Preparticipation cardiac screening in athletes

How best to prevent these tragic events is a matter of intense debate in both the USA and Europe. Most efforts have focused on identification of those at risk through preparticipation evaluation (PPE, or screening) either through history and physical (H&P) or through H&P combined with ECG. Currently, many professional sporting organisations including the International Olympic Committee mandate or recommend ECG screening.7 The US National Collegiate Athletic Association (NCAA) considered mandating ECG screening8 but ultimately decided against this.9 Recently, professional societies in both the USA and Europe have published recommendations statements regarding whether and how to best screen athletes to identify those at risk for SCD. Both the American Heart Association/American College of Cardiology (AHA/ACC)10 and the European Society of Cardiology/European Heart Rhythm Association (ESC/EHRA)7 documents extensively review the data on sensitivity, specificity and outcomes of screening methods, as well as the economic, societal and ethical contexts—each bibliography with well over 150 references, largely the same, yet reach very different conclusions. While both recommend PPE in athletes, the AHA/ACC recommends standardised H&P only, while the ESC/EHRA concludes that the data are sufficient to support addition of the ECG.

It is clear that the ECG improves yield for diagnosis of potentially life-threatening disorders such as channelopathies and cardiomyopathies.7 10–14 The H&P lacks sensitivity for asymptomatic electrical abnormalities such as Wolff-Parkinson-White or long QT syndrome, both entities in which risk stratification and appropriate treatment in the asymptomatic patient can decrease mortality,15 16 and which require an ECG for diagnosis. Cardiomyopathies also may not be identified on H&P. The ECG is not perfect—coronary artery anomalies, among the most common aetiologies of SCD in the young, are most often electrically silent.17 However, overall, the ECG will be abnormal in 80% of individuals with the conditions making up two-thirds of sudden death in young athletes.7 Complicating the use of ECG screening in athletes is the fact that many of the normal adaptations to athletic activity, such as left ventricular hypertrophy and changes in the right ventricle, can mimic pathological conditions. However, ongoing refinements of criteria to differentiate normal adaptation from pathology have improved the specificity (figure 2).18 19 In some studies, the ECG not only has improved sensitivity, but a lower false-positive rate—that is, improved specificity—than the H&P.13 Recent recommendations for ECG interpretation in athletes also describe appropriate follow-up evaluation including imaging and other testing of athletes whose ECGs fall into the abnormal category.18 While increasingly diagnostic tests can most often differentiate an athletes' heart from pathology, there remains a ‘grey area’ between normal and pathologic, as shown in figure 3.20 Furthermore, for an athlete with a highly abnormal ECG with no abnormalities found on imaging, how often this truly represents a ‘false positive’, versus latent disease not yet manifest, is unknown. Athletes whose ECG is abnormal should be followed serially over time.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

International consensus standards for ECG interpretation in athletes. International consensus standards for ECG interpretation in athletes. Reprinted with permission from Sharma et al 2017.18AV, atrioventriular block; LBBB, left bundle branch block; LVH, left ventricular hypertrophy; RBBB, right bundle branch block; RVH, right ventricular hypertrophy; PVC, premature ventricular contraction; SCD, sudden cardiac death.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

‘Grey zones’ of athletic adaptation and pathology differentiating features between physiological cardiac changes and cardiomyopathy in athletes. Reprinted with permission from D’Silva, Heart, 2017.20 ARVC, arrhythmogenic right ventricular cardiomyopathy; CMR, cardiac magnetic resonance; DCM, dilated cardiomyopathy; FH, family history; HCM, hypertrophic cardiomyopathy; LBBB, left bundle branch block; LV, left ventricle; LVH, left ventricular hypertrophy; LVNC, left ventricular non-compaction cardiomyopathy; peak VO₂, peak oxygen consumption; RV, right ventricle; RWMA, regional wall motion abnormalities; VT, ventricular tachycardia.

Whether improved sensitivity of the ECG translates into more lives saved is controversial. Just two studies have been able to address this question,21 22 both comparisons of athlete death rates before and after initiation of a national screening programme, both with much-discussed flaws, which reached very different conclusions. Differences in the interpretation of these scientific data underlie much of the difference in recommendations between European and American societies.

The seminal study whose data showed effectiveness of the ECG, published in JAMA in 2006 by Corrado et al,21 examined mortality rates in athletes and non-athletes in the Veneto region in Italy before and after institution of mandatory athlete ECG screening in Italy. Following institution of ECG screening, the sudden death rate in athletes went from 3.6/100 000 person-years to 0.4/100 000 person-years, with no change in sudden death rate in non-athletes (figure 4, purple line). Strengths of this study included the SCD endpoint ascertainment, which used an ongoing prospective regional registry with mandatory reporting to confirm the numerator, as well as use of the Sports Medicine Database of the Veneto region, which records all athletes actively participating, to confirm the denominator. Furthermore, the study included a control group of non-athletes in the same region, for whom there was no change in SCD rate, excluding the possibility of other secular trends as the reason for the decrease in athletes. A weakness of the study, however, was the availability of only 3 years of data prior to the screening programme, leaving open the possibility of random yearly variation as a possible explanation for the finding. Also, arrhythmogenic right ventricular cardiomyopathy, a common cause of death in athletes found in Italy23 but less common in other countries,4 6 is uniquely exacerbated by vigorous exercise24 and is also often manifest on the ECG. ECG screening in this population may thus have more benefit. The US screening document also notes that the prevalence of SCD in athletes after ECG screening in Veneto was similar to the prevalence of SCD without screening in Minnesota and also in Denmark. However, the differing methodologies used—insurance claims in Minnesota, death certificates in Denmark and a prospective registry in Italy, make it difficult to compare absolute incidences. Also, the Italian study considered all deaths occurring in athletes in the numerator,21 while both the Minnesota25 26 and Danish27 studies considered only those deaths occurring during organised sporting events or practices, and not all deaths in athletes occur during organised activity.28

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Incidence of sudden death after institution of mandatory ECG screening in athletes. Annual incidence of sudden cardiac death expressed per 100 000 person-years in the three studies evaluating the effects of screening on the mortality of athletes over time. The Italian study (4) (pink graph) concluded that ECG screening (started in 1982) significantly reduced the incidence of sudden cardiac death by comparing the sudden death in the 2-year prescreening period (A to B) with the postscreening period (B to F) . This study is depicted by the green graph. [The Israeli study] compared the 12 years before screening (C to E) with the 12 years after the onset of mandatory ECG screening (E to G). Had [that study] limited the comparison of the postscreening period to the 2-year period preceding the enforcement of screening in Israel (D to E vs E to G, as performed in the Italian study), [that study] would have concluded erroneously that screening saved lives of athletes in Israel. The study from Minnesota (19) (yellow graph) shows a low mortality rate in a population of athletes not undergoing systematic ECG screening. Reprinted with permission from Steinvil et al 22

Following the publication of the Italian study, Israel instituted a similar screening policy. In comparison of SCD in athletes before and after, Steinvil et al 22 found no change in death rates (figure 4, green line). This study avoided the pitfall of a short prescreening period, but had other weaknesses. Ascertainment of athlete SCD was performed solely through searches of two newspapers. However, a report from the USA of a largely media-based registry of athlete deaths shows an increase in athlete SCDs during a similar timeframe, attributed in large part to enhanced reporting,2 suggesting the possibility of a significant reporting/surveillance bias. In a study of sports-related SCD in Denmark using both a national registry with extensive death certificate, autopsy and medical record review and media searches, media searches revealed just a small percentage of sports-related SCD,29 further emphasising the poor reliability of this method of endpoint ascertainment for the numerator. Furthermore, the denominator was extrapolated from a single year of data from the Israeli Sports Authority and population estimates.

In consideration of the strengths and weaknesses of these two population-based studies, the European expert writing group concludes,

The protocol of PPE including clinical history, physical examination, and 12-lead ECG demonstrates to have superior diagnostic capability than just clinical history and physical examination. There is compelling scientific evidence that the 12-lead ECG improves substantially the diagnostic power of PPE, mostly due the capability to identity arrhythmogenic conditions at risk (cardiomyopathies and channelopathies).7

The US group, however, concludes,

It is recommended that the AHA 14-point screening guidelines and those of other societies… be used by examiners as part of comprehensive history-taking and physical examination to detect or raise suspicion of genetic/congenital and other cardiovascular abnormalities… Screening with 12-lead ECGs (or echocardiograms) in association with comprehensive history-taking and physical examination to identify or raise suspicion of genetic/congenital and other cardiovascular abnormalities may be considered in relatively small cohorts of young healthy people 12 to 25 years of age…10

The US document raises other concerns with ECG screening in addition to lack of definitive outcome data, including imperfect sensitivity and specificity, cost, difficulties with quality control and ethical issues. However, it should be noted that each of these concerns applies to screening with H&P as well. For example, while ECG screening does not meet many of the criteria of an ideal screening test as defined by the WHO,10 H&P screening does not either. As above, while sensitivity and specificity of the ECG are imperfect, both are likely better with the ECG. Quality control is another issue, but while there are difficulties with quality control both with obtaining the ECG, and with physician interpretation,30 H&P performance also falls short.31 Initiatives to train physicians in the preparticipation H&P, such as now required in some US states, are an important endeavour. Efforts to improve ECG interpretation by physicians are also underway, through the British Medical Journal, 32 which has been shown effective at improving accuracy in a study of sports medicine fellows.33

Cost of PPE screening is high, whether H&P alone, or including ECG, with cost-effectiveness studies showing widely varying results based on assumptions made. One Markov analysis compared cost-effectiveness of screening with H&P alone and combined with ECG, over no screening.34 That study used the Italian data on risk reduction with ECG screening combined with data on sensitivity and specificity of H&P with and without ECG and on secondary testing generated, and cost data from the National Center for Healthcare Statistics, and found a cost per year of life saved to be $42 900, well within what is considered societally acceptable. Because of the increased sensitivity of the ECG, the cost per year of life saved was significantly higher with H&P alone, at $76 000. A subsequent study also starting with the Italian data35 calculated a cost per life saved of over $10 million. However, that study reported the cost per life, rather than the standard cost per year of life saved, and also reported the total cost, most of which actually was generated by the H&P.36

Psychological impact is another concern.10 However, studies have not shown differences in anxiety with ECG vs H&P-alone screening,37 38 which is logical—if the false-positive rate is similar, there would be no reason why one form of screening would be more stressful than another.

Finally, both the US10 and European writing groups7 raise the ethical issue of screening just one segment of the population—athletes—while other young people are not screened. Some studies do show a higher incidence of SCD in athletes than non-athletes,23 but both are low-frequency occurrences and within the same order of magnitude—2.3 vs 0.9 per 100 000 person-years in the Italian experience23 and 1.14 vs 0.31 per 100 000 in a study of US high schools.39 Other studies have shown a higher incidence of SCD in young non-athletes than athletes.27 Ultimately, all young people may benefit from screening. While athlete screenings are generally done in specialised settings or programmes, other care-delivery models may be more feasible at a population level. Small pilot studies of ECG screening in physician’s offices40 41 have shown this model to be feasible. Whether screening should be mandatory is another ethical issue which has been raised. However, athletes are already forced to undergo screening by schools and sporting organisations, and the addition of an ECG does not seem to create an ethical difference. The competing rights of a school or sporting organisation to require screening versus the right of an athlete to refuse it would seem to apply similarly to screening whether H&P alone or including ECG.

Ultimately, however, the two sides of the Atlantic may not be that far apart. The AHA/ACC document does not recommend against inclusion of ECG with H&P in small programmes or research studies, but rather concludes that the data are not sufficient to mandate inclusion of the ECG. The ESC writing group notes that in the context of national differences in cultural, legal, social, logistic and economic frameworks, as well as differences in interpretation of available scientific evidence, it is beyond the scope of that document to suggest global mandatory PPE.

There are several conclusions with which all can likely agree. First, more data are urgently needed to assess whether ECG screening, as well as screening with H&P, saves lives or not. Given the conflicting data on outcomes, with two imperfect studies showing very different results, the current state of clinical equipoise may present a unique opportunity to determine the effectiveness of ECG screening at decreasing SCD in athletes. Prospective, two-group studies are crucial to determining the potential benefits of screening. While a randomised study would be ideal, given the overall event rate, very large studies will be needed. In a back-of-the-envelope calculation using the 1 per 43 000 person-years rate of sudden death found in NCAA athletes,3 and an online sample size calculator, to prove a reduction by three-quarters over a 4-year period would require over 180 000 athletes per group. While these numbers are large, prospective studies of high school student athletes of this magnitude have been completed and could serve as a model.39 42 Another option might be creation of larger, multicentre registries of athletes who have undergone screening with and without ECG; crucial will be the incorporation of adequate long-term follow-up.

Appropriate response to cardiac arrest stemming from adequate community preparedness, with emergency action plans in place, is also critical in decreasing death from sudden cardiac arrest (SCA) in the athlete.9 The current survival from SCA remains under 10%,43 44 and the incidence of psychopathology in the few who do survive is high.45 Research into resuscitation and postarrest care, such as therapeutic hypothermia, as well as improvements in community response, such as expanding use of automated external defibrillators44 and creation of appropriate emergency action plans, are all crucial to improving survival from SCA in athletes.

But an ounce of prevention will always be better than a pound of cure. Further research on ECG screening, considering alternative economic models, improving specificity and sufficiently large-scale research into effectiveness, are all crucial to decreasing death from SCA in athletes and potentially all young people as well.