Association between Open-Angle Glaucoma Progression and Cardiovascular Risk Predicted by a Deep Learning-Based Fundus...

Area of Science:

• Ophthalmology
• Cardiology
• Artificial Intelligence

Background:

• Glaucoma is a leading cause of irreversible blindness.
• Systemic vascular health is increasingly recognized as a factor in glaucoma progression.
• Noninvasive methods to assess systemic vascular burden are needed for risk stratification.

Purpose of the Study:

• To investigate the association between the RetiCAC score, a deep learning-derived measure of systemic atherosclerotic burden from retinal photographs, and the progression of glaucomatous visual fields (VF).

Main Methods:

• Retrospective cohort study of 192 eyes with open-angle glaucoma.
• RetiCAC scores estimated using deep learning from fundus photographs.
• Generalized estimating equations (GEE) and Kaplan-Meier analysis used to assess factors associated with VF progression.

Main Results:

• Higher RetiCAC scores, elevated peak intraocular pressure (IOP), and longer follow-up duration were independently associated with VF progression.
• Patients in the high-risk RetiCAC group had a significantly higher probability of VF progression.
• An inflection point for accelerated VF deterioration was observed at a RetiCAC score of 51.1.

Conclusions:

• The RetiCAC score is a significant, independent predictor of glaucoma progression.
• This noninvasive marker of systemic vascular health may aid in predicting glaucoma progression.
• RetiCAC may reflect vascular dysfunction contributing to the pathogenesis of glaucoma.