A computational model for retinal haemodynamics under gravitational and postural variations
Published online on March 31, 2026
Abstract
["The Journal of Physiology, Volume 604, Issue 7, Page 2698-2716, 1 April 2026. ", "\nAbstract figure legend The haemodynamic impact of gravitational and postural changes on the retina, and its relevance to spaceflight associated neuro‐ocular syndrome (SANS), remains incompletely characterized. This study assesses retinal haemodynamics from upright to head‐down tilt using a lumped‐parameter model that dynamically couples intraocular pressure (IOP) and the retinal circulation (central retinal artery, microvasculature, central retinal vein). The model is validated against experimental measurements from the literature. The figure summarizes the main conclusions: (i) head‐down tilt increases central retinal artery (CRA) peak systolic velocity and elevates CRA pressure, consistent with experimental trends; (ii) arterial compartments display the largest posture‐dependent changes; and (iii) small arterial diameter reductions (−4%) substantially modulate these responses, identifying haemodynamic loci and mechanisms relevant to SANS.\n\n\n\n\n\n\n\n\n\nAbstract\nSpaceflight‐associated neuro‐ocular syndrome presents a significant risk during long‐duration spaceflight, yet the retinal mechanisms underlying these visual changes remain poorly characterized. We developed a physiologically detailed lumped‐parameter model coupling a five‐compartment retinal vascular network with a dynamic intraocular pressure module responsive to hydrostatic shifts. Head‐down tilt experiments were reproduced to emulate cephalad fluid shifts in microgravity. Model outputs are closely aligned with experimental benchmarks. Intraocular pressure increased from 16.3 mmHg (upright) to 29.5 mmHg at −30° (81% increase), while ocular perfusion pressure remained within 2 mmHg, reflecting haemodynamic responses. Retinal blood‐flow predictions fell within the interquartile range of in vivo Doppler measurements, and peak systolic velocity trends in the central retinal artery agreed within 1% of ultrasound data after incorporating a microgravity‐induced 4% central retinal artery constriction. These findings validate the model's ability to capture the coupled influence of systemic arterial pressure, venous drainage and intraocular pressure on retinal perfusion across gravitational states. Simulations revealed that posture‐induced elevations in arterial pressure and intraocular pressure propagate non‐linearly through the retinal tree, imposing the greatest mechanical stress on the central retinal artery and arteriolar segments, critical sites for flow dysregulation and potential optic nerve damage. By integrating retinal and ocular biomechanics, the model developed in this work offers mechanistic insight into the pathogenesis of neuro‐ocular syndrome, establishes quantitative thresholds to inform countermeasure development, and provides a transferable tool for terrestrial disorders involving intraocular pressure and vascular dysregulation, such as glaucoma and hypertensive retinopathy.\n\n\n\n\n\n\n\n\n\nKey points\n\nMicrogravity and postural changes alter ocular circulation, contributing to spaceflight‐associated neuro‐ocular syndrome (SANS), yet their impact on retinal haemodynamicsremains poorly characterized.\nWe present a lumped‐parameter model that dynamically couples intraocular pressure (IOP) with retinal blood flow through a five‐compartment vascular network.\nThe model simulates posture‐induced hydrostatic changes and accurately reproduces trends in IOP, ocular perfusion pressure and retinal flow observed in head‐down tilt experiments.\nSimulations reveal that arterial compartments (especially the central retinal artery) are the most sensitive to gravitational stress and vascular constriction, identifying key sites of haemodynamic vulnerability.\nThis framework provides novel mechanistic insights into the retinal contributions to SANS and offers a validated platform for evaluating countermeasures and studying ocular pathologies.\n\n\n"]