This single-centre study was undertaken to review our experience with implantable cardioverter-defibrillator (ICD) implantation in children with relatively different aetiologies. We retrospectively reviewed the records of the paediatric patients who underwent ICD implantation between October 2001 and December 2008. The data of these patients were collected by reviewing the patients' medical records and computerized departmental pacemaker databases. A total of 28 patients who underwent ICD implantation during this period were included in this study. The median age was 12 years and median weight was 32 kg. Most of the patients had ion-channel diseases (n = 13) or cardiomyopathy (n = 11). Devices were implanted for either secondary (n = 22) or primary (n = 6) prevention. The selected ICD generator type was single chamber in 22 patients, dual chamber in 5 patients, and biventricular in 1. Nineteen patients received 122 shocks. Fifteen of 22 patients (68.2%) from the secondary prevention group and 2 of 6 patients (33.3%) from the primary prevention group experienced at least one appropriate shock during a median period of 11.3 months (range: 4 days-6.5 years). Forty-two inappropriate shocks were delivered in seven (31.8%) patients from the secondary prophylaxis group during a median period of 11.3 months. The most important reason for inappropriate shocks was T-wave oversensing. In six patients, lead-related acute or chronic complications occurred. The ICD was safe and effective in interrupting malignant arrhythmias in children and adolescents with a high risk of sudden cardiac death. However, the occurrence of lead complications is significant. The incidence of therapies delivered by the device, with appropriate and inappropriate shocks, was high and interfered with the quality of life. The most important reason for inappropriate shocks was T-wave oversensing. Careful programming is mandatory to reduce the inappropriate shocks.