Computational Fluid Dynamics Simulations Provide Insights into the Paradox of Newer Prosthetic Aortic Valves

Patient-prosthesis mismatch (PPM) is defined by the estimated orifice area divided by body surface area (BSA), or iEOA. Small iEOA predicts poor outcomes with surgical bioprosthetic valves (SBV), but balloon-expandable valves (BEV) and self-expanding valves (SEV) have more variable results.

Assess effect of geometry of the valve on overestimation of gradients and methods for adjusting for body size other than BSA.

Computational fluid dynamic simulations of SBV, BEV and SEV, with 25 different variations of anatomic and valve sizes for each valve type determined peak stenosis velocity and peak and total valve gradients. The relationship between stroke volume and body size measures was analyzed using data from the Atherosclerosis Risk In Communities study.

Doppler peak velocity correctly estimated the total gradient for SBV (slope = 1.01) but overestimated the gradient for BEV (slope = 1.19) and SEV (slope = 1.42). The gradient across the valves varied even with identical dimensions, with SBV being highest, BEV lower, and SEV lowest. BSA had a fair correlation with stroke volume (r = 0.6030), but lean body mass (r = 0.6494) and fat-free mass (r = 0.6631) were better and left ventricular outflow tract (LVOT) area was best (r = 0.7570).

Geometry of the valve has a major impact on hemodynamics with very different results with identical dimensions. LVOT area was better way to adjust for body size than BSA. Two measures based on LVOT velocity and area are proposed as an alternative to implement these results. If these findings are confirmed with in-vitro and in-vivo studies, the approach could allow more accurate prediction and evaluation of PPM.