Background: Advances in endocardial device design have been limited by the inability to visualize the device-tissue interface. The purpose of this study was to assess the validity of an isolated heart approach, which allows direct ex vivo intracardiac visualization, as a research tool for studying endocardial pacing systems. Method of approach: Endocardial pacing leads were implanted in the right atria and ventricles of intact swine under fluoroscopic guidance. After collection of pacing and sensing performance parameters, the hearts were excised with the leads intact and reanimated on the isolated heart apparatus, and parameters again recorded. Results: Atrial ex vivo parameters significantly decreased compared with in vivo measurements: -wave amplitudes by 39%, slew rates by 61%, and pacing impedances by 42% ( for each). Similarly, several ventricular ex vivo parameters decreased: -wave amplitudes by 39%, slew rates by 62%, and pacing impedances by 31%. In contrast, both atrial ( vs ; ) and ventricular thresholds increased ( vs ; for all). Three distinct phenomena were observed at the lead-tissue interface. Normal implants (70%) demonstrated minimal tissue distortion and resulted in elevated impedance and threshold values. Three implants (13%) resulted in severe tissue distortion and/or tissue wrapping and were associated with highly elevated pacing parameters. Tissue coring occurred in four implants (17%) where the lead would spin freely in the tissue after overtorquing of the lead. Conclusions: The utility of the isolated heart approach was demonstrated as a tool for the design and assessment of the performance of endocardial pacing systems. Specifically, the ability to visualize device-heart interactions allows new insights into the impact of product design and clinical factors on lead performance and successful implantation.