Abstract

Aims

We sought to examine the clinical presentation and natural history and to identify long-term prognostic predictors in patients with arrhythmogenic right ventricular cardiomyopathy (ARVC) as information concerning the natural history and risk stratification of ARVC is still incomplete.

Methods and results

A cohort of 96 ARVC patients (68% males, 35 ± 15 years) was enrolled and underwent structured diagnostic protocol and follow-up. Primary study endpoints were death and heart transplantation (HTx). Clinical and echo-Doppler data were assessed as prognostic indicators. Sixty-five per cent of patients had right ventricular (RV) systolic dysfunction (RV fractional area change < 33%) and 24% had left ventricular (LV) systolic dysfunction (LV ejection fraction <50%). During a mean follow-up of 128 ± 92 months, 20 patients (21%) experienced cardiac death or underwent HTx. At multivariate analysis (Model 1), RV dysfunction [hazard ratio (HR): 4.12; 95% confidence interval (CI): 1.01–18.0; P = 0.05], significant tricuspid regurgitation (HR: 7.6; 95% CI: 2.6–22.0; P < 0.001), and amiodarone treatment (HR: 3.4; 95% CI: 1.3–8.8; P = 0.01) resulted as predictors of death/HTx. When inserting in the model, the ‘ordinal dysfunction’ (Model 2), which considers the presence of both RV and LV dysfunctions, this variable emerged as an independent prognostic predictor (HR: 6.3; 95% CI: 2.17–17.45; P < 0.001). At the receiver operating characteristic analysis, Model 2 was significantly more accurate in predicting long-term outcome compared with Model 1 (area under the curve 0.84 vs. 0.78, respectively; P = 0.04).

Conclusion

In our tertiary referral centre ARVC population, the presence of LV dysfunction at diagnosis has an incremental power in predicting adverse outcome compared with RV dysfunction alone.

See page 1049 for the editorial comment on this article (doi:10.1093/eurheartj/ehr088)

Introduction

Arrhythmogenic right ventricular cardiomyopathy (ARVC), a primary myocardial disease that may lead to life-threatening ventricular arrhythmias and heart failure (HF), has a prevalence estimated between 1:2000 and 1:5000.1,2 Arrhythmogenic right ventricular cardiomyopathy is a genetic disease,3 and a familial trait has been reported in 30–50% of cases.4,5

Several desmosomal mutations6,7 are considered to be responsible for the progressive loss of ventricular myocytes, and their replacement with fibro-adipose tissue, the pathologic landmark of the disease.

Originally, the structural abnormalities of ARVC were described in the right ventricle (RV). The consequent ventricular electric instability was considered the main clinical characteristic of the disease. However, left ventricular (LV) involvement in ARVC has also been reported,8–12 and congestive HF can lead to death or heart transplantation (HTx).1,2,9,13–15

Although several studies13–17 have been reported in patients with ARVC, they failed to provide consistent data regarding the natural history and risk stratification of the disease. Therefore, the aims of this study were to examine the clinical presentation and natural history of the disease, as well as to identify long-term prognostic predictors in a large cohort of patients with ARVC, specifically focusing on the prognostic impact of echocardiographic parameters and LV involvement.

Methods

Study population

Data were obtained from the Registry of Myocardial Diseases of Trieste where 98 patients with ARVC were enrolled between January 1976 and January 2008. All patients who met the ARVC diagnostic criteria18 and with available follow-up (96 patients) were included in the present study. Family history, previous documented ventricular arrhythmias, age, and symptoms at onset were systematically assessed. All patients underwent clinical examination, 12-lead electrocardiogram (ECG), and echocardiography (M-mode measurements since 1976, two-dimensional since 1982, and Doppler and colour-Doppler since 1986). Signal-averaged ECG, 24 h Holter monitoring, exercise stress test, magnetic resonance imaging, coronary angiography, haemodynamic study, and endomyocardial biopsy were performed in selected cases, according to clinical indications.

Patients were treated for arrhythmias and HF. Anti-arrhythmic drug treatment was prescribed for frequent and repetitive ventricular arrhythmias, regardless of the presence of related symptoms. Implantable cardioverter-defibrillator (ICD) devices were implanted according to clinical indications.19 Patients with HF were treated with diuretics and digitalis, and since 1988 also with angiotensin-converting enzyme (ACE)-inhibitors. In selected cases, beta-blockers were used.

Available data were obtained retrospectively in the patients with an ARVC diagnosis before 1993; subsequently, data acquisition and analysis were made in a prospective manner. Only baseline clinical and laboratory data were considered in the present study.

Study design

A primary endpoint was cardiovascular death or HTx. The cause of cardiovascular death was based on a clinical history of sudden death (SD) or death due to refractory HF. Sudden death was defined as an immediate death or occurring within 1h after the onset of symptoms or during sleep in stable New York Heart Association (NYHA) I–III functional class patients. Patients who suffered a non-cardiac death were censored, whereas a death of unknown cause was considered a primary endpoint, preferring to obtain a less conservative estimate than to run the opposite risk. The end of follow-up was the last available clinical visit or the date of death or HTx. Endpoint data were obtained directly from patients during periodic evaluations, from patients' physician or relatives, or from the death registry of the patients' municipality. The disease was considered familial if there was clinical or post-mortem diagnosis of ARVC or SD at young age (<35 years) in a first-degree relative. ARVC patients were distinguished into probands (the first diagnosed case in an affected family) and family members.

Right ventricular systolic dysfunction was defined by two-dimensional echocardiography: an RV fractional area change (FAC; from apical four-chamber view) <33% (<2 standard deviations from normal values) defines RV systolic dysfunction, while RV FAC ≤ 25% (≤3 standard deviations from normal values) defines severe RV systolic dysfunction.20 Left ventricular systolic dysfunction was defined as an LV ejection fraction (EF; obtained by the biplane Simpson's method) < 50%; it was classified as mild (LVEF: 45–49%), moderate (LVEF: 36–44%), or severe (LVEF ≤ 35%). Right ventricle and LV were considered as enlarged when RV end-diastolic area > 24 cm2 and indexed LV end-diastolic volume > 76 mL/m2.21

We defined ‘ordinal dysfunction’ to arbitrarily classify the presence of ventricular dysfunction into three stages: 0 = no ventricular dysfunction, 1 = presence of RV dysfunction, and 2 = biventricular dysfunction. In patients with available colour-Doppler data, assessment of mitral regurgitation (MR) and tricuspid regurgitation (TR) was made semi-quantitatively, according to international guidelines.22 For the purpose of the analysis, MR and TR were considered as significant in the presence of a regurgitant jet area >4 cm2 on colour-Doppler.

Statistical analysis

Summary statistics of clinical and instrumental variables at enrolment were expressed as a mean and standard deviation or a count and percentage, as appropriate. Comparison between patients with and without events was made by the ANOVA test on continuous variables, using the Brown–Forsythe statistic when the assumption of equal variances did not hold, and the χ2 test for discrete variables. Survival curves were calculated by using the Kaplan–Meier method for the entire population and then stratified by the level of dysfunction; the comparison between the estimated curves was performed with the log-rank test. Univariate Cox proportional hazards models were applied to find predictors of primary endpoints. Multivariable Cox models were then estimated to determine the relationship between a subset of baseline clinical-laboratory characteristics and the long-term outcome. In particular, we compared the additive prognostic power of two different models. Model 1 included clinical as well as echo-Doppler data of the right heart, whereas in Model 2 the ‘ordinal dysfunction’ was inserted, thus including LV dysfunction. The covariates were selected from univariate analysis and by taking into account the limited number of events by means of a backward-conditional step-wise procedure. Subjects with missing data in the selected covariates were deleted case-wise. Areas the under receiver operating characteristic (ROC) curve of the estimated cumulative hazard functions were compared to check a possible increase in predictive accuracy of the estimated Cox models. The results were considered statistically significant with P < 0.05.

Inter- and intra-observer variability regarding RV areas and FAC was verified by selecting a sample size of 30 ARVC patients (four observations per subject: two different operators, double evaluation each) to achieve 90% power and, thus, to detect an intra-class correlation (ICC) of 0.8 under the null hypothesis of ICC = 0.5, by using an F-test, with a significance level of 0.05.23 The intra-observer variability was evaluated by computing the paired correlation between repeated measures of the same operator. For all the evaluated parameters the ICC values were ≥0.95 and paired correlations were ≥0.90. Statistical analyses were performed with SPSS Statistical Package 14.0 and R statistical software (version 2.7.2).

Results

The study population counted 96 patients of whom 76 (79%) were probands and 20 (21%) were family members.

The clinical and laboratory findings of the patients are summarized in Tables 1 and 2, with the frequency of major and minor criteria for ARVC diagnosis.18 In the tables, the study population was classified in two subgroups according to the presence (Group A) or the absence (Group B) of primary endpoints during follow-up.

Table 1

Clinical and laboratory data of arrhythmogenic right ventricular cardiomyopathy patients at baseline, classified according to primary endpoint at follow-up: Group A: patients with events during follow-up (cardiovascular death or heart transplantation); Group B: surviving patients

 Global population (n = 98) Group A patients (n = 20) Group B patients (n = 78) P-value 
Clinical history/physical examination 
 Males (%) 68 60 69 NS 
 Age at diagnosis (years) 34±15 35±17 34±15 NS* 
 Probands (%) 79 70 80 NS 
 Family history (%)a 46 65 40 0.04 
  Sudden death [<35 years, (%)]a 33 39 31 NS 
  Clinical diagnosis (%)a 52 46 55 NS 
  Confirmed by autopsy (%)b 14 15 14 NS 
 Duration of symptoms (months) 47±59 44±56 48±60 NS 
 Asymptomatic (%) 27 20 29 NS 
 Palpitations (%) 41 25 45 NS 
 Syncope (%) 15 19 0.04 
 Chest pain (%) NS 
 Cardiac arrest (%) NS 
 Heart failure symptoms (%) 17 35 12 0.02 
 NYHA III–IV (%) 15 0.03 

 
ECG 
 Abnormal ECG (%) 78 90 75 NS 
 Low-voltage QRS (%) 16 30 11 0.05 
 Complete RBBB (%) 10 NS 
 Epsilon waves (%)b 16 35 11 0.007 
 Localized QRS dispersion [>110 ms in right precordial leads, (%)]b NS 
 Negative anterior T-waves (%)a 51 70 46 NS 
 SVT at presentation (%) 23 35 20 NS 
 SVT LBBB pattern (%)a 19 20 19 NS 
 Atrial fibrillation/flutter 10 NS 

 
SAECG (in 39% of pts) 
 Late potentials (%)a 62 60 63 NS 

 
Holter registration (in 65% of pts) 
 PVB>1000/24 ha 48 64 43 NS 
 Couples (%) 52 71 46 NS 
 NSVT LBBB pattern (%)a 34 50 30 NS 
 Supraventricular arrhythmias (%) 19 21 18 NS 

 
Exercise test (in 67% of pts) 
 Functional capacity (W) 138±56 101±47 145±56 NS 
 Interruption for arrhythmias (%) NS 
 Exercise SVT LBBB pattern (%)a NS 
 Exercise frequent PVB (%) 37 64 32 0.04 

 
Electrophysiological study (in 27% of pts) 
 Induced MVA (%) 50 20 57 NS 
 LBBB morphology of induced MVA (%) 62 67 NS 

 
Cardiac catheterization (in 41% of pts) 
 Right atrial pressure (mmHg) 4±3 4±4 4±3 NS 
 Pulmonary artery systolic pressure (mmHg) 22±7 23±7 21±7 NS 
 Pulmonary capillary pressure (mmHg) 8±3 8±4 8±3 NS 
 Mean aortic pressure (mmHg) 87±17 84±25 88±14 NS 
 Cardiac index (L/min/m23.7±1.1 3.5±1.4 3.7±1.0 NS 

 
Magnetic resonance (in 21% of pts) 
 RV fat tissue (%) 50 50 50 NS 
 RV aneurysm (%) 38 100 29 0.05 
 LV fat tissue (%) 15 50 11 NS 

 
Right ventricular biopsy (in 26% of pts) 
 RV fibro-fatty infiltration (%)b 52 50 53 NS 

 
Treatment 
 Anti-arrhythmics (%) 58 75 54 NS 
 Amiodarone (%) 24 50 17 0.002 
 Sotalol (%) 19 15 20 NS 
 Beta-blockers [except sotalol, (%)] 20 15 21 NS 
 ACE-inhibitors (%) 17 30 13 NS 
 Digitalis (%) 15 45 0.001 
 Vasodilators (%) 12 30 0.03 
 Diuretics (%) 17 60 0.001 
 Global population (n = 98) Group A patients (n = 20) Group B patients (n = 78) P-value 
Clinical history/physical examination 
 Males (%) 68 60 69 NS 
 Age at diagnosis (years) 34±15 35±17 34±15 NS* 
 Probands (%) 79 70 80 NS 
 Family history (%)a 46 65 40 0.04 
  Sudden death [<35 years, (%)]a 33 39 31 NS 
  Clinical diagnosis (%)a 52 46 55 NS 
  Confirmed by autopsy (%)b 14 15 14 NS 
 Duration of symptoms (months) 47±59 44±56 48±60 NS 
 Asymptomatic (%) 27 20 29 NS 
 Palpitations (%) 41 25 45 NS 
 Syncope (%) 15 19 0.04 
 Chest pain (%) NS 
 Cardiac arrest (%) NS 
 Heart failure symptoms (%) 17 35 12 0.02 
 NYHA III–IV (%) 15 0.03 

 
ECG 
 Abnormal ECG (%) 78 90 75 NS 
 Low-voltage QRS (%) 16 30 11 0.05 
 Complete RBBB (%) 10 NS 
 Epsilon waves (%)b 16 35 11 0.007 
 Localized QRS dispersion [>110 ms in right precordial leads, (%)]b NS 
 Negative anterior T-waves (%)a 51 70 46 NS 
 SVT at presentation (%) 23 35 20 NS 
 SVT LBBB pattern (%)a 19 20 19 NS 
 Atrial fibrillation/flutter 10 NS 

 
SAECG (in 39% of pts) 
 Late potentials (%)a 62 60 63 NS 

 
Holter registration (in 65% of pts) 
 PVB>1000/24 ha 48 64 43 NS 
 Couples (%) 52 71 46 NS 
 NSVT LBBB pattern (%)a 34 50 30 NS 
 Supraventricular arrhythmias (%) 19 21 18 NS 

 
Exercise test (in 67% of pts) 
 Functional capacity (W) 138±56 101±47 145±56 NS 
 Interruption for arrhythmias (%) NS 
 Exercise SVT LBBB pattern (%)a NS 
 Exercise frequent PVB (%) 37 64 32 0.04 

 
Electrophysiological study (in 27% of pts) 
 Induced MVA (%) 50 20 57 NS 
 LBBB morphology of induced MVA (%) 62 67 NS 

 
Cardiac catheterization (in 41% of pts) 
 Right atrial pressure (mmHg) 4±3 4±4 4±3 NS 
 Pulmonary artery systolic pressure (mmHg) 22±7 23±7 21±7 NS 
 Pulmonary capillary pressure (mmHg) 8±3 8±4 8±3 NS 
 Mean aortic pressure (mmHg) 87±17 84±25 88±14 NS 
 Cardiac index (L/min/m23.7±1.1 3.5±1.4 3.7±1.0 NS 

 
Magnetic resonance (in 21% of pts) 
 RV fat tissue (%) 50 50 50 NS 
 RV aneurysm (%) 38 100 29 0.05 
 LV fat tissue (%) 15 50 11 NS 

 
Right ventricular biopsy (in 26% of pts) 
 RV fibro-fatty infiltration (%)b 52 50 53 NS 

 
Treatment 
 Anti-arrhythmics (%) 58 75 54 NS 
 Amiodarone (%) 24 50 17 0.002 
 Sotalol (%) 19 15 20 NS 
 Beta-blockers [except sotalol, (%)] 20 15 21 NS 
 ACE-inhibitors (%) 17 30 13 NS 
 Digitalis (%) 15 45 0.001 
 Vasodilators (%) 12 30 0.03 
 Diuretics (%) 17 60 0.001 

MVA, major ventricular arrhythmia; LBBB, left bundle branch block; LV, left ventricular; NSVT, non-sustained ventricular tachycardia; pts, patients; PVB, premature ventricular beats; RBBB, right bundle branch block; RV, right ventricular; SAECG, signal-averaged electrocardiogram; SVT, sustained ventricular tachycardia.

aMinor diagnostic criteria.

bMajor diagnostic criteria.

*Brown–Forsythe statistic.

Table 2

Echocardiographic parameters of arrhythmogenic right ventricular cardiomyopathy patients classified according to primary endpoint at follow-up

 No. of patients with available data Global population (n = 96) Group A patients (n = 20) Group B patients (n = 76) P-value 
Right heart 
 RVEDAI (cm2/m292 17±5 18±6 16±4 NS 
 RV dilation [RVEDA >24 cm2, (%)]a 92 79 78 79 NS 
 RV FAC (%) 92 29±12 23±11 30±12 0.03 
 RV dysfunction [FAC <33%, (%)]a 92 64 90 58 0.01 
 Severe RV dysfunction [FAC <26 %, (%)]b 92 46 74 38 0.006 
 RV aneurysms (%)b 92 71 65 72 NS 
 Regional RV hypokinesia (%)a 92 92 90 93 NS 
 RAAI (cm2/m292 12±6 18±8 11±5 0.003 
 Significant tricuspid regurgitation (%) 84 15 35 0.004 

 
Left heart 
 LVEDDI (mm/m296 29±4 30±4 29±4 NS 
 LVESDI (mm/m296 20±5 22±6 19±4 0.04* 
 LV dilation (%) (LVEDVI >76 mL/m292 14 20 13 NS 
 LVEDVI (mL/m292 56±19 56±24 56±18 NS* 
 LVESVI (mL/m292 26±14 31±20 25±12 NS* 
 LVEF (%) 92 56±12 47±12 58±11 0.001 
 LV dysfunction (%) 92 26 47 20 0.01 
 Mild LV dysfunction (%) 92 15 NS 
 Moderate LV dysfunction (%) 92 10 NS 
 Severe LV dysfunction (%) 92 20 0.004 
 LV wall motion abnormalities (%) 92 45 65 40 0.04 
 LV aneurysms (%) 92 NS 
 LADI (mm/m296 19±3 20±3 18±3 0.003 
 LAAI (cm2/m292 11±3 12±5 11±2 NS* 
 Significant mitral regurgitation (%) 84 NS 
 No. of patients with available data Global population (n = 96) Group A patients (n = 20) Group B patients (n = 76) P-value 
Right heart 
 RVEDAI (cm2/m292 17±5 18±6 16±4 NS 
 RV dilation [RVEDA >24 cm2, (%)]a 92 79 78 79 NS 
 RV FAC (%) 92 29±12 23±11 30±12 0.03 
 RV dysfunction [FAC <33%, (%)]a 92 64 90 58 0.01 
 Severe RV dysfunction [FAC <26 %, (%)]b 92 46 74 38 0.006 
 RV aneurysms (%)b 92 71 65 72 NS 
 Regional RV hypokinesia (%)a 92 92 90 93 NS 
 RAAI (cm2/m292 12±6 18±8 11±5 0.003 
 Significant tricuspid regurgitation (%) 84 15 35 0.004 

 
Left heart 
 LVEDDI (mm/m296 29±4 30±4 29±4 NS 
 LVESDI (mm/m296 20±5 22±6 19±4 0.04* 
 LV dilation (%) (LVEDVI >76 mL/m292 14 20 13 NS 
 LVEDVI (mL/m292 56±19 56±24 56±18 NS* 
 LVESVI (mL/m292 26±14 31±20 25±12 NS* 
 LVEF (%) 92 56±12 47±12 58±11 0.001 
 LV dysfunction (%) 92 26 47 20 0.01 
 Mild LV dysfunction (%) 92 15 NS 
 Moderate LV dysfunction (%) 92 10 NS 
 Severe LV dysfunction (%) 92 20 0.004 
 LV wall motion abnormalities (%) 92 45 65 40 0.04 
 LV aneurysms (%) 92 NS 
 LADI (mm/m296 19±3 20±3 18±3 0.003 
 LAAI (cm2/m292 11±3 12±5 11±2 NS* 
 Significant mitral regurgitation (%) 84 NS 

FAC, fractional area change; LAAI, left atrial area indexed; LADI, left atrial diameter indexed; LV, left ventricular; LVEDDI, left end-diastolic diameter indexed; LVEF, left ventricle ejection fraction; LVESDI, left end-systolic diameter indexed; LVEDVI, left end-diastolic volume indexed; LVESVI, left end-systolic volume indexed; RAAI, right atrial area indexed; RV, right ventricular; RVEDA, right ventricular end-diastolic area; RVEDAI, right ventricular end-diastolic area indexed.

aMinor diagnostic criteria.

bMajor diagnostic criteria.

*Brown–Forsythe statistic.

Approximately one-third of the patients (27%) were diagnosed during an asymptomatic phase (family screening or abnormal ECG). The most frequent clinical presentation was related to arrhythmias [palpitations (41%), syncope (15%), and cardiac arrest (3%)], followed by HF symptoms (17%). An advanced NYHA class was present in a minority of cases (5%).

ECG abnormalities were frequent (78%), most commonly characterized by negative anterior T-waves (51%). Ventricular arrhythmias (documented on ECG during the episode or on Holter monitoring) were frequent (61 patients; 61%), characterized by predominant left bundle branch block pattern: sustained ventricular tachycardia (VT) in 23%, non-sustained VT in 34%, and frequent ventricular ectopic beats in 48%. Supraventricular arrhythmias were documented in 19% (atrial fibrillation in 4%).

On two-dimensional echocardiography (available in 92 patients), RV abnormalities were present in all of them, characterized by RV hypokinesis (92%), and RV aneurysms (71%), mainly located at subtricuspid (43%) and apical (39%) areas. Right ventricular enlargement and systolic dysfunction were present in 79 and 64% of patients, respectively.

Left ventricular involvement was present in 45 patients and was characterized by the following: LV wall motion abnormalities (45%), LV systolic dysfunction (26%), and LV dilatation (14%). Left ventricular aneurysms were found in only two cases, both located at the apex. Significant MR was found in 2 of 84 patients studied with Doppler (3%), and significant TR in 14 of 84 (15%). Intra-cardiac thrombi were observed in two patients, one located within the right atrium and the other at the RV apex.

Probands, with respect to family members, were more symptomatic and more frequently presenting with HF.

Endpoint data

During a mean follow-up of 128 ± 92 months (median = 120 months), 20 patients reached a primary endpoint (Group A): 12 of them died of cardiac cause (SD: six patients, refractory HF: six patients), one died of unknown causes (included in group A, see methods), and seven underwent HTx. Two patients died of a non-cardiac cause (neoplasm), and were censored.

Twelve patients received an ICD [nine for secondary prevention following aborted SD (one patient) or sustained VT (eight patients) and three for primary prevention of SD]. Appropriate ICD interventions (for sustained VT) were observed in 3 of 12 implanted patients (25%). No aborted SD was observed during follow-up in any patient.

The long-term outcome of the whole population with ARVC is illustrated in Figure 1. Event-free-survival from cardiac death or HTx at 2, 5, and 10 years was 96, 87, and 79%, respectively.

Figure 1

Kaplan–Meier curve for cardiac death or transplant-free survival of arrhythmogenic right ventricular cardiomyopathy population (cumulative proportion of surviving patients with arrhythmogenic right ventricular cardiomyopathy from diagnosis to cardiac death or heart transplantation). D/HTx, cardiac death/heart transplantation; Pts, patients.

Figure 1

Kaplan–Meier curve for cardiac death or transplant-free survival of arrhythmogenic right ventricular cardiomyopathy population (cumulative proportion of surviving patients with arrhythmogenic right ventricular cardiomyopathy from diagnosis to cardiac death or heart transplantation). D/HTx, cardiac death/heart transplantation; Pts, patients.

The variables associated with worse prognosis at univariate analysis are shown in Table 3.

Table 3

Clinical and laboratory variables associated with cardiovascular death/HTx at univariate analysis

 HR 95% CI P-value 
Ordinal dysfunctiona 2.50 1.52–4.09 <0.001 
NYHA III–IV 8.15 2.18–30.53 0.002 
Heart failure symptoms 3.90 1.54–9.87 0.004 
Epsilon wave 4.03 1.59–10.27 0.003 
LVEF (%) 0.93 0.89–0.96 <0.001 
LV dysfunction 3.92 1.57–9.78 0.003 
Severe LV dysfunction 5.67 1.81–17.75 0.003 
LADI (mm/m25.23 1.43–19.3 0.01 
RVFAC (%) 0.95 0.91–0.99 0.02 
RV dysfunction 5.48 1.27–23.74 0.02 
Severe RV dysfunction 4.93 1.72–14.12 0.003 
Significant TR 5.59 2.17–14.41 <0.001 
ACE-inhibitors 3.16 1.18–8.43 0.02 
Digitalis 5.79 2.38–14.00 <0.001 
Diuretics 14.27 5.59–36.43 <0.001 
Amiodarone 3.46 1.42–8.38 0.006 
 HR 95% CI P-value 
Ordinal dysfunctiona 2.50 1.52–4.09 <0.001 
NYHA III–IV 8.15 2.18–30.53 0.002 
Heart failure symptoms 3.90 1.54–9.87 0.004 
Epsilon wave 4.03 1.59–10.27 0.003 
LVEF (%) 0.93 0.89–0.96 <0.001 
LV dysfunction 3.92 1.57–9.78 0.003 
Severe LV dysfunction 5.67 1.81–17.75 0.003 
LADI (mm/m25.23 1.43–19.3 0.01 
RVFAC (%) 0.95 0.91–0.99 0.02 
RV dysfunction 5.48 1.27–23.74 0.02 
Severe RV dysfunction 4.93 1.72–14.12 0.003 
Significant TR 5.59 2.17–14.41 <0.001 
ACE-inhibitors 3.16 1.18–8.43 0.02 
Digitalis 5.79 2.38–14.00 <0.001 
Diuretics 14.27 5.59–36.43 <0.001 
Amiodarone 3.46 1.42–8.38 0.006 

CI, confidence interval; FAC, fractional area change; HR, hazard ratio; HTx, heart transplantation; LADI, left atrial diameter indexed; LV, left ventricular; LVEF, left ventricle ejection fraction; NYHA, New York Heart Association functional class; RV, right ventricular.

aOrdinal dysfunction: 0, no systolic ventricular dysfunction; 1, RV dysfunction; 2, biventricular dysfunction (see text).

Group A patients had a more severe clinical presentation, characterized by a higher frequency of HF and advanced NYHA class as well as higher percentage of epsilon waves on ECG with respect to Group B patients; in addition, they were treated more frequently with amiodarone and HF drugs. At echocardiography, patients of Group A had more frequently RV as well as LV systolic dysfunctions, atrial enlargement, and significant TR on colour-Doppler.

A progressive increase of mortality and HTx rate (P = 0.003) was observed in the presence of RV and biventricular dysfunction, respectively (Figure 2).

Figure 2

Kaplan–Meier survival curves of arrhythmogenic right ventricular cardiomyopathy patients classified according to the ‘ordinal ventricular dysfunction’ (dashed line = no ventricular dysfunction; dotted line = right ventricular dysfunction; solid line = biventricular dysfunction). Cardiac death or transplant-free-survival significantly decreased with increasing the ventricular dysfunction score. D/HTx, cardiac death/heart transplantation; dysf, dysfunction; Pts, patients, RV, right ventricular.

Figure 2

Kaplan–Meier survival curves of arrhythmogenic right ventricular cardiomyopathy patients classified according to the ‘ordinal ventricular dysfunction’ (dashed line = no ventricular dysfunction; dotted line = right ventricular dysfunction; solid line = biventricular dysfunction). Cardiac death or transplant-free-survival significantly decreased with increasing the ventricular dysfunction score. D/HTx, cardiac death/heart transplantation; dysf, dysfunction; Pts, patients, RV, right ventricular.

On Cox multivariate analysis, considering Model 1 (Table 4), the three significant independent predictors of death/HTx were: significant TR [hazard ratio (HR): 7.6; 95% confidence interval (CI): 2.6–22.0; P < 0.001], RV dysfunction (HR: 4.12; 95% CI: 1.01–18.0; P = 0.05), and amiodarone treatment (HR 3.4; 95% CI: 1.3–8.8; P = 0.01).

Table 4

Independent predictors for cardiac death/heart transplantation (Model 1)

Model 1 HR 95% CI P-value AUC 
Significant tricuspid regurgitation 7.60 2.60–22.0 <0.001 0.78 
Amiodarone 3.40 1.30–8.80 0.01 
Right ventricular dysfunction 4.12 1.0–18.0 0.05 
Model 1 HR 95% CI P-value AUC 
Significant tricuspid regurgitation 7.60 2.60–22.0 <0.001 0.78 
Amiodarone 3.40 1.30–8.80 0.01 
Right ventricular dysfunction 4.12 1.0–18.0 0.05 

AUC, area under the curve; CI, confidence interval; HR, hazard ratio.

In Model 2 (Table 5), ordinal dysfunction was found to be a powerful predictor of death/HTx (HR: 6.3; 95% CI: 2.17–17.45; P < 0.001), along with significant TR and amiodarone treatment.

Table 5

Independent predictors for cardiac death/heart transplantation (Model 2)

Model 2 HR 95% CI P-value AUC 
Significant tricuspid regurgitation 5.09 1.86–13.93 <0.001 0.84 
Amiodarone 3.72 1.43–9.67 0.007 
Ordinal ventricular dysfunctiona 6.30 2.17–17.45 <0.001 
Model 2 HR 95% CI P-value AUC 
Significant tricuspid regurgitation 5.09 1.86–13.93 <0.001 0.84 
Amiodarone 3.72 1.43–9.67 0.007 
Ordinal ventricular dysfunctiona 6.30 2.17–17.45 <0.001 

AUC, area under the curve; CI, confidence interval; HR, hazard ratio.

aOrdinal ventricular dysfunction: 0, no ventricular dysfunction; 1, RV dysfunction; 2, biventricular dysfunction.

The incremental prognostic accuracy of Model 2 compared with Model 1 was shown by comparing the ROC curves (area under the curve 0.84 vs. 0.78, respectively; P = 0.04; Figure 3). No significant difference in terms of transplant-free-survival between probands and familial cases was observed.

Figure 3

Comparison of the predictive accuracies between Model 1 and Model 2. The area under the curve of receiver operating curve of Model 2 incorporating the ordinal ventricular dysfunction score was significantly higher with respect to that of Model 1 (see text for explanation). AUC, area under the curve.

Figure 3

Comparison of the predictive accuracies between Model 1 and Model 2. The area under the curve of receiver operating curve of Model 2 incorporating the ordinal ventricular dysfunction score was significantly higher with respect to that of Model 1 (see text for explanation). AUC, area under the curve.

Discussion

We analysed a large cohort of patients, consecutively enrolled in the last 30 years in a tertiary referral centre for the study of HF and cardiomyopathies. Our results permitted a comparison of the presentation and the clinical-laboratory characteristics of ARVC patients with those of previous studies.1,9,11–14 Furthermore, the natural history of the disease was observed, and early prognostic stratification models were defined.

Clinical and laboratory characteristics at the time of enrolment

This study supports some clinical and laboratory features of ARVC that already emerged from previous studies.11–13 The disease has a considerable variability of clinical presentation.1,2,4,5,9,10,14,15,24 A familial pattern was found in approximately half of the cases. This percentage is higher than previously reported4,14,15 and is probably the result of systematic family screening.

Together with RV dysfunction, our echocardiographic data showed a frequent LV involvement. This finding, previously described by our group,8 was also confirmed in other study populations.10–16 Most recently, Sen-Chowdhry et al.11 defined a new nosological entity within the ARVC group known as ‘left-dominant arrhythmogenic cardiomyopathy’ with prevalent or even exclusive LV involvement. The high proportion of LV involvement in our population could reflect an advanced disease stage typical of patient selection due to referral to a Cardiomyopathy service.

Natural history and prognosis

In our population, the overall mortality was 15% during a median 10-year follow-up. The incidence of cardiovascular deaths plus HTx was ∼1.9/100 patients per year, similar to that reported by Hulot et al.14 but higher than that found by Nava et al.17 This difference may be explained by the fact that our population comes from a tertiary centre where patients are frequently referred for advanced cardiac failure.

Role of biventricular involvement in prognostic stratification

The presence of RV dysfunction at diagnosis was found to be an independent predictor of death/HTx. Hulot et al.14 were the first to describe a worse prognosis in patients with RV dysfunction. Our study emphasizes that long-term outcome is influenced not only by the presence of RV dysfunction, a classical abnormality found in ARVC, but also by LV dysfunction, as observed by others.14,16 Furthermore, we clearly demonstrate the incremental prognostic value of LV dysfunction in the risk stratification of patients affected by ARVC.

Possible mechanisms by which right ventricular and left ventricular dysfunction may increase the risk of life-threatening arrhythmias and heart failure

Major arrhythmic events and SD can occur in both early and advanced stages of disease, albeit due to presumably different mechanisms. In the so-called ‘concealed’ phase, characterized by little or no evidence of heart disease; it was hypothesized that gap junction remodelling25 may account for the high arrhythmogenicity in the absence of ventricular structural abnormalities.

Subsequently, in the overt stage of disease, the fibro-fatty replacement of the myocardium can favour re-entry circuits leading to potentially life-threatening arrhythmias. The progressive myocyte loss and replacement with fibro-fatty tissue, due to intrinsic genetically determined structural pathology and stress, are also responsible for the gradual dilation and functional RV and LV impairment that eventually result in progressive HF.

Potential adverse prognostic significance of amiodarone treatment

In our study, anti-arrhythmic treatment with amiodarone was found to be an independent predictor of mortality.

This finding seems in contrast to the results of a recent study by Marcus et al.26 where amiodarone emerged as protective against VT in ARVC; however, in comparison with ours, in the above-mentioned study the length of follow-up was too short (1.3 ± 1.1 years) to evaluate the long-term effect of amiodarone and the potential impact of widely known, time-related adverse effects of this particular treatment.27

Most importantly, in our study, amiodarone was indicated for the secondary prevention of life-threatening arrhythmias in selected, relatively young patients. Thus, it is possible that amiodarone treatment could be selected as a predictor of poor outcome as a proxy of the complex arrhythmic burden, which indicated its administration. This was especially true in the case of patients with highly arrhythmogenic phenotype who were enrolled before the demonstration of efficacy of ICD and for whom amiodarone was the only valid anti-arrhythmic option. Furthermore, our findings underline the increasing need for appropriate selection of the ARVC patients who could benefit most by an ICD implantation in the long term.

Role of significant tricuspid regurgitation

To our knowledge, our study is the first to demonstrate an adverse prognostic impact of significant TR in ARVC. Tricuspid regurgitation, in patients with dilated and dysfunctioning RV, as in the case of MR in dilated cardiomyopathy, is usually functional and secondary to RV and right atrial remodelling and it may contribute to the worsening of HF by both increasing RV filling pressure and lowering RV forward stroke volume.

Study limitations

In the present study, RV size and systolic function were assessed by measuring two-dimensional echocardiography RV areas and FAC. This method, although correlated with RV volumes and EF,21 is potentially inaccurate due to the complex geometry of this chamber. The recent advent of three-dimensional echocardiography, which correlates favourably with magnetic resonance,28,29 may overcome this limitation and be employed in future studies with ARVC. Genetic characterization was not available in the majority of our study patients. Therefore, potential differences in clinical spectrum and prognostic impact of different genetic substrates cannot be ascertained.10 Furthermore, since our patient population with ARVC comes from a tertiary centre where patients are frequently referred for advanced HF and complex diseases, they probably do not represent an unselected ARVC population. In addition, in the present study, we analysed only clinical and laboratory data at enrolment. Since the true onset of the disease in the single patient is unknown, our evaluation was performed at different stages of the disease in different patients. In this respect, an extensive follow-up study to assess the progression of clinical and echo-Doppler RV and LV dysfunctions and their possible prognostic role is advisable.13

In our population, cardiac arrest represented the first manifestation of the disease in three patients. Only one of these patients was subsequently implanted with an ICD. The remaining two patients were diagnosed in the early 1990s when ICD was not yet available. They remained stable in anti-arrhythmic treatment, and therefore, a conservative approach was chosen. None of our patients experienced an aborted SD during the follow-up period, and appropriate ICD interventions were due to VT without cardiac arrest. Therefore, these patients were not included in Group A.

After the initial submission of this paper, revised diagnostic criteria of ARVC were published.30 Our patients were diagnosed using the traditional criteria; however, the diagnosis of ARVC was confirmed by testing the new criteria in all patients.

Conclusions

Arrhythmogenic right ventricular cardiomyopathy is characterized by a variable clinical presentation. In addition to electrical instability, in a selected patient population with advanced disease, HF and LV involvement appear to be common features of the disease. Their assessment at the time of presentation can provide prognostic stratification of patients. The main independent long-term predictors of death or HTx that emerged from our data were RV dysfunction, significant TR, biventricular involvement, and amiodarone treatment.

Therefore, a prognostic stratification that considers not only the presence of RV dysfunction but also LV dysfunction increases the power of the prognostic model. The finding that amiodarone therapy is associated with a poor long-term outcome suggests the presence of an electrical instability resistant to first-line anti-arrhythmic treatment and, thus, characterized by an intrinsic malignancy.

Acknowledgement

We thank Dr. Frank Marcus for his valuable contribution in reviewing this paper.

Conflict of interest: none declared.

References

1
Thiene
G
Corrado
D
Basso
C
Arrythmogenic right ventricular cardiomyopathy/dysplasia
Orphanet J Rare Dis
 , 
2007
, vol. 
2
 pg. 
45
 
2
Sen-Chowdry
S
Lowe
MD
Sporton
SC
McKenna
WJ
Arrhythmogenic right ventricular cardiomyopathy: clinical presentation, diagnosis and management
Am J Med
 , 
2004
, vol. 
117
 (pg. 
685
-
695
)
3
Maron
BJ
Towbin
JA
Thiene
G
Antzelevitch
C
Corrado
D
Arnett
D
Moss
AJ
Seidman
CE
Young
JB
Contemporary definitions and classification of the cardiomyopathies: an American Heart Association Scientific Statement from the Council on Clinical Cardiology, Heart Failure and Transplantation Committee; Quality of Care and Outcomes Research and Functional Genomics and Translational Biology Interdisciplinary Working Groups and Council on Epidemiology and Prevention
Circulation
 , 
2006
, vol. 
113
 (pg. 
1807
-
1816
)
4
Hamid
MS
Norman
M
Quraishi
A
Firoozi
S
Thaman
R
Gimeno
JR
Sachdev
B
Rowland
E
Elliott
PM
McKenna
WJ
Prospective evaluation of relatives for familial arrhythmogenic right ventricular cardiomyopathy/dysplasia reveals a need to broaden diagnostic criteria
J Am Coll Cardiol
 , 
2002
, vol. 
40
 (pg. 
1445
-
1450
)
5
Nava
A
Thiene
G
Canciani
B
Scognamiglio
R
Daliento
L
Buja
G
Martini
B
Stritoni
P
Fasoli
G
Familial occurrence of right ventricular dysplasia: a study involving nine families
J Am Coll Cardiol
 , 
1988
, vol. 
12
 (pg. 
1222
-
1228
)
6
Moric-Janiszewska
E
Markiewicz-Łoskot
G
Review on the genetics of arrhythmogenic right ventricular dysplasia
Europace
 , 
2007
, vol. 
9
 (pg. 
259
-
266
)
7
MacRae
CA
Birchmeier
W
Thierfelder
L
Arrhythmogenic right ventricular cardiomyopathy: moving toward mechanism
J Clin Invest
 , 
2006
, vol. 
116
 (pg. 
1825
-
1828
)
8
Pinamonti
B
Sinagra
G
Salvi
A
Di Lenarda
A
Morgera
T
Silvestri
F
Bussani
R
Camerini
F
Left ventricular involvement in right ventricular dysplasia
Am. Heart J
 , 
1992
, vol. 
123
 (pg. 
711
-
724
)
9
Corrado
D
Basso
C
Thiene
G
McKenna
WJ
Davies
MJ
Fontaliran
F
Nava
A
Silvestri
F
Blomstrom-Lundqvist
C
Wlodarska
EK
Fontaine
G
Camerini
F
Spectrum of clinicopathologic manifestations of arrhythmogenic right ventricular cardiomyopathy/dysplasia: a multicenter study
J Am Coll Cardiol
 , 
1997
, vol. 
30
 (pg. 
1512
-
1520
)
10
Sen-Chowdhry
S
Syrris
P
Ward
D
Asimaki
A
Sevdalis
E
McKenna
WJ
Clinical and genetic characterization of families with arrhythmogenic right ventricular dysplasia/cardiomyopathy provides novel insights into patterns of disease expression
Circulation
 , 
2007
, vol. 
115
 (pg. 
1710
-
1720
)
11
Sen-Chowdhry
S
Syrris
P
Prasad
SK
Hughes
SE
Merrifield
R
Ward
D
Pennell
DJ
McKenna
WJ
Left-dominant arrythmogenic cardiomyopathy: an under-recognized clinical entity
J Am Coll Cardiol
 , 
2008
, vol. 
52
 (pg. 
2175
-
2187
)
12
Lindstrom
L
Nylander
E
Larsson
H
Wranne
B
Left ventricular involvement in arrhythmogenic right ventricular cardiomyopathy—a scintigraphic and echocardiographic study
Clin Physiol Funct Imaging
 , 
2005
, vol. 
25
 (pg. 
171
-
177
)
13
Pinamonti
B
Di Lenarda
A
Sinagra
G
Silvestri
F
Bussani
R
Camerini
F
Long-term evolution of right ventricular dysplasia-cardiomyopathy
Am Heart J
 , 
1995
, vol. 
129
 (pg. 
412
-
415
)
14
Hulot
J-S
Jouven
X
Empana
J-P
Frank
R
Fontaine
G
Natural history and risk stratification of arrhythmogenic right ventricular dysplasia/cardiomyopathy
Circulation
 , 
2004
, vol. 
110
 (pg. 
1879
-
1884
)
15
Dalal
D
Nasir
K
Bomma
C
Prakasa
K
Tandri
H
Piccini
J
Roguin
A
Tichnell
C
James
C
Russell
SD
Judge
DP
Abraham
T
Spevak
PJ
Bluemke
DA
Calkins
H
Arrhythmogenic right ventricular dysplasia. A United States experience
Circulation
 , 
2005
, vol. 
112
 (pg. 
3823
-
3832
)
16
Lemola
K
Brunckhorst
C
Helfenstein
U
Oechslin
E
Jenni
R
Duru
F
Predictors of adverse outcome in patients with arrhythmogenic right ventricular dysplasia/cardiomyopathy: long term experience of a tertiary care centre
Heart
 , 
2005
, vol. 
91
 (pg. 
1167
-
1172
)
17
Nava
A
Bauce
B
Basso
C
Muriago
M
Rampazzo
A
Villanova
C
Daliento
L
Buja
G
Corrado
D
Danieli
GA
Thiene
G
Clinical profile and long-term follow-up of 37 families with arrhythmogenic right ventricular cardiomyopathy
J Am Coll Cardiol
 , 
2000
, vol. 
36
 (pg. 
2226
-
2233
)
18
McKenna
WJ
Thiene
G
Nava
A
Fontaliran
F
Blomstrom-Lundqvist
C
Fontaine
G
Camerini
F
Diagnosis of arrhythmogenic right ventricular dysplasia/cardiomyopathy. Task Force of the Working Group Myocardial and Pericardial Disease of the European Society of Cardiology and of the Scientific Council on Cardiomyopathies of the International Society and Federation of Cardiology
Br Heart J
 , 
1994
, vol. 
71
 (pg. 
215
-
218
)
19
Buja
G
Estes
M
III
Wichter
T
Corrado
D
Marcus
F
Thiene
G
Arrhythmogenic right ventricular cardiomyopathy/dysplasia: risk stratification and therapy
Prog Cardiovasc Dis
 , 
2008
, vol. 
50
 (pg. 
282
-
293
)
20
Yoerger
DM
Marcus
F
Sherrill
D
Calkins
H
Towbin
JA
Zareba
W
Picard
MH
Multidisciplinary Study of Right Ventricular Dysplasia Investigators
Echocardiographic findings in patients meeting task force criteria for arrhythmogenic right ventricular dysplasia: new insights from the multidisciplinary study of right ventricular dysplasia
. J Am Coll Cardiol
 , 
2005
, vol. 
45
 (pg. 
860
-
865
)
21
Lang
RM
Bierig
M
Devereux
RB
Flachskampf
FA
Foster
E
Pellikka
PA
Picard
MH
Roman
MJ
Seward
J
Shanewise
JS
Solomon
SD
Spencer
KT
Sutton
MS
Stewart
WJ
Recommendations for chamber quantification: a report from the American Society of Echocardiography’ Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology
J Am Soc Echocardiogr
 , 
2005
, vol. 
18
 (pg. 
1440
-
1463
)
22
Zoghbi
WA
Enriquez-Sarano
M
Foster
E
Grayburn
PA
Kraft
CD
Levine
RA
Nihoyannopoulos
P
Otto
CM
Quinones
MA
Rakowski
H
Stewart
WJ
Waggoner
A
Weissman
NJ
Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography
J Am Soc Echocardiogr
 , 
2003
, vol. 
17
 (pg. 
777
-
802
)
23
Walter
SD
Eliasziw
M
Donner
A
Sample size and optimal designs for reliability studies
Stat Med
 , 
1998
, vol. 
17
 (pg. 
101
-
110
)
24
Thiene
G
Nava
A
Corrado
D
Rossi
L
Pennelli
N
Right ventricular cardiomyopathy and sudden death in young people
N Engl J Med
 , 
1988
, vol. 
318
 (pg. 
129
-
133
)
25
Kaplan
SR
Gard
JJ
Protonotarios
N
Tsatsopoulou
A
Spiliopoulou
C
Anastasakis
A
Squarcioni
CP
McKenna
WJ
Thiene
G
Basso
C
Brousse
N
Fontaine
G
Saffitz
JE
Remodeling of myocyte gap junctions in arrhythmogenic right ventricular cardiomyopathy due to a deletion in plakoglobin (Naxos disease)
Heart Rhythm
 , 
2004
, vol. 
1
 (pg. 
3
-
11
)
26
Marcus
GM
Glidden
DV
Polonsky
B
Zareba
W
Smith
LM
Cannom
DS
Estes
NA
III
Marcus
F
Scheinman
MM
Efficacy of antiarrhythmic drugs in arrhythmogenic right ventricular cardiomyopathy
J Am Coll Cardiol
 , 
2009
, vol. 
54
 (pg. 
609
-
615
)
27
Zipes
DP
Camm
AJ
Borggrefe
M
Buxton
AE
Chaitman
B
Fromer
M
Gregoratos
G
Klein
G
Moss
AJ
Myerburg
RJ
Priori
SG
Quinones
MA
Roden
DM
Silka
MJ
Tracy
C
Smith
SC
Jr
Jacobs
AK
Adams
CD
Antman
EM
Anderson
JL
Hunt
SA
Halperin
JL
Nishimura
R
Ornato
JP
Page
RL
Riegel
B
Blanc
JJ
Budaj
A
Dean
V
Deckers
JW
Despres
C
Dickstein
K
Lekakis
J
McGregor
K
Metra
M
Morais
J
Osterspey
A
Tamargo
JL
Zamorano
JL
ACC/AHA/ESC 2006 guidelines for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: a report of the American College of Cardiology/American Heart Association Task Force and the European Society of Cardiology Committee for Practice Guidelines: developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society
Circulation.
 , 
2006
, vol. 
114
 (pg. 
385
-
484
)
28
Niemann
PS
Pinho
L
Balbach
T
Galuschky
C
Blankenhagen
M
Silberbach
M
Broberg
C
Jerosch-Herold
M
Sahn
DJ
Anatomically oriented right ventricular volume measurements with dynamic three-dimensional echocardiography validated by 3-Tesla magnetic resonance imaging
J Am Coll Cardiol
 , 
2007
, vol. 
50
 (pg. 
1668
-
1676
)
29
Kjaergaard
J
Hastrup Svendsen
J
Sogaard
P
Chen
X
Bay Nielsen
H
Køber
L
Kjaer
A
Hassager
C
Advanced quantitative echocardiography in arrhythmogenic right ventricular cardiomyopathy
J Am Soc Echocardiogr
 , 
2007
, vol. 
20
 (pg. 
27
-
35
)
30
Marcus
FI
McKenna
WJ
Sherrill
D
Basso
C
Bauce
B
Bluemke
DA
Calkins
H
Corrado
D
Cox
MG
Daubert
JP
Fontaine
G
Gear
K
Hauer
R
Nava
A
Picard
MH
Protonotarios
N
Saffitz
JE
Sanborn
DM
Steinberg
JS
Tandri
H
Thiene
G
Towbin
JA
Tsatsopoulou
A
Wichter
T
Zareba
W
Diagnosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia: proposed modification of the Task Force Criteria
Eur Heart J.
 , 
2010
, vol. 
31
 (pg. 
806
-
814
)

Supplementary data

Comments

0 Comments