Article Text

Download PDFPDF

Incidence of sudden cardiac death in athletes: where did the science go?
  1. Jonathan A Drezner1,
  2. Kimberly G Harmon1,
  3. Mats Borjesson2
  1. 1Department of Family Medicine, University of Washington, Seattle, USA
  2. 2Department of Medicine, Sahlgrenska University Hospital, Göteborg, Sweden
  1. Correspondence to Jonathan A Drezner, Department of Family Medicine, University of Washington, Box 354410, Seattle, WA 98195, USA; jdrezner{at}

Statistics from

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.

Accurate assessment of the incidence of sudden cardiac death (SCD) in athletes is necessary to shape appropriate strategies for its prevention. However, past estimates of SCD incidence vary widely and often utilise limited methodology for case identification.1 2

A recent study by Steinvil et al3 reports the incidence of SCD in Israeli athletes and claims that a mandated screening program including ECG has made no impact on the rate of SCD in athletes. The study received significant attention and has sparked further discord within the ‘ECG debate’. However, critical examination of the study methodology raises many concerns about the validity of the study conclusions.

Unfortunately, the science surrounding the incidence calculation in this study is significantly flawed. An accurate incidence calculation requires (1) a precise account of the number of SCD events during the study period (numerator) and (2) an accurate estimate of the population at risk (participating athletes per year – denominator). This study has neither.

Newspapers do not replace registries

The authors retrospectively searched media reports from two newspapers and assert that these identified all cases of SCD because the events are devastating. However, this claim is misguided and not substantiated by evidence. The study also states that the two newspapers reviewed covered 90% of the ‘readership’. Simply because the newspapers reached most of the country does not mean they would capture all or most of the cases. In fact, in a recent report by Harmon et al on the incidence of SCD in National Collegiate Athletic Association athletes, rigorous electronic searching of all media sources identified only 56% of SCD cases between 2004 and 2008.4 It is difficult to imagine that search of only two newspapers identified all cases of SCD over a 24-year period. The authors also suggest that even if the newspapers did not report all the cases, this underreporting would be the same throughout both time periods before (1985–1997) and after (1998–2009) ECG screening was mandated. Unfortunately, the authors provide no evidence to support this claim. Media coverage in 2009, with all of its technological advances, will differ significantly from media coverage in 1985. In a US report of SCD in competitive athletes, media reports were also used as the primary source of case identification; however, the authors acknowledge that more cases were identified in recent years due to heightened media attention.5 In addition, the study by Steinvil et al has only one accurate estimate of the number of athletes and that is for the last year of the study (1 of 24 years). The rest of the years are estimated by what the authors believe the ratio of athletes was to the population.

Systematic reporting systems are crucial

The only method to calculate an accurate incidence of SCD is through a systematic reporting system. Unfortunately, most countries, including the USA, do not have this. Despite the deficiencies in their study methodology, the authors criticise data from an Italian study using data from a mandatory registry for juvenile sudden death. They suggest that an abnormally high number of cases in the 2 years prior to implementing a national policy mandating ECG screening in Italian athletes may have falsely amplified the impact of the screening program when compared with later years. However, the slow (rather than abrupt) decline in the incidence of SCD in the Italian study suggests differently. It is far more likely that the Italian incidence calculations are accurate, and that the rate of SCD declined as physicians conducting the screenings gained experience in ECG interpretation to better identify individuals at risk. In fact, a separate analysis suggests that the Italian screening program effectively reduced the risk of SCD from hypertrophic cardiomyopathy, the leading cause of SCD in young athletes in the USA.6

The ‘unwritten’ mandate for preparticipation screening

ECG screening does not need to be mandatory, but it can be recommended for at-risk populations such as competitive athletes. In the USA and many other countries, young athletes are required to be screened before sports participation. There is already an unwritten mandate of sorts – schools and sports governing bodies will not allow athletes to participate unless they have passed their sports physical. However, the traditional recommendation for cardiovascular screening in athletes using history and physical examination has an extremely low sensitivity and is of questionable benefit when applied alone.7 8 Adding an ECG greatly increases the sensitivity to detect conditions at risk of SCD and has been shown to enhance the cost effectiveness of the screening process.9 10

Finally, it is tragic that only 2 of the 24 athletes survived their event – perhaps an opportunity to promote emergency planning and availability of automated external defibrillators (AEDs) at sporting venues. In the USA, >60% of high school athletes suffering sudden cardiac arrest are saved if prompt resuscitation occurs through school-based AED programs.11

The debate on ECG screening is quite polarised, and many questions remain unanswered. However, the methodology in the paper by Steinvil et al does not allow for any sound conclusions on the effectiveness of ECG screening. Given that we are 15 years from Sackett's first use of the term ‘Èvidence Based Medicine’, we are concerned that the debate on ECG screening could move so far away from real science and validated research. It is common to have study limitations – we all do – but the tone, discussion, and conclusions have to reflect the certainty of the results. To make profound claims on flawed science is simply misleading.


View Abstract


  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.