Structural and Functional Analyses in Nonarteritic Anterior Ischemic Optic Neuropathy: Optical Coherence Tomography Angiography Study

To evaluate the validity of the prevailing concept that Susac syndrome (SS), a rare microvasculopathy of the brain, retina, and inner ear, is a self-limiting disease. We performed a literature search to identify all cases of SS reported between 1973 and October 2015. If available, we determined their demographics, duration of follow-up, and the clinical course that was labeled as monocyclic or polycyclic. We attempted to determine the number of relapses and the relapse rate in patients with polycyclic disease. Our literature search yielded 185 relevant publications reporting 405 cases of SS. The duration of follow-up could be determined in 247/405 cases, with a range 0.5-312 months. The mean was 41 months but the distribution was skewed, with a median of 24 months. Defining the clinical course as monocyclic or polycyclic was possible in 102 patients who were followed for greater than 24 months; 53 were identified as having a polycyclic course. Patients labeled polycyclic were followed longer than those labeled monocyclic (median 62 vs 42 months, P < 0.001). The number or frequency of attacks per patient could not be determined. The follow-up of published cases of SS is short, creating an inherent bias toward the impression that the disease is self-limiting. Our findings suggest that stratification of SS into monocyclic, polycyclic, and chronic continuous courses may oversimplify the phenotype of SS; instead, the possibility of a relapsing-remitting course must be considered in all patients with this disorder.

Original Contribution Structural and Functional Analyses in Nonarteritic Anterior Ischemic Optic Neuropathy: Optical Coherence Tomography Angiography Study Masayuki Hata, MD, Akio Oishi, MD, PhD, Yuki Muraoka, MD, PhD, Kazuaki Miyamoto, MD, PhD, Kentaro Kawai, MD, Satoshi Yokota, MD, Masahiro Fujimoto, MD, Manabu Miyata, MD, PhD, Nagahisa Yoshimura, MD, PhD Background: Retinal and optic disc perfusion in nonarteritic anterior ischemic optic neuropathy (NAION) is incompletely understood. Our aim was to investigate the characteristics of the microvascular structures at the peripapillary area and optic disc, and their associations with retinal structure and function in patients with NAION. Methods: We conducted a prospective, observational case series study. Thirty-four eyes, consisting of 15 NAION eyes and 19 normal eyes, were included. Optical coherence tomography (OCT) angiography was used to measure the vessel densities in the peripapillary superficial retina and whole-depth mode inside the optic disc. Measurement of circumpapillary retinal nerve fiber layer (cpRNFL) thickness was performed using OCT. Sectorial division analysis of cpRNFL was performed by eliminating the influences of the difference in disc rotation between OCT images and OCT angiography images. Results: The vessel densities of peripapillary retina and inside the optic disc were significantly reduced in the NAION compared to the normal (both P , 0.001). Both the severity of visual field defect and cpRNFL thinning were significantly associated with the peripapillary vessel Department of Ophthalmology and Visual Sciences, Kyoto University, Kyoto, Japan. Supported in part by the Japan Society for the Promotion of Science (JSPS), Tokyo, Japan (Grant-in-Aid for Scientific Research, No. 21592256), Japan National Society for the Prevention of Blindness, Tokyo, Japan, and Innovative Techno-Hub for Integrated Medical Bio-Imaging of the Project for Developing Innovation Systems, from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) in Japan. A. Oishi received a grant from Alcon Japan (Tokyo, Japan). N. Yoshimura is a paid advisory board member for NIDEK, Inc, (Gamagori, Japan), and has received lecture fees from Topcon Corporation (Tokyo, Japan) and research funding from Canon, Inc, (Tokyo, Japan). The remaining authors report no conflict of interests. Address correspondence to Masayuki Hata, MD, Department of Ophthalmology and Visual Sciences, Kyoto University, 54 Shougoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan; E-mail: trj74h6@ kuhp.kyoto-u.ac.jp 140 density (P = 0.006, P = 0.046), but not with the optic disc vessel density (P = 0.981, P = 0.856). cpRNFL and peripapillary vessel density showed reduction predominantly in the superior sectors, corresponding to the visual field defect. However, the correlations showed discrepancy of the sectors. Conclusions: The microvascular structures in the peripapillary retina and optic disc were reduced, but the cpRNFL thinning was associated with vessel density only in the peripapillary retina, indicating that the vessel densities in the peripapillary retina and optic disc may be differently affected in the pathological process of NAION. Journal of Neuro-Ophthalmology 2017;37:140-148 doi: 10.1097/WNO.0000000000000470 © 2016 by North American Neuro-Ophthalmology Society N onarteritic anterior ischemic optic neuropathy (NAION) is a major nonglaucomatous optic neuropathy, characterized by a sudden, painless, decrease in vision accompanied with a visual field defect and optic disc edema (1,2). It is generally accepted that optic nerve impairment in NAION is caused by vascular insufficiency in the capillary bed of the optic disc, (3,4) as patients typically have predisposing vascular risk factors (5). However, the exact pathophysiology of NAION remains unclear. Blood is supplied to the optic disc from 3 major sources (6). The retinal arterioles supply blood to the retina, the peripapillary choroidal or short posterior ciliary arteries to the prelaminar tissue, and the centripetal branches of the short posterior ciliary arteries supply the blood to the lamina cribrosa. Studies of the ocular circulation in patients with NAION have employed various imaging modalities including fluorescein and indocyanine green angiography, and laser Doppler imaging (7-9). Results of these studies suggested an impairment of optic disc perfusion in NAION, Hata et al: J Neuro-Ophthalmol 2017; 37: 140-148 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution but the relationship between the impairment of optic disc microcirculation and the neural structure and function were not fully clarified due to limited quantitative property of these imaging modalities. Optical coherence tomography (OCT) angiography using split-spectrum amplitude-decorrelation angiography (SSADA) algorithm is a noninvasive, rapid, and accurate method for quantifying retinal and optic disc blood flow (10-12). For example, in glaucomatous optic neuropathy, OCT angiography has demonstrated reduced disc perfusion and peripapillary retinal perfusion. Moreover, disc flow index and vessel density were closely related to retinal nerve fiber damage and visual field loss in glaucoma (13,14). Quantitative analysis of the perfusion of the optic disc and peripapillary retina using OCT angiography along with visual field analysis in patients with NAION could provide some insight into the association of structural and functional impairments of the optic nerve. This analysis forms the basis of our study. METHODS Patients We performed a hospital-based prospective observational study of patients with NAION and a control group of patients who were evaluated at our hospital for cataract surgery or reasons other than optic nerve or retinal diseases. We obtained approval from by the Institutional Review Board of the Kyoto University Graduate School of Medicine and adhered to the tenets of the Declaration of Helsinki. The nature of the study and the possible risks and benefits of participation were explained to all study candidates. All participants provided written informed consent. Consecutive patients with NAION (inclusion and exclusion criteria are described below) who visited the Neuro-ophthalmology Clinic of Kyoto University Hospital (Kyoto, Japan) between September 2015 and March 2016 were recruited. Healthy control subjects examined at the same hospital were recruited during the same time period. Inclusion Criteria Inclusion criteria for patients with NAION were as follows: 1) age $40 years, 2) acute and painless visual acuity loss or visual field defect, 3) unilateral optic disc swelling on ophthalmoscopy during the acute stage, 4) an intraocular pressure (IOP) of #20 mm Hg, 5) no signs of giant-cell arteritis including elevated erythrocyte sedimentation rate and C-reactive protein levels, and 6) follow-up of at least 6 months after the acute phase. Inclusion criteria for healthy normal eyes included the following: 1) an IOP of #20 mm Hg with no history of increased IOP, and 2) the absence of glaucomatous optic disc appearance (a vertical cup-to-disc ratio $0.7, intraindividual asymmetry Hata et al: J Neuro-Ophthalmol 2017; 37: 140-148 $0.2, or presence of focal thinning, notching, and evidence of glaucomatous visual field loss). Exclusion Criteria Exclusion criteria were as follows: 1) high myopia defined as greater than 26.0 diopters (D); 2) a tilted disc, defined as an index of tilt (ratio of minimum to maximum optic disc diameter) less than 0.75; 3) any other ophthalmic disorder, including corneal opacity, vitreous opacity, diabetic retinopathy, and diseases affecting the optic disc (glaucoma, optic neuritis, uveitis, retinal or choroidal diseases, and trauma); and 4) neurologic diseases that may affect the optic nerve such as multiple sclerosis, Alzheimer disease, and Parkinson disease. Clinical Examinations All subjects underwent a comprehensive ophthalmic assessment, including visual acuity (VA) measurement with a Landolt C chart, IOP measurement using Goldmann applanation tonometry, slit-lamp examination, circumpapillary retinal nerve fiber layer (cpRNFL) measurements using the Spectralis SD OCT scanner (Heidelberg Engineering, Heidelberg, Germany), OCT angiography using the RTVue-XR Avanti scanner (Optovue Inc, Fremont, CA), and axial length measurement with an IOLMaster biometer (Carl Zeiss Meditec, Inc, Dublin, CA). Standard automated perimetry was performed with the Humphrey Visual Field Analyzer (HFA) using Swedish interactive threshold algorithm (SITA) program 30-2 (Carl Zeiss Meditec, Inc). Criteria for a visual field defect (VFD) were defined as the presence of a cluster of 3 or more adjacent nonedge points, all of which were depressed on the pattern deviation plot at a P , 5% level, and one of which was depressed at P , 1% level. Patients with NAION were divided into 2 groups based on the automated perimetry results: mild-to-moderate NAION (visual field mean deviation [MD] better than 212 dB) and severe NAION (visual field MD worse than 212 dB). OCT Angiography and Data Acquisition and Processing OCT angiography images were acquired using an 840-nm superluminescent diode at an A-scan rate of 70,000 scans per second. Each of the acquired optic disc cubes (3 · 3 mm) consisted of 304 clusters of 2 repeat B-scans containing 304 A-scans each. A SSADA algorithm was employed to improve the signal-to-noise ratio by splitting the spectrum to generate multiple repeat OCT frames from the 2 original repeat OCT frames (12). Automatic segmentation of the inner retinal borders was performed using the viewing software. Two types of segmentation algorithms were used. The surface-mode included signals from the inner limiting membrane to the posterior boundary of the nerve fiber layer and the wholedepth mode included all of the signals below the 141 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution vitreo-papillary interface. En face angiogram images of the retinal circulation were obtained from the maximum flow projections from each of the segmented areas. OCT angiography scanning was performed 2 times on the same day and the best result was chosen for analysis. In order to quantify the vessel densities in the peripapillary retina and optic disc, the optic disc boundary was determined based on the scanning laser ophthalmoscopy (SLO) images. The peripapillary region was defined as a 500-mm-wide elliptical annulus extending from the optic disc boundary, using the intrinsic software provided by OptoVue. Sectorial division was performed using the Garway-Heath regionalization (15), and the total and individual areas were used to measure the vessel densities of the peripapillary retina using the software's intensity-based thresholding technique (Fig. 1). Spectral-Domain OCT After pupillary dilation, the optic nerve was imaged using OCT. To measure the peripapillary retinal nerve fiber layer thickness, a 3.46-mm-diameter circular scan, around the optic disc center, was performed. The mean cpRNFL thickness on the OCT image was calculated by averaging 50 circular B-scans and subsequently used for analysis. Sectorial division analysis of cpRNFL thickness was performed using the following analysis. Sectorial Division Analysis of Spectral-Domain OCT Images To eliminate the influences of the difference in disc rotation between the OCT image and OCT angiography image, we performed the following analysis. First, 2 crossing sites of the retinal vessels were marked on the en face image of the OCT and the en face projection image of the surface-mode OCT angiography (Fig. 1). Next, the en face images of the OCT were overlapped with 60% transparency on the en face projection images of the surface-mode OCT angiography. If there was a deviation of the vessel crossing site on the 2 images, we rotated the OCT images around the center of the optic nerve head and matched the vessel crossing sites on the 2 images. Then, 6 quadrant thickness of cpRNFL were calculated based on a new horizontal line according to the Garway-Heath regionalization (120° , nasal # 230°, 230° , inferior nasal # 270°, 270° , inferior temporal FIG. 1. The sectorial division analysis of the optical coherence tomography (OCT) image and OCT angiography image. A. The surface-mode OCT angiography en face projection image at the peripapillary retina with sectorial divisions and (B) with marked crossing sites of the retinal vessels. C. The en face image of the OCT with marked crossing sites of the retinal vessels. D. Overlay shows no deviation of the optic disc torsion between the OCT angiography image and OCT image. 142 Hata et al: J Neuro-Ophthalmol 2017; 37: 140-148 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution # 310°, 310° , temporal # 40°, 40° , superior temporal # 80°, and 80° , superior nasal # 120°) (15). Sectorial division analysis was performed by an investigator (AH) who was masked to the other clinical examination data. Statistical Analysis All values are presented as mean ± SD. Differences in the parameters between the 2 groups were compared using the unpaired t test or Fisher's exact test. Bivariate relationships were examined using the Pearson correlation coefficient. All statistical analyses were performed using SPSS version 19.0.0 statistical software (IBM Japan, Tokyo, Japan). A P value ,0.05 was considered statistically significant. RESULTS Thirty-six eyes, consisting of 17 eyes from 13 patients with NAION and 19 eyes from 14 healthy control subjects, met the inclusion criteria of this study. Among the eyes with NAION, 2 were excluded because of insufficient quality of their OCT angiography images. As a result, 15 eyes of patients with NAION and 19 normal eyes of healthy subjects were included in this study (Table 1). The mean age was 66.4 ± 14.2 years in the NAION group and 61.3 ± 19.1 years in the control group. There were no significant differences between the NAION group and the control group in terms of age, spherical equivalent, sex, IOP, visual acuity (logMar) and axial length, whereas cpRNFL was significantly thinner in the NAION eyes compared to the normal eyes (62.6 ± 14.4 mm vs 102.5 ± 13.7 mm; P , 0.001). The microvascular network of the peripapillary retina and the optic disc was assessed using the surface-mode and whole-depth-mode OCT angiography. In the normal eyes, dense microvascular networks were observed around all discs and no focal capillary dropout was present (Fig. 2). In contrast, in the NAION eyes, focal microvascular reduction around the optic disc at the corresponding side of the visual field defects and diffuse microvascular reduction at the optic disc was apparent in most of the eyes with NAION (Fig. 3). As shown in Table 1, the vessel density of the peripapillary retina was significantly lower in the NAION eyes than in the normal eyes (50.0 ± 4.7 vs 62.0 ± 4.4; P , 0.001). Moreover, the vessel density of the optic disc was also significantly lower in the NA-AION eyes than in the normal eyes (46.6 ± 5.3 vs 56.5 ± 8.5; P , 0.001). In the sectorial analysis, compared to the normal eyes, cpRNFL was significantly thinner in all sectors except the inferior nasal region in the NAION eyes (Fig. 4). The vessel density of the peripapillary retina was lower in all sectors, especially the superior regions, in the NAION compared to the normal eyes. Thus, both cpRNFL thickness and vessel density of the peripapillary retina showed reduction predominantly in the superior sectors, which corresponded to the VFD: among 15 eyes with NAION, 9 eyes showed VFD only inferiorly, 6 eyes showed in both superior and inferior field loss, and no eye showed only superior loss. Compared to the normal eyes, 9 eyes with VFD only inferiorly did not show a reduction of vessel density of the peripapillary retina in the inferior nasal sector (56.9 ± 9.3 vs 64.4 ± 6.4, P = 0.102) and inferior temporal sector (60.8 ± 7.1 vs 65.1 ± 5.2, P = 0.158) corresponding to no VFD superiorly. These same eyes showed significantly lower vessel density of the peripapillary retina in the superior nasal sector (46.9 ± 10.6 vs 61.4 ± 5.1, P , 0.001) and superior temporal sector (47.9 ± 8.9 vs 65.3 ± 6.4, P , 0.001). Divided into 2 groups based on the visual field mean deviation (VF MD) of 212 dB (mild-to-moderate vs severe NAION), the vessel density of the peripapillary retina was significantly lower in the severe group than in the mild-tomoderate group, while that of the optic disc did not significantly differ (Table 2). In the 15 NAION eyes, cpRNFL thickness was significantly associated with the superficial vessel density of the peripapillary retina (b = 0.424, P = 0.046) and VF MD (b = 0.538, P = 0.039), and marginally associated with logMAR VA (b = 0.460, P = 0.084), but not TABLE 1. Demographic and ophthalmic characteristics of the study subjects Age, yr Sex (Male/Female) IOP, mm Hg Visual acuity, logMAR Visual field mean deviation, dB Axial length, mm Spherical equivalent, D cpRNFL, mm Optic disc vessel density on whole-depth mode, % area Peripapillary vessel density on surface-mode, % area Controls, N = 19 NAION, N = 15 P value 61.3 ± 19.1 6/8 12.8 ± 2.9 20.13 ± 0.06 n/a 24.00 ± 1.57 21.80 ± 2.65 102.5 ± 13.7 56.5 ± 8.5 66.4 ± 14.2 8/3 13.8 ± 2.4 0.46 ± 1.38 214.2 ± 7.4 24.03 ± 0.71 21.14 ± 1.72 62.6 ± 14.4 46.6 ± 5.3 0.449 0.227 0.283 0.093 n/a 0.958 0.414 ,0.001 ,0.001 62.0 ± 4.4 50.0 ± 4.7 ,0.001 Data are presented as the mean ± standard deviation. cpRNFL, circumpapillary retinal nerve fiber layer; D, diopter; db, decibel; IOP, intraocular pressure; n/a, not applicable. Hata et al: J Neuro-Ophthalmol 2017; 37: 140-148 143 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 2. Optical coherence tomography (OCT) angiography images of a normal left eye of a 48-year-old woman. A. Color photograph of the optic disc. B. The surface-mode OCT angiography en face projection image shows dense microvascular networks of the peripapillary retina. C. The whole-depth-mode OCT angiography en face projection image shows dense microvascular networks of the optic disc. D. OCT angiography image with cross-sectional angiograms overlying on B-scan images scanned at green line. E. The circumpapillary retinal nerve fiber layer (cpRNFL) shows normal thickness profile. associated with the vessel density of the optic disc (b = 0.051, P = 0.856; Table 3). There were significant correlations between cpRNFL thickness and the vessel density in the superior and inferior regions (Table 4). In the sectorial analysis of the correlations between cpRNFL thickness and vessel density of the peripapillary retina, there were some significant correlations, but most of them showed discrepancy of the sector between cpRNFL thinning and vessel density reduction: cpRNFL thinning correlated with the reduction FIG. 3. Right eye of a 60-year-old woman with nonarteritic anterior ischemic optic neuropathy (NAION). A. Color photograph shows optic atrophy predominantly superiorly. B. The surface-mode optical coherence tomography (OCT) angiography en face projection image reveals focal capillary dropout superiorly in the peripapillary retina. C. The whole-depth-mode OCT angiography en face projection image shows apparent microvascular reduction inside and around the optic disc. D. OCT angiography image with cross-sectional angiograms overlying on B-scan images scanned at green and yellow lines. E. The Humphrey Visual Field Analyzer 30-2 Swedish interactive threshold algorithm standard pattern deviation image demonstrates inferior and mild superior visual field defects (visual field mean deviation, 28.67 dB). F. There is circumpapillary retinal nerve fiber layer (cpRNFL) thinning superiorly. 144 Hata et al: J Neuro-Ophthalmol 2017; 37: 140-148 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 4. A. Comparison of the circumpapillary retinal nerve fiber layer (cpRNFL) thickness between normal subjects and those with nonarteritic anterior ischemic optic neuropathy (NAION). B. Comparison of peripapillary vessel density between normal subjects and those with NAION. **P , 0.01, ***P , 0.001. of the superficial vessel density of the peripapillary retina in the neighboring sector (Table 5). When restricted to 9 cases with VFD, the cpRNFL thickness in the superior temporal sector correlated with the vessel density of the peripapillary retina in the superior nasal sector (b = 0.737, P = 0.023) but not with that in the superior temporal sector TABLE 2. Comparison of diagnostic testing results between mild-to-moderate and severe NAION Visual acuity, logMAR Visual field mean deviation, dB PSD 30-2, dB VFI cpRNFL, mm Optic disc vessel density on whole-depth mode, % area Peripapillary vessel density on surface mode, % area Mild-to-Moderate N=7 Severe N=8 P value 20.02 ± 0.16 27.4 ± 1.7 11.2 ± 4.5 63.4 ± 30.4 73.3 ± 17.8 46.6 ± 3.3 0.87 ± 1.66 220.2 ± 5.2 9.4 ± 6.0 45.6 ± 33.5 56.0 ± 13.7 46.7 ± 6.9 0.172 ,0.001 0.525 0.303 0.064 0.981 53.3 ± 3.4 47.1 ± 3.8 0.006 Data are presented as the mean ± standard deviation. cpRNFL, circumpapillary retinal nerve fiber layer; dB, decibel; NAION, nonarteritic anterior ischemic optic neuropathy; PSD, pattern standard deviation; VFI, visual field index. Hata et al: J Neuro-Ophthalmol 2017; 37: 140-148 145 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 3. Correlation between cpRNFL thickness and other variables in eyes with NAION Variable Visual acuity, logMAR Visual field mean deviation, dB PSD 30-2, dB VFI Optic disc vessel density on whole-depth mode, % area Peripapillary vessel density on surface mode, % area Pearson Correlation Coefficient P value 20.460 0.538 0.084 0.039 0.376 0.090 0.051 0.167 0.748 0.856 0.424 0.046 cpRNFL, circumpapillary retinal nerve fiber layer; dB, decibel; NAION, nonarteritic anterior ischemic optic neuropathy; PSD, pattern standard deviation; VFI, visual field index. (b = 0.268, P = 0.486). The cpRNFL thickness in the superior nasal sector correlated with the vessel density of the peripapillary retina in the inferior nasal sector (b = 0.778, P = 0.014) but not with that in the superior nasal sector (b = 0.548, P = 0.126), although the superior cpRNFL thickness correlated with the superior vessel density of the peripapillary retina (b = 0.723, P = 0.028). DISCUSSION We investigated the OCT angiography images of the peripapillary retina and optic disc in patients with NAION. We found a decrease in the peripapillary vessel density at the corresponding location of the visual field defects that correlated with the severity of cpRNFL thinning. However, the vessel density inside the optic disc was reduced, but did not correlate with the severity of cpRNFL thinning. This may indicate that the vessel density in the peripapillary retina and optic disc are affected differently in the pathological process of NAION. In NAION, currently, there is one case reporting on the dropout of radial peripapillary capillaries (RPC), which is the most superficial of the capillary layers (16). The loss of TABLE 4. Correlation between cpRNFL thickness and peripapillary vessel density in superior and inferior regional analysis in NAION cpRNFL Thickness Peripapillary Vessel Density on Surface-Mode Superior Inferior Superior Inferior 0.518* 0.447 0.229 0.597* Pearson Correlation Coefficients. *P , 0.05. cpRNFL, circumpapillary retinal nerve fiber layer; NAION, nonarteritic anterior ischemic optic neuropathy. 146 microvasculature in the RPCs also was shown in the postmortem eyes with glaucoma in studies using India ink injection (17,18). Using OCT angiography, Liu et al (14) demonstrated that the peripapillary vessel density was reduced in eyes with glaucoma. In the current study, we observed that the peripapillary vessel density was reduced at the corresponding location of the visual field defects in the NAION eyes, which is consistent with the results of the previous study of patients with glaucoma. This fact indicates that the dropout of RPCs is not specific to glaucomatous optic neuropathy or NAION, and that it may be a secondary change accompanying retinal ganglion cell axonal loss. Interestingly, in the sectorial analysis with eliminating influences of the difference in disc rotation between the 2 images, both cpRNFL thinning and RPC dropout occurred in regions corresponding to the VFDs, but the severity of cpRNFL thinning did not coincide with that of RPC dropout in the sector. This also is true when restricted to cases with relatively narrow impairment areas. It may be because there is a sectorial difference of vulnerability between cpRNFL thickness and RPC, or that the optic disc swelling could lead to a further compression of the neighboring capillaries in NAION (19). Further study would be needed to clarify whether the microvascular reduction of the peripapillary retina is only a result of retinal ganglion cell axonal loss or whether it can be accelerated by axonal swelling. We also found that the whole-depth microvasculature of the optic disc decreased in NAION compared to that in normal eyes, but it had no relationship with VFDs nor cpRNFL thinning unlike the peripapillary vessel density. The microvascular reduction in the whole-depth OCT angiography images at the optic disc reflects blood supply to the prelaminar tissue and lamina cribrosa. The precise pathogenesis of NAION is unclear, but it is believed to result from acute ischemia of the prelaminar and laminar portions of the optic nerve head, which is mainly supplied by the short posterior ciliary circulation (20). Fluorescein and indocyanine green angiographic studies of the optic disc in NAION showed complete or sectorial delay or absence of filling of the prelaminar optic disc without impairment of the choroidal circulation (7,8). Our results support that ischemia occurs in the prelaminar and/or laminar portions of the optic nerve head in NAION. On the other hand, optic disc perfusion has no relationship with VFDs nor cpRNFL thinning unlike the peripapillary vessel density in this study. In a previous report on the quantitative analysis of optic disc perfusion using laser Doppler imaging, optic disc Doppler broadenings, which is directly proportional to blood velocity, did not correlate with severity and location of VFDs (9). Because the quantitative evaluation of the whole-depth OCT angiography images of the optic disc reflects both severity and extent (area Hata et al: J Neuro-Ophthalmol 2017; 37: 140-148 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 5. Correlation between cpRNFL thickness and peripapillary vessel density in sectorial analysis in eyes with NAION Peripapillary Vessel Density on Surface Mode N IN IT ST SN T cpRNFL thickness N IN IT ST SN T 0.667* 0.109 0.035 20.175 20.150 20.597* 0.641* 0.453 20.088 20.051 0.251 20.353 0.387 0.552* 0.533* 0.242 0.112 20.095 0.206 0.540* 20.046 0.172 0.598* 0.064 0.470 0.690† 20.112 0.021 0.508 20.227 20.142 0.088 0.757† 0.746‡ 20.414 0.344 Pearson Correlation Coefficients. *P , 0.05. † P , 0.01. ‡ P , 0.001. cpRNFL, circumpapillary retinal nerve fiber layer; N, nasal; IN, inferior nasal; IT inferior temporal; NAION, nonarteritic anterior ischemic optic neuropathy; SN, superior nasal; ST, superior temporal; T, temporal. and depths) of optic nerve head ischemia, further study is needed to confirm the correlation between retinal ganglion cell axonal loss and optic disc perfusion status in a sectorial manner and layer by layer using OCT angiography. There are several limitations to our study, including a relatively small number of patients with NAION, inclusion of bilateral cases, the lack of information on blood pressure, which could affect the intraocular perfusion pressure, and evaluation only during the atrophic phase of NAION. OCT angiography cannot obtain signals in the setting of optic disc swelling. In addition, the OCT angiography modality used has limitations in the analysis of segmented layers, especially the deeper layers, since artifacts due to vascular shadowing or projection onto the underlying tissues are unavoidable (21). In order to exclude the potential influence of such artifacts, we mainly analyzed the microvasculature reduction at the optic disc using the whole-depth-mode images. Further investigations using swept-source OCT or enhanced-depth imaging SD-OCT are required to assess the deeper vessels of the optic nerve head. In conclusion, we observed in patients with NAION a microvascular network reduction in the peripapillary retina corresponding to the defects of the cpRNFL and visual field, in a region-related fashion. Also, we found reduction of optic disc perfusion regardless of the severity of retinal ganglion cell axon impairment. Our results indicate that the use of OCT angiography can potentially lead to further breakthroughs in understanding the pathophysiology of NAION. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: M. Hata, A. Oishi, Y. Muraoka, K. Miyamoto, K. Kawai, and N. Yoshimura; b. Acquisition of data: M. Hata, A. Oishi, Y. Muraoka, K. Miyamoto, K. Kawai, S. Yokota, and M. Fujimoto; c. Analysis and interpretation of data: M. Hata, A. Oishi, Y. Muraoka, K. Miyamoto, K. Kawai, S. Yokota, M. Fujimoto, and M. Miyata. Category 2: a. Drafting the manuscript: M. Hata, A. Oishi, Hata et al: J Neuro-Ophthalmol 2017; 37: 140-148 Y. Muraoka, K. Kawai, S. Yokota, M. Fujimoto, and M. Miyata; b. Revising it for intellectual content: K. Miyamoto and N. Yoshimura. Category 3: a. Final approval of the completed manuscript: M. Hata, A. Oishi, Y. Muraoka, K. Miyamoto, K. Kawai, S. Yokota, M. Fujimoto, M. Miyata, and N. Yoshimura. ACKNOWLEDGMENTS The authors thank Akiko Hirata for her contribution to the reliability analysis of the sectorial thickness of cpRNFL on Spectralis OCT. REFERENCES 1. Johnson LN, Arnold AC. Incidence of nonarteritic and arteritic anterior ischemic optic neuropathy. Population-based study in the state of Missouri and Los Angeles County, California. J Neuroophthalmol. 1994;14:38-44. 2. Hayreh SS, Zimmerman B. Visual field abnormalities in nonarteritic anterior ischemic optic neuropathy: their pattern and prevalence at initial examination. Arch Ophthalmol. 2005;123:1554-1562. 3. Hayreh SS. Anterior ischaemic optic neuropathy. I. Terminology and pathogenesis. Br J Ophthalmol. 1974;58:955-963. 4. Hayreh SS. Anterior ischaemic optic neuropathy. III. Treatment, prophylaxis, and differential diagnosis. Br J Ophthalmol. 1974;58:981-989. 5. Hayreh SS. Management of ischemic optic neuropathies. Indian J Ophthalmol. 2011;59:123-136. 6. Hayreh SS. The blood supply of the optic nerve head and the evaluation of it-myth and reality. Prog Retin Eye Res. 2001;20:563-593. 7. Arnold AC, Hepler RS. Fluorescein angiography in acute nonarteritic anterior ischemic optic neuropathy. Am J Ophthalmol. 1994;117:222-230. 8. Oto S, Yilmaz G, Cakmakci S, Aydin P. Indocyanine green and fluorescein angiography in nonarteritic anterior ischemic optic neuropathy. Retina. 2002;22:187-191. 9. Collignon-Robe NJ, Feke GT, Rizzo JF III. Optic nerve head circulation in nonarteritic anterior ischemic optic neuropathy and optic neuritis. Ophthalmology. 2004;111:1663-1672. 10. Jia Y, Morrison JC, Tokayer J, Tan O, Lombardi L, Baumann B, Lu CD, Choi W, Fujimoto JG, Huang D. Quantitative OCT angiography of optic nerve head blood flow. Biomed Opt Express. 2012;3:3127-3137. 147 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 11. Wang X, Jia Y, Spain R, Potsaid B, Liu JJ, Baumann B, Hornegger J, Fujimoto JG, Wu Q, Huang D. Optical coherence tomography angiography of optic nerve head and parafovea in multiple sclerosis. Br J Ophthalmol. 2014;98:1368-1373. 12. Jia Y, Tan O, Tokayer J, Potsaid B, Wang Y, Liu JJ, Kraus MF, Subhash H, Fujimoto JG, Hornegger J, Huang D. Split-spectrum amplitude-decorrelation angiography with optical coherence tomography. Opt Express. 2012;20:4710-4725. 13. Wang X, Jiang C, Ko T, Kong X, Yu X, Min W, Shi G, Sun X. Correlation between optic disc perfusion and glaucomatous severity in patients with open-angle glaucoma: an optical coherence tomography angiography study. Graefes Arch Clin Exp Ophthalmol. 2015;253:1557-1564. 14. Liu L, Jia Y, Takusagawa HL, Pechauer AD, Edmunds B, Lombardi L, Davis E, Morrison JC, Huang D. Optical coherence tomography angiography of the peripapillary retina in glaucoma. JAMA Ophthalmol. 2015;133:1045- 1052. 15. Garway-Heath DF, Poinoosawmy D, Fitzke FW, Hitchings RA. Mapping the visual field to the optic disc in normal tension glaucoma eyes. Ophthalmology. 2000;107:1809-1815. 148 16. Higashiyama T, Ichiyama Y, Muraki S, Nishida Y, Ohji M. Optical coherence tomography angiography in a patient with optic atrophy after non-arteritic anterior ischaemic optic neuropathy. Neuro-Ophthalmology. 2016;40:146-149. 17. Henkind P. Symposium on glaucoma: joint meeting with the National Society for the Prevention of Blindness. New observations on the radial peripapillary capillaries. Invest Ophthalmol. 1967;6:103-108. 18. Kornzweig AL, Eliasoph I, Feldstein M. Selective atrophy of the radial peripapillary capillaries in chronic glaucoma. Arch Ophthalmol. 1968;80:696-702. 19. Hayreh SS, Zimmerman MB. Non-arteritic anterior ischemic optic neuropathy: role of systemic corticosteroid therapy. Graefes Arch Clin Exp Ophthalmol. 2008;246:1029-1046. 20. Knox DL, Kerrison JB, Green WR. Histopathologic studies of ischemic optic neuropathy. Trans Am Ophthalmol Soc. 2000;98:203-220; discussion 221-2. 21. Jia Y, Wei E, Wang X, Zhang X, Morrison JC, Parikh M, Lombardi LH, Gattey DM, Armour RL, Edmunds B, Kraus MF, Fujimoto JG, Huang D. Optical coherence tomography angiography of optic disc perfusion in glaucoma. Ophthalmology. 2014;121:1322-1332. Hata et al: J Neuro-Ophthalmol 2017; 37: 140-148 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.