AbbreviationsAITautoimmunethyroiditisCVcoefficientofvariationSEstandarderrorT1Dtype1diabetesTGthyroglobulinTPOthyroperoxidaseTSHthyroidstimulatinghormonewwwarchdischildcom FromJanuary1 ID: 937003
Download Pdf The PPT/PDF document "ORIGINALARTICLENaturalcourseofautoimmune..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
ORIGINALARTICLENaturalcourseofautoimmunethyroiditisintype1diabetes:associationwithgender,age,diabetesduration,andOKordonouri,RHartmann,DDeiss,MWilms,AGru¨ters-KieslichSeeendofarticleforauthorsaffiliationsCorrespondenceto:DrOKordonouri,KlinikfuAllgemeinePa¨diatrie,Otto-Heubner-Centrum,Charite´Universita¨tsmedizinBerlin,CampusVirchow-Klinikum,Augustenburger Abbreviations:AIT,autoimmunethyroiditis;CV,coefficientofvariation;SE,standarderror;T1D,type1diabetes;TG,thyroglobulin;TPO,thyroperoxidase;TSH,thyroidstimulatinghormonewww.archdischild.com FromJanuary1990toDecember2003,659patients(358boys,301girls)withtype1diabetesunderwentregularscreeningforthyroiddysfunction.Theirmedianageatinitialscreeningexaminationwas10.9years(range0.1 24years),andtheirdiabetesduration1.2years(0 14.8years).Afurtherpatient(16yearoldgirl)withDownssyndrome,coeliacdisease,andclinicalsignsofhyperthyroidismwasdiagnosedtohaveGravesdisease.ThispatientwasexcludedfromtheInasubgroupof126childrenandadolescents(80boys,46girls),anti-TPOandanti-TGlevelswerefollowedatyearlyintervalsfromonsetofdiabetesuptofiveyears(medianageattype1diabetesonset9.0years,range0.1 15years)inordertoevaluateantibodyfluctuationsduringthatperiod.Antibodiestothyroglobulin(anti-TG)weredeterm-inedbyradioimmunoassay(DYNOtestanti-TG,BRAHMSDiagnostica,Berlin,Germany);intra-andinter-assaycoeffi-cientsofvariation(CV)were7.5%and5.5%,respectively.Thedetectionlimitwas5.5U/ml,andtheanalyticalassaysensitivity20U/ml.Atitreexceeding20U/mlwasconsideredpositive,andabove100U/mlwasconsideredsignificantlyincreased.Anti-TGmeasurementswereavailablein637of659patients(96.7%).Antibodiestothyroperoxidase(anti-TPO)weredeterminedradioimmunometrically(DYNOtestanti-TPO,BRAHMSDiagnostica);intra-andinter-assayCVwere4.3%and9.1%,respectively.Thedetectionlimitwas5.5U/mlandtheanalyticalassaysensitivity30U/ml.Atitreexceeding30U/mlwasconsideredpositive,above100U/mlassignificantlyincreased.Anti-TPOmeasurementswereavailablein637of659patients(96.7%).Theprevalenceofanti-TPOantibodiesinhealthychildreninBerlinhasbeenfoundtobe3.4%.StatisticalanalysisDatawereanalysedusingtheStatisticalPackageforSocialSciences(SPSS11.0.1).GroupdifferencesforcontinuousvariableswereassessedusingtheMann-WhitneyU-test.Differencesinfrequenciesforcategoricalvariablesweretestedbythetest.Dataarepresentedasmedian(range).Asameasureoftheprobabilitytodevelopanautoimmunethyroiditisafterdiabetesonset,cumulativeincidencewasusedandcalculatedusingtheKaplan-Meieranalysis;resultsaregivenasmean(SE).SignificantdifferenceswereassumedforpInthetotalgroup,98of637patients(15.4%)hadincreasedanti-TPOantibodiesand92of637patients(14.4%)werepositiveforanti-TGatthestartofscreening.Girlshadmorefrequentlyincreasedanti-TPOantibodiesthanboys(58of291(19.9%)40of346(11.6%),p=0.004)aswellasincreasedanti-TGantibodies(54of291(18.6%)38of346(11.0%),p=0.007).Duringthestudyperiod,62of659patients(9.4%)requiredtreatmentwithL-thyroxineandanother40patients(6.1%)hadthyroidpositiveantibodies,butdidnotmeetthecriteriaforL-thyroxinesubstitutionafteramediandiabetesdurationof3.9years(0.2 12.4years).Furthermore,26antibodypositivepatients(3.9%)withoutL-thyroxinesubstitutionmovedtoothertreatmentcentresafteramediandiabetesdurationof7.5years(1.2 19.5years)andwerelosttofollowupforfurtheranalysis.Ofthe62patientswhohadL-thyroxinetreatment,24wereboys(39%)and38girls(61%);theirmedianageatcommencementoftreatmentwas11.7years(1 19years),andtheirmediandiabetesduration3.4years(0 12years).Inthreeofthesepatients(oneboy,twogirls),hypothyroidism(increasedTSH,decreasedTandT)wasdiagnosed6,1,and56monthsbeforediabetesonset(ageattype1diabetesonset:4.0,11.2,and12.5years),respectively.AtthestartofL-thyroxinetreatment,59of62patients(95%)werepositiveforanti-TPO,and53(85%)forbothanti-TPOandanti-TG.Ultrasoundabnormalitieswerepresentinall62patientsrequiringL-thyroxinetreatment.IncreasedTSHlevels(med-ian9.2U/ml,4.7 100.0U/ml)werepresentin44ofthosepatients,andthyroidglandenlargementandtypicalultra-soundfindingswithoutTSHincreasein18patients(TSHU/ml,1.2 3.9Thecumulativeincid
ence(SE)ofAITattheageof18yearswas0.14(0.02)inthetotalgroup(n=659),beingsignificantlyhigherinfemale(0.19(0.03),n=301)thanmale(0.09(0.02),n=358,p=0.015)patients(fig1A).Atthebeginningofpuberty,aftertheageof12years,theAITincidenceingirlsincreasedmorethaninboys(fig1B).Thecumulativeincidence(SE)ofAITat10yearsofdiabetesdurationwas0.14(0.02)inthetotalgroup(n=659)andwassignificantlyhigheringirls(0.18(0.03),n=301)thaninboys(0.10(0.02),n=358,p=0.030)(fig2).Fortyfiveof98patients(46%)withpositiveanti-TPOmeasurementsduringthestudyperiodrequiredtreatmentwithL-thyroxine.Thecumulativeincidenceat10yearsofobservationtimewas0.69(0.08)comparedwith0.12(0.05)in539patientswithnegativeanti-TPOmeasurements0.001)(fig3A).Fortyoneof92patients(45%)withpositiveanti-TGmeasurementsdevelopedAIT.Thecumula-tiveincidenceat10yearsofobservationtimewas0.79(0.10) 0.300.200.000.1025Age (years)Cumulative incidence of AIT15 10 Total group 0.400.300.200.000.1025Age (years)Cumulative incidence of AIT15 10 FemaleMaleBp Figure1Cumulativeincidence(probability)ofautoimmunethyroiditisaccordingtotheageof659childrenandadolescentswithtype1diabetes:(A)inthetotalgroup,(B)stratifiedtogenderofpatients.412Kordonouri,Hartmann,Deiss,etalwww.archdischild.com comparedwith0.12(0.04)in545patientswithnegativeanti-TGmeasurements(p0.001)(fig3B).Inthesubgroupof126patientswithrepeatedantibodymeasurementssincetype1diabetesonset,23patients(18%)hadpositivethyroidantibodyvaluesalreadyattype1diabetesonset:19patientswithpositiveanti-TPO(15.1%)and19patientswithpositiveanti-TG(15.1%).Afterthreeyearsofdiabetes,afurtherthreepatients(2.4%)becamepositiveforbothantibodies;after5yearstherewasonepatient(0.8%)foranti-TPOandthreepatients(2.4%)foranti-TG.Allpatientswithsignificantlyincreasedbaselinevaluesofanti-TPO(n=17,148 5340U/ml)andanti-TG(n=11,140 2000U/ml)attype1diabetesonsetremainedhighlypositiveduringthefirstfiveyearsofdiabetes.Inthissubgroupof30patientswithpositiveantibodies,19patients(63%)requiredtreatmentwithL-thyroxine.DISCUSSIONThesedataclearlyshowthattheincidenceofautoimmunethyroiditisrequiringtreatmentwithL-thyroxineishighlyincreasedinchildrenandadolescentswithtype1diabetescomparedtothegeneralpopulation.Thecumulativeincidenceat10yearsofdiabeteswasfoundtobe14%.Inpriorstudies,thereportedfrequencyofHashimotosthy-roiditiswas3.5%(cross-sectionalstudy)and10%(mixedcross-sectionalandlongitudinalstudy),respectively.Signsofthyroidautoimmunity,suchasthepresenceofthyroidautoantibodies,wereapparentalreadyatdiabetesonset;however,thyroiddysfunctionrequiringtreatmentwithL-thyroxinedevelopedmostlythereafter.Inourstudy,progressiontoclinicalorsubclinicalhypothyroidismrequir-ingtreatmentwithL-thyroxineoccurredduringthefirstfiveyearsofdiabetesinmostpatientswithAIT,whilenosignificantfluctuationsofthyroidantibodytitreswereAsinpreviousstudies,10 13femaleswithtype1diabetesweresignificantlypredisposedtodevelopautoimmunethyroiditis.Attheageof18years,almosteveryfifthgirlwithtype1diabeteswasdiagnosedwithanautoimmunethyroiditisrequiringtreatmentwithL-thyroxine.Inthegeneralpopulation,girlsarealsopronetodevelopthyroiddiseasemorethanboys;however,itisnoteworthy,thatmostlyfemalepatientswithtype1diabetesaremorelikelytodevelopanotherautoimmunedisease,likeautoimmunemultiplesclerosis,andcoeliacdisease.1617Giventhattype1diabetesisequallyprevalentamongmalesandfemalesinmostCaucasianpopulationsorevenahighermale-to-femaleratioisshownincountrieswithahighdiabetesincidence;thisunequalgenderdistributionsug-geststhataetiologicalriskfactors,unrelatedtothosefortype1diabetes,maybeassociatedwiththesimultaneouspresenceofmorethanoneautoimmunedisease.Inpreviousstudies,weshowedthattheprevalenceofthyroidantibodiesinpatientswithtype1diabetesincreased 0.300.200.000.1015Diabetes duration (years)Cumulative incidence of AIT10 Total group 0.400.300.200.000.1015Diabetes duration (years)Cumulative incidence of AIT10 Figure2Cumulativeincidence(probability)ofautoimmunethyroiditisaccordingtothedurationoftype1diabetesin659childrenandadolescents:(A)inthetotalgroup,(B)stratifiedtogenderofpatients
. Observation time (years) anti-TPO positiveanti-TPO negativeanti-TG positiveanti-TG negativeA Observation time (years) 8 8 6 6 4 4 2 2 Figure3Cumulativeincidence(probability)ofautoimmunethyroiditisin637patientswithtype1diabetesaccordingtothepresenceofantibodiesagainst(A)thyroperoxidase(anti-TPO)and(B)thyroglobulin(anti-TG).NaturalcourseofAITintype1diabeteswww.archdischild.com withage.1012Inthepresentstudy,thesametrendwasfoundfortheprevalenceofthyroiddysfunctionrequiringtreatmentwithL-thyroxine,whilethedurationoftype1diabetesseemedtohavenoadditionallyindependentinfluenceonthedevelopmentofclinicalthyroiditis.Interestingly,inpatientswithtype1diabetesandanotherautoimmunedisease,diabetesonsetmostlyprecedesthediagnosisoftheother1920Similarly,inthepresentstudy,exceptofthreesubjects,allpatientsdevelopedclinicallyrelevantthyroiditisafterthemanifestationofdiabetes.Thepathogeneticmechanismunderlyingthesimultaneousoccurrenceoftheseclusteredautoimmunediseaseshasnotbeenclearlyeluci-dated,eventhoughthereisevidencethatcommongeneticdeterminants,inparticularsharingofHLAriskallelesothergenesoutsidetheHLAregion(2425couldbeinvolved.Moreover,environmentalfactorsarealsoassumedtobeinvolvedinthepathogenesisofthesecomplexWhethertheinfluenceofexogenousagentsleadstoafasteronsetoftheseautoimmunediseases,andparticularlyoftype1diabetes,inpatientswithmultiplediseasescomparedtothosewithonlyonediseaseneedstobeassessedbyepidemiologicalstudies.Despitethestrikingevidencethatthyroidautoimmunitywithsubclinicalhypothyroidism(increasedTSH,normalTandT,abnormalultrasoundimages)isafrequentfindinginchildrenandadolescentswithtype1diabetes,thereisstillcontroversyconcerningthenecessityoftherapeuticinterven-tioninthesepatients.Inasmallobservationalstudyof18childrenandadolescentswithautoimmunethyroiditisandsubclinicalhypothyroidism,sevenpatientswereeuthyroid,onepatientbecamehypothyroid,and10patientscontinuedtohaveincreasedTSHlevelsafteranobservationperiodof47.3months.However,inchildrenwithtype1diabetes,thereisstrongevidencethathormonalthyroidabnormalities,evenatasubclinicalstage,mayinterferewithglycaemicmetaboliccontrolandincreaseinsulinrequirements.2829Moreover,asignificantlyreducedgrowthratewasreportedinyoungpatientswithdiabetesandsubclinicalhypothyroidismwiththyromegaly,particularlywhenTSHlevelswerehigherthan10mU/l,whilethyroidhormonereplacementledtoimprovedgrowth,especiallyinprepubertalpatients.However,uptonow,inchildrenwithsubclinicalhypothyroidism,bothwithandwithouttype1diabetes,therehasbeenalackoflongitudinalrandomisedstudyprotocols,toexaminethebenefitsofearlytreatmentwithL-thyroxine.Inconclusion,thesedatashowthatthedevelopmentofautoimmunediseaserequiringtreatmentwithL-thyroxineissignificantlyincreasedamongyoungpatientswithtype-1diabetes.Beforetheageof18years,approximately14%ofpatientswillrequiretreatmentforthyroiddisease,whilegirlsaremorepredisposedthanboys.Thefirstsignoftheseconditionsmaybethepresenceofpositivethyroidauto-antibodies,particularlyanti-TPO,alreadyatdiabetesonset.TheprogressiontoTSHincreaseand/orultrasoundabnorm-alitiesmostlyoccurswithinfiveyears.However,developmentofthyroidautoimmunitymayalsooccurduringthefurthercourseofdiabetes,particularlyduringpuberty.Therefore,wewouldrecommendthatscreeningforthyroiddiseaseshouldbeperformedatdiabetesonsetinallpaediatricpatientswithtype1diabetes.Iftheinitialthyroidscreeningispositive,yearlylaboratory(anti-TPO,anti-TG,TSH,T,andT)andultrasoundexaminationsarenecessaryinordertodetectearlythyroiddysfunctionandinitiatetreatmentwithL-thyroxine.Iftheinitialthyroidscreeningisnegative,werecommendregularmeasurementsofanti-TPOandTSH,evenintheabsenceofclinicalsigns.Particularlyfromtheageof12yearsorfromtheonsetofpuberty,TSHandantibodyscreeningshouldbeperformedatyearlyintervals.AuthorsaffiliationsOKordonouri,RHartmann,DDeiss,MWilms,AGru¨ters-Kieslich,ClinicofGeneralPediatrics,Otto-Heubner-Centrum,Charite´,CampusVirchow-Klinikum,HumboldtUniversity,Berlin,GermanyCompetinginterests:nonedeclaredREFERENCESRadettiG,PaganiniC,GentiliL,etal.FrequencyofHashimotosthyro
iditisinchildrenwithtype1diabetesmellitus.ActaDiabetol:121 4.LoriniR,dAnnunzioG,VitaliL,etal.IDDMandautoimmunethyroiddiseaseinpediatricagegroup.JPediatrEndocrinolMetab(suppl1):89 94.DayanCM,DanielsGH.Chronicautoimmunethyroiditis.NEnglJMed1996;:99 107.MarcocciC,VittiP,CetaniF,etal.Thyroidultrasonographyhelpstoidentifypatientswithdiffuselymphocyticthyroiditiswhoarepronetodevelophypothyroidism.JClinEndocrinolMetab1991;Consensusguidelinesforthemanagementoftype1diabetesmellitusinchildrenandadolescents.Zeist,Netherlands:Medforum,2000.¨tterKP,KieblerA,StachB,etal.SmallthyroidvolumesandnormaliodineexcretioninBerlinschoolchildrenindicatefullnormalizationofiodinesupply.ExpClinEndocrinolDiabetes(suppl4):46 50.KabelitzM,Liesenko¨tterKP,StachB,etal.Theprevalenceofanti-thyroidperoxidaseantibodiesandautoimmunethyroiditisinchildrenandadolescentsinaniodinerepletearea.EurJEndocrinol2003;148TunbridgeWM,BrewisM,FrenchJM,etal.Naturalhistoryofautoimmunethyroiditis.BMJ(ClinResEd)VanderpumpMP,TunbridgeWM.Epidemiologyandpreventionofclinicalandsubclinicalhypothyroidism.Thyroid2002;HollRW,Bo¨hmB,LoosU,etal.Thyroidautoimmunityinchildrenandadolescentswithtype1diabetesmellitus.Effectofage,genderandHLAtype.HormRes:113 18.HansenD,BennedbaekFN,HansenLK,etal.Thyroidfunction,morphologyandautoimmunityinyoungpatientswithinsulin-dependentdiabetesmellitus.EurJEndocrinol1999;140:512 18.KordonouriO,KlinghammerA,LangEB,etal.Thyroidautoimmunityinchildrenandadolescentswithtype1diabetesmellitus:amulticentersurvey.DiabetesCare:1346 50.KordonouriO,DeissD,DanneT,etal.Predictivityofthyroidautoantibodiesforthedevelopmentofthyroiddisordersinchildrenandadolescentswithtype1diabetes.DiabetMed:1346 50.McCanliesE,OLearyLA,FoleyTP,etal.Hashimotosthyroiditisandinsulin-dependentdiabetesmellitus:differencesamongindividualswithandwithoutabnormalthyroidfunction.JClinEndocrinolMetab1998;DormanJS,SteenkisteAR,BurkeJP,etal.Type1diabetesandmultiplesclerosis:togetheratlast.DiabetesCareValerioG,MaiuriL,TronconeR,etal.Severeclinicalonsetofdiabetesandincreasedprevalenceofotherautoimmunediseasesinchildrenwithcoeliacdiseasediagnosedbeforediabetesmellitus.2002;AceriniCL,AhmedML,RossKM,etal.CoeliacdiseaseinchildrenandadolescentswithIDDM:clinicalcharacteristicsandresponsetogluten-freediet.DiabetMed1998;KarvonenM,PitkaniemiM,PitkaniemiJ,etal.Sexdifferenceintheincidenceofinsulin-dependentdiabetesmellitus:ananalysisofrecentepidemiologicalDiabetesMetabRev:275 91.CroninCC,ShanahanF.Insulin-dependentdiabetesmellitusandcoeliacdisease.349:1096 7.PoceccoM,VenturaA.Coeliacdiseaseandinsulin-dependentdiabetesmellitus:acausalassociationActaPaediatr:1432 3.HoulstonRS,TomlinsonIP,FordD,etal.Linkageanalysisofcandidateregionsforcoeliacdiseasegenes.HumMolGenet1997;:1335 9.`kZ,KolouskovaS,CinekO,etal.05-DQB1positivitypredisposestocoeliacdiseaseinCzechdiabeticchildren.Paediatr2000;BadenhoopK,WalfishPG,RauH,etal.Susceptibilityandresistanceallelesofhumanleukocyteantigen(HLA)DQA1andHLADQB1aresharedinendocrineautoimmunediseases.JClinEndocrinolMetab1995;NaluaiAT,NilssonS,SamuelssonL,etal.TheCTLA4/CD28generegiononchromosome2q33conferssusceptibilitytocoeliacdiseaseinawaypossiblydistinctfromthatoftype1diabetesandotherchronicinflammatorydisorders.TissueAntigens2000;UedaH,HowsonJMM,EspositoL,etal.AssociationoftheT-cellregulatorygeneCTLA4withsusceptibilitytoautoimmunedisease.2003;:506 11.LeslieRD,ElliottRB.EarlyenvironmentaleventsasacauseofIDDM:evidenceandimplications.1994;MooreDC.Naturalcourseofsubclinicalhypothyroidisminchildhoodandadolescence.ArchPediatrAdolescMedFranzeseA,BuonoP,MascoloM,etal.Thyroidautoimmunitystartingduringthecourseoftype1diabetesdenotesasubgroupofchildrenwithmoreseverediabetes.DiabetesCare2000;MohnA,DiMicheleS,DiLuzioR,etal.Theeffectofsubclinicalhypothyroidismonmetaboliccontrolinchildrenandadolescentswithtype1diabetesmellitus.DiabetMed:70 3.ChaseHP,GargSK,CockerhamRS,etal.Thyroidhormonereplacementandgrowthofchildrenwithsubclinicalhypothyroidismanddiabetes.DiabetMed1990;:299 303.414Kordonouri,Hartmann,Deiss,etalwww.archdischild.c