Access to the PDF text

Free Article !

Archives of cardiovascular diseases
Volume 111, n° 5
pages 380-388 (mai 2018)
Doi : 10.1016/j.acvd.2017.10.005
Received : 14 April 2017 ;  accepted : 12 October 2017
Clinical research

Electrocardiographic patterns and long-term training-induced time changes in 2484 elite football players
Variations physiologiques et évolution de l’électrocardiogramme chez 2484 footballeurs de haut niveau

Olivier Huttin a, , Christine Selton-Suty a, Clément Venner a, Jean-Baptiste Vilain b, Pierre Rochecongar c, d, Etienne Aliot a
a Service de cardiologie, institut Lorrain du cœur et des vaisseaux Louis-Mathieu, centre hospitalier universitaire de Nancy, 4, rue du Morvan, 54500 Vandœuvre-lès-Nancy, France 
b Veltys Data analysis and statistical analysis, 75009 Paris, France 
c Department of biology and sports medicine, hôpital Pontchaillou, CHU de Rennes, 35033 Rennes, France 
d French Football Federation–Professional Football League, 75738 Paris, France 

Corresponding author.

High-level physical training induces cardiac structural and functional changes, including 12-lead electrocardiogram modifications.


The purpose of this cross-sectional longitudinal study was to establish a quantitative electrocardiographic profile in highly trained football players. Initial and serial annual electrocardiogram monitoring over subsequent years allowed us to investigate the long-term effects of exercise on cardiac conduction and electrophysiological remodelling.


Between 2005 and 2015, serial evaluations, including 12-lead electrocardiograms, were performed in 2484 elite male football players from the French Professional Football League. A total of 6247 electrocardiograms were performed (mean 2.5±1.8 electrocardiograms/player). Heart rate (beats/min), atrioventricular delay (PR, ms), intraventricular conduction delay (QRS, ms), corrected QT delay (QTc) and electrical left ventricular hypertrophy (LVH) (Sokolow-Lyon index, mm) were measured, and the fixed effect of time was evaluated using panel data analysis (β [95% confidence interval] change between two visits).


According to European Society of Cardiology and Seattle criteria, 15% of the electrocardiogram intervals were considered abnormal. We observed 17% sinus bradycardia<50 beats/min (mean heart rate 60±11 beats/min), 8% first-degree atrioventricular block>200ms (mean PR 170±27ms), 1.5% QRS>120ms (mean QRS 87±19ms) and 3% prolonged QT interval (mean QTc using Bazett's formula [QTcB] 395±42ms). Electrical LVH (mean Sokolow-Lyon index 34±10mm) was noted in 37% of players. Over time, electrocardiogram changes were noted, with a significant remodelling trend in terms of decreased heart rate (−0.41 [−0.55 to −0.26] beats/min), QRS duration (−2.4 [−2.7 to −2.1] ms) and QTcB delay (−1.2 [−1.9 to −0.5] ms) (all P <0.001).


This study describes usual electrocardiographic training-induced changes in a large series of football players over the follow-up timeframe. The most frequent outliers were electrical LVH and sinus bradycardia. These results have important implications for optimizing electrocardiogram interval measurements in initial screening and during follow-up of football players, with potential cost-effective implications.

The full text of this article is available in PDF format.

L’activité physique de haut niveau entraîne des modifications structurelles et fonctionnelles qui vont se traduire par des modifications de l’électrocardiogramme de surface. L’objectif de notre travail est d’évaluer l’évolution au cours du temps du profil électrocardiographique de footballeurs professionnels.


Nous avons analysé les aspects électrocardiographiques et leurs modifications au cours du temps sur un total de 6247 ECG collectés chez 2484 footballeurs de haut niveau (2,5±1,8 ECG par joueur). La fréquence cardiaque (FC, Bpm), le délai auriculo-ventriculaire (PR, ms), intra-ventriculaire (QRS, ms) et le QT corrigé (QTc) avec mesure de l’hypertrophie ventriculaire gauche électrique (HVG, Sokolow-Lyon, mm) ont été mesurées avec analyse de leurs variations au cours du temps.


Selon les critères de l’ESC et de Seattle, 15 % des intervalles ECGS sont considérés comme anormaux. La FC moyenne était de 60±11bpm avec 17 % de bradycardie sinusale (FC<50bpm). Le PR moyen était de 170±27ms avec 8 % de BAV du premier degré, le QRS moyen était de 87±19ms avec 1,5 % de QRS>120ms, et le QTc moyen était de 395±42ms avec 3 % de QTc>450ms. Le Sokolow moyen était de 33±9mm,>45mm chez 37 % des joueurs. Au cours du temps, des variations significatives de l’ECG sont notées en termes de diminution de la fréquence cardiaque (−0,41bpm [−0,55 ; −0,26]), largeur de QRS (−2,4ms [−2,7 ; −2,1]) et de QTc (−1,2ms [−1,9 ;−0,5]) (tous les p <0,001).


Cette étude décrit l’aspect électrocardiographique normal et les modifications à long terme induites par l’entraînement physique chez une large population de footballeurs professionnels. Ces résultats ont des implications importantes à la fois dans la détection d’anomalies électriques de conduction et de repolarisation lors du screening initial de ces professionnels mais aussi dans la prise en charge du suivi de ces sportifs.

The full text of this article is available in PDF format.

Keywords : Electrocardiogram, Football players, Athlete's heart

Mots clés : ECG, Joueurs de foot, Cœur d’athlète

Abbreviations : AV, CI, ESC, IVCD, LBBB, LV, LVH, QTc, QTcB, QTm, RBBB


A 12-lead electrocardiogram is an easy and reproducible screening method that significantly improves the sensitivity of the detection of serious cardiac diseases in athletes, including athletic heart syndrome. Accurate interpretation of an athlete's electrocardiogram time intervals is crucial to minimize the risk of unnecessary investigations based on electrical phenotype, which are mandatory to rule out primary cardiomyopathy. The prevalence of abnormal electrocardiogram findings is known to be related to age, sex, ethnicity, level of activity and sporting discipline [1]. International teams of experts have established the Seattle and European Society of Cardiology (ESC) criteria for the interpretation of electrocardiograms and the definition of outliers [2, 3]. However, these recommendations arose from studies involving heterogeneous athlete populations or simply used ranges defined in the general population [4]. Moreover, most electrocardiogram-based studies focused on repolarization, including ST-segment patterns and T-wave morphologies. Yet participation in regular intensive exercise is also associated with conduction and repolarization delays, the physiopathology of which remains unclear.

Long-term variations in atrioventricular (AV) and intraventricular conductions, including repolarization duration, are not specifically assessed, and may discriminate normal from pathological remodelling. Prospective longitudinal studies evaluating specific training-induced changes have focused on cardiac function and stature, but fell short on electrocardiogram data and long-term follow-up [5]. Serial electrocardiogram evaluations with longitudinal follow-up allow the repetition of time-interval and voltage measurements, and facilitate the assessment of changes throughout a player's career.

In this large nationwide multicentre population of elite football players, we aimed to define the normal ranges of standard measures, such as PR, QRS, QT intervals and electrocardiogram left ventricular hypertrophy (LVH); and compare serial measurements over time, to evaluate long-term electrocardiographic adaptations that may occur during a player's career.

Participating subjects

All football players participating in the French Professional Football League underwent a yearly medical examination with cardiovascular evaluation, including an electrocardiogram recording and echocardiography. Evaluations included comprehensive family history, presence of cardiac symptoms and physical examination, and all study subjects were judged to be free of structural heart disease (normal echocardiography) and participated in national competitions.

The present study analysed 6433 electrocardiograms from 2484 football players aged 16–40 years, collected through the French Football Federation (Fédération Française de Football, Paris, France) between 2005 and 2015. All data were collected and processed using an electronic online case report form (Almerys©, Paris, France). All participants gave written informed consent, and the study was approved by the board of the French Football Federation and its ethical committee.

Electrocardiogram recording

Twelve-lead electrocardiograms were recorded at 25mm/s with an amplification of 0.1mV/mm in the supine position during quiet respiration, and were interpreted by trained cardiologists. A well-defined protocol for electrode placement was used throughout the study by a physiologist. Electrocardiograms were interpreted using the ESC recommendations [2] and the Seattle criteria [3]. For this study, only quantitative (intervals and voltage) variables were considered. Electrocardiograms were recorded and analysed in each centre, following international guidelines. The automated measures recorded were considered and validated by the cardiologist, with a blind reading for serial evaluation.

Continuous electrocardiogram interval analysis

The RR, PR and QRS intervals were measured in lead II to the nearest 2ms from the average. Heart rate was calculated using the 1/RR interval. The QT interval was measured in lead II or V5 (whichever provided best delineation of the T wave). The highest value was quoted as measured QT interval (QTm). Using the preceding RR interval, the corrected QT interval (QTc) was obtained by Bazett's formula (QTcB=QTm/√RR) and Fridericia's formula (QTcF=QTm/√RR3). The R- and S-wave amplitudes were measured to the nearest μV as the mean of the highest amplitudes of QRS complexes; Sokolow-Lyon voltage criteria were used to define LVH (S wave in V1+R wave in V5 or V6, whichever was greater).

Categorical electrocardiogram interval analysis

Based on the normal range for an electrocardiogram, and according to the cut-off value of defined by the ESC and Seattle criteria, we defined the percentage of abnormal interval duration. Sinus bradycardia was defined as heart rate < 50 beats/min, and was significant at<40 beats/min. First-degree AV block was defined as PR interval>200ms, and was significant at>240ms. Short PR was defined as a PR interval<120ms, with or without identification of a delta wave. Intraventricular conduction delay (IVCD) was defined as incomplete if the QRS duration was measured between 110–120ms, and as complete if>120ms. Complete left bundle branch block (LBBB) morphology was defined as a predominantly negative QRS complex in lead V1 (QS or rS pattern) and an upright monophasic R wave in leads I and V6. Right bundle branch block (RBBB) morphology was considered in the presence of rsR’ in V1  or was defined as rsR’ pattern if the QRS duration was<110ms. A prolonged QRS duration (>110ms) not fulfilling the criteria for either LBBB or RBBB was referred to as a non-specific intraventricular conduction disturbance [6]. Sokolow-Lyon voltage criteria were used to define electrical LVH (SV1 +RV5 35mm), which was considered severe if>45mm. QTc was considered to be elongated if QTcB was>440ms (ESC criteria) or>470ms (Seattle criteria). QTc was considered to be short if QTcB was<340ms (ESC criteria) or320ms (Seattle criteria).

Further evaluation and follow-up

Follow-up started from the date of the initial screening, and lasted until October 2015. Overall, 1495 (60%) athletes (5258 electrocardiograms) underwent repeated clinical and electrocardiographic evaluations. Hence, an electrocardiogram follow-up (median, years) was available for some players: 534 subjects had two electrocardiograms; 371 had three electrocardiograms; 238 had four electrocardiograms; and 146 had five electrocardiograms (Figure 1). All athletes were alive and in good physical shape as of October 2015.

Figure 1

Figure 1. 

Distribution and time information regarding the number of electrocardiograms per visit.


Statistical analysis

All statistical analyses were conducted using Stata® software, version 14 (StataCorp, College Station, TX, USA). Descriptive analysis of electrocardiogram continuous variables (RR, PR, QRS and QT duration and QRS voltage) were presented as analyses of central tendency (means±standard deviations, medians [interquartile ranges]) and as categorical analyses. Categorical data were defined as numbers and percentages of patients meeting or exceeding the predefined upper limit values. As the data were unbalanced (some players had more electrocardiograms than others), we only kept one electrocardiogram per player in our descriptive analysis. The electrocardiogram that was retained for each player was the most “complete” electrocardiogram in terms of data.

For the longitudinal evaluation, we first presented descriptive statistics of players who underwent at least four electrocardiograms (subsample of 590 players, 23.8%). We compared means and standard deviations of electrocardiogram continuous variables for these players over the four visits. In a second step, we use a fixed-effects model that is better suited to panel data analysis (subsample of 1495 players with at least two electrocardiograms; 5258 electrocardiograms). With fixed effects, all players with at least two electrocardiograms were part of the sample, and the individual variability was accounted for through the fixed effects. For QT, PR and QRS, the RR interval can be an important determinant of the variations from one visit to the next. Therefore, we introduced the RR interval as a control variable. We reported β values of QT, PR and QRS measurements, along with their standard errors, P values and 95% confidence intervals (CIs).

We selected 521 digitally recorded electrocardiograms from routine visits in 2014 and 2015 (n =521 players). Intraclass correlation coefficients were used to quantify both inter-rater and intrarater assessments of the five electrocardiogram measures (Appendix A).


Quantitative and qualitative analyses of electrograms are shown in Table 1, Table 2, respectively. At least one training-related electrocardiogram anomaly, according to the specified quantitative Seattle electrocardiogram criteria, was noted in 345 players (14%). All subjects had sinus node rhythm, and no rhythm disturbance. On average, the athletes had been followed and trained for 2.0±2.4 years (Figure 1).

Heart rate

The mean heart rate for the overall population was 59±11 beats/min. We observed 54% sinus bradycardia<60 beats/min, 17%<50 beats/min and 2%<40 beats/min (Figure 2). Over time, we observed a slight, but significant, change in heart rate, with a decrease of −0.41 (95% CI −0.55 to −0.26) beats/min (P <0.001) per visit consistently, with a mean increase in the RR delay of 6ms (95% CI 3.6 to 8.5) per visit (P <0.001). Introducing the RR interval as a control variable in the panel data model did not affect QTm, PR or QRS changes during follow-up visits (Table 3).

Figure 2

Figure 2. 

Distribution of heart rate, PR and QRS delay and Sokolow-Lyon index for the overall population (2484 football players).


AV conduction defects

Mean PR duration was 169±28ms. First-degree AV block (PR>200ms) was observed in 8% of the representative sample, with a significant increase (PR>240ms) in 2% and a short PR interval (PR<120ms) in 2%. There was no significant change in AV conduction at the successive visits (β=−0.0004, 95% CI −1.53 to 1.53; not significant).

Intraventricular conduction defects

Mean QRS duration was 87±18ms. We observed 1.5% of athletes with an IVCD>120ms (RBBB pattern in 0.5% and non-specific in 1.0%). The high prevalence of incomplete IVCD was mainly driven by incomplete RBBB (3%). There was no LBBB. Over time, we observed a significant decrease in QRS interval of −2.4ms (95% CI −2.7 to −2.1) per visit (P <0.001).

QT interval

The mean QT interval was measured at 400±33ms. After correction for heart rate, the mean QTcB was 395±41ms and the mean QTcF was 396±34ms. Significant prolonged QT interval was observed in 3% using Bazett's formula (QTcB>470ms) and in 1% using Fridericia's correction (QTcF>480ms) (Figure 3). Longitudinal analysis of QTcB was consistent, with a shortened delay of −1.2ms (95% CI −1.9 to −0.5) per visit (P <0.001).

Figure 3

Figure 3. 

Distribution of measured QT interval (QTm) and corrected QT interval (QTc) obtained by Bazett's formula (QTcB=QTm/√RR) and Fridericia's formula (QTcF=QTm/√RR3) for the overall population (2484 football players).


Electrical LVH

The Sokolow-Lyon voltage criteria for electrical LVH were present in 37% athletes, and 11% showed importantly increased values>45mm. A significant, but not clinically relevant, decrease in the Sokolow-Lyon index was observed during follow-up (β=−0.31mm, 95% CI −0.52 to −0.09; P =0.005).


Our study indicates that, similar to endurance athletes, elite football players frequently have conduction and repolarization delays; the abnormalities included not only voltage changes, but also a relatively high prevalence of IVCD. In addition, we confirmed the lack of specificity of the current guidelines regarding an appropriate definition of normal training-related electrocardiogram changes, especially for QTc duration. Finally, over time, follow-up of electrocardiograms revealed significant electrical remodelling, with progressive shortening of QRS and QT intervals.

Bradycardia and AV conduction delay

In highly trained athletes, resting sinus bradycardia is a common finding, resulting from an adaptive change to the autonomic nervous system. In healthy adults, but also in athletes, heart rate values<60 beats/min are considered as sinus bradycardia, which was observed in 58% of our population. Depending on the type of sport activity and the level of training, the heart rate drops lower in endurance sports, such as long-distance running. Extreme values of heart rate<40 beats/min were punctually noted, in the absence of symptoms such as fatigue, dizziness or syncope. Therefore, sinus bradycardia>30 beats/min could be considered normal in a highly trained athlete.

Combined AV conduction delay is also common, and is caused by increased vagal tone. First-degree AV block is considered as a common training-related electrocardiogram change in the ESC classification (group 1) [2]. First-degree AV block is defined as a prolonged PR interval>200ms, but relies on a heterogeneous condition depending on the duration of the block. Severity of AV block is focused on its association with IVCD, but not with PR duration, especially highly prolonged PR. Overall, 2% of players had marked PR interval values>240ms, while a lengthy PR interval is observed in only 0.2% of the general population. This delay in AV nodal conduction in athletes may be the consequence of increased vagal activity or some degree of intrinsic AV node changes, but tends to shorten with increasing heart rate during practice. An association between prolonged AV conduction (PR>220ms) and LVH has also been suggested.

Intraventricular conduction delay

Complete bundle branch block is rare in the general population, particularly in younger groups (aged<40 years), occurring in 0.6% of males [2]. RBBB and LBBB in athletes must be considered as being unrelated to training (group 2; ESC), and warrants further clinical investigation, as it can be considered an indicator of possible structural heart disease [2].

No player had an LBBB, and only 0.5% had a complete RBBB, but 1% had non-specific IVCD (i.e. 1.5% of players had QRS>120ms). An incidence of complete LBBB or RBBB of 0.4% was reported by Pelliccia et al. in a large athletic population of highly trained competitors [4].

RBBB is supposed to be related to structural RV remodelling, characterized by ventricular dilation, a reduction in systolic function at rest and interventricular dyssynchrony [7]. Complete RBBB is uncommon in athletes, and is a potential marker of serious underlying cardiovascular disease [3]. The prevalence of incomplete RBBB has been estimated as being higher (35–50%) in highly trained athletes, particularly those engaged in endurance training and mixed-sport disciplines [4]. It has been suggested that the RV conduction delay is not an intrinsic delay within the His-Purkinje system itself, but is caused by the enlarged RV cavity and the resultant increased conduction time [8].

QRS duration tends to depend upon required cardiovascular work, and thus upon the type of sport, but also, in sports such as football, on the player's position on the field [9]. A previous study noted that athletes with greater left ventricular (LV) mass had a slightly longer QRS duration, and that the QRS duration was shorter in the African/American subgroup compared with in the white/other subgroup [10]. It is possible to suggest a correlation between QRS duration and LV mass and ethnicity [10]. Unfortunately, such data could not be obtained in our study for confidentiality reasons.

QRS voltage criteria for LVH

In athletes, intensive conditioning is associated with morphological cardiac changes of increased cavity dimensions and wall thickness, which are reflected on the electrocardiogram. The most commonly used and recommended voltage criterion for LVH is the Sokolow-Lyon index, which is associated with LV remodelling (cavity dimensions and wall thickness). Prevalence of isolated increased QRS voltage in athletes can be as high as 45%. However, isolated increased QRS voltage in the absence of another anomaly is not a reliable indicator of LVH, and does not mandate further investigation [2]. Several limits to LVH interpretation must be kept in mind (placement of electrodes, sex and ethnicity), and it should be interpreted in light of the athlete's discipline (increased for endurance sports) [11].

Delayed repolarization and QT interval

The QT interval is longer in athletes than in non-athletic controls, because of the lower heart rate at rest compared with a non-athletic population, while QTc remains within normal range, yet close to the upper limit [12].

A prolonged QTc value should raise concerns regarding either acquired or congenital long QT syndrome, but several limitations should be considered: heart rate variability; sinus bradycardia (<50 beats/min); prolonged QRS duration; LVH; T-wave morphology (flat, biphasic or bifid); and the presence of a U wave. Bazett's formula is currently the method of choice, according to the ESC and Seattle recommendations. However, to improve screening, a more appropriate method and correction for this interval estimation have to be considered [13]. Measurement of QTc using Fridericia's formula shows less dependence towards heart rate, and might be preferred in such cases [14]. In our cohort, 6–15% of the athletes presented with prolonged QTcB according to the Seattle and ESC recommendations, while only 3–5% met the criteria using Fridericia's formula, and were subsequently reclassified as normal. Among different correction factors, Pickham et al. demonstrated that Fridericia's had the lowest residual dependence on heart rate [14]. Up to 75% of athletes with an uncorrected QT interval>99% of the population are in the normal range after Fridericia's correction. QT was evaluated in professional soccer players, and it was shown that different degrees of cardiac hypertrophy and changes in heart rate and ST repolarization change may influence QT duration and have important confounding effects.

On the other hand, short QT syndrome in athletes is a debated condition that merits careful consideration. Shortening of repolarization accounts for short myocardial refractoriness, which predisposes to life-threatening ventricular arrhythmias. According to measurements and lower limits of QT duration, between 2% and 5% are affected.

Prevalence of electrocardiogram abnormalities, and change over the time

As the debate over the sensitivity of contemporary electrocardiogram interpretation guidelines and the cost-effectiveness of annual monitoring rages, the detecting and refining of criteria for interval changes, without compromising sensitivity to detect serious cardiac disease, is most welcome for cardiac screening of athletes.

Cardiac remodelling supposedly increases with intensive and prolonged athletic conditioning, and may be reflected on a 12-lead electrocardiogram. The need for improved specificity of electrocardiogram criteria has been reported previously in highly trained athletes. This was underlined by a reduction of 29–11% in electrocardiogram abnormalities in soccer players when the ESC recommendations were exchanged for the new Seattle criteria.

Occupational or recreational exercise reduces mortality from cardiovascular disease in a global population. But the long-term effect of high-level training implies potential remodelling mechanisms with unevaluated long-term electrocardiogram changes. During exercise, the response related to heart rate increase is shortening of the PR, QRS and QT intervals. Increased vagal modulation of heart rate after exercise training is sufficient to achieve a modest hypotensive effect from decreasing vascular resistance. However, more prolonged and intense training does not necessarily lead to greater enhancement of circulatory control and, therefore, may not provide an added cardiovascular protective benefit via autonomic mechanisms.

We have demonstrated, for the first time, electrical remodelling based on 12-lead electrocardiogram analysis. A long-term decrease in RR interval is associated with altered action potential duration, conduction velocity and contractile velocity, resulting in electrocardiogram changes. In our longitudinal follow-up, the QRS complex duration shortened progressively, with a mean decrease of 2.4ms per visit. For instance, prolongation of the QRS complex and new onset of IVCD occur with cardiovascular aging or are associated with myocardial disease. There are no data in the recent literature on shortening of QRS complex duration in the setting of long-term training. One possible explanation for the apparently accelerated conduction may be an increase in the number of Purkinje fibres or the degree of penetration of these fibres across the ventricular wall. Short-term beat-to-beat variability in the QT interval was shown to be more preponderant in athletes compared with age-matched non-athletes [15]. Regarding change in repolarization delay, we also suggest that the decrease in QT duration could be linked with intrinsic shortening of QRS duration. This long-term longitudinal study with evaluation of chronic electrocardiogram training-induced changes gives a first insight into the impact of training on the regulation of electrocardiogram time intervals, which may help our understanding of individual susceptibility to ventricular arrhythmia.

Study limitations

The systematic strategy of investigating a longitudinal cohort with a descriptive approach is useful as an exploratory step, but has two limitations. The number of observations is restricted to players with a small number of electrocardiograms. Also, a potential concern with the interpretation of electrocardiogram changes over time is reproducibility, as intrareader and inter-reader reproducibility might be limited. Therefore, players compared over the four periods showed disparity. In order to deal with these two issues, we use a fixed-effects model that is better suited to panel data analysis. A relatively high number of subjects had missing values for PR interval (first visits), precluding results of AV conduction changes overtime. Finally, data relating to age or ethnicity could not be obtained for confidentiality reasons, precluding subgroup analysis.


Comprehensive analyses of electrocardiogram conduction delay and amplitudes allow the characterization of abnormal electrocardiogram training-induced changes. This original large cross-sectional and longitudinal evaluation illustrates the need to adopt a global approach regarding the definition of athletes’ normal and abnormal electrocardiogram findings. Our study provided specific insights into electrocardiogram profiles in a large and representative population of elite football players. These findings have important implications for preparticipation screening of young athletes and their subsequent follow-up. Despite limited variations in baseline 12-lead electrocardiograms, electrical remodelling in an athlete's heart justifies annual monitoring to detect time interval changes.

Sources of funding


Disclosure of interest

The authors declare that they have no competing interest.


We would like to acknowledge: the French Football Federation (Professional Football League); all the cardiologists of the football clubs involved in screening and electrocardiogram and echocardiogram data collection; and Almerys©, Paris, France.

Appendix A. Supplementary data

(13 Ko)

Magalski A., Maron B.J., Main M.L., and al. Relation of race to electrocardiographic patterns in elite American football players J Am Coll Cardiol 2008 ;  51 : 2250-2255 [cross-ref]
Corrado D., Pelliccia A., Heidbuchel H., and al. Recommendations for interpretation of 12-lead electrocardiogram in the athlete Eur Heart J 2010 ;  31 : 243-259 [cross-ref]
Drezner J.A., Ackerman M.J., Anderson J., and al. Electrocardiographic interpretation in athletes: the ‘Seattle criteria’ Br J Sports Med 2013 ;  47 : 122-124 [cross-ref]
Pelliccia A., Maron B.J., Culasso F., and al. Clinical significance of abnormal electrocardiographic patterns in trained athletes Circulation 2000 ;  102 : 278-284 [cross-ref]
Baggish A.L., Wang F., Weiner R.B., and al. Training-specific changes in cardiac structure and function: a prospective and longitudinal assessment of competitive athletes J Appl Physiol (1985) 2008 ;  104 : 1121-1128 [cross-ref]
Surawicz B., Childers R., Deal B.J., and al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part III: intraventricular conduction disturbances: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology Circulation 2009 ;  119 : e235-e240
Kim J.H., Noseworthy P.A., McCarty D., and al. Significance of electrocardiographic right bundle branch block in trained athletes Am J Cardiol 2011 ;  107 : 1083-1089 [inter-ref]
Kim J.H., Baggish A.L. Electrocardiographic right and left bundle branch block patterns in athletes: prevalence, pathology, and clinical significance J Electrocardiol 2015 ;  48 : 380-384 [cross-ref]
Fagard R. Athlete's heart Circulation 2001 ;  103 : E28-E29
Uberoi A., Sadik J., Lipinski M.J., Van Le V., Froelicher V. Association between cardiac dimensions and athlete lineup position: analysis using echocardiography in NCAA football team players Phys Sports Med 2013 ;  41 : 58-66 [cross-ref]
Hancock E.W., Deal B.J., Mirvis D.M., and al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part V: electrocardiogram changes associated with cardiac chamber hypertrophy: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Endorsed by the International Society for Computerized Electrocardiology J Am Coll Cardiol 2009 ;  53 : 992-1002 [cross-ref]
Kapetanopoulos A., Kluger J., Maron B.J., Thompson P.D. The congenital long QT syndrome and implications for young athletes Med Sci Sports Exerc 2006 ;  38 : 816-825 [cross-ref]
Chiladakis J., Kalogeropoulos A., Arvanitis P., Koutsogiannis N., Zagli F., Alexopoulos D. Heart rate-dependence of QTc intervals assessed by different correction methods in patients with normal or prolonged repolarization Pacing Clin Electrophysiol 2010 ;  33 : 553-560
Pickham D., Hsu D., Soofi M., and al. Optimizing QT interval measurement for the preparticipation screening of young athletes Med Sci Sports Exerc 2016 ;  48 : 1745-1750 [cross-ref]
Corrado D., Basso C., Schiavon M., Thiene G. Screening for hypertrophic cardiomyopathy in young athletes N Engl J Med 1998 ;  339 : 364-369 [cross-ref]

© 2017  Elsevier Masson SAS. All Rights Reserved.
EM-CONSULTE.COM is registrered at the CNIL, déclaration n° 1286925.
As per the Law relating to information storage and personal integrity, you have the right to oppose (art 26 of that law), access (art 34 of that law) and rectify (art 36 of that law) your personal data. You may thus request that your data, should it be inaccurate, incomplete, unclear, outdated, not be used or stored, be corrected, clarified, updated or deleted.
Personal information regarding our website's visitors, including their identity, is confidential.
The owners of this website hereby guarantee to respect the legal confidentiality conditions, applicable in France, and not to disclose this data to third parties.
Article Outline