With the advent of smaller biphasic transvenous implantable Cardioverter defibrillators (ICDs) and the experience gained over the years, it is now feasible for electrophysioiogists to implant them safely in the abdominal or pectoral area without surgical assistance. Throughout the years, general anesthesia has been used as the standard technique of anesthesia for these procedures. However, use of local anesthesia combined with deep sedation only for defibrillation threshold (DFT) testing might further facilitate and simplify these procedures. The purpose of this study was to test the feasibility of using local anesthesia and compare it with the standard technique of general anesthesia, during implantation of transvenous ICDs performed by an electrophysiologist in the electrophysiology laboratory. For over 4 years in the electrophysiology laboratory, we have implanted transvenous ICDs in 90 consecutive patients (84 men and 6 women, aged 58 ± 15 years). Early on, general anesthesia was used (n = 40, group I), but in recent series (n = 50, group II) local anesthesia was combined with deep sedation for DFT testing. Patients had coronary (n = 58) or valvular (n = 4) disease, cardiomyopathy (n = 25) or no organic disease (n = 3), a mean left ventricular ejection fraction of 35%, and presented with ventricular tachycardia (n = 72) or fibrillation (n = 16), or syncope (n = 2). One-lead ICD systems were used in 74 patients, two-lead systems in 10 patients, andan AVICD in 6 patients. ICDs were implanted in abdominal (n = 17, all in group I) or more recently in pectoral (n = 73) pockets. The DFT averaged 9.7 ± 3.6 J and 10.2 ± 3.6 J in the two groups, respectively (P = NS) and there were no differences in pace-sense thresholds. The total procedural duration was shorter (2.1 ± 0.5 hours) in group II (all pectoral implants) compared with 23 pectoral implants of group I (2.9 ± 0.5 hours) (P < 0.0001). Biphasic devices were used in all patients and active shell devices in 67 patients; no patient needed a subcutaneous patch. There were six complications (7%), four in group I and two in group II: one pulmonary edema and one respiratory insufficiency that delayed extubation for 3 hours in a patient with prior lung resection, both probably related to general anesthesia, one lead insulation break that required reoperation on day 3, two pocket hematomas, and one pneumothorax. There was one postoperative arrhythmic death at 48 hours in group I. No infections occurred. Patients were discharged at a mean time of 3 days. All devices functioned well at predischarge testing. Thus, it is feasible to use local anesthesia for current ICD implants to expedite the procedure and avoid general anesthesia related cost and possible complications.