Abstract Background Tetralogy of Fallot (TOF) is the most common cyanotic congenital heart disease, and sudden cardiac death represents an important mode of death in these patients. Data evaluating the implantable cardioverter defibrillator (ICD) in this patient population remain scarce. Purpose We aimed to describe long-term follow-up of patients with TOF and ICD through a large nationwide registry. Methods Nationwide Registry including all TOF patients with an ICD initiated in 2010. The primary outcome was the first appropriate ICD therapy. Secondary outcomes included ICD-related complications, heart transplantation, and death. Clinical events were centrally adjudicated by a blinded committee. Cox proportional hazard models were used to identify predictors of appropriate ICD therapies and ICD-related complications. Results A total of 165 patients (mean age 42.2±13.3 years, 70.1% males) were included from 40 centers, including 104 (63.0%) in secondary prevention. During a median (IQR) follow-up of 6.8 (2.5–11.4) years, 78 (47.3%) patients received at least one appropriate ICD therapy, giving an annual incidence of 10.5% (7.1% and 12.5% in primary and secondary prevention, respectively, p=0.03). Overall, 71 (43.0%) patients presented with at least one complication, including inappropriate ICD shocks in 42 (25.5%) patients and lead/generator dysfunction in 36 (21.8%) patients. Among 61 (37.0%) primary prevention patients, the annual rate of appropriate ICD therapies was 4.1%, 5.3%, 9.5%, and 13.3% in patients with respectively no, one, two, or ≥ three guideline-recommended risk factors. In our cohort, QRS fragmentation was the only independent predictor of appropriate ICD therapies (HR 4.34, 95% CI 1.42–13.23), and its integration in a model with current criteria increased the area under the curve from 0.61 to 0.72 (p=0.006). No patient with left ventricular ejection fraction (LVEF) ≤35% without at least one other risk factor had appropriate ICD therapy. Patients with congestive heart failure and/or reduced LVEF had a higher risk of non-sudden death or heart transplantation (HR=11.01, 95% CI: 2.96–40.95). Conclusions Our findings demonstrate high rates of appropriate therapies in TOF patients with an ICD, including in primary prevention. The considerable long-term burden of ICD-related complications, however, underlines the need for careful candidate selection. A combination of easy-to-use criteria might improve risk stratification beyond low LVEF. Freedom from appropriate ICD therapy Funding Acknowledgement Type of funding source: None