The pro arguments

Accurate epidemiologic data on the risk of SCD in any given population are necessary for evaluating the cost effectiveness of screening and providing adequate recommendations based on appropriate risk-to-benefit considerations. Estimates of the incidence of SCD vary widely from country to country, mostly as a consequence of different methods of collecting data. The reported incidences of cardiovascular fatalities during sport are rare and range from ∼4 per 100 000 Italian competitive athletes per year (before implementation of universal preparticipation screening) to <1 per 100 000 participants per year among US high school and college athletes.2 ,3 While Italian estimates are derived from a prospective investigation with rigorous data collection and systematic postmortem investigation, the lower prevalence reported in other countries, particularly the USA, are explained by a lack of a mandatory reporting system for juvenile sudden death, with most studies relying solely on surveys, media reports, and insurance claims. Therefore, reasonable concerns exist regarding the reliability of these studies for the estimation of SCD rates due to unavoidable limitations inherent in the data collection and the retrospective study design.

The rationale for the preparticipation screening programme is based on the demonstration that young competitive athletes represent a special subset of the general population who are at higher risk for SCD. Available data coming from the prospective Italian study show a three times greater risk of SCD in young competitive athletes compared with their sedentary counterparts.4 More recently, Marijon et al5 reported that the relative risk of sport related SCD in France is 4.5 times higher in young competitive athletes (age 10–35 years) compared with non-competitive sports participants of the same age.

The public health relevance of a disease/event to deserve population based screening is not necessarily linked to its high occurrence. The ultimate goal of the preparticipation examination is to make athletic activity as safe as possible by minimising the associated risk to health. Athletic screening does not aim to identify those very few athletes who will die suddenly, but to detect silent cardiovascular abnormalities that can lead to SCD and to prevent fatal events during sports by appropriate interventions. From a screening perspective, the prevalence of cardiovascular conditions with the potential for SCD among adolescents and young adults is of the utmost important. In contrast to the disparities on SCD rates, the overall prevalence of cardiovascular diseases that predispose to SCD in the general population of young athletes has been consistently estimated worldwide within a narrow range of 0.2–0.3%.6

The con arguments

The highly publicised fatalities occurring in sports competitors generate the false idea that sport related SCD occurs frequently, while in fact it is a very rare event. The uncommon occurrence of SCD in competitive sports underlines the limitations implicit in structuring productive and cost effective, broad based preparticipation screening strategies. Trying to identify the extremely rare young athlete who will die suddenly during sports activities among millions of competitors is a clinical challenge with a prohibitive cost.

The only data supporting the increased risk of SCD in athletes come from the Italian studies. Recent epidemiologic observations from other countries indicate that the risk of SCD in the athletic population may not be as high as reported in Italian competitors and that the general non-athlete population may have a similar or even higher risk of SCD compared with athletes.

Maron et al7 reported an annual incidence of SCD among US high school and college athletes from Minnesota (USA) of 0.93 per 100 000 athletes annually, and argued that SCD is a low event rate phenomenon which is unlikely to be influenced by preparticipation cardiovascular evaluation.

Holst et al8 reported an annual rate of SCD in the Danish general population (age range 12–35 years) of 3.76 per 100 000 persons, while the estimated SCD annual rates of sports related SCD was 1.21 per 100 000 athletes (defined as individuals doing physically demanding sports and taking part in competition). The conclusions were that, in Denmark, sport related SCD in the young is a rare occurrence and its incidence is approximately three times lower than that of SCD unrelated to competitive sports activity. The authors concluded that the mortality rate in unscreened Danish athletes is comparable to that found in the Italian athletic population which has been exposed to systematic preparticipation screening for >25 years and, thus, inferred that preparticipation athletic screening is likely to be of low value.

It should be recognised that, in the Danish study, the numerator (ie, the number of SCD victims over the study period) was obtained by retrospective analysis of death certificates and public media reports on fatal events occurring during or within 1 h after moderate-to-high intensity exercise, and diagnosis of the cause of death on postmortem investigation by forensic pathologists. Moreover, the denominator data did not reflect the real number of active athletes in each year but was an estimate of the competitive athlete population, which was based on an interview of people stratified according to geographic as well as socioeconomic parameters.

The pro arguments

The most common mechanism of SCD during sports activity is an abrupt ventricular fibrillation as a consequence of a wide spectrum of cardiovascular diseases (box 1, figure 1).12 Such diseases are often clinically silent and unlikely to be suspected or diagnosed on the basis of spontaneous symptoms. This explains why a screening protocol such as that recommended by the AHA, based solely on the athlete's history and physical examination, is of marginal value for identification of athletes at risk for SCD. Cardiovascular preparticipation evaluation offers the potential to identify athletes with life threatening cardiovascular abnormalities before the onset of symptoms. A mass preparticipation screening programme, essentially based on the ECG, has been in practice in Italy since 1982 (figure 2). This screening programme has shown that ECG, in addition to history and physical examination, has a substantial incremental value for early identification (ie, increase of screening sensitivity) of asymptomatic athletes who have potentially lethal heart disorders distinctively manifesting with ECG abnormalities, such as cardiomyopathies, pre-excitation syndromes, and cardiac ion channel disorders (table 1).13 Based on published series from the USA and Italy, these ECG detectable conditions are present in approximately two-thirds of young competitive athletes with SCD.9

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Leading causes of sudden cardiovascular death in young competitive athletes. (A) Hypertrophic cardiomyopathy: short axis cut of the heart specimen showing asymmetric septal hypertrophy with multiple septal scars (top); histology of the interventricular septum revealing typical myocardial disarray with interstitial fibrosis (bottom) (Heidenhain trichrome). (B) Arrhythmogenic right ventricular cardiomyopathy/dysplasia: section of the heart specimen along the right ventricular infundibulum (left); panoramic histological view of the right infundibular free wall showing wall thinning with fibro-fatty replacement (right) (Heidenhain trichrome). (C) Atherosclerotic coronary artery disease: histology of the proximal tract of the left anterior descending coronary artery showing a non-obstructive fibrous plaque complicated by luminal thrombosis due to endothelial erosion (Heidenhain trichrome). (D) Congenital coronary anomaly: gross view of the aortic root showing both coronary ostia located in the right coronary sinus, pointing to an anomalous left coronary artery arising from the right aortic sinus of Valsalva and running between the aorta and the pulmonary trunk. Modified from Corrado et al.6

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Flow diagram illustrating the Italian screening work-up. The initial cardiovascular protocol includes family and personal history, physical examination (with determination of blood pressure), and basal 12 lead ECG. Additional tests are requested only for athletes who have a positive family history of cardiac disease/premature sudden death, relevant symptoms, abnormal physical findings, or group 2 ECG abnormalities. In athletes with a negative familial/personal history and a normal physical examination, the finding of ECG changes due to cardiac adaptation to physical exertion (group 1) should allow participation in competitive sport activity without additional investigations. Athletes diagnosed with clinically relevant cardiovascular abnormalities are managed according to available guidelines for assessing athletic risk. Group 1 and group 2 ECG abnormalities are reported in figure 3. EMB, endomyocardial biopsy; EPS, electrophysiologic study. Modified from Corrado et al.14

A common objection to ECG screening is the concern for a high false-positive rate, leading to either unnecessary secondary evaluations or restriction from sporting activity. Early studies demonstrated ECG ‘abnormalities’ in up to 40% of athletes, causing many physicians to dismiss the idea of ECG screening. However, it is critical to recognise that the false-positive rate for ECG screening is largely affected by the criteria used to define ‘abnormal’. Significant advances have been made in our understanding of how to interpret an athlete's ECG. Modern standards have been proposed to improve the ability to distinguish physiologic cardiac adaptations in athletes from findings suggestive of underlying pathology (ie, increase of screening specificity) (figure 3).14 Recent studies applying modern ECG criteria have resulted in substantially lower false-positive rates.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Classification of ECG abnormalities in the athlete according to European Society of Cardiology recommendations. Common ECG abnormalities: up to 80% of trained athletes exhibit ECG changes such as sinus bradycardia, first degree atrioventricular (AV) block, early repolarisation, incomplete right bundle branch block (RBBB), and pure increase of QRS voltages (group 1). Such common ECG changes are the consequence of the physiologic cardiovascular adaptation to sustained physical exertion and do not reflect the presence of an underlying cardiovascular disease. Therefore, they are not associated with an increase of cardiovascular risk and allow eligibility to competitive sports without additional evaluation. Uncommon ECG abnormalities: this subset includes uncommon ECG patterns (<5%) such as ST segment and T wave repolarisation abnormalities, pathological Q waves, intraventricular conduction defects, and ventricular arrhythmias (group 2). These ECG abnormalities are unrelated to athletic conditioning and should be regarded as an expression of possible underlying cardiovascular disorders, notably cardiomyopathies and cardiac ion channel diseases, and, thus, are associated with an inherent increased risk of sudden arrhythmic death. LBBB, left bundle branch block. Modified from Corrado et al.14

The con arguments

The value of the ECG as a discriminatory test is not well established. The ideal screening test should distinctively detect cardiovascular conditions responsible for SCD in all affected athletes (ie, high sensitivity) and rule out these disorders in healthy athletes (ie, high specificity). However, the low disease prevalence and non-adequate estimates of sensitivity and specificity of the 12 lead ECG result in inefficient screening strategies with unjustified socioeconomic costs. The ECG does not detect all disorders predisposing to SCD. It is unable to identify premature coronary artery disease (CAD) and congenital coronary anomalies, which account for approximately 20% of SCD in young athletes. In addition, the ECG may miss approximately 5% of patients with hypertrophic cardiomyopathy (HCM) and does not identify aortic root disorders. Because of the low disease prevalence and inherent inaccuracy of tests (including ECG), screening for rare disorders will lead to many false-positive results that enable additional diagnostic tests and, possibly, unnecessary treatments and disqualification from sports activity. The safety, efficacy, and acceptability of specific therapies and recommendations for prevention of SCD in asymptomatic young competitive athletes detected by screening is often uncertain and based on expert consensus.

Whereas ECG screening has been proven to be lifesaving in young competitive athletes (12 to 35 years), in whom SCD is mostly caused by ECG detectable genetic cardiomyopathies, it appears to be a non-accurate test for detecting CAD, which is the most common cause of SCD in middle-aged/senior athletes engaged in leisure time sports.6 Despite its low diagnostic accuracy, exercise testing is the currently recommended test to screen for CAD in the general population of asymptomatic individuals in this age group before an exercise programme. Screening of children (<12 years) is expected to have a lower sensitivity for detection of genetic cardiomyopathies and cardiac ion channel diseases (except for long QT syndrome), whose clinical and ECG findings may not manifest until adolescence. Available studies indicate that up to a third of healthy athletes of Afro-Caribbean descent exhibit pronounced repolarisation abnormalities, which might make preparticipation ECG screening more problematic. Box 2 summarises the arguments in favour and against including the 12 lead ECG at rest for the preparticipation check-up.

The pro arguments

The importance of early identification by ECG screening of athletes with arrhythmogenic cardiovascular disorders at a presymptomatic stage relies on the concrete possibility of SCD prevention by lifestyle modification, such as restriction of competitive sports activity. Italian athletes who did not obtain eligibility for competition because of cardiovascular reasons were found to have a good long term clinical course. In particular, no deaths occurred among asymptomatic athletes who were identified and disqualified because of HCM during a long term follow-up period of ∼8 years.13 Furthermore, preparticipation identification of an athlete with a disease at risk of SCD is ‘lifesaving’ when disqualification from competitive sports is followed by subsequent close follow-up and clinical management. The former athlete becomes a patient who enters a clinical programme of risk stratification and prophylactic antiarrhythmic treatment, including antiarrhythmic drugs, β-blocker drugs, and implantable cardioverter-defibrillator therapy.

The con arguments

Disqualification from competitive sports activity cannot be considered an effective preventive strategy for all athletes with an early diagnosis of risky cardiac diseases. It is unrealistic to expect that disqualified athletes will completely refrain from physical exercise. In addition, the threshold to consider physical exercise ‘safe’ has not been defined and there are reports of fatalities occurring in athletes who continued to exercise against medical advice. Furthermore, if high risk athletes are correctly identified and they actually stop exercising, this may not be enough to eliminate the risk of SCD because emotional and stress related ventricular arrhythmias may also precipitate cardiac arrest in individuals with susceptible heart muscle disorders.

The pro arguments

The long term Italian experience with preparticipation screening has provided the most compelling evidence that initiation of treatment (ie, restriction of sports activity and medical treatment) before the onset of symptoms may alter the natural course of at-risk cardiovascular disease in the athlete and improve prognosis. Comparison of incidence data for SCD in Italian athletes and non-athletes aged 12–35 years in the Veneto region of Italy, before and during this screening programme, demonstrated that such a systematic preparticipation evaluation, coupled with athletic restriction, resulted in a decline in deaths among screened athletes from 3.6 per 100 000 person-years in the 2 years before the implementation of screening to 0.4 per 100 000 person-years two decades later (ie, 89% mortality reduction); in contrast, there was no change in deaths among unscreened non-athletes, suggesting that this reduction was not due to changes in the population death rate.2 Most of the decrease in death was due to fewer deaths attributable to cardiomyopathies, and the number of asymptomatic athletes diagnosed and disqualified because of cardiomyopathies increased during the same period. Of the 42 386 screened athletes, 9% required additional cardiovascular testing and 2% were ultimately prohibited from athletic participation.

Advocates of ECG screening argue that the true incidence of SCD is not sufficiently appreciated in other countries such as the USA, with data relying on media reports underestimating the true incidence of SCD. A recent study on National Collegiate Athletic Association (NCAA) athletes identified a much greater SCD rate in US athletes than previously reported. At the same time, the study demonstrated that only 56% of SCD events were actually identified by public media reports, despite weekly systematic internet searches.15

The con arguments

Antagonists stressed that the Italian study was not a controlled comparison of screening versus non-screening of young competitive athletes but an observational population based investigation.16 The study was not a randomised trial, and unequivocal conclusions that the reduced mortality was solely the consequence of the screening process cannot be drawn. In addition, it has been argued that the annual death rate of Italian competitive athletes before mandatory screening was higher and the late screening annual death rate was roughly similar to that reported in US high school and college athletes who were not screened with ECG. It has thus been suggested that no screening process or the less formal screening process practised in the USA might have been as effective as the Italian programme.7

Moreover, Steinvil et al17 compared SCD rates in the 12 year periods before and after mandatory ECG screening was initiated in Israel and found that the average yearly incidence of athletic SCD increased from 2.54/100 000 to 2.66/100 000 athlete-years after the screening programme was initiated. In this study, SCD rates were calculated based on media reports and estimates of the number of Israeli athletes. Reliance on news reports unavoidably underestimates SCD numbers and also biases the data toward a greater number of SCD cases in more recent years (due to the general increase in the ability to distribute news reports over the internet). Furthermore, assumptions regarding the denominator (ie, the number of screened athletes) were only estimates.

The pro arguments

Former athletes diagnosed with a genetic disease at risk of arrhythmic cardiac arrest will survive for many decades with normal or nearly normal life expectancy, due to restriction from competition and prophylactic therapy against life threatening arrhythmias. The large amount of life-years saved influences cost effectiveness analysis and explains favourable cost estimates per year of life saved by screening. According to the Italian experience on mortality reduction after screening implementation (from 3.6 to 0.4 per 100 000 person-years), the expected absolute survival benefit is ∼4 lives saved per 100 000 person-years of screening.2 Therefore, 25 000 athletes must be screened to save one life, with an estimated cost of ∼€1 000 000 (∼£850 000, ∼US$1.35 million). If one assumes that former young competitive athletes saved from SCD because of identification and disqualification from competitive sports will live at least an additional 20 years, the cost per year of life saved can be estimated at ∼€50 000 (∼£43 000, ∼$67 000) which is the traditional threshold to consider a health intervention cost effective. According to Myerburg and Vetter,18 as a result of the large number of cardiac conditions that have a genetic background, diagnosis of these abnormalities by ECG screening in adolescents and young adults enables cascade family screening which, in turn, may result in subsequent identification of other affected family members, with a further increase of screening cost effectiveness and decrease of overall mortality.

The con arguments

The Italian data suggest that, even in the most experienced hands, mandatory screening of all competitive athletes would save few lives at large a cost.19 This money could be better spent on secondary prevention of sport related SCD, including enforced availability of external automatic defibrillators in every sports location. Chaitman20 argued that because of the huge number of individuals involved in a network of sporting activity, the lack of research and the high cost of tests, implementing universal ECG screening for all competitive athletes in the USA is not feasible to do in a cost effective manner. The author pointed out that universal preparticipation screening removes resources from the healthcare system that might be better used for more urgent health problems such as the emerging risk of diabetes mellitus and obesity in US adolescents and young adults.