Objectives: We have compared mortality and risk factors for late deaths in patients with tetralogy of Fallot undergoing surgical repair in 1972 and 1982 in a Japanese multicentric study, examining in particular the impact of time of repair. Background: There is limited information on the effect that time of repair, and our constantly changing approach to it, has on late outcome in repaired tetralogy of Fallot. Methods: We analysed the Japanese registry of deaths occurring after surgical repair of tetralogy of Fallot. We studied two postoperative 1-year cohorts of survivors of surgery performed in 12 centers. Of the patients, 122, aged 29 ± 12 years, had undergone repair in 1972, their age at repair being 9.6 years. An additional 186 patients, aged 23 ± 8.7 years, had been repaired in 1982 at the age of 7.7 years. Results: Annual mortality, as judged per 100,000 population of patients with tetralogy of Fallot, declined from 0.387 in 1972 to 0.196 in 1982. Significant differences were deaths following surgery (27% vs. 13%, p < 0.001), patching of the subpulmonary outflow tract (48% vs. 89%, p < 0.001), and transjunctional patching (13% vs. 63%, p < 0.001). Late death was observed in 6 vs. 3 patients (9/308, 2.9%). The actuarial rate of survival calculated over 14 years was 97% vs. 98%. Reoperation was performed in 5 vs. 9 patients (14/308, 4.5%). Risk factors for late death were age at repair (p = 0.01), and history of reoperation (p < 0.001). Transjunctional patching (p = 0.01) proved to be associated with late mortality only in patients repaired in 1972. Conclusions: Late survival was excellent, with a low incidence of reoperations in both groups of patients. The era of repair has a big influence on total and operative mortality, but has only a small impact on late mortality. Ongoing analysis of follow-up will possibly reveal subsequent changes with time.