27 July 2021
AperTO - Archivio Istituzionale Open Access dell'Università di Torino
Original Citation:
Electrocardiograms of Children and Adolescents Practicing Non-competitive Sports: Normal Limits and Abnormal Findings in a Large European Cohort Evaluated by Telecardiology
Published version:
DOI:10.1007/s40279-016-0609-7
Terms of use:
Open Access
(Article begins on next page)
Anyone can freely access the full text of works made available as "Open Access". Works made available under a Creative Commons license can be used according to the terms and conditions of said license. Use of all other works requires consent of the right holder (author or publisher) if not exempted from copyright protection by the applicable law.
Availability:
This is the author's manuscript
This full text was downloaded from iris - AperTO: https://iris.unito.it/
iris - AperTO
University of Turin’s Institutional Research Information System and Open Access Institutional Repository
This is the author's final version of the contribution published as:
Molinari, Giuseppe; Brunetti, Natale Daniele; Biasco, Luigi; Squarcia,
Sandro; Cristoforetti, Yvonne; Bennicelli, Riccardo; Del Vecchio, Cecilia;
Viacava, Cecilia; Giustetto, Carla; Gaita, Fiorenzo. Electrocardiograms of
Children and Adolescents Practicing Non-competitive Sports: Normal Limits
and Abnormal Findings in a Large European Cohort Evaluated by
Telecardiology. SPORTS MEDICINE. None pp: 1-9.
DOI: 10.1007/s40279-016-0609-7
The publisher's version is available at:
http://link.springer.com/content/pdf/10.1007/s40279-016-0609-7
When citing, please refer to the published version.
Link to this full text:
Electrocardiograms of Children and Adolescents Practicing Non-competitive Sports: Normal Limits and Abnormal Findings in a Large European Cohort Evaluated by Telecardiology
Giuseppe Molinari, Natale Daniele Brunetti, Luigi Biasco, Sandro Squarcia, Yvonne Cristoforetti, Riccardo Bennicelli, Cecilia Del Vecchio, Cecilia Viacava, Carla Giustetto, Fiorenzo Gaita
____________
Abstract Objective
The objective of this study was to derive normal electrocardiographic values and to report the abnormal findings in a large contemporary European cohort of physically active children and young adolescents.
Methods
In a 3-month period, data derived from subjects aged between 3 and 14 years and referred to the Telecardiology Centre (Genoa, Italy) for electrocardiogram (ECG) evaluation as pre-participation screening for non-competitive sports were analyzed.
Results
A total of 2060 ECGs were recorded. Of those, 1962 did not show any morphological abnormality and were used to derive normality ranges for heart rate, PR interval, QRS duration, corrected QT interval, and voltage of R wave as measured in V1 according to age and sex.
Findings and clinical implications of the 98 ECGs with abnormal findings were also reported. Abnormal ECG findings were not as uncommon as expected in this population, being manifest in about 5 % of subjects. However, major ECG anomalies (diffuse negative T-waves, preexcitation) were present in just ten subjects (0.5 %). Lower mean heart rate values (from 90–100 bpm at 3 years of age to 80–85 bpm at 14 years of age) and lower rates of the prevalence of negative T-waves in the V3 lead (from 55–60 % at 3 years of age to 8–10 % at 14 years of age) were observed with increasing age.
Conclusions
This is the first work reporting derived normal limits and abnormal ECG findings in a large contemporary European cohort of children and adolescents aged 3–14 years practicing non-competitive sports. Clear pathological alterations are extremely uncommon, deserving, when encountered, additional examinations. Even in a physically active population, the common features of an adult athlete’s ECG are absent.
__________________
Background
Interpretation of electrocardiograms (ECGs) in infants and children may be challenging owing to dynamic changes mirroring the gradual transition of right-ventricular preponderance to the left-ventricular dominance. An additional major confounder is the possible presence of exercise-related changes in those subjects practicing noncompetitive sports, leading to a combination that may puzzle the reporting physician. Traditional normal limits for ECG interpretation, derived several years ago from Canadian and Dutch cohorts, when the practice of sport among children was not so
widespread as today, did not evaluate the impact of physical activity and its influence on exercise-related ECG changes [1, 2]. Thus, given the need to provide physicians with an up-to-date reference in the common situation of interpreting ECGs of children practicing non-competitive sports, we aimed to derive normal electrocardiographic values and to report the abnormal findings in a large contemporary European cohort of physically active children and young adolescents.
2 Methods
2.1 Study Population
Data derived from subjects aged between 3 and 14 years referred for ECG analysis as a pre-participation screening for non-competitive sports at the ‘‘Telemedico’’ Telecardiology Centre (Genoa, Italy) between 1 September and 30 November, 2014 were digitally stored and analyzed. ECGs were recorded and sent by a network of ‘spoke’ centers located all over Italy (Fig. 1); a ‘hub’ center, located in Genoa, received and reported back on ECGs. The age of the subjects enrolled in the study (3–14 years) was chosen based on the expected range for compulsory education in Italy. For all subjects, besides registry data, a detailed clinical and pharmacological history was obtained by the referring center through an electronic form. Only good-quality, artifact-free recordings of subjects with complete clinical data and without known congenital or acquired heart disease, not receiving long-term therapy or medical treatment in the 48 h preceding the exam, were used in this study. The study was performed in accordance with the Declaration of Helsinki [3] and approved by an ethical committee.
2.2 ECG Recording and Transmission
For each subject, a 12-lead ECG was recorded using Cardiette Microtel 1 (Cardioline SpA, Trento, IT) or Cardiette Microtel 2 at a sampling rate of 500 Hz and with a frequency response range of 0.05–150 Hz. ECGs recorded by trained healthcare professionals (general practitioners,
pediatricians, sport physicians, nurses, or pharmacists) in spoke centers widespread all over Italy were digitally sent to our center for quality assessment and interpretation. ECGs were interpreted by one or more cardiologists, trained in ECG interpretation and available 24 h/7 days a week, as
previously reported [4, 5].
2.3 ECG Measurements
The digitally stored ECGs were processed using a wellvalidated computer tool (Cardioline Record; IT Medical Devices SpA, Milano, IT) and manually analyzed. All ECGs were analyzed by two independent reviewers. In ambiguous cases, adjudication was achieved by consensus with a third reviewer. For all voltage measurements, the isoelectric line was considered to be the PR segment. Heart rate was calculated as a mean considering the frequent physiological sinus arrhythmia; PR interval, QRS duration, and QT interval were measured in lead II. Correction of the QT interval was obtained using Bazett’s formula (QTc = QT/HRR). For QT measurement, the tangent method was used. Left-atrial enlargement, firstand second-degree type I atrioventricular block, left-anterior fascicular block, left-posterior fascicular block, complete/ incomplete, and right and left bundle block were defined according to the American Heart Association/ American College of Chest Physicians/Heart Rhythm Society recommendations [6–8]. The ECG was deemed consistent with early repolarization if a J-point elevation C1 mm associated with a ST segment elevation was evident in two or more contiguous leads [9]. The presence of abnormal ECG findings according to the classification of abnormalities of the athlete’s ECG was searched and reported [10]. Moreover, additional morphological parameters were also evaluated: voltage of the R-wave was measured in
lead V1, while the presence of inverted T-waves in leads V2 and V3 was recorded as a dichotomous parameter.
2.4 Statistical Analysis
Continuous variables are reported as means ± standard deviation or 95 % confidence intervals (CIs) and median (with interquartile range or 5th and 95th percentiles), and dichotomous variables as percentages. Trends and differences between sexes were analyzed with multiple regression analysis. A p value\0.05 was considered statistically significant.
3 Results
In a 3-month period, 2120 children participating in noncompetitive sports, aged 3–14 years, had an ECG referred to our center for analysis. Twenty were excluded because of artifacts limiting proper interpretation, while 40 were excluded because of concomitant medical treatment, leading to a final cohort of 2060 subjects. Of those, 1962 did not show any morphological abnormality and were used to derive normality ranges, while 98 showed ECG abnormalities.
Table 1 reports the clinical characteristics of the normal population, while Table 2 reports clinical and ECG characteristics of subjects with abnormal findings. Inter-observer agreement in ECG interpretation was characterized by a 0.98 kappa value. Of 1962 subjects with normal ECGs, 887 (45 %) were female and 1075 (55 %) were male.
Figure 2 shows the age distribution of subjects while Table 1 reports the type of sport practiced according to the Comitato Organizzativo Cardiologico per l’Idoneita` allo Sport (Italian Committee for sport eligibility) (COCIS) classification [11]. The age distribution of ‘abnormal’ ECGs is given in Fig. 3.
Figure 4 reports means and 95 % CIs for heart rate, PR interval, QRS duration, corrected QT
interval (QTc), and voltage of R-wave as measured in V1 according to age and sex. Charts reporting the median values, and 5th and 95th percentile values according to age and sex are also provided as a supplementary image.
3.1 Heart Rate
As known, a significant inverse correlation was observed between age and heart rate with a progressively decreasing trend from early childhood to adolescence (p\0.001). While from the age of 3–5 years this trend shows a steep curve, from the age of 6 years, the heart rate remains
substantially unchanged with male subjects showing a numerically, not statistically significant, lower heart rate during early adolescence. Sex-related differences were also evident with male subjects showing a significantly lower heart rate (p\0.001).
3.2 Atrioventricular Conduction
The PR segment increases slightly during childhood and early adolescence with a duration between 135 and 145 ms, with the trend reaching statistical, but not clinical, significance (p\0.001). No sex-related differences were observed.
3.3 Ventricular Conduction
For the PR segment, QRS duration also showed a slight increase throughout childhood and
adolescence with a stable duration between 85 and 95 ms. Furthermore, the trend reached statistical significance (p\0.001). Significant differences were also detectable between male and female subjects (p\0.001), with the latter showing a shorter QRS duration.
3.4 QTc Duration
During childhood, QTc shows a stable duration p-value not significant, approximately between 420 and 430 ms, without sex-related differences during childhood. Male subjects aged 13 and 14 years showed a trend toward a reduction in QTc, which may be related to the increased androgenic and adrenergic tone characteristic of adolescence.
3.5 R-wave’s Voltage (V1)
Voltage of R-wave as measured in V1 showed a progressive reduction from childhood to early adolescence (p\0.001), almost halving its value at the age of 14 years. Female subjects show a significantly lower R-wave voltage when compared with male subjects (p\0.001).
3.6 Negative T-waves in Leads V2–V3
As known, a progressive shift from negative to positive T-waves is common among children, reflecting the reduced workload of the right ventricle. While at the age of 3 years, negative T-waves in V2 and V3 are evident respectively in almost 70 and 60 % of children, a progressive transition to positive T-waves in both leads is observed (p\0.001 for both V2 and V3) with V3 showing positive T-waves first. Nonetheless, in both male and female subjects, negative T-waves in both leads may persist until adolescence in about 10 % of cases (Figs. 5, 6). No sex-related differences were observed.
3.7 Abnormal ECG Patterns
Abnormal ECG patterns were evident in 98 (4.7 %) cases; details are reported in Table 2. In this large series, rhythm disturbances were extremely rare with only 5 out of 2020 (0.2 %) showing other than sinus rhythm. None of those showing rhythms other than sinus rhythm needed further clinical evaluations. In two cases, sinus bradycardia was evident on the ECG, deeming additional evaluation with Holter ECG monitoring and an exercise test to rule out the possible presence of sick sinus syndrome. None of our subjects showed any delay of the atrioventricular conduction; in six cases, a Wolff Parkinson White pattern with ventricular pre-excitation (0.3 %) was evident. Incomplete right bundle branch block was the most common ECG abnormality encountered, with complete right bundle branch block being extremely rare (three subjects, 0.1 %). No subject showed either left bundle branch block nor voltage criteria for left-ventricular hypertrophy as defined by the Sokolow Lyon index. One child aged 6 years showed a QTc duration of 502 ms while two girls aged 3 and 6 years showed QTc values shorter than age-corrected derived limits (respectively, 412 ms with a 99 % CI age of 3 years of 411.1–429.5 ms; and 422 ms with a 99 % CI age of 8 years of 422.4–439.5 ms). No subjects showed a type I or II Brugada pattern. In one subject, a suspicious pattern with a J point elevation of 1 mm and a downsloping ST segment with negative T-waves was referred for further evaluations. Three subjects showed diffuse negative T-waves in leads other than V1–V3 and were referred for further investigations. Even in subjects with anomalous ECGs,
alterations were of none or minor clinical importance in 84; nonetheless, this electrocardiographic screening pinpointed 14 of 2060 (0.7 %) children with abnormal ECG findings worthy of additional investigations.
4 Discussion
Besides reporting updated, contemporary, Europeanderived normal electrocardiographic values for children and adolescents aged 3–14 years practicing non-competitive sports, several findings of this work deserve particular attention.
Previous studies have reported dynamic changes occurring in normal ECGs from birth to childhood. More than 30 years ago, Davignon et al. conducted a study on 2141 white children aged 0–16 years in Quebec, Canada, reporting normal limits for 39 variables [1]. Normal values for male and female subjects were not distinct.
More recently, Rijnbeek et al. have reported data from 1912 healthy Dutch children, aged 11 days to 16 years with limits differing substantially from those of Davignon, questioning the external validity of the previous study [2]. This latter contribution showed a skewed distribution of subjects favoring age groups from 8 to 16 years with a low representation of younger children.
To the best of our knowledge, our work is the first reporting ECG findings in a contemporary large European population of children and adolescents practicing noncompetitive sports. As evident from Fig. 1, the age of enrolled subjects shows a normal distribution with a balance between those aged younger and older than 8 years.
In line with previous reports, heart rate shows a progressive decrease in both male and female subjects with a diverging trend by sex (lower heart rates in male subjects, with significant
differences in the age class 8–10 years). When compared with the population reported by Rijnbeek et al., despite being physically active, our population shows a trend toward a higher heart rate with their mean falling outside of our tenth CI limit for ages older than 12 years.
Both atrioventricular and ventricular conduction showed a substantial stability among age and sex groups. Previous works showed that both PR and QRS duration increase as a part of the transition between pediatric and adult ECGs, reflecting the longer conduction time associated with growth, but with those phenomena reaching adult-like durations at the age of 3 years. Both PR and QRS durations reported in this work are in line with normal limits reported by Rijnbeek [2].
Interestingly, QTc duration in male and female subjects starts to diverge during early adolescence with male subjects showing a shorter QTc interval. This is common in adults, being probably related to the increased androgenic and adrenergic tone present in this age period.
While European and American guidelines on the preparticipation screening of competitive athletes diverge on the need for an ECG in addition to clinical history and physical examination as a useful tool to rule out the presence of cardiovascular abnormalities [12, 13], no experience exists in the setting of children practicing noncompetitive sports. Traditionally, in Italy, a country with considerable experience in pre-participation screening with a structured approach dating back to 1982, children and adults practicing recreational physical activities were screened on a clinical basis only by a pediatrician and general practitioner until 2013. Since then, the Italian Government
mandated by law that all subjects involved even in recreational physical activities should have an ECG recorded and analyzed to obtain eligibility. This approach may have some intrinsic limitations and, as in the case of competitive athletes, may show its effect later in time.
The issue is still a matter of debate. Despite current guidelines issued by the European Society of Cardiology advocating pre-participation screening of young competitive athletes with ECG [12], this approach is not shared in USA [13], where the cost-analysis studies calculated a cost per life saved with a pre-participating ECG screening in athletes of US$10–14 million dollars [14]. Other studies even questioned the real burden of sudden arrhythmic death during sport among young competitive athletes, estimated at 50–75 per year in USA [15].
The overall burden of sports-related sudden death, however, was 4.6 cases per million population per year in the general population, considerably more common than previously suspected [16].
Nonetheless, this report highlights that in this new, large, and common group of subjects referred for ECG evaluation, abnormalities requiring further investigations are not so rare, being manifest in about 1 every 140 screened subjects.
Despite being physically active, our children and adolescents did not show all the classical ECG features characteristic of trained adults such as bradycardia, slowed atrioventricular conduction, complete right bundle branch block, increased voltage of the QRS complexes, inverted T-waves in lateral leads, or early repolarization. Thus, when interpreting the ECG of physically active,
noncompetitive children and adolescents, caution should be exercised when labeling even non-specific alterations as exercise-induced abnormalities because no evidence of this was reported in this study.
4.1 Limitations
This is an exclusively electrocardiographic study; no data are available on clinical follow-up and further instrumental or genetic examinations. Although an ‘abnormal’ ECG is not always by itself a disease, at the same time, several disease conditions may be undetected by an ECG.
5 Conclusions
This is the first study reporting derived normal limits and abnormal ECG findings in a large contemporary European cohort of children and adolescents aged 3–14 years practicing non-competitive sports. This works substantially confirms previously reported normal limits for ECG interpretation by Rijnbeek et al. and shows that abnormal ECG findings are not as uncommon as expected, being manifest in less than 5 % of subjects considered otherwise healthy. Clear
pathological alterations such as atrioventricular blocks or left bundle branch block or significant repolarization abnormalities (inverted T-waves in lateral leads, QTc shortage or prolongation) are absent or extremely uncommon, deserving, when encountered, additional examinations. Even in a physically active population, the common features of an adult athlete’s ECG are absent.
_____________________________________
Tables and Figures
Table 1 Baseline characteristics of subjects with normal ECGs
Characteristics n = 1962 Age (years) 8 (6–10) Male, n (%) 1075 (52) Ethnicity, n (%) Caucasian 1889 (96.3) Asian 22 (1.1) African 26 (1.3) Latin American 25 (1.2) Practiced sportsa, n (%) A 188 (9.6) B 163 (8.3) C 322 (16.4) D1 754 (38.4) D2 535 (27.3) ECGb Heart rate (bpm) 85 (75–96) PR interval (ms) 140 (130–154) QRS duration (ms) 92 (82–100)
Sokolow Lyon index (mV) 25 (22–30) QT duration (ms) 358 (340–376) QTc (Bazett) (ms) 426 (407–444)
________________
ECGs electrocardiograms, COCIS Comitato Organizzativo Cardiologico per l’Idoneita` allo Sport
a According to the COCIS classification [A: low cardiac involvement (e.g., golf), B: moderate cardiac involvement (horse riding), C: cardiac involvement with isometric strain (body building), D1: moderate to high cardiac involvement (e.g., football), D2: moderate to high cardiac involvement with persistent increase in heart rate (marathon)] b Data are expressed as median (interquartile range)
Table 2 Characteristics of subjects with clearly abnormal ECG tracings
n (%) 12 (12.2)
QT interval (ms) 358 (340–376) Corrected QT interval (ms) 426 (407–444) Sokolow Lyon index (mm) 24 (23–29)
QT interval exceeding age corrected normal limits n (%)
QT interval shorter than age corrected normal limits, n (%)
1 (1.0)
2 (2.0)
Type I or II Brugada pattern 0 (0) Suspicious Brugada pattern 1 (1.0) Negative T-wavesa
, n (%) 3 (1.8) Early repolarization pattern, n (%) 6 (6.1) Wolf Parkinson White pattern 6 (6.1) Other non-specific alterations, n (%) 12 (12.2
Continuous variables are shown as median (interquartile range) AV atrioventricular, ECG electrocardiogram, HR heart rate, LBB left
ECG parameters N=98 Rhythm Sinus rhythm, n (%) 93 (94.8) Atrial/junctional rhythm, n (%) 5 (6.2) HR (bpm) 85 (75–96) Sinus tachycardia (HR [100 bpm), n (%) 7 (7.1) Sinus bradycardia (HR \50 bpm), n (%) AV conductiona 2 (2.0) PR interval (ms) 140 (130–154) Normal, n (%) 98 (100)
Any degree of AV block, n (%) 0 (0) Ventricular conduction
QRS duration (ms) 92 (82–100) Left axis deviation, n (%) 2 (2.0) Right axis deviation, n (%) 1 (1.0) Complete RBB, n (%) 3 (3.0) Incomplete RBB, n (%) 43 (40.8) Complete LBB, n (%) 0 (0) Repolarization
bundle block, RBB right bundle block a In peripheral leads other than V1–V3
Fig 1
Fig. 1 Location of centers participating in the study and sending children and adolescents’
electrocardiograms to a telemedicine ‘hub’ center where electrocardiograms were interpreted and sent back [number per Italian district (province)]
Fig. 2
Fig. 3
Fig. 4
Fig. 4 Means and confidence intervals for heart rate, PR interval, QRS duration, corrected QT interval (QTc), and voltage of R-wave as measured in V1 according to age and sex
Fig. 5
Fig. 6
Fig. 6 Two electrocardiogram tracings of a girl recorded at the age of 12 and 14 years, showing transition from negative to positive T-waves in leads V2 and V3.
References 1.
Davignon A, Rautaharju P, Boiselle E, Soumis F, Megelas M, Choquette A. Normal ECG standards for infants and children. Pediatric Cardiology. 1979;1:123–131.
2.
Rijnbeek PR, Witsenburg M, Schrama E, et al. New normal limits for the paediatric electrocardiogram. Eur Heart J. 2001;22:702–11.
3.
World Medical Association. Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310:2191–4.
4.
Molinari G, Reboa G, Frascio M, et al. The role of telecardiology in supporting the decision making process of general practitioners during the management of patients with suspect cardiac event. J Telemed Telecare. 2002;8:97–101.
5.
Molinari G, Valbusa A, Terrizzano M, et al. Nine years’ experience of telecardiology in primary care. J Telemed Telecare. 2004;10:249–53.
6.
Hancock EW, Deal BJ, Mirvis DM, et 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 AHA
Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the ACC Foundation; and the HRS. Endorsed by the International Society for Computerized
Electrocardiology. J Am Coll Cardiol. 2009;53:992–1002. 7.
Rautaharju PM, Surawicz B, Gettes LS, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part IV: the ST segment, T and U waves, and the QT interval: a scientific statement from the AHA Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the ACC Foundation; and the HRS. Endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol. 2009;53:982–91.
8.
Surawicz B, Childers R, Deal BJ, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part III: intraventricular conduction disturbances: a scientific statement from the AHA Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the ACC Foundation; and the HRS. Endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol. 2009;53:976–81.
Biasco L, Cristoforetti Y, Castagno D, et al. Clinical, electrocardiographic, echocardiographic characteristics and long-term follow-up of elite soccer players with J-point elevation. Circ Arrhythm Electrophysiol. 2013;6:1178–84.
10.
Corrado D, Pelliccia A, Heidbuchel H, et al. Section of Sports Cardiology, European Association of Cardiovascular Prevention and Rehabilitation. Recommendations for interpretation of 12-lead electrocardiogram in the athlete. Eur Heart J. 2010;31:243–59.
11.
Faina M, Veicsteinas A, Biffi A. Classificazione degli sport in relazione all’impegno cardiovascolare. Med Sport. 2010;63:9–13.
12.
Corrado D, Pelliccia A, Bjørnstad HH, et al. Study Group of Sport Cardiology of the Working Group of Cardiac Rehabilitation and Exercise Physiology and the Working Group of Myocardial and Pericardial Diseases of the European Society of Cardiology. Cardiovascular pre-participation screening of young competitive athletes for prevention of sudden death: proposal for a common European protocol: consensus statement of the Study Group of Sport Cardiology of the Working Group of Cardiac Rehabilitation and Exercise Physiology and the Working Group of Myocardial and Pericardial Diseases of the European Society of Cardiology. Eur Heart J. 2005;26:516–24. 13.
Maron BJ, Thompson PD, Ackerman MJ, American Heart Association Council on Nutrition, Physical Activity, and Metabolism, et al. Recommendations and considerations related to
preparticipation screening for cardiovascular abnormalities in competitive athletes: 2007 update: a scientific statement from the American Heart Association Council on nutrition, physical activity, and metabolism. Circulation. 2007;115:1643–55.
14.
Halkin A, Steinvil A, Rosso R, et al. Preventing sudden death of athletes with electrocardiographic screening: what is the absolute benefit and how much will it cost? J Am Coll Cardiol.
2012;60:2271–6. 15.
Maron BJ, Doerer JJ, Haas TS, et al. Sudden deaths in young competitive athletes: analysis of 1866 deaths in the United States, 1980–2006. Circulation. 2009;119:1085–92.
16.
Marijon E, Tafflet M, Celermajer DS, et al. Sports-related sudden death in the general population. Circulation. 2011;124:672–81
