ICDs provide protection against sudden cardiac death in patients with life-threatening arrhythmias. Nevertheless, efficacy of defibrillation remains an important issue to guarantee the future safety of patients who receive an ICD. There is a significant number of patients who need an additional subcutaneous lead to obtain a defibrillation safety margin of at least 10 J between the maximum output of the ICD and the energy needed for ventricular defibrillation. However, few data exists about the long-term performance of different types of subcutaneous leads. Therefore, the aim of this study was to analyze the long-term experience with three different types of subcutaneous leads. The study included 132 patients (109 men, 23 women; mean age 59.8 years [SD ± 10.7 years]). All of them received a subcutaneous lead in addition to a single chamber or dual chamber ICD between October 1990 and April 2002. Two patients received a second subcutaneous lead after the first lead had been removed so that a total of 134 subcutaneous leads were evaluated. Inclusion criteria for the implantation of an additional subcutaneous lead were (1) unsuccessful ventricular defibrillation at implant without a subcutaneous lead, (2) insufficient safety margin (< 10 J) between the maximum output of the ICD and the energy needed for ventricular defibrillation, or (3) clinical evaluation of a new subcutaneous lead (Medtronic 13014). There were no significant differences between the three study groups with regard to age, sex, underlying cardiac disease, left ventricular ejection fraction, NYHA class assessment and clinical arrhythmia. The results of the DFT testing during follow-up (prehospital discharge test and 1 and 3 years) were compared to the baseline value obtained during the implantation procedure. All lead related complications were analyzed. Eighty-two single element subcutaneous array electrodes (SQ-A1), 31 subcutaneous three-finger electrodes (SQ-A3), and 21 subcutaneous patch electrodes (SQ-P) were implanted during the study period. The median follow-up was 1,499 days (25th percentile: 798 days, 75th percentile: 1,976 days) in the SQ-A1 group, 2,209 days (25th percentile: 1,242 days, 75th percentile: 2,710 days) in the SQ-A3 group, and 1,419 days (25th percentile: 787 days, 75th percentile: 2,838 days) in the SQ-P group. None of the three groups had a significant change of the DFT during follow-up compared to baseline. Major complications occurred in six (7.3%) patients in group SQ-A1 and in two (9.5%) patients in group SQ-P. There were no major complications in group SQ-A3. Kaplan-Meier curves analyzing freedom from subcutaneous lead related complications did not show a significant difference between the three study groups (P = 0.16). SQ-A1, SQ-A3, and SQ-P leads provide stable DFTs during long-term follow-up. Major complications are rare. However, a careful follow-up including chest radiographs at regular intervals is needed to detect potentially fatal complications like lead fractures.