Introduction

Despite the proven long-term benefits of regular physical activity, the act of exercise places considerable stress on the body and cardiovascular system. Albert et al. (¹) demonstrated an 11-fold increased risk of sudden death during or immediately after exercise even in habitually exercising individuals.

While most cases of sudden death in middle-aged athletes are due to occult coronary artery disease (^29,30,36), there are several uncommon conditions that can trigger life-threatening arrhythmias in young but susceptible athletes.

The sudden death of a high school student, college athlete, or marathon participant is always a newsworthy event. Nevertheless, the lack of a mandatory reporting requirement in many countries has prevented precise estimates of rates of sudden death. A recent study by Marijon et al. (²⁶) found that the incidence of sports-related sudden cardiac death (SCD) in the general population during recreational sports is more common in France than previously reported but still relatively rare, i.e., 4.6 cases per million population per year. Because of the greater intensity of competitive sports, the risk of SCD in National Collegiate Athletic Association (NCAA) athletes is considerably higher and likewise has been underestimated in studies to date, with recent estimates suggesting an overall incidence of 1:43,770 participants per year. In Division I NCAA basketball players, however, the rate was 10-fold higher at 1:3,100 per year (¹⁹).

In order to develop a better understanding of the risk of SCD, we will review the effects of exercise on the electrophysiological system of the heart. The conditions that affect cardiac electrophysiology that cause or increase the risk for cardiac events then will be discussed. Finally, we will discuss approaches to screening of athletes to reduce the risk of cardiac events and SCD.

Cardiovascular Effects of Exercise

Vigorous exercise has profound effects on the heart. The autonomic nervous system exerts extensive control over cardiovascular hemodynamics during exercise, producing a short-term state of sympathetic augmentation and parasympathetic withdrawal (⁹) (Fig. 1). Norepinephrine principally is released from sympathetic neurons that innervate the heart, while epinephrine is secreted by the adrenal medulla (¹⁷). Both neurotransmitters cause a significant improvement in left ventricular function, and removal of parasympathetic tone further augments stroke volume and ejection fraction (⁴⁶).

Effects of Exercise on the Cardiac Myocyte

The cardiac myocyte is the smallest unit of cardiac mechanical action, and stimulation by neurally released norepinephrine of the β-adrenergic receptor on the surface of the myocytes is the most important mediator of cardiac performance during exercise. Stimulation of the β-adrenergic receptor leads to phosphorylation of a broad array of calcium-handling proteins by activated protein kinase A (PKA), particularly the membrane calcium channel (I_Ca,L), which then allows for increased Ca²⁺ entry into the muscle cell during the plateau phase of the action potential (Fig. 2). Parasympathetic stimulation of muscarinic receptors attenuates the effects of sympathetic stimulation, preventing excess Ca²⁺ entry and overload, a potentially arrhythmogenic event (⁵²). This localized concentration of calcium triggers a subsequent release of significantly greater quantities of calcium from the sarcoplasmic reticulum, a process known as calcium-induced calcium release. This process results in a >50-fold increase in intracellular calcium, an essential step for excitation-contraction coupling in myocardial cells through binding of calcium to troponin C (⁴). β-adrenergic stimulation also causes faster calcium reuptake by the sarcoplasmic reticulum and extrusion from the cell by the membrane-bound sodium-calcium exchanger (⁵¹). This combination of events allows for faster diastolic relaxation of cardiac muscle. The rate of spontaneous depolarization of the sinus node also seems to be determined by calcium release from the sarcoplasmic reticulum coupled to sodium-calcium exchange in a manner roughly analogous to the process that drives excitation and contraction coupling in muscle (⁴⁹). This process similarly is modulated by β-adrenergic and muscarinic stimulation. Finally, depolarization of cells of the cardiac conduction tissue in the atrioventricular (AV) node is mediated by phosphorylation-activated I_Ca,L channels and therefore modulated by sympathetic and parasympathetic activity. Thus, β-adrenergic stimulation has profound effects on the increased rate of contractility, peak force, increased rate of relaxation, and heart rate.

Normal homeostatic mechanisms, including negative feedback by cytosolic calcium on I_Ca,L and sarcoplasmic release channels, parasympathetic suppression of PKA activity, and activation of balancing outward membrane currents, are required to prevent β-adrenergic stimulation from inducing calcium-associated arrhythmias (⁴). Catecholaminergic polymorphic ventricular tachycardia (CPVT) is an inherited condition associated with adrenergically mediated spontaneous calcium release from the sarcoplasmic reticulum resulting in exercise-associated arrhythmias (⁵⁰). Activation of β-adrenergic receptors leads to production of cyclic adenosine monophosphate (cAMP) and PKA-dependent phosphorylation of the I_Ca,L channel, resulting in an inward current that would prolong the action potential if unopposed. PKA also phosphorylates the slow component of the delayed rectifier current, I_Ks, whose outward current acts to shorten the action potential and QT interval. Imbalance between the competing actions of these I_Ks and I_Ca,L channels will destabilize both the cellular action potential and the QT interval measured on the electrocardiogram (ECG). During exercise, healthy individuals experience a progressive shortening of the QT, but drugs, underlying electrolyte deficiencies, or congenital abnormalities in channel function may result in a failure of the QT to shorten appropriately with exercise and a significant risk of arrhythmia.

Resting bradycardia is the rule in conditioned athletes, and this reflects heightened parasympathetic tone. In the resting state, a conditioned athlete maintains cardiac output at lower heart rates due to higher stroke volumes. They will manifest a reduction in resting sympathetic tone with an increase in parasympathetic (vagal) tone. Both exercise and deconditioning reverse this trend and result in increased sympathetic tone. Heightened vagal tone explains many innocent arrhythmias and conduction alterations that can be present in the conditioned athlete but only rarely may result in symptomatic bradycardia necessitating a period of deconditioning (³⁸).

The Athletic Heart

In the conditioned athlete, a spectrum of cardiac morphologic adaptations is seen in response to the volume and pressure effects on the heart over time (²⁷). Chronic elevations in cytosolic calcium activate signaling pathways that cause cardiac hypertrophy in animal models, and likely explain some of the adaptations seen in athletes (³⁵). The extent of adaptation varies with the type of exercise (dynamic vs isometric), as well as the duration and frequency of training. Cycling and rowing, which combine aspects of both exercise types, result in the greatest increases in left ventricular chamber volume and wall thickness (⁴⁰). While some degree of chamber enlargement is common, only 2% of elite athletes manifest wall thickness consistent with significant hypertrophy, i.e., >13 mm (⁴⁰). These morphologic changes do not play a direct role in the processes that lead to SCD, but they potentially may contribute to some of the common ECG findings in athletes. Overall, it is estimated that 40% of highly trained athletes have abnormal baseline ECG (³⁹), and many of these are reflective of the underlying response to vagal tone and athletic conditioning. Differentiating pathologic from physiologic hypertrophy may require ultrasound evaluation by an experienced echocardiographer (⁴⁵) and, in unusual cases, may necessitate a period of deconditioning to confirm the reversibility of increases in chamber dimensions. Palpitations caused by premature ventricular contractions are very common in trained athletes and seem to have little prognostic significance if less than 2,000 per 24-h period in the absence of nonsustained ventricular tachycardia (NSVT) and structural heart disease (⁵). In the case of NSVT, expert cardiologic evaluation is indicated, and a period of deconditioning also may be required to differentiate benign arrhythmias from those associated with a poor prognosis. Premature beats in the presence of significant structural heart disease are not benign and may trigger life-threatening arrhythmias.

Electrophysiologic Abnormalities, Arrhythmias, and SCD

Separate from normal adaptations in the conditioned athlete, pathologic conditions may affect the normal physiologic process of the cardiac exercise response adversely. Many of these conditions present with distinctive electrical patterns that are detectable on a standard 12-lead ECG. When present, they increase the risk of a cardiovascular event and SCD during exercise. Cardiac events and SCD associated with exercise typically originate as arrhythmias. The origins of cardiac arrhythmias can be divided into two general categories: disorders of electrical impulse formation (abnormal automaticity and triggered activities) and disorders of impulse conduction (reentry).

Triggered activities or abnormal depolarizations that are provoked by a previous beat are infrequently the cause of arrhythmias but, when present, may be associated with a potentially life-threatening diagnosis. Triggered arrhythmias are classified as either early after-depolarizations (EAD) or delayed after-depolarizations (DAD).

As the name suggests, EAD occur immediately following a depolarization but before repolarization is complete. They occur during phase 2 or 3 of the action potential (Fig. 2). Prolonged action potential duration (and consequently the QT interval) predisposes toward developing EAD. “Torsades de pointes,” a polymorphic ventricular tachycardia (VT) occurring in the presence of QT prolongation, is the feared consequence of EAD. Although torsades usually is self-terminating, it can result in recurrent syncope and can be a fatal rhythm if not recognized and treated properly. Any condition that prolongs the QT interval on a 12-lead ECG thus increases the risk for EAD and torsades de pointes.

DAD occur after repolarization of the cell and thus following the QT interval on an ECG. They occur as a result of elevated intracellular calcium levels and spontaneous release from the sarcoplasmic reticulum. Inherited disorders such as CPVT therefore lead to DAD. This could be exacerbated by tachycardia and sympathetic activity (such as with exercise) because these both further increase intracellular calcium levels.

Myocardial reentry is the underlying mechanism of arrhythmias in diseases associated with an abnormal structure of the heart such as Wolff-Parkinson-White (WPW), hypertrophic cardiomyopathy (HCM), arrhythmogenic right ventricular cardiomyopathy (ARVC), and infiltrative diseases (sarcoidosis, myocarditis, tumors). Reentrant circuits require the presence of heterogeneous refractoriness, as well as an area of slow conduction. In this setting, a premature stimulus pathway can initiate tachycardia if it falls in the refractory period of one area but in the excitable period of another. Typically, the resulting wave of depolarization travels down slow the pathway then propagates retrograde around the previously refractory region that has regained excitability during the delay afforded by the slow conduction. The slow conduction pathways are created through changes in myocardial architecture or electrophysiology. The sequence is the usual mechanism for initiation of ventricular and supraventricular tachycardias.

Clinical Disorders

Accessory Pathways

Accessory pathways are abnormal congenital electrical pathways that connect the atria to the ventricles outside of the normal conduction system. The presence of a potentially dangerous accessory pathway usually can be determined by a 12-lead ECG. The diagnosis of WPW syndrome is made when such a pattern, i.e., a PR interval less than 0.12 s and a “delta wave” (Fig. 3), is present in association with tachyarrhythmias. A relatively common problem, a WPW pattern is found in up to 0.3% of the population. In the absence of symptoms, the ECG usually is referred to as a WPW pattern. Competitive athletes manifesting the WPW pattern should be referred to a cardiologist for further evaluation because symptomatic patients are at increased risk of cardiac arrest due to a risk of atrial fibrillation with very rapid conduction to the ventricles via the accessory pathway (Fig. 4). In over 95% of cases, WPW syndrome can be cured effectively by catheter-based techniques (⁴⁴). Although definitive studies are lacking, current evidence suggests that SCD is rare (^{16,18,22,37,48}).

In cases of WPW pattern on ECG and absence of symptoms, the Bethesda Conference (^28,32) encourages electrophysiology testing only if moderate to severe activity is proposed. In the setting of symptoms likely attributable to WPW syndrome, additional testing is obligatory regardless of the sport. The European Society of Cardiology (ESC) guidelines take a more aggressive stance in general, mandating a thorough evaluation including invasive testing in all competitive athletes with the WPW pattern on ECG (¹¹).

Channelopathies

Clinical disorders characterized by abnormal function of cardiac ion channels are referred to as channelopathies. These disorders typically are inherited in an autosomal dominant manner but may be sporadic, so a family history of sudden death may or may not be present.

The understanding of this class of disorders has increased since the initial description of the molecular mechanism of the ion channel genetic mutation for long QT syndrome type 3 (LQT3) syndrome in 1995 (³). Multiple channelopathies have been identified, including the long QT syndrome (LQTS), short QT syndrome, CPVT, and Brugada syndrome.

Congenital LQTS are familial diseases with significant genotypic and phenotypic variation. The long QT interval reflects abnormal cardiac repolarization and prolonged action potential duration. There is potential for life-threatening arrhythmia, the characteristic polymorphic VT being torsades de pointes. All afflicted patients present with a similar clinical picture of an abnormally long QT interval on a standard 12-lead ECG. Three LQTS clinical syndromes are most common, accounting for >70% of cases. Long QT syndrome type 1 (LQT1), a defect in the gene encoding I_Ks, is most relevant to athletes. LQT1 manifests QT prolongation and a high incidence of arrhythmic events during exercise. Long QT syndrome type 2 (LQT2) is the result of loss of function mutations in outward potassium current channel I_Kr. Unlike I_Ks, the I_Kr current is not modulated by norepinephrine, and so, arrhythmias and QT prolongation do not occur typically with exercise. In LQT3, a defect in the SCN5A cardiac sodium channel gene leads to a gain of function mutation resulting in increased inward sodium current. LQT3 has a long ST isoelectric segment and SCD events that occur during sleep. Interestingly, “loss of function” mutations in this same gene are implicated in many cases of Brugada syndrome. The Bethesda Conference guidelines advocate the rate-corrected QT interval (QT/√RR) interval cutoff points of 0.47 s in male subjects and 0.48 s in female subjects; sports participation should be restricted to only low-intensity activities when these intervals are exceeded (²⁸). The ECS recommends a cutoff value of 0.44 s in males and 0.46 s in females and restricts asymptomatic gene carriers from competition (¹¹). The Bethesda Conference found insufficient evidence to substantiate competitive sport restrictions in clinically silent carriers of long QT syndrome (LQT) mutations.

CPVT occurs through defects in the process of calcium-induced calcium release from the sarcoplasmic reticulum of cardiac myocytes. This results in increased calcium leak during heightened sympathetic activity, such as what occurs with exercise. As previously discussed, elevated cytosolic calcium levels can lead to DAD triggering arrhythmias. The clinical presentation is one of exercise-induced syncope or cardiac arrest with a normal ECG and QT interval at rest (³⁴). The diagnosis is made by observing VT or high-grade ectopy during a stress test or adrenaline infusion. Genetic analysis is positive in 65% of patients with clear-cut phenotypes (⁴⁷). CPVT has a relatively high level of penetrance and expressivity, such that the majority of those with a mutation will have syncope or cardiac arrest by age 40. In a retrospective analysis, CPVT was the most commonly identified cause of aborted cardiac arrest, accounting for half of otherwise unexplained cardiac arrests (²³).

Brugada syndrome is an autosomal dominant disorder caused by mutations leading to abbreviation of repolarization. Symptomatic Brugada syndrome presents with a characteristic ECG pattern of >2 mm coved ST elevation in two or more right precordial leads (Fig. 5), and symptoms include syncope or cardiac arrest, frequently occurring during sleep. Sympathetic tone seems to suppress the Brugada pattern, and increased parasympathetic tone provokes greater ST elevation. Syncope or cardiac arrest therefore may occur immediately following exercise during recovery (¹⁵). The most common genetic finding is a “loss of function” mutation in the cardiac sodium channel gene SCN5A, leading to shortening of the action potential and slowing of cardiac conduction in the right ventricular outflow tract. SCN5A mutations represent approximately 25% of confirmed cases of Brugada syndrome, however, and so, the underlying defect remains a mystery in the majority. The prevalence is not well understood given the relatively recent description of the disease (⁷). Three primary ECG patterns have been identified, but only the type 1 pattern is considered diagnostic. In a multicenter study of more than 1,000 patients, a history of syncope, cardiac arrest, or a spontaneous (as opposed to drug induced) type 1 ECG was predictive of subsequent cardiac events (⁴²).

The diagnoses of CPVT and Brugada syndrome are a cause for significant exercise restrictions according to the Bethesda Conference, with the allowance for low-intensity sports in those with the latter diagnosis. Genotypic positivity alone should be a cause for restriction in these two entities. Conversely, these diagnoses are grounds for avoidance of all competitive sports, according to the European guidelines, even if there is only isolated genotypic positivity in the absence of ventricular tachyarrhythmias or symptoms.

Heritable Cardiomyopathies

Heritable cardiomyopathies predispose to SCD through alterations in the myocardial structure, which ultimately create a substrate conducive to reentry as previously described. Two of the most common of these conditions are HCM and ARVC.

HCM is the most common heart condition leading to SCD in young athletes in the United States. Maron et al. (²⁹) analyzed registry data collected over a 27-year period from 1980 to 2006, which included 1,866 athlete deaths. Of those, 56% were due to cardiovascular disease, and the most common finding was HCM (36%). The electrophysiologic manifestations of the disease are secondary to the morphologic changes in the myocardium from increased muscle mass, fibrosis, and fiber disarray leading to an increased risk for reentrant tachyarrhythmias. The ECG is abnormal in 75% to 95% of HCM cases, and it can identify asymptomatic individuals who are at risk for exercise-induced sudden death; affected individuals with a normal ECG seem to have a benign prognosis (³³). In Italy, where ECG screening is required for all athletes, the prevalence of HCM among athletes experiencing SCD was 2% (¹⁰) compared with 24% in the United States (⁸), where ECG screening is not routine. Given that the prevalence of HCM in the population is similar in the two countries (0.07% in Italy vs 0.1% in the United States), the fact that fewer athletes are dying with HCM suggests that ECG screening is an effective technique for identifying HCM and preventing SCD in athletes. It warrants mentioning that there is increased recognition that the QRS voltage in isolation correlates poorly with left ventricular mass in young athletes, and therefore, in the absence of other changes suggestive of hypertrophy, such as repolarization changes, atrial abnormalities, increased QRS width, and axis changes, this finding should not prompt further evaluation (^2,20,43).

ARVC is associated with myocyte necrosis and fibrofatty replacement of the right ventricular myocardium (and occasionally left ventricular myocardium), which leads to ventricular dilatation (²⁴). The replacement of myocytes with nonconducting elements creates an ideal substrate for reentrant tachyarrhythmias, and spontaneous VT and fibrillation significantly are increased in frequency. ARVC is the most common cause of SCD among athletes in the Veneto region of Italy, accounting for 22% of cases (³⁴). The diagnosis of ARVC is complicated by significant variability in penetrance and expressivity of the disease, and there is frequently a prolonged delay (average = 1 year) between presentation with palpitations or syncope and diagnosis (¹³). Ventricular arrhythmias with left bundle-branch block morphology (indicating a right ventricular source) are common and precipitated by exercise, with a clinical picture of syncope during exercise. Electrocardiographic diagnosis of ARVC is much more difficult than HCM. ARVC is associated with slow conduction through the right ventricle, and this can be associated with the presence of a so-called epsilon wave, deflections found between the QRS complex and the T wave in the right precordial leads. There is a clear association between ARVC and exercise-induced SCD, in which VT can degenerate into ventricular fibrillation (VF). Individuals with documented HCM or ARVC are excluded from most competitive sports, with the possible exception of those characterized by low static and low dynamic intensity, such as bowling or golf (^28,41).

Coronary Artery Disease

Atherosclerotic coronary artery disease accounts for the majority of SCD overall in the United States but is much less common in the young athlete. Age and traditional risk factors often identify those at risk for events. Anomalous coronary arteries, however, are found in a significant percentage of athletes experiencing cardiac arrest during exercise and in half of military recruits dying during basic training (¹⁴). A history of exercise-associated chest pain should prompt the clinician to consider coronary anomalies. Coronary ischemia leads to a state of increased intracellular calcium and enhanced heterogeneity of repolarization. Ischemic cells exhibit membrane depolarization, slowed conduction, and heterogeneous repolarization phases of the action potential depending on cellular adenosine triphosphate (ATP). Through the previously described process of reentry, a premature stimulus can trigger VT degrading to VF.

Trauma

Commotio cordis does not result from underlying cardiac abnormalities but is a common cause of SCD in young athletes. It is responsible for more than 20% of deaths in young athletes (²⁹). Blunt nonpenetrating chest blows may trigger VF if the blow is timed before the peak of the T wave. Awareness and recognition of the potential condition are important because survival has been related to prompt access to cardiopulmonary resuscitation and defibrillation (³¹).

Screening

In 2007, the American Heart Association (AHA) Council on Nutrition, Physical Activity, and Metabolism updated their original 1996 document (³²) regarding recommendations and considerations related to preparticipation screening for cardiovascular abnormalities in competitive athletes. These include taking a focused history screening for exertional pain or discomfort, unexplained syncope or near syncope and excessive exertional dyspnea and fatigue, and a family history of sudden death and performing a physical examination. A positive response to at least one item should be considered sufficient grounds for referral to a cardiovascular specialist. Conspicuously absent was a requirement for a screening 12-lead ECG. Unlike the AHA, the ESC (¹¹) and International Olympic Committee (^6,21) have recommended that all young athletes be screened with an ECG prior to competitive sports. The ECG has variable sensitivity and specificity depending on the condition being screened but provides significant additional data to the history and physical examination. It may provide unique diagnostic information in asymptomatic subjects in a number of the conditions responsible for SCD in athletes (Table). Indeed, the Italian experience suggests that ECG screening resulted in a highly significant 89% reduction in sudden death among athletes (from 3.6 per 100,000 person-years in 1979 to 1980 to 0.4 per 100,000 person-years in 2003 to 2004, P for trend < 0.001) (¹²).

Reasons for omitting the ECG from routine screening in the United States include the poor sensitivity for detecting LQTS, ARVC, and CPVT, as well as poor specificity for differentiating pathologic hypertrophy from training-related increases in QRS voltage. In addition, concerns have been raised regarding the sheer size of the population in question, as well as cost, logistics, and the lack of a cohort of health professionals available to perform and interpret the resultant ECG. Nevertheless, a recent report of more than 32,000 Illinois high school students subjected to ECG screening found a modest 2.5% incidence of abnormal findings (²⁵). These studies were performed at a cost of $8.67 per ECG. On the basis of this report, it seems that ECG may be applied to large populations in a manner that is cost-effective and avoids large numbers of unnecessary medical investigations. However, it must be acknowledged that depending on the criteria used to define an ECG as abnormal in the young athletic population, as well as the physician fee structure, among other considerations, the costs of a routine ECG screening program can vary significantly. The model described by Marek et al. (²⁵) is exemplary given its use of more strict ECG criteria and highly trained medical professional volunteers.

Role of the Clinical Cardiac Electrophysiologist

Cardiac electrophysiologists undergo 3 years of cardiology fellowship followed by 1 to 2 years of additional training in the diagnosis and therapy of arrhythmias. They have special expertise in treating the unusual conditions that can result in sudden death in athletes. Sports medicine physicians should consider referral to a cardiac electrophysiologist if any ECG abnormality listed in the Table is detected, in any subject with a history of exercise-associated syncope, if exercise-associated palpitations lasting more than seconds in duration are present, if there is a family history of unexplained sudden death or cardiac arrest attributable to an inheritable channelopathy or cardiomyopathy, or if there are palpitations the sports medicine clinician is not comfortable evaluating. In addition to the usual noninvasive testing common to all cardiologists, cardiac electrophysiologists are able to induce reentrant and triggered arrhythmias through direct stimulation of atrial and ventricular tissue. Exercise-associated arrhythmias may be induced using isoproterenol, epinephrine, or actual upper body exercise while the heart is mapped electrically. In many common conditions, such as WPW, other supraventricular tachycardias, and VT without structural heart disease, cardiac electrophysiologists are capable of curing the underlying problem with an invasive procedure aimed at the ablation of the triggering focus or reentrant circuit causing the arrhythmia. This approach succeeds in allowing the athlete to return to full activity in over 90% of cases (provided there is no residual structural heart disease or inducible arrhythmia) (³²).

Conclusions

Exercise has profound effects on cardiac electrophysiology, both acutely and chronically. The positive impact of neurally released norepinephrine on the flow of calcium into the myocardial cell, resulting in greater systolic calcium release from the sarcoplasmic reticulum, explains both the improved cardiac performance in healthy hearts and much of the propensity toward arrhythmogenesis in subjects with inherited channelopathies or cardiomyopathies.

 The views expressed in this article reflect the opinions of the authors only and not the official policy of the United States Army, United States Navy, Uniformed Services University, or the Department of Defense. The author declares no conflict of interest and does not have any financial disclosures.