Parapapillary Deep-Layer Microvasculature Dropout in Primary Open-Angle Glaucoma Eyes With a Parapapillary γ-Zone

Abstract

Purpose: To explore the clinical characteristics of parapapillary deep-layer microvasculature dropout (MvD) associated with parapapillary γ-zone in primary open-angle glaucoma (POAG).

Methods: A total of 150 eyes with POAG and 75 age- and axial length-matched control eyes, both having parapapillary γ-zone as determined by spectral-domain (SD) optical coherence tomography (OCT), were included. Parapapillary MvD was defined as a focal sectoral capillary dropout without any visible microvascular network identified in deep-layer en-face images obtained using swept-source OCT angiography (OCTA). Optic nerve heads were imaged using SD-OCT to examine the microstructure of β-parapapillary atrophy, and to measure areas of the β-zones (area with intact Bruch's membrane) and γ-zones (area devoid of Bruch's membrane). Factors associated with the presence of MvD were determined using logistic regression analyses.

Results: Parapapillary deep-layer MvD was identified in 117 of 150 POAG eyes (78.0%) with a parapapillary γ-zone, whereas none of the healthy control eyes showed MvD. MvD was universally identified within the γ-zone and additionally involved the β-zone in eyes with both γ- and β-zones. Multivariate logistic regression analyses showed significant influence of thinner global retinal nerve fiber layer (P = 0.001), worse visual field mean deviation (P < 0.001), as well as larger areas of parapapillary γ-zones (P ≤ 0.010) and β-zones (P ≤ 0.005) on the presence of MvD.

Conclusions: MvD was frequently identified in the γ-zone in POAG eyes, but not in healthy eyes. Worse visual field loss and larger areas of γ- and β-zones were associated with the presence of MvD in POAG eyes with a parapapillary γ-zone.

Glaucoma is characterized by the progressive loss of retinal ganglion cells and their axons. Although intraocular pressure (IOP)-related mechanical stress is a principal risk factor, ocular blood flow may play a pathogenic role in the development and progression of glaucoma.¹ The microvasculature of the peripapillary area has been of particular interest because of its close association with the optic nerve head (ONH) structure and the fact that it is downstream from the short posterior ciliary artery,^2–6 which also perfuses the prelaminar tissue and lamina cribrosa.^4–6 

Fluorescein^3,7,8 and indocyanine green angiographic studies² have revealed perfusion disturbance in areas with parapapillary atrophy (PPA). In agreement with these findings, recent studies using optical coherence tomography (OCT) angiography (OCTA) reported regional microvasculature dropout (MvD) within the PPA in glaucoma patients.^9,10 Akagi et al.⁹ showed that regional MvD, which was found in the PPA area in glaucomatous eyes, was topographically correlated with hemifield visual field (VF) defects. Suh et al.¹⁰ also observed MvD within the PPA area, and reported that the presence of MvD was associated with a lower diastolic blood pressure (BP) and a thinner choroid. However, neither study considered the microstructure of the PPA. 

The PPA is divided into areas that contain the Bruch's membrane (BM) and the underlying choroid (β-zone), and areas devoid of the BM and choroid (γ-zone). The parapapillary β-zone is histologically characterized by atrophy of the retinal pigment epithelium and closure of the choriocapillaris,^11,12 and has been considered an age-related choroidal atrophy.¹³ On the other hand, the γ-zone is associated with the long axial length and oblique border tissue of the Elschnig, and is devoid of the BM and choriocapillaris.^13,14 These findings suggest that the γ-zone is the exposed inner surface of the peripapillary scleral ring, which results from scleral stretching associated with eyeball elongation.^13–15 Therefore, although both the β- and γ-zones are supplied by the same vascular system (i.e., short posterior ciliary arteries), it is possible that the stress implied on the microvessels within each zone differs, and so the pathogenesis of MvD differs between eyes in the β- and γ-zones. 

Studies exploring the pathogenesis of MvD should be performed separately for eyes with the β- and γ-zones since the two structures are derived through different mechanisms. This study explored the clinical characteristics of MvD associated with the γ-zone to increase our understanding of the pathogenesis of MvD in eyes with the γ-zone. 

Methods

This study investigated the peripapillary deep-layer circulation using OCTA in consecutive primary open-angle glaucoma (POAG) patients and healthy control subjects who were enrolled in the Investigating Glaucoma Progression Study, which is an ongoing prospective study at the Glaucoma Clinic of Seoul National University Bundang Hospital Glaucoma Clinic. Written informed consent to participate was obtained from all of the subjects. The study protocol was approved by the Institutional Review Board of Seoul National University Bundang Hospital and it followed the tenets of the Declaration of Helsinki. 

All of the participants underwent comprehensive ophthalmic examinations that included best-corrected visual acuity (BCVA), Goldmann applanation tonometry, a refraction test, slit-lamp biomicroscopy, gonioscopy, stereo disc photography, red-free fundus photography (EOS D60 digital camera; Canon, Utsunomiyashi, Tochigiken, Japan), central corneal thickness measurement (Orbscan II; Bausch & Lomb Surgical, Rochester, NY, USA), axial length measurement (IOLMaster version 5; Carl Zeiss Meditec, Dublin, CA, USA), corneal curvature measurement (KR-1800; Topcon, Tokyo, Japan), spectral-domain OCT (SD-OCT) scanning of the circumpapillary retinal nerve fiber layer (RNFL) and the optic disc using enhanced depth imaging (EDI) mode (Spectralis; Heidelberg Engineering, Heidelberg, Germany), standard automated perimetry (Humphrey Field Analyzer II 750, 24-2 Swedish interactive threshold algorithm; Carl Zeiss Meditec), and swept-source OCTA (DRI OCT Triton; Topcon). The clinical history was also obtained from all of the participants including demographic characteristics, the presence of cold extremities or migraine, and other systemic diseases. Systolic and diastolic BP were measured at the time of OCTA. Mean arterial pressure (MAP) and mean ocular perfusion pressure (MOPP) were calculated based on the following equations: MAP = diastolic BP + 1/3 (systolic BP–diastolic BP) and MOPP = MAP–IOP at the time of OCTA. 

POAG was defined as the presence of an open iridocorneal angle, signs of glaucomatous optic nerve damage (i.e., neuroretinal rim thinning, notching, or an RNFL defect), and a glaucomatous VF defect. A glaucomatous VF defect was defined as a defect conforming with one or more of the following criteria: outside normal limits on a glaucoma hemifield test; three abnormal points with a P < 5% probability of being normal and one abnormal point with a P < 1% by pattern deviation; or a pattern standard deviation of P < 5% confirmed on two consecutive reliable tests (fixation loss rate ≤ 20% and false-positive and false-negative error rates ≤ 25%). Healthy controls had an IOP ≤ 21 mm Hg, no history of increased IOP, an optic disc with a normal appearance, and a normal VF. Healthy control eyes were selected by matching them with POAG eyes in terms of age and axial length. The specific parameter values used to indicate a match were as follows: age within 2 years and axial length within 0.1 mm. Each healthy eye was matched with two POAG eyes (i.e., 2:1 matching). To be included in this study, both POAG and healthy control eyes were required to have a parapapillary γ-zone (see below). 

Exclusion criteria included eyes with a BCVA worse than 20/40, a spherical equivalent <−9.0 D or >+3.0 D, a cylinder correction <−3.0 D or >+3.0 D, a history of intraocular surgery with the exception of uneventful cataract surgery or trabeculectomy, or retinal or neurologic diseases. When both eyes were eligible, one eye was randomly selected for inclusion in the study. 

Eyes included in the POAG group were required to have a record of untreated IOP, which was measured prior to the initiation of ocular hypotensive treatment or was identified in the referral notes. In patients with an untreated IOP ≤ 21 mm Hg, diurnal variation was measured during office hours (9 AM to 5 PM). In patients undergoing treatment with ocular hypotensive medication at the time of the initial visit, diurnal variation was measured after a 4-week washout period. 

OCTA and Determination of the Presence of MvD

The optic nerve and peripapillary area were imaged using a commercially available swept-source OCTA device (DRI OCT Triton; Topcon) with a central wavelength of 1050 nm, an acquisition speed of 100,000 A-scans per second, and an axial and transversal resolution of 7 and 20 μm in tissue, respectively. Scans were taken from 4.5 × 4.5 mm cubes, with each cube consisting of 320 clusters of four repeated B-scans centered on the optic disc. 

Deep-layer microvasculature in the peripapillary area was evaluated on the en-face images of the peripapillary deep-layer generated based on the automated layer segmentation performed using OCT instrument software.^16,17 The en-face images of the deep-layer were derived from an en-face slab, extending from the retinal pigment epithelium to the outer border of the sclera. 

MvD was defined as a focal sectoral capillary dropout without any visible microvascular network identified in the deep-layer en-face images, as previously described.^10,16,18 When the circumferential width of the area with capillary dropout appeared more than two times greater than the width of visible juxtapapillary microvessels with its border adjoining the optic disc margin, an MvD was defined. Two independent observers (E.J.L. and J.-A.K.) identified MvDs while being blinded to the clinical information of the subjects. Disagreements between the two observers were resolved by a third adjudicator (T.-W.K.). All OCT B-scan images had to have an image quality score ≥30, according to the manufacturer's recommendation. When the quality of the OCTA images was poor (i.e., blurred images that hampered the delineation of the MvD), the eye was excluded from the analysis. 

OCT Scanning of the Peripapillary Area and Investigation of the Structure of the PPA

EDI OCT radial scanning of the optic disc was performed using Spectralis OCT to assess the structure of the PPA. A potential magnification error was removed by entering the corneal curvature of each eye into the Spectralis OCT system before performing EDI scanning. It has been shown that MvD is correlated with PPA,¹⁰ which can be divided into an area with BM and underlying choroid (β-zone) and an area devoid of BM and choroidal tissue (γ-zone).^14,15 This study only included eyes with a γ-zone wider than the width set for this study (radial width >200 μm in at least one meridian¹³ and radial width at the MvD location >100 μm for eyes with MvD). To our experience, this criterion works best for differentiating the γ- and β-zones, and characterizing the differed clinical nature and the rate of glaucoma progression between the groups. 

The radial widths and areas of the β- and γ-zones were manually measured using a built-in caliper tool of the Spectralis OCT system (Heidelberg Eye Explorer software version 1.7.0.0; Heidelberg Engineering). To do this, radial B-scan images were obtained using a 12-radial line scan centered on the optic disc. The scan angle spanned 20 degrees and the distance between each scan was 15 degrees. Only eyes with acceptable scans with good-quality images (i.e., quality score ≥ 15) obtained for all of the scans were included in the analysis. 

The maximal radial width of the γ-zone and radial width of the γ-zone at the MvD location (in eyes with MvD) were determined by investigating the 24 peripapillary meridians in the 12 radial scan images, using the display window of the Spectralis viewer (Fig. 1). The location of the MvD in the Spectralis infrared image was estimated by overlapping the en-face OCTA image on the infrared image (Fig. 1). The outer borders of the PPA and the margins of the BM opening and optic disc were then determined on the infrared images by simultaneously investigating the co-registered B-scans. Areas of the β- and γ-zones were defined as the areas outlined by the outer borders of the PPA and the BM opening, and those outlined by the BM opening and optic disc margin, respectively, and were measured on the infrared images in mm². Measurements were performed by two independent observers (E.J.L. and J.-A.K.), who were blinded to the clinical information of the subjects. The average of the two measurements from each observer was used for the analyses. 

Figure 1

 

Determination of the radial widths of the γ-zone and areas of the β- and γ-zones. Radial widths of the γ-zone were determined at 24 meridians defined on the 12 radial scan images (A). Black, red, and yellow dots and arrows indicate outer margin of β-zone, outer margin of γ-zone (BM opening), and clinical optic disc margin, respectively. Insets are radial B-scan images obtained at the location where the width of the γ-zone (horizontal distance between the red and yellow arrows) was the largest (a) and at the location with the MvD (b). The location of the MvD in the infrared image was determined by overlapping the en-face OCTA image on the infrared image (B). Lines indicating the outer margin of the β-zone, outer margin of the γ-zone (BM opening), and clinical optic disc margin were connected, and the areas of the β- and γ-zones were measured on the infrared images in mm² (C).

Data Analysis

The interobserver agreements regarding the presence of the MvD and measurements of the areas of the β- and γ-zones were determined using kappa statistics (i.e., κ value) and intraclass correlation coefficients, respectively. Comparison between groups was performed using the Student's t-test or 1-way analysis of variance with the LSD post hoc test for continuous variables and the χ² test for categorical variables. Factors associated with the presence and area of the MvD were assessed using univariate and multivariate logistic regression analyses. Multivariate analyses were performed in multiple ways to avoid multicollinearity. All of the statistical analyses were performed using SPSS statistical software for Windows, version 19.0 (SPSS, Chicago, IL, USA). Data are expressed as the mean ± standard deviation except when stated otherwise, and the cutoff for statistical significance was set at P < 0.05. 

Results

The study initially included 157 eyes with POAG and 98 control eyes, both of which had the γ-zone in the parapapillary area. Of these eyes, seven POAG and three control eyes were excluded because of poor-quality angiography images. Twenty control eyes were then additionally excluded after matching with the POAG group in terms of age and axial length, resulting in 150 and 75 eyes being finally included in the POAG and control groups, respectively. Table 1 gives the clinical characteristics of the participants. 

Table 1

 

Comparison of Clinical Characteristics Between POAG and Control Groups

 

An MvD was identified in 117 of 150 POAG eyes (78.0%), whereas none of the control eyes showed an MvD in the deep-layer OCTA images. The area of the γ-zone did not differ between groups, whereas the β-zone was significantly larger in eyes with POAG (P = 0.036). POAG eyes had a lower systolic and diastolic BPs, a lower MAP, and a lower MOPP (all P ≤ 0.037). There was excellent interobserver agreement regarding the detection of MvD, with κ = 0.957. The intraclass correlation coefficients for interobserver reproducibility in measuring areas of the β- and γ-zones were 0.965 and 0.987, respectively. 

Of the 117 eyes with an MvD, 11 and 72 had single MvDs in the superior and inferior hemispheres, respectively, and seven had large single MvDs involving both hemispheres. Two had double individual MvDs in the inferior hemispheres and 25 had double MvDs in both the superior and inferior hemispheres, resulting in a total of 144 MvDs. Table 2 provides the topographic relationship between the locations of MvD and glaucomatous VF defects, and shows that all MvDs were associated with a VF defect in the corresponding hemifield. 

Table 2

 

Topographic Correlation Between the Hemispheric Locations of MvD and the Hemifield Location Glaucomatous VF Defect

 

The MvD was identified in the PPA area and its inner and outer margins adjoined the optic disc margin and PPA margin, respectively. In eyes with a γ-zone only, the MvD was confined within the γ-zone (Fig. 2). In eyes with β- and γ-zones, the MvD involved both zones (Fig. 2). 

Figure 2

 

Eyes with POAG with a parapapillary deep-layer MvD in the γ-zone. Red, white, and yellow arrowheads indicate outer margin of the β-zone, outer margin of the γ-zone (BM opening), and clinical optic disc margin, respectively. In eyes with only a γ-zone, the MvD was confined within the γ-zone (A–C), while in eyes containing both β- and γ-zones, the MvD involved both zones (E–G). Note that the location of the MvD topographically corresponds to the hemifield VF defect (D, H).

Table 3 gives a comparison of clinical characteristics between POAG patients with and without MvD and healthy control subjects. POAG eyes with MvD had more severe VF damage and larger areas of β- and γ-zones (all P ≤ 0.011) than eyes without MvD. Compared to healthy control subjects, POAG patients with MvD had larger areas of β- and γ-zones, and larger degree of optic disc tilt. No difference was observed for the respective values between POAG patients without MvD and control subjects. Diastolic BP, MAP, and MOPP were significantly lower in POAG eyes with MvD, and systolic and diastolic BPs were significantly lower in POAG eyes without MvD, compared to control eyes. However, no significant differences were observed in the systemic characteristics between POAG eyes with and without MvD. 

Table 3

 

Comparison of Clinical Characteristics Between POAG Groups With and Without an MvD and Healthy Control Group

 

Univariate logistic regression analyses performed in the POAG group showed that a smaller global RNFL thickness (P < 0.001), a worse VF mean deviation (MD) (P < 0.001), and larger β- (P = 0.001) and γ-zones (P = 0.016) were associated with the presence of MvD. Multivariate analyses were performed in two ways, accounting for the multicollinearity between global RNFL thicknesses and VF MD, which showed a significant association of global RNFL thickness (P = 0.001), VF MD (P < 0.001), and areas of β- (P ≤ 0.005) and γ-zones (P ≤ 0.010) with the presence of a MvD. 

Discussion

This study showed that parapapillary MvD is often observed within the γ-zone in POAG eyes, but not in healthy eyes. The presence of MvD was associated with more severe structural and functional optic nerve damage and larger areas of β- and γ-zones. To the best of our knowledge, there is no study published in the literature exploring the factors associated with the presence of MvD in eyes with γ-zone. 

A previous study showed that a lower diastolic BP was associated with MvD.¹⁰ The finding is in agreement with a study by Memarzadeh et al.¹⁹ that showed an association between lower diastolic BP and a higher prevalence of POAG, suggesting that decreased diastolic perfusion pressure may be involved in the pathophysiology of glaucoma. In this study, however, none of the systemic factors including diastolic BP, systolic BP, MAP, and MOPP were associated with MvD. This result suggests that MvD in the γ-zone is not affected by decreased perfusion pressure. 

We found that the γ-zone area was larger in POAG eyes with parapapillary MvD than in eyes without MvD and healthy control subjects. In addition, a larger γ-zone area was significantly associated with the presence of MvD among POAG eyes. These results suggest that structural changes within the γ-zone are associated with the development of MvD. Considering that the γ-zone is likely the result of peripapillary scleral stretching, it is possible that microvessels within the scleral flange are damaged due to long-standing tensile stress. This damage may lead to the occlusion of microvessels. It is expected that the occlusion more likely occurs at the location under highest stress. Previous studies have suggested that ONH axons in myopic eyes are often under large tensile stress in the inferotemporal and superotemporal locations in relation with the direction of disc torsion.^20–22 Since the inferotemporal torsion is more common than the superotemporal torsion,^21,23 it is likely that the inferotemporal region would be under the greatest tensile stress. Therefore, MvD is expected to be most commonly found in the inferotemporal region. Indeed, in this study, MvD was most prevalent in the inferotemporal sector, followed by the superotemporal sector. 

Larger β-zone area was also associated with the presence of MvD in the γ-zone. The mechanism that explains how the β-zone is related to the presence of MvD in the γ-zone remains unclear. However, it is known that the retinal pigment epithelium is damaged or absent in the β-zone.¹¹ Because the retinal pigment epithelium comprises the outer blood–retinal barrier,²⁴ its attrition or absence in the β-zone region can be regarded as the area of disrupted blood retinal barrier. Based on this, it may be considered that vasoactive or toxic substances may be released from the β-zone^25,26 into the adjacent γ-zone, inducing a harmful influence.^27,28 This may facilitate the damage of the vessels under tensile stress within the nearby γ-zone. 

The MvD was only found in the POAG group. Furthermore, MvD was associated with more severe structural and functional damage. This result can be interpreted in two ways. First, MvD may provide additional insults (i.e., ischemia) to the ONH other than IOP-related mechanical stress, thereby increasing the susceptibility of ONH to glaucomatous damage. Alternatively, the finding suggests that MvD may be a secondary change due to glaucomatous damage: metabolic requirements of the ONH decreases in glaucoma resulting in decreased perfusion to the ONH and subsequent obliteration of microvessels supplying the ONH. Whichever the case is, this result indicates that MvD in the γ-zone may be potentially implicated in the glaucoma pathogenesis. 

Axial length was not a significant factor influencing MvD. Given the varied elastic properties among individual eyes, the response of peripapillary tissues to axial elongation may differ between eyes. Thus, the area of the γ-zone may not necessarily represent the amount of axial elongation. It is possible that rather than the amount of axial elongation, the presence and extent of the γ-zone better represents the stress imposed on peripapillary tissues including the retinal ganglion cell axons, scleral flange, and microvasculature within it. 

This study had several limitations. First, because of its cross-sectional design, this study could not determine the causal relationship between the MvD and glaucomatous damage. Second, flow projection artifacts from superficial blood vessels could have hampered the precise evaluation of MvD.^29,30 This might have resulted in the underestimation of the prevalence of MvD. 

In conclusion, MvD was frequently identified in POAG eyes with parapapillary γ-zones. Larger areas of the parapapillary β- and γ-zones and more severe glaucomatous damage were associated with the presence of MvD. 

Table 4

 

Factors Associated With the Presence of MvD in the POAG Group (n = 150)

 

Acknowledgments

Supported by Grant No. 06-2016-174 from the Seoul National University Bundang Hospital Research Fund. 

Disclosure: E.J. Lee, None; T.-W. Kim, None; J.-A. Kim, None; J.-A. Kim, None