OBJECTIVES: This study aims to evaluate our 45-year experience with the Fontan procedure and to identify risk factors for late mortality and morbidity. METHODS: Demographic, preoperative, perioperative and postoperative characteristics were retrospectively collected for all patients who underwent a Fontan procedure in a single centre between 1972 and 2016. RESULTS: The study included 277 Fontan procedures (44 atriopulmonary connections, 28 Fontan-Björk, 42 lateral tunnels and 163 extracardiac conduits). Early failure occurred in 17 patients (6.1%). Median follow-up of the study cohort was 11.9 years (Q1-Q3 7.6-17.5). Longest survival estimates were 31% [95% confidence intervals (CI) 18-44%] at 35 years for atriopulmonary connection/Björk, 87% (95% CI 63-96%) at 20 years for lateral tunnel and 99% (95% CI 96-100%) at 15 years for extracardiac conduit. Estimated freedom from Fontan failure (death, heart transplant, take-down, protein-losing enteropathy, New York Heart Association III-IV) at 15 years was 65% (95% CI 52-76%) for atriopulmonary connection/Björk, 90% (95% CI 73-97%) for lateral tunnel and 90% (95% CI 82-94%) for extracardiac conduit. The development of tachyarrhythmia was an important predictor of Fontan failure [hazard ratio (HR) 2.6, 95% CI 1.2-5.8; P = 0.017], thromboembolic/neurological events (HR 3.6, 95% CI 1.4-9.4; P = 0.008) and pacemaker for sinus node dysfunction (HR 3.7, 95% CI 1.4-9.6; P = 0.008). Prolonged pleural effusion (>21 days) increased the risk of experiencing protein-losing enteropathy (HR 4.7, 95% CI 2.0-11.1; P < 0.001). CONCLUSIONS: With modern techniques, survival and freedom from Fontan failure are good. However, Fontan patients remain subject to general attrition. Tachyarrhythmia is an important sign for an adverse outcome. Prevention and early treatment of tachyarrhythmia may, therefore, be paramount in improving the long-term outcome.