Aortic valve disease is known to be the most frequent valvular disease in the elderly and aortic valve replacement is often the best therapeutic strategy. Hemodynamic performance of prostheses is critical in this subset of patients to ensure an optimal quality of life. Moreover, old patients with small aortic ostia are getting more and more common in clinical practice, making often necessary to implant small prostheses. If a significant pressure drop is not achieved, hypertrophy persists and left ventricular function may not improve. Such conditions have not yet been extensively studied in the elderly. The aim of this study was firstly to assess echocardiographically the performance of aortic prosthetic heart valves in old patients (> or = 70 years) and compare the results obtained in patients with prostheses of different type and size, and secondly to evaluate the postoperative changes in left ventricular hypertrophy and function in a subset of patients with isolated or prevalent aortic stenosis. One hundred fifty-one patients were initially considered; global mortality was 9.3% at 20 +/- 12 months from intervention. In the 75 patients with a postoperative echocardiogram, transprosthetic gradient was 27 +/- 12 (max) and 15.1 +/- 6.6 (mean) mmHg. Mean functional prosthetic area (FPA) was 1.5 +/- 0.5 cm2. No statistically significant differences could be demonstrated between mechanical and biological prostheses. Three groups were identified, according to prosthetic size (Group 1: diameter < 23 mm, Group 2: diameter 23 mm, Group 3: diameter > 23 mm). Among groups, max and mean gradients as well as FPA were found to be significantly different. Respectively max gradient was 33.2 +/- 13, 26 +/- 11, 20.2 +/- 7.2 mmHg (p < 0.05), mean gradient was 17.2 +/- 6.1, 15.4 +/- 7.6, 11.7 +/- 4.3 mmHg (p < 0.01) and FPA was 1.2 +/- 0.3, 1.5 +/- 0.3, 1.8 +/- 0.7 cm2 (p < 0.05 between Group 1 and Group 3). In a subgroup of 31 patients with isolated or prevalent aortic stenosis, a significant interventricular septal thickness reduction was found postoperatively (14.3 +/- 2.3 vs 12.6 +/- 8.0 mm, p < 0.001). Posterior wall thickness decreased similarly, but to a lesser extent; left ventricular diameters and myocardial mass also significantly decreased (left ventricular mass: 186 +/- 45 vs 146 +/- 38 g/m2, p < 0.001). When prosthetic size was considered, septal thickness reduction was more evident in Group 1 and Group 2 (p < 0.05 and p < 0.01). On the contrary, a significant improvement in left ventricular diameters was observed only in Group 3 (p < 0.05). Left ventricular mass decreased significantly in Group 2 and Group 3 (p < 0.01 and p < 0.05). Such improvements could be demonstrated only in those patients (79%) who showed at least a 50% reduction in the transvalvular gradient. In this subset, left ventricular function also significantly improved (fractional shortening: 29 +/- 0.7 vs 33 +/- 0.7%, p < 0.02). In conclusion, aortic valve replacement in the elderly is a safe and effective therapeutic strategy. In patients with small aortic prostheses, the transvalvular gradient was found to be slightly but significantly higher as compared to that of larger prostheses. However, left ventricular function was good and similar in all subgroups. No significant differences were found between mechanical and biological prostheses. In old patients with isolated or prevalent aortic stenosis a significant reduction in left ventricular hypertrophy and mass is observed within 2 years from intervention. An increase in myocardial contractility can also be expected, if at least a 50% reduction in transvalvular gradient is obtained.