Article Text

Download PDFPDF

Incidence and aetiology of sudden cardiac death in young athletes: an international perspective
  1. M Borjesson1,
  2. A Pelliccia2
  1. 1
    SU/Ostra, Sahlgrenska Academy, Goteborg University, Goteborg, Sweden
  2. 2
    Institute of Sports Medicine and Science, Rome, Italy
  1. Correspondence to Associate Professor Mats Borjesson, Department of Medicine, Sahlgrenska University Hospital/Ostra, Goteborg University, Goteborg, Sweden; mats.brjesson{at}


The incidence of sudden cardiac death (SCD) among young athletes is estimated to be 1–3 per 100 000 person years, and may be underestimated. The risk of SCD in athletes is higher than in non-athletes because of several factors associated with sports activity that increase the risk in people with an underlying cardiovascular abnormality. A clear gender difference in the incidence of SCD exists in young athletes, with the risk in male athletes being up to 9 times higher than in female athletes. The most common causes of SCD in young athletes is underlying inherited/congenital cardiac disease, such as cardiomyopathies, congenital coronary anomalies and ion channelopathies. Blunt chest trauma also may cause ventricular fibrillation in a structurally normal heart, known as commotio cordis. Although geographical differences in the causes of SCD in young athletes have been reported, these disparities are more likely to be related to the type and implementation of pre-participation screening leading to the identification of athletes at risk, rather than reflecting a truly different ethiology. More studies are needed to clarify the role of ethnicity in the prevalence of diseases known to cause SCD in young athletes.

Statistics from

The favourable effect of regular physical activity for primary and secondary prevention of cardiovascular disease is well recognised.1 2 However, regular vigorous exercise is also associated with the occurrence of adverse events, including sudden cardiac death (SCD) in young and adult people with underlying cardiovascular disease.3 4 This review focuses on the incidence and aetiology of SCD in young athletes, with particular emphasis on the differences between countries, genders and ethnic origin and in relation to non-athletes.

Incidence of SCD

SCD is usually defined as death occurring within 1 h of the onset of symptoms in someone without a previously recognised cardiovascular abnormality,5 excluding respiratory, cerebrovascular and drug-related deaths. In many studies, however, a time frame of 24 h from the onset of symptoms until death is considered.6 7 8

The incidence of SCD in the general population (⩾35 years of age) is estimated to be 1 in 1000 persons per year.9 In young people (<35 years of age), the incidence of all sudden deaths (including non-cardiac causes) is 1.5–6.5 per 100 000 persons per year, and the incidence of SCD is 0.3–3.6 per 100 000 persons per year.7 8 10 11 12 13 14 15 The incidence varies between studies, probably due to the differences in the definition of SCD, study populations and death registration, making direct comparisons difficult. Specifically, in US high-school and college athletes (age 12–24 years), the incidence of SCD has been found to be 0.5/100 000 participants per year,14 whereas, in Italian competitive athletes (age 14–35 years), the incidence was found to be 3.6/100 000.15 According to the National Center for Catastrophic Sport Injury Research in the USA, the rate of SCD is fivefold higher for male than female high-school and college athletes (0.75 vs 0.13 per 100 000 athletes per year).14 Moreover, the estimated death rate is twice as high in college athletes (age range 20–24 years) as in their high school (age range 12–19 years) counterparts (1.45 vs 0.66 per 100 000 athletes per year).14

The discrepancy between the Italian and US figures may be explained in part by the different mechanisms used to identify cases of SCD in young athletes. In US studies, case identification has been predominantly through review of public media reports and other available electronic resources, making underestimation of the true incidence of SCD likely. The Italian data were collected by a prospective registry of juvenile sudden death but in a relatively limited geographic area. In addition, differences in the populations with regard to age (older athletes in the Italian series) and gender (larger proportion of female athletes in the US series) may have contributed to the differences, as will be discussed below.

Aetiology of SCD

In young athletes (<35 years of age), a broad spectrum of underlying cardiovascular abnormalities is responsible for SCD. The most common are cardiomyopathies such as hypertrophic cardiomyopathy (HCM), arrhythmogenic right ventricular cardiomyopathy (ARVC) and dilated cardiomyopathy (DCM), and congenital coronary artery anomalies (CCAAs).3 6 14 16 To a lesser extent, other abnormalities are responsible, including aortic rupture in the context of Marfan syndrome, myocarditis, valvular disease (aortic stenosis, mitral valve prolapse) and ion channelopathies (long QT syndrome (LQTS), Brugada syndrome, catecholaminergic polymorphic ventricular tachycardia), as well as blunt chest trauma causing malignant arrhythmia (commotio cordis).6 14 16 17

In athletes >35 years, the primary cause of death, as in the general population, is coronary artery disease.3

Structural cardiac abnormalities

Hypertrophic cardiomyopathy

HMV is the single most common cardiovascular cause of SCD in young athletes in the USA.17 It is a primary and familial cardiac disease with heterogeneous clinical and morphological expression and a diverse clinical course. The prevalence is about 1/500 in the general population18; in screened athletes, the prevalence is estimated to be 0.07%19 to 0.08%.20 Morphologically, HCM in young people is characterised by a spectrum of alterations, including progressive left ventricular hypertrophy and myocardial disarray, and, clinically, by a variety of symptoms such as pre-syncope, syncope and palpitations. However, sudden death is most common in young people with HCM (<age 30 years) who have usually been asymptomatic (or only mildly symptomatic).21 SCD from HCM is usually caused by primary ventricular tachycardia/fibrillation and can be the first clinical manifestation of the disease, usually in the context of exercise and sport participation.

Congenital coronary artery anomaly (CCAA)

CCAAs are a group of congenital disorders with variable clinical manifestations. The CCAAs most commonly responsible for SCD occur in the coronary arteries originating from the wrong sinus of Valsalva,22 23 namely, the left coronary artery originating from the right sinus, or the right coronary artery originating from the left sinus, which represent <1% of consecutive patients studied by angiography.23 Young people with CCAA may die suddenly as the first manifestation of their abnormality, although some may experience angina, syncope or palpitations. The vast majority of these symptoms are related to exertion. Myocardial ischaemia is precipitated by exercise because of critically impaired coronary flow due to an abnormal, slit-like ostium of the anomalous coronary artery, compression of the anomalous artery between the pulmonary artery and ascending aorta or, possibly, coronary spasm triggered by endothelial dysfunction.23

Arrhythmogenic right ventricular cardiomyopathy/dysplasia

ARVC is an inherited heart muscle disease that predominantly affects the right ventricle and, less commonly, the left ventricle.24 Pathologically, ARVC is characterised by progressive death of myocardial cells with subsequent fibro-fatty replacement, and, clinically, by electrical instability leading to ventricular tachycardia and fibrillation responsible for SCD.25 In patients with ARVC, the risk of SCD is 5.4 times higher during competitive sports than during sedentary activity.16 In fact, ARVC is implicated in a quarter of fatal athletic field events recorded in Italy.16 Triggering mechanisms include increased afterload during exercise, which stretches the diseased myocardium resulting in ventricular arrhythmia and/or favouring re-entrant arrhythmic mechanisms. Also, “supersensitivity” to catecholamines due to damage of sympathetic nerve trunks by fibro-fatty replacement may contribute to ventricular arrhythmias.25 26

Dilated cardiomyopathy

DCM is a myocardial disorder characterised by left ventricular dilatation and impaired systolic function. It can be familial or genetic in origin, secondary to infection, inflammation, toxic substances or metabolic disorders, or related to long-standing uncontrolled hypertension. The prevalence of DCM in the general population is 4/10 000 persons.27 28 Morphologically, the left ventricular cavity is disproportionately increased in size and round-shaped, with impaired systolic function and reduced stroke volume, as well as increased end-diastolic chamber pressure.

Life-threatening ventricular and supraventricular tachyarrhythmias usually represent the clinical presentation of this disorder, before left ventricular dysfunction is clinically evident,27 28 and SCD is most common in young people with DCM engaged in physical exercise and sports.


Myocarditis is usually caused by viral infections of the myocardium, accounting for 3–7% of SCDs related to sports.14 15 17 This entity was originally considered to be the main cause of SCD in young athletes and suggested as the cause of an upsurge of SCDs among Swedish orienteers in the 1980s.29 At present, myocarditis is more often found in military recruits (10–20%) than in the general population.8 30 These findings may reflect the specific environment of the soldiers, possibly carrying an increased risk of, and proclivity for, concomitant infections.


Less common structural cardiac abnormalities responsible for sudden death in young athletes (accounting for 3–8%)3 14 15 include aortic dissection and rupture (usually in the context of Marfan syndrome), sarcoidosis, valvular heart disease (eg, mitral valve prolapse or aortic valve stenosis) and atherosclerotic coronary artery disease.

Electrical disorders

Wolff–Parkinson–White syndrome

Wolff–Parkinson–White syndrome is defined as the presence of paroxysmal arrhythmias in a patient with ventricular pre-excitation due to an accessory pathway with anterograde conduction.31 The prevalence of pre-excitation in athletes is 0.1–0.3%, similar to that of the general population.32 Most patients with pre-excitation remain asymptomatic, but, when symptoms do occur, they are usually secondary to tachyarrhythmias. Patients with ventricular pre-excitation may develop atrial fibrillation, which could lead to ventricular fibrillation and SCD as a result of rapid anterograde conduction over the accessory pathway. Determining the electrical properties of the accessory pathway is crucial for establishing the risk of SCD in the single athlete.31


Different arrhythmogenic diseases defined as channelopathies (congenital long or short QT syndrome, Brugada syndrome, catecholaminergic polymorphic ventricular tachycardia) have in common the inherent risk of life-threatening arrhythmias in affected young people.33 34 Pathophysiological mechanisms include mutations in transmembrane ion channels or proteins involved in intracellular calcium handling.

Congenital long QT syndrome

LQTS is manifested by symptoms (syncope is the most common) and electrocardiographic changes due to mutations in ion-channels involved in the maintenance of the cardiac action potential plateau and repolarisation. Electrocardiographically, LQTS is defined as a corrected QT interval ⩾440 ms in men and ⩾460 ms in women,35 but intervals of ⩾470 ms and ⩾480 ms have been suggested,34 illustrating the difficulty of defining an optimal cut-off value. Therefore, other ECG and clinical factors are usually taken into consideration, using various scoring systems, for diagnosis. Extreme QT prolongation (usually >500 ms) predisposes to torsade de pointes and ventricular fibrillation.36 The number of cases of SCD in young athletes estimated to be due to LQTS varies from 0.5% to 8%.15 17 37

Brugada syndrome

This is an autosomal dominant, cardiac sodium channelopathy, which clinically presents with syncope (due to polymorphic ventricular tachycardia) or less commonly as ventricular fibrillation leading to SCD.38 An ECG pattern of J-point and ST-segment elevation in the right precordial leads is typically seen.39 Mixed phenotypic expressions of the disease, ranging from distinct repolarisation abnormalities to subclinical cardiac conduction defects, also occur. Increased vagal tone induced by chronic athletic conditioning may enhance the propensity to die at rest,36 and hyperthermia during exercise may potentially trigger arrhythmias during strenuous activity.

Catecholaminergic polymorphic ventricular tachycardia

This involves mutations in the ryanodine receptor, calsequestrin or the ankyrin-B protein, which predispose to arrhythmias and recurrent syncope typically provoked by exercise or stress.33 The baseline ECG is unremarkable, but the exercise test may show multifocal ventricular premature beats or ventricular tachycardia with alternating QRS axis, ie, “bidirectional ventricular tachycardia”.36 This adrenergically dependent arrhythmia is highly lethal, typically before the age of 20–30 years, and cases are estimated to comprise a substantial number of patients with familial arrhythmias without QTc prolongation or Brugada-type ECG abnormalities.

Commotio cordis

Blows to the chest, specifically to the precordium, can trigger ventricular fibrillation without causing structural injury to the ribs, sternum or the heart itself (ie, commotio cordis).40 Commotio cordis is typically caused by projectiles used in sports, such as hockey pucks or lacrosse balls, forcefully striking the chest, but also by more modest force (eg, a pitched baseball).41 Similarly, bodily contact (eg, a karate blow or players colliding) can be implicated. Experimentally, a single blow directly over the heart area, timed at 10–30 ms before the T-wave peak (ie, during the vulnerable phase of repolarisation), may induce ventricular fibrillation.42

SCD of unknown cause

A considerable subset of SCD, however, occurs in the absence of an evident structural or electrical cardiac disorder. In the study of Burke et al,6 6/34 sports-related SCDs were of unknown cause, and other studies found a “normal heart” in 1–7% of SCD cases.14 15 43 As channelopathies do not present any distinct morphological abnormalities, it is possible that these electrical disorders may be responsible for a number of SCDs previously considered as unexplained.37

Differences in SCD between athletes and non-athletes


The incidence of SCD is higher in competitive athletes than in non-athletes, with an estimated relative risk of 2.8.16 It is worth clarifying that the sport activity itself is not responsible for the greater incidence of SCD in athletes, rather it may be the combination of intense exercise and an underlying cardiovascular disease that triggers arrhythmias leading to cardiac arrest. The relative risk associated with sport is different according to the pathological substrate and appears to be greater in people with cardiomyopathies (eg, HCM or ARVC) or CCAAs.16

Several factors, in the presence of an underlying cardiovascular disease, may contribute to the increased risk of SCD in athletes, including the high release of catecholamines, increased platelet aggregation/adhesion, dehydration and electrolyte disturbances associated with exercise, as well as the potential concomitant use of drugs/doping.44 Hence, it is likely that cardiovascular risk may be higher in sport activities that require greater cardiovascular work.


Compared with non-athletes, cardiomyopathies and coronary anomalies are more common causes of SCD in athletes.6 16 37 Conversely, atherosclerotic heart disease is seldom the cause of SCD in young athletes, but is the most common cause in adult and senior athletes.6 37 46 A large subset of SCD in young people occurs without evidence of structural cardiac abnormalities, indicating a high incidence of arrhythmogenic disorders.37

These observations are the rationale for the implementation of pre-participation cardiovascular screening in young athletes, as structural cardiac diseases such as cardiomyopathies are detectable during life, if appropriately searched for.47 48 The importance of timely identification by screening of asymptomatic athletes with cardiomyopathies is illustrated by the real possibility of preventing SCD by lifestyle modification, including restriction of competitive sports activity,15 concomitant prophylactic treatment or implantable cardioverter defibrillator therapy.49

Differences in the incidence of SCD between different sports

Some sports are more often implicated in sports-related SCD than others. For instance, in the USA, basketball is the sport most often associated with SCD,3 whereas in the rest of the world soccer is most often involved.50 51 The reason for this difference is probably related to the size of the athletic population participating in various sports in different areas. An additional explanation may include the level of cardiac work required by different sports. It is reasonable that sports with higher cardiovascular requirement—for example, soccer compared with riflery—confer a higher risk. Finally, more male than female athletes participate in certain sports, which also may contribute to the different incidence of SCD (see below).

Geographical differences in the incidence of SCD

According to available studies, cardiac diseases responsible for SCD seem to differ according to the geographical area. For example, HCM is more common in US than Italian studies (30% and 1% of SCDs, respectively).3 16 Conversely, ARVC is much more common in Italian series, causing 22% of SCDs19 and only 1–3% in the USA.3 14 Comparisons between Italian and US findings, however, show a similar prevalence of HCM in non-sport-related SCDs.3 15 Consistently, the prevalence of HCM in the general population is similar in Italy and the USA.18 19

The lower incidence of SCD due to HCM in Italian studies therefore cannot be ascribed to ethnic and geographic differences in HCM; rather, it is reasonably explained by the exposure of the athletic population in Italy to systematic pre-participation screening, which leads to identification and disqualification of people with HCM,15 thereby reducing the proportion of SCDs on the athletic field due to HCM. As a consequence, other cardiovascular conditions, such as ARVC and CCAAs, have become responsible for a greater proportion of all sudden deaths in Italian athletes. In other countries in Europe without mandatory screening, such as Sweden52 and France,53 HCM is still a very common cause of SCD.

Differences in the incidence of SCD between ethnic groups

Some of the underlying cardiovascular abnormalities, such as HCM, are more common in the black population (0.24% in black people vs 0.10% in the Caucasian population),18 and therefore the occurrence of HCM-related SCD is expected to be higher in black athletes. For a long time, however, little was known about the rate of SCD in athletes of different ethnicity. Basavarajaiah et al54 recently reported that black athletes have greater cardiac hypertrophy than Caucasian athletes, as well as associated ECG changes, suggesting a possible proclivity for a greater incidence of SCD. There is also evidence that Brugada syndrome may be more common in Asian than Caucasian populations,55 but systematic studies addressing the causes of SCD in Asian, African and South American populations are still lacking.

Differences in SCD related to gender


The incidence of SCD is higher in male than female athletes.6 11 19 43 56 For instance, Corrado et al16 reported an incidence of 2.62 per 100 000 male athletes per year and only 1.07 per 100 000 female athletes per year. Likewise, van Camp et al14 found a fivefold higher incidence of sudden death in male than female high school and college athletes. Also, only a small proportion (7–9%) of SCD cases were found to be female athletes by Sack.57

The gender difference may be related to several factors, including a lower prevalence of underlying cardiac abnormalities and/or different cardiovascular adaptations to intense exercise training in women.58 Furthermore, in a global context, women participate less in high-level competitive sports, which may explain part of the difference.


The cause of SCD may differ between men and women, which may reflect a different prevalence of underlying cardiac abnormalities, such as HCM (0.26% incidence in men vs 0.09% in women).18 In addition, women have a lower prevalence of “athletes heart”,58 and it is not clear if the lower extent of cardiac remodelling, in the presence of underlying cardiac disease, may in fact represent a measure of protection, lowering the risk of SCD.

Role of doping?

The widespread use of doping among athletes has to be considered in the light of the high proportion of SCD of unknown origin in young people.37 However, the incomplete information that we have on the true dimensions of drug misuse and doping among athletes limits our ability to draw any clear conclusions. Many of the substances commonly misused (anabolic steroids, growth hormone, erythropoietin) have the potential to increase cardiovascular risk in athletes by inducing potentially lethal ventricular tachyarrhythmias, especially in extreme circumstances such as physical exhaustion, difficult environmental conditions and extreme stress.59


The incidence of SCD among young athletes is estimated to be 1–3 per 100 000 person years, is higher than in non-athletes, and may possibly still be underestimated.60 Several factors associated with sports activity increase the risk of SCD in people with an underlying cardiovascular abnormality. The most common cause of SCD in young athletes is underlying inherited/congenital cardiac disease, such as cardiomyopathies, congenital coronary anomalies and ion channelopathies. There is a clear gender difference in the incidence of SCD in young athletes, with male athletes having a higher risk than female athletes. Geographical differences in SCD in young athletes are more likely to be related to differences in pre-participation screening programmes rather than a truly different incidence of cardiovascular disease. Finally, more studies are needed to clarify the role of ethnicity and the risk of different diseases that cause SCD in young athletes.

What is already known on this topic

  • The incidence of sudden cardiac death (SCD) in young athletes is 1–3/100 000 person years.

  • The most common cause of SCD in athletes <35years of age is underlying congenital/inherited cardiac disease.

What this study adds

  • There is a clear gender difference in the incidence of SCD in young athletes, with male athletes having a higher risk.

  • Previously reported geographical differences in the incidence of SCD may be due to differences in pre-participation programmes and study designs.



  • Competing interests None.

  • Provenance and peer review Commissioned; not externally 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.