Harms and benefits of screening young people to prevent sudden cardiac death
BMJ 2016; 353 doi: https://doi.org/10.1136/bmj.i1156 (Published 20 April 2016) Cite this as: BMJ 2016;353:i1156
- Hans Van Brabandt, researcher,
- Anja Desomer, researcher,
- Sophie Gerkens, economist,
- Mattias Neyt, health economist
- Correspondence to: H Van Brabandt hans.vanbrabandt{at}kce.fgov.be
- Accepted 15 February 2016
Sudden cardiac death of a young person on a sports field is a devastating event. Often these deaths are due to an unrecognised underlying heart condition, and screening has been proposed as a method to prevent them. However, disagreement remains about its benefits and harms.
In Italy, screening before participation in competitive sport has been mandatory since the 1970s. But the Netherlands abandoned its mandatory programme in 1984 because of the poor diagnostic performance of screening tests.1 In the UK, Cardiac Risk in the Young (CRY), a charity offering support and counselling to affected families, runs screening events, but the National Screening Committee decided last year not to recommend a population screening programme.2 Here, we summarise the findings of our literature review on the benefits and harms of pre-participation screening for non-professional athletes aged 14-34 years.3 We defined athletes as anyone participating in an organised team or individual sport that involves regular competition against others, places a high premium on excellence and achievement, and requires some form of systematic training.4
How common is sudden cardiac death in young healthy athletes?
Sudden cardiac death may be caused by a panoply of rare genetic and acquired heart disorders, present in 0.3% of the population.5 Most affected people are not aware of the disease and lead a normal life, but some develop symptoms such as dizziness, fainting, or breathlessness, which leads to a diagnosis. In around 1% of young athletes with unrecognised heart disease, sudden death will be the first and only manifestation of the disease,6 7 8 and it is these people that screening aims to detect. Although it is uncertain whether sudden cardiac death occurs more often during exercise than at rest, research has predominantly focused on athletes because exercise is considered a potential trigger.9
The exact frequency of sudden cardiac death in young athletes is not known. Numbers obtained from heterogeneous populations of general, college, and high school athletes indicate incidences varying between 1 per 917 000 and 1 per 3000.10 A recent review by the UK National Screening Committee suggested an annual incidence of 1 per 100 000 people aged 12-35.11 Most of those affected are men.12One UK study among athletes aged 12-35 years reported 64 deaths a year.13 In the UK National Audit of Sudden Arrhythmic Death Syndrome, a yearly average of 36 cases are registered.14 The charity CRY claims “at least 600” cases every year based on data from the Office for National Statistics,15 but this includes all people aged 1-34 years, whether or not they were athletes or had known heart disease.
Hypertrophic cardiomyopathy is the most common underlying cause of sudden cardiac death in young people, but the absolute risk of sudden death in people affected by the disease is low.16 17 In a prospective study in England, the annual incidence of sudden death due to hypertrophic cardiomyopathy was around 0.5-1 per million inhabitants.6
How accurately can pre-participation screening identify those at risk?
The most basic type of screening, as recommended by the American Heart Association,18 comprises taking a personal and family history in combination with a physical examination. However, very few people at risk of sudden cardiac death can be detected in this way.19 Of 115 young athletes who died suddenly and who had had a standard pre-participation medical evaluation, only four (3%) were suspected of having heart disease, and the abnormality responsible for the death was correctly identified in only one athlete (0.9%).20 To increase the diagnostic yield of screening, some experts advocate including a resting electrocardiogram (ECG).21 This may allow for better prediction of certain arrhythmias, such as those associated with the Wolff-Parkinson-White syndrome. However, athletes with a congenital anomaly of the coronary arteries or with premature coronary artery disease will not be identified by electrocardiography. Overall, 25% of people affected by a disease that may lead to sudden cardiac death would remain undetected.3 16 In other words, the sensitivity of a screening examination that includes electrocardiography is 0.75.
Correctly identifying athletes with a disease that may lead to sudden cardiac death (true positives) is only one side of the diagnostic medal. Screening may also lead to inappropriately suspecting disease in healthy people (false positives). In a recent US study of 1339 students, 916 (68%) reported at least one positive response on a questionnaire with heart health questions. After review by a physician, 421/1339 (31.4%) had at least one relevant response that required additional cardiac evaluation.22 Electrocardiography will pick up only some of the possible causes of signs and symptoms. For example, the resting ECG may be normal in someone with a heart murmur or who mentions that a syncope, palpitations, or chest discomfort has ever occurred during or shortly after exercise. Yet, these clinical findings may be a sign of a disease that can lead to sudden death and can therefore not be ignored. Based on data from small case series we estimated a specificity for screening including history, physical examination, and electrocardiography, ranging from 0.70 to 0.95.3 If screening is performed by experienced physicians, even with the application of the most stringent “Seattle” criteria for electrocardiography, 5% of healthy people will be suspected of having disease.23 24 25 In a more realistic scenario, up to 30% of those screened may be referred for additional testing after screening.19
The picture is further complicated by the clinical presentation. Most people with a cardiac disease are asymptomatic or have no specific clinical signs.26 27 28 Symptoms that may indicate serious disease (such as dizziness, syncope, palpitations, and chest discomfort) are aspecific and more often caused by other more benign conditions. Furthermore, there is a large overlap of the normal ECG and the ECG of someone with one of the index cardiac diseases. This overlap is even more pronounced in highly trained athletes, whose ECGs may show physiological alterations that mimic disease.
What is the benefit of pre-participation screening?
To reliably assess whether screening can reduce the number of sudden cardiac deaths in young athletes requires a randomised controlled trial.29 30 Unfortunately, no such study exists. The only evidence that screening saves lives comes from a population based study in Italy.5 31 A 90% reduction in the incidence of sudden cardiac death in young athletes was observed in the Veneto region over 26 years after mandatory screening was introduced in 1979 (from 3.6/100 000 person years to 0.4/100 000 in 2004; fig 1).31
Some experts have argued that this study does not provide evidence that screening was the cause of the fall in incidence.32 Since there is no unscreened control population, the fall may, for example, have been due to improved resuscitation or to an increasing proportion of female athletes, who are known to be much less prone to sudden cardiac death. Furthermore, the remarkably high incidence at the start of the study (3.6/100 000 person years) was calculated on no more than 14 cases and might represent mere random variation.33
The fact that the incidence of sudden cardiac death in countries with no mandatory screening such as the US (Minnesota) and France is no different from the most recently reported data from Italy (fig 1) further suggests that the observed fall in Italy was not the result of screening.12 34
What are the harms of screening?
Based on the diagnostic performance of screening discussed above, we can assume that out of one million young people screened, 50 000 to 300 000 will be identified as being suspected of having a disease that may lead to sudden cardiac death (see supplementary data on thebmj.com).
They will need additional cardiovascular testing, which may cause anxiety and psychological harm, unnecessary and unwarranted restrictions and disqualifications from sports, and impediments to insurability or employment opportunities.32 35 According to bayesian logic,36 a hyperoptimistic scenario assuming exceptional performance of downstream investigations (sensitivity of 1.0 and a specificity of 0.99 as shown in the supplementary table on thebmj.com) will end up with up to 5200 out of a million people screened being labelled for the rest of their life with a potentially lethal diagnosis. In the Italian pre-participation screening programme, 2% of people screened (20 000/million) were ultimately disqualified from participation in competitive sports.31
The most prevalent diseases detected at screening are the Wolff-Parkinson-White syndrome (550-2180/million),37 congenital anomalies of the coronary arteries (1000/million),16 38 and hypertrophic cardiomyopathy (100-790/million).37 Most people with these diseases will never experience any symptoms and lead a normal life if the disease is left undetected.26 38 39 There is no consensus on treatment in asymptomatic individuals. Some argue that they do not need to be treated; others will proceed to catheter ablation or the insertion of an implantable defibrillator. However, the risk of death associated with these treatments is similar to that of sudden cardiac death in asymptomatic people.4 8 40 41
Call for open data
During our study of the literature on screening, we noticed that the Italian researchers who published the Veneto study passionately promote screening. Through repetition of their study results in secondary papers, and their participation in guideline development groups and consensus meetings, they have had an important role in convincing sports physicians and sports federations around the world to introduce pre-participation screening.31 42 43 44 45 In their seminal paper they concluded that “The incidence of SCD in young competitive athletes has substantially declined in the Veneto region of Italy since the introduction of a nationwide systematic screening.”31 However, over the years, and with no new evidence, the wording has gradually changed and increasingly suggests a causal relation. In a 2013 paper they wrote: “Pre-participation screening based on a 12-lead ECG is effective in identifying athletes with potentially lethal cardiovascular disease and saves lives.”43 In addition, the Italian investigators claim that guidelines from the European Society of Cardiology (ESC) recommend pre-participation screening with electrocardiography.44 46 47 48 However, the document they refer to is in fact a “proposal for a common European protocol” of which they are the principal authors.21 Furthermore, they suggest that screening is cost effective by referring to economic models that used their data as input variables.42
Some of the concerns about the Veneto study might be clarified if the Italian investigators would provide access to additional unpublished data. Although pre-participation screening is mandatory throughout Italy, only data from one region (Veneto) have been published. Furthermore, since screening in Italy became mandatory in 1971, data collected before 1979 should be available showing whether the unexpectedly high initial incidence was simple random variation. Italian data obtained after 2004 have also not been published. These missing data would allow better understanding of the consistency of the initial findings.
Researchers from the UK tried to obtain additional data from the Italian screening programme through a formal request from Jeremy Hunt, the secretary of state for health for England, to the minister of health in Italy, to no avail.11 Our requests to the Italian researchers in June 2014 and January 2015 also remained unanswered.
Conclusion
The effectiveness of pre-participation screening in reducing the number of sudden cardiac deaths among young athletes has not been substantiated by solid evidence. Its potential to reduce deaths is likely to be low because of the poor detection rate and the uncertain effectiveness of the management of the diseases thus identified in asymptomatic people. Pre-participation screening induces harm because of the high number of false positive test results leading to temporary or lifelong disqualification from competitive sports, psychological and financial harm, and medical follow-up and treatment with unknown benefit.3 18 36 In addition, it leads to young people avoiding exercise known to be beneficial to their overall health. As long as those at high risk of sudden death cannot reliably be identified and appropriately managed, young athletes should not be submitted to pre-participation screening.
Key messages
An estimated 0.001% of young athletes die suddenly every year, and pre-participation screening aims to detect heart disease that puts them at risk
Up to 30% of those screened may be referred for additional testing, leading to unnecessary anxiety, overdiagnosis, overtreatment, and exclusion from sport
Screening would not detect around 25% of those at risk of sudden death and no randomised trial has examined its effectiveness
On balance cardiovascular screening in young athletes is likely to be more harmful than beneficial
Footnotes
Contributors and sources: This article is the summary of a Health Technology Assessment on cardiovascular pre-participation screening in young athletes, produced by the Belgian Health Care Knowledge Centre (KCE) and commissioned by the Belgian authorities. The full text can be accessed online (https://kce.fgov.be/publication/report/cardiovascular-pre-participation-screening-in-young-athletes). HVB is a cardiologist with 30 years’ clinical experience. He is a former director of the Belgian Centre for Evidence-Based Medicine and Cochrane Belgium. AD has a master’s degree in rehabilitation sciences and has been involved as researcher in health services research projects and in the development of clinical practice guidelines. SG is involved in health technology assessments and health services research projects and is a member of the Health Systems and Policy Monitor network. MN contributed to the development of the Belgian guidelines for economic evaluations and budget impact analyses and the European guidelines for health related quality of life and utility measures. He is cofounder of ME-TA, which offers expertise to foreign non-profit organisations and provides training and education in health technology assessment and economic evaluations.
Competing interests: We have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.
Provenance and peer review: Not commissioned; externally peer reviewed.