The prevalence of atrial arrhythmias in Arrhythmogenic Right
Christian F. Camm*, BM BCh, Cynthia A. James†, PhD, Crystal Tichnell†, MGC, Brittney Murray†,
MS, Aditya Bhonsale†, MD, Anneline S.J.M te Riele†‡, MD, Daniel P. Judge†, MD, Harikrishna
Tandri†, MD, Hugh Calkins†, MD FHRS
* New College, University of Oxford, Oxford, United Kingdom
† Department of Medicine, Division of Cardiology, Johns Hopkins University School of Medicine,
Baltimore, MD, USA
‡ Department of Medicine, Division of Cardiology, University medical Center Utrecht, Utrecht, the
Short Title: The prevalence of atrial arrhythmias in ARVD/C
Word Count: 5188
Address for Correspondence:
Hugh Calkins, MD, Sheikh Zayed Tower – Room 7125R, The Johns Hopkins Hospital, 1800 Orleans
Street, Baltimore, MD, USA 21287. Phone: 410-502-1953; Fax 410-614-1345; Email: firstname.lastname@example.org.
Sources of Funding:
The authors wish to acknowledge funding from the Alexandre Suerman Stipend (to ASJMtR), the
National Heart, Lung, and Blood Institute (K23HL093350 to HT), the St. Jude Medical Foundation, and
Medtronic Inc. The Johns Hopkins ARVD/C Program (ARVD.com) is supported by the Bogle
Foundation, the Healing Hearts Foundation, the Campanella family, and Wilmerding Endowments, and
the Dr. Francis P. Chiaramonte Private Foundation.
Disclosures: Dr. Calkins receives research support from Medtronic and St. Jude Medical. The other
authors report no conflict of interest.
Arrhythmogenic Right Ventricular Dysplasia/Cardiomyopathy (ARVD/C) is an inherited
cardiomyopathy, characterized by right ventricular dysfunction and ventricular arrhythmias. Limited
information is available concerning atrial arrhythmias in ARVD/C.
The purpose of this study was to characterize spontaneous atrial arrhythmias in a large registry population
of ARVD/C patients.
Patients (n=248) from the Johns Hopkins ARVD/C registry, meeting the diagnostic criteria and having
undertaken genotype analysis, were included. Medical records of each were reviewed to ascertain
incidence and characteristics of atrial arrhythmia episodes. Detailed demographic, phenotypic, and
structural information were obtained from registry data.
Thirty-five patients with ARVD/C (14%) experienced one or more types of atrial arrhythmia during a
median follow-up of 5.78 (IQR=8.52) years. Atrial fibrillation was the most common atrial arrhythmia,
occurring in 80% of ARVD/C patients with atrial arrhythmias. Patients developed atrial arrhythmias at a
mean age of 43.0±14.0 years. Atrial arrhythmia patients obtained a total of 22 inappropriate ICD shocks
during follow-up. Older age at last follow-up (p<0.001) and male gender (p=0.044) were associated with
atrial arrhythmia development. Patients with atrial arrhythmias had a higher occurrence of death
(p=0.028), heart failure (p<0.001), and left atrial enlargement on echocardiography (p=0.004).
Atrial arrhythmias are common in ARVD/C and present at a younger age than the general population.
They are associated with male gender, increasing age, and left atrial enlargement. Atrial arrhythmias are
clinically important as they are associated with inappropriate ICD shocks, and increased risk of both death
and heart failure.
Key Words: Arrhythmogenic Right Ventricular Dysplasia; Arrhythmogenic Right Ventricular
Cardiomyopathy; Atrial arrhythmia; Atrial Fibrillation;
AF Atrial Fibrillation
ARVD/C Arrhythmogenic Right Ventricular Dysplasia/Cardiomyopathy
ECG 12-lead Electrocardiogram
ICD Implantable Cardioverter-Defibrillator
IQR Inter-Quartile Range
PVC Premature Ventricular Complex
SVT Supraventricular Tachycardia
TFC Task Force Criteria
VF Ventricular Fibrillation
VT Ventricular Tachycardia
Arrhythmogenic right ventricular dysplasia/cardiomyopathy (ARVD/C) is an inherited cardiomyopathy
characterized by right ventricular dysfunction, ventricular arrhythmias, and an increased risk of sudden
cardiac death.(1–4) The severity and progression of ARVD/C is highly variable, ranging from
asymptomatic disease to severe heart failure.(5) Starting with the seminal description of plakoglobin
mutations,(6) we now know that ARVD/C is often a disease of the cardiac desmosome.(7) This structure
is essential for cardiac integrity and aids electromechanical coupling of cardiomyocytes.
A large number of studies have documented the high incidence of life-threatening ventricular arrhythmias
in patients with ARVD/C.(4,8–11) In contrast, relatively little attention has been focused on
supraventricular arrhythmias in patients with ARVD/C,(12–14) and existing studies are limited by sample
size, and a lack of genotype data. While unlikely to cause sudden death, atrial arrhythmias are nonetheless
associated with increased risk of mortality,(15) and may also be associated with an increased risk of
morbidity.(16) Therefore, the purpose of this study was to examine the prevalence and characteristics of
supraventricular arrhythmias in patients with ARVD/C. Particular attention is focused on determining the
relationship between the development of atrial arrhythmias, patient age, and the severity of ARVD/C.
The Johns Hopkins ARVD/C Program was established in 1999 with a goal of studying ARVD/C and
providing care for patients with this condition. Patients with definite or possible ARVD/C are enrolled in
a prospective registry. The study population was identified from this registry. Inclusion criteria for this
study were registry patients who had undertaken genotype analysis, and met the 2010 Task Force Criteria
(TFC) for ARVD/C.(17) All registry participants provided written informed consent; the study protocol
was approved by the Johns Hopkins School of Medicine Institutional Review Board.
Detailed clinical information regarding demographics, presentation, symptom onset, non-invasive, and
invasive studies was obtained for each patient. The medical history of each subject was obtained through
review of medical records, clinical evaluation, and patient interview. A detailed family history was
obtained through patient interview by genetic counsellors with a special interest in ARVD/C.
Comprehensive mutation testing was performed, using commercial or research genetic testing, on all
patients included in this study.(18) Genetic testing included sequencing of five ARVD/C-associated
desmosomal genes (PKP2, DSC2, DSG2, DSP, JUP) and the PLN and TMEM43 genes. The diagnosis of
ARVD/C was based on the presence of major and minor diagnostic criteria according to the revised TFC
2010.(17) The proband was defined as the first person in a family in whom ARVD/C diagnosis was
confirmed (i.e. fulfilling TFC for ARVD/C irrespective of family history).
Severity and extent of structural abnormalities were determined through review of echocardiography,
cardiac magnetic resonance imaging, and right ventricular angiography reports. Mitral and tricuspid
regurgitation was considered if rated as at least moderately severe on echocardiography reports.
Enlargement of the right and left atrial was defined as either present or absent based on the atrial diameter
The primary outcome was presence of atrial arrhythmias, ascertained from medical records, including
records of ICD interrogations. Atrial arrhythmias were defined as a narrow complex tachycardias with at
least one episode lasting ≥30 seconds. Where available, traces were examined and adjudicated in addition
to reports from the treating physician. Atrial arrhythmias appearing only during electrophysiology studies
were excluded. Type of arrhythmia, symptoms, and detection method were recorded. Atrial arrhythmias
were categorized into three groups; atrial fibrillation (AF), atrial flutter, and all other supraventricular
Secondary outcomes were the development of stage C heart failure, a sustained ventricular arrhythmic
event, ICD placement, and death. Stage C heart failure is defined as patients with current or past
symptoms of heart failure (e.g. dyspnea, fatigue) associated with underlying structural heart disease.(19)
A sustained arrhythmic event is a composite measure of the occurrence of sudden cardiac death,
spontaneous sustained VT/VF, or an appropriate ICD intervention for a sustained ventricular
arrhythmia.(9) Echocardiographic reports were reviewed for moderate-severe tricuspid regurgitation,
mitral regurgitation, and right and left atrial diameter.
Continuous variables are expressed as either mean +/- standard deviation or median (Inter-Quartile Range
(IQR)) and compared across groups using the independent Student t-test or Mann-Whitney U-test
respectively. Categorical variables are reported as frequency (percentage) and compared between groups
using the chi-squared or Fisher exact test. The cumulative probability of survival free of an atrial
arrhythmia was determined by the Kaplan Meier method, and differences in survival between groups
evaluated with the log-rank test. Binary logistical regression was used to control for age at last follow-up
and sex. A p-value <0.05 was considered significant. SPSS Version 19 (IBM, Chicago, Illinois) was used
for all statistical analysis.
Shown in Table 1 are the baseline demographic data of the 248 patients enrolled in this study. The mean
age at last follow up was 41.6±14.0 years, 131 (52.8%) were male, and 136 (54.8%) had known ARVD/C
associated genetic mutations (77.9% PKP2, see Supplementary Table 1).
Prevalence of supraventricular arrhythmias:
Among the 248 patients, 35 (14.1%) experienced one or more types of supraventricular arrhythmias (47
total arrhythmias) over a median follow-up period of 5.78 years (IQR=8.52). The distribution of
supraventricular arrhythmias are shown in Figure 1. AF was most common, observed in 28 of 35 patients
(80%), followed by atrial flutter (11, 31%) and other SVTs (8, 23%). As shown in Figure 1, 11 patients
experienced more than one type of atrial arrhythmia with one patient experiencing all three types.
The mean age at time of first atrial arrhythmia of 43±14 years. Thirty-two (68%) of the sentinel
arrhythmic events occurred while the patient was on anti-arrhythmic medications. Five of forty-seven
sentinel atrial arrhythmic events (11%) occurred prior to the first documented ventricular event (4 AF, 1
Atrial Flutter). There were a total of 22 inappropriate shocks recorded for atrial arrhythmias in 10
patients; six responding to AF (four patients), seven for Atrial Flutter (three patients), and nine for SVT
(four patients). Diagnosis method used for these atrial arrhythmias is shown in Figure 2. A comparison of
the clinical features of ARVD/C patients with AF, atrial flutter, and other types of supraventricular
arrhythmias is shown in Table 2. No difference was observed in the age at sentinel atrial arrhythmic
event, time since ARVD/C presentation, gender, gene carrier status, proportion on anti-arrhythmic
medication at time of sentinel atrial arrhythmic event, or proportion with an ICD at last follow-up
between different types of atrial arrhythmias.
AF accounted for 28 of 47 supraventricular arrhythmias experienced by our patient cohort (60%).
Twenty-seven of 28 arrhythmias (96%) were paroxysmal with a single patient suffering from persistent
AF. The average CHADS2 score was 0.68 (SD 0.84); 13 individuals had a CHADS2 score >0 (46%).
Anticoagulation was prescribed in 18 patients; warfarin (11), aspirin (5), and dabigatran (2). There were
no recorded adverse events related to anticoagulation in this patient group. The following thrombo-
emoblic events were recorded in the AF group: right ventricular thrombus (2), transient ischemic attack
(1), subclavian vein thrombosis (1), pulmonary embolism (1), and deep vein thrombosis (1). All but one
of these patients is currently anti-coagulated.
Comparison of patients with and without atrial arrhythmias
We compared clinical characteristics of ARVD/C patients with and without at least one supraventricular
arrhythmic event at last follow-up (Table 1). Patients experiencing atrial arrhythmias were older at the
date of last follow-up (49.7 ± 14.6 vs 40.3 ± 13.5 years respectively, p<0.001) and more likely to be male
(68.6% and 50.2% respectively, p=0.044) than patients without atrial arrhythmias. In bivariate analysis,
older age at diagnosis and longer duration of follow-up were also associated with having at least one atrial
arrhythmic event. However, binary logistic regression controlling for age at last follow-up and gender,
showed no significant association with remaining demographic features, including duration of follow-up.
Survival curves outlining the development of atrial arrhythmias illustrate that men have a lower lifetime
survival free from atrial arrhythmias than women (Figure 3).
Relationship Between Supraventricular Arrhythmias and Markers of Disease
Echocardiographic reports were available for 210 study participants (34 with atrial arrhythmias, 176
without). Moderate or severe tricuspid regurgitation was significantly associated with the presence of
atrial arrhythmias (29.4% vs. 13.6%, p=0.022). There was no association seen between atrial arrhythmias
and moderate or severe mitral regurgitation (Table 3). Binary logistical regression controlling for gender
and age at last follow-up removed the statistical significance of tricuspid regurgitation (B=0.750,
SE=0.470, p=0.110). Both right (52.9% vs. 27.8%, p=0.008) and left (38.2% vs. 10.2%, p<0.001) atrial
enlargement were significantly associated with the presence of atrial arrhythmias; following binary
logistical regression both right (B=0.929, SE=0.405, p=0.022) and left (B=1.352, SE=0.463, p=0.004)
atrial enlargement remained significant. Left ventricular enlargement was present in five patients with
atrial arrhythmias (14.3%). No statistically significant difference was seen between the rate of left
ventricular enlargement in those with left atrial enlargement (3/12, 25.0%) and those without (2/21,
9.5%), p=0.233. Left ventricular ejection fractions were similar between those with left atrial enlargement
(54.7%, SD = 8.1%, n=13) and those without (54.7%, SD=10.8%, n=21), p=0.994. The rate of
hypertension >140mmHg systolic or 90mmHg diastolic was not significantly different between those
with left atrial enlargement (3/21, 14.3%), and those without left atrial enlargement (3/12, 25.0%),
The presence of atrial arrhythmias was associated with heart failure (34.3% vs. 9.3% respectively,
p<0.001) and death (11.4% vs. 2.8% respectively, p=0.037). Additionally, patients with atrial arrhythmias
were significantly more likely to be placed on anti-arrhythmic therapies (82.9% vs. 41.3% respectively,
p<0.001). No other markers of disease severity were significantly associated (Table 4). Following binary
logistical regression analysis, controlling for age at last follow-up and gender, the presence of atrial
arrhythmias were significantly associated with heart failure (B=1.646, SE=0.455, p<0.001) and death
(B=1.576, SE=0.719, p=0.028).
The results of this study reveal that atrial arrhythmias are not uncommon in patients with ARVD/C,
occurring in 14% of patients. ARVD/C patients who develop atrial arrhythmias are, on average, older and
more likely to be male. The development of atrial arrhythmias was associated with structural changes
including left and right atrial enlargement.
Prior Studies of Atrial Arrhythmias in Patients with ARVD/C
The presence of atrial arrhythmias in ARVD/C was highlighted soon after the initial recognition of this
cardiomyopathy and has been confirmed in prior cases studies and case series.(12–14,20,21) Tonet et al.
were the first group to assess the presence of atrial arrhythmias in patients with ARVD/C.(14) Evaluation
of 72 ARVD/C patients presenting with ventricular arrhythmias demonstrated a 24% prevalence rate with
atrial fibrillation the most common arrhythmia (59%). The presence of right atrial enlargement and
tricuspid regurgitation were both suggested by this group as factors in the development of atrial
arrhythmias. In a similar population of 47 patients, Brembilla-Perrot et al. demonstrated a 15%
prevalence rate,(13) and an increased susceptibility to the development of atrial arrhythmias under
programmed electrical stimulation. Jaoude et al. assessed the progression of ECG tracings during the
follow-up of 74 patients with ARVD/C.(22) This yielded a 4% prevalence of atrial arrhythmias (three
patients). However, this study did not further categorize the arrhythmias or the patient presenting with
Recently, Chu et al. highlighted a prevalence rate of 42% in a retrospective analysis of 36 ARVD/C
patients undergoing ablation for ventricular arrhythmias.(12) There was a non-significant trend towards
older age in the atrial arrhythmia group; however, there was no difference in gender. Tricuspid
regurgitation was significantly associated with the development of atrial arrhythmias; in contrast,
association with right atrial enlargement was not significant.
Atrial Arrhythmias in the Johns Hopkins ARVD/C Registry
The results of the present study confirm and extend the findings of prior studies of atrial arrhythmias in
patients with ARVD/C. This is the first study directly assessing the presence and nature of atrial
arrhythmias in a large research database. Previous studies characterizing atrial arrhythmia type and
prevalence have focused on particular sub-populations of ARVD/C patients; those diagnosed with
ventricular arrhythmias,(14) or patients undergoing ventricular ablation.(12) The current study provides
novel insights using data from a large, specialist registry and highlights key features important to the
practicing physician regarding the development of atrial arrhythmias in this group.
The first important finding of this study is the prevalence of atrial arrhythmias (AF, Atrial Flutter, and
other SVTs) in patients with ARVD/C was 14%. This rate is similar to that seen by Brembilla-Perrot et
al. (14.9%).(13) The lower prevalence in this study when compared with Chu et al. (42%) and Tonet et
al. (24%) may relate to population selection. In these prior studies, patients were limited to those
undergoing ventricular ablation therapy and those presenting with ventricular arrhythmias
respectively,(12,14) both markers of more severe disease. Given the high proportion of paroxysmal atrial
fibrillation in our current study (27 of 28; 96%), the low rate of atrial arrhythmias demonstrated by Jaoude
et al. (4.1%) likely reflects the lack of comprehensive assessment of patient records examining for atrial
arrhythmias, which were instead identified solely through ECG assessment.(22)
Age and Gender
Atrial fibrillation, and other atrial arrhythmias, are age dependent and more likely to be seen in male
patients;(23,24) AF has a population prevalence of 21% by age 80.(23) Atrial arrhythmias were shown to
be similarly age and gender dependent within our ARVD/C population. Additionally, the mean age of
43.0±14.0 years illustrates a high disease burden in an age group that normally has a AF prevalence of
<0.5%.(25) Furthermore, our study has shown that older ARVD/C patients were more likely to have
experienced an atrial arrhythmia; this supports the earlier findings of Chu et al.(12) The mean age of 43
years in this population is lower than seen with previous studies - 51 and 52 years.(12,13) The lower age
range seen in our population may, in part, represent increased recognition of ARVD/C as a condition over
the past decade with better diagnostic clarification following improved Task Force Criteria published in
2010. Increased use of genetic screening in family members,(7) as suggested by Heart Rhythm Society
guidelines,(26) following proband diagnosis (and hence their meeting TFC at an earlier stage of disease)
may explain the significant association with negative genetic screening results and atrial arrhythmias (due
to the potential diagnosis of such individuals at an earlier stage of disease progression); an association that
was not maintained when regression controlled for age at last follow-up and gender. Unlike age, gender
has not been shown to be a significant predisposing factor in previous studies in this area.(12) These
results suggest the possibility that atrial arrhythmias are occurring in older individuals with a
predisposition to AF that is triggered by the presence of ARVD/C.
Diagnosis method And ICD
A large proportion of atrial arrhythmias were diagnosed in this study using implantable cardiac
defibrillators (both during routine follow-up and as a result of inappropriate shocks), and Holter monitors.
This suggests that many atrial arrhythmias in this population may be concealed with only 34.0% of all
atrial arrhythmias being first demonstrated using an ECG tracing. This is consistent among different atrial
arrhythmia types and has not been shown previously. Tonet et al. showed that the majority of their atrial
arrhythmias (82%) were diagnosed using a standard 12-lead ECG tracing.(14) The increased diagnostic
prevalence of other techniques likely results from their increased utilization in the diagnosis and
management of ARVD/C over the past two decades. However, the fact that 17% of all atrial arrhythmias
were initially diagnosed following an inappropriate ICD shock and the presence of 22 inappropriate
shocks within our atrial arrhythmia population highlights the need for better awareness and control of
It has been previously suggested that atrial arrhythmias may be associated with the presence or
development of more severe disease.(12,14) In this study, the presence of atrial arrhythmias showed a
variable association with factors suggestive of disease severity. Death and heart failure were both more
prevalent in the atrial arrhythmia group. However, it is not clear whether either atrial arrhythmias or heart
failure played a causative role in the development of the other. Although statistically significant, the small
number of deaths in our study population suggests that the clinical significance of this finding is less
clear. The increased utilization of anti-arrhythmic agents in the atrial arrhythmia group is potentially a
result of the atrial arrhythmia burden rather than suggestive of increased ventricular arrhythmia burden.
Although atrial arrhythmias in general, and AF in particular, are associated with an increased mortality
rate in the general population,(15) the increased rate of heart failure in this population likely added to this.
The pathophysiology behind the development of atrial arrhythmias in ARVD/C is far from clear. Atrial
enlargement is a known risk factor for atrial arrhythmias, especially atrial fibrillation.(27) Several groups
have previously suggested a role for the right atrial dilatation often seen in ARVD/C in the development
of atrial arrhymias.(12,14) The limited availability of cardiac MRI data within this population makes
structural analysis of the atria difficult. However, we demonstrated a correlation between atrial
arrhythmias and tricuspid regurgitation using echocardiographic imagining, in support of previous
studies.(12) However, this finding cannot fully explain the prevalence of atrial arrhythmias within the
ARVD/C population. Although, a correlation with both left and right atrial enlargement and atrial
arrhythmias was also identified, the known limitations of echocardiography when viewing the atria limit
the validity of these results.. Left atrial enlargement is a known effect of atrial fibrillation;(28) this
15 Download full-text
suggests that the structural changes seen, both in this study, and by previous groups may be a result of the
atrial arrhythmias rather than a causative factor.
Desmosome mutations, and resulting dysfunction, are known to be a major pathological mechanism in
ARVD/C. Desmosomes are found throughout the cardiac system including the atria. Evidence for a
pathological role of the atria in ARVD/C remains limited, however, support for a desmosomal basis to
atrial arrhythmias in this population is multi-factorial. A small autopsy case series (two patients with
ARVD/C) by Morimoto et al. showed the presence of fatty tissue within the sino-atrial node of both
patients.(29) Direct atrial involvement with ARVD/C is further suggested by the finding of atrial
involvement in 17% of cats in an ARVD/C model.(30) Additionally, the post-natal loss of desmoplakin
within the cardiac conduction system has been shown to lead to sino-atrial node dysfunction in mice.(31)
Furthermore, work by Platonov et al. has demonstrated evidence for altered electrical conduction within
the atria of ARVD/C patients.(32) Despite this suggestive evidence, data directly relating to atrial
involvement in patients with ARVD/C is lacking. The thin walled nature of the atria makes biopsy
unwise. Furthermore, despite cardiac MRI proving itself to be an excellent imaging modality for
ventricular involvement,(33) resolution of the atria remains poor. Without further evidence relating to a
direct role of the atria in ARVD/C, it remains uncertain whether atrial arrhythmias are due to desmosomal
dysfunction within the atria, right atrial enlargement as a result of right ventricular dysfunction, or a
combination of both mechanisms.
An atrial arrhythmia prevalence of 14% in this ARVD/C population highlights the need for vigilance
among treating clinicians. This is further supported by increased mortality and heart failure prevalence
associated with these patients. Furthermore, our study demonstrates that atrial arrhythmias are a common
cause of inappropriate shocks in patients with ARVD/C. This finding has important implications for
programming of ICDs to minimize the risk of inappropriate shocks should an atrial arrhythmia occur. The