Inherited cardiomyopathies and sports participation

Competitive sports activity is associated with an increased risk of sudden cardiovascular death (SCD) in adolescents and young adults with clinically silent cardiovascular disorders [2]. Inherited cardiomyopathies are recognised leading causes of sports-related cardiac arrest in young competitive athletes, with hypertrophic cardiomyopathy (HCM) accounting for one-third of fatal cases in the USA and arrhythmogenic right ventricular cardiomyopathy for approximately one-fourth in Italy [3‐5]. Preparticipation evaluation may allow the identification of asymptomatic athletes who have potentially lethal cardiomyopathies and protect them from the risk of SCD through restriction from competitive sports (Fig. 1; [6, 7]). However, many affected athletes aspire to continue practicing leisure-time sports activities or at least maintain a physically active lifestyle despite their non-eligibility to engage in competitive sports.

This article will review the available data on the sports-related risk of SCD and the recommendations regarding the eligibility to engage in sports activity, either competitive or recreational, of individuals affected by inherited cardiomyopathies. We will focus on major inherited cardiomyopathies which have been implicated as leading causes of SCD during sport activity and/or require a specific clinical work-up for differential diagnosis with physiologic remodelling of the athlete’s heart. We refer for sports participation with inherited channelopathies to Panhuyzen et al. in this same issue [8].

Hypertrophic cardiomyopathy, the most common inherited heart muscle disease, is caused by mutations of genes encoding for sarcomeric proteins. The phenotype of HCM is characterised by left ventricular (LV) hypertrophy, either symmetric or asymmetric, which occurs in the absence of abnormal loading conditions. The diagnosis traditionally relies on the demonstration by echocardiography of otherwise unexplained diffuse or segmental LV wall and/or septal thickening ≥15 mm; a lower cut-off value (≥13 mm) is used in the context of family members of affected individuals, especially if they are mutation carriers [9, 10]. Cardiac magnetic resonance imaging may provide additional diagnostic and prognostic information given its ability to better detect segmental LV hypertrophy in some regions, such as anterior free wall, posterior septum or apex, which are not visualised adequately by echocardiography, to provide more accurate wall thickness measurements and to identify areas of intramyocardial late gadolinium enhancement/fibrosis. Affected patients are at an increased risk of ventricular arrhythmias/SCD, heart failure, and thromboembolism [11].

Left ventricular hypertrophy may develop as a consequence of heart adaption to physical exercise (athlete’s heart) and is particularly prominent in athletes involved in endurance sports activities, such as rowing and cycling [12], but maximal wall thickness rarely exceeds 13 mm [13‐15]. A minority of highly trained athletes, particularly males of African/Afro-Caribbean descent, may exhibit more pronounced LV hypertrophy (13–16 mm), which requires an accurate clinical work-up for differential diagnosis with HCM [13‐15]. In addition, in African/Afro-Caribbean athletes the electrocardiogram (ECG) often demonstrates a variant of the anterior early repolarisation pattern characterised by J‑point/ST-segment elevation and T‑wave inversion in the anterior precordial leads (V1 to V4) which may raise the suspicion of an underlying cardiomyopathy (Fig. 2; [16, 17]). A LV wall thickness >16 mm should be considered diagnostic of HCM even in highly trained athletes, irrespective of ethnicity [18, 19]. In athletes with LV hypertrophy falling into the ‘grey zone’ (13–16 mm), differential diagnosis between HCM and athlete’s heart is based on several parameters including imaging features (cavity size, distribution of hypertrophy, outflow tract obstruction, mitral valve abnormalities, diastolic function, and late enhancement at cardiac magnetic resonance imaging with a characteristic pattern of patchy involvement [20]) and ECG findings (ST-segment depression, T‑wave inversion in infero-lateral leads, pathological Q waves and conduction disturbances) (Fig. 3; [18, 19]).

Hypertrophic cardiomyopathy is one of the leading causes of SCD among athletes in the USA where it has been reported to account for more than one third of fatal cases [3, 4]. By contrast, in European studies HCM is reported to cause less than 10% of SCDs ([5, 21]; Table 1). Besides ethnic and genetic differences between the populations, this discrepancy may be explained by the preventive identification of HCM through systematic preparticipation screening including ECG, which is abnormal in up to 90% of athletes with HCM (Fig. 4; [1, 22‐24]). ECG screening is common practice in most European countries and is compulsory in Italy, but it is not routinely performed in the USA.

High intensity competitive sports may promote life-threatening ventricular arrhythmias even in the absence of traditional risk factors and there is no evidence that algorithms developed to predict the risk of SCD in the general population of patients with HCM can be extrapolated to athletes. Accordingly, both European and American recommendations for sports eligibility agree that athletes with HCM should be restricted from competitive sports activity. Instead, participation in disciplines at low cardiovascular demand (such as bowling, golf and brisk walking) may be allowed to asymptomatic patients with mild LV hypertrophy (wall thickness <20 mm), no ventricular arrhythmias at exercise testing or ambulatory ECG monitoring and negative family history for SCD ([29, 30]; Table 2).

Arrhythmogenic right ventricular cardiomyopathy (ARVC) is an inherited cardiomyopathy characterised by progressive loss of myocytes and replacement by fibro-fatty tissue, which creates the substrate for life-threatening ventricular arrhythmias. In its classical form the phenotypic manifestations predominantly involve the right ventricle, but biventricular and left-dominant variants exist. Molecular genetic studies showed that ARVC is a genetic disorder resulting from defective desmosomal proteins, most often with an autosomal dominant pattern of inheritance [31, 32].

The clinical manifestations of the disease usually occur between the second and forth decade of life and consist of ECG depolarisation and repolarisation changes, typically localised in the right precordial leads (V1–V3/V4), structural abnormalities, such as global or regional dilatation/dysfunction of the right ventricle and, most importantly, ventricular arrhythmias with a left bundle branch block morphology, ranging from isolated premature ventricular beats to sustained ventricular tachycardia, which can degenerate into ventricular fibrillation [33, 34].

The diagnosis of ARVC requires a combination of different criteria from various categories such as 1) histopathological abnormalities at endomyocardial biopsy; 2) morpho-functional abnormalities consisting of regional right ventricular wall motion abnormalities (regional akinesia, dyskinesia or bulging) plus right ventricular dilatation or global right ventricular dysfunction; 3) depolarisation abnormalities such as epsilon waves, delayed S‑wave upstroke in V1–V3 and late potentials on signal-averaged ECG; 4) T-wave inversion; 5) ventricular arrhythmias and 6) positive family history of ARVC or SCD and/or identification of a pathogenetic mutation by molecular genetic testing [35].

Arrhythmogenic cardiomyopathy has been demonstrated to be the leading cause of SCD in athletes of the Veneto region of Italy (14% of cases), whereas in the USA it accounts for only 6% of cases ([3‐5]; Table 1). This finding may depend on the experience of the pathologist or coroner who performs post-mortem investigation as ARVC is rarely associated with cardiomegaly and may spare the left ventricle, so that the affected heart may be erroneously diagnosed as normal. Of note, in a British study on SCD in athletes, whose hearts were referred for post-mortem to a tertiary centre, the prevalence of ARVC was similar (13%) to that found in the Italian series [21].

The ‘left dominant’ variant of ARVC is characterised by an early and predominant LV involvement, as a result of a specific genetic background [36]. At variance with the classic ‘right-dominant’ variant (Fig. 5), the power of traditional investigations including routine ECG and standard echocardiography for the diagnosis of ‘left-dominant’ ARVC is limited because repolarisation abnormalities and left ventricular systolic dysfunction, either regional or global, are observed in a minority of affected patients. The reason is that the fibro-fatty scarring process predominantly involves the sub-epicardial and mid-mural myocardial layers, which contribute less to ventricular contraction, and the disease lesion can be only identified in the form of late gadolinium enhancement using contrast-enhanced cardiac magnetic resonance imaging (Fig. 6; [37]). Not surprisingly, the incidence of SCD in patients with the classic ARVC variant has markedly decreased since the introduction of ECG preparticipation screening, while the difficult-to-diagnose ‘left dominant’ variant is now an increasingly emergent substrate at post-mortem [38, 39].

Competitive sports activity increases the risk of SCD by 5‑fold in adolescent and young adults with ARVC [5]. Sports has also been implicated as the most important environmental factor promoting ARVC progression and worsening of the disease arrhythmic substrate (Fig. 7). Actually, genetically determined impairment of cell-to-cell adhesion may predispose to myocyte detachment and death, mostly under condition of increased mechanical wall stress such as that occurring during competitive sports activity [41]. Kirchhof et al. [42] demonstrated that in heterozygous plakoglobin-deficient mice, endurance training accelerated the development of right ventricular abnormalities and ventricular arrhythmias. James et al. [43] and Saberniak et al. [44] confirmed in clinical studies that endurance sports and intense physical exercise increase age-related penetrance, risk of ventricular tachyarrhythmias and occurrence of heart failure in ARVC desmosomal gene carriers. Ruwald et al. [45] reported that the absolute risk of malignant arrhythmic events in ARVC patients who participated in competitive sports reached 61% at 40 years of age. On the other hand, early identification of affected patients by preparticipation screening and disqualification from competitive sports activity offers the potential to prevent SCD [1]. Accordingly, both European and American recommendations for sports eligibility in patients with heart diseases and the International Task Force consensus document on ARVC treatment agree that restriction from competitive sports activity of patients with ARVC should be regarded as a therapeutic measure aimed to reduce the risk of SCD (Table 2; [29, 30, 46]).

Dilated cardiomyopathy (DCM) is characterised by dilatation and systolic dysfunction of the left ventricle which is genetically determined in approximately one third of cases. Up to 40 genes have been identified, which affect proteins of a wide variety of cellular structures such as the sarcomere, the nuclear envelope, the cytoskeleton, the sarcolemma and the intercellular junction [48]. Differential diagnosis between athlete’s heart and DCM may be challenging as a sizeable proportion of athletes, mostly those engaged in endurance sports activities, exhibit an enlarged left ventricle and a subset of them show a mild reduction in ejection fraction, i. e. between 45 and 55% [49, 50]. The diagnosis of DCM is suggested by the presence of the following features: depressed systolic function with ejection fraction below 45%, associated regional wall motion abnormalities, concomitant right ventricular dysfunction, positive family history of SCD, cardiac arrest or inherited cardiac disease, and evidence of ECG abnormalities, such as T‑wave inversion or intraventricular conduction defects, complex ventricular arrhythmias at 24-hour Holter monitoring or stress testing and late gadolinium enhancement at contrast-enhanced cardiac magnetic resonance. In borderline cases, demonstration of a significant (>10%) increase of the ejection fraction during exercise echocardiography indicates a preserved contractile reserve and may support the diagnosis of athlete’s heart versus DCM [51].

The American Heart Association consensus statement on definition and classification of cardiomyopathies considers the left ventricular noncompaction (LVNC) as a genetic heart muscle disorder [52], while a European Society of Cardiology consensus document reports it as ‘unclassified cardiomyopathy’ [53]. This inconsistency in the classification of LVNC can be explained by the uncertain nature of the disease, either a separate condition or a phenotypic variant of other cardiomyopathies. Pathogenic mutations have been identified in up to 41% of the affected subjects, in a significant proportion of cases (29%) as sarcomere gene mutations [54, 55]. Although current imaging criteria for the diagnosis of LVNC are based on the ratio between the compact and non-compact LV layers, the thickness of the non-compacted myocardium or the number of trabeculations, these morphologic features are non-specific and can be observed in up to 8% of highly trained athletes (particularly of African/Afro-Caribbean ethnicity). These morphologic changes have been interpreted as a result of the chronic increase in the LV preload occurring during intense training and exercise. In this regard, hypertrabeculation has also been observed in healthy pregnant women (with complete resolution or marked reduction of trabeculation after delivery) or in patients with chronic anaemia [56‐58]. Hypertrabeculation is rarely associated with LV systolic dysfunction, ECG repolarisation abnormalities, or late gadolinium enhancement, suggesting a ‘LVNC cardiomyopathy’ [58]. Hence, hypertrabeculation in isolation can be part of the adaptive heart changes to exercise and should not be considered a cardiac disease [59].

Dilated cardiomyopathy is an uncommon cause of SCD in young competitive athletes. In the Veneto region of Italy, it accounted for 4% of fatalities with cardiovascular origin [5], while in the USA, Maron et al. found it in 2.5% [3] and Harmon et al. in 8% [26] of SCD victims. Finocchiaro et al. reported a 1% prevalence of fatal DCM in British athletes [21]. It is noteworthy that none of the above-mentioned studies reported isolated LVNC as a cause of SCD in athletes (Table 1).

Athletes affected by either DCM or LVNC are considered at significant risk of SCD during exercise, and both European and American recommendations agree that affected athletes are not eligible to engage in competitive sports activity. Conversely, asymptomatic athletes with isolated LVNC, i. e. with a normal LV systolic function, no ECG abnormalities and/or ventricular arrhythmias, and no evidence of late enhancement at cardiac magnetic resonance imaging can participate in competitive sports ([29, 30]; Table 2).

At present, there is no scientific evidence demonstrating that competitive sports activity increases the risk of disease development or sudden death in athletes who carry a pathogenetic mutation but who do not show any sign of phenotypic manifestation (i. e., genotype-positive/phenotype-negative individuals). However, ARVC associated with desmosome gene mutation constitutes an important exception: in this condition, physical exercise and sports activity have been implicated as a key environmental factor for promoting the development and progression of the disease phenotype, both in animal models [42] and in clinical studies on gene mutation carriers [43‐45].

Patients with an ICD are traditionally considered not eligible to engage in competitive sports activity, except for disciplines characterised by a low cardiovascular demand which do not expose to the risk of traumatic damage of the device [60]. However, there is growing evidence that ICD patients may practice sports safely. Lampert et al. [61] reported on 372 US athletes with ICDs (age 10–60 years) participating in either competitive (N = 328) or dangerous (N = 44) sports. During a median follow-up of 31 months, 46 athletes had appropriate therapies (25 during physical exertion) and 29 inappropriate interventions (25 during physical exertion). There were no deaths, resuscitated cardiac arrests or arrhythmia related injuries during sports activity. Freedom from lead malfunction was 97% at 5 years and 90% at 10 years from ICD implant. The authors concluded that their results on efficacy and safety of ICD therapy in the athletic population did not differ significantly from available data in non-athletic ICD patients and, thus, do not support restriction of sports activity for ICD patients. In line with these new data, the most recent recommendations of the American Heart Association regarding the eligibility and disqualification of athletes with cardiovascular abnormalities to engage in competitive sports stated that competitive sports may be allowed in selected athletes with an ICD [62]. The new subcutaneous ICD system appears a valuable therapeutic option for young patients with cardiomyopathy to avoid the risk of transvenous lead damage. However, we should emphasise that the reasons for competitive sports restriction in young patients with cardiomyopathy who have received an ICD for the prevention of sudden cardiac death go beyond the increased risk of unsuccessful shock, inappropriate interventions, injury to the patient, or damage of the device. In fact, as discussed above for patients with ARVC, sports participation may play a key role in the heart muscle disease progression, worsening of the arrhythmic substrate and adverse outcome.

Recreational sports include a wide range of physical activities, from modest to vigorous in intensity, performed either in a regular or inconsistent basis not requiring systematic training or pursuit of excellence, nor involving the same psychological pressure to surpass other participants, which is characteristic of competitive sports.

Although there is a paucity of scientific data regarding the risk of SCD in patients with cardiomyopathies engaging in recreational sports activity, it is likely that a certain degree of risk will remain. However, it seems reasonable to state that this potential risk of exercise should not entirely deprive cardiomyopathy patients of the many benefits offered by regular exercise [63]. A recent study in a small cohort of HCM patients with mild phenotype suggested that moderate-intensity physical activity may provide benefits in terms of fitness with no significant arrhythmic risk [64]. Ruwald et al. [45] reported that the absolute risk of ventricular tachyarrhythmias/death at 40 years after birth was high (33%) in ARVC patients practicing recreational sports, but did not significantly differ from that of physically inactive ARVC patients (22%).

With regard to the prescription of recreational exercise programmes in patients with a diagnosis of cardiomyopathy, the following criteria which characterise a ‘low risk’ of SCD should be taken into account: 1) no symptoms; 2) negative family history for SCD; 3) ‘mild’ structural abnormalities; 4) normal or near-normal response to exercise (i. e. no ST-segment depression and normal increase of blood pressure during stress testing); 5) absence of clinically relevant arrhythmias at exercise testing or during 24-hour ambulatory ECG monitoring. When all these conditions are fulfilled, the physician can reassure the patient with respect to his ability and safety to sustain a regular exercise programme; if one or more risk criteria are present, only low intensity physical activities should be prescribed (Table 3).

Patients should be educated to start an exercise session with a warm-up period and specifically avoid peak exertion, characterised by high-intensity interval exercise. At the end of the session, an appropriate cool-down period is also recommended. The clinician should teach patients to take control over the level of exertion by assessing heart rate with commercially available devices. Exercise in extremely adverse environmental conditions, including very hot, humid, or extremely cold weather should be avoided. Patients should be carefully informed on the specific risk profile of their disease, the potentially dangerous sports activities, mostly in the presence of a history of syncope, and the warning symptoms that may occur in association with exercise. Finally, the availability of an automated external defibrillator in the athletic field should be considered as an additional ‘back-up’ measure to prevent SCD.