Clinical predictors of cardiac events in patients with isolated syncope and negative electrophysiologic study

Background: Patients with syncope or near syncope of unknown etiology represent a great challenge to cardiologists. An initial symptomatic episode triggers a series of diagnostic analysis which remain unsatisfactory when negative. More invasive tools such as electrophysiologic testing yield only partial answers to risk stratification while the complementary implantable holter diagnostics are not usually considered until a recurrent episode is documented. Objective: This study targets predictors of significant cardiac rhythmic events in patients with a reported episode of syncope or near syncope presenting with negative diagnostics and electrophysiologic study results (EPS). A significant cardiac rhythmic event was defined as a combined end-point of (1) symptomatic AV block; (2) symptomatic conduction abnormalities requiring pacemaker therapy; (3) symptomatic sustained ventricular arrhythmia; and (4) sudden death. Methods: All patients undergoing EPS after a first episode of syncope or presyncope between January 1997 and December 2001 were included for analysis. The study population consisted of 329 pts (42.6% women), 21 to 96 years old (mean 70 +/- 15 years) referred for an EP study for syncope or near syncope. Results: Of the 329 patients who underwent EPS, 305 (92.7%) had follow-up data. The population, mean age 70 (+/- 15 years) and composed of 42% women, presented with hypertension (51.5%), diabetes mellitus (14.4%), hypercholesterolemia (30%), tobacco use (35%), a familial history of coronary heart disease (22%), history of stroke (4%), history of MI (12%), history of atrial fibrillation (10%), structural heart disease (17.4%), left ventricular ejection fraction 61 (11%) and ECG abnormalities (37%). These anomalies included right (RBBB) or left (LBBB) bundle branch blocks, left anterior fascicular block (LAFB), left posterior fascicular block (LPFB), bifascicular block (RBBB+LAFB) and traces of myocardial infarction. The mean follow-up was 31 +/- 20 months with 5% of patients recording significant cardiac rhythmic events (15/305): AV block requiring pacemaker therapy in 7 patients, sinus dysfunction in 4. sudden death in 3 and ventricular tachycardia in 1. Univariate analysis reveals structural heart disease, ECG abnormalities and LVEF associated with the risk of significant cardiac rhythmic events defined by the combined end-point. Multivariate analysis using a Cox model found that the only independent predictor of events was an ECG abnormality. The long-term risk of significant event in the subset with ECG abnormalities is of 10.6% (12/113). If unexplained syncope recurrence was included in the combined endpoint, ECG abnormality and LVEF were both determinants with a 13.3% (15/113) risk of a arrhythmic events analysis in the subset of patients presenting with ECG abnormalities and Cox model found ECG abnormality as the only independent predictor of event. Conclusions: This study demonstrated that an ECG abnormality is the only predictive variable associated with a significant arrhythmic event in patients with a lone episode of syncope or near syncope and a negative EPS. (c) 2005 Elsevier Ireland Ltd. All rights reserved.