Derivation and sensitivity analysis of a novel one-dimensional model of coronary blood flow accounting for vessel taper and boundary slip

Computational models of coronary flow allow quantification of virtual fractional flow reserve (vFFR) and absolute flow, reducing the need for invasive testing in supporting a diagnosis of ischemic heart disease. One-dimensional (1-D) models of coronary flow are an emerging tool, which provide rapid assessment of coronary physiology. However, they currently do not fully account for side branch flow, the importance of which for clinically indeterminate cases is disputed. In the current study, we derive a novel 1-D model of coronary flow from the Navier-Stokes equations, which accounts for vessel taper, convective acceleration, and boundary slip. Using 15 idealized coronary arterial anatomies derived from patient data, we compared vFFR results from our model with those from a preexisting 1-D model of coronary flow. The techniques showed strong correlation (<i>r</i> = 0.99, <i>P</i> [Formula: see text]) with close agreement. Through a novel output-constrained global sensitivity analysis (GSA), we demonstrated that anatomical parameters, particularly the magnitude of simulated side branches, significantly influence vFFR in the diagnostic gray zone (0.75-0.85) and absolute coronary flow regardless of stenosis severity. This challenges the commonly accepted dominance of stenosis parameters. We present a more mathematically and anatomically consistent 1-D system for modeling coronary flow, which demonstrated modest differences with established models. GSA highlighted the influence of the flow diameter (Murray) scaling exponent in all cases when computing flow and in intermediate epicardial stenoses when quantifying vFFR. This work highlights the importance of accurately quantifying side branch flow when computing virtual coronary physiology.<b>NEW & NOTEWORTHY</b> A new 1-D model for blood flow is derived accounting for vessel taper, side branch flow, and boundary slip. This model is applied to a novel output-constrained GSA, highlighting the influence of anatomical and flow input parameters within clinically relevant regions of virtual fractional flow reserve. Notably, we reveal the significant influence of the flow diameter scaling exponent in the virtual fractional flow reserve gray zone (0.75-0.85) compared with the conventionally accepted stenosis severity.