The extent to which Doppler echocardiography information can be used in the assessment of prosthesis hemodynamic performance is still controversial. The goals of our study were to assess the importance of valve design and size both on Doppler echocardiography findings and on pressure recovery in a fluid mechanics model. We performed Doppler and catheter measurements in the different orifices of the bileaflet St Jude (central and side orifices), the monoleaflet Omnicarbon (major and minor orifices), and the stented Biocor porcine prosthesis. Net pressure gradients were predicted from Doppler flow velocities, assuming either independence or dependence of valve size. The peak Doppler estimated gradients (mean +/- SD for sizes 21 to 27) were 21 +/- 10.3 mm Hg for St Jude, 18 +/- 9.3 mm Hg for Omnicarbon, and 37 +/- 14.5 mm Hg for Biocor (P <.05 for St Jude and Omnicarbon vs Biocor). The pressure recovery (proportion of peak catheter pressure) was 53% +/- 8.6% for central-St Jude, 29% +/- 8. 9% for side-St Jude, 20% +/- 5.6% for major-Omnicarbon, 23% +/- 7.4% for minor-Omnicarbon, and 18% +/- 3.6% for Biocor (P <.05 for central-St Jude and side-St Jude vs Omnicarbon and Biocor). Valve sizes (x) significantly influenced pressure recovery (y in percentage) (central-St Jude: y = 3.7x - 35.9, r = 0.88, P =.0001; major-Omnicarbon: y = 2.1x - 30.3, r = 0.85, P =.0001). By assuming dependence of valve size, Doppler was able to predict net pressure gradients in St Jude with a mean difference between net catheter and Doppler-predicted gradient of -3.8 +/- 2.5 mm Hg. In conclusion, prosthetic valve design and size influence the degree of pressure recovery, making Doppler gradients potentially misleading in both the assessment of hemodynamic performance and the comparison of one design with another. The preliminary results indicate that net gradient can be predicted from Doppler gradients.