Detection of the dengue nonstructural 1 antigen in cerebral spinal fluid samples using a commercially available enzymelinked immunosorbent assay

JournalofVirologicalMethods177 (2011) 128–131

ContentslistsavailableatScienceDirect

Journal

of

Virological

Methods

jo u r n al h om epa ge :w w w . e l s e v i e r . c o m / l o c a t e / j v i r o m e t

Short

communication

Detection

of

the

dengue

non-structural

1

antigen

in

cerebral

spinal

fluid

samples

using

a

commercially

available

enzyme-linked

immunosorbent

assay

F.M.C.

Araújo

,

R.S.N.

Brilhante

,

L.P.G.

Cavalcanti

,

M.F.G.

Rocha

,

R.A.

Cordeiro

,

A.C.B.

Perdigão

,

I.S.

Miralles

,

L.C.

Araújo

,

R.M.C.

Araújo

,

E.G.

Lima

,

J.J.C.

Sidrim

a_{CentralPublicHealthLaboratoryofCeará,Brazil} b_{StateHealthSecretariatofCeará,Brazil}

c_{PostgraduatePrograminMedicalSciences,FederalUniversityofCeará,Brazil} d_{SpecializedCenterofMedicalMycology,Ceará,Brazil}

e_{NortheastBiotechnologyNetwork,StateUniversityofCeará,Brazil} f_{FederalUniversityofCeará,Brazil}

Articlehistory: Received5January2011

Receivedinrevisedform8April2011 Accepted11July2011

Available online 20 July 2011

Keywords: Denguevirus Diagnosis NS1antigen Cerebralspinalfluid ELISA

a

b

s

t

r

a

c

t

Theinvolvementofthecentralnervoussystemindengueinfectionshasbeenreportedincountries wherethediseaseinendemic.Thepurposeofthisstudywastodeterminewhetheranenzyme-linked immunosorbentassaykitdesignedtodetectthedengueNS1antigeninserumwasabletodetectthis antigenincerebralspinalfluid(CSF)samplesfrompatientswithfataloutcomes.Toevaluatethesensitivity ofthekit,26dengue-positiveCSFsampleswereused.ThePan-EDengueEarlykitwasabletodetectthe NS1antigenin13of26dengue-positiveCSFsamples,resultinginasensitivityof50%(95%confidence interval,29.9–70.1%)andspecificityof100%(95%confidenceinterval,75.3–100%).Thekitwasableto detecttheNS1antigeninCSFofindividualswhohaddiedofdengue.Whenusedincombinationwith IgM,thedetectionrateroseto92.3%.Thisstudyreportsamethodforrapidlydetectingthedenguevirus inCSF,therebyincreasingthediagnosisofdenguefevercaseswithunusualneurologicalmanifestations.

© 2011 Elsevier B.V. All rights reserved.

Denguefever is a mosquito-bornedisease caused byone of fourdengueviruses(DENV),DENV-1,DENV-2,DENV-3and DENV-4,whichbelongtotheFlavivirusgenusoftheFlaviviridaefamily (Gubler,2002;ICTV,2006).Itsprevalencehasbeenincreasingin recentdecades.Currently itisendemicinover100countriesin Africa,theAmericas,EasternMediterranean,SoutheastAsiaand WesternPacific(WHO,2009).

InBrazil,dengueepidemicshavebeendescribedsince1986, ini-tiallywithinvolvementofDENV-1,followedbyDENV-2in1990, and DENV-3 after 2000(Nogueira et al., 2007).In thestate of Cearáin NortheastBrazil,thediseaseisendemic andis caused byserotypes1,2and3,withcasesreportedeveryyearand peri-odicepidemics.Thehighestnumberofcaseswasobservedin1987, 1994,2001and2008.In2003,asevereDENV-3epidemicoccurred, anddenguehemorrhagicfever(DHF)incidencewashighamong adults(Cavalcantietal.,2010).

Dengue’s symptoms can vary from mild fever, the most common form,to potentially fatal forms, suchas dengue DHF fever and dengue shock syndrome (DSS). Unusual

manifesta-∗_{Correspondingauthorat:Av.BarãodeStudart2405,PostalCode:60120-002,}

Fortaleza,CE,Brazil.Tel.:+558531011498;fax:+558531011493. E-mailaddress:[email protected](F.M.C.Araújo).

tions, such as myocardiopathy, hepatic insufficiency, fulminant hepatitis,encephalopathyandencephalitis,havealsobecome com-mon (Nogueira et al.,2007).Thesefindings aredue perhaps to the increase of dengue diagnosis. Theinvolvement of the cen-tralnervoussystem(CNS)indenguepatientshasbeenreported incountrieswherethediseaseisendemic.Thereisnoantiviral therapy orvaccine approvedfor useagainstdengue, sopatient managementrequiresgoodlaboratoryandclinicalsupport. Spe-cificandrapiddiagnosiscanhelpdirectingthepropertreatmentto patients(WHO,2009).

TheDENVhasthreegenesthatencodeforstructuralproteins: envelope(E),membrane(M)andcapsid(C),andsevengenesthat encodefornon-structuralproteins:NS1,NS2A,NS2B,NS3,NS4A, NS4BandNS5.TheglycoproteinNS1ishighlyconserved,secreted bycellsinfectedwithDENVinvitroandinvivo,butits biologi-calactivityisstillnotwellunderstood.Duringinvitroinfection, NS1proteinisexpressedinaformassociatedwiththe intracellu-larmembrane,whichisessentialforviralreplication,orassociated withthecellsurface,whichcanbeinvolvedin signal transduc-tion(Mackenzieetal.,1996).NS1protein,insolution,circulates and accumulatesin theplasma ofpatientsinfected withDENV throughout the clinical phase, and can be correlated with the developmentofmoreseriousformsofthedisease(Libratyetal., 2002).

F.M.C.Araújoetal./JournalofVirologicalMethods177 (2011) 128–131 129

Table1

DetectionrateforthemethodsusedtodiagnosedengueaccordingtotheavailablesamplesofpatientswithDENVinfection.

Clinicalsample RT-PCRa_{P/S(%) VirusIsolationP}b_/Sc_{(%) Serotypedetected IgM}d_{P/S(%) NS1Ag}e_P/S(%)

CSFg _{9/26 3/26 3DENV-2 19/26 13/26}

(34.6) (11.5) 4DENV-3 (73.1) (50)

2DENV-1

Bloodorserum 2/20 2/26 2DENV-31 4/11 –f

(10.0) (7.7) DENV-1 (36.4)

Total 11/46 5/52 3DENV-2 23/37 13/26

(23.9) (9.6) 6DENV-33DENV-1 (62.2) (50)

a_{Reversetranscriptase-polymerasechainreaction(RT-PCR).} b _Positive(P).

c _Studied(S).

d _{ImmunoglobulinM(IgM).} e_{Non-structuralantigen(NS1Ag).} f _{Notdone(–).}

g_{Cerebralspinalfluid(CSF).}

SeveralnewcommercialassaysfordetectionoftheNS1 anti-genhavebeendeveloped.Theemploymentoftheenzyme-linked immunosorbentassay(ELISA)todetectNS1proteinfromDENVin serumorplasmaofpatientswithacutediseaseservesasa supple-mentarymethodforuseinassociationwithotherdiagnostictests (Dussartetal.,2008;Blackselletal.,2008).However,thereareno reportsoftheuseofthesetestsonthecerebrospinalfluid(CSF).This applicationwouldbeveryusefultodiagnosecasesofnon-classical dengue.

Thepurposeofthisstudywastoevaluatetheperformanceofthe immunosorbentassayPan-EDengueEarlyELISA(Panbio Diagnos-tics,Brisbane,Australia)(NS1Early),indetectingthepresenceof theNS1antigen(NS1Ag)inCSFsamplesobtainedduringautopsies. Epidemiologicalandclinicaldatawereobtainedfromanational databasesystemthatprovidesinformationonageandsexofeach patient,CSFcollectionandthedatewhensymptomsoccurred. Clin-icaldataandCSFsampleswerecollectedfrompatientswhodiedof adengue-likeillness,andwereautopsiedatthemunicipalcoroner’s officeinthecontextofthedenguesurveillanceactivityoftheCeará StateHealthSecretariat.CSFspecimens frompeoplewithother diagnoseddiseases(HIVinfection,leptospirosis,visceral leishma-niasis,pneumonia, fungalmeningitis, meningococcalmeningitis andotherbacterialformsofmeningitis)wereusedtoevaluatethe specificityoftheassaykit.ContaminationofCSFwithbloodwasan exclusioncriterionforthesamples.

CSF sampleswere tested for thepresence of DENV by viral isolation in C6/36 cells, as previously described (Gubler et al., 1984);genomedetectionwasbyRT-PCR(Lanciottietal.,1992); andIgMdetectionwasbyIgM-capturewiththeEnzyme-Linked ImmunosorbentAssay (ELISA) (Panbio, Brisbane, Australia). All assayswereconductedaccordingtothemanufactures‘instructions, usinga1:2sampledilution(Soaresetal.,2006).Caseswere consid-eredpositiveforDENVinfectionwhenCSFsampleswerepositive accordingtooneormore ofthesetestsand patientspresented dengue-likesyndrome.CaseswereconsiderednegativewhenCSF sampleswerenegativeforallofthesetestsandwerediagnosedas sufferingfromanotherdisease.

Thestudywasapprovedbytheethicsandresearchcommittee ofSãoJoséInfectiousDiseaseHospital(protocol005/2009,number CAAE:0005.0.042.000-09).

TheNS1 Early systemis anantigen-capture ELISAthat pro-videsqualitative,non-serotype-specificdetectionofDENVNS1Ag. Alltestswereperformedinaccordance withthemanufacturers’ instructions.Briefly,100␮Lofdiluted(1:2)samples,positive con-trol,negativecontrolandcalibratorwereaddedtomicrowellsthat werepre-coatedwithapolyclonalcaptureanti-NS1antibodyand thenincubatedfor1h,at37◦_{C.Eachplatewaswashedsixtimes} and incubated for 1hat, 37◦_{C, following the addition of}

HRP-conjugatedanti-NS1Mab.Aftersixwashes,antibodycomplexes weredetectedbyaddingTMBandincubatingtheplatefor10min atroomtemperature.Thereactionwasstoppedbyaddingastop solution,andtheplatewasreadat450nm,witha630nmreference filter.Asampleratiowasdeterminedforeachsamplebydividing theaverageopticaldensity(OD)ofthetestsamplebythe aver-ageODofthecutoffcontrol.Sampleratiosof<0.5,0.5to<1.0,and ≥1wereindicativeofnegative,inconclusiveandpositiveresults, respectively.Inconclusivesampleswereconsiderednegativeafter thetestswererepeatedandremainedinconclusive.

Toevaluatetheperformanceoftheteststatisticalmeasuresof sensitivityand specificity,aconfidence interval(CI)of95% was used.

CSF samples from 26 patients who presented fatal dengue infectionand60negativefordenguewereselectedtoassessthe performanceoftheNS1Earlykit.Allthese26positivesampleswere positivefordengueinCSFpreviouslybyatleastoneother diagnos-ticmethod.Thekitshowed73negativeand13positiveresultsfor thepresenceofNS1Ag.The13positivesampleswerepreviously positiveforDENVinCSF.Thedetectionrateforthemethodsusedto diagnosedengue,accordingtothesamplesofpatientswithDENV infection,isshowninTable1.

Theresultsofsensitivityandspecificityinrelationtothe detec-tionofNS1AginCSFcanbeseeninTable2.TwoCSFsamplesthat gaveambiguousresultswereretestedandconsiderednegative.The percentageofdetectionoftheNS1EarlytestinCSFofpatientswith laboratorydiagnosisofdenguefever,accordingtotheIgM detec-tionmethodology,isshowninTable3.Thedetectionrateofthe combinationofIgMantibodiesandNS1Agwas92.3%(Table4).

Inpatientswithdengue,theaveragebetweentheonsetof symp-tomstodeathwassixdays,rangingfrom1to14days.Thesignsand symptomspresentedwere:fever(77%);headache(42.3%); vomit-ing(30.7%);asthenia(26.9%);myalgia(26.9%);dyspnea(19.2%); mental confusion (19.2%); abdominal pain (15.4%); agitation (11.5%);hemorrhage (11.5%); somnolence(7.7%); splenomegaly (7.7%);chills (7.7%);cough(7.7%);diarrhea(7.7%);coma(7.7%); and,atalowerpercentage(3.8%),anorexia,abdominalbloating,

Table2

DiagnosticaccuracyscorefortheNS1detectionbyELISAkitinCSFsamples.

Diagnosis Positive Negative Total

Dengue 13 13 26

Nondengue 0 60 60

Total 13 73 86

Sensitivity 50%(95%CIa_{-29.9–70.1)} Specificity 100%(95%CI-94.0–100)

130 F.M.C.Araújoetal./JournalofVirologicalMethods177 (2011) 128–131

Table3

DetectionrateofNS1AgaccordingtothepresenceofIgMin26CSFsamples.

PresenceofIgMa _{AbsenceofIgM Total}

NS1+b _{9 4 13}

NS1−c 10 3 13

Total 19 7 26

Detection(%) 9/26(34.6) 4/26(15.4) 13/26(50)

a_{ImmunoglobulinM(IgM).}

b_{Non-structural1antigenpositive(NS1+).} c _{Non-structural1antigennegative(NS1−).}

limbstiffness,earinfection,cyanosis,nausea,chestpain, pancy-topenia,hypotension,sweating,acuterenalfailure,septicshock, intra-orbital swelling, dizziness, paresthesia, hallucinations and convulsion.

Ofthe26denguepatients,9hadco-morbidities:onehada res-piratoryinfection,threehadbacterialmeningitis,onehadWilson’s disease,onehadinfectiousenteritis,onehadpneumonia,onehad anearinfectionandonehadavaricella-zostervirusinfection.Six patientswereclassifiedasDHF,oneasDSSandtheother19,who didnotmeettheWorldHealthOrganizationcriterionforDHF clas-sification,wereconsideredashavingseveredengue.

Themeanageofpatientswas27years,rangingfrom2months to84years,and26.9%(7/24)ofthemwereyoungerthan7yearsold. Femalesaccountedfor62.5%ofthecasesandmalesfortheother 37.5%.

Theincreasingtransmissionofthedengueviruscontinuestobe aglobalpublichealthproblem,especiallyindevelopingcountries, where accesstopreventiveand diagnosticresourcesis limited. SincetheappearanceofthisvirusinBrazil,specificallyinthestate ofCeará,in1986,thenumberofseriouscasesofthediseasehas beenincreasingsteadilyinlinewithothertropicaland subtropi-calregionsoftheworld,wherethediseasehasbecomeendemic, withcyclicalvariation:yearsofsubstantialepidemicsfollowedby non-epidemicyears(WHO,2009).

Thesevereformsofthediseasehavecausedunusual manifesta-tionsofdengue,suchasneurologicalsigns.DENVvirus,thoughnot consideredneurotropic,hasbeenisolatedoritsviralantigenshave beenobservedinhumanCNS(Miagostovichetal.,1997;Ramos etal.,1998).Thisalsohasbeendemonstratedbyviralisolationand detectionofviralRNAinCSFanddemonstrationofintrathecal syn-thesisofspecificdengueantibodies(Chenetal.,1991;Nogueira etal.,2002;Puccioni-Sholeretal.,2009).However,Rosenetal. (1999),couldnotevidencevirusreplicationinthebraininfatal humaninfections.

TheresultsofthisstudyshowthatNS1AgcanbedetectedinCSF byusingacommerciallyavailablekit:NS1Early,eventhoughthis kitwasnotstandardizedfortestingCSF.Thereisnokitavailable designedtodetectIgMantibodiesinCSFeither.Instead,this detec-tionisgenerallydoneusingakitwithstandardizedserum.Soares etal.(2006)foundthisantibodyfunctioningwithCSFina1:2 dilu-tion.However,todetectIgMthepatientmusthavehadthedisease formorethanfivedays,whereasviralantigenscanbedetectedonly duringviremia,thusenablingdiagnosis.

Table4

DetectionofIgMandNS1AginCSFsamples.

Test Positive Negative Detection(%)

IgMa _{19 7 73.1}

NS1b _{13 13 50}

IgM-NS1(combined)c _{24 2 92.3}

a_{ImmunoglobulinM(IgM).} b_{Non-structuralantigen(NS1Ag).} c _{IgMand/orNS1Agpositive(combined).}

NS1Agwasdetectedin50%ofthecaseswhereCSFhadbeen foundpositivebyothermethods.ThesensitivityofserumNS1Ag detectionbythiskit rangesfrom60.4%(Dussartet al.,2008)to 91.6%(Serakanetal.,2007).ItisknownthatthequantityofIgM inCSFislowerthaninserum(Chenetal.,1991;Puccioni-Sholer etal.,2009).Maybethesamephenomenonoccursinthepresence ofNS1AginCSF.Takingintoaccountthattherearenopublished parametersforcomparingthisfindingofNS1AginCSF,itshould beconsideredonemoremethodforthediagnosisofdengueinthis typeofmaterial,thusopeningthediscussiononitsapplicabilityin thistypeofsample.However,theuseofNS1AgcombinedwithIgM detectionincreaseddenguediagnosisinCSFto92.3%,consistent witharecentassessmentofthesetwokitsinserumreportedby Blackselletal.(2008).

Thespecificityvaluesof100%fortheNS1Earlykitfoundwhen testingCSFweresimilartootherreportsintheliteraturewiththe useofserum(Blackselletal.,2008;Dussartetal.,2008;Limaetal., 2010).However,Guzmanetal.(2010)found90%specificitywhen testedinhealthyblooddonorsandpatientswithotherconfirmed diagnoses.

Astudyusing theNS1Earlykitshowedthat whenIgM was presentintheserum,thesensitivityofNS1Agdetectionfellfrom 91.6%to48.3%(Serakanetal.,2007).ThisalsohappenedintheCSF samplesstudiedhere,butdespitetheaverageofsixdaysbetween theonsetofthediseaseandthesamplecollection,thedetectionrate decreasedslightlyinrelationtothepresenceofIgM,from57.1%to 47.4%fortheNS1Earlykit.Thisfactisprobablyduetothetypeof patient,sinceallthesamplesweretakenfrompatientswithfatal outcomes.TheycouldhavehadhigherlevelsoftheNS1 glycopro-teincirculatingintheplasmaandhigherlevelsofviralRNA,which couldbeoneofthereasonsforthedevelopmentofseriousforms ofthedisease,assuggestedbyLibratyetal.(2002),eventhough manypatientswithhighlevelsofviremianeverdevelopclinical complications.However,theseveresyndromehasbeenobserved inpatientswhopresentcirculatingheterotypicdengueantibodies athighconcentrations(Halstead,2009).

ThisevaluationshowsthattheNS1Earlykitfordetectionof NS1AginserumcanbeusedonCSFsamplesfrompatients diag-nosedwithdengue.Serologicaltestsaresimple,rapid andeasy toperform,buttheaveragelifespanofIgMantibodies,whichcan beuptotwomonths,confusesthediagnosisincaseswherethe dateofdiseaseonsetisunknown,leavingquestionsaboutwhether theinfectionisacuteorrecent(Innis,1997).Thedetectionofthe NS1antigencombinestheaccuracyandspeedofRT-PCRwiththe practicalityoftheELISAtechnique,providingareliableresultthat facilitatesclinicalmanagementofpatientswithneurological man-ifestations.

TheuseofteststodetecttheNS1Agallowsstudyingthe involve-mentoftheCNSinpatientssuspectedofbeinginfectedwithDENV, enablingbettersupportforpatientswithsevereformsofthe dis-ease.Therefore, theresultsofthis studysuggesttheuseof the Pan-EDengueEarlyELISAassaytodetectviralNS1AginCSF,to diagnosesevereformsofdenguethatpresentneurological man-ifestationsinendemicregionsfordengue.However,itshouldbe usedinassociationwithDENVIgMantibodydetectioninCSF,in ordertoincreasethesensitivityofthediagnosis.Besidesthis,more studies,includingduringnon-epidemicperiodsand prospective studies,areneededtoassessbettertheaccuracyofthekitonCSF samples.

Acknowledgements

F.M.C.Araújoetal./JournalofVirologicalMethods177 (2011) 128–131 131

References

Blacksell,S.D.,MammenJr.,M.P.,Thongpaseuth,S.,Gibbons,R.V.,Jarman,R.G., Jen-jaroen,K.,Nisalak,A.,Phetsouvanh,R.,Newton,P.N.,Day,N.P.,2008.Evaluation ofthePanbiodenguevirusnonstructural1antigendetectionand immunoglobu-linMantibodyenzyme-linkedimmunosorbentassaysforthediagnosisofacute dengueinfectionsinLaos.Diagn.Microbiol.Infect.Dis.60,43–49.

Cavalcanti,L.P.,Coelho,I.C.,Vilar,D.C.,Holanda,S.G.,Escóssia,K.N.,Souza-Santos, R.,2010.Clinicalandepidemiologicalcharacterizationofdenguehemorrhagic fevercasesinnortheasternBrazil.Rev.Soc.Bras.Med.Trop.43,355–358. Chen,W.J.,Hwang,K.P.,Fang,A.H.,1991.DetectionofIgMantibodiesfrom

cere-brospinalfluidandseraofdenguefeverpatients.SoutheastAsianJ.Trop.Med. PublicHealth.22,659–663.

Dussart,P.,Petit,L.,Labeau,B.,Leduc,A.,Moua,D.,Matheus,S.,Baril,L.,2008. Eval-uationoftwonewcommercialtestsforthediagnosisofacutedenguevirus infectionusingNS1antigendetectioninhumanserum.PlosNegl.Trop.Dis.2, e280.

Gubler,D.J.,Kuno,G.,Sather,G.E.,Velez,M.,Oliver,A.,1984.Mosquitocellcultures andspecificmonoclonalantibodiesinsurveillancefordengueviruses.Am.J. Trop.Med.Hyg.33,158–165.

Gubler,D.J.,2002.Epidemicdengue/denguehemorrhagicfeverasapublichealth, socialandeconomicprobleminthe21st_{century.TrendsMicrobiol.10,100–103.}

Guzman,M.G.,Jaenisch,T.,Gaczkowski,R.,Hang,V.T.T.,Sekaran,S.D.,Kroeger,A., Vazquez,S.,Ruiz,D.,Martinez,E.,Mercado,J.C.,Balmaseda,A.,Harris,E.,Dimano, E.,Leano,P.S.A.,Yoksan,S.,Villegas,E.,Benduzu,H.,Villalobos,I.,Farrar,J., Sim-mons,C.P.,2010.Multi-Countryevaluationofthesensitivityandspecificityof twocommercially-availableNS1ELISAassaysfordenguediagnosis.PlosNegl. Trop.Dis.4,e811.

Halstead,S.B.,2009.Antibodiesdeterminevirulenceindengue.Ann.N.Y.Acad.Sci. 1171,E48–E56.

Innis,B.L.,1997.Antibodyresponsestodenguevirusinfection.In:Gubler,D.J.,Kuno, G.(Eds.),DengueandDengueHemorrhagicFever.CABInternational,Cambridge, MA,pp.221–243.

International Committee on Taxonomy of Viruses (ICTV), 2006. http://www.ncbi.nlm.nih.gov/ICTVdb/ICTVdB/.

Lanciotti,R.S.,Calisher,C.H.,Gubler,D.J.,Vorndam,A.V.,1992.Rapiddetectionand typingofdenguevirusfromclinicalsamplesbyusingreverse transcriptase-polymerasechainreaction.J.Clin.Microbiol.30,545–551.

Libraty,D.H.,Young,P.R.,Pickering,D.,Endy,T.P.,Kalayanarooj,S.,Green,S.,Vaughn, D.W.,Nisalak,A.,Ennis,F.A.,Rothman,A.L.,2002.Highcirculatinglevelsofthe denguevirusnonstructuralproteinNS1earlyindengueilnesscorrelatewith thedevelopmentofdenguehemorrhagicfever.J.Infect.Dis.186,1165–1168. Lima,M.R.Q.,Nogueira,R.M.R.,Schtzmayer,H.G.,Santos,I.B.,2010.Comparison

ofthreecommerciallyavailabledengueNS1antigencaptureassaysforacute diagnosisofdengueinBrazil.PlosNegl.Trop.Dis.4,e738.

Mackenzie,J.M.,Jones,M.K.,Young,P.R.,1996.Immunolocalizationofthedengue virusnonstructuralglycoproteinNS1suggestsaroleinviralRNAreplication. Virology220,232–240.

Miagostovich,M.P., Ramos, R.G., Nicol, A.F.,Nogueira, R.M.R., Cuzzi-maya, T., Oliveira,A.V.,Marchevsky,R.S.,Mesquita,R.P.,Schatzmayr,H.G.,1997. Retro-spectivestudyondenguefatalcases.Clin.Neuropathol.16,204–208. Nogueira,R.M.R.,Araújo,J.M.G.,Schatzmayr,H.G.,2007.DenguevirusinBrazil,

1986-2006.Rev.Panam.SaludPublica.22,358–363.

Nogueira,R.M.R.,Filippis,A.M.B.,Coelho,J.M.O.,Sequeira,P.C.,Schatzmayr,H.G., Paiva,F.G.,Ramos,A.M.O.,Miagstovich,M.P.,2002.Denguevirusinfectionof thecentralnervoussystem(CNS):acasereportfromBrazil.SoutheastAsianJ TropMed.PublicHealth.33,68–77.

Puccioni-Sholer,M.,Soares,C.N.,Papaiz-Alvarenga,R.,Castro,M.J.C.,Faria,L.C., Per-alta,J.M.,2009.Neurologicdenguemanifestationsassociatedwithintrathecal specificimmuneresponse.Neurology73,1413–1417.

Ramos,C.,Sanchez,G.,Pando,R.H.,Baquera,J.,Hernández,D.,Mota,J.,Ramos,J., Flo-res,A.,Llausás,E.,1998.Denguevirusinthebrainofafatalcaseofhemorrhagic denguefever.J.Neurovirol.4,465–468.

Rosen,L.,Drowet,M.T.,Deubel,V.,1999.DetectionofdenguevirusRNAbyreverse transcriptionpolymerasechainreactionin theliverandlymphoidorgans butnot inthebraininfatal humaninfection.Am. J.Trop.Med.Hyg.61, 720–724.

Serakan,S.D.,Lan,E.C.,Mahesawarappa,B.K.,Appanna,R.,Subramaniam,G.,2007. EvaluationofadengueNS1captureELISAassayfortherapiddetectionofdengue. J.Infect.Dev.Ctries.1,182–188.

Soares,C.N.,Faria,L.C.,Puccioni-Sohler,M.,Peralta, J.M.,Freitas,M.R.G.,2006. Dengueinfection:neurologicalmanifestationsandcerebrospinalfluid(CSF) analysis.J.Neurol.Sci.249,19–24.