Keywords

1 Anatomy and Histology for Macular Imaging by OCT

The retina is a multi-layered structure made up of two synaptic (plexiform) layers sandwiched between three nuclear layers: the outer nuclear layer (ONL), inner nuclear layer (INL), and ganglion cell layer (GCL). The ONL contains the cell bodies of photoreceptors (rods and cones), and the INL contains the cell bodies of bipolar, horizontal, amacrine, and radial glial (Müller) cells. The GCL contains displaced amacrine and retinal ganglion cells (RGC). These three nuclear layers are separated by two synaptic (plexiform) layers. The outer plexiform layer (OPL) is located between the ONL and the INL, and is the place where the photoreceptors, horizontal and bipolar cell dendrites interact. The inner plexiform layer (IPL) lies between the INL and GCL and is where the bipolar cell axons, amacrine, and ganglion cells interact. The retinal nerve fiber layer (RNFL) consists primarily of RGC axons.

Optical coherence tomography (OCT) technology has attained improved axial resolution to 5–8 μm. This makes it possible to correlate OCT images in-vivo with histological features of the retina. Despite the fact that the OCT features of the outer retina are less well understood and remain controversial, the features of the inner retina appear to be well correlated with histology. Within the retina, the RNFL and the plexiform layers (both inner and outer) are seen as hyperreflective, while the GCL and the nuclear layers (both inner and outer) are hyporeflective. The retinal blood vessels can be seen in OCT images as circular hyperreflective structures in the inner retina, with a reduced reflectivity that extends into the deeper layers (vertical shadow). A representative OCT image of a human retina with annotated layers is provided in Fig. 1.

Fig. 1
figure 1

Human macular structure. Retinal layers (Left) and representative human macular structure imaged by swept-source optical coherence tomography (SS-OCT) (Middle and Right). Retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), and inner plexiform layer (IPL) are annotated in green solid lines. The foveal structure in the yellow inset is magnified and each retinal layer is annotated. RNFL, retinal nerve fiber layer; GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium; BM, Bruch’s membrane

Full size image

The macula is an oval-shaped pigmented area (macula lutea) in the center of the retina, about 5.5 mm in diameter in humans. It covers the region surrounding the fovea with the highest density of RGCs (over 30% of RGCs) despite occupying less than 2% of the retinal area. The macular structure is characterized by less variation among individuals, and therefore, is relatively less affected by ocular or demographic factors than is the peripapillary RNFL. Glaucoma is characterized by progressive loss of RGCs, with typical changes in the optic nerve head, resulting in progressive visual field defect in the corresponding area. It has long been well documented that early-glaucomatous damage can affect the macula (Anctil and Anderson 1984; Heijl and Lundqvist 1984). However, clinicians paid less attention to the macula until accurate measurements of each of the retinal layers of the macula became possible in-vivo. Lately, several commercially available OCT devices have made available various types of macular parameters. The two representative macular parameters are the combinations of measurements of the RNFL, GCL, and IPL: the ganglion cell-inner plexiform layer (GCIPL) thickness is the sum of the thicknesses of the GCL and IPL; the ganglion cell complex (GCC) thickness is the sum of the thicknesses of the RNFL, GCL, and IPL. These two parameters are the most commonly used and obviously essential for early detection and monitoring of glaucoma.

2 Topographic Correlations among Optic Nerve Head, RNFL, and Macula

The axons of the RGC lie in the RNFL in bundles and enter the optic disc. The RNFL bundles each have their own course of travel to the optic disc. The RNFL at the temporal side of the fovea travels to the optic disc in an arcuate shape around the fovea. The axons of the RGC do not cross the horizontal meridian, so that a raphe with relatively few axons is formed in the temporal area (temporal raphe). According to this anatomy, most macular change in glaucomatous eyes appears in an arcuate-to-crescent shape, and is located mainly in the temporal macular regions along the horizontal raphe. The macular change also is usually located within the same hemifield on the continuum of the corresponding RNFL defect and optic disc damage (Fig. 2).

Fig. 2
figure 2

Macular damage in glaucoma. The axons of the retinal ganglion cells (RGC) travel in the retinal nerve fiber layer (RNFL) in bundles and enter the optic disc. The RNFL at the temporal side of the fovea travels to the optic disc in an arcuate shape around the fovea. The RGC axons do not cross the horizontal meridian, so that a raphe with relatively few axons is formed in the temporal area (temporal raphe, red dotted line). RGC axons in the superior macula enter through the temporal border of the optic disc while RGC axons in the inferior macula enter through the inferotemporal border of the optic disc. This topographical relationship between the macula and the optic disc results in an overlap of the inferior macular region with the inferotemporal region of the optic disc, where glaucomatous defects dominate. Hood et al. designated this region the “macular vulnerability zone (MVZ)” of the optic disc (blue arrow). Case of 40-year-old male with primary open-angle glaucoma (POAG) in right eye. Note the inferotemporal and superotemporal RNFL defects in the red-free RNFL photography. Thickness and deviation maps for the ganglion cell-inner plexiform layer (GCIPL) from Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, CA, USA) are provided. Macular damage is manifest in the inferotemporal macular region (inferotemporal GCIPL thickness = 66 μm), in an arcuate shape that does not cross the horizontal raphe. The superior RNFL defect does not invade the superior macular region, with the result that the superior macula is not damaged

Full size image

It is widely accepted that glaucomatous defect is most prevalent in the inferotemporal and superotemporal regions of the optic disc. The neural canal opening is connected to the superonasal region of the eyeball. This means that the center of the optic disc lies about 6° above the horizontal midline through the center of the fovea. Therefore, the RGC axons in the superior macula enter through the temporal border of the optic disc, while RGC axons in the inferior macula enter through the inferotemporal border of the optic disc. This topographical relationship between macula and optic disc results in an overlap of the inferior macular region with the inferotemporal region of the optic disc, where glaucomatous defects dominate. Hood et al. designated this region as the “macular vulnerability zone (MVZ)” of the optic disc (Fig. 2) (Hood et al. 2013; Hood 2017). Most of the RGCs in the inferior macular region project to the MVZ, which is the commonly affected region in glaucoma.

Kim et al. (2014a) further investigated the topographical relationship between localized peripapillary RNFL defect and macular GCIPL defect. They utilized the MATLAB program to construct a “GCIPL deviation frequency map” from 140 eyes of 140 open-angle glaucoma patients showing localized RNFL defect in one hemifield. According to the analysis from six different clock-hour locations of RNFL defect, the GCIPL defects had an arcuate shape that appeared as a continuation of the RNFL defect in the same hemisphere. The temporal macular region was the most frequently damaged site in either hemifield, and was larger in the inferior hemifield than in the superior (Fig. 3).

Fig. 3
figure 3

Topographic relationship between localized peripapillary RNFL and macular GCIPL defects. A “GCIPL deviation frequency map” was constructed by superimposing the ganglion cell-inner plexiform layer (GCIPL) defects onto the deviation map corresponding to peripapillary retinal nerve fiber layer (RNFL) defects at different clock-hour locations (from 6 to 12 o’clock). The color-coded scale presents the frequency of GCIPL defects: the red-colored region indicates the highest GCIPL defect frequency, while the blue-colored region is close to zero. The GCIPL defects had an arcuate shape that appeared as a continuation of the RNFL defect in the same hemisphere. The temporal macular region was the most frequently damaged site in either hemifield, and was larger in the inferior hemifield than in the superior (adapted and modified from Kim et al. 2014a)

Full size image

3 Diagnostic Accuracy of Macular Imaging

The macular parameters are known to have a high diagnostic utility for perimetric glaucoma comparable to those of peripapillary RNFL thickness or optic nerve head (ONH) parameters. In one study, the minimum (defined as the GCIPL thickness on the meridian showing the lowest average measurement) and inferotemporal GCIPL showed the highest sensitivity (94.8%) among GCIPL parameters, with comparable specificities (87.9 and 85.8%, respectively) (Mwanza et al. 2012). In another study, the GCC also showed good diagnostic utility for glaucoma comparable to that of peripapillary RNFL thickness, where global loss volume (percentage of global GCC loss over the entire GCC map) was found to be the best parameter (Tan et al. 2009).

3.1 Diagnostic Accuracy for Preperimetric and Early Glaucoma

Consistent with earlier studies investigating early-glaucomatous damage affecting the macula, the macular parameters have exhibited excellent RNFL-comparable diagnostic power for discriminating early-stage glaucoma (glaucoma suspect or preperimetric glaucoma). Mwanza et al. (2012) for the first time reported that minimum GCIPL thickness was found to be the best GCIPL parameter for discriminating early glaucoma (including both preperimetric and perimetric) from normal eyes. In a Korean study that had enrolled 92 preperimetric glaucoma patients and 92 age-matched controls, the inferotemporal GCIPL had comparable diagnostic power for glaucoma (area under receiver operating characteristic curves [AUROC] = 0.823), which results were not significantly different from those of the best parameter for peripapillary RNFL (7 o’clock sector, AUROC = 0.764) and ONH (rim area, 0.767) (Kim et al. 2014b). As the angular distance between the fovea and an RNFL defect increases, the sensitivity of macular GCIPL parameters for detection of RNFL defects deteriorates (Kim et al. 2014b; Hwang et al. 2014).

3.2 Diagnostic Accuracy in Myopic Eyes

Myopic eyes have divergent optic disc shapes such as are manifested in tilted disc, peripapillary atrophy, or posterior staphyloma. These structural variations make diagnosis and monitoring of glaucoma challenging in myopic eyes (Tan et al. 2019). Macular inner retinal thickness is known to be less affected by the degree of myopia and myopia-related ONH structural change relative to peripapillary RNFL thickness (Kim et al. 2015a; Jeong et al. 2016). In this context, macular GCIPL thickness can be an effective alternative method to evaluate glaucomatous damage in myopic eyes. A Korean study that included open-angle glaucoma patients with either highly myopic or non-highly myopic eyes reported that the best parameters for discriminating normal eyes from glaucomatous ones were inferior RNFL (0.906) and inferotemporal GCIPL (0.852) thickness in the highly myopic group, and average RNFL (0.920) and minimum GCIPL (0.908) thickness in the non-highly myopic group (Choi et al. 2013). Another Korean study further investigated inferotemporal GCIPL thickness’s diagnostic utility in a myopic preperimetric glaucoma population, and found it to be the best parameter, having a significantly greater diagnostic power than the other parameters including average RNFL or GCIPL thickness, rim area, inferior RNFL thickness, and minimum GCIPL thickness (Seol et al. 2015). A recent study using swept-source OCT (SS-OCT, deep range imaging [DRI] OCT, Topcon, Tokyo, Japan) demonstrated that inferotemporal GCL+ (identical with GCIPL thickness) and GCL++ (identical with GCC thickness) had the greatest AUROC for myopic glaucoma (Kim et al. 2020). In this study, macular GCL++ thickness (87.6%) and GCL+ thickness (87.5%) showed greater AUROC than did macular GCIPL thickness based on spectral-domain OCT (Cirrus HD-OCT, Carl Zeiss Meditec, Dublin, CA, USA).

However, it should be noted that the macular parameters are not always dispositive in myopic eyes. The current normative database in most commercially available OCT devices does not fully represent the myopic population, with the result that the prevalence of false-positive red signs can reach 40.4% (Kim et al. 2015b). Myopic macular degeneration can cause abnormal thinning (due to patch atrophy) or thickening (due to retinoschisis) of the retinal layers, and these structural abnormalities may deteriorate the accuracy of OCT devices’ automated segmentation (Tan et al. 2019; Ruiz-Medrano et al. 2019). Therefore, before interpreting macular parameters in glaucoma diagnostics, it would be prudent to carefully examine the entire macular structure and OCT scan quality rather than rely solely on the color-coded map and thickness profile.

4 Instrument-Specific Maps and Interpretation

4.1 Cirrus HD-OCT

Cirrus HD-OCT is an SD-OCT with a speed of 68,000 scans/s (latest model: Cirrus HD 6000) (Cirrus HD-OCT User Manual). It provides GCIPL thickness in a Ganglion Cell Analysis (GCA) printout. The GCIPL thickness is measured within a horizontally oval 4.8 × 4.0 mm area excluding a central perifoveal ellipse of 1.2 × 1.0 mm area: global average, minimum, and 6 wedge-shaped sectoral GCIPL thicknesses are provided in the GCA printout (Fig. 4). The normative database for macular measurements consists of 282 normal subjects aged between 19 and 84 years (mean: 46.5 years). Only 28 subjects aged 70–79 years and 3 subjects above age 80 are included therein, so caution is needed when interpreting patient data in these age ranges (Cirrus HD-OCT User Manual). There is no normative database for subjects younger than 19 years. Cirrus HD-OCT additionally provides an “Asian normative database” consisting of 315 individuals from Japan, China, and India (aged 19–79). All of these normative databases are adjusted only by age, and the color-coded maps in the GCA printouts do not take into account differences that may be present due to ethnicity, axial length, refraction, optic disc area, or signal strength (Cirrus HD-OCT User Manual).

Fig. 4
figure 4

Cirrus HD-OCT: Ganglion Cell Analysis. Typical printout of Ganglion Cell Analysis (GCA) from Cirrus HD-OCT. The thickness map (horizontally oval 4.8 × 4.0 mm area excluding central perifoveal ellipse 1.2 × 1.0 mm sized area) is provided in the first row. In the second row are the deviation maps showing regions where GCIPL thickness fell below the 5% limit based on the normative database with the yellow- (percentile values <5%) and red-color superpixels (percentile values <1%). Thickness measurements (average, minimum, and six sectoral measurements) with color-coded maps based on the internal normative database are provided in the center

Full size image

Besides thickness measurements, the GCA printout provides additional GCIPL thickness and deviation maps. The thickness map provides the raw measurement thickness profiles on a color-scaled map. The deviation map shows regions wherein GCIPL thickness has fallen below the 5% limit based on the normative database with the yellow- (percentile values <5%) and red-color superpixels (percentile values <1%).

4.1.1 Temporal Raphe Sign

Temporal raphe sign (or GCIPL hemifield test) is determined to be positive in subjects who have a straight line on the horizontal raphe longer than one-half of the length between the inner and outer annulus in the temporal elliptical area of the GCIPL color thickness map (Fig. 5). This sign is known to be more effective than other macular parameters in discriminating early-glaucomatous changes (AUROC 0.967 in preperimetric glaucoma and 0.962 in perimetric glaucoma) (Kim et al. 2015c). This method is easy to apply in clinical practice and is advantageous for discrimination of glaucomatous change in highly myopic eyes that have a depigmented fundus and where RNFL defects are difficult to find (Kim et al. 2016). Temporal raphe sign positivity at the baseline has been associated with faster conversion to glaucoma in elderly subjects with a large cup-to-disc ratio (Ha et al. 2020).

Fig. 5
figure 5

Temporal Raphe sign. Temporal raphe sign (or GCIPL hemifield test) is determined to be positive in subjects who have a straight line longer than one-half of the length between the inner and outer annulus in the temporal elliptical area of the GCIPL color thickness map (red arrow). In this case, the patient has an inferotemporal retinal nerve fiber layer (RNFL) defect and a corresponding inferotemporal ganglion-cell inner plexiform layer (GCIPL) defect

Full size image

4.1.2 PanoMap

Cirrus HD-OCT integrates the optic disc cube and macular cube scans into a single image and provides a combined wide-field OCT map, the so-called PanoMap (Fig. 6). The integration of the RNFL and GCIPL maps is useful for understanding the topographic patterns of glaucomatous structural damage in the early stages of glaucoma. By this method, Kim et al. (2017a) demonstrated that no single case of peripapillary RNFL defect in the MVZ was found without inferior macular GCIPL loss. However, there were a few cases of inferior macular GCIPL loss without peripapillary RNFL defect in the MVZ. This finding suggests that inferior macular GCIPL loss may precede peripapillary RNFL defect in the MVZ. In Kim et al.’s (2017b) subsequent longitudinal observation of deviation maps of macular GCIPL and peripapillary RNFL from 151 eyes with early-stage glaucoma, macular GCIPL change was frequently detected before corresponding peripapillary RNFL change (Fig. 6). Another study, this one from Australia, further demonstrated an association between lower intraocular pressure and glaucomatous structural change manifesting at the macular GCIPL before manifesting at the peripapillary RNFL (Marshall et al. 2019). These findings highlight the importance of macular imaging in glaucoma patients, as peripapillary RNFL analysis alone can overlook macular damage. The PanoMap is also known to be helpful in the detection of early-stage structural progression by inspecting various progression patterns of RNFL and GCIPL from a single combined image (Lee et al. 2018a, b).

Fig. 6
figure 6

PanoMap: combined GCA and RNFL deviation map. Cirrus HD-OCT PanoMap Analysis integrates data from optic disc cube and macular cube scans into a single image and provides a wide-field perspective for comprehensive posterior segment analysis. A case of 27-year-old male with myopia (spherical equivalence –5 diopters, axial length = 25.65 mm) and open-angle glaucoma in his right eye. At baseline (2012-12-07), the patient had inferoinferior retinal nerve fiber layer (RNFL) and inferotemporal ganglion-cell inner plexiform layer (GCIPL) defects. The automated perimetry showed a few superior paracentral scotomas (mean deviation, –2.14 dB). Eight years later (2020-09-08), the patient showed progressive thinning of the RNFL in the macular vulnerability zone (MVZ) (blue arrow) with increased nasal and paracentral scotomas

Full size image

4.2 Spectralis OCT

Spectralis OCT (Heidelberg Engineering GmbH, Heidelberg, Germany) has Glaucoma Module Premium Edition (GMPE) software that provides a Posterior Pole Asymmetry Analysis consisting of 61 horizontal B-scans, each comprised of 768 A-scans that are acquired along the Bruch’s membrane opening (BMO)-fovea axis (Heidelberg Engineering). It provides a thickness map, a thickness deviation map, and a color-coded classification chart for the total retina, RNFL, GCL, and IPL (Fig. 7). On the thickness deviation map, the red and yellow areas represent percentile values <1% and <5%, respectively, while the blue and purple areas represent percentile values >95% and >99%, respectively. The green areas represent a percentile value of 5–95%. There is no RNFL color-coded classification chart, because the macular RNFL is anatomically thin, which limits the reliability of measurements in this location, and also because RNFL defects are typically most prominent beyond the confines of the GCL-optimized grid (Heidelberg Engineering).

Fig. 7
figure 7

Spectralis OCT: thickness and deviation maps. Spectralis OCT provides a thickness map, thickness deviation map, and color-coded classification chart for the total retina, retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), and inner plexiform layer (IPL). In the thickness deviation map, the red and yellow areas represent percentile values <1% and <5%, respectively, while the blue and purple areas represent percentile values >95% and >99%, respectively. The green areas represent a percentile value of 5–95%. There is no RNFL color-coded classification chart, because the macular RNFL is anatomically thin, which limits the reliability of measurements in this location, and also because RNFL defects are typically most prominent beyond the confines of the GCL-optimized grid

Full size image

Total retinal thickness is further displayed on a color-coded map (Fig. 8). A thickness asymmetry graph between the inferior and superior macular hemispheres of each eye, and between the right and left eyes, is also available. The darker the square, the larger the difference in thickness between that square and the corresponding square in the opposite hemisphere or eye, with jet-black squares indicating a difference >30 μm.

Fig. 8
figure 8

Posterior pole asymmetry analysis. Spectralis OCT provides a retinal thickness asymmetry graph between the inferior and superior macula hemisphere of each eye. A case of 28-year-old male with open-angle glaucoma on his left eye. The patient had mild superotemporal and more severe inferotemporal retinal nerve fiber layer (RNFL) defects. The thickness asymmetry graph (bottom right) revealed asymmetrical retinal thinning in the inferior region

Full size image

4.3 DRI OCT Triton

DRI OCT Triton (Topcon, Tokyo, Japan) is a swept-source OCT using a longer, 1,050 nm wavelength light-source that enables a deeper imaging range and better tissue penetration. It provides a faster scanning speed of 100,000 A-scans/s, which significantly reduces motion artifacts with short capture times and collects more OCT data in a single scan. The 3D Wide scan in DRI OCT Triton provides a 12 × 9 mm wide-field thickness map, a thickness surface map, and SuperPixel maps for the RNFL, GCL++, and GCL+ layers. The thickness surface map (second row) is displayed in 3D for selected layers (RNFL, GCL++, or GCL+) and can be rotated or zoomed-in and -out by mouse operation (Fig. 9). The SuperPixel-200 map consists of 26 × 26 grids within a 5.2 × 5.2 mm2 peripapillary area and 30 × 30 grids within a 6.0 × 6.0 mm2 macular area, thereby providing a significance map (identical with the deviation map in Cirrus OCT) based on the built-in normative database. The side length of each grid is 200 μm. Uncolored pixels indicate the normal range, whereas yellow- and red-colored pixels indicate abnormality at P = 1–5% and P < 1% of the normal level, respectively.

Fig. 9
figure 9

DRI OCT Triton: wide-field map. The 3D Wide scan in DRI OCT Triton provides a 12 × 9 mm wide-field thickness map (first row), a thickness surface map (second row), and SuperPixel maps (third row) for the retinal nerve fiber layer (RNFL) (second column), GCL++ (third column), and GCL+ layers (fourth column). SuperPixel maps can be displayed as a combination of RNFL with either GCL++ or GCL+ (first column). The thickness surface map (second row) is displayed in 3D for selected layers (RNFL, GCL++, or GCL+, respectively), and can be rotated or zoomed-in and -out by mouse operation. The SuperPixel-200 map consists of 26 × 26 grids within a 5.2 × 5.2 mm2 peripapillary area and 30 × 30 grids within a 6.0 × 6.0 mm2 macular area, thereby providing a significance map based on the built-in normative database. The side length of each grid is 200 μm. The uncolored pixels indicate the normal range, whereas the yellow- and red-colored pixels indicate abnormality at P = 1–5% and P < 1% of the normal level, respectively

Full size image

The inferior GCL++ (AUROC = 0.809) and inferotemporal GCL+ thickness (AUROC = 0.809) showed the greatest diagnostic ability for detection of preperimetric glaucoma from healthy eyes (Lee et al. 2017a). In addition, inferotemporal GCL++ (AUROC = 0.865) and GCL+ thicknesses (AUROC = 0.865) showed the best diagnostic power for early-perimetric glaucoma (Lee et al. 2017a). The wide-field thickness map showed comparable diagnostic power for preperimetric and early-perimetric glaucoma with Cirrus HD-OCT (Lee et al. 2018c). A recent study that enrolled 150 myopic primary open-angle glaucoma (POAG) eyes and 100 healthy myopic eyes reported that the RNFL/GCL++/GCL+ wide-field thickness (thickness surface) map showed better accuracy for glaucomatous defects in both the superotemporal and inferotemporal regions than did the Cirrus HD-OCT thickness map (Kim et al. 2020). This possibly was due to the difference in the measurement regions between the two devices. The measurement area of the macular parameter is within a 6 mm diameter circle for DRI OCT Triton and within a 4.8 × 4.0 mm elliptical annulus for Cirrus HD-OCT. A wider scan range in SS-OCT may be more advantageous for diagnosis of POAG in myopic eyes. Furthermore, the difference in the segmentation algorithms between the two OCT devices may influence the diagnostic power for myopic POAG (Kim et al. 2020; Pierro et al. 2012).

5 Detecting Glaucoma Progression

Monitoring of macular parameters facilitates detection of structural progression in glaucomatous eyes and prediction of subsequent visual field defect. The rate of GCIPL thinning reportedly is significantly faster in glaucomatous eyes with progression (Lee et al. 2017b, c). The mean rate of GCIPL thinning has been found to be significantly faster, even, in pseudoexfoliative glaucoma (–1.46 μm/year) than in open-angle glaucoma (–0.49 μm/year) or normal eyes (–0.31 μm/year) (Lee et al. 2019). A trend-based analysis (>5 years) of macular GCIPL and peripapillary RNFL from 163 POAG patients revealed that progressive macular GCIPL and peripapillary RNFL thinning were mutually predictive, and that both were indicative of visual field progression (Hou et al. 2018). Another retrospective cohort study, this one from Korea, also demonstrated that eyes with progressive GCIPL and RNFL thinning showed a significant higher risk of developing visual field defects (Shin et al. 2020).

Trend-based analysis of GCIPL is also known to be useful in detecting progression even in advanced-glaucoma eyes. Peripapillary RNFL is likely to reach the measurement floor in advanced-glaucoma eyes, which makes it less valuable for detection of progression in these eyes. However, recent studies reported significant rates of change for macular GCIPL but no significant changes for peripapillary RNFL or rim area in advanced-glaucoma eyes (Shin et al. 2017; Lavinsky et al. 2018). This suggests that damage impacting the macular GCIPL occurs at a different rate than in the case of the peripapillary RNFL.