Abstract
A theoretical model of the human retina is simulated using two distinct vascular network geometries to predict the impact of heterogeneity in vascular network structure on retinal tissue oxygenation. Each vascular network is modeled as a combined heterogeneous representation of retinal arterioles and compartmental representation of capillaries, small venules, and large venules. A Green's function method is used to model oxygen transport in the arterioles, and a Krogh cylinder model is used in the capillaries and venules. Identical input arterial blood saturation (0.92), arteriolar pressure drop (16 mmHg), and arteriolar diameters by vessel order ( 117, 73, 44, 32, and 22 μm ) are assumed for both networks. The model shows that 12% of the arteriolar tissue in Branch 1 has a PO(2) less than 25 mmHg, while only 1% of the arteriolar tissue in Branch 2 has a PO(2) less than 25 mmHg. However, downstream of the capillaries, Branch 2 was predicted to exhibit lower tissue PO(2) than Branch 1. The model also predicted increased oxygen extraction fraction as oxygen demand increased or capillary density decreased. Even with identical initial conditions for saturation, pressure drop, and diameter, variations in network geometry led to significantly different regions of low PO(2), indicating a wide range of potential oxygenation outcomes for individual patients. This study therefore demonstrates that regional heterogeneity in vessel branching architecture may significantly impact oxygen saturation and ultimately retinal ganglion cell functionality, motivating the need for creating patient-specific vascular networks.