Carotid stenosis affects the structure and microvascular of the retina. However, whether microvascular changes of the retina reflect brain hemodynamic changes in patients with carotid stenosis remains unknown. Our study shows a novel association of microvascular changes in SVC and SVP with brain perfusion among patients with unilateral carotid stenosis of more than 50%. We suggest that the retinal microvasculature may reflect early hemodynamic changes caused by carotid stenosis. Therefore, OCTA may be used as a potential noninvasive quantitative screening tool for the assessment of early cerebral hemodynamic compromise, which may have a prospect for identifying subgroups with a high risk of stroke in patients with moderate and severe ICA stenosis.
ICA stenosis causes retinal hemodynamics19. However, whether microvascular changes in the retina could be indicator of early changes in brain hemodynamics among patients with ICA stenosis remains unclear. To the best knowledge, our current report is the first to show a significant correlation between perfusion changes in the retina and brain in ICA stenosis patients. Given the homology between the retinal and cerebral microvasculature, concomitant cerebral hemodynamic changes in ICA stenosis might extend to the retina, causing changes in the microvascular network. Similarly, reports using different OCTA machines showed patients with ICA stenosis have significantly decreased superficial vasculature densities compared with controls20,21. It suggested the superficial vessels of the retina are sensitive to ischemic injury21, which are the main blood flow channel and responsible for the arterial circulation of the retina22,23,24. That may help to explain why we only found microvascular changes in superficial vascular layer(SVC and SVP) significantly related to brain hemodynamic changes.
Previous reports using different imaging modalities have shown that the superficial vessels of the retina are significantly altered in patients with cerebrovascular diseases25–27 and may be a potential risk indicator of ischemic stroke28. Since ICA stenosis causes cerebral hypoperfusion29, accounting for 30%-50% ischemic strokes in Asia30, it is plausible that changes in the superficial vessels may give clues to the occurrence of ischemic stroke.
In our study, patients with severe ICA stenosis showed a higher MTT and lower CBF in the ipsilateral side compared to moderate ICA stenosis(Fig. 3). It is consistent with others reports that the severity of ICA stenosis is related to cerebral hypoperfusion, which may have a higher risk of ischemic stroke or recurrence31,32. However, our data did not show any significant differences between severity of ICA stenosis and retinal microvascular changes(Fig. 4). It may suggest brain hemodynamics occurs earlier than retinal changes in patients with moderate and severe ICA stenosis. Reports on retinal microvasculature changes in patients with carotid artery stenosis compared with controls have been inconsistent33–35. Little research has investigated the retinal perfusion changes in different severity of ICA stenosis, especially between moderate and severe ICA stenosis. Longitudinal studies with larger sample sizes are needed to investigate this.
In addition, we did not find any significant differences in retinal microvascular between ipsilateral and contralateral eyes. During stenosis, ipsilateral circulation may rely on contralateral circulation for support (a compensatory mechanism), which may explain the insignificant difference between the contralateral and ipsilateral retinal microvasculature. Given the small sample size of the subjects in the study, future studies with larger sample sizes are needed.
We are aware that our study labors under several limitations. Since our study was a cross-sectional study, it is also not certain when the retinal perfusion change occurs, whether the retinal perfusion changes in patients with ICA stenosis could be an indicator for predicting a broad spectrum of neurological disorders such as ischemic stroke, transient ischemic attack, and vascular dementia. Future studies with follow-up data were needed to explore this. In addition, bias might be caused by the small number of patients in the moderate stenosis group. The future study will expand the enrollment group, especially those in the mild and moderate stenosis group.