Article Text

Download PDFPDF

ESC criteria for ECG interpretation in athletes: better but not perfect
  1. S Sharma,
  2. S Ghani,
  3. M Papadakis
  1. Department of Cardiac and Vascular Sciences. St George's University of London, London, UK
  1. Correspondence to Professor Sanjay Sharma, Department of Cardiovascular Sciences, St George's University of London, Cranmer Terrace, London SW17 0RE, UK; ssharma21{at}

Statistics from

The sudden cardiac death (SCD) of an apparently healthy young person (<35 years) has a devastating impact on the family and peers. Poignant newspaper articles and video footage showing the athletic prowess of the youth, apparent epitome of health and circumstantial paradox send ripples of emotion within the lay community. Over 80% of all exercise-related SCDs in young athletes are attributed to inherited or congenital cardiovascular disorders.1 2

Most causes of SCD are identifiable during life and several therapeutic strategies are available to minimise the risk of a SCD. Whereas most health professionals are staunch advocates of protecting young athletes on humanitarian grounds, the feasibility of implementing preparticipation screening (PPS) for cardiac disease specifically is frequently met with resistance and remains a heavily debated subject. The low incidence of SCD in sport (1 in 50 000) and need for multiple investigations to identify all implicated disorder raise issues pertinent to cost-effectiveness and are a persistent ‘Achilles heel’ for proponents of PPS. Furthermore, athletic training is associated with electrocardiographic patterns that may resemble those seen in patients with incomplete or morphologically mild expressions of primary cardiomyopathies and ion channelopathies. These false-positive results raise concerns about unnecessary investigations, erroneous disqualification and psychological harm to the athlete. Conversely, SCDs in sport are highly visible, claiming young lives.3

There is general agreement that a form of cost-effective PPS should be implemented; however, the precise methodology is contested. Most European countries do not offer state-sponsored PPS. In the USA, the American Heart Association implemented a PPS programme in 1996 to identify serious cardiovascular disorders in high school and intercollegiate athletes through a 12-point health questionnaire and physical examination.4 The American approach appears pragmatic but lacks sensitivity because prodromal symptoms are experienced by a minority of athletes before death, family history is rarely ascertained since most hereditary disorders have low event rates and concealed forms may not be exposed in sedentary relatives. Importantly, most disorders implicated in SCD do not exhibit abnormal physical signs.

In Italy a unique state-sponsored, national PPS programme has been in operation since 1982, evaluating several million young athletes annually with history, examination and 12-lead ECG. Athletes with abnormalities are investigated further and those with potentially serious cardiovascular abnormalities are disqualified. Prospective evaluation of the Italian PPS system demonstrated a 90% reduction in SCDs from 3.6/100 000 to 0.4/100 000 person-years over a 25-year period since its implementation and was predominantly achieved by fewer cases of SCD from cardiomyopathy. In contrast, SCD from anomalous coronary origins and premature atherosclerotic coronary disease remains unchanged, exposing a major shortcoming of the inability of the 12-lead ECG to singly identify all causes of SCD.5

Despite these limitations, and acknowledgement that most countries do not possess the resources to develop a similar de novo PPS system in all sporting individuals, the success of the Italian model in achieving its ultimate goal of reducing SCD, has led major sporting bodies, including the International Olympic Committee, to adopt the model and endorse its implementation in evaluation of athletes competing at regional and national level. Several financially endowed sporting disciplines in Europe and the USA have followed suit by funding PPS through private medical organisations.6

In 2005, the sport cardiology study section of the Working Group of Cardiac Rehabilitation and Exercise Physiology devised a common protocol for PPS in young athletes for conditions predisposing to SCD based on the Italian model.7 The incentive attempted to facilitate a unified approach in differentiating ECGs representative of physiological adaptation from those associated with cardiac pathology. Common ECG manifestations of an athlete's heart including first-degree atrioventricular block, early repolarisation and incomplete right bundle branch block (RBBB) were excluded; these variants were subsequently shown to collectively comprise 60% of all ECGs in a population of 32 652 recreational athletes.8

In 2010, the ESC guidelines for interpretation of ECG in young athletes were revised, and isolated QRS voltage criteria for left ventricular hypertrophy (LVH) were also excluded.9 In retrospect, the initial inclusion of isolated QRS voltage criteria for LVH as markers of abnormality might be considered naive since these are common normal variants in young athletes. Moreover, large QRS voltages in isolation are a rare phenotypic expression of hypertrophic cardiomyopathy (HCM) (<5%). In a comprehensive study of 1005 highly trained young athletes the prevalence of false-positive ECGs was 40% based predominantly on the expression of QRS voltage criteria for LVH.10 Such high false-positive rates have denigrated the role of ECG and fuelled arguments against its inclusion in PPS.11

In their paper published in Heart, Weiner et al12 compare the performance of the 2005 ECG criteria with the modified criteria published in 2010 in 508 university students (see page 1573). The study was prompted by an earlier investigation by the group in the same cohort, which demonstrated that addition of ECG to history and examination was associated with a significant increase in sensitivity from 45% to 90%; however, inclusion of ECG was associated with an unacceptable false-positive rate of almost 16%.13 The current study identified 11 athletes with cardiac abnormalities; five with congenital valvular heart disorders and six with marked LVH or chamber dilatation. After further investigation 3/11 athletes were disqualified owing to potentially fatal abnormalities. History and examination alone identified 5/11 abnormalities and addition of the ECG increased the yield to 10/11. In comparison with the 2005 criteria, use of the 2010 criteria was associated with a significant decline in the number of false-positive ECGs from 16% to almost 10%. The reduction was solely attributed to reclassification of isolated QRS voltage criteria from abnormal to normal, resulting in increased specificity from 83% to almost 90% without compromising sensitivity (91%). The results are encouraging but warrant comment pertinent to potential globalisation of PPS in nationally ranked athletes.

The study cohort was small when one considers that prevalence rates of disorders implicated in SCD in young athletes ranges from 1 in 500 (HCM) to 1 in 5000 (arrhythmogenic right ventricular cardiomyopathy (ARVC)). The identification of two individuals with cardiomyopathy and one with moderate pulmonary stenosis in just over 500 individuals is fortuitous and embellishes the potential for identifying a potentially serious disorder in 0.6% of athletes, a figure which is threefold greater than the Italian experience (0.2% of over 42 000 athletes).5 The small numbers of athletes examined do not permit the deserving accolade of the ECG in revealing ion channel disorders, accessory pathways, concealed forms of ARVC and early manifestations of HCM, which are associated with normal physical examination and would have demonstrated an even lower sensitivity rate of disease detection using a health questionnaire and physical examination alone.

Improved specificity aside, a persisting false-positive rate of 10% is not trivial and continues to raise concerns about the potentially deleterious impact of ECG inclusion for PPS, particularly in countries with large populations of elite athletes and those with limited financial resources. However, systematic analysis of the precise nature of ECG abnormalities observed by the investigators is of interest and offers optimism in reducing the false-positive rate further. Of the 49 athletes with abnormal ECGs, 28 exhibited voltage criteria for left atrial enlargement alone or in combination with high QRS complexes and 13 exhibited RBBB. Our own experience of evaluating over 20 000 athletes would consider these variants as normal14 and with these considerations, the true false-positive rate approximates to a readily acceptable 2%.

Finally, ESC criteria for ECG interpretation in athletes are derived solely from an adult Caucasian population. In this respect, the investigators used an appropriate cohort to test performance. All athletes were adult and almost 70% were Caucasian. Although the qualitative ECG changes in adult male and female athletes are similar, the potential pandemonium and diagnostic dilemmas associated with extrapolation of these criteria to the adolescent or African/Afro-Caribbean (black) athlete population cannot be underestimated. The issue is imperative since both groups comprise an increasing population of sportsmen competing at national level that will be subjected to PPS. T-wave inversion in the anterior leads resembling ARVC are present in approximately 4% of all adolescent athletes15 and notched T waves mimicking long QT syndrome type 2 are even commoner in this age group, yet the differentiation from sinister pathology is prudent since 40% of all SCD in sport affects adolescent athletes. Almost a quarter of adult black male and 12% adult black female athletes exhibit T-wave inversion, a recognised indication for further investigation in a Caucasian athlete. Although, the 2010 ESC criteria make recommendations that T-wave inversion confined to V1–V4 may not warrant assessment, this still leaves 10% of black athletes with T-wave inversion in other leads requiring further assessment based on this single criterion alone.16

The ESC criteria for interpretation of the athlete's ECG in adult Caucasian athletes are commendable. Modifications of the criteria in 2010 have strived for perfection and improved specificity. In a cost-prohibitive financial climate the ECG is probably the most effective tool for identifying conditions predisposing to SCD in young athletes. However, there is room for improvement by examining the true relevance and the requirement for many criteria such as left atrial enlargement, axis deviations and RBBB with large-scale studies; exclusion of these criteria may reduce the false-positive rate significantly without affecting sensitivity. Furthermore, inclusion of more detailed guidelines for ECG interpretation in adolescents and non-Caucasians is crucial before the criteria can be considered replete for global use.


View Abstract


  • Linked article 227330.

  • Disclosures SG is funded by a grant from the charitable organisation, Cardiac Risk in the Young, which supports preparticipation screening of athletes.

  • Guarantor SS is guarantor.

  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles