**3.4 Optical coherence tomography angiography features of the choroidal microvasculature in glaucoma**

According to vascular theory, changes in the choroidal microvasculature may play a key role in the pathophysiology of glaucoma as it supplies nutrition to both the optic nerve and the outer layers of retina. In a study evaluating parapapillary choroidal microvasculature dropout (MVD), Kim et al. found that MVD was associated with progressive RNFL thinning and worsening VF parameters [36]. The authors proposed that progressive retinal ganglion cell damage decreases metabolic demand and leads to reduced vascular perfusion and more extensive MVD.

Recently, Chatziralli et al. found that the peripapillary and MVD were lower in eyes with pseudoexfoliative glaucoma (PXG) [37]. This indicates optic nerve hypoperfusion may play a greater role in patients with PXG.

## **3.5 Using optical coherence tomography angiography to monitor disease progression in glaucoma**

OCTA may have a role in the early detection and monitoring of the progression of glaucoma progression. AlSalem et al. found that whole image ONH VD, whole image macular VD, average RNFL thickness, average mGCC thickness, and cataract status were predictive of worsening VF MD [9]. They also determined that structural properties and VD were equally effective at determining the glaucomatous stage. This was supported by the findings of Geyman et al. and Chen et al. who also determined that structural properties and VD values were equally effective at determining the glaucoma stage [38, 39]. Tracking the previously described OCTA parameters over time may enable physicians to objectively gauge glaucoma progression.

## **3.6 Use of optical coherence tomography angiography in healthy, glaucoma suspect, and eyes with POAG**

Using a standard OCTA protocol, typical qualitative and quantitative findings in healthy, glaucoma suspects, and eyes with POAG are shown in **Figure 5**. A, E, and I are images of the ONH and corresponding CDR. The B, F, and J show a map of the superficial retinal vessels and VD in different zones around the macula. C, G, and K show the nerve fiber layer thickness in the corresponding zones. D, H, and L are area maps of the radial peripapillary capillary VD.

The healthy eye (5A-D) has normal retinal morphology and OCTA parameter values. The CDR is 0.35, the retinal VD is 46%, the RNFL thickness is 113 μm, and the peripapillary capillary density is 53%. In a glaucoma suspect eye (5E-H), the enlarged CDR (0.80) alone may mislead the clinician. However, OCTA measurements of the VD (43%), RNFL (115 μm), and peripapillary capillary density (54%) are in the normal range. The OCTA scans of an eye with POAG (5I-L) show several abnormal findings characteristic of glaucoma. There is a severe degree of cupping (CDR 0.85), the overall retinal VD is reduced to 34%, the RNFL thickness is 55 μm, and the radial peripapillary VD is 34%.

#### **3.7 Using optical coherence tomography angiography to characterize myopia**

In clinical practice, detecting glaucoma in myopic patients has always been extremely challenging. As myopia increases, findings such as ONH tilt, increased ovality of the ONH, and peripapillary atrophy become more prominent [40]. The peripapillary RNFL thickness peaks also shift in myopic eyes, leading to difficulties in detecting thinning (40. Examples of changes seen in a patient with moderate myopia are shown in **Figure 6**.

In patients with high myopia, OCTA can be helpful in identifying the narrowing of vessel diameters which leads to reduced retinal VD and indirectly implying impaired blood flow. Suwan et al. found that patients with myopia alone had lower peripapillary VD compared to healthy controls [41]. Furthermore, patients with both myopia and glaucoma had an even greater reduction in VD that allowed the researchers to distinguish these patients from healthy controls, myopia only patients, and patients with glaucoma alone [41].

Recent work by Chang et al. found that both peripapillary and macular perfusion density were significantly reduced in patients with high myopia (HM) compared to non-high myopic (NHM) healthy controls [42]. However, macular perfusion
