Risk Stratification in Arrhythmic Right Ventricular Cardiomyopathy Without Implantable Cardioverter-Defibrillators

Abstract

Objectives

The primary objective of this study is risk stratification of patients with arrhythmic right ventricular cardiomyopathy (ARVC).

Background

There is a need to identify those who need an automatic implantable defibrillator (ICD) to prevent sudden death.

Methods

This is an analysis of 88 patients with ARVC from three centers who were not treated with an ICD.

Results

Risk factors for subsequent arrhythmic deaths were pre-enrollment sustained or nonsustained ventricular tachycardia (VT) and decreased left ventricular function.

Conclusion

These factors serve as proposed guidelines for implantation of an ICD in patients with ARVC to prevent sudden death.

Objectives

Ventricular arrhythmias are a characteristic of ARVC. Patients may present with palpitations due to ventricular premature beats or with VT.¹ A well-documented feature of this disease is that patients may have ventricular fibrillation resulting in sudden cardiac death (SCD) as the first or an early manifestation or during the course of the disease.² This latter feared complication is a major basis for the recommendation of ICD implantation. However, there is a significant complication rate of ICDs that are implanted for many years.³ Therefore, risk stratification to identify patients who need an ICD to prevent SCD would be useful clinical information.

Introduction

There have been a number of studies that have risk stratified patients based on appropriate ICD therapy as an indication of risk.^4-11 However, it is well documented that counting the number of patients who receive ICD treatment may greatly overestimate the efficacy of ICDs in preventing SCD. Of the 458 patients with non-ischemic cardiomyopathy who were randomized to ICD treatment vs no ICD implantation, the number of shocks delivered was greatly in excess of lives saved as compared with controls without ICDs.^12-13 Estimate of excess of lives saved by ICD therapy may be even greater in patients with ARVC who have normal left ventricular function and can tolerate VT at rates of > 200 BPM for many hours without syncope or cardiac arrest. This data was confirmed in the National Institutes of Health (NIH) sponsored study of ARVC, since there were 14 patients with clinical VT at a rate of 215 ± 21 BPM.¹⁴ An alternative approach to risk stratification is to study patients with a definite diagnosis of ARVC who were not implanted with an ICD and to analyze their presenting clinical features and subsequent risk of arrhythmic death. This is the approach used in this study.

Methods

Data from 88 probands with definite ARVC based on the modified Task Force Criteria¹⁵ who did not receive ICDs was analyzed and their clinical features evaluated to determine risk stratification. These were patients from the Multidisciplinary Study of Right Ventricular Dysplasia, funded by the NIH,¹⁴ the Familial Cardiomyopathy Registry of Trieste, Italy,¹⁶ and a registry from the University Medical Center of Utrecht, The Netherlands.¹⁷ The 88 probands, included 60 males, and 28 females. Medical and family history, physical examination, electrocardiogram (ECG), Holter monitoring ECG, and echocardiogram were performed on all index patients. Arrhythmias were recorded either on ECG or 24/48 hour Holter monitoring: The data was analyzed separately for each center due to the fact that there was a marked difference in the percent of patients who received ICDs among the centers. The analysis was also performed with the combined data.

The NIH sponsored Multidisciplinary Study of Right Ventricular Dysplasia was a prospective study from 2000 to 2008 that enrolled 137 probands with newly diagnosed ARVC.¹⁴ Twenty-eight patients (20%) did not receive an ICD. Of those, 22 probands were included in the study. Six other enrolled probands without an ICD met the original Task Force Criteria¹⁸ but did not meet the Modified Task Force Criteria¹⁵ and were excluded from this analysis. The average age at enrollment was 38.4 ± 15.9 years. The mean follow-up was 2.7 ± 1.5 years. The pre-enrollment clinical characteristics included syncope (3) sustained VT (4) nonsustained VT (5) or absence of documented arrhythmia (10) (Table 1 and Table 2) Antiarrhythmic drugs were prescribed in 4 and beta blockers in 10. One patient was prescribed a combination of antiarrhythmic drugs and beta blockers. Seven of the 22 received no antiarrhythmic or beta blocker therapy. None of the patients had an ablation procedure.

Table 1.

Clinical Features of Patients who Died Suddenly Without an ICD

Code/ Date of Enrollm ent	Age at Enro ll- ment	F/u (m)	Gend er	Rhyt hm	Inverte d T waves in the precor dial leads	Q wave Inf. Lead s	SAE CG	Sync ope	NYH A Class	LVE F baseli ne (%)	LVE DD baseli ne (cm)	2D- echo RVF S (%)	Ind ex VT	NSus VT	AAD	Death
CV 3/6/1989	30	89	M	SR	Yes	Yes	No	No	1	35	5.3	20	Sus VT	Yes	Amiodar one	06/20/1 996
FM- 8/3/1976	16	74	M	SR	Yes	Yes	No	No	1	N/A	5	40	Sus VT	Yes	No	9/1/198 2
MG- 8/29/198 7	48	30	M	SR	Yes	Yes	No	No	1	51	6.1	58	Sus VT	Yes	Amiodar one	2/22/19 90
PC- 2/1/1997	25	16	F	SR	Yes	Yes	Yes	No	1	37.7	4.9	32	NS us VT	Yes	Sotalol	06/03/1 998
NLUTR2 _1 03/16/19 50	35	264	M	SR	V1-4	Unkn own	No	Yes	1	NA	NA	NA	Sus VT	Yes	Disopyr amide/ Sotalol	06/07/2 007

Legend: F/u, follow up in months; M, male; F, female; SR, sinus rhythm; inverted T waves V2-3 leads; NYHA, New York Heart Association, SAECG, signal-averaged electrocardiography; LVEF, left ventricular ejection fraction at baseline; LEDD, Left Ventricular End Diastolic Diameter; RVFS, right ventricular fractional shortening; NSus VT, non-sustained ventricular tachycardia; Sus VT, sustained ventricular tachycardia AAD, antiarrhythmic drug, N/A, not available.

Table 2.

Pre-enrollment Risk Factors Related to Survival Tabulated by Center

Pre enrollment risk factors	Survived N = 76	Died N = 12 (14%)
Syncope with no index VT	3a
Sustained VT and no syncope	4a 13b 5c	3 arrhythmicb 1 non cardiacb 1 heart failurec
Sustained VT and syncope	2b 3c	1 arrhythmicc
Non-sustained VT and no syncope	5a 12b 3c	1 arrhythmicb 3 heart failureb 1 non cardiacb
Non-sustained VT and syncope	2b 1c
Transient VF; no syncope	1b
No index VT and no syncope	10a 23b
Unknown ventricular arrhythmia (No syncope)	1b	1 heart failureb

^aMultidisciplinary Study of Arrhythmogenic Right Ventricular Dysplasia

^bFamilial Cardiomyopathy Registry, Trieste, Italy

^cUniversity Medical Center, Utrecht

There were 82 probands from the Familial Cardiomyopathy Registry, Trieste, Italy.¹⁶ Fifteen of the patients received ICDs. Of the 67 patients (82%) who did not receive ICDs, 13 were excluded from further analysis. Of these, 9 did not meet the Modified Task Force Criteria,¹⁵ one patient died less than two months after enrollment due to progressive heart failure and three were lost to follow up. The remaining 54 patients did not receive an ICD because they were enrolled before the ICDs became available, or before the current guidelines were published, or because they underwent heart transplantation. Details of the clinical characteristics of the patients without an ICD who died suddenly are reported in Table 1. Pre-enrollment arrhythmias included sustained VT in 15 of whom two had syncope. There were 14 patients with pre-enrollment nonsustained VT, including two with nonsustained VT and syncope. There were no patients with syncope without sustained or nonsustained VT. One patient had transient VF on the ECG monitor and one patient had non-specified VT on ECG. There were 23 patients with no index VT or syncope and one with an unknown arrhythmia and no syncope. The average age at enrollment was 35.9 ± 14.8 years. The mean follow up was 12 ± 8.3 years. Antiarrhythmic drugs alone were used in 27, beta blockers in 5, a combination of antiarrhythmics and beta blockers were used in 9. There were 12 who were not treated with antiarrhythmics or beta blockers. One patient with sustained VT was successfully treated with ablation. There were no clinical arrhythmias after ablation.

The University Medical Center at Utrecht had 49 probands, 36 with an ICD and 13 (27%) without an ICD.¹⁶ One of the 13 had no follow-up. There were 12 probands with ARVC who met Modified Task Force Criteria and who did not receive an ICD. The average age at enrollment was 47.5 ±12.2 years. The average follow up was 9.48 ± 5.9 years. Pre enrollment sustained VT and no syncope (5), non-sustained VT and no syncope (3), sustained VT and syncope (3), and nonsustained VT and syncope (1) was present in the 12 probands. Five were treated with antiarrhythmic drugs, one with beta blockers, and one with neither. Five were treated with both antiarrhythmics and beta blockers. One of the patients from The University Medical Center at Utrecht had an arrhythmic cause of death. There were four patients who had cardiac ablation for sustained VT that was present at enrollment. Two patients had ablation for VT at enrollment while two had ablation for recurrent VT during follow up. The ablation procedure was unsuccessful in the latter two patients who had recurrent VT after ablation. None of the patients received ICDs after ablation. All four patients were treated medically with different antiarrhythmics and beta blockers. One patient died 18 years after VT ablation due to hemodynamically unstable sustained VT during exercise and secondary refractory cardiogenic shock. Three patients were stable with medical treatment without sustained arrhythmias.

Statistical analysis

Cox Proportional Hazards Model Regression with survival data was used to identify variables associated with mortality in patients diagnosed with ARVC was A “frailty analysis” with a random effect was used to account for possible differences among sites. SAS/STAT 9.4 (SAS Institute, Cary, NC) and the SPSS statistical package, version 20, were used to calculate the various statistics (IBM Corp. Released 2011. IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY: IBM Corp.). Two-tailed tests were employed in all cases with an alpha level of 0.05.

Results

During the average follow up of 9.1 ± 7.7 years, there were 12 deaths (14%) in the 88 probands without an ICD, 10 males and 2 females.

Of the 36 patients who did not have sustained or nonsustained VT prior to enrollment in their respective registries, there were no deaths. There were 5 deaths attributed to ventricular arrhythmias.

There were no deaths during follow up in the Multidisciplinary Study of Arrhythmogenic Right Ventricular Dysplasia. There were 10 deaths during follow up of 12 ± 8.3 years in the Familial Cardiomyopathy Registry, Trieste, Italy: 4 were arrhythmic, 4 were from progressive heart failure and 2 were from non-cardiac causes. In addition, 4 received heart transplant during follow up. During 9.48 ± 5.9 years of follow up at the University Medical Center at Utrecht, there was 1 arrhythmic death and 1 death due to progressive heart failure.

The pre-enrollment risk factors related to survival from each of the three enrolling registries is shown in Table 2 and the data is summarized in Table 3. The relationship between index VT and all-cause mortality was close but not statistically significant (^χ2(0.999)=3.76, p=0.052), although the overall Likelihood Test was significant (^χ2(1.08)=6.56, p=0.012). The random effect for site was small (0.015) and also not significant (^χ2(0.077)=0.08, p=0.099). The survival curves for patients with and without an index VT are shown in Figure 1. The odds for VF and unknown types of VT were difficult to estimate due to the low incidence in this sample. Thus, of the 52 patients who had an index episode of VT (sustained or nonsustained VT) there were 5 arrhythmic deaths. Of the 36 patients who did not have an index arrhythmia, there were no arrhythmic causes of death. One of these patients died of congestive heart failure. Among the 5 patients who had an arrhythmic death the left ventricular ejection fraction (LVEF) at baseline was significantly lower (p=0.04) than those who did not have an arrhythmic death (Table 3).

Table 3.

Relation of Ventricular Arrhythmias to Arrhythmic Death by Center

Pre-enrollment Arrhythmias
Total number of patients	NIH Study N=22	Trieste N=54	Utrecht N=12		Total N=88	Arrhythmic Deaths N=5
No Ventricular Arrhythmias	13	23	0		36	0
Non Sustained Ventricular Tachycardia	5	14	4		23	1
Sustained Ventricular Tachycardia	4	17*	8		29	4

^*Includes one patient with self-terminating ventricular fibrillation on ECG monitor and one patient with ventricular tachycardia unspecified.

Figure 1.

Survival based on the presence of the index non-sustained or sustained VT. The all-cause mortality of ARVC patients with VT (either sustained or nonsustained) at enrollment was of borderline significance (p=.052).

For the survival analysis with the other variables of interest, there was no statistically significant relation between mortality and syncope prior to diagnosis (p = 0.560), sex (p = 0.768), age at diagnosis (p=0.911), inducibility at EPS (p = 0.555), extent of T-wave inversion (T-wave V1 to V5; all p > 0.05), index QRS duration, (p = 0.447) or beta blockers (p = .122) for the 76 survivors and 12 who died during follow up.

Discussion

The major finding of this study is that there is a considerable risk of death in patients with ARVC who have decresed LVEF. Sustained or nonsustained VT was also a risk factor. From this data, it would seem reasonable to advise ICD implantation in these patients. An equally significant finding is that there were no arrhythmic deaths in patients who did not have clinical sustained or nonsustained VT. This suggests that these patients may not need an ICD to prevent SCD since arrhythmic deaths were not observed in these patients. Left ventricular function should be considered in the discussion of whether patients may need an ICD to prevent an arrhythmic death.

There have been several early reports of the long term outcome of patients without ICDs prior to the accepted practice of implanting ICDs in ARVC patients. Leclercq et al reported data on 65 patients with presumed ARVC and sustained VT¹⁹. During a mean follow-up of 81 ± 40 months, there were 3 deaths including 1 sudden death. Canu followed 21 patients with ARVC for an average of 10.7 years.²⁰ All had clinical sustained VT. An ICD was implanted in one patient. Twenty of the patients had recurrent VT and there were 3 sudden deaths. They concluded that clinical sustained VT and/or right heart failure are poor prognostic factors. Hulot reported the outcome of 72 patients with ARVC during a mean follow-up of 8.1 ± 7.8 years.²¹ Only 10 patients received an ICD. Of these, six had aborted SCD or sustained VT with hemodynamic compromise prior to ICD insertion. Twenty-four patients died during follow-up (18.5%) including 7 SCDs. Clinical signs of right and left heart failure were independently associated with the cardiovascular deaths. Patients without VT had the best prognosis.

Wichter et al, reviewed data from 60 patients treated with ICDs that were inserted primarily for resuscitated cardiac arrest or sustained VT.²² Of these, 18% had appropriate treatment for recurrent VT. They concluded that there was a beneficial role for ICD therapy in selected high risk patients with ARVC. They documented that there were 53 adverse events attributed to the implanted ICDs, including 37 of the 60 patients (62%) who had ICD related infection or malfunction and/or failure of the ICD system. In a series of 313 patients with ARVC, Peters found that the clinical course of ARVC is uneventful in primarily asymptomatic patients.⁶

Pinamonti et al, also documented that ARVC patients with severe RV and LV dysfunction had a poor long term prognosis for cardiac death.¹⁶ A multi-center study of 132 patients with ARVC who had ICDs implanted.⁴ found that a history of cardiac arrest or VT with hemodynamic compromise and younger age as well as LV involvement were independent predictors of VF/flutter. A subsequent study by Corrado et al, of 106 patients with ARVC treated with an ICD.¹⁰ concluded that prophylactic ICDs may not be indicated in asymptomatic patients because of their low arrhythmic risk regardless of familial sudden death and programmed ventricular stimulation findings. In their study programmed ventricular stimulation had a low predictive accuracy for ICD treatment.

Finally, Bhonsale et al, reported that, of 84 patients with definite or probable ARVC who had ICDs implanted for primary prevention, prediction of appropriate ICD therapy included induced VT at electrophysiological study.¹¹ Nonsustained VT and PVC count > 1000/24 hours by Holter monitor were also significant predictors of ICD therapy. The same authors reported risk stratification in 215 probands and family members who all had a desmosomal mutation. Of these, ICDs were implanted in 105 patients. Risk factors for sustained ventricular arrhythmias included the extent of precordial T-waves inversions, PVC count on Holter monitor, proband status and male sex.²³ An analysis of 69 patients with ARVC, and pathogenic mutations concluded that absence of significant electrical abnormalities at baseline by ECG and Holter monitoring predicted the absence of sustained ventricular arrhythmias during a mean follow-up of 5.8 ± 4.4 years.²⁴

Data from these studies indicates that risk factors for appropriate ICD therapy include patients with a history of SCD and those with sustained or nonsustained VT, as well as decreased LV function. However, the number and need for ICD treatment to prevent arrhythmic death may be overestimated due to the inclusion of patients who can tolerate rapid VT for long periods. Many of the VTs in patients with these characteristics may be self-terminating. Other established risk factors include decreased LV function and markedly decreased RV function.

A major incentive for attempting to find alternative therapies to ICDs is that the disease is slowly progressive and many patients may have a nearly normal life span. When ICDs are inserted in patients who are in their 30s or 40s, one can anticipate a number of required changes of the ICDs as well as other complications including pocket and lead infection, lead malfunction and inappropriate shocks, not to mention the psychological impact of living with an ICD.^22,25-26

Risk stratification for ARVC may change due to the recently reported success of epicardial ablation that can eliminate VT in a large percentage of patients with sustained VT.^27-29 Restriction of exercise, particularly a decrease in competitive endurance exercise may also decrease arrhythmic risk.³⁰

Non-sustained VT was found to be a factor for risk stratification. In addition, a recent analysis by Link et al analyzed the predictors of ventricular arrhythmias by interrogation of ICDs in the Multidisciplinary Study of Arrhythmogenic Right Ventricular Dysplasia.³¹ In this report, pre-enrollment spontaneous sustained VT or VF were significant factors to predict life threatening arrhythmias. Other risk factors included decreased LVEF, abnormal signal averaged ECG, T-wave inversion in the inferior leads, and induced ventricular arrhythmias. As previously mentioned, VT/VF induced during electrophysiological study is controversial as a risk factor. Potential risk factors that were not considered include the presence of premature ventricular beats by Holter monitor, and VT during exercise testing.

The present study is unique since it is the only report of risk stratification in patients who have ARVC by the modified Task Force Criteria and did not have ICDs. This approach was utilized because of the known deficiencies of risk stratification in patients with ARVC based on appropriate ICD therapy.

One interesting finding in this study was the difference in the percent of ARVC patients who did not receive ICDs. In the Multi-Center Study of Right Ventricular Dysplasia, 20% did not receive ICDs. In the medical center in Utrecht, 27% of the patients were not treated with an ICD and in the Familial Cardiomyopathy Registry in Trieste, 82% of the patients did not receive an ICD because they were enrolled before ICD became available or before the current guidelines.

Limitations of the study

There is the possibility that antiarrhythmic drugs or beta blocker use may have skewed the results of this analysis. However, Marcus et al, found that neither sotalol nor beta blockers were associated with an increased or decreased risk of ventricular arrhythmias in an analysis of 95 patients with ARVC.³²

A strength, but also a limitation of this study, is the analysis of outcomes in a highly selected population. The present analysis was limited to probands who met modified Task Force Criteria for ARVC and did not have ICDs implanted. Family members with ARVC were not included because the anticipated number of arrhythmic events including sudden death have reported to be few in this category and would therefore require a considerably larger number of patients to identify risk predictors.

The reasons the ICDs were not implanted was not documented systematically. As mentioned in the Methods, 54 patients from the Trieste did not receive ICDs because they were enrolled before ICDs were available or because the current guidelines were available or before the current guidelines were published or because they underwent heart transplantation. Some of the patients from the Netherlands did not receive an ICD because this was an exceptional and expensive treatment that was not considered indicated in patients with hemodynamically stable sustained VTs. Only 20% of patients from the NIH Registry did not have an ICD implanted. The specific reasons are not known. At least some of the patients refused this therapy.

Risk stratification based on spontaneous nonsustained VT and sustained VT may change with time during the course of the disease and should be reassessed to determine the presence of these risk factors. It is not known how frequently this should be done but every three years could be considered. In addition, since exercise has been well documented to exacerbate the signs and symptoms of ARVC, patients should avoid prolonged or competitive exertion.

The results of this analysis need to be confirmed since the numbers of patients are relatively few even though the data were obtained from three centers that enrolled a large number of patients with ARVC.

Conclusions

In this analysis of 88 patients with ARVC who met modified Task Force criteria but did not have ICDs, the risk factors for arrhythmic death were the baseline presence of decreased LV function as well as the presence of nonsustained or sustained VT. There were no arrhythmic deaths in patients who did not have these arrhythmias. This information should assist in risk stratification.

Table 4.

Evaluation of Risk Factors Based on Left Ventricular and Right Ventricular Size and Function by Echocardiogram of the 88 ARVC Patients at Baseline who did not Receive an ICD.

Arrhythmic Death	No Arrhythmic Death	P-value
Mean Age 29 ± 5 years	Mean Age 36 ± 15 years	(p=0.4)
RV Fractional Shortening 37 ± 19%	RV Fractional Shortening 27 ± 11%	(p=0.19)
LV Ejection Fraction 41 ± 8%	LV Ejection Fraction 56 ± 12%	(p=0.04)
LVEDD 5.3 ± 0.5cm	LVEDD 5.3 ± 0.7cm	(p=0.9)

LVEDD = Left Ventricular and End Diastolic Diameter

Clinical perspectives.

Traditional treatment to prevent sudden death in patients with ARVC is implantation of an automatic defibrillator. However a large percentage of these patients will not have sudden death. The consequences of ICD therapy that would be present for many years is to treat VT that is well tolerated and may resolve spontaneously. Evaluation of patients with ARVC who receive appropriate ICD therapy may greatly overestimate the number of patients who would have ventricular arrhythmias that do not result in sudden death.

Translational outlook.

This study provides guidelines to define which patients with ARVC are likely to have sudden death and to treat them appropriately. These data require confirmation.

Condensed Abstract.

This is an analysis of 88 patients from three centers who met modified Task Force Criteria for arrhythmogenic right ventricular cardiomyopathy (ARVC) and were not treated with implanted cardioverter defibrillators (ICDs). Result risk factors for subsequent arrhythmic deaths were pre-enrollment decreased left ventricular function and possible pre-enrollment sustained or nonsustained ventricular tachycardia (VT). This information provides a basis for risk stratification for patients with ARVC.

Abbreviations and Definitions

ARVC

arrhythmogenic right ventricular cardiomyopathy

EPS

electrophysiology study

ICDs

implantable cardioverter defibrillators

LV

left ventricle

LVEF

left ventricular ejection fraction

RV

right ventricle

SCD

sudden cardiac death

VF

ventricular fibrillation

VT

ventricular tachycardia