BACKGROUND: The aim of the study was to compare early differences in reversal of LV dilatation between patients with mechanical prosthesis = group A (n = 51: Carbomedics = 40, Tekna/Edwards = 11) and biological procedures = group B (n = 75: pulmonary autograft = 36, aortic valve repair = 29, homograft = 10). METHODS: Since 1,990,126 consecutive patients younger than 50 years who had surgical correction of isolated aortic incompetence underwent echocardiographic examinations preoperatively, at discharge, and at one-year follow-up. Left-ventricular (LV) diameters were measured (LVEDD, LVESD) and matched to body surface area (LVESDI, LVEDDI and fractional shortening (FS) was calculated. Aortic peak flow velocities were assessed by Doppler technique and gradients were calculated. RESULTS: There were no significant differences preoperatively in aortic incompetence, NYHA classification, LVEDDI, LVESDI, and FS. In group B there was a significant decrease of LVESDI (p < 0.002) and LVEDDI (p < 0.001) but no change in FS at discharge. In group A a significant reduction of FS (p < 0.05) without any significant changes in LV size was observed. No patient died perioperatively or during the first year. At one-year follow-up (complete in 97.6% patients) there were no significant differences in LV diameters but group B had better ventricular function (p < 0.05) resulting in better NYHA classification (p < 0.05). Only group B had normal aortic valve gradients at discharge and at follow-up (A: 25.2 +/- 4.3 vs B: 10.2 +/- 2.4 mmHg). CONCLUSIONS: Normal aortic valve gradients in patients after aortic valve repair or allograft replacement for chronic aortic incompetence lead to early recovery from ventricular dilatation and significantly better ventricular function at discharge. One year postoperatively they had improved ventricular function and NYHA class in comparison with patients in whom a mechanical prosthesis was implanted.