OUP user menu

Desmoplakin disease in arrhythmogenic right ventricular cardiomyopathy: early genotype–phenotype studies

Srijita Sen-Chowdhry , Petros Syrris , William J. McKenna
DOI: http://dx.doi.org/10.1093/eurheartj/ehi343 1582-1584 First published online: 7 June 2005

This editorial refers to ‘Clinical profile of four families with arrhythmogenic right ventricular cardiomyopathy caused by dominant desmoplakin mutations’ by B. Bauce et al., on page 1666

Arrhythmogenic right ventricular cardiomyopathy (ARVC) is an inherited heart muscle disease that may result in ventricular arrhythmia and sudden death. Myocyte loss with fibrofatty replacement is the typical pathological finding, with inflammatory infiltrates in up to 67% of cases.1 Insight into the pathogenesis of ARVC awaited identification of the first disease-causing gene, plakoglobin, in a recessive variant of ARVC known as Naxos disease.2 Naxos disease is distinguished from the more common autosomal dominant disease form by the presence of palmoplantar keratoderma and woolly hair in all genetically affected individuals.

Plakoglobin is a key component of desmosomes, the specialized intercellular junctions of cardiac and epithelial tissues. Desmosomes provide mechanical coupling between the intermediate filaments and the cytoplasmic membranes of adjacent cells. Tissue exposed to shear or frictional stress is dependent on the strength of this supporting network. A defect in a major desmosomal component may compromise its adhesive function, pre-disposing to cellular detachment and death. Significant myocyte necrosis may be accompanied by an inflammatory response. The regenerative capacity of the myocardium is limited, necessitating repair by fibrofatty substitution.

The identification of plakoglobin as the cause of Naxos disease paved the way for the search for other desmosomal mutations. Shortly thereafter, a homozygous mutation in desmoplakin was isolated in Carvajal disease, a cardiocutaneous syndrome bearing close resemblance to Naxos disease, with two exceptions: early and prominent left ventricular involvement and pure fibrosis without adipose replacement.3 In 2002, Rampazzo et al.4 reported a missense mutation in desmoplakin (S299R) in a large Italian family with autosomal dominant ARVC. The subsequent recognition that mutations in plakophilin-2 are common in ARVC lent further support to the desmosomal model.5 The phenotypic characteristics of families with mutations in plakophilin remain to be established.

Bauce et al.6 report the first systematic analysis of genotype–phenotype correlations in dominant ARVC. Their study population includes four families with varying defects in desmoplakin. Desmoplakin is a 210–250 kDa protein expressed in two isoforms generated by alternative splicing and differing in the length of the coiled-coil rod domains that mediate homodimerization. The N-terminus of desmoplakin is essential for localization to the desmosomal plaque and additionally contains binding sites for plakoglobin and plakophilin. At its C-terminus, desmoplakin binds to the intermediate filament desmin.

The original S299R mutation modifies a putative phosphorylation site in the N-terminal plakoglobin-binding domain. The group expand on their earlier work by describing two additional missense mutations in the rod domain and the c.423-1G>A mutation that alters the acceptor splicing site of intron 3. The latter encodes a truncated protein of 200 amino acids in length. Although in vitro expression studies are currently lacking, the predicted effect of the splicing mutation is sufficiently disparate from the remainder to generate variations in phenotype.

Left ventricular involvement is traditionally considered an end-stage complication of ARVC, occurring after the onset of marked right ventricular dilation and dysfunction. The authors report that half of the patients in their study population with echocardiographic evidence of ARVC had left ventricular abnormalities and highlight the limitations of existing diagnostic criteria in this respect.6 The Task Force Guidelines were proposed over a decade ago to facilitate diagnosis of a then little-known disease. Minimal or absent left ventricular disease was a frequent stipulation of the criteria, intended to improve specificity and ensure distinction from dilated cardiomyopathy. However, as the authors indicate, strict application of the Task Force criteria would indeed preclude diagnosis in many cases of ARVC with left ventricular involvement.

Of note, variants of ARVC with a predilection for the left ventricle are increasingly recognized. The salient pathological finding is fibrofatty replacement of the left ventricle with variable right-sided involvement.7 Clinical manifestations of left-sided ARVC include dilation and systolic impairment of the left ventricle and inverted T-waves confined to the inferolateral leads.8 This entity may be differentiated from progressive ARVC by the occurrence of left ventricular involvement early in the disease course. Carvajal syndrome may be regarded as an autosomal recessive form of left-sided ARVC. As noted by Bauce et al.,6 the presence of prominent repolarisation abnormalities, ventricular arrhythmia, and right ventricular aneurysms distinguish it from dilated cardiomyopathy.

Interestingly, features of early left-sided disease were observed in conjunction with the c.423-1G>A mutation. The variation is best illustrated by a review of the echocardiographic data. In the other families, patients with left ventricular ejection fractions of <55% invariably had severe right ventricular dilation and impairment. Conversely, two patients carrying the splicing mutation had left ventricular systolic dysfunction in the context of preserved or minimally impaired right ventricular function. Furthermore, one of these patients demonstrated isolated lateral T-wave inversion characteristic of left-sided ARVC, although the documented sustained ventricular tachycardia was of right ventricular origin. The authors further report low voltage QRS complexes and/or inferior Q-waves in a significant proportion of their patients, whereas other groups have highlighted prolonged upstroke of the S-wave9 and R-wave reduction in the right pre-cordial leads. Incorporation of these features into future revisions of the Task Force criteria may enhance diagnostic sensitivity.

The propensity for early left ventricular involvement in Carvajal syndrome and the family with the splicing mutation may have a genetic basis. The 7901delG mutation of Carvajal syndrome produces a premature stop codon with consequent truncation of the desmoplakin protein and absence of its C-terminus. The final effect of the c.423-1G>A mutation is comparable. Loss of the C-terminus of desmoplakin will compromise normal attachment of the intermediate filaments. Although impaired cell adhesion may impact predominantly on the thin-walled distensible right ventricle, myocytes in the high-pressure left ventricle may be dependent on a strong intermediate filament cytoskeleton. We can hypothesize that mutations resulting in premature truncation of desmoplakin may predispose to early left ventricular involvement; however, detailed phenotyping of large kindreds will be required to evaluate this supposition.

Assessment of left ventricular disease is particularly important from a prognostic standpoint. Retrospective follow-up studies have highlighted left ventricular involvement as a risk factor for recurrent arrhythmic events.10 Of the six patients who suffered ventricular fibrillation (VF), four had clinical and/or pathological evidence of left ventricular involvement, including posterior hypokinesia and repolarization abnormalities of the left ventricular leads. Inferior Q-waves were found to be associated with fibrosis of the inferior left ventricular wall in one patient who died suddenly at the age of 15. Cardiac evaluation 2 years before had been unremarkable; post-mortem studies revealed acute myocyte necrosis and inflammation without cavity enlargement or aneurysms.6 This sudden and unexpected death is worthy of closer examination, as is the atypical clinical presentation of another patient who suffered VF arrest.

The latter patient was admitted to hospital at the age of 12 with prolonged chest pain, ST segment elevation, and myocardial enzyme rise. Angiography revealed unobstructed coronary arteries with wall motion abnormalities in both ventricles. Chest pain is a common complaint among ARVC patients and has been attributed to thickening of the media of distal coronary vessels.11 However, acute chest pain may be related to acceleration of the underlying disease process, with myocyte death and accompanying inflammation. The so-called ‘hot phase’ may be clinically silent in some individuals, whereas others will present with symptoms of myocarditis or worsening ventricular arrhythmia. Sudden death may be a rare and tragic manifestation, with myocardial necrosis and inflammation as the dominant features on autopsy. We can speculate that a proportion of deaths attributed to acute myocarditis may represent activation of previously dormant ARVC.

The authors conclude that genetic screening is mandatory for early identification of asymptomatic carriers in families with a genotyped index case. Although private mutations may predominate in familial ARVC, the relatively frequent involvement of desmoplakin and plakophilin suggests that genotyping will be feasible in a significant proportion of families.5 Cascade screening of relatives, diagnostic clarification in borderline cases, and reassurance of gene-negative individuals are likely to be the major benefits. However, the identification of asymptomatic carriers raises a number of challenging issues. In contrast to hypertrophic cardiomyopathy, a reliable algorithm for primary prevention of sudden death in familial ARVC is currently lacking. Recent advances in prognostic assessment have focused primarily on probands; risk stratification in asymptomatic relatives has not been systematically addressed. In describing the clinical profile of families with desmoplakin mutations, the authors underscore the difficulties in predicting arrhythmic events; sudden death may occur at rest, without pre-monitory symptoms, and in the absence of prominent clinical findings.6 Thus, a management strategy based on frequent follow-up and exclusion from competitive sports will not prevent a significant proportion of events. Conversely, universal placement of implantable cardioverter-defibrillators (ICDs) is not a tenable option. The ICD represents the ultimate safeguard against sudden death but is far from being hazard-free,12 particularly in young patients. In view of the incomplete penetrance of most ARVC mutations and the benign disease course in the majority, the risks of device placement will outweigh the benefits in most asymptomatic gene carriers. Until the dilemma of risk prediction in this often-unpredictable disease can be resolved, early identification of asymptomatic gene carriers will remain of limited prognostic value.


This ongoing work is supported by the British Heart Foundation and European Community Research Contract no. QLG1-CT-2000-01091.


  • doi:10.1093/eurheartj/ehi341

  • The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.