We studied the outcome of 2327 patients after aortic valve replacement from October 1962 to December 1993. 1840 mechanical (mostly STJ bi-leaflet and BS tilting disc valves) and 487 biological (IS and CE valves) prostheses were implanted. The mean follow-up period of 1458 surviving patients was 7.1 years. There were 688 non-survivors and 181 lost cases. The mean age was 50.1 years, 73% of the patients were male and 27% were female. Early mortality was about 20% in the 1960's and about 4.5% in the last years. Mortality following valve replacement was influenced by preoperative NYHA classification, cardiac index, pressure gradient and simultaneous CHD. The long term results of all valves showed a survival rate of 80% after 5 years, 73% after 10 and 60% after 15 years following operative treatment. There was no significant difference in survival rates between all mechanical and biological valves. Non-lethal complications of all mechanical valves showed no significant difference but there was a clearly lower rate of paraprosthetic leakages, haemolysis and thromboembolism in biografts. The biological valves showed a high rate of degeneration (2.7%/pty). Reoperation was performed in 170 patients. Most valve changes consisted of biological to mechanical and mechanical to mechanical valves. The main reason of reoperation was degeneration (biological), paraprosthetic leakage and haemolysis (mechanical). The cause of death of the 688 non-survivors was valve related in 17.9%. 37% of these were due to thromboembolism and 38% due to bleeding. 55.6% of survivors (group 2) could be ascribed to NYHA class III and 17.2% to class IV prior to operation. Postoperative outcome demonstrated an improvement in NYHA classification in about 80%. Of surviving patients 80% pronounced an increase of physical activity after operation.