Role of Exercise as a Modulating Factor in Arrhythmogenic Cardiomyopathy

Purpose of Review The review addresses the role of exercise in triggering ventricular arrhythmias and promoting disease progression in arrhythmogenic cardiomyopathy (AC) patients and gene-mutation carriers, the differential diagnosis between AC and athlete’s heart and current recommendations on exercise activity in AC. Recent Findings AC is an inherited heart muscle disease caused by genetically defective cell-to-cell adhesion structures (mainly desmosomes). The pathophysiological hallmark of the disease is progressive myocyte loss and replacement by fibro-fatty tissue, which creates the substrates for ventricular arrhythmias. Animal and human studies demonstrated that intense exercise, but not moderate physical activity, may increase disease penetrance, worsen the phenotype, and favor life-threatening ventricular arrhythmias. It has been proposed that in some individuals prolonged endurance sports activity may in itself cause AC (so-called exercise-induced AC). Summary The studies agree that intense physical activity should be avoided in patients with AC and healthy gene-mutation carriers. However, low-to-moderate intensity exercise does not appear detrimental and these patients should not be entirely deprived from the many health benefits of physical activity.


Introduction
Arrhythmogenic cardiomyopathy (ACM) is an inherited heart muscle disease caused by genetically defective cell-to-cell adhesion structures (mainly desmosomes) [1]. The pathophysiological mechanism of ACM is a progressive myocyte loss and substitution by fibro-fatty scar, which represents the substrate for re-entrant ventricular tachycardia (VT) and myocardial dysfunction [2]. Although wall motion abnormalities resulting from myocardial scarring are usually more evident in the thinner right ventricle (RV), cardiac magnetic resonance (CMR) studies demonstrated that ACM is often biventricular [3, 4•]. However, only in a minority of patients the degree of myocyte loss is severe enough to cause symptomatic ventricular dysfunction. Life-threatening ventricular arrhythmias (VAs), usually ventricular fibrillation (VF), and heart failure may also occur abruptly, during the so-called hot-phases, characterized by acute bouts of myocyte necrosis and inflammation similarly to acute myocarditis [5].
Sports activity is a well-known arrhythmic trigger and ACM is a leading cause of sudden death (SD) in the athlete [6][7][8][9][10]. More recently, intense exercise has been recognized as a modulating factor promoting phenotypic penetrance and disease progression both in animal studies and in familial ACM [11-15, 16••, 17-23]. Observation that some longterm endurance athletes may develop disease features or even This article is part of the Topical Collection on Myocardial Disease overt ACM in the absence of gene mutations has led to the hypothesis that ACM may be entirely "exercise-induced" [24]. On the other hand, there is no demonstration that even low-to-moderate intensity exercise may be detrimental in patients with ACM or in their gene-positive family members and recent recommendations suggest not to deprive entirely these subjects from the many benefits of physical activity [25].
In the present article, we will review the role of exercise in triggering ventricular arrhythmias and promoting disease progression. We will also briefly address the differential diagnosis between ACM and athlete's heart. Finally, we will discuss current recommendations about competitive sports and leisure time exercise in ACM patients.

Animal Studies
The role of physical exercise as a trigger of arrhythmic events or disease progression factor has been evaluated in different animal models.
In 2006, Kirchhof and coworkers investigated the effects of age and endurance training on heterozygous plakoglobindeficient (JUP+/−) mice [11]. They found that RV dilatation and dysfunction as well as VAs were exacerbated by daily swimming, supporting that endurance training could accelerate disease progression. However, neither cardiomyocyte abnormalities or fibrofatty replacement nor ultrastructural changes of desmosomes or adherens junctions were observed in trained mice. Further studies from the same group demonstrated that a load-reducing therapy (furosemide and nitrates) prevents training-induced disease development and progression as well as VAs in JUP-deficient (+/−) mice; however, these results have yet to be confirmed in the clinical setting [12].
The effect of endurance training was also assessed in a mouse model overexpressing a nonsense plakophilin-2 (PKP2) gene mutation [13]. Endurance training was associated with abnormal localization and distribution of the Cx43 and exacerbated by 5-fold the risk of developing RV dysfunction. In another study, endurance training was evaluated related to the gradient overexpression of the transgene carrying another nonsense PKP2 gene-mutation [14]. In this case, exercise was associated with RV outflow tract dilatation but no electrocardiographic changes or higher inducibility of VAs. Noteworthy, also transgenic PKP2 animals, similarly to the JUP-deficient mice, showed no histological/ultrastructural changes typical of ACM regardless training. Both these models suggest that exercise may induce ventricular dilation/ dysfunction and favor VAs in mice with abnormal desmosomal proteins but do not cause the structural and ultrastructural changes typical of ACM, suggesting that the deleterious effects of exercise are potentially reversible.
In contrast, a recent study on a conditional heterozygous desmoplakin (DSP) mouse model showed that treadmill exercise partially rescued ACM phenotype by normalizing twothirds of the aberrant regulated genes in sedentary haploinsufficient DSP mice, particularly those related to inflammation, epithelial-mesenchymal transition, and oxidative phosphorylation [15]. According to this model, treadmill exercise did not worsen cardiac systolic function or electrophysiologic parameters but was associated with eccentric cardiac hypertrophy and significant reduction of the apoptotic index in the presence of normally distributed intercalated discs proteins.

ACM as a Cause of Exercise-Related Sudden Death
The recognition of ACM as a cause of exercise-related SD dates back to 1988. Thiene et al. found that 12 of 60 (20%) young (<35 year-old) victims of SD in the Veneto region of Italy had morphologic features of RV cardiomyopathy. Ten of the subjects had died during exertion [6]. Further studies on larger populations confirmed that ACM is one of the leading causes of SD in athletes of the Veneto region of Italy (14% of cases) and that athletes with ACM had 5.4 times higher risk of dying suddenly than their sedentary counterpart [7]. In a British study on SD in athletes, whose hearts were referred for post-mortem to a tertiary center, the prevalence of ACM was similar (13%) to that found in the Italian series [9]. Conversely, ACM accounted for only 6% of SD cases in the USA [26]. Such discrepancy may depend on one hand on the experience of the pathologist or coroner performing postmortem analysis, who may miss this rare and peculiar disease, and on the fact that the US series included many black athletes who are rarely affected by ACM.
Systematic preparticipation screening has demonstrated to reduce the risk of SD from classic ACM variants (rightdominant or biventricular) because of its ability to identify early asymptomatic patients by investigating abnormal electrocardiogram (ECG) findings or VAs at exercise testing [8]. However, a subset of patients exhibits an ACM variant that predominantly affects the left ventricle (left-dominant). ECG abnormalities such as T-wave inversion in the lateral leads V4-V6 and low QRS voltages in the limb leads and left ventricular regional systolic function abnormalities may be observed, but standard clinical investigations are often normal and the diagnosis requires demonstration of subepicardial/ midmyocardial late enhancement on cardiac magnetic resonance ( Fig. 1) [28]. For this reason, the incidence of SD from classic ACM has markedly decreased since the introduction of pre-participation screening, while the difficulty to diagnose "left dominant" variant had led to the fact that this clinical entity is now an increasingly reported substrate of SD [29].
According to recent studies, exercise testing for evaluation of VAs may provide additional value for raising the suspicion of an underlying left-dominant ACM [30]. In particular, the occurrence of exercise-induced repetitive premature ventricular beats with a right-bundle-branch block and superior axis pattern (suggesting an origin from the infero-lateral left ventricular wall) may deserve more in-depth investigation by cardiac magnetic resonance [31]. Table 1 summarizes the design and the results of the studies that addressed the role of exercise activity in modifying the clinical course of patients with ACM and in genotype positive/phenotype negative individuals [16••, 17-23] The studies agree that intense and prolonged physical activity: (1) favors the development of the disease in genotype positive/phenotype negative patients; (2) worsen the degree of ventricular dysfunction in patients with overt ACM; (3) increases the likelihood of sustained ventricular arrhythmias and implantable cardioverter-defibrillator (ICD) interventions (Figs. 2 and 3). Common limitations of these studies included their retrospective nature, the fact that the type and amount of exercise was self-reported and that the definition of type and amount of exercise was heterogeneous.

Role of Sports Activity in Promoting Disease Penetrance and Progression
Many young patients with ACM may wish to participate in recreational and leisure time exercise activity, given the recognized beneficial effects of a physically active lifestyle. On the other hand, a sedentary behavior may carry adverse consequences (e.g., obesity, depression, social deprivation, increased risk of coronary artery disease) that potentially outweigh the benefits. It is difficult to infer from the majority of the studies whether there is a dose-response correlation between exercise and adverse outcome (i.e., ideally ACM patients should not exercise at all) or if there is a threshold of exercise type, intensity and/or duration that can be considered safe. However, the investigations that specifically addressed the outcome of ACM patients engaged in low-to-moderate intensity exercise found that their outcome was similar to that of sedentary patients. These findings suggest that not all exercise is the same and that patients with ACM may be able to derive the benefits of leisure time exercise without excess risk.
The Theory of Exercise-Induced ACM So far, we have discussed the potential role of exercise in modifying the natural history of patients with ACM or of their genotype-positive/phenotype-negative relatives. It has been proposed that intense and prolonged endurance exercise may cause ACM in the absence of gene mutations, so-called exercise-induced ACM. This theory was first advanced by Heidbuchel et al. in 2003, who described a series of 46  Patients were asked what type of exercise level they performed both before and after diagnosis, with the following pre-specified fixed options: (i) inactive; (ii) recreational; ACM patients engaged in competitive sport had: 1. Significantly higher risk of VT/death 2. Symptoms developed at an earlier age In ACM patients, competitive sport was associated with a twofold increased risk of VT/death, and earlier presentation of symptoms, when compared with inactive patients, and to patients who participated in recreational sport The major arrhythmic event-free survival was shorter and the occurrence of severe RV dysfunction was more probable in the high-intensity exercise group, followed by the moderate intensity group and by the low-intensity/inactive group A dose-dependent relationship between a history of dynamic physical activity and phenotypic severity based on the early onset of major arrhythmic events and deterioration of RV contractility was observed.  [33]. Subsequent genotype-phenotype studies showed that ACM features in subjects who engaged in high-intensity exercise were typically not associated with a negative molecular genetics analysis for desmosomal gene mutations (so-called gene-elusive patients) ACM [17,34]. In the "John Hopkins registry," gene-elusive patients, particularly those <25 yearold, reported performing significantly more intense exercise prior to their diagnosis compared to desmosomal-gene mutation carriers [17]. Another analysis of the same registry, which focused on the effect of physical activity on ICD carriers, found that exercise restriction reduced the arrhythmic risk more in gene-elusive patients than in those with desmosomal mutations [35]. These observations further support the hypothesis that ACM may be acquired through intense exercise. Others reported that athletes may exhibit arrhythmogenic ventricular abnormalities typical of ACM variants such as isolated non-ischemic left ventricular scar or a particular form of RV outflow tract tachycardia associated with epicardial scar [36][37][38]. Taken together, these findings suggest that even in the absence of desmosomal-gene mutations, the increased hemodynamic load induce by prolonged intense exercise may be enough to cause permanent myocardial damage in some cases. However, it has to be noted that this phenomenon is rare as many studies demonstrated a lack of long-term adverse remodelling in large series of top-level athletes [24]. For this reason, it is plausible that a combination of a genetic predisposition that remains unknown and environmental factors (including exercise) is needed for an athlete with no mutations in known ACM-causing genes to develop the disease.

Sports Activity in ACM Patients with an ICD
The risk of SD during exercise is the main reason why patients with ACM are advised against engaging in sport activities. The ICD may offer protection against SD and may theoretically allow safe sport participation of athletes with ACM. Traditionally, there has been concern that intense exercise may influence the efficacy and safety of ICD therapy and increase the risk of inappropriate shocks, damage to the device, and lead failure [39] (Fig. 4). A multinational registry that recruited 372 athletes provided reassuring data about sports participation of ICD carriers by showing no arrhythmic deaths, resuscitated cardiac arrests, or shock-related injuries during follow-up and no increased risk of technical malfunction. On the other hand, the study confirmed the detrimental effect of competitive sports on the arrhythmic outcome of ACM patients by showing that an underlying ACM was the only variable associated with exercise-induced ICD shock [40•]. A sub-analysis on the arrhythmic outcome of ICD carriers enrolled in the European arm of the Registry who engaged in leisure time physical activity (mostly endurance disciplines such as running or cycling) showed that the risk of receiving appropriate ICD interventions was lower (6.3% vs. 20.2%) than their counterpart of competitive athletes. Also, in  [32] this subgroup, however, 2 of 3 patients that experienced appropriate ICD interventions during exercise were affected by ACM [41].
The association between continuing to practice high intensity exercise after ICD implantation and worse arrhythmic outcome was demonstrated by a study on 129 patients enrolled in the Johns Hopkins ACM registry, of whom 80% were athletes engaged in intense sports activity before the diagnosis (>936 METs*Hours yearly) [35]. More than half of patients (66/129) received appropriate ICD therapy during a median follow-up of 5.1 years. The chance of experiencing an appropriate intervention was directly related with the reduction of physical activity compared with baseline, and the correlation between exercise dose reduction and arrhythmic outcome was strengthen after correction for demographics, exercise dose before presentation, and established arrhythmic risk factors. The hazard ratio of 0.14 at multivariate analysis translated into an 86% lower risk of ICD therapy of those who reduced the exercise dose the most compared to those who reduced the least, independently from other variables. Of note, reducing the exercise "dose," which takes into account the intensity of the activity, demonstrated higher effect in decreasing arrhythmic risk than just maintaining the same intensity for a shorter duration.
Finally, it should be emphasized that the reasons for competitive sports restriction in young patients with a prophylactic ICD go beyond the increased arrhythmic risk. As discussed, intense exercise plays a role in disease progression by accelerating the process of myocyte death and fibro-fatty replacement, thus worsening ventricular dysfunction.
In conclusion, considering the above-discussed evidence, there is general agreement that the implant of an ICD should not be considered a justification for participation of patients with ACM in competitive sports or high-intensity noncompetitive physical activities, particularly in high-risk patients. On the other hand, compliance or not to exercise restriction Fig. 3 Representative example of the role of endurance exercise in worsening the arrhythmogenic cardiomyopathy phenotype. A 38-year-old runner received a diagnosis of arrhythmogenic cardiomyopathy after investigation of ventricular arrhythmias and ECG abnormalities at preparticipation screening. Genetic testing revealed the presence of a plakophilin-2 gene mutation. At the time of diagnosis, cardiac magnetic resonance 4-chamber view on cine sequences found dilatation of the right ventricle with mild dysfunction (a) and diffuse right ventricular lateenhancement on post-contrast sequences (arrowheads) (b). Although according to the Italian law he was considered not eligible for competitive sports activity, he continued to practice highintensity endurance training. After 4 years, a new cardiac magnetic resonance showed a more enlarged right ventricle, an apical aneurysm (arrow), and a moderate right ventricular dysfunction (c). Post-contrast sequences confirmed the presence of a diffuse right ventricular late enhancement (arrowheads) (d) should be considered as an important factor in the decision to implant an ICD in intermediate-risk patients.

Differentiating ACM and Athlete's Heart
The long-running Italian experience with pre-participation screening demonstrated that early identification of affected individuals and restriction from competitive sports activity can substantially reduce the risk of SD [8,42]. The electrical and structural remodelling of the athlete's heart may sometimes require differential diagnosis with ACM. Although a detailed analysis of all elements useful for differentiating the two conditions goes beyond the scope of the review, some key points are: 1) ACM is an inherited disorder and family history of cardiomyopathies, and premature SD should be extended beyond first-degree family members. It is noteworthy that end-stage ACM can be confused with dilated cardiomyopathy [43]. 2) Structural remodelling of the athlete's heart may cause RV dilation (mainly the main body rather than the outflow tract) that parallels left ventricular dilation (right/left   [39] ventricular ratio remains <1) and is never accompanied by regional wall motion abnormalities. Mild global dysfunction may be observed at rest in highly trained athletes, but exercise echocardiography demonstrates normalization of contractility [44][45][46]. 3) At cardiac magnetic resonance, only the presence of lateenhancement at the insertion points between the right and left free wall and the septum (junctional late enhancement) can be considered normal in athletes [37]. 4) VAs, at rest or during exercise, are observed in a minority of athletes versus the majority of ACM patients [47][48][49][50]. Arrhythmias at exercise testing were also observed in the majority of asymptomatic desmosomal gene-mutation carriers without structural changes at cardiac magnetic resonance suggesting that VAs may be the first manifestation of ACM [51]. Besides its diagnostic value, exercise testing appears particularly useful for evaluating the threshold of VAs onset when prescribing moderateintensity exercise in ACM patients. 5) Although multiple ECG abnormalities may be observed in ACM, T-wave inversion in the anterior leads V1-V4 is the most common and should always raise the suspicion of an underlying disease. In the athlete, anterior T-wave inversion should be considered normal only before pubertal development ("juvenile pattern of repolarization") or in black individuals when preceded by J-point/ST-segment elevation (early repolarization variant typical of Afro-Caribbean athletes) [52]. 6) The evaluation of an athlete should be multi-parametric: although no single test is diagnostic of ACM, the association of more than one abnormality at family history, resting, and exercise ECG and imaging increases the probability of a disease (Fig. 5).
Competitive Sport, Leisure Time Physical Activity and ACM: Current Guidelines Table 2 summarizes the recommendations for competitive sports participation and leisure time physical activity according to recent guidelines and consensus documents. They all agree that patients with ACM should not engage in competitive sport, with the possible exception of skills disciplines, or high-intensity leisure time physical exercise. According to the   [53] 2015 American Heart Association/American College of Cardiology guidelines, healthy gene carriers are not precluded from participation in competitive sports [54]. Conversely, more recent documents are more restrictive because of emerging evidence that regular exercise training and competitive sports can play a role in triggering cellular mechanisms leading to the development and progression of the disease phenotype in the presence of a predisposing gene abnormality [25,[55][56][57].

Conclusions
There is compelling evidence from both animal and human studies that exercise is an important modulating factor in ACM that may favor disease penetrance in genotype positive/phenotype negative patients, worsen ventricular dysfunction, and trigger arrhythmias. For these reasons, there is unanimous consensus that ACM patients should be advised against participation in competitive sports (with the possible exception of skill disciplines at low-cardiovascular demands in selected cases) and high-intensity leisure time physical activity. On the other hand, ACM patients, particularly young with a mild disease, should not be deprived from the many health benefits of low-to-moderate intensity physical activity. In ACM, exercise is a medicine with a narrow therapeutic range and should be prescribed by experienced physicians on a case-by-case basis.

Declarations
Conflict of Interest The authors declare that they have no conflict of interest.
Human and Animal Rights and Informed Consent This article does not contain any studies with human or animal subjects performed by any of the authors.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.  [52] • No competitive sports (with the possible exception of skill sports) for possible, borderline or definite ACM • ICD placement for permitting participation in high-intensity sport not recommended Not addressed 2015 International Task Force Consensus Statement on the Treatment of ARVC [54] • No competitive sports in patients with definite ACM • Restriction from competitive sports activity may be considered in ACM family members with a negative phenotype, either healthy gene carriers or with unknown genotype • Low-intensity exercise may be allowed 2019 EAPC recommendations for participation in competitive and leisure time sport in athletes with cardiomyopathies [55] • No competitive sports in patients with definite or borderline ACM and gene positive/phenotype negative patients • Low-intensity exercise recommended 2019 HRS guidelines on ACM [56] • No competitive sports in ACM patients and gene positive/phenotype negative patients • Frequent high-intensity endurance exercise not advised in ACM patients and gene positive/phenotype negative patients • Preference for low-intensity activities 2020 ESC guidelines sports cardiology and exercise in patients with cardiovascular disease [25] • No competitive sports of any type in ACM patients and gene positive/phenotype negative patients • Low-intensity exercise, 150 min/week, for all • Moderate intensity exercise only in low-risk patients* • High intensity not advised ACM arrhythmogenic cardiomyopathy, ACC American college of cardiology, AHA American heart association, EAPC European association of preventive cardiology, ESC European society of cardiology, ICD implantable cardioverter defibrillator, HRS heart rhythm society *Low risk = no history of cardiac arrest or ventricular tachycardia, low arrhythmic burden (<500 PVBs/day, no exercise-induced arrhythmias), minimal structural abnormalities