Transfusion-associated circulatory overload and transfusion ...

Transfusion-associated circulatory overload (TACO) and transfusion-related acute lung injury (TRALI) are syndromes of acute respiratory ... SkiptoMainContent Advertisement Close BloodBloodAdvancesHematologyASH-SAPASHClinicalNewsTheHematologistASHNewsDailyInternationalBloodChineseEditionBloodJapaneseEditionBloodItalianEditionBloodLatinAmericaEditionBloodSpanishEditionASHASHHomeResearchEducationAdvocacyMeetingsPublicationsASHStore Cart UserTools Cart SignIn Search navsearch searchinput Searchinputautosuggest searchfilter AllContentAllJournalsBlood Search ToggleMenuMenu Issues CurrentIssue AllIssues Firstedition Abstracts 2021AnnualMeeting 2020AnnualMeeting 2020LateBreaking 2019AnnualMeeting 2019LateBreaking AllMeetingAbstracts Collections Collections SpecialCollections Multimedia Alerts AuthorCenter Submit AuthorGuide StyleGuide WhySubmittoBlood? About AboutBlood EditorialBoardandStaff Subscriptions PublicAccess Copyright Alerts BloodClassifieds SkipNavDestination ContentMenu Close Abstract Historicalandcurrentperspective Clinicalpresentationanddiagnosis Pathophysiology Potentialtherapiesandfuturedirections Acknowledgments Authorship REFERENCES ArticleNavigation TransfusionMedicine| April25,2019 Transfusion-associatedcirculatoryoverloadandtransfusion-relatedacutelunginjury JohnW.Semple, JohnW.Semple 1DivisionofHematologyandTransfusionMedicine,LundUniversity,Lund,Sweden;2KeenanResearchCentreforBiomedicalScience,St.Michael'sHospital,Toronto,Canada;and3DepartmentofPharmacology,4DepartmentofMedicine,and5DepartmentofLaboratoryMedicineandPathobiology,UniversityofToronto,Toronto,Canada   https://orcid.org/0000-0002-1510-0077 Searchforotherworksbythisauthoron: ThisSite PubMed GoogleScholar JohanRebetz, JohanRebetz 1DivisionofHematologyandTransfusionMedicine,LundUniversity,Lund,Sweden; Searchforotherworksbythisauthoron: ThisSite PubMed GoogleScholar RickKapur RickKapur 1DivisionofHematologyandTransfusionMedicine,LundUniversity,Lund,Sweden;   https://orcid.org/0000-0002-1608-876X Searchforotherworksbythisauthoron: ThisSite PubMed GoogleScholar Blood(2019)133(17):1840–1853. https://doi.org/10.1182/blood-2018-10-860809 Articlehistory Submitted: October17,2018 Accepted: December10,2018 FirstEdition: February26,2019 ConnectedContent Arelatedarticlehasbeenpublished: Bloodgroupgenotyping Arelatedarticlehasbeenpublished: Noninfectioustransfusion-associatedadverseeventsandtheirmitigationstrategies Arelatedarticlehasbeenpublished: Preventionoftransfusion-transmittedinfections Viewmore Arelatedarticlehasbeenpublished: Introductiontoareviewseriesontransfusionmedicine Arelatedarticlehasbeenpublished: Transfusion-relatedredbloodcellalloantibodies:inductionandconsequences Viewless Split-Screen ShareIcon Share Twitter LinkedIn ToolsIcon Tools RequestPermissions CiteIcon Cite SearchSite PDFLinkPDF Citation JohnW.Semple,JohanRebetz,RickKapur;Transfusion-associatedcirculatoryoverloadandtransfusion-relatedacutelunginjury.Blood2019;133(17):1840–1853.doi:https://doi.org/10.1182/blood-2018-10-860809 Downloadcitationfile: Ris(Zotero) ReferenceManager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex toolbarsearch Search navsearch searchinput Searchinputautosuggest searchfilter AllContentAllJournalsBlood Search Abstract Transfusion-associatedcirculatoryoverload(TACO)andtransfusion-relatedacutelunginjury(TRALI)aresyndromesofacuterespiratorydistressthatoccurwithin6hoursofbloodtransfusion.TACOandTRALIaretheleadingcausesoftransfusion-relatedfatalities,andspecifictherapiesareunavailable.Diagnostically,itremainsverychallengingtodistinguishTACOandTRALIfromunderlyingcausesoflunginjuryand/orfluidoverloadaswellasfromeachother.TACOischaracterizedbypulmonaryhydrostatic(cardiogenic)edema,whereasTRALIpresentsaspulmonarypermeabilityedema(noncardiogenic).Thepathophysiologyofbothsyndromesiscomplexandincompletelyunderstood.A2-hitmodelisgenerallyassumedtounderlieTACOandTRALIdiseasepathology,wherethefirsthitrepresentstheclinicalconditionofthepatientandthesecondhitisconveyedbythetransfusionproduct.InTACO,cardiacorrenalimpairmentandpositivefluidbalanceappearfirsthits,whereassuboptimalfluidmanagementorothercomponentsinthetransfusedproductmayenablethesecondhit.Remarkably,otherfactorsbeyondvolumeplayaroleinTACO.InTRALI,thefirsthitcan,forexample,berepresentedbyinflammation,whereasthesecondhitisassumedtobecausedbyantileukocyteantibodiesorbiologicalresponsemodifiers(eg,lipids).Inthisreview,weprovideanup-to-dateoverviewofTACOandTRALIregardingclinicaldefinitions,diagnosticstrategies,pathophysiologicalmechanisms,andpotentialtherapies.MoreresearchisrequiredtobetterunderstandTACOandTRALIpathophysiology,andmorebiomarkerstudiesarewarranted.Collectively,thismayresultinimproveddiagnosticsanddevelopmentoftherapeuticapproachesfortheselife-threateningtransfusionreactions. Subjects: ReviewArticles, ReviewSeries, TransfusionMedicine Topics: transfusionassociatedcirculatoryoverload, transfusion-relatedacutelunginjury, antibodies, transfusion Historicalandcurrentperspective Theoccurrenceoftransfusion-associatedcirculatoryoverload(TACO)firstreceivedattentioninthe1930s.In1936,Plummerreported5fatalitiesduetocirculatoryoverloadafterbloodtransfusion1 followedbymorepublicationsbyPygott2 andDeGowin3 inthe1930s,andDrummond4 andPelnerandWaldman5 inthe1940sand1950s.Pulmonaryedemawasobservedinthesecases,andpatientswithleft-sidedheartdiseasewerefoundtobeatparticularrisk.4,5 Despitethisearlyawareness,itwasnotuntilthe1990swhenTACOreceivedrenewedattentionandwasseenasadistinctclinicalentity.6 Currently,TACOisthemostfrequentpulmonarycomplicationoftransfusion,anditisanindependentriskfactorforin-hospitalmorbidityandmortalitywithincreasedincidenceinamixedintensivecareunitpopulation.7 TheestimatedfrequencyofTACOvariesfrom1%inhemovigilancereports,upto8%inpostoperativeelderlypatients,andupto11%incriticallyillpatients.8-10 Usingaprospectiveactivesurveillancealgorithm,theincidenceofTACOwasfoundtobe1%.11 Inapediatricintensivecarepopulationconsistingof136patients,theincidenceofTACOwasestimatedtobebetween1.5%and11%.12 TheNationalBloodCollectionandUtilizationSurveyreportdescribedTACOtooccurin1:14 000transfusedcomponents.13 TACOappearstobeoneoftheleadingcausesoftransfusion-relatedfatalities,with44.1%ofthereportedtransfusion-relatedfatalitiesbeingfromTACO(60/136reportedtransfusion-relateddeathsfrom2010to2017)accordingtotheSeriousHazardsofTransfusion(SHOT)report.14 TheFoodandDrugAdministration(FDA)reportedthat30%ofthereportedtransfusion-relatedfatalitieswereduetoTACO(56/186reportedtransfusion-relateddeathsfrom2012to2016).15  Transfusion-relatedacutelunginjury(TRALI)wasfirstreportedin1951byBarnard,whodescribedapatientwithacuteleukemiawhodieduponbloodtransfusionduetoanacutepulmonaryreactionduetoahypersensitivityresponse.16 In1957,BrittinghamalsodescribedaTRALIcaseofaseverepulmonaryreactionduetotransfusedbloodcontaininghigh-titerleukoagglutinins.17 In1966,Phillipsetalreported3transfusedpatientswhosufferedfrompulmonaryedemawithoutfluidoverfload.18 In1970,Wardobservedtheoccurrenceofnoncardiogenicpulmonaryedemapossiblyduetoantileukocyteantibodies,19 andin1971,ThompsonetalpublishedacasereportconcerningafatalpulmonaryreactiontoHLA-incompatiblebloodtransfusion.20 Itwasnotuntil1983whenTRALIwasofficiallyrecognizedasaseparatediseaseentitybecauseitwasdescribedtooccurduetopassivetransferofantileukocyteantibodies.21 TheNationalBloodCollectionandUtilizationSurveyreportedTRALItooccur1:64 000transfusedcomponents,13 andthefrequencyofTRALIhasbeenestimatedtobe∼0.08%to15.1%perpatientand0.01%to1.12%perproduct.22 TheFDAhasreportedTRALItobetheleadingcauseoftransfusion-relatedfatalitiesformanyyearswith34%ofthereportedtransfusion-relatedfatalitiesbeingfromTRALI(64/186reportedtransfusion-relateddeathsfrom2012to2016).15 Incontrast,the2017SHOTreportpublishedthat∼4%ofthereportedtransfusion-relatedfatalitieswereduetoTRALI(5/136reportedtransfusion-relateddeathsfrom2010to2017).14 SurvivalofTRALIincriticallyillpatientshasbeenreportedtobeaslowas53%comparedwith83%inacutelunginjurycontrolpatients.23  Clinicalpresentationanddiagnosis TACOandTRALIarelife-threateningtransfusionreactions,anditremainschallengingtoaccuratelydiagnoseanddistinguishbothsyndromes.BothTACOandTRALIpresentwiththeonsetofacuterespiratorydistress(hypoxemia)within6hoursofabloodtransfusionanddemonstrateinfiltratesonafrontalchestradiographindicativeofthepresenceofpulmonaryedema(Figure1).Clinicaldefinitionshavebeenestablishedforbothsyndromes(Table1).ForTACO,thisincludestheNationalHealthcareSafetyNetworkdefinitionspublishedin201624 orthe2011InternationalSocietyofBloodTransfusion(ISBT)criteria,25 whereasforthediagnosisofTRALI,the2005NationalHeart,Lung,andBloodInstituteWorkingGroupdefinitions26 orthe2004CanadianConsensusConferencecriteria27 arecommonlyused(Table1).TheclinicaldefinitionsofTACOmayincludeevidenceofpositivefluidbalanceorcardiogenicinvolvement,whichmaymanifestasleftheartfailure,elevatedbloodpressure,ortachycardia.Incontrast,TRALIisstrictlynoncardiogenic,withoutevidenceofleftarterialhypertension,andthus,circulatoryoverloadmustbeexcluded.Inaddition,notemporalrelationshiptoanalternativeriskfactorforacutelunginjury(eg,pneumonia,sepsis,aspiration,multipletrauma,andacutepancreatitis)maybepresentforthediagnosisofTRALI;otherwise,ithastobeclassifiedas“possibleTRALI.”IthasbeensuggestedthatthetransfusionitselfmayhaveaminorcontributioninpossibleTRALIandthatthistermshouldthereforebereplacedby“transfusedacuterespiratorydistresssyndrome(ARDS).”28,29 Others,however,argueagainstthisbystatingthatPossibleTRALIisadistinctclinicalentity,whichisespeciallyevidentincriticallyillandinjuredpatients.30 DespitethesimilarityinclinicalpresentationbetweenTACOandTRALI,keydiagnosticfeatureshavebeenidentifiedthatmayaidinclinicaldiagnosisanddifferentiationofTACOvsTRALI(Table2).31 Unfortunately,thetoolstoperformthesediagnosticanalysesmaynotbewidelyavailable.Thesekeydiagnosticfeaturesincludeassessmentofhydrostaticpulmonarypressure(increasedinTACO),proteinlevelsinedemafluid(proteinpoorinTACO),responsetodiuretics(mayoccurinTACO),andcardiogenicnonlaboratoryparameters,whichmaybeimpairedinTACO(eg,decreasedsystolicinjectionfraction,increasedsystolicbloodpressure,increasedvascularpediclewidth,andincreasedcardiothoracicratioonchestradiograph,inTACO).IncreasedventricularfillingormyocardialstretchingmaybeassessedaswellinthediagnosisofTACObyanalyzingthelevelsofBNPorN-terminalpro-BNPbecausebothhavedemonstratedasignificantpositivecorrelationbetweenpre-andposttransfusionlevelsinTACOpatients.32,33 Caution,however,isrequiredwiththeuseofnatriureticpeptidesasadiagnosticmeasurefordistinguishingTACOfromTRALI,becausetheselevelsmayalsobesignificantlyincreasedincriticallyillpatientssufferingfromTRALI.34 Importantly,cardiacischemia(ischemicchangesonelectrocardiographyorincreasednewtroponinTlevels)mustbeexcludedindiagnosingTACO.SupportiveclinicalandlaboratoryfeaturesmayperhapsfurtherhelpinthediagnosisofTACOorTRALI(Table3).35 Thisincludesphysicalexaminationofneckveins(distendedinTACO),auscultation(rales,S3inTACO),andbloodpressuremeasurement(increasedinTACO).Regardingbodytemperature,TRALIpatientsarefrequentlyfebrile,butfevermaybepresentinone-thirdofTACOpatientsaswell.36 Supportivelaboratoryinvestigationsmayincludedeterminationofwhitebloodcellcounts(transientleukopeniaandmildthrombocytopeniainTRALI),37-40 thedetectionofantibodiesagainstrecipientHLAclassI/IIand/orhumanneutrophilantigen(HNA)antigensintheinvolvedblooddonor(in∼80%ofTRALIcases),41 andanalysisofcytokines,asisdiscussedlaterinmoredetail.Itremains,however,unknownifbloodcellcountsareaffectedandifantileukocyteantibodiescouldbepresentinTACOpatients.Notably,thereareongoinginternationaleffortstofurtherdefineTACO,whichincludethe2017ISBTworkingpartyonHaemovigilancetogetherwiththeInternationalHaemovigilanceNetwork,whichhaveproposedrevisedTACOcriteriatoestablishasurveillancedefinitionforreportingandtrackingpurposes.42  Figure1.ViewlargeDownloadPPTChestradiographsofaTACOandTRALIpatient.PatientattimeofTACOoccurrence(A),andthesamepatientafterresolutionofTACO(B).Pretransfusionpatient(C),andthesamepatientattimeofTRALIoccurrence(D).Normalchestradiographswithoutsignsofpulmonaryedema(B-C);infiltrativechangesindicativeofpulmonaryedema(A,D).AdaptedfromAgnihotriandAgnihotri110 andVlaarandJuffermans22 withpermission.Figure1.ViewlargeDownloadPPTChestradiographsofaTACOandTRALIpatient.PatientattimeofTACOoccurrence(A),andthesamepatientafterresolutionofTACO(B).Pretransfusionpatient(C),andthesamepatientattimeofTRALIoccurrence(D).Normalchestradiographswithoutsignsofpulmonaryedema(B-C);infiltrativechangesindicativeofpulmonaryedema(A,D).AdaptedfromAgnihotriandAgnihotri110 andVlaarandJuffermans22 withpermission. Table1.ClinicaldefinitionsforTACOandTRALINHSN2016-TACOdefinition24  . ISBT2011-TACOdefinition25  . NHLBIWorkingGroup2005-TRALIdefinition26  . CanadianConsensusConference2004-TRALIdefinition27  . Newonsetorexacerbationof≥3ofthefollowingwithin6hoftransfusion: Any4ofthefollowingoccurringwithin6hofcompletionoftransfusion: Patientswithoutacutelunginjury(ALI)riskfactor(s)otherthantransfusion: ALI:  (a)Acuterespiratorydistress(dyspnea,orthopnea,andcough)  (a)Acuterespiratorydistress  InpatientswithnoALIimmediatelybeforetransfusion,atemporalassociationoftransfusionandALIismadeifthereis:  (a)Acuteonset  (b)Evidenceofpositivefluidbalance  (b)Tachycardia   (a)NewALI,96 and  (b)Hypoxemia:PaO2/FiO2≤300,orSpO2<90%onroomair(orotherclinicalevidenceofhypoxemiainanonresearchsetting)  (c)Elevatedbrainnatriureticpeptide  (c)Elevatedbloodpressure   (b)Theonsetofsymptomsorsignsisduringorwithin6haftertheendoftransfusionof1ormoreplasma-containingbloodproducts  (c)Bilateralinfiltratesonfrontalchestradiograph  (d)Radiographicevidencepulmonaryedema  (d)Acuteorworseningpulmonaryedemaoffrontalchestradiograph   (d)Noevidenceofleftatrialhypertension(ie,circulatoryoverload)  (e)Evidenceofleftheartfailure  (e)Evidenceofpositivefluidbalance  AsthereisnootherALIriskfactor,thenewALIisinferredtobemechanisticallyrelatedtotransfusion(ie,TRALI) NopreexistingALIbeforetransfusion  (f)Elevatedcentralvenouspressure  PatientswithALIriskfactor(s)otherthantransfusion: Duringorwithin6hoftransfusion    InpatientswithnoALIimmediatelybeforetransfusion,atemporalassociationoftransfusionandALIismadeifthereis: NotemporalrelationshiptoanalternativeriskfactorforALI     (a)NewALI,96 and IfcleartemporalrelationshiptoanalternativeriskfactorforALI(incombinationwithpreviouslistedcriteria),thendefinedas“possibleTRALI”     (b)Theonsetofsymptomsorsignsisduringorwithin6haftertheendoftransfusionof1ormoreplasma-containingbloodproducts     Byassessingthepatient’sclinicalcourse,thenewALIiseither:      (a)TRALI,andthenewALIisinferredtobemechanisticallyrelatedtothetransfusion,orboththetransfusionandthealternativeriskfactor,or      (b)NotTRALI,andthenewALIismechanisticallyrelatedtothealternativeALIriskfactoralone,whereasthetransfusioniscoincidental  NHSN2016-TACOdefinition24  . ISBT2011-TACOdefinition25  . NHLBIWorkingGroup2005-TRALIdefinition26  . CanadianConsensusConference2004-TRALIdefinition27  . Newonsetorexacerbationof≥3ofthefollowingwithin6hoftransfusion: Any4ofthefollowingoccurringwithin6hofcompletionoftransfusion: Patientswithoutacutelunginjury(ALI)riskfactor(s)otherthantransfusion: ALI:  (a)Acuterespiratorydistress(dyspnea,orthopnea,andcough)  (a)Acuterespiratorydistress  InpatientswithnoALIimmediatelybeforetransfusion,atemporalassociationoftransfusionandALIismadeifthereis:  (a)Acuteonset  (b)Evidenceofpositivefluidbalance  (b)Tachycardia   (a)NewALI,96 and  (b)Hypoxemia:PaO2/FiO2≤300,orSpO2<90%onroomair(orotherclinicalevidenceofhypoxemiainanonresearchsetting)  (c)Elevatedbrainnatriureticpeptide  (c)Elevatedbloodpressure   (b)Theonsetofsymptomsorsignsisduringorwithin6haftertheendoftransfusionof1ormoreplasma-containingbloodproducts  (c)Bilateralinfiltratesonfrontalchestradiograph  (d)Radiographicevidencepulmonaryedema  (d)Acuteorworseningpulmonaryedemaoffrontalchestradiograph   (d)Noevidenceofleftatrialhypertension(ie,circulatoryoverload)  (e)Evidenceofleftheartfailure  (e)Evidenceofpositivefluidbalance  AsthereisnootherALIriskfactor,thenewALIisinferredtobemechanisticallyrelatedtotransfusion(ie,TRALI) NopreexistingALIbeforetransfusion  (f)Elevatedcentralvenouspressure  PatientswithALIriskfactor(s)otherthantransfusion: Duringorwithin6hoftransfusion    InpatientswithnoALIimmediatelybeforetransfusion,atemporalassociationoftransfusionandALIismadeifthereis: NotemporalrelationshiptoanalternativeriskfactorforALI     (a)NewALI,96 and IfcleartemporalrelationshiptoanalternativeriskfactorforALI(incombinationwithpreviouslistedcriteria),thendefinedas“possibleTRALI”     (b)Theonsetofsymptomsorsignsisduringorwithin6haftertheendoftransfusionof1ormoreplasma-containingbloodproducts     Byassessingthepatient’sclinicalcourse,thenewALIiseither:      (a)TRALI,andthenewALIisinferredtobemechanisticallyrelatedtothetransfusion,orboththetransfusionandthealternativeriskfactor,or      (b)NotTRALI,andthenewALIismechanisticallyrelatedtothealternativeALIriskfactoralone,whereasthetransfusioniscoincidental  NHLBI,NationalHeart,LungandBloodInstitute;NHSN,NationalHealthcareSafetyNetwork.ViewLarge Table2.KeydiagnosticfeaturesdifferentiatingTACOfromTRALIKeydiagnosticfeature . Specificdiagnosticreadout . TACO . TRALI . Acuteonsetofrespiratorydistresssymptoms Onset<6huponbloodtransfusion Yes Yes Hypoxemia SpO2<90%orPaO2/FiO2<300mmHgonroomair Yes Yes Pulmonaryedema Bilateralinfiltratesonchestradiograph Yes Yes AlternativeriskfactorsforALI eg,pneumonia,sepsis,aspiration,multipletrauma,acutepancreatitis No No Yes:possibleTRALI Hydrostaticpulmonarypressureincreased Pulmonaryarteryocclusionpressure>18mmHg Yes No Protein-pooredemafluid Edemaorplasmaproteinconcentration<0.65attheonsetofacuterespiratoryfailure Yes No Increasedventricularfilling/myocardialstretching32-34  B-typenatriureticpeptide(BNP)>250orpre-/posttransfusionBNPratio>1.5orN-terminalpro-BNP>1000pg/mL Yes Yes*/No Responsetodiuretics Rapidandsignificantimprovement Yes No Cardiogenicnonlaboratoryevidenceforcirculatoryoverload Systolicejectionfraction<45andnoseverevalvularheartdiseaseonechocardiography Yes No Systolicbloodpressure>160 Vascularpediclewidth>65mmandcardiothoracicratio>0.55onchestradiograph Cardiacischemia NewischemicchangesonelectrocardiographyornewtroponinTlevelsof>0.05 No No Keydiagnosticfeature . Specificdiagnosticreadout . TACO . TRALI . Acuteonsetofrespiratorydistresssymptoms Onset<6huponbloodtransfusion Yes Yes Hypoxemia SpO2<90%orPaO2/FiO2<300mmHgonroomair Yes Yes Pulmonaryedema Bilateralinfiltratesonchestradiograph Yes Yes AlternativeriskfactorsforALI eg,pneumonia,sepsis,aspiration,multipletrauma,acutepancreatitis No No Yes:possibleTRALI Hydrostaticpulmonarypressureincreased Pulmonaryarteryocclusionpressure>18mmHg Yes No Protein-pooredemafluid Edemaorplasmaproteinconcentration<0.65attheonsetofacuterespiratoryfailure Yes No Increasedventricularfilling/myocardialstretching32-34  B-typenatriureticpeptide(BNP)>250orpre-/posttransfusionBNPratio>1.5orN-terminalpro-BNP>1000pg/mL Yes Yes*/No Responsetodiuretics Rapidandsignificantimprovement Yes No Cardiogenicnonlaboratoryevidenceforcirculatoryoverload Systolicejectionfraction<45andnoseverevalvularheartdiseaseonechocardiography Yes No Systolicbloodpressure>160 Vascularpediclewidth>65mmandcardiothoracicratio>0.55onchestradiograph Cardiacischemia NewischemicchangesonelectrocardiographyornewtroponinTlevelsof>0.05 No No ModifiedfromGajicetal.31 *Intransfusedcriticallyillpatients.ViewLarge Table3.SupportiveclinicalandlaboratoryfeaturesfurtherdifferentiatingTACOfromTRALISupportiveclinicalandlaboratoryfeatures . TACO . TRALI . Neckveins Distended Normal Auscultation Rales,S3 Rales,NoS3 Bloodpressure Hypertension Hypotension Bodytemperature Fevermaybepresent(1/3ofpatients)36  Febrile Whitebloodcellcount Unknown Transientleukopenia(infrequent)andmildthrombocytopenia37-40  Leukocyteantibodies Unknown Maybepresent(±80%)41 : Anti-HLAclassIantibodies Anti-HLAclassIIantibodies Anti-HNAantibodies Posttransfusioncytokines49-51,59  IL-6increased IL-6increased IL-8notincreased IL-8increased IL-10increased IL-10notincreased NochangesinTNF-α,GM-CSF NochangesinTNF-α,GM-CSF Supportiveclinicalandlaboratoryfeatures . TACO . TRALI . Neckveins Distended Normal Auscultation Rales,S3 Rales,NoS3 Bloodpressure Hypertension Hypotension Bodytemperature Fevermaybepresent(1/3ofpatients)36  Febrile Whitebloodcellcount Unknown Transientleukopenia(infrequent)andmildthrombocytopenia37-40  Leukocyteantibodies Unknown Maybepresent(±80%)41 : Anti-HLAclassIantibodies Anti-HLAclassIIantibodies Anti-HNAantibodies Posttransfusioncytokines49-51,59  IL-6increased IL-6increased IL-8notincreased IL-8increased IL-10increased IL-10notincreased NochangesinTNF-α,GM-CSF NochangesinTNF-α,GM-CSF ModifiedfromSkeateetal.35 ViewLarge Pathophysiology Two-hitmodel ThepathogenesesofTACOandTRALIareincompletelyunderstood.Asageneralpathophysiologicalframework,a2-hitmodelhasbeensuggestedforbothTACO7 andTRALI.43,44 Thefirsthitrepresentstheunderlying,preexistingclinicalconditionofthepatient(transfusion-recipientriskfactors),whereasthesecondhitisconveyedbythetransfusedbloodproduct.BothhitsarerequiredfordevelopmentofTACOorTRALI.ThefirsthitinTACOmayberepresentedbythepooradaptabilityforvolumeoverload.Arecentlargeprospectivestudyenrolling200patientswithTACOidentifiedcongestiveheartfailure,cardiomegalyonchestradiograph,pretransfusiondiureticuse,elevatedbloodpressure,acutekidneyinjury,chronickidneydisease,plasmatransfusion,andemergencysurgerytoberiskfactorsassociatedwithTACO.45 Also,arecentretrospectivecohortstudyof66TACOpatientsfoundcardiacfailure,renalfailure,andthedegreeofpositivefluidbalanceasriskfactorsforTACOdevelopment.46 Previously,congestiveheartfailureandrenaldysfunctionwereidentifiedascommonfeaturesinTACOinaretrospectivestudyof98TACOpatients,which,inaddition,alsoidentifiedanageofover70yearsasariskfactorforTACOdevelopment.47 AdvancedagewasalsoshowntoincreasetheincidenceofTACOinperioperativenoncardiacsurgerypatients.48 ThesecondhitinTACOmaybereflectedbysuboptimalfluidmanagementandinappropriateinfusionpractices(suchasrapidinfusionrates),whichhavefrequentlybeenrelatedtotheonsetofTACO.47 Inaperioperativesettingofnoncardiacsurgery,theincidenceofTACOwasfoundtobeincreasedwiththevolumetransfusedandthetotalintraoperativefluidbalance.48 Remarkably,itwasrecentlyobservedthatthedegreeofpositivefluidbalanceappearedtobeassociatedlesswiththedevelopmentofTACOthanwiththedevelopmentofcirculatoryoverloadintheabsenceoftransfusion.46 ThislowerassociationofpositivefluidbalancewiththedevelopmentofTACOmaybeduetotheincreaseincolloid-osmoticpressureduetothetransfusionproductstimulatingfluidsfromtheextravascularspaceintotheintravascularspaceinanattempttocontributeamoreeffectivecirculatingvolume.Alternatively,itmayindicatethatotherfactorsinthetransfusedbloodproductbesidesthetransfusionvolumecouldplayaroleintheonsetofTACO.Combinedwiththefirsthit,thesefactorsmaypossiblycontributetoinflammationinthetransfusedrecipientresultinginTACO,aswillbediscussedlater. InTRALI,first-hitriskfactorsincludechronicalcoholabuse,shock,liversurgery,currentsmoking,higherpeakairwaypressurewhileundergoingmechanicalventilation,positiveintravascularfluidbalance,49 lowinterleukin-10(IL-10)levels,50-53 andsystemicinflammation.53 Systemicinflammationmaybereflectedintheplasmacytokineprofiles(increasedIL-6andIL-8levelsasdiscussedlater)butalsoviaincreasedlevelsofC-reactiveprotein(CRP).CRPisanacute-phaseproteinthatrapidlyincreasesduringacuteinfectionsandinflammationandiswidelyusedinclinicasaninflammatorybiomarker.54 CRPwasshowntobeelevatedinTRALIpatients55 andfunctionallyenabledthefirsthitinthedevelopmentofTRALIinamurinemodelbyincreasingthelevelsoftheneutrophil(PMN)-chemoattractantmacrophageinflammatoryprotein(MIP)-2(murinehomologofIL-8),resultinginincreasedpulmonaryPMNaccumulation.56 ThesecondhitinTRALImaybeconveyedbyantileukocyteantibodiesorotherfactorspresentinthetransfusionproduct.In∼80%ofcases,anti-HLAclassIorIIoranti-HNAantibodiesareimplicatedtobeinvolvedintriggeringTRALI,41 althoughthismaybeevenhigherdependingonthedetectionmethodsused.57 Intheremaining20%ofTRALIcases,non–antibodyfactors/biologicalresponsemodifiersaresuggestedtocontributethesecondhit,andthesemaypossiblyincludelipidmediators,extracellularvesicles,andagedbloodcells(asdescribedin“Non-antibody–mediatedTRALI”).58  Cytokineprofiles CytokineprofilesmayshedlightonthepathophysiologicenvironmentofTACOandTRALI;however,inTACO,thishasrarelybeeninvestigated.Anestedcasecontrolstudyinvolving29TACOpatients,147controlpatients,aswellas70TRALIpatientshasinitiatedtheseimportantinvestigations.50 Inthisstudy,theproinflammatorycytokineIL-6wasfoundtobesignificantlyelevatedinposttransfusionTACOpatientscomparedwithmatchedcontrols.LevelsoftheproinflammatorycytokineIL-8,ontheotherhand,werenotelevatedinpre-orposttransfusionTACOpatients,whereasinTRALIpatients,bothIL-6andIL-8levelswereelevatedpre-aswellasposttransfusioninthesamestudy.50 Furthermore,theanti-inflammatorycytokineIL-10wasfoundtobeincreasedinbothpre-andposttransfusionTACOpatients,whereasIL-10levelsremainedlowinpre-andposttransfusionTRALIpatients.50 Nochangeswerefoundinpre-orposttransfusionTACOorTRALIpatientsregardinggranulocytemacrophage-colonystimulatingfactor(GM-CSF)ortumornecrosisfactor-α(TNF-α).50 OtherstudiesexaminingcytokinelevelsinTRALIhavereportedincreasedlevelsofpre-andposttransfusionIL-6andIL-8incardiacsurgerypatients,59 increasedlevelsofpretransfusionIL-8,49 andlow/nonelevatedlevelsofposttransfusionIL-10.51 Onestudy,however,foundIL-10levelstobeincreasedinTRALIpatients.60 Thismaybeduetothefactthatthisstudyinvestigatedfoldchangesofpairedsamplespriortoandfollowingtransfusion,incontrasttothecomparisonofposttransfusionTRALIsamplesvstransfusedcontrols.50,51 TheposttransfusioncytokinelevelsinTACOandTRALIbasedonthesestudiesaresummarizedinTable3.Thedatasofarappearpromising,especiallyconsideringthattheymayfacilitatethechallengingdistinctionbetweenTACOandTRALI(inasymptomaticsettingposttransfusion).LowIL-8levelsincombinationwithelevatedIL-10levelsmaypotentiallyhelpsupportadiagnosisofTACO,whereasincreasedIL-8levelsincombinationwithlowIL-10levelsmaypotentiallyaidinthediagnosisofTRALI.Itwillbeveryimportant,however,tovalidatethesecytokineprofilesinlargerpatientcohortsofTRALIandespeciallyTACO. Pathophysiologicalmechanisms TACO ResearchintothepathophysiologyofTACOhasbeensurprisinglylimitedovertheyears.7 ThepathophysiologyofTACOwasinitiallythoughttobeidenticaltothewell-studiedmechanismsofcongestiveheartfailureorpulmonaryhydrostaticedema,butisnowrecognizedtobemorecomplex.Itremainsunknownwhichpathwaysareinvolvedingeneratingthepulmonaryhydrostaticedema(protein-pooredemafluid)inTACO.Recently,studieshavestartedtoinvestigateriskfactorsfordevelopmentofTACOand,aspreviouslydiscussed,cardiacfailure,renalfailure,andpositivefluidbalancearefirst-hitcomponentsresponsibleforthepooradaptabilityforvolumeoverload.Secondhitsmaybesuboptimalfluidmanagementorothercomponentspresentinthetransfusionproduct.SecretionofIL-10,whichwasfoundtobeincreasedinTACOpatients,50 isknowntorequirespecificstimulationsuchasbymicrobialproductsorspecificantibodies.52,61 Itisthereforepossiblethatfactorsinthetransfusedbloodproductcouldperhapsbemediatingthiseffect.RegardingIL-10levelsinTACO,however,animportantrolefortherecipient’sstatus(comorbiditiesorpostoperativestatus)mustbeconsideredaswell.IL-10levelswerealsofoundtobeincreasedinpretransfusionTACOpatients,althoughapparentlytoalowerextentthaninposttransfusionTACOpatients.50 Furthermore,1studyfoundthattheoccurrenceofTACOdecreased∼50%bytheintroductionofuniversalleukoreducedproducts,62 indicatingthatotherfactorsbesidesvolumeoverloadmaybeinvolved.Asthiswasonlyasingleretrospectivestudy,furthervalidationinanimalmodels,inprospectiveclinicaltrials,andviahemovigilancesystemsisrequiredtoinvestigatewhetherTACOratesindeeddecreaseduetoleukoreduction.Similarly,ithasbeenobservedthatinperioperativenoncardiacpatients,theratesofTACOincidencedecreasedfrom2004to2011,48 whichmayberelatedtotheexclusionoffemaleplasmadonorsfortransfusionsinthisperiod.Alternatively,itmaybewellpossiblethatthisdecreaseinTACOincidencecouldalsohavebeenrelatedtochangesinclinicalpractice(suchasdeclinesintransfusionrelatedtopatientbloodmanagement).Otherfactorsthatcouldpotentiallybeinvolvedmayberelatedtoinflammationorcytokinelevels.TheproinflammatorycytokineIL-6wasfoundtobeincreasedinTACOpatients,50 andfeverwasdescribedtooccurinone-thirdofthepatientsdevelopingTACO.36 Itwillbeimportanttofurtherresearchthecontributionofthetransfusionproductbeyondthevolumealone,includingtheroleofinflammation.BecauseoflackofexperimentalanimalmodelsforTACO,almostnothingisknownaboutthecellularmoietiesinvolvedinthepathophysiologyofTACO.Onepaperreportedthepresenceofintra-alveolarneutrophils(withouttheformationofneutrophilextracellulartraps[NETs])inthelungsofaTACOpatient,63 whichadditionallysupportstheinvolvementofaninflammatorycomponentinTACO.ThecurrentknowledgeregardingthepathophysiologyofTACOisdepictedinFigure2. Figure2.ViewlargeDownloadPPTPathophysiologicalmechanismsofTACOandTRALIinthelungs.PathwaysA-Naresystematicallydiscussedinthemaintext.sCD40L,solubleCD40ligand;VEGF,vascularendothelialgrowthfactor.ThisfigurewasinpartcreatedwithimagesadaptedfromServierMedicalArtbyServier,whichislicensedunderaCreativeCommonsAttribution3.0UnportedLicense.Figure2.ViewlargeDownloadPPTPathophysiologicalmechanismsofTACOandTRALIinthelungs.PathwaysA-Naresystematicallydiscussedinthemaintext.sCD40L,solubleCD40ligand;VEGF,vascularendothelialgrowthfactor.ThisfigurewasinpartcreatedwithimagesadaptedfromServierMedicalArtbyServier,whichislicensedunderaCreativeCommonsAttribution3.0UnportedLicense. TRALI AlthoughmuchmoreisknownregardingthepathophysiologyofTRALIascomparedwithTACO,itremainscomplexandstillincompletelyunderstood.TRALIresultsinnoncardiogenicpermeabilityedema(protein-richedemafluid).MostofthedataontheTRALIpathophysiologicalmechanismsarederivedfromanimalmodelsorfrominvitroculturesofhumanlungendothelialcells.AnimalmodelsofTRALIhavesignificantlyfacilitatedresearchintotheTRALIpathophysiology,64 butcontroversieshavearisen.Thecomplexitycomesfromtheuseofvariousanimalmodels,varioustriggers(differenttypesofantibodiesorbiologicalresponsemodifiers),anddifferentexperimentalconditions.WhenassessingcellularinvolvementinthepathophysiologyofTRALI,whichissummarizedinTable4,itappearsthatPMNsarethekeypathogeniccellsthatmediatelungdamageinTRALI,65 whereasCD4+CD25+FoxP3+T-regulatory(Treg)cellsanddendriticcellsappeartobethekeyprotectivecells.52 Furthermore,itseemsthatmonocytesand/ormacrophagesandredbloodcellsconveyapathogenicroleinTRALI(Table4).AsfurtherdescribedinTable4,theroleofplateletsappearspathogenicinTRALI;however,reportshavealsostronglyarguedagainstthispathogenicinvolvement.Similarly,CD8+TlymphocytesandBlymphocytesdonotappeartohaveasignificantroleinthepathophysiologyofTRALI(Table4).SuggestedTRALIpathophysiologicmechanismsalsodependonthetypeofantibodies(anti-HLAclassI,anti-HLAclassII,oranti-HNAantibodies)orbiologicalresponsemodifiers(suchaslipidsfromstoredplateletsorredbloodcells)involved. Table4.OverviewofevidenceforandagainsttheinvolvementofspecificcelltypesinthepathogenesisofTRALI,basedonbothTRALIanimalmodelsand/orhumandataCelltype . SupportingevidenceforinvolvementinTRALI . EvidenceagainstinvolvementinTRALI . OveralleffectinTRALI . PMNs •PMNsareobservedinTRALIlungsuponautopsy86,87 •InvitroTRALImodelshaveshownPMN-inducedendothelialcelldamage.Antibody-mediatedsecondhits:anti-HNA-3aantibody,74 anti-HLAantibody,70 lowIgMserum.97 Non-antibody–mediatedsecondhits:lyso-PC,98 solubleCD40ligand,85 plateletmicroparticles82 •NETsdetectedinplasma(n=14)andlungsofTRALIpatientsandinmurineplasmaandlungmicrocirculationofmiceundergoinganti-MHCclassIantibody–mediatedTRALI(uponLPSpriming)63 Suggestedmechanism:PlateletsinduceNETformationandNETsinducelunginjuryviadirecttoxicitytopulmonaryendothelialcells•NETbiomarkersfoundinserumofTRALIpatients(n=5)andinmurinelungalveoliusingananti-MHCclassIantibody–mediatedTRALImodel(uponLPSpriming)73 Suggestedmechanism:NETsformedthroughdirectprimingofPMNsbyanti-HNA-3aantibodies73 •PMNswereshowntointeractwithvonWillebrandfactorviaCTL-2,enablinganti-HNA-3atosignalviaCD11b/CD18,resultinginPMN-activationandagglutination99 •PulmonaryPMNinfiltrationhasfrequentlybeenobservedinmultipleanimalmodelsofTRALI,bothantibodymediated52,56,63,66,68,75,76,99-103 andnon-antibodymediated103,104 •PMN-FcγRswerefoundtobeessentialinamurineanti-MHCclassIantibody–mediatedTRALImodel(withoutLPSpriming)66 Suggestedmechanism:Anti-MHCclassIantibodybindstothepulmonaryendotheliumandsequestersPMNsviatheirFcγRs,resultinginPMNactivationandTRALIinduction66 •PMNdepletionfullypreventedTRALIoccurrenceusingmurineanti-MHCclassIantibody–mediatedTRALImodels(withoutLPSpriming)52,66,100 •PMNandROSarecriticallyrequiredforTRALIinductioninamurinemodelofanti-MHCclassIantibody–mediatedTRALI(modelbasedonCD4+T-celldepletionandwithoutLPS,usingC57BL/6gp91phoxknock-outmice)52 •Ina2-eventinvivoratTRALImodel,inflammatoryprimingPMNsinducedMHCclassIIsurfaceexpressionandpulmonaryendotheliumactivation.Anti-MHCclassIIantibodiessubsequentlytargetedsequesteredPMNsresultinginTRALI71  •TRALIreportedinaneutropenicpatient105 •NoalveolarPMNinfluxobserveduponlungtissuehistologyanalysisof2TRALIpatients106 •OccurrenceofTRALIafterPMNdepletioninamurineanti-HNA-3aTRALImodel(despiteadecreaseddiseaseseverity)75 •NodependenceonPMNs(andpartialindependenceofFcγRs)butoncomplementcomponentC5ainamurineanti-MHCclassIantibody–mediatedTRALImodel67  Pathogenic Monocytesand/ormacrophages •Anti-MHCclassIantibody–mediatedmurine(SCIDmice,withoutLPSpriming)TRALIinductionwascompletelyabrogateduponmonocytedepletion(andrestoreduponrepletionwithpurifiedmonocytes)68 Suggestedmechanism:Anti-MHCclassIantibodybindingtomonocytesinducesthesecretionofthePMN-chemoattractantMIP-2(murinehomologofIL-8),resultinginpulmonaryPMNrecruitmentandTRALI68 •Depletionorinactivationofmonocytes/macrophagesinvivofullysuppressedTRALIinmurinemodelofanti-MHCclassIantibody–mediatedTRALI(withoutLPSpriming)67 Suggestedmechanism:Anti-MHCclassIantibodybindingtoendothelialcellsactivatescomplementwithproductionofC5a,whichthenbindstotheC5aRonmonocytes/macrophages,attractingthesecellstothelungsandinducingthemtoproduceROS,damagingtheendotheliumandinducingTRALI67 •Anti-HLAclassIIantibodiesinducemonocyteactivationinanexvivorodentmodelofTRALI,whichsubsequentlyresultsinactivationofPMNs,whichthendamagethepulmonaryendothelium72 •TRALImonocyteswereactivatedinvitrobyseraimplicatedinTRALI,resultinginincreasedproductionofTNF-α,IL-1β,andTissueFactorovera4-hperiodcomparedwithcontrolsera107   Pathogenic Platelets •OccurrenceofthrombocytopeniainTRALIpatients37-40 •Reportedthrombocytopeniainmurineanti-MHCclassIantibody–mediatedTRALImodelbasedonCRPinfusion56 •Plateletdepletion(plateletdepletingantibodyinjected4hpriortoTRALIinduction)protectedmicefromTRALIinamurineanti-MHCclassI–mediatedTRALImodel(withLPSpriming)100 Suggestedmechanism:PlateletsinduceNETformation(whichinducelunginjuryviadirecttoxicitytopulmonaryendothelialcells)andtargetingplateletactivationwithaspirinoraglycoproteinIIb/IIIainhibitordecreasedNETformationandTRALI63 •Storedplatelet-derivedvascularendothelialgrowthfactorpotentiallymediatesincreasedlungvascularpermeabilityinisolated-perfusedratlungs79 •PlasmaandlipidsfromstoredplateletsenableTRALIinanLPS-primedratmodel78 •Lyso-PCfromstoredplateletscausespulmonaryandsystemiccoagulopathyinanLPS-primedratmodel,viaPMN-priminginvitro81 •Platelet-derivedmicroparticlesenablePMN-mediatedpulmonaryendothelialcelldamageinvitro82 •AgedplateletsinducelunginjuryviaacidsphingomyelinaseinanLPS-primedmurineTRALImodel80  •Nothrombocytopeniadetectedinmurinemodelofanti-MHCclassIantibody–mediatedTRALI(withoutLPSpriming)66 •Noeffectofinducingthrombocytopeniaorofpharmacologicalandgenetictargetingofplateletfunctions,onthedevelopmentofTRALIinamurineanti-MHCclassIantibody–mediatedTRALImodel(withLPSpriming)108 •Plateletdepletion(withplateletdepletingantibodyinjected24and48hpriortoTRALIinductionorbyinjectionofneuraminidase)didnotsuppressTRALIdevelopmentinamurinemodelofanti-MHCclassI–mediatedTRALI(withoutLPSpriming)67  Pathogenic?Dispensable? Redbloodcells •PlasmaandlipidsfromredbloodcellsandleukoreducedredbloodcellsprimedPMNsandinducedALIinratmodelwithLPSpriming84 •Redbloodcell–derivedmicroparticlesmayprimePMNsandenableTRALIinLPS-primedmice83 •LipidsandsupernatantsfromstoredredbloodcellsactivatepulmonaryendotheliumthroughtheBLT2receptorandproteinkinaseCactivation,andpredisposetoALI77 •TransfusionofsupernatantofagedredbloodcellsinducedlunginflammationandcoagulopathyinLPS-primedrats109 •Injectionofanti–redbloodcellantibody1dbeforeinductionofanti-MHCclassIantibody–mediatedmurineTRALIresultedintotalsuppressionofTRALIdevelopmentinBALB/cmice67  •Transfusionof35-dstoredautologousredbloodcellsinLPS-primedhumanvolunteers(withconfirmedsepsis/endotoxemia)didnotresultinlunginjury89  Pathogenic CD4+CD25+FoxP3Tregsanddendriticcells •DepletionofCD4+CD25+FoxP3+TregsorofCD11c+dendriticcellsresultedinlowIL-10levels,whichenabledanti-MHCclassIantibody–mediatedmurineTRALI(protectionagainstanti-MHCclassIantibody–mediatedTRALIwasassociatedwithincreasedIL-10levels)52 Suggestedmechanism:DepletionofCD4+CD25+FoxP3+TregsorofCD11c+dendriticcellsinmiceresultsinlowIL-10levels,whichenablesincreasedanti-MHCclassIantibody–mediatedplasmaMIP-2levels(murinehomologofIL-8),pulmonaryPMNinfiltration,ROSproduction,andTRALIdevelopmentwithimpairedlungfunction52   Protective CD8+Tcells  DepletionofCD8+Tcellsdidnotsignificantlyresultintheonsetofanti-MHCclassIantibody–mediatedmurineTRALI52  Nosignificantinvolvement Bcells  DepletionofBcellsdidnotsignificantlyresultintheonsetofanti-MHCclassIantibody–mediatedmurineTRALI52  Nosignificantinvolvement Celltype . SupportingevidenceforinvolvementinTRALI . EvidenceagainstinvolvementinTRALI . OveralleffectinTRALI . PMNs •PMNsareobservedinTRALIlungsuponautopsy86,87 •InvitroTRALImodelshaveshownPMN-inducedendothelialcelldamage.Antibody-mediatedsecondhits:anti-HNA-3aantibody,74 anti-HLAantibody,70 lowIgMserum.97 Non-antibody–mediatedsecondhits:lyso-PC,98 solubleCD40ligand,85 plateletmicroparticles82 •NETsdetectedinplasma(n=14)andlungsofTRALIpatientsandinmurineplasmaandlungmicrocirculationofmiceundergoinganti-MHCclassIantibody–mediatedTRALI(uponLPSpriming)63 Suggestedmechanism:PlateletsinduceNETformationandNETsinducelunginjuryviadirecttoxicitytopulmonaryendothelialcells•NETbiomarkersfoundinserumofTRALIpatients(n=5)andinmurinelungalveoliusingananti-MHCclassIantibody–mediatedTRALImodel(uponLPSpriming)73 Suggestedmechanism:NETsformedthroughdirectprimingofPMNsbyanti-HNA-3aantibodies73 •PMNswereshowntointeractwithvonWillebrandfactorviaCTL-2,enablinganti-HNA-3atosignalviaCD11b/CD18,resultinginPMN-activationandagglutination99 •PulmonaryPMNinfiltrationhasfrequentlybeenobservedinmultipleanimalmodelsofTRALI,bothantibodymediated52,56,63,66,68,75,76,99-103 andnon-antibodymediated103,104 •PMN-FcγRswerefoundtobeessentialinamurineanti-MHCclassIantibody–mediatedTRALImodel(withoutLPSpriming)66 Suggestedmechanism:Anti-MHCclassIantibodybindstothepulmonaryendotheliumandsequestersPMNsviatheirFcγRs,resultinginPMNactivationandTRALIinduction66 •PMNdepletionfullypreventedTRALIoccurrenceusingmurineanti-MHCclassIantibody–mediatedTRALImodels(withoutLPSpriming)52,66,100 •PMNandROSarecriticallyrequiredforTRALIinductioninamurinemodelofanti-MHCclassIantibody–mediatedTRALI(modelbasedonCD4+T-celldepletionandwithoutLPS,usingC57BL/6gp91phoxknock-outmice)52 •Ina2-eventinvivoratTRALImodel,inflammatoryprimingPMNsinducedMHCclassIIsurfaceexpressionandpulmonaryendotheliumactivation.Anti-MHCclassIIantibodiessubsequentlytargetedsequesteredPMNsresultinginTRALI71  •TRALIreportedinaneutropenicpatient105 •NoalveolarPMNinfluxobserveduponlungtissuehistologyanalysisof2TRALIpatients106 •OccurrenceofTRALIafterPMNdepletioninamurineanti-HNA-3aTRALImodel(despiteadecreaseddiseaseseverity)75 •NodependenceonPMNs(andpartialindependenceofFcγRs)butoncomplementcomponentC5ainamurineanti-MHCclassIantibody–mediatedTRALImodel67  Pathogenic Monocytesand/ormacrophages •Anti-MHCclassIantibody–mediatedmurine(SCIDmice,withoutLPSpriming)TRALIinductionwascompletelyabrogateduponmonocytedepletion(andrestoreduponrepletionwithpurifiedmonocytes)68 Suggestedmechanism:Anti-MHCclassIantibodybindingtomonocytesinducesthesecretionofthePMN-chemoattractantMIP-2(murinehomologofIL-8),resultinginpulmonaryPMNrecruitmentandTRALI68 •Depletionorinactivationofmonocytes/macrophagesinvivofullysuppressedTRALIinmurinemodelofanti-MHCclassIantibody–mediatedTRALI(withoutLPSpriming)67 Suggestedmechanism:Anti-MHCclassIantibodybindingtoendothelialcellsactivatescomplementwithproductionofC5a,whichthenbindstotheC5aRonmonocytes/macrophages,attractingthesecellstothelungsandinducingthemtoproduceROS,damagingtheendotheliumandinducingTRALI67 •Anti-HLAclassIIantibodiesinducemonocyteactivationinanexvivorodentmodelofTRALI,whichsubsequentlyresultsinactivationofPMNs,whichthendamagethepulmonaryendothelium72 •TRALImonocyteswereactivatedinvitrobyseraimplicatedinTRALI,resultinginincreasedproductionofTNF-α,IL-1β,andTissueFactorovera4-hperiodcomparedwithcontrolsera107   Pathogenic Platelets •OccurrenceofthrombocytopeniainTRALIpatients37-40 •Reportedthrombocytopeniainmurineanti-MHCclassIantibody–mediatedTRALImodelbasedonCRPinfusion56 •Plateletdepletion(plateletdepletingantibodyinjected4hpriortoTRALIinduction)protectedmicefromTRALIinamurineanti-MHCclassI–mediatedTRALImodel(withLPSpriming)100 Suggestedmechanism:PlateletsinduceNETformation(whichinducelunginjuryviadirecttoxicitytopulmonaryendothelialcells)andtargetingplateletactivationwithaspirinoraglycoproteinIIb/IIIainhibitordecreasedNETformationandTRALI63 •Storedplatelet-derivedvascularendothelialgrowthfactorpotentiallymediatesincreasedlungvascularpermeabilityinisolated-perfusedratlungs79 •PlasmaandlipidsfromstoredplateletsenableTRALIinanLPS-primedratmodel78 •Lyso-PCfromstoredplateletscausespulmonaryandsystemiccoagulopathyinanLPS-primedratmodel,viaPMN-priminginvitro81 •Platelet-derivedmicroparticlesenablePMN-mediatedpulmonaryendothelialcelldamageinvitro82 •AgedplateletsinducelunginjuryviaacidsphingomyelinaseinanLPS-primedmurineTRALImodel80  •Nothrombocytopeniadetectedinmurinemodelofanti-MHCclassIantibody–mediatedTRALI(withoutLPSpriming)66 •Noeffectofinducingthrombocytopeniaorofpharmacologicalandgenetictargetingofplateletfunctions,onthedevelopmentofTRALIinamurineanti-MHCclassIantibody–mediatedTRALImodel(withLPSpriming)108 •Plateletdepletion(withplateletdepletingantibodyinjected24and48hpriortoTRALIinductionorbyinjectionofneuraminidase)didnotsuppressTRALIdevelopmentinamurinemodelofanti-MHCclassI–mediatedTRALI(withoutLPSpriming)67  Pathogenic?Dispensable? Redbloodcells •PlasmaandlipidsfromredbloodcellsandleukoreducedredbloodcellsprimedPMNsandinducedALIinratmodelwithLPSpriming84 •Redbloodcell–derivedmicroparticlesmayprimePMNsandenableTRALIinLPS-primedmice83 •LipidsandsupernatantsfromstoredredbloodcellsactivatepulmonaryendotheliumthroughtheBLT2receptorandproteinkinaseCactivation,andpredisposetoALI77 •TransfusionofsupernatantofagedredbloodcellsinducedlunginflammationandcoagulopathyinLPS-primedrats109 •Injectionofanti–redbloodcellantibody1dbeforeinductionofanti-MHCclassIantibody–mediatedmurineTRALIresultedintotalsuppressionofTRALIdevelopmentinBALB/cmice67  •Transfusionof35-dstoredautologousredbloodcellsinLPS-primedhumanvolunteers(withconfirmedsepsis/endotoxemia)didnotresultinlunginjury89  Pathogenic CD4+CD25+FoxP3Tregsanddendriticcells •DepletionofCD4+CD25+FoxP3+TregsorofCD11c+dendriticcellsresultedinlowIL-10levels,whichenabledanti-MHCclassIantibody–mediatedmurineTRALI(protectionagainstanti-MHCclassIantibody–mediatedTRALIwasassociatedwithincreasedIL-10levels)52 Suggestedmechanism:DepletionofCD4+CD25+FoxP3+TregsorofCD11c+dendriticcellsinmiceresultsinlowIL-10levels,whichenablesincreasedanti-MHCclassIantibody–mediatedplasmaMIP-2levels(murinehomologofIL-8),pulmonaryPMNinfiltration,ROSproduction,andTRALIdevelopmentwithimpairedlungfunction52   Protective CD8+Tcells  DepletionofCD8+Tcellsdidnotsignificantlyresultintheonsetofanti-MHCclassIantibody–mediatedmurineTRALI52  Nosignificantinvolvement Bcells  DepletionofBcellsdidnotsignificantlyresultintheonsetofanti-MHCclassIantibody–mediatedmurineTRALI52  Nosignificantinvolvement ViewLarge Antibody-mediatedTRALI Antibody-mediatedmechanismsofTRALIaredepictedinpathwaysA-L(Figure2),andnon-antibody–mediatedmechanismsareillustratedinpathwaysMandNofFigure2.Usingananti–majorhistocompatibilitycomplex(MHC)classIantibody–mediatedmurinemodel,itwassuggestedthattheanti-MHCclassIantibodybindstothepulmonaryendotheliumandsequestersPMNsviatheirFcγRs,resultinginPMNactivation.66 Subsequently,ithasbeendescribedthatplateletsmayinduceNETformation,causingtheNETstoinducedirecttoxicitytothepulmonaryendotheliumenablingTRALI63 (Figure2pathwayA).Anothermechanismmaybethatanti-MHCclassIantibodiesmaybindthepulmonaryendothelialcellsactivatingthecomplementcascade,resultingintheproductionofC5a.67 C5amaythenbindtotheC5areceptoronmonocytes/macrophages,attractingthesecellstothelungsandinducingthemtoproducereactiveoxygenspecies(ROS),whichthendamagesthepulmonaryendothelium,resultinginTRALI67 (Figure2pathwayB).Importantly,ROSwasshowntobecriticallyrequiredforanti-MHCclassIantibody–meditatedmurineTRALIaswasdemonstratedusinggp91phoxknock-outmice.52 Anti-MHCclassIantibodiesmayalsotargetmonocytes,inducingsecretionofMIP-2,resultinginpulmonaryPMNrecruitmentandTRALI.68 CRPtogetherwithanti-MHCclassIantibodycanalsocauseasynergisticincreaseinbothMIP-2levelsandpulmonaryPMNaccumulation,enablingTRALI.56 CRPmayalsodirectlytargettheendothelium,69 andCRPhasbeendescribedtobeelevatedinTRALIpatients55 (Figure2pathwayC).Anti-HLA-A2antibodieswerealsoshowntoactivatePMNsandinduceendothelialcelldamageinvitro70 (Figure2pathwayD).InflammatoryprimingwasshowntoinduceMHCclassIIsurfaceexpressiononPMNsandactivationofthepulmonaryendothelium,enablinganti-MHCclassIIantibodiestodirectlytargetsequesteredPMNsinducingTRALIinaratmodel71 (Figure2pathwayE).Furthermore,anti-HLAclassIIantibodieswereshowntoinducemonocyteactivationinanexvivorodentTRALImodel,whichsubsequentlyresultedinPMNactivationthatthendamagedthepulmonaryendotheliumcausingTRALI72 (Figure2pathwayF).Anti–HNA-3aantibodieswereshowntodirectlyprimePMNs,resultingintheformationofNETs73 (Figure2pathwayG)aswellastocausePMNprimingenablingPMN-mediatedpulmonaryendothelialcelldamageinvitro74 (Figure2pathwayH).Alternatively,anti–HNA-3aantibodieshavebeenshowntodirectlytargetlungendotheliumcells,causingbarrierdysfunction,althoughthepresenceofPMNsdidaggravatethisendothelialbarrierdysfunction75 (Figure2pathwayI).Usingananti-MHCclassI–mediatedmurineTRALImodel,CD4+CD25+FoxP3+TregsanddendriticcellswerefoundtobekeyprotectingcellsinTRALIviaupregulationofIL-10.52 ThiswasdemonstratedbyinvivoTreganddendriticcelldepletions,whichcorrespondedtolowIL-10levelsandsubsequentlyenabledanti-MHCclassIantibody–mediatedTRALI.Inaddition,thiswasverifiedbyusingIL-10knock-outmice,whichwereshowntobehypersensitivetoanti-MHCclassIantibody–mediatedmurineTRALIinduction52 (Figure2pathwayJ).Redbloodcellsmayalsobepathogenicinantibody-mediatedTRALIbecauseRBCdepletionwithinjectionwithananti-RBCantibodypreventedtheonsetofanti-MHCclassIantibody–mediatedTRALIinBALB/cmice67 (Figure2pathwayK).Thegastrointestinalmicrobiotawasalsoshowntocontributetothedevelopmentofantibody-mediatedmurineTRALI.76 Micehousedinamoresterileenvironment(specificpathogen-freemice)wereshowntoberesistanttoanti-MHCclassI–mediatedTRALI,whereaslesssterilehousedmice(barrier-freemice)weresensitivetoanti-MHCclassIantibody–mediatedTRALIbutbecomeresistanttoTRALIupongutfloradepletionusingbroad-spectrumantibiotics.76 Interestingly,fecaltransplantsfromTRALI-susceptiblemiceintothesterile-housedTRALI-resistantmicewereabletorestoretheTRALIresponses.76 TheexactinterplaybetweenthegastrointestinalfloraandTRALIinductionmechanismswillneedtobefurtherinvestigated(Figure2pathwayL). Non-antibody–mediatedTRALI Non-antibody–mediatedmechanismsofTRALImayincludedirecttargetingofbiologicalresponsemodifierstowardthepulmonaryendothelium.LipidsfromstoredredbloodcellsmayactivatethepulmonaryendotheliumviatheBLT2receptorandproteinkinaseCpredisposingtoacutelunginjury.77 Inaddition,plasmaandlipidsfromstoredplateletsmayenableTRALIinanlipopolysaccharide(LPS)-primedratmodel,78 andstoredplatelet-derivedvascularendothelialgrowthfactormayalsopotentiallyincreaselungvascularpermeabilityinratlungs.79 AcidsphingomyelinasefromagedplateletshasalsobeenshowntoinduceTRALIinLPS-primedmice80 (Figure2pathwayM).BiologicalresponsemodifiersmayalsomediateTRALIviaPMNs,suchaslyso-PC,fromstoredplatelets,whichmaycausepulmonaryandsystemiccoagulopathyinLPS-primedmiceinvitroviaPMNpriming.81 Furthermore,bothplatelet-andredbloodcellmicroparticleshavebeendemonstratedtoprimePMNsandmediateTRALIinhumanpulmonarymicrovascularendothelialcellsinvitroandinmice,respectively.82,83 Inaddition,lipidsfromredbloodcellswereshowntoprimePMNsandinduceacutelunginjuryinLPS-primedmice.84 SolubleCD40ligand(sCD40L)hasbeendemonstratedtoaccumulateinstoredbloodcomponentsandenabledPMN-mediatedendothelialcelldamageinvitro85 (Figure2pathwayN).Notably,regardingtheroleofPMNsinthe2-hitTRALImodels,thesecellsmaydifferentiallyaccumulateinthelungsatvariousstagesbetweenthefirstandsecondhitsofTRALIinduction(Figure2pathwayC,E,J,orN). ImplicationsforhumanTRALIdisease ThedatasupportingthesemechanismsinhumancasesofTRALIarelimited.SupportingdatathatarederivedfromTRALIpatientsdirectlyincludethediscussedplasmacytokineprofiles(seeprevioussection),increasedlevelsofCRP,55 detectionofNETsinplasmaandlungsofTRALIpatients,63 presenceofNETbiomarkersinserumofTRALIpatients,73 andtheobservationofPMNsinTRALIlungsuponautopsy.86,87 Notably,thesuggestedmechanismsofnon-antibody–mediatedTRALIhavenotbeenconfirmedtooccurinhumanstudies.88,89  Potentialtherapiesandfuturedirections Unfortunately,forbothTACOandTRALI,onlysupportivemeasuresareavailable,andspecifictherapiesarelacking.SupportivemeasuresforTACOmayincludediuresis,oxygen,andintubation.ForTRALI,supportivemeasuresmayincludeoxygen,intubation,andjudiciousfluidandpressormanagementtomaintainhemodynamics.PreventivestrategiesforTRALI(TRALImitigation)includedonordeferralbasedonantibodyscreening(anti-HNAandanti-HLAantibodies),donordeferralbasedonhistoryofpregnancyorhistoryoftransfusion,anddeferralofallfemaledonors(useofmale-onlyplasmadonors).90 TheseTRALImitigationstrategieshavebeensuccessfulinsignificantlyreducingtheincidenceofTRALI.Therewasareductioninplasma-inducedTRALIcasesinat-riskpatientpopulations(surgery,intensivecareunit)andadecreasedtendencyofplasma-inducedTRALIcasesinageneralpatientpopulation.90 Inaddition,theFDAreported35TRALIcasesin2006,anduponTRALImitigation,thiswasreducedto8casesin2016.15 TheSHOTreportalsodescribedastrongdeclinefrom24casesin2003thatdroppedto3casesin2017afterintroductionofTRALImitigationstrategies.14 Furthermore,usingprospective,activesurveillance,ToyetalobservedaTRALIincidenceof2.57(95%confidenceinterval,1.72-3.86)in2006vs0.81(95%confidenceinterval,0.44-1.49)in2009per10 000transfusedunits(P=.002).49 Similarly,theincidenceofTACOinperioperativenoncardiacsurgerypatientswasobservedtobe5.5%(119/2162)in2004vs3.0%(57/1908)in2011(P<.001 recently figure3.viewlargedownloadpptpotentialtralitherapies.summaryofpotentialtherapeutictargetsintrali.ivig regardingpotentialtherapiesfortaco overall thepublicationcostsofthisarticleweredefrayedinpartbypagechargepayment.therefore acknowledgments thisworkwassupportedbygrantsfromlunduniversity authorship contribution:j.w.s.wroteandeditedthemanuscript conflict-of-interestdisclosure:theauthorsdeclarenocompetingfinancialinterests. thecurrentaffiliationforr.k.isdepartmentofexperimentalimmunohematology correspondence:johnw.semplelunduniversity references addcomment closecommentformmodal submitacomment name affiliations commenttitle comment youmustacceptthetermsandconditions. youhaveenteredaninvalidcode submit cancel thankyouforsubmittingacommentonthisarticle.yourcommentwillbereviewedandpublishedatthejournal close thisfeatureisavailabletosubscribersonly signin or createanaccount closemodal volume133 april252019 previousarticle nextarticle advertisement citations viewmetrics citedby webofscience googlescholar emailalerts articleactivityalert latestissuealert currentissue firstedition allissues collections abstracts authors submittoblood aboutblood subscriptions publicaccess permissions alerts contactus bloodclassifieds advertisinginblood termsandconditions twitter americansocietyofhematology washington tel fax ashpublications blood bloodadvances hematology ashclinicalnews ash-sap thehematologist ashhome research education advocacy meetings publications ashstore copyright privacypolicy cookiepolicy termsofuse signinorcreateanaccount x>