BACKGROUND Implantable cardioverter defibrillators (ICDs) are intended to prevent sudden cardiac death yet also impose a risk of morbidity. This study describes the outcomes of ICDs in a pediatric and congenital heart disease (CHD) population from a single center. METHODS Retrospective cohort study of all patients with an ICD followed at the University of Michigan Congenital Heart Center from 2005-2013. The primary outcome was ICD system revision for any reason excluding routine generator change for battery depletion. RESULTS There were 191 ICD systems in 131 patients, including 57 with CHD, 24 with hypertrophic cardiomyopathy, and 45 with structurally normal hearts. Median age was 16 years at initial implant. Total follow-up was 850 patient-years; median 4.9 years/patient. There were 43 (33%) patients who required 60 ICD revisions; 70 revisions/1,000 patient-years of follow-up. Revisions included 25 lead extractions with replacement, 21 lead additions, five lead repositions, and four full system revisions. Kaplan-Meier (K-M) median time to appropriate shock was similar to the median time to system revision. K-M time to system revision was significantly affected by recalled lead performance. CONCLUSIONS The need for ICD system revision is high in this pediatric and CHD population and occurs at a rate similar to the rate of receiving appropriate therapy. These results highlight the need for judicious implant criteria and improved device longevity.