Implantable cardioverter-defibrillators (ICDs) with nonthoracotomy lead systems are widely available, and are implanted either in the electrophysiology laboratory or the operating room. The purpose of this study was to prospectively evaluate the safety and efficacy of nonthoracotomy ICD implantation in an electrophysiology laboratory versus an operating room. During a 7-month period, 62 consecutive ICDs with nonthoracotomy lead systems were implanted in patients in an electrophysiology laboratory. During the next 10 months, 110 consecutive ICDs were implanted in patients in a surgical operating room. All ICD implantations were performed under general anesthesia by electrophysiologists. There were no differences in age (58 +/- 14 vs 62 +/- 12 years, p = 0.06), gender distribution (p = 0.3), frequency of structural heart disease (97% vs 97%, p = 0.9), ejection fraction (0.31 +/- 0.15 vs 0.29 +/- 0.13, p = 0.3), or presentation with cardiac arrest (65% vs 53%, p = 0.2) between patients undergoing ICD implantation in the electrophysiology laboratory and operating room, respectively. The rate of successful implantation and of complications for systems implanted in the electrophysiology laboratory (95% and 13%, respectively) and in the operating room (98% and 14%, respectively) were similar (p = 0.4 and p = 0.8, respectively). Specifically, the rate of infection (0% vs 4%, p = 0.3) and hematoma formation (2% vs 4%, p = 0.8) were not statistically significantly different. Three patients who had undergone ICD implantation in an operating room died within 30 days. ICDs with nonthoracotomy lead systems can be implanted with a similarly high rate of success and acceptable complication rate in the electrophysiology laboratory and in the operating room.