OBJECTIVE The Ross procedure (pulmonary autograft) has since the 1980s attracted growing interest as an alternative to the widely practised insertion of a prosthetic aortic valve. The 12-year experience of a consecutive series from one centre are reported here. METHODS Between February 1990 and January 2002 a Ross procedure, predominantly with the subcoronary technique, was performed in 244 consecutive patients with aortic valve disease (244 men, 54 women, mean age 46 +/- 13.5 years). Annual follow-up clinical examinations (mean postoperative period 32.9 +/- 29.5 months in 99 % of the cohort) were performed. RESULTS Perioperative mortality was 0.8 % (n=2), and there were two late deaths unrelated to the aortic valve disease. Seven patients had to be re-operated for failure of the homograft (n=4) or autograft (n=4). According to clinical criteria, 99 % of the followed-up patients were in New York Heart Association (NYHA) functional class I or II, only two patients, with pulmonary comorbidity, were in class III. Echocardiography demonstrated autografts with nearly normal transvalvular gradient (mean maximal pressure gradient 6.5 +/- 3.3 mmHg), while nine patients had second-degree aortic regurgitation. The mean maximal gradient across the homograft valve in the pulmonary position was 12.0 +/- 6.9 mmHg, while ten patients had second-degree and one had third-degree pulmonary regurgitation. CONCLUSIONS The technically demanding Ross procedure produced excellent clinical and hemodynamic mid-term results. It is thus an appealing alternative to the widely used replacement by a prosthetic valve. Definitive assessment awaits further long-term follow-up.