Comparison of epicardial and endocardial programmed stimulation in patients at risk for ventricular arrhythmias after cardiac surgery.

Programmed ventricular stimulation was performed on 36 patients after recent cardiac surgery using implanted right ventricular epicardial temporary wires and with catheters positioned percutaneously at two right ventricular endocardial sites. Patients were followed for a mean of 18.5 months (range 3 to 36 months). Epicardial wires were nonfunctional in 10 patients (28%) due to excessively high pacing thresholds. Overall, 22 patients (61%) had inducible sustained ventricular tachycardia; epicardial wires were functional in 15 (68%) of these patients. Six patients without inducible ventricular tachycardia with epicardial stimulation were inducible using endocardial stimulation. Of the 24 patients in whom epicardial and endocardial ventricular stimulation could be performed, concordant results were obtained in only 17 (71%), despite similar epicardial and endocardial ventricular effective and functional refractory periods. A total of 14 arrhythmic events occurred during the follow-up period. Of the 22 patients with an inducible sustained ventricular tachycardia, 12 (64%) had subsequent arrhythmic events. Only 2 of the 14 noninducible patients had follow-up arrhythmic events, one of which was caused by medication proarrhythmia. Endocardial ventricular stimulation had a superior sensitivity (83% versus 30%, P < 0.0001) and an improved negative predictive value (86% versus 61%, P < 0.05) compared with epicardial ventricular stimulation. These results indicate that noninducibility using epicardial programmed ventricular stimulation does not reliably portend a low risk for recurrent ventricular tachyarrhythmias. Epicardial programmed stimulation, used alone, may be inadequate for postoperative electrophysiological evaluation of patients at risk for ventricular arrhythmias.