Listen to this manuscript’s

audio summary by

JACC Editor-in-Chief

Dr. Valentin Fuster.

J O U R N A L O F T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y VO L . 6 9 , N O . 1 , 2 0 1 7

ª 2 0 1 7 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N

P U B L I S H E D B Y E L S E V I E R

I S S N 0 7 3 5 - 1 0 9 7 / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 6 . 1 0 . 0 4 4

ORIGINAL INVESTIGATIONS

Anterior T-Wave Inversion in YoungWhite Athletes and NonathletesPrevalence and Significance

Aneil Malhotra, MBBCHIR, MA, MSC,a Harshil Dhutia, MBBS, BSC,a Sabiha Gati, MBBS, PHD,a Tee-Joo Yeo, MD,a

Helder Dores, MD,b Rachel Bastiaenen, MBBS, PHD,a Rajay Narain, MBBS,a Ahmed Merghani, MBBS, BMEDSCI,a

Gherardo Finocchiaro, MD,a Nabeel Sheikh, MBBS, BSC,a Alexandros Steriotis, MD, PHD,a

Abbas Zaidi, MBBS, MD, BSC, MD,a Lynne Millar, MBBS, BSC,a Elijah Behr, MBBS, MD,a Maite Tome, MD, PHD,a

Michael Papadakis, MBBS, MD,a Sanjay Sharma, MBBS, BSC, MDa

ABSTRACT

Fro

Ca

Ma

zat

ha

Ma

BACKGROUND Anterior T-wave inversion (ATWI) on electrocardiography (ECG) in young white adults raises the pos-

sibility of cardiomyopathy, specifically arrhythmogenic right ventricular cardiomyopathy (ARVC). Whereas the 2010

European consensus recommendations for ECG interpretation in young athletes state that ATWI beyond lead V1 warrants

further investigation, the prevalence and significance of ATWI have never been reported in a large population of

asymptomatic whites.

OBJECTIVES This study investigated the prevalence and significance of ATWI in a large cohort of young, white adults

including athletes.

METHODS Individuals 16 to 35 years of age (n ¼ 14,646), including 4,720 females (32%) and 2,958 athletes (20%),

were evaluated by using a health questionnaire, physical examination, and 12-lead ECG. ATWI was defined as T-wave

inversion in $2 contiguous anterior leads (V1 to V4).

RESULTS ATWI was detected in 338 individuals (2.3%) and was more common in women than in men (4.3% vs. 1.4%,

respectively; p < 0.0001) and more common among athletes than in nonathletes (3.5% vs. 2.0%, respectively;

p < 0.0001). T-wave inversion was predominantly confined to leads V1 to V2 (77%). Only 1.2% of women and 0.2% of

men exhibited ATWI beyond V2. No one with ATWI fulfilled diagnostic criteria for ARVC after further evaluation. During a

mean follow-up of 23.1 � 12.2 months none of the individuals with ATWI experienced an adverse event.

CONCLUSIONS ATWI confined to leads V1 to V2 is a normal variant or physiological phenomenon in asymptomatic

white individuals without a relevant family history. ATWI beyond V2 is rare, particularly in men, and may warrant

investigation. (J Am Coll Cardiol 2017;69:1–9) © 2017 by the American College of Cardiology Foundation.

T here is general agreement that T-waveinversion (TWI) in the inferior or lateralleads in young individuals warrants

further investigation for cardiac disease, particularly

m the aDivision of Cardiovascular Sciences, St. George’s University of

rdiology, Universidade Nova de Lisboa, Hospital das Forças Armadas, Luz

lhotra, Dhutia, Yeo, Steriotis, Finocchiaro, Narain, and Millar were suppo

ion Cardiac Risk in the Young. Dr. Merghani was supported by the British H

ve no relationships relevant to the contents of this paper to disclose.

nuscript received July 13, 2016; revised manuscript received October 1, 2

cardiomyopathy (1). It is also well established thatadolescent athletes (2–6) and black adult athletes (7)frequently exhibit TWI in the anterior leads as partof the normal physiological or ethnic spectrum

London, United Kingdom; and the bDepartment of

Saúde, NOVA Medical School, Lisbon, Portugal. Drs.

rted by a research grant from the charitable organi-

eart Foundation. All authors have reported that they

016, accepted October 11, 2016.

ABBR EV I A T I ON S

AND ACRONYMS

ARVC = arrhythmogenic right

ventricular cardiomyopathy

ATWI = anterior T-wave

inversion

CMRI = cardiac magnetic

resonance imaging

ECG = electrocardiogram

Jt = J point

SCD = sudden cardiac death

TWI = T-wave inversion

Malhotra et al. J A C C V O L . 6 9 , N O . 1 , 2 0 1 7

ATWI in Young White Athletes and Nonathletes J A N U A R Y 3 / 1 0 , 2 0 1 7 : 1 – 9

2

respectively. However, the generalconsensus for the significance of anteriorTWI (ATWI), defined as T-wave inversionin $2 contiguous anterior leads (V1 to V4) inwhite adults varies among expert recommen-dations for the interpretation of the athlete’selectrocardiogram (ECG). Whereas the Euro-pean Society of Cardiology recommendationssuggest further evaluation of athletes withTWI beyond lead V1 (8), more recent recom-mendations from the Seattle criteria advo-cate investigation only if TWI extendsbeyond V2 (9).

Both of the consensus panels have relied on datafrom unselected (10) or small cohorts of athletes (11);however recent studies reveal that TWI in leads V1 toV2/V3 is detected in up to 6%of endurance athletes (12).Conversely, ATWI in V1 to V2/V3 is a recognized repo-larization abnormality in a significant proportion ofpatients with arrhythmogenic right ventricular car-diomyopathy (ARVC) and a minority of patients withhypertrophic cardiomyopathy (7), which collectivelyaccounts for >40% of all sudden cardiac deaths (SCDs)in young athletes (13). Differentiation of potentiallypathological ATWI from a pattern that represents anormal variant or physiological remodeling in whiteadult athletes is essential to minimize the risk andconsequences of an erroneous diagnosis (6,14).

SEE PAGE 10

Because the prevalence of ATWI has been reportedin black athletes, controls of both sexes, and theadolescent population, the present study focused onthe prevalence and significance of ATWI in a largecohort of apparently healthy white adults including alarge proportion of athletes.

METHODS

SETTING. The United Kingdom does not support anationally sponsored screening program for cardiacdisease in young asymptomatic individuals in theabsence of a family history of inherited cardiac dis-ease or premature SCD. Several elite sporting orga-nizations finance the evaluation of their athletesthrough the charitable organization CRY (Cardiac Riskin the Young). These organizations include premierleague football clubs, the Lawn Tennis Association,and the English Institute of Sport. Up to 1,000 ath-letes are tested annually at their specific clubs ornational training camps, usually using history, ex-amination, and ECG. Financially endowed organiza-tions such as the Football Association and the Lawn

Tennis Association also incorporate echocardiogra-phy as a standard test.

CRY also offers cardiac screening to all young in-dividuals (14 to 35 years of age) who wish to beassessed, even in the absence of symptoms, historyof cardiac disease, or a family history of inheritedcardiac diseases or SCD. Such screenings are con-ducted at community centers and high schools and arelimited to history, examination, and ECG with referralfor further assessment only in those with abnormalpreliminary findings or if participating as controls forresearch studies. Screening events are advertisedthrough the local media and on the CRY website.Individuals from the general population, includingthose from local high schools, self-present toscreening events whereas competitive athletes attendspecified screening events mandated by their relevantsporting bodies. The CRY screening program is su-pervised by our principal investigator (S.S.).

SUBJECTS. Between 2007 and 2013, 14,646 young,white adults between 16 and 35 years of age under-went cardiac evaluation through CRY at varioustesting centers in England. Ethnicity was self-reported through the questionnaire that includedterms such as white British, white Irish, white Euro-pean, and white other.Athletes . The study included 2,958 athletes (20.2%)competing at regional, national, or international levelswho performed $8 h of exercise per week. Sportingdisciplines were categorized as predominantlyendurance or strength. Endurance sports were definedas those typically resulting in >70% maximal oxygenuptake (VO2max) (15) and included badminton, basket-ball, canoeing, cycling, hockey, middle- and long-distance running, rowing, rugby, soccer, squash,swimming, tennis, and triathlon. All other sports weredeemed strength disciplines, including cricket, diving,sailing, volley ball, water polo, weight lifting, andwrestling.Nonath letes . The nonathlete cohort consisted of11,688 individuals (79.8%) whose primary inclusioncriterion was a sedentary lifestyle (#2 h organizedphysical activity per week). Individuals with symp-toms suggestive of cardiac disease, history or familyhistory of cardiac disease or premature cardiac dis-ease or SCD (<50 years of age) were excluded.

ELECTROCARDIOGRAPHY. A standard 12-lead ECGwas performed while patient were in a supine posi-tion, using a Marquette Hellige recorder (Marquette-Hellige Medical Systems, Milwaukee, Wisconsin) at apaper speed of 25 mm/s. Wave voltages P, Q, R, S, andT and ST-segments QRS, PR, and QT intervals weremeasured in each lead as described elsewhere (16).

FIGURE 1 ST-Segment Morphology

(A) The vertical solid line and the vertical dashed line define the M interval, which has a duration of 100 ms. The horizontal dashed line through the onset of the QRS

complex provides the reference point for the measurement of J point (Jt). Jt is elevated at 0.2 mV with a convex appearance. ST-segment morphologies with anterior

T-wave inversion in chest leads V2 and V3 are shown as the following: (B) ascending convex, (C) isoelectric, (D) ascending concave, and (E) depressed.

J A C C V O L . 6 9 , N O . 1 , 2 0 1 7 Malhotra et al.J A N U A R Y 3 / 1 0 , 2 0 1 7 : 1 – 9 ATWI in Young White Athletes and Nonathletes

3

Leads V1 to V4 were classified as anterior precordialleads. T-wave deflection $�0.1 mV in these leadswas regarded as abnormal TWI. Deep TWI wasdefined as a T-wave amplitude $�0.2 mV. In cases ofbiphasic T-waves, we applied the definition to thenegative component of the T-wave. In cases of ATWI,the ECG was repeated to ensure that the leads werecorrectly positioned according to standard recom-mendations. In women, the ECG electrodes wereplaced under the breast tissue according to AmericanHeart Association recommendations (17). Partial rightbundle branch block was defined as a QRS duration>0.1 but <0.12 s, with rSR0 morphology in lead V1 andQRS in V6 (18). Individuals with TWI and completeright bundle branch block (QRS: $0.12 s) wereexcluded from the ATWI group. Additional ECGmarkers compatible with ARVC were also sought,including terminal activation duration of the QRScomplex $55 ms in lead V1, V2, or V3 and the epsilonwave (19).

The amplitude of the J point (Jt) (20) was measuredat the end of the QRS complex (the onset of the ST-segment) with reference to the onset of the QRScomplex. The Jt was considered elevated if Jt $0.1 mVor depressed if Jt #�0.1 mV. The morphology of theST-segment in the anterior leads was ascertained inthe interval, M (during the 100 ms following Jt) (20).The ST-segment at the onset of the M interval (i.e., Jt)was considered elevated if it was above Jt, depressedif it was below Jt, and isoelectric if it was in line

with Jt. Ascending ST-segments were categorized asascending convex or ascending concave (Figure 1).

ECHOCARDIOGRAPHY. Two-dimensional (2D) trans-thoracic echocardiography was performed in all sub-jects with ATWI by using Sonos 7500 system (modelCPX50, iE33, Philips, Baltimore, Maryland) and Vivid I(GE, Tiral, Israel) machines. Standard views were ob-tained, and dimensions of cavities and wall thicknessmeasurements and pulsed color and tissue Dopplermeasurements were made in accordance with estab-lished guidelines (21–23). Right ventricular assess-ment was performed as outlined previously (14). Rightventricular regional wall-motion abnormalities weredefined as akinetic, dyskinetic, or aneurysmal, inaccordance with diagnostic criteria for ARVC (19).

Echocardiography was also performed as standardtesting in 1,079 athletes and 769 nonathletes withoutATWI of similar age and sex proportion whose exam-inations had normal results and ECGs. The echocar-diogram was part of the mandatory pre-participationcardiac evaluation in athletes, although the echocar-diogram was conducted as part of research in volun-teering nonathletes. These cohorts served ascomparative groups for athletes and nonathletes withATWI, respectively.

All ECG and echocardiograms were performed bynationally-accredited cardiac physiologists. Echocar-diography was conducted by physiologists blinded tothe ECG findings. All ECG and echocardiogram images

TABLE 1 Demographics and ECG Characteristics of Individuals With and

Without ATWI

PopulationWith ATWI(n ¼ 338)

PopulationWithout ATWI(n ¼ 14,308) p Value

Demographics

Age, yrs 21.1 � 5.4 21.7 � 5.3 0.0398

Female 60.1 31.6 <0.0001

Athlete 30.5 20.0 0.0003

BSA, m2 1.81 � 0.3 1.91 � 0.2 <0.0001

Blood pressure, mm Hg 121/66 � 12/7 123/80 � 10/6 0.0003

ECG parameters

Heart rate, beats/min 64 � 14 66 � 14 <0.0001

PR, ms 150 � 25 151 � 32 <0.0001

QRS, ms 93 � 12 92 � 13 <0.0001

QTc, ms 421 � 28 412 � 20 <0.0001

Incomplete RBBB 17.7 5.5 <0.0001

LBBB 0 0.02 0.77

LVH 17.9 10.1 <0.0001

RVH 1.2 1.1 0.93

ER 15.1 9.7 0.0018

Pathological Q waves 0.5 0.3 <0.0001

LA enlargement 3.3 1.4 0.002

RA enlargement 1.4 0.6 0.1

LAD 1.0 1.2 0.83

RAD 0.7 0.4 0.49

Pre-excitation 0.5 0.5 0.93

Values are mean � SD or % overall population.

ATWI ¼ anterior T-wave inversion; BSA ¼ body surface area; ECG ¼ 12-lead electrocardiogram;ER ¼ early repolarization; LA ¼ left atrial; LAD ¼ left axis deviation; LBBB ¼ left bundle branchblock; LVH ¼ left ventricular hypertrophy; RA ¼ right atrial; RAD ¼ right axis deviation; RBBB ¼right bundle branch block; RVH ¼ right ventricular hypertrophy; TWI ¼ T-wave inversion.

Malhotra et al. J A C C V O L . 6 9 , N O . 1 , 2 0 1 7

ATWI in Young White Athletes and Nonathletes J A N U A R Y 3 / 1 0 , 2 0 1 7 : 1 – 9

4

were reviewed by 2 independent cardiologists withthe principal investigator (S.S.) adjudicating anyqueries.

FURTHER INVESTIGATIONS. All subjects withATWI underwent additional investigations to detectthe broader phenotypic features of a primarycardiomyopathy, particularly ARVC, hypertrophiccardiomyopathy and dilated cardiomyopathy. Pre-determined diagnostic criteria for ARVC were basedon the 2010 Modified Task Force criteria (19). Hyper-trophic cardiomyopathy was considered in in-dividuals with left ventricular hypertrophy whereseptal or wall thickness measured $15 mm in anymyocardial segment in the absence of another condi-tion capable of producing left ventricular hypertrophyof the same magnitude (24,25). Dilated cardiomyopa-thy was considered in individuals with a dilated LV(men >59 mm and women >53 mm) when accompa-nied by a reduced ejection fraction (<52%) (26). Thevast majority of further investigations (n ¼ 1,396; 95%)were performed at our institution.Ambulatory ECG monitor ing . Ambulatory 24-hECG recording (Lifecard CF Holters, Spacelabs

Healthcare, Snoqualmie, Washington) was used todetect ventricular arrhythmias. Subjects wereencouraged to continue day-to-day activitiesincluding exercise during monitoring.

Exerc i se test ing . Exercise testing was performedupright on a treadmill using the standard Bruceprotocol (27). Subjects were exercised to volitionalexhaustion and assessed for cardiac symptoms,ischemic changes, attenuated blood pressureresponse, or arrhythmias.

S ignal -averaged ECG. Signal-averaged ECG(SAECG) was acquired according to accepted meth-odology using the same machines used for standardECG, with use of a 40 Hz high-pass bidirectional filter(28). Late potentials were defined as abnormal valuesin 1 or more of the parameters in accordance thediagnostic criteria for ARVC (19).

Card iac magnet ic resonance imaging . Cardiacmagnetic resonance imaging (CMRI) was performedusing a Philips Achiever 3.0T TX scanner (Philips,Amsterdam, the Netherlands). Delayed gadoliniumenhancement images were acquired as previouslydescribed (29). Ventricular volumes and functionwere measured for both ventricles using standardtechniques and analyzed using semi-automatedsoftware (Extended MR workspace, Philips) (30). Allmeasures were indexed to body surface area.

ETHICAL APPROVAL. Ethics approval was grantedby the National Research Ethics Service, Essex 2Research Ethics Committee in the United Kingdom.Written consent was obtained from all subjects.

STATISTICAL ANALYSIS. Data are mean � SD orpercentages, as appropriate, and were analyzed usingSPSS version 20 software (SPSS, Chicago, Illinois).Comparisons between groups were performed usingStudent t test for continuous variables with adjust-ment for unequal variance if needed and chi-squaretests or Fisher exact tests for categorical variables.Univariate analyses were performed to determinevariables (sex, age, athletic status, left ventricularend diastolic diameter, and right ventricular outflowtract size [parasternal long- and short-axis measure-ments]) associated with ATWI. Multivariate logisticregression analyses were used to determine the in-dependence of these associations. Significance wasdefined as p < 0.05.

RESULTS

DEMOGRAPHICS. The mean age of the cohort was21.7 � 5.4 years of age. Of the 14,646 subjects, 9,926were men (67.8%); 2,063 men (20.8%) and 895women (19.0%) were athletes. Athletes exercised for

CENTRAL ILLUSTRATION Prevalence of Anterior T-Wave Inversion in the Adult White Population

Malhotra, A. et al. J Am Coll Cardiol. 2017;69(1):1–9.

The overall prevalence of anterior T-wave inversion (ATWI) in adult white individuals (16 to 35 years of age) was 2.3%. ATWI was more common in women and in

athletes. The prevalence of ATWI beyond V2 was rare, falling to 0.2% in male nonathletes.

J A C C V O L . 6 9 , N O . 1 , 2 0 1 7 Malhotra et al.J A N U A R Y 3 / 1 0 , 2 0 1 7 : 1 – 9 ATWI in Young White Athletes and Nonathletes

5

an average of 15.7 � 5.1 h/week compared with non-athletes, who exercised 1.8 � 0.6 h/week.

PREVALENCE OF ATWI. A total of 338 individuals(2.3%) exhibited ATWI. Individuals with ATWI were ofsimilar age and had a similar mean body surface areacompared with those without ATWI (Table 1). ATWIwas more common in women than in men (n ¼ 203[4.3%] vs. n ¼ 135 [1.4%], respectively; p < 0.0001) andwas more common in athletes than in nonathletes(n ¼ 103 [3.5%] vs. n ¼ 235 [2%], respectively;p < 0.0001) in both sexes (women: n ¼ 58 [6.5%] vs.n ¼ 145 [3.8%]; p ¼ 0.0005, and men: n ¼ 45 [2.1%] vs.n ¼ 90 [1.1%]; p¼ 0.0004) (Central Illustration). Amongathletes, ATWI was more prevalent in those engagingin endurance sports than in strength sports athletes(n ¼ 82 [5.6%] vs. n ¼ 41 [2.8%]; respectively;p < 0.0001). The prevalence of ATWI among those16 to 21 years of age was similar to that in those 21years of age and older (2.28% vs. 2.46%, respectively;p ¼ 0.52).

DISTRIBUTION OF ATWI. A total of 260 individuals(1.8%) revealed TWI confined to leads V1 to V2. Thosein whom TWI was confined to V1 to V2 constituted77% of all ATWI cases. Only 78 individuals (0.5%)

demonstrated TWI beyond V2, which was present in56 women (1.2%) versus 22 men (0.2%) (p < 0.0001).Among athletes, TWI in leads V1 to V3 was detected in19 women (2.1%) versus 7 men (0.3%) (p < 0.0001)(Central Illustration). Four women but none of themen showed TWI extending to V4, which equated to2% of all ATWI in females.

Deep ATWI was more common in males than infemales (55.6% vs. 33%, respectively; p ¼ 0.0166) butdid not differ between athletes and nonathletes. Fiftyindividuals with ATWI (14.8%) exhibited incompleteright bundle branch block that never extendedbeyond V2.

JT ELEVATION AND ST-SEGMENT MORPHOLOGY

PRECEDING ATWI. Among individuals with ATWI, Jtelevation was more common in athletes than innonathletes (49% vs. 29%, respectively; p ¼ 0.0008)and more common in men than in women regardlessof athletic status (athletes: 71.1% vs. 31.0%, respec-tively; p ¼ 0.0004; nonathletes: 58.9% vs. 10.3%,respectively; p < 0.0001). None of the individualswith ATWI demonstrated a depressed Jt.

Men frequently showed an elevated ST-segment,that of ascending convex morphology (42%),

FIGURE 2 Bar Graph Shows ST-Segment Morphology Type Preceding ATWI in Healthy

Individuals and ATWI According to Sex

60

50

40

30

20

10

0

%

AscendingConvex

AscendingConcave

Isoelectric

ATWI males ATWI females

Depressed

An ascending convex and an ascending concave ST-segment morphology were more

common in men than women. Women more commonly demonstrated an isoelectric

ST-segment. ATWI ¼ anterior T-wave inversion.

Malhotra et al. J A C C V O L . 6 9 , N O . 1 , 2 0 1 7

ATWI in Young White Athletes and Nonathletes J A N U A R Y 3 / 1 0 , 2 0 1 7 : 1 – 9

6

followed by an ascending concave morphology (33%)and an isoelectric pattern (25%). In women withATWI, the ST segment was most commonly isoelectric(57%), followed by ascending convex (24%) andascending concave (19%) morphologies. Only 1 indi-vidual with ATWI demonstrated a depressed ST-segment (Figure 2).

CARDIAC STRUCTURE AND FUNCTION IN

INDIVIDUALS WITH ATWI. The echocardiographicresults of all 338 individuals with ATWI (103 athletes,235 nonathletes) were compared with the results of1,848 individuals without ATWI (1,079 athletes, 769nonathletes). Athletes revealed larger ventricular di-mensions than nonathletes regardless of ATWI. Therewere no differences in left or right ventricular di-mensions or function in individuals (athletes andnonathletes) with ATWI compared to those withoutAWTI (Table 2).

Cardiac MRI was performed in 250 subjects(74%) with ATWI. Athletes demonstrated larger leftand right ventricular volumes and masses thandid nonathletes (Table 2). Following gadolinium-enhanced MRI, there was no evidence of lateenhancement in any subject. None of the individualswith ATWI showed unequivocal diagnostic featuresof ARVC, hypertrophic cardiomyopathy, or dilatedcardiomyopathy.

SAECG. A total of 316 individuals (93%) with ATWIunderwent SAECG, and 21 (7%) showed an abnor-mality in 1 of the 3 parameters. The most commonabnormality was filtered QRS prolongation (60%), a

phenomenon that has been reported previously inhealthy individuals (31,32).

EXERCISE STRESS TESTING AND AMBULATORY ECG

MONITORING. A total of 274 individuals (81%) withATWI underwent an exercise stress test, and 293(87%) had 24-h ECG monitoring. None of the in-dividuals with ATWI exhibited arrhythmia duringexercise, other than occasional isolated ventricularectopic beats (n ¼ 10; 3%) of right or left ventricularorigin in the early stages of exercise. Similarly, noneshowed >500 ventricular ectopic beats or runs ofnonsustained ventricular tachycardia during Holtermonitoring (19).

DETERMINANTS OF ATWI. Univariate predictors ofATWI were female sex and athletic status. Stepwiselogistic regression identified female sex (odds ratio[OR]: 3.1; 95% confidence interval [CI]: 1.96 to 4.90;p < 0.001) and athletic status (OR: 3.3; 95% CI: 1.91to 5.63; p ¼ 0.001) as being independently associ-ated with ATWI in the screened adult population,regardless of age.

DETECTION OF CARDIAC PATHOLOGY. Followingcomprehensive clinical evaluation of 274 individuals(81%) with ATWI (including echocardiography in all338 individuals and a mean follow-up period of23.1 � 12.2 months), we could not diagnose ARVC orany other cardiomyopathy. However, 16 athletesand 10 nonathletes with ATWI fell into the gray zonein which structural changes attributed to physiolog-ical adaptation needed to be differentiated fromprimary cardiomyopathies. These subjects included 2athletes and 3 nonathletes with an indexed para-sternal right ventricular outflow tract dimension$19 mm/m2; 1 athlete with a maximal left ventricularwall thickness of 13 mm; and 6 nonathletes whoinitially demonstrated an absolute LVEDD (left ven-tricular end diastolic diameter) above the upperlimit of normal (>59 mm LVEDD in male nonathletes,>53 mm in women [3], and LVEDD of >60 mm inathletes [33]).

IDENTIFICATION OF MINOR CARDIAC PATHOLOGY.

Echocardiography in all 338 subjects with ATWIfailed to show akinetic segments or regional wallmotion abnormalities affecting the right ventricle. Asmall proportion revealed minor pathology in 5 sub-jects (1.5%) including bicuspid aortic valve (n ¼ 2;0.6%), mitral valve prolapse with moderate mitralregurgitation (n ¼ 1; 0.3%), atrial septal defect (n ¼1;0.3%), and patent ductus arteriosus (n ¼ 1; 0.3%).Seven individuals (2%) had patent foramen ovalenoted, and pectus excavatum was noted in 2 cases(0.6%).

TABLE 2 Echocardiographic and CMRI Measurements of Athletes and Nonathletes With

and Without ATWI

Athletes (n ¼ 1,182) Nonathletes (n ¼ 1,004)

With ATWI(n ¼ 103)

Without ATWI(n ¼ 1,079) p Value

With ATWI(n ¼ 235)

Without ATWI(n ¼ 769) p Value

Ao, mm 28.0 � 4.3 28.6 � 4.5 0.2306 27.1 � 3.6 27.4 � 3.8 0.2839

LA, mm 33.1 � 6.1 33.5 � 4.9 0.4394 31.0 � 4.3 31.3 � 5.5 0.443

LVEDD, mm 50.8 � 5.8 51.5 � 5.6 0.2272 48.1 � 4.5 48.5 � 5.8 0.3315

LVESD, mm 34.0 � 6.1 33.9 � 4.8 0.844 31.1 � 3.9 31.6 � 5.1 0.1666

MLVWT, mm 9.2 � 2.3 8.9 � 2.1 0.1699 8.5 � 2.2 8.3 � 1.5 0.1126

LVMI, g/m2 105 � 15 103 � 16 0.2233 94 � 8 95 � 9 0.1267

% EF 60 � 9 59 � 8 0.231 66 � 8 67 � 9 0.1267

RVOTplax, mm 29.9 � 5.4 29.8 � 4.8 0.8417 26.0 � 3.7 25.4 � 3.6 0.1072

RVOTplax, mm/m2 16.7 � 2.1 16.8 � 3.7 0.7871 15.2 � 2.5 14.8 � 3.1 0.2041

RVOTpsax, mm 31.6 � 4.9 32.3 � 5.6 0.221 28.2 � 5.8 29.1 � 5.7 0.1266

RVOTpsax, mm/m2 17.8 � 2.5 17.5 � 2.9 0.3106 16.0 � 2.1 16.1 � 2.5 0.6945

RVOT2, mm 25.8 � 4.8 25.1 � 4.4 0.1263 22.6 � 3.7 23.2 � 4.3 0.0537

RVD1, mm 41.1 � 6.6 40.6 � 5.8 0.4093 35.3 � 4.9 35.6 � 5.5 0.4534

RVD2, mm 34.1 � 6.4 33.3 � 5.5 0.165 28.2 � 4.5 28.7 � 5.7 0.2181

RVD3, mm 84.2 � 10.5 82.0 � 13.2 0.1008 73.5 � 11.4 74.9 � 11.1 0.093

RVWT, mm 4.8 � 1.5 4.6 � 1.3 0.1416 4.2 � 0.8 4.1 � 1.0 0.1613

TAPSE, mm 23.4 � 5.3 23.5 � 4.2 0.8219 22.8 � 4.7 22.9 � 4.5 0.768

PASP, mm Hg 17.6 � 7.7 15.9 � 6.5 0.0128 17.8 � 3.3 18.3 � 4.0 0.0816

TV E/A 1.9 � 0.5 2.0 � 0.4 0.0181 1.9 � 0.6 2.0 � 0.8 0.0771

TV S0, cm/s 14.8 � 2.6 14.9 � 2.5 0.6992 14.6 � 2.8 14.2 � 2.8 0.0556

TV E0, cm/s 13.9 � 3.4 14.1 � 3.1 0.5353 14.9 � 2.9 15.1 � 3.5 0.426

RAA, cm2 19.2 � 3.4 18.8 � 3.7 0.2915 15.2 � 2.8 15.1 � 4.8 0.7613

% RV FAC 38.7 � 4.9 39.6 � 4.8 0.0698 36.2 � 4.1 37.3 � 6.2 0.0108

CMRIATWI Athletes

(n ¼ 76)ATWI Nonathletes

(n ¼ 174)

LVM-I, g/m2 105 � 15 94 � 9 <0.05

LVEDVI, ml/m2 105.8 � 15 94.3 � 14.0 <0.05

% RVEF 52.5 � 5.1 55.5 � 5.9 <0.05

RVEDVI, ml/m2 105.3 � 14.0 94.3 � 14.0 <0.05

% LGE 0 0 –

Values are mean � SD.

Ao ¼ aorta; ATWI ¼ anterior T-wave inversion; CMRI ¼ cardiac magnetic resonance imaging; EF ¼ ejectionfraction according to Simpson’s biplane; FAC ¼ fractional area change; LA ¼ left atrial; LGE ¼ late gadoliniumenhancement; LVEDD ¼ left ventricular end-diastolic diameter; LVEDV ¼ left ventricular end-diastolic volume;LVEDVI ¼ left ventricular end-diastolic volume index; LVESD ¼ left ventricular end-systolic diameter; LVM-I ¼left ventricular mass index; MLVWT ¼ maximum left ventricular wall thickness; PASP ¼ pulmonary artery systolicpressure; RAA ¼ right atrial area; RV ¼ right ventricular; RVD1 ¼ right ventricular basal dimension; RVD2 ¼ rightventricular midventricular dimension; RVD3 ¼ right ventricular longitudinal dimension; RVEDV¼ right ventricularend-diastolic volume; RVEDVI ¼ right ventricular end-diastolic volume index; RVEF ¼ right ventricular ejectionfraction; RVOT1 ¼ proximal right ventricular outflow tract dimension; RVOT2 ¼ distal right ventricular outflowtract dimension; RVOTplax ¼ right ventricular outflow tract dimension (parasternal); RVOTsax ¼ right ventricularoutflow tract dimension (short axis); RVWT ¼ right ventricular free wall thickness; S0 ¼ peak systolic velocity;TAPSE ¼ tricuspid annular plane systolic excursion; TV ¼ tricuspid valve; TV E0 ¼ tricuspid valve early myocardialrelaxation velocity.

J A C C V O L . 6 9 , N O . 1 , 2 0 1 7 Malhotra et al.J A N U A R Y 3 / 1 0 , 2 0 1 7 : 1 – 9 ATWI in Young White Athletes and Nonathletes

7

DISCUSSION

Detection of lateral or inferolateral TWI in youngblack or white individuals has a relatively high yieldfor the diagnosis of cardiomyopathy (1). AlthoughATWI is a benign variant in healthy adolescents of allethnic origins and in black adolescent and adult ath-letes, its significance in asymptomatic white adults isunknown. However, between 50% and 60% of pro-bands with ARVC show ATWI in leads V1 to V3 (34).This study of almost 15,000 healthy, white adults,including 4,720 women and almost 3,000 athletes,showed that ATWI beyond V1 was present in a smallproportion of individuals (2.3%) and that this preva-lence fell to just 0.5% beyond V2. ATWI was morecommon in women than in men regardless of athleticstatus and validates data from much smaller studiesperformed 6 to 7 decades ago (35,36). Several postu-lations for these sex differences have been proposed,including various levels of sympathetic innervationand anatomical differences in chest wall structure,specifically breast tissue. Based on the fact thatthe prevalence of ATWI is almost identical in prepu-bertal boys and girls (3), we suspect that sex differ-ences in adults are likely to reflect differences in leadplacement as a result of increased breast tissue inwomen.

PREVALENCE OF ATWI IN ATHLETES. Athletesdemonstrated a greater prevalence of ATWI thandid nonathletes, particularly those engaging in >15h/week of exercise. Such intense exercise regimens,particularly in endurance sports, place a greater he-modynamic load on the right ventricle that maymanifest on the ECG as ATWI. Our study, however,was unable to demonstrate any structural differencesbetween the right ventricles of individuals with ATWIand those of individuals without.

SIGNIFICANCE OF EXTRAPOLATING DATA FROM

PROBANDS WITH CARDIOMYOPATHY TO LOW-RISK

POPULATIONS. There are justifiable concerns aboutthe association of ATWI with an underlying cardio-myopathy such as ARVC or hypertrophic cardiomy-opathy. While isolated ATWI is rare in hypertrophiccardiomyopathy (2), ATWI beyond V2 in probandswith ARVC is common and classified as a majorrepolarization abnormality in the Revised Task ForceCriteria for ARVC (19). In this study, none of theathletes with ATWI in leads V2/V3 fulfilled diagnosticcriteria for ARVC based on a combination of healthquestionnaire, ECG, and echocardiography in 100% ofcases, signal-averaged ECG in 93%, 24-h ECG in 87%,exercise testing in 81%, and cardiac magnetic reso-nance imaging in 74% of subjects. This observation

highlights the fact that data derived from probandswith ARVC for generating diagnostic criteria lackspecificity in low-risk populations (29). However,TWI beyond V2 was present in just 1 in 200 whiteadult athletes and could justify detailed assessmentto exclude ARVC or any other cardiomyopathy.Our data support the consensus-based Seattle rec-ommendations, which pragmatically suggest thatonly TWI beyond V2 in asymptomatic white athletesrequires further evaluation (9). However, these

PERSPECTIVES

COMPETENCY IN MEDICAL KNOWLEDGE: ATWI

confined to leads V1 and V2 may be a normal variant or

physiological phenomenon in asymptomatic white

individuals without a family history of relevant heart

disease.

TRANSLATIONAL OUTLOOK: Further research is

needed to understand the clinical significance of TWI

beyond lead V2, which occurs in approximately 1 in

200 young, white athletes and in an unknown pro-

portion of nonathletes.

Malhotra et al. J A C C V O L . 6 9 , N O . 1 , 2 0 1 7

ATWI in Young White Athletes and Nonathletes J A N U A R Y 3 / 1 0 , 2 0 1 7 : 1 – 9

8

recommendations are at odds with recommendationsfrom the European Society of Cardiology and therecently published refined criteria (8,37). Given thepotentially sinister ramifications of false negativetests with regard to ARVC in particular, more robustdata are necessary before such criteria can be adoptedwith more certainly in future updates for ECG inter-pretation in athletes. This comprehensive study of alarge population of athletes with ATWI providessupport for the Seattle consensus.

POTENTIAL MARKERS OF DISEASE IN INDIVIDUALS

WITH ATWI. In athletes with TWI beyond V1, infor-mation from the preceding Jt or ST-segment mayprovide valuable diagnostic information whenconsidering ARVC. Based on comparisons between 45athletes with ATWI and 35 patients with ARVC, wehave previously reported that a Jt and ST-segment inline with the onset of the QRS complex or a depressedST-segment preceding ATWI is a powerful discrimi-nator between the 2 entities (29). Moreover, a recentstudy examining ATWI as a marker of cardiomyopa-thy in a small cohort of athletes of black andwhite ethnicity showed that Jt elevation ($0.1 mV)preceding the TWI excluded ARVC (34). Our largestudy of almost 15,000 white individuals providesvalidation for these concepts in males but revealsthat Jt may be in line with the onset of the QRScomplex in as many as 50% of healthy females withATWI. Importantly, only 1 athlete demonstratedATWI with preceding ST-segment depression andnone of the individuals with ATWI showed Jtdepression suggesting that the presence of suchelectrical markers may be pointers for cardiacpathology.

There remains the possibility that ATWI confinedto V1 to V2 may be a manifestation of ARVC. We haveexamined our own cohort of 35 probands with ARVCand identified ATWI in V1 to V2 alone in 6%. All ofthese patients either expressed symptoms or otherelectrical features diagnostic of ARVC (29).

STUDY LIMITATIONS. This study was cross-sectionalin nature, and although there were no adverse clin-ical events in the ATWI group during a follow-up ofnearly 2 years, we cannot be certain whether ATWImay precede the development of ARVC by severalyears. Familial evaluation was not performed in anyof the individuals with ATWI because none fulfilledovert criteria for a cardiomyopathy. However, weconcede that such practice may have highlightedsome individuals with incomplete expression of dis-ease. A small proportion of ATWI individuals werelost to follow-up due to logistical difficulties thatcould not be overcome (e.g., emigration). CMRI is the

recognized gold standard for the investigation ofprimary cardiomyopathies but was only performed in250 (74%) individuals with ATWI. However, 81% of allindividuals with ATWI had all of ECG, echocardiog-raphy, SAECG, Holter, and exercise stress test, whichare sufficient to diagnose ARVC according to modifiedtask force criteria (19). Voluntary cardiac screeningprograms of nonathletes in the community conductedthrough organizations such as CRY do have a poten-tial for inherent selection bias though given the largenumbers included in this study of nearly 15,000 par-ticipants, the potential of any such bias is signifi-cantly mitigated.

CONCLUSIONS

ATWI is present in 2.3% of the young white popula-tion and is more common in women and in athletes.Almost 80% of ATWI is confined to leads V1 to V2 andhas a poor diagnostic yield for cardiac pathology,implying that this ECG pattern could be considered anormal phenomenon in asymptomatic individualswithout a family history of cardiomyopathy or pre-mature SCD. In contrast, TWI extending beyond V2 ispresent in only 1% of females and 0.2% in men andmay justify further evaluation in white individuals,particularly when preceded by Jt depression or ST-segment depression.

ACKNOWLEDGMENTS The authors thank the CRYorganization for providing the ECG machines andportable echocardiography facilities used for thestudy in the United Kingdom.

REPRINT REQUESTS AND CORRESPONDENCE:

Professor Sanjay Sharma, St. George’s University ofLondon, Cranmer Terrace, London SW17 0RE, UnitedKingdom. E-mail: sasharma@sgul.ac.uk.

J A C C V O L . 6 9 , N O . 1 , 2 0 1 7 Malhotra et al.J A N U A R Y 3 / 1 0 , 2 0 1 7 : 1 – 9 ATWI in Young White Athletes and Nonathletes

9

RE F E RENCE S

1. Schnell F, Riding N, O’Hanlon R, et al.Recognition and significance of pathologicalT-wave inversions in athletes. Circulation 2015;131:165–73.

2. Papadakis M, Basavarajaiah S, Rawlins J, et al.Prevalence and significance of T-wave inversionsin predominantly Caucasian adolescent athletes.Eur Heart J 2009;30:1728–35.

3. Migliore F, Zorzi A, Michieli P, et al. Prevalenceof cardiomyopathy in Italian asymptomatic chil-dren with electrocardiographic T-wave inversionat preparticipation screening. Circulation 2012;125:529–38.

4. Sharma S, Whyte G, Elliott P, et al. Electrocar-diographic changes in 1,000 highly trained juniorelite athletes. Br J Sports Med 1999;33:319–24.

5. Rawlins J, Carre F, Kervio G, et al. Ethnic dif-ferences in physiological cardiac adaptation tointense physical exercise in highly trained femaleathletes. Circulation 2010;121:1078–85.

6. Sheikh N, Papadakis M, Carre F, et al. Cardiacadaptation to exercise in adolescent athletes ofAfrican ethnicity: an emergent elite athletic pop-ulation. Br J Sports Med 2013;47:585–92.

7. Papadakis M, Carre F, Kervio G, et al. Theprevalence, distribution, and clinical outcomes ofelectrocardiographic repolarization patterns inmale athletes of African/Afro-Caribbean origin.Eur Heart J 2011;32:2304–13.

8. Corrado D, Pelliccia A, Heidbuchel H, et al.Recommendations for interpretation of 12-leadelectrocardiogram in the athlete. Eur Heart J2010;31:243–59.

9. Drezner J, Ackerman MJ, Anderson J, et al.Electrocardiographic interpretation in athletes: the“Seattle criteria.” Br J Sports Med 2013;47:122–4.

10. Pelliccia A, Culasso F, Di Paolo F, et al. Prev-alence of abnormal electrocardiograms in a large,unselected population undergoing pre-participation cardiovascular screening. Eur HeartJ 2007;28:2006–10.

11. Uberoi A, Stein R, Perez M, et al. Interpretationof the electrocardiogram of young athletes.Circulation 2011;124:746–57.

12. Brosnan M, La Gerche A, Kalman J, et al.Comparison of frequency of significant electro-cardiographic abnormalities in endurance versusnonendurance athletes. Am J Cardiol 2014;113:1567–73.

13. Corrado D, Basso C, Rizzoli G, Schiavon M,Thiene G. Does sports activity enhance the risk ofsudden death in adolescents and young adults?J Am Coll Cardiol 2003;42:1959–63.

14. Zaidi A, Ghani S, Sharma R, et al. Physiologicalright ventricular adaptation in elite athletes of

African and Afro-Caribbean origin. Circulation2013;127:1783–92.

15. Mitchell JH, Haskell W, Snell P, Van Camp SP.Task Force 8: classification of sports. J Am CollCardiol 2005;45:1364–7.

16. Friedmann H. Diagnostic Electrocardiographyand Vectorcardiography. New York, NY: McGraw-Hill, 1971.

17. Kligfield P, Gettes LS, Bailey JJ, et al. Recom-mendations for the standardization and interpretationof the electrocardiogram: part I: The electrocardio-gram and its technology: a scientific statement fromthe American Heart Association Electrocardiographyand Arrhythmias Committee, Council on Clinical Car-diology. Circulation 2007;115:1306–24.

18. Surawicz B, Knilans TK. Right bundlebranch block. In: Andjelkovic N, editor. Chou’sElectrocardiography in Clinical Practice. 6th ed.Philadelphia, PA: Elsevier Saunders, 2008:95–107.

19. Marcus FI, McKenna WJ, Sherrill D, et al. Diag-nosis of arrhythmogenic right ventricular cardio-myopathy/dysplasia: proposed modification of thetask force criteria. Circulation 2010;121:1533–41.

20. Macfarlane PW, Antzelevitch C, HaissaguerreM,et al. The early repolarization pattern: a consensuspaper. J Am Coll Cardiol 2015;66:470–7.

21. Lang RM, Bierig M, Devereux RB, et al. Rec-ommendations for chamber quantification: areport from the American Society of Echo-cardiography’s Guidelines and Standards Commit-tee and the Chamber Quantification Writing Group,developed in conjunction with the European As-sociation of Echocardiograph. J Am Soc Echo-cardiogr 2005;18:1440–63.

22. Quiñones M, Otto CM, Stoddard M, et al. Recom-mendations for quantification of Doppler echocardi-ography:A report fromtheDopplerquantification taskforceof thenomenclatureandstandards committeeofthe American Society of Echocardiography. J Am SocEchocardiogr 2002;15:167–84.

23. Nagueh SF, Appleton CP, Gillebert TC, et al.Recommendations for the evaluation of left ven-tricular diastolic function by echocardiography.J Am Soc Echocardiogr 2009;22:107–33.

24. Elliott PM, Anastasakis A, Borger M, et al. 2014ESC Guidelines on diagnosis and management ofhypertrophic cardiomyopathy: the Task Force forthe Diagnosis and Management of HypertrophicCardiomyopathy of the European Society of Cardi-ology (ESC). Eur Heart J 2014;35:2733–79.

25. Gersh BJ, Maron BJ, Bonow RO, et al.2011 ACCF/AHA guideline for the diagnosis andtreatment of hypertrophic cardiomyopathy: areport of the American College of CardiologyFoundation/American Heart Association TaskForce on Practice Guidelines. J Am Coll Cardiol2011;58:e212–60.

26. Lang RM, Badano LP, Mor-Avi V, et al. Rec-ommendations for cardiac chamber quantificationby echocardiography in adults: an update from theAmerican Society of Echocardiography and theEuropean Association of Cardiovascular Imaging.J Am Soc Echocardiogr 2015;28:1–39.e14.

27. Bruce RA. Exercise testing of patients withcoronary heart disease. Principles and normalstandards for evaluation. Ann Clin Res 1971;3:323–32.

28. Cain M, Anderson J, Arnsdorf M, et al. Signal-averaged electrocardiography. ACC expertconsensus document. J Am Coll Cardiol 1996;27:238–49.

29. Zaidi A, Sheikh N, Jongman JK, et al. Clinicaldifferentiation between physiological remodelingand arrhythmogenic right ventricular cardiomy-opathy in athletes with marked electrocardio-graphic repolarization anomalies. J Am Coll Cardiol2015;65:2702–11.

30. Grothues F, Moon JC, Bellenger N, Smith GS,Klein HU, Pennell DJ. Interstudy reproducibility ofright ventricular volumes, function, and mass withcardiovascular magnetic resonance. Am Heart J2004;147:218–23.

31. Bauce B, Frigo G, Benini G, et al. Differencesand similarities between arrhythmogenic rightventricular cardiomyopathy and athlete’s heartadaptations. Br J Sports Med 2010;44:148–54.

32. Biffi A, Verdile L, Ansalone G, et al. Lack ofcorrelation between ventricular late potentialsand left ventricular mass in top-level male ath-letes. Med Sci Sports Exerc 1999;31:359–61.

33. Pelliccia A, Culasso F, Di Paolo FM, Maron BJ.Physiologic left ventricular cavity dilatation in eliteathletes. Ann Intern Med 1999;130:23–31.

34. Calore C, Zorzi A, Sheikh N, et al. Electrocar-diographic anterior T-wave inversion in athletes ofdifferent ethnicities: differential diagnosis be-tween athlete’s heart and cardiomyopathy. EurHeart J 2016;37:2515–27.

35. Suarez RM, Suarez RM Jr. The T-wave of theprecordial electrocardiogram at different agelevels. Am Hear J 1946;32:480–93.

36. Gottschalk C, Craige E. A comparison of theprecordial S-T and T waves in the electrocardio-grams of 600 healthy young Negro and whiteadults. South Med J 1956;49:453–7.

37. Sheikh N, Papadakis M, Ghani S, et al. Com-parison of ECG criteria for the detection of cardiacabnormalities in elite black and white athletes.Circulation 2014;129:1637–49.

KEY WORDS anterior T-wave inversion,arrhythmogenic right ventricularcardiomyopathy, ECG screening, ethnicity