ICD for Syncope in Dilated Cardiomyopathy
Abstract & Commentary
Synopsis: Implantable defibrillator implantation should be considered given the high prevalence of appropriate shocks, which are considered to be surrogates for sudden death.
Source: Knight BP, et al. J Am Coll Cardiol 1999; 33:1964-1970.
Knight and colleagues prospectively collected data in patients with nonischemic dilated cardiomyopathy who were referred to the electrophysiology laboratory at their center with a diagnosis of unexplained syncope and then received implantable cardioverter defibrillators (ICDs). Over a six-year period, 14 such patients were identified. Their mean age was 53 and their mean ejection fraction was 26 ± 11%. Each patient underwent a complete history and physical exam before receiving the ICD. Seven of the 14 underwent tilt table testing with a normal response. Patients were monitored in the hospital and nine of the 14 had asymptomatic nonsustained ventricular tachycardia detected. Electrophysiologic studies were performed. None of the patients had inducible sustained monomorphic ventricular tachycardia (VT) or other marked abnormalities in their conduction system. These 14 patients were then compared to a second group of 19 patients with nonischemic cardiomyopathy, each of whom had suffered a documented cardiac arrest with ventricular fibrillation. These patients also received ICDs. All of the patients in the syncope group received ICDs that stored RR intervals or intracardiac electrograms. Twelve of the patients in the arrest group also received ICDs with storage capability, but seven received devices before this feature was available. Devices in both groups were usually programmed as single-zone devices with relatively high rate cutoffs for arrhythmia detection. Back-up VVI pacing was available in most of the ICDs implanted but three patients in each group also had separate, dedicated pacemakers. The primary end point for the comparison between the two groups was an appropriate defibrillator shock either confirmed as ventricular tachycardia after analysis of intra-cardiac electrograms or RR intervals or by symptoms of syncope, presyncope, or rates of more than 200 bpm in those with devices without electrogram storage.
Appropriate defibrillator shocks occurred in seven of the 14 patients in the syncope group over a follow-up period of 24 ± 13 months. In the arrest group, eight of 19 patients during a mean follow-up of 47 ± 41 months received defibrillator shocks. The actuarial incidence of appropriate shocks at one and two years was 36% and 43% in the syncope group and 10% and 21% in the arrest group. Inappropriate shocks for supraventricular arrhythmias were seen in seven of 14 patients in the syncope group compared to four of 19 patients in the arrest group. There was no significant difference in mortality between the two groups. Twenty-eight percent of patients in the syncope group and 32% of the patients in the arrest group died during follow-up. Most of the deaths were due to progressive heart failure. One patient in each group underwent transplantation. Both of these latter patients had received appropriate shocks.
Among patients in the syncope group, shocks were associated with lower ejection fractions and higher New York Heart Association functional class. However, the number of patients in the two groups was too small to allow identification of meaningful risk predictors.
Knight et al note that recent practice guidelines from the American College of Cardiology and the American Heart Association did not consider syncope of undetermined cause in the absence of inducible ventricular tachycardia to be an indication for defibrillator implantation. They argue that their data suggest that such implantation should be considered given the high prevalence of appropriate shocks, which they considered to be surrogates for sudden death, in their population.
Comment by John P. DiMarco, MD, PhD
Unexplained syncope in patients with structural heart disease is a frequently perplexing problem. In some cases, an obvious etiology can be identified from the initial history and physical examination. More often, one is left with a broad differential diagnosis that includes both relatively benign and potentially life-threatening conditions. Electrophysiologic studies have been proposed to be useful in patients with syncope, and induction of sustained VT in such patients has been accepted as an indication for ICD implant. The yield of programmed ventricular stimulation in patients with a clinical episode of sustained monomorphic VT and a prior myocardial infarction is quite high and this serves as justification for this recommendation in patients with unexplained syncope. In patients with nonischemic myopathies, responses to stimulation are more variable, and a negative study in such patients with syncope has less predictive accuracy. The data presented here by Knight et al would suggest that empiric ICD implant is the appropriate strategy to follow in these patients.
The data presented here clearly show that these patients were a high-risk group with a prognosis similar to patients with a history of cardiac arrest. Unfortunately, it’s impossible to estimate from the data presented whether this approach should be used in all patients with syncope and nonischemic cardiomyopathy. We don’t know what factors led to these patients getting an ICD. Syncope is common in patients with advanced cardiomyopathy and it has many causes. We cannot know how many patients with cardiomyopathy and syncope were not referred for EP study nor are we told how many patients, with or without inducible VT, were treated with other forms of therapy. Without these data, it will remain difficult to make informed decisions for these patients.
Several clinical trials in patients with ischemic and nonischemic cardiomyopathies are now being conducted to investigate the potential role of ICD implantation for the primary prevention of sudden death. The results of these studies should provide accurate estimates of the benefits of ICD implantation in patients without prior sustained arrhythmias. It is likely that these data will also be directly applicable to patients with syncope in whom ventricular function and functional class may still be the most important predictors of prognosis.
Unexplained syncope in dilated nonischemic cardiomyopathy patients is best treated with:
a. prophylactic pacemaker.
b. beta blockers.
c. ICD placement.