Detection of subclinical atrial dysfunction by two-dimensional echocardiography in patients with overt hyperthyroidism

Availableonlineat

www.sciencedirect.com

CLINICAL

RESEARCH

Detection

of

subclinical

atrial

dysfunction

by

two-dimensional

echocardiography

in

patients

with

overt

hyperthyroidism

Détection

d’une

dysfonction

auriculaire

infra-clinique

par

échocardiographie

bidimensionnelle

chez

des

patients

hyperthyroïdiens

Selim

Ayhan

,

Serkan

Ozturk

,

O˘

guz

Dikbas

,

Alim

Erdem

,

Mehmet

Fatih

Ozlu

,

Davut

Baltaci

,

Aytekin

Alc

¸elik

,

Mehmet

Tosun

,

Mehmet

Ozyasar

,

Mehmet

Yazici

a_{DepartmentofCardiology,FacultyofMedicine,AbantIzzetBaysalUniversity,14280Golkoy,}

Bolu,Turkey

b_{DepartmentofEndocrinology,FacultyofMedicine,AbantIzzetBaysalUniversity,Golkoy,}

Bolu,Turkey

c_{DepartmentofFamilyMedicine,FacultyofMedicine,DuzceUniverstity,Duzce,Turkey}

d_{DepartmentofInternalMedicine,FacultyofMedicine,AbantIzzetBaysalUniversity,}

Golkoy,Bolu,Turkey

e_{DepartmentofBiochemistry,FacultyofMedicine,AbantIzzetBaysalUniversity,Golkoy,}

Bolu,Turkey

Received29May2012;receivedinrevisedform5July2012;accepted6July2012 Availableonline5October2012

KEYWORDS Atrialfunction; Atrialconduction time;

Summary

Background.—Hyperthyroidismisanimportantcardiovascularriskfactorinthedevelopment ofatrialfibrillationandheartfailure.Increasedatrialelectromechanicalintervalsareusedto predictatrialfibrillation,measuredbytissueDopplerimaging(TDI).

Aims.—Toevaluateatrialelectromechanicaldelay(EMD)andleftatrial(LA)mechanical func-tioninpatientswithoverthyperthyroidism.

Abbreviations:2D,Two-dimensional;Am,Latediastolicvelocity;BMI,Bodymassindex;BSA,Bodysurfacearea;ECG,

Electrocardiogram;Em,Earlydiastolicvelocity;EMD,Interatrialelectromechanicaldelay;fT3,FreeT3;fT4,FreeT4;LA,Leftatrial;LV,

Leftventricular;PA,Atrialelectromechanicalcoupling;Sm,Peaksystolicvelocity;TDI,TissueDopplerimaging;TSH,Thyroid-stimulating

hormone;Vmax,LAmaximumvolumeattheend-systolicphase;Vmin,LAminimumvolume

attheend-diastolicphase;Vp,LAvolumebeforeatrialsystole.

∗_{Correspondingauthor.Fax:+903742534615.}

E-mailaddress:drserkan69@hotmail.com(S.Ozturk).

1875-2136/$—seefrontmatter©2012ElsevierMassonSAS.Allrightsreserved.

Interatrialdelay; Hyperthyroidism

Methods.—Thirty-fourpatientswithoverthyperthyroidismand34controlswereincluded.A diagnosisofoverthyperthyroidismwasreachedwithdecreasedserumthyroid-stimulating hor-mone(TSH)andincreasedfreeT4(fT4) concentrations.UsingTDI, atrialelectromechanical coupling(PA) wasobtained fromthelateralmitralannulus(PA lateral),septalmitral annu-lus(PAseptum)andrightventriculartricuspidannulus(PAtricuspid).LAvolumes(maximum, minimumandpresystolic)weremeasured by thedisks methodinapical four-chamberview andindexedtobodysurfacearea.LAactiveandpassiveemptyingvolumesandfractionswere calculated.

Results.—LAdiameterwassignificantlyhigherinhyperthyroidpatients(P=0.001).LApassive emptyingvolumeandfractionweresignificantlydecreasedinhyperthyroidpatients(P=0.038 andP<0.001).LAactiveemptyingvolumeandfractionweresignificantlyincreasedin hyper-thyroidpatients (P<0.001 and P<0.001).Left and right intra-atrial(PA lateral—PA septum andPAseptum—PAtricuspid)andinteratrial(PAlateral—PAtricuspid)EMDsweresignificantly higherinhyperthyroidpatients (29.2±4.4vs 18.1±2.6,P<0.001;18.7±4.3vs 10.6±2.0,

P<0.001;and10.5_±2.9vs7.1_±1.2,P<0.001,respectively).Stepwiselinearregression anal-ysisdemonstratedthatfT4andTSHconcentrationswereindependentpredictorsofinteratrial EMD(␤=0.436,P<0.001and␤=—0.310,P=0.005,respectively).

Conclusion.—ThisstudyshowedprolongedatrialelectromechanicalintervalsandimpairedLA mechanicalfunctioninpatients with overthyperthyroidism, whichmay be anearlysignof subclinicalcardiacinvolvementanddysrhythmiasinoverthyperthyroidism.

©2012ElsevierMassonSAS.Allrightsreserved.

MOTSCLÉS Fonctionauriculaire; Tempsdeconduction atrial; Délaiinter-atrial; Hyperthyroïdie Résumé

Objectif.— L’hyperthyroïdie est un facteur de risque important de développement d’une fibrillationatriale(FA)etd’insuffisancecardiaque.L’augmentationdesintervalles électromé-caniquesatriaux,prédisantlaFAsontmesuréesenDopplertissulaire(DTI).L’objectifdecette étude est d’évaluer le délai électromécanique auriculaire (EMD) et la fonction mécanique auriculairegauche(OG)chezdespatientshyperthyroïdiens.

Méthode.—Trentre-quatrepatientsenhyperthyroïdie et34témoins ontétéinclus. Le diag-nosticd’hyperthyroïdieaété retenusurunediminution dela concentrationenTSH etune augmentationdelaT4libre(fT4).EnutilisantleDTI,lecouplageélectromécaniqueauriculaire (PA)aétémesuréàl’anneaumitrallatéral(PAlatéral),àl’anneaumitralseptal(PAseptum) etàl’anneautricuspide(PAtricuspide).Lesvolumesauriculairesgauches(maximum,minimum etprésystolique)ontétémesurésparlaméthodedesdisques,enincidenceapicaledesquatre cavités,etindexésàlasurfacecorporelle.Lesvolumesdevidangedel’oreillettegauchelors desphasesactiveetpassiveainsiquelesfractionsontétécalculées.

Résultats.—Lediamètredel’oreillettegaucheestsignificativementplusimportantchezles patients hyperthyroïdiens (p=0,001).Le volume devidange passif auriculaire gauche et la fraction étaient significativement diminués chez les patients hyperthyroïdiens (p=0,038et

p<0,001).Levolumedevidangeactifauriculairegaucheetlafractionétaientsignificativement augmentés chezleshyperthyroïdiens (p<0,001et p<0,001).Les délaisintra-atriaux auricu-lairegaucheetdroit (PAlatéral—PAseptumetPAseptum—PAtricuspide,respectivement) ainsique lesdélais inter-atriaux (PAlatéral— PA tricuspide).Les délais électromécaniques auriculairesétaientsignificativementplusélevéschezleshyperthyroïdiens(29,2±4,4versus 18,0±2,6, p<0,001; 18,7±4,3versus 10,6±2,0, p<0,001; et 10,5±2,9versus 7,1±1,3,

p<0,001).L’analyseparrégression linéaireamontré quela T4libre etla concentration en TSHétaientdesprédicteursindépendantsdudélaiélectromécaniqueatrialdanssacomposante inter-auriculaire(_{␤ =}0,388,p=0,001et_{␤ =}0,547,p<0,001).

Conclusion.— Cette étude montre que les intervalles électromécaniques auriculaires sont prolongésetquelafonctionmécaniqueauriculairegaucheestaltéréechezlepatient hyper-thyroïdien. L’allongement des intervalles électromécaniques et l’altération de la fonction mécaniqueauriculairegauchepourraientêtreunsigneprécoced’uneatteintecardiaque infra-cliniqueetderisqueaccrud’arythmiechezdespatientshyperthyroïdiens.

©2012ElsevierMassonSAS.Tousdroitsréservés.

Background

Thyroid hormone receptors are highly expressed in the myocardium[1]. Consequently, overt hyperthyroidismhas

profound cardiac effects, including increased heart rate, arrhythmias, LV diastolic dysfunction [2], LV systolic dys-function[3],decreasedsystemicvascularresistance[4]and thedevelopmentofatrialfibrillationandheartfailure[5,6].

Hyperthyroidismcanleadtodecreaseddiastolicfunctionas aresultofimpairedmyocardialrelaxation[1,2].Moreover, hyperthyroidismis associatedwithLVdiastolicdysfunction duetodecreasedmyocardialrelaxationandinhibitedrapid ventricularfilling,bothatrestandwithexercise[1,7];thisis frequentlyrelatedtoachangeableimpairmentofLVsystolic function. Thus, hyperthyroidism may result in significant cardiovascular alterations, includingatrial fibrillation and heartfailure[5,6].

Inastudy,theatrialsystolictimeinterval,atrialejection timeandatrialpre-ejectionperiodwereusedtoevaluate atrialfunction.Inaddition,therelationshipbetween hyper-thyroidismandatrialfunctionwasexamined[8].LAvolume and mechanical function has recently been identified as a potential indicator of cardiac disease and arrhythmias

[9—11]. Inter- and intra-atrial conduction disorders are

well-knownelectrophysiologicaldistinctionsof atriaprone tofibrillation [12]. Unlike LA size,atrial conduction time reflectsthelevelsofbothelectricalandstructural remod-elling of the atria. All of these variables are similar and resultinimpairedatrialfunction[12—14].

Alterations in thyroid status may lead to changes in bothventricularandatrialfunction.However,LA mechani-calfunction and atrial conduction abnormalities have not been investigated in overt hyperthyroidism. The aim of the present study was toevaluate atrial electromechani-caldelayandLAmechanicalfunctioninpatientswithovert hyperthyroidism.

Methods

Study

population

Westudied34newly treatedoruntreatedpatients, previ-ouslydiagnosedwithoverthyperthyroidism(25womenand nine men; mean age 45.2±9.3years). The control group consisted of 34 sex- and age-matched healthy subjects (24womenandtenmen;meanageof43.6±8.0years)who werefreeofendocrinological,inflammatory,connective tis-sue, cardiovascular, pulmonary and other known systemic disease,andwereadmittedtohospitalfor acheck-up.All subjectsinthestudypopulationhadnormalsinusrhythmon ECG.Thestudywasdesignedascross-sectional.

A diagnosis of overthyperthyroidismwasreached with decreased serum TSH concentrations, increased fT4 con-centrationsand/or increasedfT3concentrations infasting blood samples (normal values in our laboratory were: 0.4—4.0mIU/mL for TSH; 0.89—1.76ng/dL for fT4; and 1.80—5.00pg/mL for fT3). Demographic characteristics, biochemicalvariables,lipidvaluesandECGswereobtained forthe entirestudypopulation. Exclusioncriteriawereas follows: subclinical hyperthyroidism, acute coronary syn-drome, prior myocardial infarction and coronary artery disease,congestiveheartfailure,LVhypertrophy,prolonged QRS duration (≥120ms), reduced LV ejection fraction (<55%),atrialflutterorfibrillation,significantvalvularheart disease,pacemakerimplantation,frequentventricular pre-excitation and atrioventricular conduction abnormalities, diabetesmellitus,arterialhypertension(restingblood pres-sure≥140/90mmHg), medications knownto alter cardiac conduction, peripheral vascular disease, congenital heart

disease, pulmonary or neurological disease, pericarditis, peripheralneuropathy,alcoholabuse,renalorhepatic dis-easeandpoorechocardiographicimaging.Approvalforthe studywasobtainedfromthelocalethicscommitteeandall subjectsgaveinformedconsent.

Standard

echocardiography

Atstudyentry,allpatientswereevaluatedbytransthoracic, M-mode, 2D, pulsed-wave, continuous-wave, colour-flow andTDI. Echocardiographic examinationswere performed with the GE Vivid-7 system (GE Vingmed, Horten, Nor-way)witha2—4MHztransduceratadeptof16cm.During echocardiography,a continuous single-lead ECG recording wasobtained.All patientswereimagedintheleftlateral decubitusposition.2Dandconventional Doppler examina-tions were obtained in the parasternal and apical views accordingtotheguidelinesoftheAmericanSocietyof Echo-cardiography [15]. LV diameters and wall thickness were measured by M-mode echocardiography. LV ejection frac-tionwascalculatedusingtheapicaltwo-andfour-chamber views by Simpson’s method, according to American Soci-ety of Echocardiography guidelines[15]. The mitral valve inflowpattern(E-wave,A-wave,E-wavedecelerationtime, E/Aratio and isovolumic relaxation time)were measured usingpulsed-wave Doppler. LV mass index wascalculated usingtheDevereuxequation[16].LAvolumeswereobtained echocardiographicallyfromtheapical four-chamberviews bythe disksmethod [17,18].LA Vmaxat the end-systolic phase(onsetof themitral opening),LA Vminat the end-diastolicphase(onset of themitral closure) andVp were measuredatthebeginningofatrialsystole(onsetofPwave onECG)andindexedtoBSA.LAfunctionvariableswere cal-culatedasfollows:LApassiveemptyingvolume=Vmax—Vp; LApassive emptying fraction=([Vmax—Vp]/Vmax)×100%; LAactiveemptyingvolume=Vp—Vmin;LAactiveemptying fraction=([Vp—Vmin]/Vp)×100%[12].

Atrial

electromechanical

interval

TDIwasperformed usingtransducer frequenciesof 3.5 to 4.0MHz,adjustingthespectralpulsedDopplersignalfilters toacquiretheNyquistlimitof15to20cm/sandusingthe minimaloptimalgain.MyocardialTDIvelocities(Sm,Emand Amvelocities)weremeasuredviaspectralpulsedDopplerof theLVfreewallfromtheapicalfour-chamberview[15].The ultrasoundbeamwaspositionedasparallelaspossibletothe myocardialsegmenttoacquiretheoptimalangleof imag-ing.Themonitorsweepspeedwassetat50to100mm/sto optimizethespectraldisplayofmyocardialvelocities.The timeintervalfromthePwaveonsetonthesurfaceECGto thebeginningof theAm isdefined asPA; itwasobtained fromthelateralmitralannulus, theseptalmitralannulus andtheright ventriculartricuspidannulus,andnamed PA lateral(Fig.1),PA septum,andPAtricuspid,respectively. The difference between PA lateral and PA tricuspid was definedastheEMD,thedifferencebetweenPAlateralandPA septumwasdefinedastheleftintra-atrialEMDandthe dif-ferencebetweenPAseptumandPAtricuspidwasdefinedas therightintra-atrialEMD[12—15].Allmeasurementswere repeatedthreetimesandaveragevalueswereobtainedfor eachoftheatrialconductiondelaytimes.Allmeasurements

Figure1. ThetimeintervalfromthePwaveonsetonthesurface

electrocardiogramtothebeginningofthelatediastolicwave(Am),

whichisdefinedasatrialelectromechanicaldelay.

wereperformedbytwoexperiencedinvestigatorswhowere unawareofthesubject’sclinicalstatus.

Reproducibility

Intraobserver variability was evaluated in 20subjects selected randomly from the study population by repeat-ingthemeasurementsunderthesamebasalconditions.To testinter-observervariability,themeasurementswere per-formedofflinefromvideorecordingsbyasecondobserver. ReproducibilityofatrialelectromechanicalcouplingandLA volumesobtainedby2D echocardiographywereevaluated bythecoefficientofvariationbetweenmeasurements.

Intraobservervariabilitywas4.1%forPAlateral,4.5%for PAseptum,4.8%forPAtricuspid,4.7%forVmax,4.3%forVp and4.7%forVmin.Inter-observervariabilitywas4.0%forPA lateral,4.3%forPAseptum,4.5%forPAtricuspid,4.9%for Vmax,4.6%forVpand4.5%forVmin.

Statistical

analysis

All analyses were performed using the SPSS (SPSS for Windows 15.0) software package. Continuous variables arepresented asmeans±standarddeviations. Categorical variablesarepresentedaspercentages.Thechi-squaretest wasusedforcategorical variablesandtheunpairedttest was used for continuous variables. Pearson’s correlation coefficientwasusedforcorrelationanalysis.Astepwise mul-tipleregressionanalysiswasusedtorecognizethesignificant determinantsofinteratrialEMD,whichincorporated varia-blesthatcorrelatedwithavalueofP<0.1inthecorrelation analysis.AvalueofP<0.05wasconsideredstatistically sig-nificant.

Results

Patient

characteristics

The clinical andlaboratory characteristics and echocardi-ographicfindingsforthetwogroupsareshowninTable1. Age, sex, smoking, systolic and diastolic blood pressure,

LVend-diastolicandend-systolicdiameters,LVmassindex and LV ejection fraction were similar in the two groups (P>0.05). BMI and BSA were significantly lower in hyper-thyroidpatients(P=0.023andP=0.001,respectively).So, asexpected,hyperthyroidpatients hadsignificantlylower TSHandhigherfT4andfT3concentrationscomparedwith controls. Heart rate and LA diameter were significantly higher in hyperthyroidpatients than in controls (P=0.001 andP=0.001,respectively).Also,Amvelocity,Em/Amand E/E’ratioweresignificantlylowerinhyperthyroidpatients (P=0.035,P=0.004andP=0.002,respectively).

Left

atrial

mechanical

function

Measurements of LA volume and mechanical function are presentedinTable2.Bothgroupsweresimilarinterms of Vmax andVmin (P=0.212 andP=0.203, respectively) but Vp was significantly higher in hyperthyroid patients than in controls (P=0.001). Also, LA passive emptying volume and fraction were significantly decreased in hyperthyroid patients(P=0.038andP<0.001,respectively).Additionally, LAactiveemptyingvolumeandfractionweresignificantly increased inhyperthyroid patients compared within con-trols(P<0.001andP<0.001,respectively).

Atrial

electromechanical

intervals

Atrialelectromechanicalcouplingintervalsarepresentedin

Table3.PAlateral,PAseptum,PAtricuspidandintra-and

interatrial EMDs were significantly prolongedin hyperthy-roidpatientscomparedwithinhealthycontrols(P<0.001,

P<0.001, P=0.001, P<0.001 and P<0.001, respectively). Interatrial EMD and left intra-atrial EMD were positively correlatedwithfT4concentration (r=0.628, P<0.001 and r=0.772,P<0.001,respectively)andwerenegatively cor-related with TSH concentration (r=—0.379, P=0.035 and r=—0.726,P<0.001,respectively;Fig.2).Additionally,left intra-atrialEMDwascorrelatedwithEm/AmandE/E’ratio (r=—0.282,P=0.029andr=0.324,P=0.011,respectively). Also,therewasaslightpositivecorrelationbetweenright intra-atrialEMDandfT4concentration(r=0.365,P=0.043) but right intra-atrial EMD was not correlated with TSH concentration(r=—0.260,P=0.158).

Stepwise linear regression analysis demonstrated that LA passive and active emptying fractions and fT4 and TSHconcentrationsweresignificantlyrelatedtointeratrial EMD (␤=—0.155, P=0.032; ␤=0.189, P=0.007; ␤=0.436,

P<0.001and␤=—0.310,P=0.005,respectively).However, therewasnorelationshipbetweenage,BMI,BSA,heartrate, LApassiveandactiveemptyingvolumes,LAdiameter,fT3 orEm/AmratioandinteratrialEMD(Table4).

Discussion

The present study demonstrates that patients with overt hyperthyroidism display lengthened intra- and interatrial EMDasmeasuredbyTDIandimpairedLAmechanical func-tion.Furthermore,wehaveshownthatintra-andinteratrial EMDs correlate with fT4 concentration. Additionally, we found that fT4 and TSH concentrations are independent

Table1 Patientdemographics,laboratorycharacteristicsandechocardiographicfindings. Hyperthyroid (n=34) Control (n=34) P Age 45.2±9.3 43.6±8.0 0.328 Women 25(73.5) 24(70.5) 0.749 Smoker 10(26.3) 7(23.3) 0.620 BMI(kg/m2_{) 26.8±2.0 28.1±2.4 0.023} BSA 1.71±0.13 1.88±0.15 0.001

Heartrate(beats/minute) 83.5±8.9 74.3±6.5 0.001

Systolicbloodpressure(mmHg) 119.8±8.7 118.3±7.9 0.301

Diastolicbloodpressure(mmHg) 77.8±7.4 78.1±8.0 0.359

Glucose 95.6±8.1 97.7±11.8 0.417 Haemoglobin(g/dL) 12.9±1.2 13.4±1.2 0.531 Creatinine(mg/dL) 0.80±0.17 0.81±0.16 0.412 TSH(␮IU/mL) 0.02±0.04 2.04±0.43 <0.001 FreeT3(pg/dL) 5.48±1.75 2.63±0.41 <0.001 FreeT4(ng/dL) 2.08±0.42 1.06±0.16 <0.001 LVEDD(mm) 49.0±3.5 48.7±2.9 0.435 LVESD(mm) 30.8±3.9 30.2±2.8 0.513 Ejectionfraction(%) 63.9±6.7 64.2±5.1 0.679 LVmassindex(g/m2_{) 82.5±21.8 78.9±14.5 0.342} Septumthickness(mm) 9.9±1.1 9.7±0.8 0.401

Posteriorwallthickness(mm) 8.9±0.9 8.7±1.0 0.389

Decelerationtime(ms) 180±35 186±45 0.821

Em/Amratio 0.91±0.26 1.14±0.32 0.004

E/E’ratio 7.2±0.9 9.2±1.0 0.002

LAdiameter(mm) 35.5±2.5 33.5±1.8 0.001

Dataarenumber(%)ormean_±standarddeviation.BMI:bodymassindex;BSA:bodysurfacearea;LA:leftatrium;LVEDD:leftventricle

end-diastolicdiameter;LVESD:leftventricleend-systolicdiameter;TSH:thyroid-stimulatinghormone.

Table2 Measurementsofleftatrialmechanicalfunction. Hyperthyroid (n=34) Control (n=34) P Vmax(mL/m2_{) 30.3±8.4 27.7±7.5 0.212} Vmin(mL/m2_{) 9.6±4.4 8.3±3.2 0.203} Vp(mL/m2_{) 19.0±6.1 14.1±5.0 0.001}

LApassiveemptyingvolume(mL/m2_{) 11.4±3.5 13.4±3.6 0.038}

LApassiveemptyingfraction(%) 37.5±8.5 47.4±8.0 <0.001

LAactiveemptyingvolume(mL/m2_{) 9.4±3.2 5.7±2.4 <0.001}

LAactiveemptyingfraction(%) 49.6±11.5 39.8±5.9 <0.001

Dataaremean±standarddeviation.LA:leftatrium;Vmax:LAmaximumvolume;Vmin:LAminimumvolume;Vp:LAvolumebeforeP

wave.

predictorsof interatrialEMD inpatientswithovert hyper-thyroidism.

The myocardium is well understood in tissues that includethyroidhormonereceptors[4].Thecardiaceffects of hyperthyroidism are caused by increased metabolic demandsandthedirectchronotropicandinotropiceffects of excess thyroid hormones on the myocardium. Thyroid hormoneschange theaction potentialperiodand repolar-izationspeedofatrialandventricularmyocytes[19,20].An increasedheart rate andsupraventricular ectopicactivity areobserved in patients withoverthyperthyroidism [21].

Similarly,wereportthatpatientswithoverthyperthyroidism haveincreasedheartrates.Komiyaetal.[22]showed dif-ferencesintheatrialeffectiverefractoryperiodandatrial conduction delay in patients with hyperthyroidism. Simi-larly, we observed prolonged intra- and interatrial EMDs inpatientswithoverthyperthyroidisminthis study. Addi-tionally,our results suggest that impaired LA mechanical function occurs in overt hyperthyroid patients. Previous studiesreported lengtheningof theatrial electromechan-icalcoupling interval and impaired atrial function in the developmentofsupraventriculararrhythmias[13,14].Asa

Table3 AtrialelectromechanicalintervalfindingsmeasuredbytissueDopplerimaging. Hyperthyroid (n=34) Control (n=34) P PAlateral(ms) 70.4±8.6 55.3±3.9 <0.001 PAseptum(ms) 51.7±5.8 44.4±2.7 <0.001 PAtricuspid(ms) 40.2±4.7 37.2±2.2 0.001 PAlateral—PAtricuspid(ms)a _{29.2±4.4 18.1±2.6 <0.001} PAlateral—PAseptum(ms)b _{18.7±4.3 10.6±2.0 <0.001} PAseptum—PAtricuspid(ms)c _{10.5±2.9 7.0±1.2 <0.001}

Dataaremean_±standarddeviation.PA:theintervalmeasuredbytissueDopplerimagingfromtheonsetofthePwaveonthesurface

electrocardiogramtobeginningofthelatediastolic(Am)wave.

a_{Interatrialelectromechanicaldelay.}

b _{Leftintra-atrialelectromechanicaldelay.}

c _{Rightintra-atrialelectromechanicaldelay.}

Figure2. ApositivecorrelationbetweeninteratrialdelayandfreeT4,andanegativecorrelationbetweeninteratrialdelayand

thyroid-stimulatinghormone(TSH).

consequence,lengthenedintra-andinteratrialEMDsmaybe relatedtoanincreasedriskofarrhythmiasinpatientswith overthyperthyroidism.

Themajorityofpreviousstudiesfocusedonthe relation-ship between ventricular function and thyroid hormones. Alterationsinthyroidstatusmayleadtochangesnotonly in ventricular function but also in atrial function. Gun-tekinetal. [23]observed prolongedPwave durationand dispersioninpatients withhyperthyroidism. Shenoyetal.

[24] demonstrated the effects of thyroid hormones on sarcoplasmic reticulum calcium transporters in rat atria. Nevertheless, LA mechanical function has not been eval-uatedin patientswithoverthyperthyroidism. When atrial mechanicalfunctionwasinvestigatedinthepresentstudy, we found that LA mechanical function was significantly

impairedinpatientswithoverthyperthyroidism.LApassive emptyingvolumeandfractionsignificantlydecreased,while LAactiveemptyingvolumefractionsignificantlyincreased inhyperthyroidpatients.Also,wefoundthattheLVEm/Am and E/E’ ratio were significantly lower in hyperthyroid patientsthanincontrols.Therefore,impairedLA mechani-calfunctionisrelatedtoincreasedwalltensionasaresult of increased LV filling pressure in hyperthyroid patients. LA mechanical functions include passive emptying, active emptyingandreservoirfunctionsatdifferentstagesofthe cardiac cycle. Reservoir functions arise during ventricu-lar systole, passive emptying functions occur during early diastole and active emptying functions take place during ventriculardiastoleinthepresenceofsinusrhythm.When left ventricular diastolic dysfunction develops, the left

Table4 Stepwiselinearregressionanalysesofvariables associatedwithinteratrialelectromechanicaldelay.

␤ t P Age 0.098 1.651 0.104 BMI 0.001 0.010 0.992 BSA 0.019 0.269 0.732 Heartrate 0.053 0.747 0.458 LAsize(mm) 0.076 1.091 0.280 Em/Amratio —0.075 —1.253 0.223 LApassiveemptying volume(mL) —0.016 0.244 0.802 LApassiveemptying fraction(%) —0.155 —2.243 0.032 LAactiveemptying volume(mL) 0.018 0.201 0.841 LAactiveemptying fraction(%) 0.189 2.797 0.007 FreeT3 0.078 0.694 0.490 FreeT4 0.436 4.055 <0.001 TSH —0.310 —2.950 0.005

Am: latediastolic velocity;BMI:bodymass index; BSA:body

surfacearea;Em:earlydiastolicvelocity;LA:leftatrium;TSH:

thyroid-stimulatinghormone.

atrium maypreserve an adequate cardiac output through the regulation of the reservoir and booster pump func-tions [11,25].Conversely, atrial function intenselyaffects heart function. This effect is more prominent in patients withreduced LVfunction [11,25].Therefore, impaired LA functionmayresultinthedevelopmentofheartfailurein patientswithoverthyperthyroidism.

Study

limitations

The major limitation of this study was the size of the studypopulation,whichwasrelativelysmall.Patientscould notbefollowedupforarrhythmicepisodes.Therefore,we do not know whether prolongation of intra- and intera-trial EMD and impaired LA mechanical function occursin patientswithoverthyperthyroidismandwhetherthese fac-torscanbeusedforthepredictionofarrhythmiasandheart failure.Forthesereasons, long-termfollow-up and large-scale prospective studieswith clinicalevents and theuse ofnewdeformationtoolsareneededtochangeourclinical behaviour.

Conclusion

Thecurrentstudyshowedprolongationofintra-and inter-atrial electromechanical intervals and impairment of LA mechanicalfunctioninpatientswithoverthyperthyroidism. Thestudyalsoshowedthatlengthenedintra-andinteratrial delays were related to TSH and fT4 concentrations. Pro-longedatrial electromechanical intervals andimpaired LA mechanicalfunctionmaybeearlysignsofsubclinicalcardiac involvementanddysrhythmiasinoverthyperthyroidism.

Disclosure

of

interest

Theauthorsdeclarethattheyhavenoconflictsofinterest concerningthisarticle.

References

[1]KleinI, OjamaaK. Thyroidhormone and thecardiovascular

system.NEnglJMed2001;344:501—9.

[2]Klein I. Endocrine disorders and cardiovascular

dis-ease. In: Zipes DP, Libby P, Bonow RO, et al., editors.

Braunwald’s heart disease: a textbook of cardiovascular

medicine. 7thed. Philadelphia:Elsevier Saunders; 2005. p.

2056—63.

[3] Forfar JC, Muir AL, Sawers SA, et al. Abnormal left

ventricularfunction inhyperthyroidism:evidence fora

pos-sible reversible cardiomyopathy. N Engl J Med 1982;307:

1165—70.

[4]KleinI,Ojamaa K. Thyrotoxicosisand theheart. Endocrinol

MetabClinNorthAm1998;27:51—62.

[5]SiuCW,PongV,ZhangX,etal.Riskofischemicstrokeafter

new-onsetatrialfibrillationinpatientswithhyperthyroidism.

HeartRhythm2009;6:169—73.

[6]SiuCW,YeungCY,LauCP,etal.Incidence,clinical

character-istics andoutcome ofcongestive heartfailureas theinitial

presentationinpatientswithprimaryhyperthyroidism.Heart

2007;93:483—7.

[7] Yue WS, Chong BH, Zhang XH, et al.

Hyperthyroidism-induced leftventriculardiastolicdysfunction: implicationin

hyperthyroidism-related heart failure. ClinEndocrinol (Oxf)

2011;74:636—43.

[8]DazaiY.Leftatrialsystolictimeintervalinhyperthyroidism.

Angiology1999;50:591—8.

[9]Abecasis J, Dourado R, Ferreira A, et al. Left atrial

volume calculated by multi-detector computed

tomogra-phy may predict successful pulmonary vein isolation in

catheter ablation of atrial fibrillation. Europace 2009;11:

1289—94.

[10]HofI,ChilukuriK,Arbab-ZadehA,etal.Doesleftatrial

vol-umeandpulmonaryvenousanatomypredicttheoutcomeof

catheterablation ofatrialfibrillation? JCardiovasc

Electro-physiol2009;20:1005—10.

[11]PrioliA,MarinoP,LanzoniL,etal.Increasingdegreesofleft

ventricularfillingimpairmentmodulateleftatrialfunctionin

humans.AmJCardiol1998;82:756—61.

[12]Acar G, Akcay A, Sokmen A, et al. Assessment of atrial

electromechanical delay, diastolic functions, and left atrial

mechanicalfunctionsinpatientswithtype1diabetesmellitus.

JAmSocEchocardiogr2009;22:732—8.

[13]Cui QQ, Zhang W, Wang H, et al. Assessment of atrial

electromechanical coupling and influential factors in

non-rheumaticparoxysmalatrialfibrillation.ClinCardiol2008;31:

74—8.

[14]DogduO,YarliogluesM,KayaMG,etal.Assessmentofatrial

conductiontimeinpatientswithsystemiclupus

erythemato-sus.JInvestigMed2011;59:281—6.

[15]LangRM,BierigM,DevereuxRB,etal.Recommendationsfor

chamberquantification:areportfromtheAmericanSocietyof

Echocardiography’sGuidelinesand StandardsCommitteeand

theChamberQuantificationWritingGroup,developedin

con-junctionwiththeEuropeanAssociationofEchocardiography,

a branchof the European Societyof Cardiology. J Am Soc

Echocardiogr2005;18:1440—63.

[16] Devereux RB, Reichek N. Echocardiographic determination

ofleft ventricularmass inman. Anatomicvalidation ofthe

[17]QuinonesMA,OttoCM,StoddardM,etal.Recommendations

forquantificationofDopplerechocardiography:areportfrom

theDopplerQuantificationTaskForceoftheNomenclatureand

StandardsCommitteeoftheAmericanSocietyof

Echocardiog-raphy.JAmSocEchocardiogr2002;15:167—84.

[18]Tiryakioglu SK, Tiryakioglu O, Ari H, et al. Left

ventricu-larlongitudinalmyocardialfunctioninoverthypothyroidism:

a tissue Doppler echocardiographic study. Echocardiography

2010;27:505—11.

[19]JohnsonPN,FreedbergAS,MarshallJM.Actionofthyroid

hor-moneonthetransmembranepotentialsfromsinoatrialnode

cellsandatrialmusclecellsinisolatedatriaofrabbits.

Cardi-ology1973;58:273—89.

[20] Sun ZQ, Ojamaa K, Coetzee WA, et al. Effects of

thy-roid hormone on action potential and repolarizing currents

in ratventricular myocytes.Am JPhysiolEndocrinol Metab

2000;278:E302—7.

[21]WustmannK,Kucera JP, ZanchiA, etal.Activation of

elec-tricaltriggersofatrialfibrillationinhyperthyroidism.JClin

EndocrinolMetab2008;93:2104—8.

[22]Komiya N, Isomoto S, Nakao K, et al. Electrophysiological

abnormalitiesofthe atrialmusclein patientswith

paroxys-mal atrial fibrillation associated with hyperthyroidism. Clin

Endocrinol(Oxf)2002;56:39—44.

[23]GuntekinU, GunesY, Simsek H,et al.Pwavedurationand

dispersioninpatientswithhyperthyroidismandtheshort-term

effectsofantithyroidtreatment.IndianPacingElectrophysiol

J2009;9:251—9.

[24]Shenoy R, Klein I, Ojamaa K. Differential regulation of SR

calciumtransportersbythyroidhormoneinratatriaand

ven-tricles.AmJPhysiolHeartCircPhysiol2001;281:H1690—6.

[25]MatsudaY,TomaY,OgawaH,etal.Importanceofleftatrial

function in patientswith myocardial infarction. Circulation