Outcome of patients with nonischemic dilated cardiomyopathy and unexplained syncope treated with an implantable defibrillator

Patient characteristics

A total of 33 patients with a clinical diagnosis of NICM were included in the study population. Since 1993, all patients at our institution with NICM and unexplained syncope who had a nondiagnostic electrophysiology test have been treated with an implantable defibrillator. During this time, 14 consecutive patients were prospectively identified who had at least one episode of unexplained syncope (mean number of syncopal episodes 1.6 ± 0.9; range 1 to 3) and a negative electrophysiology test and who underwent implantation of a defibrillator (Syncope Group). During the same time period, no patients with NICM and unexplained syncope who were referred to our institution for electrophysiologic testing had inducible ventricular tachycardia. Twelve of the 14 patients with syncope (86%) underwent cardiac catheterization, and 10 patients had no obstructive coronary artery disease; 2 patients had a single stenosis of >50%, diffuse myocardial dysfunction that was disproportionate to the severity of coronary artery disease, and no prior history of myocardial infarction. One of the two patients who did not undergo cardiac catheterization was 17 years old and had no history of myocardial infarction. The remaining patient had severe global left ventricular dysfunction and no inducible ischemia by dobutamine stress echocardiography, consistent with NICM.

Using a prospective institutional database that includes all defibrillator recipients, 19 consecutive patients who had a cardiac arrest and underwent defibrillator implantation between 1988 and 1998 were retrospectively identified to serve as a control group (Arrest Group). Documentation that a cardiac catheterization was performed was available in 15 patients (79%), and in none of these patients was there obstructive coronary disease. There were no identifiable differences in the clinical characteristics of the patients in the Syncope Group compared to the Arrest Group (Table 1).

Antiarrhythmic drug therapy was discontinued in all patients after defibrillator implantation except for three patients in the Arrest Group in whom amiodarone therapy was continued at a dose of 200 mg/day. Angiotensin-converting enzyme inhibitor therapy was prescribed at the time of discharge in 13 (93%) of the patients in the Syncope Group compared to 13 (68%) of the patients in the Arrest Group (p = 0.2). Beta-blockers were prescribed at the time of discharge for one patient (7%) in the Syncope Group, compared to four of the patients (21%) in the Arrest Group (p = 0.4).

Evaluation for syncope and cardiac arrest

Patients in the Syncope Group were hospitalized after the episode of syncope that preceded defibrillator implantation and were included in this study only if the etiology of syncope remained undetermined after a thorough evaluation. Each patient underwent a complete history and physical examination, including carotid sinus massage, routine blood measurements and electrocardiography. No patient had nausea or fatigue associated with the syncopal episode to suggest vasodepressor syndrome (9). Seven patients had a negative head-up tilt-table test. In addition, each patient underwent continuous electrocardiographic (ECG) monitoring for 4.9 ± 4.0 days before defibrillator implantation, and asymptomatic, nonsustained ventricular tachycardia was present in nine (64%) patients. Twenty-four-hour ambulatory ECG monitoring was not performed. All patients in the Syncope Group underwent complete electrophysiologic testing. Standard assessments of sinus and atrioventricular nodes and His-Purkinje function were performed, including sinus node recovery time, atrioventricular node block cycle length, atrioventricular node effective refractory period, ventriculoatrial block cycle length, and His-ventricular conduction time. Programmed electrical stimulation of the right ventricle was performed using a protocol consisting of four extrastimuli at two right ventricular sites and three basic drive cycle lengths, as previously described (10). Programmed stimulation was repeated at one right ventricular site during infusion of 2 μg/min of isoproterenol. Patients were included in the Syncope Group if they had a normal electrophysiology test or demonstrated nonspecific abnormalities considered to be nondiagnostic (11).

Evaluation for patients who survived cardiac arrest included hospitalization and a complete history and physical examination. Potentially reversible causes of cardiac arrest were excluded. Fifteen patients (79%) in the Arrest Group underwent a complete electrophysiology test. Sustained monomorphic ventricular tachycardia was inducible in two of the patients (13%).

Implantable defibrillator systems

Each patient in the Syncope Group underwent implantation of a nonthoracotomy defibrillator that provided stored R-R intervals or intracardiac electrograms (Guidant, St. Paul, Minnesota; models 1705, 1720, 1740, 1762 and 1763), or both. The mean implant defibrillation energy requirement was 13 ± 5 J, and there were no implant-related complications.

Twelve patients (63%) in the Arrest Group underwent implantation of a defibrillator that provided stored R-R intervals and/or intracardiac electrograms (Guidant, St. Paul, Minnesota; models 1625, 1705, 1715, and 1720; Medtronic, Minneapolis, Minnesota; models 7219, 7223; St. Jude Medical, Sunnyvale, California; Ventritex models V100 and 7223), and seven patients (37%) received a non-event-recording defibrillator (Guidant; models 1500, 1520, 1550, and 1600). Four patients received an epicardial lead system and 15 patients received a nonthoracotomy lead system. The mean implant defibrillation energy requirement was 15 ± 8 J in the Arrest Group. Implantation was complicated by a pneumothorax in one patient. Seven patients underwent elective generator replacement during follow-up for battery depletion.

Implantable defibrillator programming

In patients in the Syncope Group, the defibrillator was programmed to a one-zone device in 11 patients, with a tachycardia detection rate cutoff of 180 ± 11 beats/min. In the remaining three patients, the defibrillator was programmed to a two-zone device to allow for empiric antitachycardia pacing therapy in the first zone, with a first zone rate cutoff of 180 ± 18 beats/min and a second zone rate cutoff of 207 ± 13 beats/min. Ventricular demand (VVI) pacing was programmed to 40 beats/min in 13 patients and 50 beats/min in 1 patient. The percent of time patients received bradycardia pacing from the defibrillator was available in 10 patients and was reported in increments of 1%.

Among the patients in the Arrest Group, the defibrillator was programmed to a one-zone device in 17 patients, with a tachycardia detection rate cutoff of 174 ± 9 beats/min. In the remaining two patients, the defibrillator was programmed to a two-zone device, with a first zone rate cutoff of 138 ± 4 beats/min and a second zone rate cutoff of 175 ± 7 beats/min, with empiric antitachycardia pacing programmed in the first zone. The VVI pacing was programmed to 40 beats/min in 12 patients, 50 beats/min in 1 patient and was not available for the remaining 6 patients.

Concomitant pacemaker therapy

Two patients in each group were previously treated with a dedicated pacemaker in addition to an implantable defibrillator 5.3 ± 3.9 months before defibrillator implantation. In addition, one patient in the Syncope Group received a pacemaker for bradycardia-dependent oversensing by the defibrillator (12), and one patient in the Arrest Group received a pacemaker for carotid sinus hypersensitivity after defibrillator implantation.

Follow-up

The duration of follow-up was calculated from the time of defibrillator implantation. Patients were followed in clinic every four months. Defibrillator function, device therapy history, and occurrence of syncope or presyncope were documented and evaluated during each clinic visit. Surviving patients or their primary physicians were contacted by telephone in May 1998 if they had not been seen in the University of Michigan Defibrillator Clinic within the previous three months. The medical records of all patients who died during follow-up were reviewed.

Analysis of defibrillator therapies

Among the seven patients in the Arrest Group with non-event-recording implantable defibrillators, an appropriate shock was defined as one associated with syncope, presyncope, or a rate over 200 beats/min. Among the remaining 12 patients in the Arrest Group, and for all patients in the Syncope Group, the stored intracardiac electrograms or R-R intervals of all implantable defibrillator events were reviewed. The rhythms that were associated with defibrillator therapy were categorized as sinus tachycardia, atrial fibrillation, or ventricular arrhythmias, based on visual inspection of intracardiac electrograms or analysis of the R-R intervals, a comparison with the electrogram obtained during the baseline rhythm, and the response to therapy (13). Monomorphic ventricular tachycardia was defined as a regular tachycardia with an electrogram morphology that differed from the baseline rhythm and was distinguished from atrial fibrillation by an R-R interval variability of <60 ms. Ventricular fibrillation and polymorphic ventricular tachycardia were defined as irregular tachycardias with polymorphic electrogram morphology. Appropriate defibrillator shocks were defined as shocks delivered for a ventricular tachyarrhythmia.

Statistical analysis

Continuous variables are expressed as mean ± SD and were compared using the Student ttest or Mann-Whitney U nonparametric test, as appropriate. Nominal variables were compared using the chi-square or Fisher exact test, as appropriate. Kaplan-Meier survival curves were constructed for each patient group to express overall morality and the time from defibrillator implantation to first appropriate shock. Patients who underwent cardiac transplantation were censored from the survival analysis at the time of the operation. A log-rank test was used to compare Kaplan-Meier curves. Univariate analysis was performed on clinical variables to determine predictors of appropriate defibrillator shocks. A p value of <0.05 was considered statistically significant.

Appropriate defibrillatory therapies

Appropriate defibrillator shocks occurred in 7 of 14 patients (50%) during a mean follow-up of 24 ± 13 months in the Syncope Group compared with 8 of 19 patients (42%) during a mean follow-up of 47 ± 41 months in the Arrest Group (Fig. 1; p = 0.1). Using Kaplan-Meier analysis, the mean duration from device implantation until the first appropriate shock was 32 ± 7 months (95% CI, 18 to 45 months) in the Syncope Group compared to 72 ± 12 months (95% CI, 48 to 96 months) in the Arrest Group (p = 0.1). The actuarial incidence of appropriate shocks at one and two years was 36% and 43%, respectively, in the Syncope Group, and 10% and 21%, respectively, in the Arrest Group.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Kaplan-Meier survival curves depicting freedom from appropriate defibrillator shocks for patients in the Syncope Group (solid line)compared to the Arrest Group (dotted line). Each vertical mark represents a censored event.

Among patients who received appropriate shocks, the mean time from defibrillator implantation to the first appropriate shock was 10 ± 14 months in the Syncope Group, compared with 48 ± 47 months in the Arrest Group (p = 0.06). The rate of the ventricular arrhythmia for the first appropriate shock was 257 ± 39 beats/min in the Syncope Group, compared with 258 ± 46 beats/min in the Arrest Group (p = 1.0). One patient each group received >10 shocks during follow up. Among the remaining patients who received appropriate shocks, there was a total of 2.2 ± 1.5 appropriate shocks in the Syncope Group, compared with 1.4 ± 0.8 shocks in the Arrest Group (p = 0.3). The mean duration of the ventricular arrhythmia resulting in a defibrillator shock was 11.8 ± 0.5 s in the Syncope Group. No patient in either group received antitachycardia pacing for a ventricular arrhythmia.

Seven patients (50%) in the Syncope Group received shocks for atrial fibrillation or sinus tachycardia, compared to four patients (21%) in the Arrest Group (p = 0.08). One patient in the Arrest Group received inappropriate shocks due to an epicardial sensing lead fracture and underwent endocardial lead placement. One of the two patients in each group who had a pacemaker before defibrillator implantation received appropriate shocks.

Five of the eight patients in the Arrest Group who received appropriate shocks had initially undergone implantation of a device without R-R intervals or stored electrograms. However, three of the five patients underwent generator replacement with a device that had retrievable R-R intervals before they received their first appropriate shock. The remaining two patients had not undergone generator replacement before their first shock but experienced presyncope with the shock.

One patient in the Syncope Group was treated with bradycardia pacing from the defibrillator for 9% of the time after amiodarone therapy was initiated for ventricular tachycardia. No other patient received bradycardia pacing for more than 0.5% of the time during follow up.

Mortality

There was no significant difference in mortality between the Syncope Group (4 of 14 patients, 28%) and the Arrest Group (6 of 19 patients, 32%; p = 0.8; Fig. 2). Using Kaplan-Meier analysis, the mean survival time was 40 ± 5 months (95% CI, 30 to 50 months) in the Syncope Group compared with 86 ± 13 months (95% CI, 61 to 111 months) in the Arrest Group (p = 0.08). In the Syncope Group, three of the four patients died of progressive congestive heart failure 6.3 ± 2.1 months after defibrillator implantation. The other patient died suddenly in an emergency department 36 months following implantation. In the Arrest Group, three of the six patients died of progressive congestive heart failure 12.3 ± 17.1 months after defibrillator implantation. A fourth patient died of asystole in an emergency department 17 months post-implantation and three hours after receiving three shocks for atrial fibrillation. The remaining two patients in the Arrest Group died of unclear reasons 9 and 13 months after defibrillator implantation. Two of the four patients in the Syncope group who died received at least one appropriate defibrillator shock three and five months before death, compared with none of the six patients in the Arrest Group who died. One patient in the Syncope Group underwent successful cardiac transplantation 14 months after defibrillator implantation and 2 months after receiving two appropriate shocks. One patient in the Arrest Group underwent cardiac transplantation 19 months after implantation and 7 months after receiving an appropriate shock.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Kaplan-Meier survival curves depicting overall mortality for patients in the Syncope Group (solid line)compared to the Arrest Group (dotted line). Each vertical mark represents a censored event.

Recurrent syncope

Among patients in the Syncope Group, an appropriate shock associated with syncope or presyncope occurred in one and four patients, respectively. One of the four patients who had presyncope associated with a shock also had episodes of presyncope that were not associated with device therapy. In this patient, a continuous loop recorder correlated the episodes with up to 18 beats of nonsustained monomorphic ventricular tachycardia at a rate of 200 beats/min. An additional patient had unexplained presyncope three years after defibrillator implantation that was not associated with a shock.

Patients with syncope and appropriate defibrillator shocks

Among patients in the Syncope Group, those who received appropriate shocks had a lower left ventricular ejection fraction (0.20 ± 0.07 vs. 0.31 ± 0.12; p = 0.04) and a worse New York Heart Association (NYHA) functional class (2.7 ± 0.8 vs. 1.4 ± 0.5; p = 0.02) compared to patients who did not receive appropriate shocks (Table 2). None of the three patients with an ejection fraction >35% received an appropriate shock during follow-up. However, no cutoff point could be identified that distinguished the patients who received shocks from those who did not. All six patients with NYHA class III heart failure received appropriate defibrillator shocks. The remaining patient who received appropriate shocks had NYHA class I symptoms. There were no significant differences in the clinical characteristics among patients in the Arrest Group who did and did not receive appropriate shocks.

Main findings

This study found that 50% of patients with NICM, unexplained syncope, and a negative electrophysiology test who were treated with an implantable defibrillator received appropriate defibrillator shocks for ventricular arrhythmias during a mean follow-up of two years, and that recurrent syncope and presyncope were primarily due to ventricular tachyarrhythmias. The incidence of appropriate shocks was similar to patients with NICM and a history of cardiac arrest.

Pacemaker therapy

An alternative treatment option to an implantable defibrillator in patients with NICM, unexplained syncope, and a nondiagnostic electrophysiology test includes empiric pacemaker therapy. Although a bradycardia may be responsible for sudden death in some patients with advanced heart failure (14), pacing therapy has not been shown to improve survival (15). In the present study, the implantable defibrillators in each patient in the Syncope Group provided fixed-rate backup bradycardia pacing, yet half of these patients still received defibrillator shocks for ventricular arrhythmias. In addition, only one patient in the Syncope Group received a significant amount of bradycardia pacing. However, the possibility that brief episodes of ventricular pacing therapy prevented recurrent syncope or sudden death in some patients cannot be excluded.

Predictors of appropriate defibrillator shocks

The ability to identify patients with NICM and unexplained syncope who will receive appropriate shocks after defibrillator implantation would be useful. All patients in the Syncope Group with NYHA class III functional status prior to defibrillator implantation received appropriate shocks during follow-up. One patient with class I symptoms, however, also received appropriate shocks. No patient in the Syncope Group with an ejection fraction >35% received an appropriate shock. However, only 3 of the 14 patients (21%) had an ejection fraction >35% and there was no cutoff point to differentiate patients who received appropriate shocks from those who did not. Therefore, although it appears that patients with more severe cardiomyopathy are more likely to receive appropriate shocks, the usefulness of clinical characteristics as predictors of appropriate defibrillator shocks among patients with NICM and unexplained syncope may be limited.

Prior studies

Electrophysiologic testing is frequently used to determine the cause of syncope (11,16–19). However, electrophysiologic testing may be less useful in selected subgroups of patients, including patients with dilated cardiomyopathy (3–6,11). The inability of electrophysiologic testing to risk-stratify asymptomatic patients with dilated cardiomyopathy (4–6), in contrast to patients with ischemic cardiomyopathy (20), highlights the limitations of programmed electrical stimulation. For patients with NICM who present with syncope, there are minimal data regarding the use of electrophysiologic testing (3,11). A natural history study of 99 patients with unexplained syncope and a nondiagnostic electrophysiologic test included only 6 patients with dilated cardiomyopathy (11). A study of 60 patients with advanced heart failure and syncope included 33 patients (55%) with NICM, and found the one-year risk of sudden death to be 45% irrespective of the etiology of syncope (3). These results suggest that electrophysiologic testing in patients with NICM has a low negative predictive value.

Two studies of the outcome of patients with NICM treated with an implantable defibrillator have been reported (21,22). Each study included a limited number of patients who presented with syncope. Ten patients in the study by Fazio et al. (21)presented with syncope. However, each patient had syncope that was documented to be due to ventricular tachycardia, and only two of the patients were noninducible at electrophysiology testing. Six of the 49 defibrillator recipients in the study by Grimm et al. (22)presented with syncope. It was not reported if spontaneous sustained ventricular tachycardia occurred in these patients. In addition, the results of electrophysiology testing were not specifically reported for the patients with syncope. The actuarial incidence of shocks in the present study among patients in the Syncope Group (45% at two years) was comparable to the results of both Fazio et al. (21)(42% at 18 months) and the results of Grimm et al. (22)(49% at three years).

Study limitations

A limitation of this study is the small sample size. A second limitation is the uncertainty associated with categorizing a defibrillator shock as appropriate in patients with devices that do not provide stored R-R intervals or electrograms. However, all patients in the Syncope Group initially had event-recording devices implanted compared to 63% of the patients in the Arrest Group. Therefore, if the incidence of appropriate shocks was overestimated, this was more likely to have occurred in the Arrest Group.

Clinical implications

Recent practice guidelines from the American College of Cardiology and the American Heart Association do not consider syncope of undetermined cause in the absence of inducible ventricular tachyarrhythmias to be an indication for defibrillator implantation (23). The present study does not prove that implantable defibrillators reduce mortality in patients with NICM and unexplained syncope. However, occurrences of syncope and presyncope largely in association with ventricular arrhythmias, and the similar incidence of appropriate shocks compared to patients with NICM and a previous cardiac arrest, support the use of defibrillators in this setting. The benefit of defibrillators compared to antiarrhythmic drugs in patients with documented life-threatening arrhythmias (24,25)and the low risk associated with nonthoracotomy defibrillator implantation (26,27)provide additional rationale for this approach. Further study is required to determine the effect of this approach on overall mortality and its cost-effectiveness (28).