dis 2002; 186:
1297306.
50 Verweij PE, Weemaes CM, Curfs JH, Bretagne S, Meis JF.
Failure todetect circulating Aspergillus markers in a patient with
chronicgranulomatous disease and invasive aspergillosis. J Clin
Microbiol2000; 38: 390001.
51 Husain S, Kwak EJ, Obman A, et al. Prospective assessment
ofPlatelia Aspergillus galactomannan antigen for the diagnosis
ofinvasive aspergillosis in lung transplant recipients. Am J
Transplant2004; 4: 796802.
52 Kwak EJ, Husain S, Obman A, et al. Efficacy of
galactomannanantigen in the Platelia Aspergillus enzyme immunoassay
for diagnosisof invasive aspergillosis in liver transplant
recipients. J Clin Microbiol2004; 42: 43538.
53 Fortun J, Martin-Davila P, Alvarez ME, et al. Aspergillus
antigenemiasandwich-enzyme immunoassay test as a serodiagnostic
method forinvasive aspergillosis in liver transplant recipients.
Transplantation2001; 71: 14549.
54 Marr KA, Laverdiere M, Gugel A, Leisenring W. Antifungal
therapydecreases sensitivity of the Aspergillus galactomannan
enzymeimmunoassay. Clin Infect Dis 2005; 40: 176269.
620 http://infection.thelancet.com Vol 5 October 2005
-
Review
55 Maertens J, Theunissen K, Verbeken E, et al. Prospective
clinicalevaluation of lower cut-offs for galactomannan detection in
adultneutropenic cancer patients and haematological stem cell
transplantrecipients. Br J Haematol 2004; 126: 85260.
56 Siemann M, Koch-Dorfler M, Gaude M. False-positive results
inpremature infants with the Platelia Aspergillus sandwich
enzyme-linked immunosorbent assay. Mycoses 1998; 41: 37377.
57 Ansorg R, van den Boom R, Rath PM. Detection of
Aspergillusgalactomannan antigen in foods and antibiotics. Mycoses
1997; 40:35357.
58 Viscoli C, Machetti M, Cappellano P, et al.
False-positivegalactomannan platelia Aspergillus test results for
patients receivingpiperacillin-tazobactam. Clin Infect Dis 2004;
38: 91316.
59 Sulahian A, Touratier S, Ribaud P. False positive test for
Aspergillusantigenemia related to concomitant administration of
piperacillinand tazobactam. N Engl J Med 2003; 349: 236667.
60 Walsh TJ, Shoham S, Petraitiene R, et al. Detection
ofgalactomannan antigenemia in patients receiving
piperacillin-tazobactam and correlations between in vitro, in vivo,
and clinicalproperties of the drug-antigen interaction. J Clin
Microbiol 2004; 42:474448.
61 Singh N, Obman A, Husain S, Aspinall S, Mietzner S, Stout
JE.Reactivity of platelia Aspergillus galactomannan antigen
withpiperacillin-tazobactam: clinical implications based on
achievableconcentrations in serum. Antimicrob Agents Chemother
2004; 48:198992.
62 FDA. Practice Caution in Using Bio-Rad Platelia Aspergillus
EIA.http://www.fda.gov/cdrh/oivd/laboratory.html#tip8 (accessed
Aug17, 2005).
63 Pazos C, Ponton J, Del Palacio A. Contribution of
(1->3)-beta-D-glucan chromogenic assay to diagnosis and
therapeutic monitoringof invasive aspergillosis in neutropenic
adult patients: a comparisonwith serial screening for circulating
galactomannan. J Clin Microbiol2005; 43: 299305.
64 Obayashi T, Yoshida M, Mori T, et al. Plasma
(1>3)-beta-D-glucanmeasurement in diagnosis of invasive deep
mycosis and fungalfebrile episodes. Lancet 1995; 345: 1720.
65 Mori T, Ikemoto H, Matsumura M, et al. Evaluation of plasma
(1-3)-beta-D-glucan measurement by the kinetic turbidimetric
Limulustest, for the clinical diagnosis of mycotic infections. Eur
J Clin ChemClin Biochem 1997; 35: 55360.
66 Miyazaki T, Kohno S, Mitsutake K, et al. Plasma
(1>3)-beta-D-glucan and fungal antigenemia in patients with
candidemia,aspergillosis, and cryptococcosis. J Clin Microbiol
1995; 33:311518.
67 Odabasi Z, Mattiuzzi G, Estey E, et al. Beta-D-glucan as a
diagnosticadjunct for invasive fungal infections: validation,
cutoff,development, and performance in patients with acute
myelogenousluekemia and myelodysplastic syndrome. Clin Infect Dis
2004; 39:199205.
68 Rylander R, Norrhall M, Engdahl U, Tunsater A, Holt
PG.Airways inflammation, atopy, and (1>3)-beta-D-glucanexposures
in two schools. Am J Respir Crit Care Med 1998; 158:168587.
69 Denning DW, Riniotis K, Dobrashian R, Sambatakou H.
Chroniccavitary and fibrosing pulmonary and pleural aspergillosis:
caseseries, proposed nomenclature change, and review. Clin Infect
Dis2003; 37 (suppl 3): S26580.
70 Young RC, Bennett JE. Invasive aspergillosis. Absence of
detectableantibody response. Am Rev Respir Dis 1971; 104:
71016.
71 Chan CM, Woo PC, Leung AS, et al. Detection of antibodies
specificto an antigenic cell wall galactomannoprotein for
serodiagnosis ofAspergillus fumigatus aspergillosis. J Clin
Microbiol 2002; 40:204145.
72 Sambatakou H, Guiver M, Denning D. Pulmonary aspergillosis in
apatient with chronic granulomatous disease: confirmation
bypolymerase chain reaction and serological tests, and
successfultreatment with voriconazole. Eur J Clin Microbiol Infect
Dis 2003; 22:68185.
73 Kappe R, Schulze-Berge A, Sonntag HG. Evaluation of eight
antibodytests and one antigen test for the diagnosis of invasive
aspergillosis.Mycoses 1996; 39: 1323.
74 Schonheyder H. Pathogenic and serological aspects of
pulmonaryaspergillosis. Scand J Infect Dis Suppl 1987; 51: 162.
75 Centeno-Lima S, de Lacerda JM, do Carmo JA, Abecasis
M,Casimiro C, Exposto F. Follow-up of anti-Aspergillus IgG and
IgAantibodies in bone marrow transplanted patients with
invasiveaspergillosis. J Clin Lab Anal 2002; 16: 15662.
76 Holdom MD, Lechenne B, Hay RJ, Hamilton AJ, Monod
M.Production and characterization of recombinant
Aspergillusfumigatus Cu,Zn superoxide dismutase and its recognition
byimmune human sera. J Clin Microbiol 2000; 38: 55862.
77 Weig M, Frosch M, Tintelnot K, et al. Use of
recombinantmitogillin for improved serodiagnosis of Aspergillus
fumigatus-associated diseases. J Clin Microbiol 2001; 39:
172130.
78 Yuen KY, Chan CM, Chan KM, et al. Characterization of AFMP1:
anovel target for serodiagnosis of aspergillosis. J Clin Microbiol
2001;39: 383037.
79 Latge JP. Aspergillus fumigatus and aspergillosis. Clin
Microbiol Rev1999; 12: 31050.
80 Strandberg GW. D-mannitol metabolism by Aspergillus candidus.
J Bacteriol 1969; 97: 130509.
81 Wong B, Brauer KL, Tsai RR, Jayasimhulu K. Increased
amountsof the Aspergillus metabolite D-mannitol in tissue and serum
ofrats with experimental aspergillosis. J Infect Dis 1989;
160:95103.
82 Lewis RE, Wiederhold NP, Chi J, et al. Detection of gliotoxin
inexperimental and human aspergillosis. Infect Immun 2005;
73:63537.
83 Yoo JH, Choi JH, Choi SM, et al. Application of nucleic
acidsequence-based amplification for diagnosis of and monitoring
the clinical course of invasive aspergillosis in patients with
hematologic diseases. Clin Infect Dis 2005; 40: 39298.
84 Loeffler J, Hebart H, Cox P, Flues N, Schumacher U, Einsele
H.Nucleic acid sequence-based amplification of Aspergillus RNA
inblood samples. J Clin Microbiol 2001; 39: 162629.
85 Bretagne S. Molecular diagnostics in clinical parasitology
andmycology: limits of the current polymerase chain reaction
(PCR)assays and interest of the real-time PCR assays. Clin
Microbiol Infect2003; 9: 50511.
86 Sanguinetti M, Posteraro B, Pagano L, et al. Comparison of
real-time PCR, conventional PCR, and galactomannan antigendetection
by enzyme-linked immunosorbent assay usingbronchoalveolar lavage
fluid samples from hematology patients fordiagnosis of invasive
pulmonary aspergillosis. J Clin Microbiol2003; 41: 392225.
87 Rantakokko-Jalava K, Laaksonen S, Issakainen J, et
al.Semiquantitative detection by real-time PCR of
Aspergillusfumigatus in bronchoalveolar lavage fluids and tissue
biopsyspecimens from patients with invasive aspergillosis. J
ClinMicrobiol 2003; 41: 430411.
88 Hendolin PH, Paulin L, Koukila-Kahkola P, et al. Panfungal
PCRand multiplex liquid hybridization for detection of fungi in
tissuespecimens. J Clin Microbiol 2000; 38: 418692.
89 Willinger B, Obradovic A, Selitsch B, et al. Detection
andidentification of fungi from fungus balls of the maxillary sinus
bymolecular techniques. J Clin Microbiol 2003; 41: 58185.
90 Paterson PJ, Seaton S, McLaughlin J, Kibbler CC. Development
ofmolecular methods for the identification of aspergillus
andemerging moulds in paraffin wax embedded tissue sections.
MolPathol 2003; 56: 36870.
91 Costa C, Costa JM, Desterke C, Botterel F, Cordonnier
C,Bretagne S. Real-time PCR coupled with automated DNAextraction
and detection of galactomannan antigen in serum byenzyme-linked
immunosorbent assay for diagnosis of invasiveaspergillosis. J Clin
Microbiol 2002; 40: 222427.
92 Loeffler J, Hebart H, Brauchle U, Schumacher U, Einsele
H.Comparison between plasma and whole blood specimens fordetection
of Aspergillus DNA by PCR. J Clin Microbiol 2000; 38:383033.
93 Garcia ME, Blanco JL, Caballero J, Gargallo-Viola
D.Anticoagulants interfere with PCR used to diagnose
invasiveaspergillosis. J Clin Microbiol 2002; 40: 156768.
94 Loffler J, Hebart H, Schumacher U, Reitze H, Einsele
H.Comparison of different methods for extraction of DNA of
fungalpathogens from cultures and blood. J Clin Microbiol 1997;
35:331112.
http://infection.thelancet.com Vol 5 October 2005 621
-
Review
95 Muller FM, Werner KE, Kasai M, Francesconi A, Chanock SJ,
Walsh TJ. Rapid extraction of genomic DNA from medicallyimportant
yeasts and filamentous fungi by high-speed celldisruption. J Clin
Microbiol 1998; 36: 162529.
96 Loeffler J, Hebart H, Bialek R, et al. Contaminations
occurring infungal PCR assays. J Clin Microbiol 1999; 37:
120002.
97 Nierman W, Pain A, Anderson MJ, et al. Genomic sequence of
thepathogenic and allergenic Aspergillus fumigatus. Nature (in
press).
98 Chen SC, Halliday CL, Meyer W. A review of nucleic
acid-baseddiagnostic tests for systemic mycoses with an emphasis
onpolymerase chain reaction-based assays. Med Mycol 2002; 40:
33357.
99 Spiess B, Buchheidt D, Baust C, et al. Development of a
LightCyclerPCR assay for detection and quantification of
Aspergillus fumigatusDNA in clinical samples from neutropenic
patients. J Clin Microbiol2003; 41: 181118.
100 Melchers WJ, Verweij PE, van den Hurk P, et al. General
primer-mediated PCR for detection of Aspergillus species. J Clin
Microbiol1994; 32: 171017.
101 Spreadbury C, Holden D, Aufauvre-Brown A, Bainbridge B,
Cohen J.Detection of Aspergillus fumigatus by polymerase chain
reaction. J Clin Microbiol 1993; 31: 61521.
102 Walsh TJ, Francesconi A, Kasai M, Chanock SJ. PCR and
single-strand conformational polymorphism for recognition of
medicallyimportant opportunistic fungi. J Clin Microbiol 1995; 33:
321620.
103 Rath PM, Ansorg R. Identification of medically important
Aspergillusspecies by single strand conformational polymorphism
(SSCP) of thePCR-amplified intergenic spacer region. Mycoses 2000;
43: 38186.
104 Hopfer RL, Walden P, Setterquist S, Highsmith WE. Detection
anddifferentiation of fungi in clinical specimens using polymerase
chainreaction (PCR) amplification and restriction enzyme analysis.
J MedVet Mycol 1993; 31: 6575.
105 Martin C, Roberts D, van Der Weide M, et al. Development of
a PCR-based line probe assay for identification of fungal
pathogens. J ClinMicrobiol 2000; 38: 373542.
106 Turenne CY, Sanche SE, Hoban DJ, Karlowsky JA, Kabani
AM.Rapid identification of fungi by using the ITS2 genetic region
and anautomated fluorescent capillary electrophoresis system. J
ClinMicrobiol 1999; 37: 184651.
107 de Aguirre L, Hurst SF, Choi JS, Shin JH, Hinrikson HP,
Morrison CJ. Rapid differentiation of Aspergillus species from
othermedically important opportunistic molds and yeasts by
PCR-enzymeimmunoassay. J Clin Microbiol 2004; 42: 3495504.
108 Forbes BA. Introducing a molecular test into the
clinicalmicrobiology laboratory: development, evaluation, and
validation.Arch Pathol Lab Med 2003; 127: 110611.
109 Dimech W, Bowden S, Brestovac B, et al. Validation of
assemblednucleic acid-based tests in diagnostic microbiology
laboratories.Pathology 2004; 36: 4550.
110 Einsele H, Hebart H, Roller G, et al. Detection and
identification offungal pathogens in blood by using molecular
probes. J ClinMicrobiol 1997; 35: 135360.
111 Kami M, Fukui T, Ogawa S, et al. Use of real-time PCR on
bloodsamples for diagnosis of invasive aspergillosis. Clin Infect
Dis 2001;33: 150412.
112 Bretagne S, Costa JM, Marmorat-Khuong A, et al. Detection
ofAspergillus species DNA in bronchoalveolar lavage samples
bycompetitive PCR. J Clin Microbiol 1995; 33: 116468.
113 Bretagne S, Costa JM, Bart-Delabesse E, Dhedin N, Rieux
C,Cordonnier C. Comparison of serum galactomannan antigendetection
and competitive polymerase chain reaction for diagnosinginvasive
aspergillosis. Clin Infect Dis 1998; 26: 140712.
114 Verweij PE, Meis JF, van den Hurk P, Zoll J, Samson RA,
Melchers WJ. Phylogenetic relationships of five species of
Aspergillusand related taxa as deduced by comparison of sequences
of smallsubunit ribosomal RNA. J Med Vet Mycol 1995; 33: 18590.
115 OSullivan CE, Kasai M, Francesconi A, et al. Development
andvalidation of a quantitative real-time PCR assay using
fluorescenceresonance energy transfer technology for detection of
Aspergillusfumigatus in experimental invasive pulmonary
aspergillosis. J Clin Microbiol 2003; 41: 567682.
116 Bowman JC, Abruzzo GK, Anderson JW, et al. Quantitative
PCRassay to measure Aspergillus fumigatus burden in a murine model
ofdisseminated aspergillosis: demonstration of efficacy of
caspofunginacetate. Antimicrob Agents Chemother 2001; 45:
347481.
117 Henry T, Iwen PC, Hinrichs SH. Identification of Aspergillus
speciesusing internal transcribed spacer regions 1 and 2. J Clin
Microbiol2000; 38: 151015.
118 Luo G, Mitchell TG. Rapid identification of pathogenic fungi
directlyfrom cultures by using multiplex PCR. J Clin Microbiol
2002; 40:286065.
119 Zhao J, Kong F, Li R, Wang X, Wan Z, Wang D. Identification
ofAspergillus fumigatus and related species by nested PCR
targetingribosomal DNA internal transcribed spacer regions. J Clin
Microbiol2001; 39: 226166.
120 Herbrecht R, Denning DW, Patterson TF, et al. Voriconazole
versusamphotericin B for primary therapy of invasive aspergillosis.
N Engl JMed 2002; 347: 40815.
121 Loeffler J, Kloepfer K, Hebart H, et al. Polymerase chain
reactiondetection of aspergillus DNA in experimental models of
invasiveaspergillosis. J Infect Dis 2002; 185: 120306.
122 Albelda SM, Talbot GH, Gerson SL, Miller WT, Cassileth PA.
Role offiberoptic bronchoscopy in the diagnosis of invasive
pulmonaryaspergillosis in patients with acute leukemia. Am J Med
1984; 76:102734.
123 Kahn FW, Jones JM, England DM. The role of
bronchoalveolarlavage in the diagnosis of invasive pulmonary
aspergillosis. Am J ClinPathol 1986; 86: 51823.
124 Einsele H, Quabeck K, Muller KD, et al. Prediction of
invasivepulmonary aspergillosis from colonisation of lower
respiratory tractbefore marrow transplantation. Lancet 1998; 352:
1443.
125 Levy H, Horak DA, Tegtmeier BR, Yokota SB, Forman SJ. The
valueof bronchoalveolar lavage and bronchial washings in the
diagnosis ofinvasive pulmonary aspergillosis. Respir Med 1992; 86:
24348.
126 von Eiff M, Zuhlsdorf M, Roos N, Hesse M, Schulten R, van de
Loo J.Pulmonary fungal infections in patients with
hematologicalmalignanciesdiagnostic approaches. Ann Hematol 1995;
70:13541.
127 Crystal RG, Reynolds HY, Kalica AR. Bronchoalveolar lavage.
Thereport of an international conference. Chest 1986; 90:
12231.
128 McWhinney PH, Kibbler CC, Hamon MD, et al. Progress in
thediagnosis and management of aspergillosis in bone
marrowtransplantation: 13 years experience. Clin Infect Dis 1993;
17:397404.
129 Brown MJ, Worthy SA, Flint JD, Muller NL. Invasive
aspergillosis inthe immunocompromised host: utility of computed
tomography andbronchoalveolar lavage. Clin Radiol 1998; 53:
25557.
130 Musher B, Fredricks D, Leisenring W, Balajee SA, Smith C,
Marr KA. Aspergillus galactomannan enzyme immunoassay
andquantitative PCR for diagnosis of invasive aspergillosis
withbronchoalveolar lavage fluid. J Clin Microbiol 2004; 42:
551722.
131 Yu VL, Muder RR, Poorsattar A. Significance of isolation
ofAspergillus from the respiratory tract in diagnosis of
invasivepulmonary aspergillosis. Results from a three-year
prospective study.Am J Med 1986; 81: 24954.
132 Hebart H, Loffler J, Meisner C, et al. Early detection of
Aspergillusinfection after allogeneic stem cell transplantation by
polymerasechain reaction screening. J Infect Dis 2000; 181:
171319.
133 Jones BL, McLintock LA. Impact of diagnostic markers on
earlyantifungal therapy. Curr Opin Infect Dis 2003; 16: 52126.
622 http://infection.thelancet.com Vol 5 October 2005
-
Reproduced with permission of the copyright owner. Further
reproduction prohibited without permission.