advertisement
The paradigm of calculating risk for the development and progression of glaucomatous optic neuropathy has undergone a significant evolution over the past several decades. While reduction of intraocular pressure (IOP) remains the only current therapeutic approach, the disease has been shown to be multifactorial with highly varied progression patterns and risk clusters. Ocular hemodynamic contributions to the disease process have been identified as a significant consideration in many individuals, with higher vascular involvement in certain populations including persons of African descent and patients with normal-tension glaucoma. Determining total risk is additionally complicated by diurnal variations in physiological biomarkers, including fluctuations in IOP, blood pressure (BP), ocular perfusion pressures (estimated by BP and IOP) and blood flow. In consideration of this, Baek and coauthors utilized swept-source optical coherence tomography angiography (OCTA) to assess retinal vessel density (RVD) variation between 8:00 AM and 8:00 PM in primary open-angle glaucoma and healthy subjects. The authors found diurnal changes of IOP, mean ocular-perfusion pressure and RVD were significantly greater in glaucoma patients compared to healthy controls. The authors identified the largest difference between groups at 8:00 PM where macular RVD increased to the highest level in the healthy group while glaucoma patients conversely demonstrated their lowest levels. These findings highlight the difficulty in estimating risk from biomarkers assessed only during normal clinical office hours. The study utilizes OCTA which allows for visualization of RVD at levels of specificity not previously possible and the authors demonstrated high levels of intra-visit repeatability (0.755-0.943) and inter-visit reproducibility (0.843- 0.986). One significant limitation is the inclusion of patients taking ocular hypotensive medication which both influence IOP variation and possibly ocular hemodynamics. Additionally, other vascular beds not assessed with OCTA in this study may be important to consider, such as those seen within the retrobulbar blood vessels. Nocturnal variation was also not assessed, and the authors correctly suggest targeting nocturnal variation as a needed follow up for future research. Moving forward, the use of mathematical modeling and artificial intelligence (AI) in understanding the complex interactions of risk biomarkers and their variation both within and across individuals may improve risk calculation and lead to individually tailored approaches for precision medicine.