Hemodynamic parameters and vessel tortuosity: an investigation with a mesenterial vascular network

Purpose: The effect of hemodynamic parameters on vessel tortuosity remains un- clear. Here we investigate the correlation of tortuosity with a set of hemodynamic parameters in a mesenterial vascular network. Methods: A mesenterial vascular network of 389 vessels (131 arteries, 132 veins, and 126 capillaries) was imaged. Eleven hemodynamic parameters were measured (pressure, wall shear stress, diameter, blood velocity and flow, viscosity, haematocrit, partial oxygen saturation, oxygen saturation, wall thickness, and local vessel density). Tortuosity was assessed quantitatively with a validated algorithm and correlation computed with subsets of hemodynamic parameters selected by a lasso regressor. Results: Results suggest that tortuosity is related to pressure, wall shear stress, diameter, blood velocity, viscosity, partial but not full oxygen saturation, and wall thickness for the arteries; diameter, blood flow, hematocrit, and density for the veins; and viscosity (but not hematocrit), partial and full oxygen saturation, and density for the capillaries. The combination of hemodynamic parameters correlating best with tortuosity is the set of all parameters except density (r = 0.64, p < 0.01), using as tortuosity definition the set of tortuosity features (geometric measures) correlating best with a single hemodynamic factor for the arteries. Conclusion:  This pilot suggests two general conclusions. First, the quantitative definition of tortuosity (i.e., the set of geometric features adopted) should be tuned to the specific data and problem considered. Second, tortuosity is caused by a combination of hemodynamic factors, not a single one.