Prescribed Flow in 3D Printed Models for Determination of Coronary Ischemia: A Comparison to a Computational Fluid Dynamic-based Assessment

Introduction: Advances in coronary computed tomography angiography (CCTA) imaging and 3D printing have enabled fabrication of physical models of coronary arterial geometry. Computational fluid dynamics (CFD) applied to CCTA allows determination of fractional flow reserve—or the hyperemic pressure ratio (P d /P a ) across a stenosis from a typically acquired CT (FFR CT ), and has been demonstrated to be a non-invasive “gold standard” for determining ischemia. We evaluated the influence of prescribed flow rates on P d /P a by 3D printed models compared to CFD evaluation. Methods: Prescribed flows were evaluated using an in vitro experimental flow loop with a pump circulating blood analog fluid at steady-state flow (Fig 1A). A 50% diameter stenosis (DS) model with a smooth Gaussian shape was fabricated using VeroClear rigid material in an Objet260 Connex printer. Mean proximal pressure (P a ) was maintained at 90 mm Hg using a linear flow needle valve. P d /P a at differing prescribed flow rates (Q) was titrated by a flow meter. A prescribed flow ratio of P d /P a ≤0.80 was considered a diagnostic of coronary ischemia. CFD simulation studies with a 50% DS model were performed using the Multi-physics finite element solver (MUPFES). The experimental data was used as input boundary conditions for the CFD analysis (Fig 1B). Results: The P d /P a -Q curve (Fig 1C) followed a quadratic trend with decreasing slope. The corresponding flow rate for determining ischemia at 0.8 threshold was at 6.12 and 6.43 cc/sec for in vitro and CFD, respectively. The pressure drop (ΔP) between the experimental and CFD simulations showed a significant linear correlation (r=0.99). The mean differences (bias) in ΔP, as shown by Bland-Altman plot (Fig 1D), was at 1.05±0.79 mm Hg (95% limit of agreement = 2.63 to -0.50 mm Hg). Conclusions: As compared to a CFD-based standard, prescribed flow using 3D printed coronary artery models enables identification of specific flow conditions that are associated with coronary ischemia.