Article Text

other Versions

PDF

Screening of competitive athletes to prevent sudden cardiac death: Think Twice
  1. Anne Mackie
  1. Correspondence to Dr Anne Mackie, Imperial College Healthcare NHS Trust, Mint Wing, Centre Block G, South Wharf Road, London W2 1NY, UK; Anne.mackie{at}imperial.nhs.uk

Statistics from Altmetric.com

Sudden cardiac death (SCD) in young competitive athletes is rare but has led to calls for screening. The rationale seems simple: if only we had known we might have been able to prevent the catastrophe. Yet even if this were true, it would require screening of many hundreds of thousands of cases to identify high-risk individuals and respond appropriately. Hypertrophic cardiomyopathy (HCM) is the single most common cause of SCD in young people, contributing to around a quarter of cases, and was looked at by the UK's National Screening Committee (NSC) in 2010, with more recent attention given to other causes of SCD.1 A consultation document is now in preparation, but meanwhile I present some of the issues that require consideration when proposing a national screening programme.

First it is prudent to refer to some of the classical screening appraisal criteria set out in the 1950s by Wilson and Jungner.2 (1) How common is the problem? (2) What is the natural history of the disorder and the appropriate timing of intervention? (3) Is there an accurate test? (4) Once diagnosed, is there effective treatment? (5) Are services in place to deliver the required number of diagnostic tests, interpret the results and manage those who are test positive? (6) Is the screening programme affordable and acceptable? (7) Has the effectiveness of the proposed screening programme been adequately tested, ideally in a randomised trial?

Applying these criteria to SCD in young competitive athletes reveals the profound difficulty of implementing an effective screening programme. To begin with, we are dealing with a problem that attracts attention more because of its emotive power than the frequency with which it occurs. Studies on recreational runners indicate approximately one fatality for every one million man (or woman)-hours of exercise per year, or roughly one death per 25,000 to 50,000 per year. Given the fact that the majority of these deaths in athletes aged >40 are caused by coronary artery disease, the screening programmes proposed for the younger age groups that comprise the majority of competitive athletes are targeting proverbial needles in haystacks. Thus, the most recent UK NSC report estimated that HCM accounts for only 15% of deaths in people aged <55, and even if it was the only cause of SCD in competitive athletes, its natural history is so variable—and often entirely benign—that its identification by screening in asymptomatic individuals provides no certainty that it will have adverse consequences. Indeed, many patients with HCM—asymptomatic or symptomatic—live long lives, eventually succumbing to other disorders. But HCM is by no means the only cause of SCD in competitive athletes; there is a long list of other causes which includes aortic valve disease, myocarditis, dilated cardiomyopathy, arrhythmic right ventricular cardiomyopathy, long QT syndrome and channelopathies.3 The screening requirement for an accurate test cannot therefore be easily met, while the requirement for effective treatment involves only the withdrawal of the athlete from sport, even though there is, in most cases, no certainty that this will provide protection against SCD. It should not be forgotten that athletes who screen positive have invariably dedicated years of training to their chosen sport without adverse consequences. Before wholesale screening is adopted, therefore, it needs to be demonstrated that the health benefits exceed those that can be delivered by better prevention through, for instance, deployment of defibrillators in sporting arenas and training staff in their use, by referral and treatment of symptomatic cases or by targeted screening of athletes in whom there is a family history of sudden death. It should not be overlooked that screening is very prone to unintended consequences, as evidenced by the neuroblastoma screening programme in Japan which was associated with the deaths of more children than before the implementation of the programme.

The statistical arguments that need resolving before implementation of a screening programme for competitive athletes are well illustrated by the deliberations of the UK NSC. Based on the premise that screening is designed to exclude at-risk athletes and thereby avoid SCD, and using the following assumptions in their most recent review, the programme would need to target the estimated 500 000 young people aged 10–19 who are active in UK sport at school or club level, among whom 20 SCDs would be expected. Suppose the 12-lead ECG is chosen as the screening test, the programme would immediately generate 500 000 new ECGs per year, of which an estimated 9% (54 000) might be ‘positive’ mostly with abnormalities consistent with, not diagnostic of, a disease related to SCD. Applying data from Italy, an estimated 2% of the total screened population totalling about 1000 per year would find themselves disqualified from competitive sport, all of whom need advice and treatment and most of whom will lead long event-free lives. The expectation of the screening programme must be that these 1000 young people disqualified from sport will include the 20 who would have experienced SCD and, moreover, that disqualifying them from sport will protect them from SCD. Yet we know from high profile cases such as that of Fabrice Muamba who, like all Premiership football players, had undergone full cardiac screening before he collapsed during a match earlier this year, that ECG and other screening tests are not perfect and it is possible to be screened several times and still to collapse and die during sport or relaxation. Conversely, screening in the UK can be expected to exclude from sport more than 900 young people per year in whom there was never any risk of SCD.

It is clear from the assumptions presented in the previous paragraph that, in straightforward feasibility and effectiveness terms, so little is known about many of the disorders potentially responsible for SCD and even less about the diagnostic accuracy of simple screening tests that a screening programme that meets the classic criteria of Wilson and Jungner is simply not possible. But that is not the end of the matter. How do we define an athlete and how do we obtain consent? Could a club insist that all athletes are screened and what would be the consequences of withholding consent in this situation? Once a screening result is available, who will have access to that information? Would the results be the exclusive property of the athlete or would they be passed on to parents, coaches, sporting bodies and team doctors? Will insurers expect to be informed?

Given the uncertainty surrounding the quality of the screening process and the consequences for young people, would it not be better for athletes to take their chances and accept the minute risk of SCD? This has a compelling rationale since the screening procedures currently on offer are, at an individual level, more likely than not to result in unnecessary disqualification from the sport. Moreover, the public health argument that, at the population level, SCD in competitive sport might be made even more rare by comprehensive screening programmes, is diluted by the fact that disqualifying so many people from sport is to rob them of one of the most effective means of protection against coronary artery disease which dwarfs all other causes of SCD in middle and later life.

Before we use millions of pounds of public resources and give (slightly) educated guesses about individuals’ futures, perhaps we should learn more about the array of conditions that contribute to SCD in young people and apply this knowledge to develop screening programmes that will reliably identify only those people for whom participation in competitive sport is dangerous. Meanwhile, the UK NSC is re-examining the case for screening and will consult on its findings shortly.

References

View Abstract

Footnotes

  • 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