In vitro hydrodynamic comparison of mitral valve prostheses at high flow rates.

A pulse duplicator system for evaluating the hemodynamic performance of mitral prostheses is described. Under conditions stimulating normal resting physiology, all valves tested had measurable but acceptably small pressure drops. Under conditions simulating exercise, all were moderately to severely stenotic. Valves with nearly equal mounting diameters were compared. The Hancock, Beall, and Starr-Edwards valves (Group A) were found to be significantly more stenotic than the Björk-Shiley, Cutter-Cooley, Ionescu-Shiley, and Lillehei-Kaster valves (Group B). In the 29 to 30 mm. mounting diameter size at cardiac outputs of 5 and 9 L. per minute, Group A had average pressure drops of 3.2 and 10.5 mm. Hg and Group B, pressure drops of 1.6 and 5.3 mm. Hg, respectively. In the 24 to 26 mm. mounting diameter size, at cardiac outputs of 9 L. per minute, all the valves had critically large pressure drops (9 to 17.6 mm. Hg). The standard Gorlin formula is inappropriate for computing the orifice area of prosthetic valves. The discharge coefficient for a valve (a measure of how well the valve uses its primary flow area) and a performance index (a measure of how well the valve uses its mounting area) have been computed from a knowledge of the orifice size, without the necessity of assuming a value for the discharge coefficient required by the Gorlin formula. The biological valves (Hancock and Ionescu-Shiley) provide an efficient orifice for fluid flow at the free leaflet margins and have large discharge coefficients. On the basis of the fluid dynamic equation of motion, steady flow, root mean square (RMS) flow, and peak flow, combined with the appropriate transvalvular gradients, were all shown to yield equally accurate characterizations of valvular hydrodynamic performance. Mean flow, unfortunately the only value obtainable clinically, yielded effective orifice areas 10 percent smaller than either of the other three flow values.