Background

Although Doppler echocardiography has been shown to be accurate in assessing stenotic orifice areas in native valves, its accuracy in evaluating the prosthetic valve orifice area remains undetermined.

Methods and results

Doppler-estimated valve areas were studied for their agreement with catheter-derived Gorlin effective orifice areas and their flow dependence in five sizes (19/20-27 mm) of St. Jude, Medtronic-Hall, and Hancock aortic valves using a pulsatile flow model. Doppler areas were calculated three ways: using the standard continuity equation; using its simplified modification (peak flow/peak velocity); and using the Gorlin equation with Doppler pressure gradients. The results were compared with Gorlin effective orifice areas derived from direct flow and catheter pressure measurements. Excellent correlation between Gorlin effective orifice areas and the three Doppler approaches was found in all three valve types (r = 0.93-0.99, SEE = 0.07-0.11 cm2). In Medtronic-Hall and Hancock valves, there was only slight underestimation by Doppler (mean difference, 0.003-0.25 cm2). In St. Jude valves, however, all three Doppler methods significantly underestimated effective orifice areas derived from direct flow and pressure measurements (mean difference, 0.40-0.57 cm2) with differences as great as 1.6 cm2. In general, the modified continuity equation calculated the largest Doppler areas. When orifice areas were calculated from the valve geometry using the area determined from the inner valve diameter reduced by the projected area of the opened leaflets, Gorlin effective orifice areas were much closer to the geometric orifice areas than Doppler areas (mean difference, 0.40 +/- 0.31 versus 1.04 +/- 0.20 cm2). In St. Jude and Medtronic-Hall valves, areas calculated by either technique did not show a consistent or clinically significant flow dependence. In Hancock valves, however, areas calculated by both the continuity equation and the Gorlin equation decreased significantly (p less than 0.001) with low flow rates.

Conclusions

Doppler echocardiography using either the continuity equation or Gorlin formula allows in vitro calculation of Medtronic-Hall and Hancock effective valve orifice areas but underestimates valve areas in St. Jude valves. This phenomenon is due to localized high velocities in St. Jude valves, which do not reflect the mean velocity distribution across the orifice. Valve areas are flow independent in St. Jude and Medtronic-Hall prostheses but decrease significantly with low flow in Hancock valves, suggesting that bioprosthetic leaflets may not open fully at low flow rates.