evaluation of atrial electromechanical delay and diastolic functions in patients with...

Evaluation of Atrial Electromechanical Delay andDiastolic Functions in Patients with Hyperthyroidism

Abdullah Sokmen, M.D.,* Gurkan Acar, M.D.,* Gulizar Sokmen, M.D.,* Ahmet Akcay, M.D.,*Murat Akkoyun, M.D.,* Sedat Koroglu, M.D.,† Alper Bugra Nacar, M.D.,‡ and Mesut Ozkaya, M.D.§

*Department of Cardiology, Faculty of Medicine, Kahramanmaras Sutcu Imam University, Kahramanmaras,Turkey; †Cardiology Clinic, Afsin State Hospital, Kahramanmaras, Turkey; ‡Cardiology Clinic, Elbistan StateHospital, Kahramanmaras, Turkey; and §Faculty of Medicine, Department of Endocrinology, GaziantepUniversity, Gaziantep, Turkey

Aim: Hyperthyroidism is a well-known cause of atrial fibrillation (AF) which is associated with increasedmorbidity and mortality. Atrial electromechanical delay (EMD) is a significant predictor of AF. The aimof this study was to assess the atrial EMD and diastolic functions in subclinical and overt hyperthyroid-ism by using tissue Doppler imaging (TDI). Methods and Results: The study population consisted of 3groups: group I (30 healthy subjects), group II (38 patients with subclinical hyperthyroidism), andgroup III (25 patients with overt hyperthyroidism). Atrial electromechanical coupling was measuredwith TDI. Standard echocardiographic measurements and parameters of diastolic function wereobtained by conventional echocardiography and TDI.Intra- and inter-atrial EMD were significantly prolonged in subclinical and overt hyperthyroidism com-

pared with control group (P = 0.03 and P < 0.001 for intra-atrial EMD; P < 0.001 for inter-atrial EMD).In groups II and III, mitral A velocity (P = 0.005 and P = 0.001) and mitral E-wave deceleration time(P < 0.001 and P = 0.02) were significantly increased, and mitral E/A ratio (P = 0.005 and P = 0.001)was significantly decreased compared with the control group. The lateral mitral Em/Am ratio in group IIand group III was significantly lower than controls (P = 0.001). Mitral Em/Am ratio (b = �0.32,P = 0.002) and thyroid stimulating hormone (TSH) level (b = �0.27, P = 0.009) were negatively andindependently correlated with inter-atrial EMD. Conclusion: This study showed that intra- and inter-atrial electromechanical intervals were prolonged and diastolic function was impaired in both overt andsubclinical hyperthyroidism. TSH level and mitral Em/Am ratio were found as independent predictors ofatrial EMD. (Echocardiography 2013;30:1194-1201)

Key words: atrial electromechanical delay, hyperthyroidism, tissue Doppler imaging

Hyperthyroidism is a common disorder withprominent cardiovascular effects including sinustachycardia, systolic hypertension, changes inventricular systolic and diastolic function, andpredisposition to dysrhythmias, especially atrialfibrillation (AF).1 Thyroid hormones have majoreffects on cardiac pacemaker function and onmyocardial contractility through effects on myo-cardial structural and regulatory gene transcrip-tion. The resulting chronotropic, inotropic, andlusitropic effects of excess thyroid hormone onthe heart in overt hyperthyroidism are present toa lesser extent in subclinical hyperthyroidism.2

In the adult population, AF is the most com-mon type of tachyarrhythmia encountered in

clinical practice and is associated with increasedmorbidity and mortality.3 It is the most prevalentdysrhythmia after sinus tachycardia in patientswith hyperthyroidism. Between 10% and 15% ofhyperthyroid patients, clinical or subclinical,develop AF.4,5

The prolongation of intra- and inter-atrialconduction time and the inhomogeneous prop-agation of sinus impulses are well-known elec-trophysiologic characteristics of the atria proneto fibrillation. Electromechanical delay (EMD)has been defined as the temporal delaybetween the detected onset of electrical activityand the realization of force in the myocardium.This issue has been evaluated noninvasively bysimple electrocardiographic markers such asmaximum P-wave duration and P-wave disper-sion as well as by tissue Doppler imaging (TDI).It is well known that increased atrial EMD onTDI is a predictor of the development of AF.6–11

Address for correspondence and reprint requests: AbdullahSokmen, M.D., Faculty of Medicine, Department of Cardiol-ogy, Kahramanmaras Sutcu Imam University, 46050 Kahr-amanmaras, Turkey. Fax: +90-344-2217239;E-mail: [email protected]

1194

© 2013, Wiley Periodicals, Inc.DOI: 10.1111/echo.12277 Echocardiography

Several studies have been conducted about theeffects of hypothyroidism and subclinicalhypothyroidism on atrial electromechanicalfunctions, but there is a paucity of data aboutatrial electromechanical coupling in patientswith hyperthyroidism.12,13 The aim of this studywas to evaluate atrial EMD and diastolic func-tions by TDI, and also to investigate clinical andechocardiographic characteristics affecting atrialEMD in patients with overt and subclinicalhyperthyroidism.

Materials and Methods:Study Population:Thirty healthy subjects (group I: 4 male, 26female; mean age 38.2 � 9.7 years), 38 patientswith subclinical hyperthyroidism (group II: 5 male,33 female; mean age 38.4 � 10.8 years), and 25patients with overt hyperthyroidism (group III: 3male, 22 female; mean age 40.6 � 11.3 years)were included in the study. Patients with coronaryartery disease, systemic hypertension, valvularheart disease, left ventricular (LV) systolic dysfunc-tion, conduction abnormalities, rhythm distur-bances, diabetes mellitus (DM), electrolytedisorders, renal failure, pulmonary disease, orother associated systemic diseases, patients receiv-ing drugs known to influence the electrophysiol-ogy of the heart, and patients with poorechocardiographic view were excluded from thestudy. The institutional ethics committeeapproved the study protocol. Written informedconsent was obtained from each subject.

Assessment of Thyroid Status:The evaluation of thyroid morphology and func-tion was performed by clinical examination,assessment of serum free thyroid hormones (FT3and FT4) and thyroid stimulating hormone (TSH)levels, and ultrasound examination of the thyroidgland. A total of 5 mL of venous blood sampleswere drawn and then centrifuged at 2000 rpmfor 10 minutes in a refrigerated centrifuge toseparate serum. Serum FT3, FT4, and TSH levelswere measured using the commercial Immulitekits, which are solid-phase, two-site chemilumi-nescent immunumetric assay in Immulite auto-analyser (IMMULITE 2000 XPI, SiemensHealthcare, Erlangen, Germany). The normal ref-erence intervals for FT3 was 1.8–5.2 ng/dL,for FT4 was 0.80–1.9 ng/dL, and for TSH was0.40–4.0 mIU/mL.

Subclinical hyperthyroidism was diagnosed bynormal serum FT3 and FT4 concentrations anddecreased or undetectable serum TSH concentra-tions. Overt hyperthyroidism was diagnosedclinically and confirmed by high FT3 and/orFT4 concentrations with suppressed serum TSHconcentrations.2

Conventional Echocardiographic Examination:All subjects were performed two-dimensional,M-mode, pulsed, and color flow Dopplerechocardiographic examinations (Vivid 7 Pro,GE, Horten, Norway, 2–4 MHz phased-arraytransducer) by a cardiologist who was blind tothe clinical details of each subject. During echo-cardiography, 1 lead electrocardiogram wasrecorded continuously. M-mode measurementsand conventional Doppler echocardiographicexaminations were performed according to thecriteria of the American Society of Echocardiogra-phy guidelines.14 LV diastolic function was evalu-ated by mitral inflow velocities, namely peak E(early diastolic) and peak A (late diastolic), E/Aratio and also by deceleration time of the E-wave(DT) and isovolumic relaxation time (IVRT). Allmeasurements were recorded as an average of 3cardiac cycles. Left atrium (LA) dimension, LVend-systolic and end-diastolic dimensions, dia-stolic ventricular septal thickness, and diastolicthickness of LV posterior wall were measured inparasternal long-axis view. LV ejection fraction(EF) was estimated by Simpson’s rule. LV masswas calculated with the Devereux Formula andindexed to body surface.15 LA volume wasmeasured by the method of discs in the apicalfour-chamber view and indexed to the bodysurface area (BSA).

Atrial Electromechanical Coupling and TissueDoppler Imaging:Tissue Doppler imaging was performed withtransducer frequencies of 3.5–4.0 MHz, adjust-ing the spectral pulsed Doppler signal filters untila Nyquist limit of 15–20 cm/sec, and using theminimal optimal gain. The monitor sweep speedwas set at 50–100 mm/sec to optimize the spec-tral display of myocardial velocities. In the apicalfour-chamber view, the pulsed Doppler samplevolume was subsequently placed at the level ofthe LV lateral mitral annulus, septal mitral annu-lus, and right ventricular (RV) tricuspid annulus.The pulsed wave cursor was aligned as parallel aspossible to the myocardial segment of interest toensure the optimal angle of imaging. Time inter-vals from the onset of P-wave on surface ECG tothe beginning of A-wave (PA) representing atrialelectromechanical duration were obtained fromlateral mitral annulus, septal mitral annulus, andRV tricuspid annulus and named as PA lateral, PAseptal, and PA tricuspid, respectively. Values wereaveraged over 3 consecutive beats. The differ-ence between PA lateral and PA tricuspid wasdefined as inter-atrial EMD, and the differencebetween PA septum and PA tricuspid was definedas intra-atrial EMD.9

Myocardial velocities named as Sm (peaksystolic), Em (early diastolic), and Am (late

1195

Atrial Electromechanical Delay in Thyroid Disease

diastolic) were obtained at the lateral mitral andlateral tricuspid annulus. LV Em/Am, LV E/Em, andRV Em/Am ratios were calculated. The cases withLV Em/Am < 1 were accepted to have LV diastolicdysfunction.16

Reproducibility:Intra-observer variability was assessed in 20 sub-jects selected randomly from the patient groupby repeating the measurements under the samebasal conditions. To test the inter-observer vari-ability, the measurements were performed offline from video recordings by a second observer.Reproducibility of echocardiographic measure-ments was assessed by coefficient of variationbetween measurements. Intra-observer and inter-observer coefficients of variation for echocardio-graphic measurements were found to be <5%and nonsignificant.

Statistical Analysis:All analyses were conducted using SPSS 15.0(SPSS for Windows 15.0, Chicago, IL, USA).Continuous variables were expressed asmean � standard deviation; categoric variableswere defined as percentages. Categorical datawere compared with the chi-square test. All thenumerical variables of the study groups pre-sented a normal distribution. Thus, mean valuesof continuous variables were compared byusing analysis of variance (ANOVA) and posthoc Scheffe test was used for the comparisonof groups. Pearson’s correlation coefficientswere used to assess the strength of the relation-ship between continuous variables. A stepwise,multiple regression analysis was used to identifysignificant determinants of inter-atrial EMD.Thus, all predetermined independent variablesthat correlated with a P-value of <1 in the Pear-son’s correlations were inserted into a stepwise,multiple regression analysis. P < 0.05 was con-sidered significant.

Results:Clinical, laboratory, and echocardiographic dataof 3 groups are shown in Table I. Age, gender,body mass index, BSA, smoking status, systolicand diastolic blood pressure, LV end-diastolicdimension, LV end-systolic dimension, and LV EFwere similar between 3 groups (P > 0.05). LVmass index was slightly increased in subclinicaland overt hyperthyroidism, but the differencebetween the groups was not statistically signifi-cant. Heart rate was found to be higher in groupIII compared with group I (83.9 � 10.7 bpm vs.78.0 � 8.2 bpm, P = 0.04). LA dimension wassignificantly higher in group III than the controls(36.8 � 5.4 mm vs. 32.1 � 4.6 mm, P = 0.002),but there was no significant difference between

group II and the other groups. TSH levels weresignificantly lower in group II (0.19 � 0.11 lIU/mL)and group III (0.06 � 0.11 lIU/mL) than group I(1.56 � 0.99 lIU/mL, and P < 0.001 for both).Although TSH level of group III was lower thanthat of group II, the difference did not reach sta-tistical significance. The FT3 and FT4 levels weresignificantly higher in group III than group I(5.41 � 2.73 vs. 3.18 � 0.57 for FT3, and2.00 � 1.23 ng/dL vs. 1.21 � 0.22 ng/dL forFT4, respectively, P < 0.001) and group II (FT3,3.84 � 0.60, P < 0.001; FT4, 1.37 � 0.22 ng/dL,P = 0.002, respectively), and FT4 level in group IIwas significantly higher than group I (1.37 �0.22 vs. 1.21 � 0.22, P < 0.002).

Comparison of conventional and tissue Dopp-ler variables between groups is demonstrated inTable II. Mitral E velocity and IVRT were similarbetween the 3 groups (P > 0.05). Mitral A veloc-ity in group III and group II was significantlyhigher than controls (77.5 � 16.9 cm/sec and73.3 � 17.0 cm/sec vs. 60.8 � 10.6 cm/sec;P = 0.001 and P = 0.005, respectively). MitralE/A ratio in group III and group II wassignificantly lower than controls (1.18 � 0.27and 1.25 � 0.30 vs. 1.47 � 0.26; P = 0.001 andP = 0.005, respectively). The DT in group IIIand group II was significantly higher than thecontrol group (204.8 � 24.5 msec and 210.8 �29.2 msec vs. 186.9 � 9.9 msec; P = 0.02and P < 0.001, respectively). In short, mitral Avelocity, E/A ratio, and DT in groups II and IIIwere significantly different from the controlgroup.

The LV lateral Am in group III and group IIwas significantly higher than the control group(12.6 � 2.6 cm/sec and 12.2 � 2.4 cm/sec vs.9.8 � 2.4 cm/sec; P < 0.001 for both). Thelateral Em/Am ratio in group III and group II wassignificantly lower than controls (1.22 � 0.47and 1.28 � 0.43 vs. 1.76 � 0.58, P = 0.001).Number of subjects with LV diastolic dysfunctionwas 1 (3.3%), 10 (26%), and 11 (44%) in groupI, group II, and group III, respectively. Consider-ing diastolic dysfunction, there was significantdifference between patient groups (groups II andIII) and the control group (P < 0.002), but thedifference between groups II and III was not sta-tistically significant. There was also no significantdifference between groups II and III for LV lateralSm, Em, Am, and Em/Am. The RV Am in group IIIwas significantly higher than controls (17.0 � 4.7cm/sec vs. 13.2 � 3.4 cm/sec, P = 0.001). Othertissue Doppler variables were similar between the3 groups (P > 0.05).

The atrial electromechanical coupling param-eters of different sites measured by TDI areshown in Table III. PA lateral in group III(65.3 � 7.3 msec, P < 0.001 vs. controls) and

1196

Sokmen, et al.

group II (62.5 � 6.4 msec, P < 0.001 vs.controls) was significantly higher than controls(52.4 � 6.6 msec). PA septum in group III(46.1 � 4.6 msec, P = 0.002 vs. controls) andin group II (45.3 � 5.5 msec, P = 0.004 vs. con-trols) was significantly higher than controls(40.9 � 5.9 msec). PA tricuspid was not differ-ent between groups. The inter-atrial EMD ingroup III (25.5 � 7.0 msec, P < 0.001 vs. con-trols) and group II (22.0 � 6.3 msec, P < 0.001vs. controls) was significantly higher than con-trols (14.1 � 4.2 msec). The intra-atrial EMD ingroup III (6.3 � 3.2 msec, P < 0.001 vs. con-trols) and group II (4.8 � 3.7 msec, P = 0.03vs. controls) was significantly higher than con-trols (2.6 � 2.0 msec). There was no significantdifference between group II and group III con-sidering atrial EMD.

Bivariate and multivariate relationships of theinter-atrial EMD with clinical and echocardio-graphic variables are shown in Table IV. Inter-atrial EMD was positively correlated with age(r = 0.28, P = 0.007), systolic blood pressure(SBP) (r = 0.30, P = 0.004), diastolic blood pres-sure (r = 0.31, P = 0.003), LV mass index(r = 0.37, P = 0.000), LA volume index (r = 0.21,

P = 0.04), FT3 (r = 0.22, P = 0.04), and nega-tively correlated with tricuspid Em/Am (r = �0.23,P = 0.03), mitral Em/Am ratio (r = �0.43,P = 0.000), and TSH level (r = �0.38,P = 0.000). In stepwise linear regression analyses,only mitral Em/Am ratio (b = �0.32, P = 0.002)and TSH level (b = �0.27, P = 0.009) wereweakly but significantly related with inter-atrialEMD. Correlation of TSH with clinical and echo-cardiographic parameters revealed that TSH wasnegatively correlated with age (r = �0.22,P = 0.03), SBP (r = �0.28, P = 0.007), DT(r = �0.33, P = 0.001), intra-atrial EMD(r = �0.28, P = 0.007), and inter-atrial EMD(r = �0.38, P = 0.000), and positively correlatedwith mitral E/A (r = 0.35, P = 0.000) and mitralEm/Am (r = 0.34, P = 0.000). In stepwise linearregression analyses, mitral Em/Am (b = 0.22,P = 0.04), mitral E/A (b = 0.25, P = 0.02), DT(b = �0.24, P = 0.02), and inter-atrial EMD(r = �0.38, P = 0.000) were weakly but signifi-cantly correlated with TSH (Table V).

Discussion:The major findings of this study were as follows:first, patients with subclinical and overt

TABLE I

Clinical, Laboratory, and Echocardiographic Data of the Subjects

Controls (group I, n = 30)

Patients with SubclinicalHyperthyroidism(group II, n = 38)

Patients with OvertHyperthyroidism(group III, n = 25)

Age (years) 38.2 � 9.7 38.4 � 10.8 40.6 � 11.3Male/female (n/n) 4/26 5/33 3/22BMI (kg/m2) 27.9 � 5.0 29.0 � 5.0 28.4 � 4.5BSA (m2) 1.74 � 0.15 1.73 � 0.15 1.74 � 0.14Smoking (n) 4 5 4SBP (mmHg) 112.7 � 8.9 113.8 � 13.0 115.8 � 12.2DBP (mmHg) 70.8 � 6.2 73.0 � 6.9 74.8 � 7.3Heart rate (beats/min) 78.0 � 8.2 81.0 � 9.6 83.9 � 10.7*

LV end-diastolic dimension (mm) 46.4 � 4.9 46.6 � 3.6 47.6 � 3.8LV end-systolic dimension (mm) 29.0 � 3.7 29.0 � 3.3 29.1 � 3.3Septum thickness (mm) 9.0 � 1.2 9.3 � 1.2 9.6 � 1.0Posterior wall thickness (mm) 8.0 � 1.1 8.4 � 1.3 8.7 � 1.1LV ejection fraction (%) 67.3 � 5.3 68.0 � 5.5 68.8 � 6.2LV mass index (g/m2) 85.5 � 20.8 90.1 � 20.3 97.5 � 21.5LA dimension (mm) 32.1 � 4.6 34.8 � 4.4 36.8 � 5.4†

LA volume index (mL/m2) 26.3 � 6.9 29.2 � 5.6 30.7 � 10.7FT3 (ng/dL) 3.18 � 0.57 3.84 � 0.60¶ 5.41 � 2.73§

FT4 (ng/dL) 1.21 � 0.22 1.37 � 0.22‡ 2.00 � 1.23§

TSH (lIU/mL) 1.56 � 0.99 0.19 � 0.11§ 0.06 � 0.11§

Values are mean � standard deviation or number.*P = 0.04 versus controls.†P = 0.002 versus controls.‡P = 0.002 versus patients with overt hyperthyroidism.§P < 0.001 versus controls.¶P < 0.001 versus patients with overt hyperthyroidism.BMI = body mass index; BSA = body surface area; SBP = systolic blood pressure; DBP = diastolic blood pressure; LV = leftventricular; TSH = thyroid stimulating hormone; LA = left atrium.

1197


hyperthyroidism had prolonged intra- and inter-atrial EMDs (EMD) measured with TDI; second,diastolic function parameters were impaired inthese patients, and finally, the mitral Em/Am ratioand serum TSH level were independent factorsrelated with inter-atrial EMD.

Thyroid hormone affects the structure andfunction of the heart. It alters relaxation-contrac-tion rates changing gene expression at thenuclear level through changes in the myosinheavy chain a and b ratio, intracellular calciumlevel and calcium transport rate in the cell mem-brane. Therefore, mild changes in serum levels ofthyroid hormones have many adverse effects oncontractility and function of the myocardium.1,17

Overt hyperthyroidism induces a hyperdynamiccardiovascular state with high cardiac output,low systemic vascular resistance, enhanced LVsystolic function, LV hypertophy, impaired ven-tricular relaxation, and increased atrial ectopicactivity.18 The cardiovascular consequences ofsubclinical hyperthyroidism, defined by sup-pressed serum TSH levels despite normal free T4and T3 concentrations, are less well established.However, the most consistent findings in subclin-ical hyperthyroidism include an increased heartrate, supraventricular arrhythmias, increased LV

mass with a slightly enhanced systolic function,and impaired diastolic function due to slowedmyocardial relaxation.18–22

In this study, we found significant impairmentin LV diastolic function parameters comparedwith normal subjects as documented by thereduced ratio between early and late diastolicpeak myocardial velocities and the prolonged DTin both subclinical and overt hyperthyroidism.Our findings about overt hyperthyroidism werein line with the earlier studies, but there are con-flicting data on whether diastolic functionparameters are impaired in patients with subclini-cal hyperthyroidism. Many investigators reportimpairment of diastolic function, and others haveseen no significant change in subclinical hyper-thyroidism.20,23,24 These conflicting results mightbe explained by the differences in demographicfeatures of the study population, degree of TSHsuppression, and duration and the cause of thedisease. It has been suggested that diastolic dys-function in subclinical hyperthyroidism resultsfrom increased LV mass, but exact mechanism isunclear. Although, in our study, LV mass wasslightly increased in patients with subclinical andovert hyperthyroidism, the difference was notstatistically significant. It is more likely that

TABLE II

Comparison of Conventional and Tissue Doppler Variables between Patients and the Controls

Controls(group I, n = 30)

Patients withSubclinical Hyperthyroidism

(group II, n = 38)


Conventional Doppler parametersMitral E velocity (cm/sec) 87.9 � 13.0 87.4 � 13.5 88.5 � 16.9Mitral A velocity (cm/sec) 60.8 � 10.6 73.3 � 17.0* 77.5 � 16.9†

E/A 1.47 � 0.26 1.25 � 0.30* 1.18 � 0.27†

DT (msec) 186.9 � 9.9 210.8 � 29.2¶ 204.8 � 24.5§

IVRT (msec) 86.4 � 8.8 88.2 � 19.6 84.8 � 11.2Tissue Doppler parametersLV lateral annulusSm (cm/sec) 10.9 � 1.5 11.4 � 2.7 12.2 � 2.4Am (cm/sec) 9.8 � 2.4 12.2 � 2.4¶ 12.6 � 2.6¶

Em (cm/sec) 16.1 � 3.0 14.9 � 3.7 14.4 � 4.0Em/Am 1.76 � 0.58 1.28 � 0.43† 1.22 � 0.47†

E/Em 5.6 � 1.2 6.2 � 2.1 6.5 � 1.9RV lateral annulusSm (cm/sec) 14.3 � 2.7 15.0 � 2.6 15.4 � 3.9Am (cm/sec) 13.2 � 3.4 15.3 � 3.0 17.0 � 4.7†

Em (cm/sec) 13.6 � 3.1 13.9 � 2.9 14.6 � 4.9Em/Am 1.09 � 0.33 0.94 � 0.27 0.89 � 0.30

Values are mean � standard deviation.*P = 0.005 versus controls.†P = 0.001 versus controls.§P = 0.02 versus controls.¶P < 0.001 versus controls.DT = mitral E-wave deceleration time; IVRT = isovolumic relaxation time; LV = left ventricular; Sm = systolic myocardial velocity;Em = early myocardial diastolic velocity; Am = late myocardial diastolic velocity; RV = right ventricular.

1198

Sokmen, et al.

biochemical effects of thyroid hormones on myo-cardial function are involved, unrelated to LVmass index.19,22,25,26

Atrial fibrillation is the most common arrhyth-mia encountered in clinical practice, and thesecond common rhythm disorder in patientswith hyperthyroidism. It is associated with signifi-cant morbidity and mortality due to hemody-namic impairment and thromboembolic events.In many studies, it has been demonstrated thatatrial EMD was independently related to new

onset AF and recurrence of AF. EMD can be mea-sured by invasive or noninvasive methods. Simpleelectrocardiographic markers, maximum P-waveduration and P-wave dispersion, and someechocardiographic variables have been used fornoninvasive evaluation of intra- and inter-atrialconduction times and inhomogeneous conduc-tion of sinus impulses.6–11 Initially, M-modeechocardiographic evaluation of intra-atrialconduction time was demonstrated.27 With theprogress in echocardiographic techniques, thetime interval from P-wave onset to DopplerA-wave was studied with Doppler echocardiogra-phy.10 Recently, atrial EMD measured by TDI has

TABLE III

Comparison of the Atrial Electromechanical Coupling Parameters Measured by Tissue Doppler Imaging

Controls(group I, n = 30)

Patients withSubclinical

Hyperthyroidism(group II, n = 38)


PA lateral (msec) 52.4 � 6.6 62.5 � 6.4* 65.3 � 7.3*

PA septum (msec) 40.9 � 5.9 45.3 � 5.5† 46.1 � 4.6‡

PA tricuspid (msec) 38.3 � 5.3 40.5 � 5.4 39.8 � 4.2Inter-atrial EMD (msec) 14.1 � 4.2 22.0 � 6.3* 25.5 � 7.0*

Intra-atrial EMD (msec) 2.6 � 2.0 4.8 � 3.7§ 6.3 � 3.2*

Values are mean � standard deviation.*P < 0.001 versus controls.†P = 0.004 versus controls.‡P = 0.002 versus controls.§P = 0.03 versus controls.PA = time interval from the onset of P-wave on surface ECG to the beginning of Am wave interval with tissue Doppler echocardiog-raphy; EMD = electromechanical delay.

TABLE IV

Bivariate and Multivariate Relationship of Inter-Atrial EMDwith Clinical and Echocardiographic Variables

Inter-Atrial EMD

PearsonCorrelationCoefficient, r P-Value

StandardizedRegression

Coefficient, b P-Value

Age 0.28 0.007 0.07 0.50SBP 0.30 0.004 0.13 0.20DBP 0.31 0.003 0.14 0.18LV mass

index0.37 0.000 0.19 0.08

LA volumeindex

0.21 0.04 0.02 0.83

FT3 0.22 0.04 0.06 0.54Tricuspid

Em/Am

�0.23 0.03 �0.03 0.83

MitralEm/Am

�0.43 0.000 �0.32 0.002

TSH �0.38 0.000 �0.27 0.009

Bold indicates statistical significance. SBP = systolic bloodpressure; DBP = diastolic blood pressure; LV = left ventricular;LA = left atrium; TSH = thyroid stimulating hormone.

TABLE V

Bivariate and Multivariate Relationship of TSH with Clinicaland Echocardiographic Variables

TSH

PearsonCorrelationCoefficient, r P-Value

StandardizedRegression

Coefficient, b P-Value

Age �0.22 0.03 �0.09 0.39SBP �0.28 0.007 �0.16 0.14Intra-atrial

EMD�0.28 0.007 �0.08 0.5

Mitral E/A 0.35 0.000 0.25 0.02Mitral

Em/Am

0.34 0.001 0.22 0.04

DT �0.33 0.001 �0.24 0.02Inter-atrial

EMD�0.38 0.000 �0.38 0.000

Bold indicates statistical significance. SBP = systolic bloodpressure; TSH = thyroid stimulating hormone; DT = decelera-tion time; EMD = electromechanical delay.

1199


become a promising noninvasive technique.Earlier studies found that P-wave duration wascorrelated well with atrial EMD assessed byTDI.28 Recently, Deniz et al.29 compared theatrial conduction times measured by electrophys-iolgical methods and tissue Doppler echocardi-ography, and reported that TDI could be used toevaluate atrial conduction times. Earlier studiesrevealed that atrial EMD obtained by this methodwas prolonged in various diseases such as type 1DM, Behcet disease, familial Mediterranean fever,ankylosing spondylitis, scleroderma, and hypo-thyroidism.12,30–35

There are scarce data about the effects ofhyperthyroidism on atrial electromechanicalduration in the literature. Nacar et al.13 reportedthat subclinical hyperthyroidism was associatedwith prolonged atrial conduction time that mightbe improved by achievement of euthyroid state.In a recent study, Ayhan et al.36 also reportedprolonged atrial EMD in overt hyperthyroidismcompared with control subjects. To the best ofour knowledge, this study is the first one evaluat-ing atrial EMD in patients with subclinical andovert hyperthyroidism as compared to eachother and normal subjects. The study revealedthat intra- and inter-atrial EMD was prolonged insubclinical and overt hyperthyroidism. Moreover,the prolongation was more prominent in overthyperthyroidism. Higher serum levels of thyroidhormones in overt hyperthyroidism may be anexplanation for this, as thyroid hormone affectsthe action potential duration and shortens atrialrefractory period that may facilitate atrialconduction delay.18,37

Detection of prolonged intra- or inter-atrialEMD in both subclinical and overt hyperthyroid-ism may be an earlier sign of atrial dysfunctionpreceding AF. We also found that TSH level andmitral Em/Am ratio were independently correlatedwith inter-atrial EMD, that might implicate thatmore severe the disease, greater impairment indiastolic function, and more likelihood of AF. Thefindings of the study also add to understandingthat subclinical hyperthyroidism is not just a labo-ratory condition, but a disorder that can causesubclinical cardiac dysfunction similar to overthyperthyroidism. As the use of antithyroidtreatment in subclinical hyperthyroidism remainscontroversial, these noninvasive and easilyobtainable echocardiographic parameters mayhelp in clinical judgement to determine therequirement for treatment in such patients.

Study Limitations:The main limitation of this study is that this is across-sectional study and we could not follow-upthe patients prospectively for the development ofarrhythmic episodes. Also, the size of the study

population was relatively small. Therefore, furtherprospective studies need to be conducted with alarger number of patients and a longer follow-uptime to increase the accuracy of the results andto determine the predictive value of thesemeasurements for future development of AF.

Conclusion:This study revealed that intra- and inter-atrial elec-tromechanical intervals were prolonged anddiastolic function was impaired in subclinical andovert hyperthyroidism. Furthermore, TSH levelandmitral Em/Am ratio were found as independentpredictors of atrial EMD. These findings may bemarkers of subclinical cardiac involvement andtendency towards AF. Although, more prominentin overt hyperthyroidism, presence of these altera-tions in subclinical hyperthyroidism may help inthe decision of this controversial condition.

References1. Klein I, Ojamaa K: Thyroid hormone and the cardiovascu-

lar system. N Engl J Med 2001;344:501–509.2. Cooper DS, Biondi B: Subclinical thyroid disease. Lancet

2012;379:1142–1154.3. Camm JA, Kirchhof P, Lip GY, et al: Guidelines for the

management of atrial fibrillation. Eur HeartJ 2010;31:2369–2429.

4. Sawin CT, Geller A, Wolf P, et al: Low serum thyrotropinconcentrations as a risk factor for atrial fibrillation in olderpersons. N Engl J Med 1994;331:1249–1252.

5. Auer J, Scheibner P, Mische T, et al: Subclinical hyperthy-roidism as a risk factor for atrial fibrillation. Am HeartJ 2001;142:838–842.

6. Dilaveris P, Gialafos EJ, Sideris S, et al: Simple electrocar-diographic markers for the prediction of paroxysmal idio-pathic atrial fibrillation. Am Heart J 1998;135:733–738.

7. Aras D, Maden O, Ozdemir O, et al: Simple electrocar-diographic markers for the detection of paroxysmal atrialfibrillation in hyperthyroidism. Int J Cardiol 2005;99:59–64.

8. Omi W, Nagai H, Takamura M, et al: Doppler tissue anal-ysis of atrial electromechanical coupling in paroxysmalatrial fibrillation. J Am Soc Echocardiogr 2005;18:39–44.

9. De Vos CB, Weijs B, Crjins HJGM, et al: Atrial tissue Dopp-ler imaging for prediction of new onset atrial fibrillation.Heart 2009;95:835–840.

10. Ozer N, Yavuz B, Can I, et al: Doppler tissue evaluation ofintra-atrial and interatrial electromechanical delay andcomparison with P-wave dispersion in patients withmitral stenosis. J Am Soc Echocardiogr 2005;18:945–948.

11. Rein AJ, O’Donnell CP, Colan SD, et al: Tissue velocityDoppler assessment of atrial and ventricular electrome-chanical coupling and atrioventricular time intervals innormal subjects. Am J Cardiol 2003;92:1347–1350.

12. Ozturk S, Dikbas O, Ozyasar M, et al: Evaluation of leftatrial mechanical functions and atrial conduction abnor-malities in patients with clinical hypothyroid. CardiolJ 2012;19:374–380.

13. Nacar AB, Acar G, Yorgun H, et al: The effect of antithy-roid treatment on atrial conduction times in patients withsubclinical hyperthyroidism. Echocardiography 2012;29:950–955.

14. Quinones MA, Otto CM, Stoddard M, et al: Recommen-dations for quantification of Doppler echocardiography:A report from the Doppler Quantification Task Force ofthe Nomenclature and Standards Committee of the

1200

Sokmen, et al.

American Society of Echocardiography. J Am Soc Echocar-diogr 2002;15:167–184.

15. Devereux RB, Reichek N: Echocardiographic determina-tion of left ventricular mass in men. Circulation1977;55:613–618.

16. Ommen SR, Nishimura RA, Appleton CP, et al: Clinicalutility of Doppler echocardiography and tissue Dopplerimaging in the estimation of left ventricular filling pres-sures: A comparative simultaneous Doppler-catheteriza-tion study. Circulation 2000;102:1788–1794.

17. Biondi B, Palmieri EA, Lombardi G, et al: Effects of thyroidhormone on cardiac function: The relative importance ofheart rate, loading conditions, and myocardial contractil-ity in the regulation of cardiac performance in humanhyperthyroidism. J Clin Endocrinol Metab 2002;87:968–974.

18. Bielecka-Dabrowa A, Mikhailidis DP, Rysz J, et al: Themechanism of atrial fibrillation in hyperthyroidism.Thyroid Res 2009;2:4. doi:10.1186/1756-6614-2-4.

19. Fazio S, Palmieri EA, Lombardi G, et al: Effects of thyroidhormone on the cardiovascular system. Recent Prog HormRes 2004;59:31–50.

20. Biondi B, Palmieri EA, Lombardi G, et al: Effects of sub-clinical thyroid dysfunction on the heart. Ann Intern Med2002;137:904–914.

21. Biondi B, Palmieri EA, Fazio S, et al: Endogenous subclini-cal hyperthyroidism affects quality of life and cardiacmorphology and function in young and middle agedpatients. J Clin Endocrinol Metab 2000;85:4701–4705.

22. Kahaly GJ, Dillmann WH: Thyroid hormone action in theheart. Endocr Rev 2005;26:704–728.

23. Biondi B, Palmieri EA, Klain M, et al: Subclinical hyperthy-roidism: Clinical features and treatment options. EurJ Endocrinol 2005;152:1–9.

24. Petretta M, Bonaduce D, Spinelli L, et al: Cardiovascularhaemodynamics and cardiac autonomic control inpatients with subclinical and overt hyperthyroidism. EurJ Endocrinol 2001;145:691–696.

25. Smit JW, Eustatia-Rutten CF, Corssmit EP, et al: Reversiblediastolic dysfunction after long-term exogenous subclini-cal hyperthyroidism: A randomized, placebo-controlledstudy. J Clin Endocrinol Metab 2005;90:6041–6047.

26. Klein I, Danzi S: Thyroid disease and the heart. Circulation2007;116:1725–1735.

27. Wang K, Xiao HB, Fujimoto S, et al: Atrial electromechan-ical sequence in normal subjects and patients with DDDpacemakers. Heart 1995;74:403–407.

28. Dabrowska-Kugacka A, Lewicka-Nowak E, Rucinski P,et al: Relationship between P-wave duration and atrialelectromechanical delay assessed by tissue Doppler echo-cardiography. PACE 2011;34:23–31.

29. Deniz A, Sahiner L, Aytemir K, et al: Tissue Doppler echo-cardiography can be a useful technique to evaluate atrialconduction time. Cardiol J 2012;19:487–493.

30. Karabag T, Aydin M, Dogan SM, et al: Investigation ofthe atrial electromechanical delay duration in Behcetpatients by tissue Doppler echocardiography. Eur HeartJ Cardiovasc Imaging 2012;13:251–256.

31. Acar G, Akcay A, Sokmen A, et al: Assessment of atrialelectromechanical delay, diastolic functions, and leftatrial mechanical functions in patients with type 1 diabe-tes mellitus. J Am Soc Echocardiogr 2009;22:732–738.

32. Akcay A, Acar G, Suner A, et al: Effects of slow coronaryartery flow on P-wave dispersion and atrial electrome-chanical coupling. J Electrocardiol 2009;42:328–333.

33. Acar G, Akcay A, Sayarlioglu M, et al: Assessment of atrialconduction time in patients with familial Mediterraneanfever. PACE 2009;32:308–313.

34. Acar G, Sayarlioglu M, Akcay A, et al: Assessment of atrialelectromechanical coupling characteristics in patientswith ankylosing spondylitis. Echocardiography 2009;26:549–557.

35. Aktoz M, Yilmaztepe M, Tatli E, et al: Assessment ofventricular and left atrial mechanical functions, atrial elec-tromechanical delay and P wave dispersion in patientswith scleroderma. Cardiol J 2011;18:261–269.

36. Ayhan S, Ozturk S, Dikbas O, et al: Detection of subclini-cal atrial dysfunction by two-dimensional echocardiogra-phy in patients with overt hyperthyroidism. ArchCardiovasc Dis 2012;105:631–638.

37. Komiya N, Isomoto S, Kakao K, et al: Electrophysiologicalabnormalities of the atrial muscle in patients with parox-ysmal atrial fibrillation associated with hyperthyroidism.Clin Endocrinol 2002;56:39–44.

1201


evaluation of atrial electromechanical delay and diastolic functions in patients with...

Documents