Improved Statistical Characterization of Prosthetic Heart Valve Hydrodynamics Using a Performance Index and Regression Analysis
Sandy F. C. Stewart PhD, Harry F. Bushar PhD

In-vitro hydrodynamic testing is a fundamental step in the development of new heart valve prostheses. Typically, three samples of each size valve are tested, and only descriptive statistics are reported for each valve size and cardiac output (CO). In this study, a regression analysis performed on an entire valve dataset was shown to provide a better statistical assessment of valve behavior. The analysis uses a dimensionless performance index (PI) (equal to the effective orifice area divided by the tissue annulus area) as the dependent variable, with valve size and CO as independent variables. The standard errors so calculated were generally lower, and the confidence intervals tighter, than those calculated using descriptive statistics, thus allowing a better comparison to a performance criterion. This method should be considered for the next version of the ISO 5840 Standard for heart valve prostheses, which is now undergoing revision.