Prevalence of Macular Microcystoid Lacunae in Autosomal Dominant Optic Atrophy Assessed With Adaptive Optics

Background: To assess the prevalence of macular microcystoid lacunae in patients with autosomal dominant optic atrophy (ADOA) and its association with visual function and inner retinal morphology. Methods: The study included 140 participants with ADOA, with a mean age of 44 (SD ±19, range 7-82) years. Study participants with a genetically verified sequence variant in the OPA1 gene were examined with best-corrected visual acuity, contrast sensitivity, optical coherence tomography (Spectralis, Heidelberg) and adaptive optics fundus photography (rtx1, Imagine Eyes). Optically empty microcystoid spaces in the ganglion cell layer and inner plexiform layer were mapped by inspection of the 2 sets of images. Data were analyzed with a mixed model adjusted for age and sex with family and individual as random effect. Results: Microcystoid lacunae were present in 32 of 140 participants (23%) including 18 males and 14 females. Microcystoid lacunae were associated with younger age ( P = 0.0503) and a smaller nerve fiber layer volume ( P = 0.035). No association was found between presence of microcystoid lacunae and visual acuity ( P = 0.2), contrast sensitivity ( P = 0.8), axial length ( P = 0.7), or ganglion cell layer volume ( P = 0.2). The analysis showed moderately reduced visual acuity in patients with microcystoid lacunae. Normal and severely impaired visual function were seen only in participants without microcystoid lacunae. Conclusion: In ADOA, macular microcystoid lacunae were found in 23% of the study participants and tended to be present in younger participants with moderate visual acuity reduction and a smaller nerve fiber layer volume. Further studies are needed to investigate whether cavities left by dead ganglion cells are predictors of decrease in visual function.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Prevalence of Macular Microcystoid Lacunae in Autosomal Dominant Optic Atrophy Assessed With Adaptive Optics Christina Eckmann-Hansen, MSc, Toke Bek, MD, DMSc, Birgit Sander, PhD, Karen Grønskov, PhD, Michael Larsen, MD, DMSc Background: To assess the prevalence of macular microcystoid lacunae in patients with autosomal dominant optic atrophy (ADOA) and its association with visual function and inner retinal morphology. Methods: The study included 140 participants with ADOA, with a mean age of 44 (SD ±19, range 7–82) years. Study participants with a genetically verified sequence variant in the OPA1 gene were examined with best-corrected visual acuity, contrast sensitivity, optical coherence tomography (Spectralis, Heidelberg) and adaptive optics fundus photography (rtx1, Imagine Eyes). Optically empty microcystoid spaces in the ganglion cell layer and inner plexiform layer were mapped by inspection of the 2 sets of images. Data were analyzed with a mixed model adjusted for age and sex with family and individual as random effect. Results: Microcystoid lacunae were present in 32 of 140 participants (23%) including 18 males and 14 females. Microcystoid lacunae were associated with younger age (P = 0.0503) and a smaller nerve fiber layer volume (P = 0.035). No association was found between presence of microcystoid lacunae and visual acuity (P = 0.2), contrast sensitivity (P = 0.8), axial length (P = 0.7), or ganglion cell layer volume (P = 0.2). The analysis showed moderately reduced visual acuity in patients with microcystoid lacunae. Normal and Department of Ophthalmology, Rigshospitalet, Glostrup (CE-H, BS, ML); Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen (CE-H, ML); Department of Ophthalmology, Aarhus University Hospital, Aarhus (TB); and Department of Genetics, Rigshospitalet, Copenhagen (KG) Denmark. The study was funded by the Synoptik Foundation, Fight for Sight Denmark, and the Danish Eye Research Foundation. The study was also supported by the European Union, Horizon 2020, EU.2.1.1., project ID 780989 (MERLIN). M. Larsen serves as a consultant for Stoke Therapeutics. The remaining authors report no conflicts of interest. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www. jneuro-ophthalmology.com). Address correspondence to Christina Eckmann-Hansen, MSc, Department of Ophthalmology, Center for Research, Rigshospitalet Valdemar Hansens Vej 1-23, 2600 Glostrup, Denmark; E-mail: [e-mail]. 328 severely impaired visual function were seen only in participants without microcystoid lacunae. Conclusion: In ADOA, macular microcystoid lacunae were found in 23% of the study participants and tended to be present in younger participants with moderate visual acuity reduction and a smaller nerve fiber layer volume. Further studies are needed to investigate whether cavities left by dead ganglion cells are predictors of decrease in visual function. Journal of Neuro-Ophthalmology 2022;42:328–333 doi: 10.1097/WNO.0000000000001592 © 2022 by North American Neuro-Ophthalmology Society A utosomal dominant optic atrophy (ADOA) is characterized by bilateral visual acuity loss with a centrocecal scotoma, temporal optic disc pallor, attenuation of the retinal ganglion cell and nerve fiber layers, and color vision anomaly—notably tritanopia (1). Penetrance is approximately 90%, and expressivity is highly variable within and between families (2). Optically hyporeflective microcystoid lacunae have been demonstrated in the inner nuclear layer of the retina by optical coherence tomography (OCT) in ADOA, other forms of optic atrophy, and neurodegenerative diseases such as multiple sclerosis, glaucoma, optic disc drusen, and ischemic optic neuropathy (3,4,20). Adaptive optics imaging of the retina in ADOA has shown sharply outlined optically empty hyporeflective oval or round cavities that appear like oil droplets within the retinal matrix (5). Given that there is no evidence of an epithelial lining or expansion of the retina and because they appear in the context of atrophy, the term microcystoid lacunae seems to be a more precise description than microcystoid edema (6,7). The pathogenesis is obviously related to neuronal cell loss, but the fine structure of the retina in ADOA, including the involvement of glial cells, the role of retrograde transsynaptic degeneration, etc. are unknown (6,8). There is no evidence of vascular leakage on fluorescein angiography (9) Eckmann-Hansen et al: J Neuro-Ophthalmol 2022; 42: 328-333 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution and it has been proposed that the formation of lacunae could be the result of Müller cell dysfunction (8). The purpose of this study was to assess the prevalence and the morphological appearance of macular microcystoid lacunae in ADOA and the relation to visual function and retinal atrophy using adaptive optics fundus photography and OCT. METHODS This observational cross-sectional study investigated the clinical phenotype of 145 participants with a documented OPA1 variant associated with ADOA. Clinical details of the participants have previously been described (10). Five participants were excluded from the present analysis due to missing adaptive optics recordings of acceptable quality, because of insufficient, poor fixation. The analysis included 140 participants, 71 males and 69 females, with a mean age of 44 (SD ±19, range 7–82) years, from 47 families. Participants were identified and invited from a national ophthalmic register. Written, informed consent was obtained from adults and adolescents aged 15 years or older and from parents or legal guardians if the participants were younger than age 15 years. The study was approved by the local medical ethics committee, the Danish Data Protection Agency, and the Danish Patient Safety Authority, and adhered to the tenets of the Declaration of Helsinki. The participants underwent clinical examination at Rigshospitalet in Copenhagen (88 participants) or at Aarhus University Hospital (52 participants). All participants were examined by the same investigator (C.E.-H.). Clinical examinations included full refractioning and bestcorrected visual acuity assessment according to the Early Treatment Diabetic Retinopathy Study (ETDRS) protocol (11), contrast sensitivity testing (Pelli-Robson), biometry (IOL Master 500, software version 7.1.2.042; or IOL master 700, software version 1.50, both Carl Zeiss Meditec, La Jolla, CA), fundus photography (TRC-50DX, IMAGEnet i-base version 3.23.0, Topcon, Tokyo, Japan), spectral domain OCT (HRA + OCT Spectralis OCT2, HRA2 software version 6.12.3.0; or HRA + OCT Spectralis OCT1, HRA2 version 6.9.4.0, software version 6.9a, both Heidelberg Engineering, Heidelberg, Germany), microperimetry (MAIA, Centervue, Padova, Italy, software version 2.5.1), and flood-illumination adaptive optics fundus photography with a resolution of 2 mm and a wavelength of 850 nm (rtx1e, AO Image version 3.3; or rtx1, AO Image version 3.3, Imagine Eyes, Orsay, France). Adaptive optics fundus imaging was made according to a standard protocol using images spanning 4 · 4° (corresponding to 1.3 · 1.3 mm) of the fundus. Images of the photoreceptor layer were centered on the fovea and 2° and 4° nasal of the foveal center, as well as 2 images centered 4° inferiorly and 4° temporal and nasal of the foveal center. The layers of the retina in front of the photoreceptors were Eckmann-Hansen et al: J Neuro-Ophthalmol 2022; 42: 328-333 examined by live imaging to identify microcystoid lacunae, if present, which can be found at different distances from the photoreceptor layer. Because microcystoid lacunae are located mainly in the nasal, superior, and inferior perimacula (4,5), a broader search in live imaging mode and still image recording was made in these locations, using adjustment of focus depth and site of pupil entry as per operator discretion to identify clusters of lacunae. Analysis of nerve fiber layer and ganglion cell layer volumes was made using OCT data analysis software (Heidelberg Eye Explorer, version 1.10.4.0), for segmentation and calculation of averages within the fields of the fovea-centered ETDRS grid. Source data consisted of a 20 · 30° fovea-centered block of 61 parallel horizontal, posterior B-scans with real-time averaging of 25 B-scans per line. Automated segmentation was fully reviewed by investigator CEH and found to be acceptable, except in one case, where manual correction was applied. One eye in one participant was excluded from analysis because a prominent epiretinal membrane was found to be associated with pathological thickening and deformation of the retina that made both automated and manual segmentation of the inner layers unreliable. In addition to the protocolized OCT scans, supplementary ad hoc scans were made as per investigator discretion. Statistical analysis was conducted using the statistical software RStudio version 1.2.5001. All descriptive measures were reported as an average of data for the 2 eyes, as ADOA is a bilateral disease that is expected to affect both eyes equally (12). In 3 cases of unilateral reduction of vision caused by amblyopia or other unilateral ocular abnormality, the visual acuity of the worseseeing eye was excluded. Data in the 2 groups were compared using Student unpaired t test or x2-test. Mixed-model analysis was applied using the nlme package (R Core Team 2020, version 3.1–150). All mixed-model analyses included data for both eyes and were corrected for age and sex and for family and individual as random effects. P-values below 0.05 were considered statistically significant. RESULTS Microcystoid retinal lacunae were found by adaptive optics imaging in 32 (23%, 18 males and 14 females) of 140 participants. Participants with lacunae were younger, on average, than participants without microcystoid lacunae (P = 0.054), but the overlap was considerable (see Supplemental Digital Content 1, Figure, http://links.lww.com/ WNO/A585, which illustrates the association between age and visual acuity in the 2 groups). Nerve fiber layer volume was significantly smaller in participants with microcystoid lacunae (P = 0.006), whereas there was no difference in ganglion cell layer volume between the 2 groups (P = 0.6; Table 1). 329 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 1. Characteristics of participants with and without microcystoid lacunae Yes (n = 32) No (n = 108) Microcystoid lacunae Mean (SD)/n Range Mean (SD)/n Range P value Age (years)* Sex (male/female)† BCVA (ETDRS letters)* Contrast sensitivity (log CS)* Axial length (mm)* Nerve fiber layer volume (mm3)* Ganglion cell layer volume (mm3)* 38 (±18) 18/14 55 (±14) 1.25 (±0.2) 24.5 (±1.3) 0.52 (±0.07) 0.6 (±0.07) 8–76 N/A 24–82 0.55–1.65 22.24–27.48 0.41–0.67 0.43–0.75 46 (±20) 53/55 58 (±22) 1.23 (±0.3) 24.3 (±1.6) 0.57 (±0.1) 0.6 (±0.2) 7–82 N/A 3–99 0.1–1.65 20.41–31.05 0.4–1 0.37–1.03 0.054 0.6 0.4 0.6 0.5 0.006 0.6 *Student T test. † 2 x test. BCVA, best-corrected visual acuity. The lesions were present on adaptive optics, but not on OCT in 11 of the 32 participants (34%). No lacunae were found on OCT without them also being visible on adaptive optics fundus imaging. Best-corrected visual acuity (BCVA) was numerically 3 ETDRS letters lower in participants with lacunae than in participants without lacunae (P . 0.5) but more notably, the distribution was irregular, with participants with lacunae having predominantly midrange BCVA (Fig. 1). Contrast sensitivity was numerically 0.02 log units better in the patients with lacunae (P . 0.5, see Supplemental Digital Content 2, Figure, http://links.lww.com/WNO/A586, which illustrates the distribution of the contrast sensitivity in the 2 groups). There was no difference between the sexes. The genotypes of participants with and without microcysts can be seen in Supplemental Digital Content 3 (see Table, http://links.lww.com/WNO/A587), which lists the included genotypes in the study. If variants are divided into 3 groups: variants expected to cause a truncation of OPA1, variants expected to cause a defect in RNA splicing, and others (missense, in-frame deletions, and extension), there is a clear tendency that there are more patients with splice variants in the group with microcystoid lacunae, whereas truncating variants are more frequent in the group without microcystoid lacunae. Given that the adaptive optics camera has a high numerical aperture, the lacunae are visible only at a fraction of the depth range of camera focus (see Supplemental Digital Content 4, Figure, http://links.lww.com/WNO/A588, which demonstrates the visibility of the microcystoid lacunae at 2 different depths of focus). The angle of the light entering the pupil during imaging is important when detecting microcystoid lacunae with adaptive optics. A difference in optical reflectivity is seen when moving the instrument corresponding to different entries through the pupil (see Supplemental Digital Content 5, Figure, http:// links.lww.com/WNO/A589, which demonstrates the visibility of the microcystoid lacunae when entering light through 2 different angles through the pupil). Mixed-model analysis showed that the presence of microcystoid lacunae in patients with ADOA was significantly associated with younger age (P = 0.05) and a lower nerve fiber layer volume (P = 0.036), but not with BCVA (P = 0.2), contrast sensitivity (P = 0.8), axial length (P = 0.7), or ganglion cell layer volume (P = 0.2; Table 2). A single participant who was followed for 2 years presented with microcystoid lacunae and BCVA 20/50 at the age of 60 years and continued to have lacunae for 1 year, whereas very few lacunae were found after 2 years. Meanwhile, BCVA decreased from 20/50 to 20/63 (Fig. 2). An ancillary observation made during live observation of OCT scans, as they were recorded, could be seen to gradually gain in smoothness on the monitoring screen as they were averaged in real time. It occasionally showed that lacunae could be seen on the initial granulated images, only to disappear or lose contrast as additional scans were averaged (Fig. 3). DISCUSSION FIG. 1. Visual acuity in participants with (filled bars) and without (open bars) microcystoid lacunae. 330 Microcystoid lacunae, a form of lacunar degeneration of the inner nuclear layer of the retina, was found in 23% of participants with ADOA who had marginally thinner nerve Eckmann-Hansen et al: J Neuro-Ophthalmol 2022; 42: 328-333 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 2. Association of microcystoid lacunae and descriptive parameters P Estimate (95% CI) BCVA (ETDRS letters) Contrast sensitivity (log CS) AL (mm) Nerve fiber layer volume (mm3) Ganglion cell layer volume (mm3) 24.8 20.02 0.12 20.05 20.03 (213 to 3.4) (20.1 to 0.09) (20.5 to 0.7) (20.1 to 20.004) (20.08 to 0.02) 0.2 0.8 0.7 0.036 0.2 All analyses were conducted as mixed model with family and individual as random effects, corrected for age and sex. BCVA, best-corrected visual acuity. fiber layers, were marginally younger, and had midrange visual acuity compared with participants without microcystoid lacunae. Underlying these seemingly unremarkable findings, however, were 2 different distributions of characteristics, best exemplified by that of BCVA, where participants with lacunae were clustered in the midrange, whereas participants without lacunae were found predominantly at the top and the bottom of the BCVA range. Previous reports of macular microcystoid lacunae prevalence in ADOA have ranged from 4% to 75% (4,13). The prevalence of 23% found in this study is markedly higher than the 4% found in the study by Rönnbäck et al (7) and the 5.2% found by Carbonelli et al (13), both by B-scan or block B-scan OCT, but markedly lower than the prevalence of 75% found by Wolff et al using en face OCT covering 5.79 · 5.79 mm (20 · 20°) of the macula (4). This study, having shown that lacunae are better seen with adaptive optics imaging covering 1.3 · 1.3 mm (4 · 4°) than with B-scan OCT, suggests that projection and fundus area coverage may explain some of the differences between study prevalence findings. There are no obvious explanation why splice variants are more frequent in the group of patients with microcystoid lacunae while truncating variants are more frequent in the group without microcystoid lacunae, but often splice variants cause a mixture of mis-spliced gene product together with wild-type spliced gene product, whereas the truncating variants in this study are all predicted to undergo nonsense-mediated decay. The relevance of studying microcystoid lacunae is underlined by findings of prevalence of 4.7%–40% in multiple sclerosis (3,14) and 3%–6% in primary open angle glaucoma (15,16). This study confirmed the previous finding of thin retinal nerve fiber layers in ADOA with microcystoid lacunae (4), whereas no abnormality was found in ganglion cell layer volume, in contrast to an earlier study that also found a deficit in ganglion cell–inner plexiform layer thickness (13). Compared with reported microcystoid lacunae in other optic neuropathies, the microcystoid lacunae found in ADOA appears morphologically similar regarding location and appearance of size and shape. In this study, presence of microcystoid lacunae was not associated with visual acuity. Gelfand et al (3) assessed this association in multiple sclerosis and found that eyes with no microcystoid lacunae had a significant better visual acuity. This might be due to multiple sclerosis being widely dispersed in time and location, whereas ADOA is much more uniform and presumably with a steady progression rate. Our demonstration of microcystoid lacunae being associated with intermediate ADOA is consistent with the demonstrations by Abegg et al (17) of a decrease in microcystoid lacunae in a patient over a 7-month period and the de novo appearance of lacunae in a patient 4 months after diagnosis of optic neuropathy, as well as our own observation of a case who was followed over 2 years. So far, there has been no report of coming back once they have disappeared. Altogether, these FIG. 2. Reduction in extent of microcystoid lacunae (dotted lines) in autosomal dominant optic atrophy over 2 years. Eckmann-Hansen et al: J Neuro-Ophthalmol 2022; 42: 328-333 331 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution predominantly in participants with midrange BCVA and lower nerve fiber layer thickness. It found that microcystoid lacunae were better detected with adaptive optics than with OCT as the latter was used in this study. The demonstration that averaging of OCT scans can blur the imaging of microcystoid lacunae may help inform clinical practice and guide data analysis methodology toward giving higher priority to lateral resolution. This study also suggests that systematic mapping of microcystoid lacunae may help detect the effect of interventions aimed at rescuing retinal cells in ADOA. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: C. Eckmann-Hansen and M. Larsen; b. Acquisition of data: C. Eckmann-Hansen; c. Analysis and interpretation of data: C. Eckmann-Hansen. Category 2: a. Drafting the manuscript: C. Eckmann-Hansen; b. Revising the manuscript for intellectual content: C. Eckmann-Hansen. Category 3: a. Final approval of the completed manuscript: C. Eckmann-Hansen, T. Bek, B. Sander, K. Grønskov, and M. Larsen. REFERENCES FIG. 3. Disappearance of microcystoid lacunae with scan averaging of OCT B-scans, progressing from an average of 2 scans (A), over 5 scans (B), 10 scans (C), and to 20 scans (D). data suggest that the development of microcystoid lacunae may be a transient, midphase phenomenon in ADOA and that it seems to accompany a period of demonstrable visual loss. Microcystoid macular edema is therefore of fundamental interest for the planning of intervention trials in ADOA. Control of the depth of focus is crucial in adaptive optics imaging of microcystoid lacunae, as is the angle of observation. The change in optical reflectivity has previously been described in patients with macular edema due to acute macular neuroretinopathy (18). Split detection in adaptive optics scanning laser ophthalmoscopy (AOSLO) might be a beneficial method of examining microcystoid lacunae in the future, as living ganglion cells have previously been imaged with this technique (19). This study is limited by its size, its cross-sectional design, and the clustering into families and few participants per variant group. Nevertheless, this study is larger than comparable studies and combining adaptive optics imaging with OCT. In conclusion, microcystoid lacunae was found in 32 of 140 (23%) of our study participants with ADOA and 332 1. Lenaers G, Hamel C, Delettre C, Amati-Bonneau P, Procaccio V, Bonneau D, Reynier P, Milea D. Dominant optic atrophy. Orphanet J Rare Dis. 2012;7:46. 2. Cohn AC, Toomes C, Potter C, Towns KV, Hewitt AW, Inglehearn CF, Craig JE, Mackey DA. Autosomal dominant optic atrophy: penetrance and expressivity in patients with OPA1 mutations. Am J Ophthalmol. 2007;143:656–662. 3. Gelfand JM, Nolan R, Schwartz DM, Graves J, Green AJ. Microcystic macular oedema in multiple sclerosis is associated with disease severity. Brain. 2012;135:1786–1793. 4. Wolff B, Azar G, Vasseur V, Sahel JA, Vignal C, Mauget-Faÿsse M. Microcystic changes in the retinal internal nuclear layer associated with optic atrophy: a prospective study. J Ophthalmol. 2014;2014:395189. 5. Gocho K, Kikuchi S, Kabuto T, Kameya S, Shinoda K, Mizota A, Yamaki K, Takahashi H. High-resolution en face images of microcystic macular edema in patients with autosomal dominant optic atrophy. Biomed Res Int. 2013;2013:1. 6. Sigler EJ. Microcysts in the inner nuclear layer, a nonspecific SD-OCT sign of cystoid macular edema. Invest Ophthalmol Vis Sci. 2014;55:3282–3284. 7. Rönnbäck C, Milea D, Larsen M. Author reply. Ophthalmology. 2014;121:30–31. 8. Mahroo OA, Shalchi Z, Kisimbi J, Sanyiwa AJ, Mohamed MD, Plant GT. Re: Abegg et al.: microcystic macular edema: retrograde maculopathy caused by optic neuropathy (Ophthalmology 2014;121:142–9). Ophthalmology. 2014;121:3212. 9. Wolff B, Basdekidou C, Vasseur V, Mauget-Faÿsse M, Sahel JA, Vignal C. Retinal inner nuclear layer microcystic changes in optic nerve atrophy: a novel spectral-domain oct finding. Retina. 2013;33:2133–2138. 10. Eckmann-Hansen C, Bek T, Sander B, Larsen M. Vision-related quality of life and visual ability in patients with autosomal dominant optic atrophy. Acta Ophthalmol. 2022 February 10 (doi: 10.1111/aos.15102). 11. Ferris FL III, Kassoff A, Bresnick GH, Bailey I. New visual acuity charts for clinical research. Am J Ophthalmol. 1982;94:91–96. 12. Yu-Wai-Man P, Griffiths PG, Burke A, Sellar PW, Clarke MP, Gnanaraj L, Ah-Kine D, Hudson G, Czermin B, Taylor RW, Horvath R, Chinnery PF. The prevalence and natural history of Eckmann-Hansen et al: J Neuro-Ophthalmol 2022; 42: 328-333 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 13. 14. 15. 16. dominant optic atrophy due to OPA1 mutations. Ophthalmology. 2010;117:1538–1546. Carbonelli M, La Morgia C, Savini G, Cascavilla ML, Borrelli E, Chicani F, do V F Ramos C, Salomao SR, Parisi V, Sebag J, Bandello F, Sadun AA, Carelli V, Barboni P. Macular microcysts in mitochondrial optic neuropathies: prevalence and retinal layer thickness measurements. PLoS One. 2015;10:e0127906. Green AJ, McQuaid S, Hauser SL, Allen IV, Lyness R Ocular pathology in multiple sclerosis: retinal atrophy and inflammation irrespective of disease duration. Brain. 2010;133:1591–1601. Brazerol J, Iliev ME, Höhn R, Fränkl S, Grabe H, Abegg M. Retrograde maculopathy in patients with glaucoma. J Glaucoma. 2017;26:423–429. Hasegawa T, Akagi T, Yoshikawa M, Suda K, Yamada H, Kimura Y, Nakanishi H, Miyake M, Unoki N, Ikeda HO, Yoshimura N. Microcystic inner nuclear layer changes and Eckmann-Hansen et al: J Neuro-Ophthalmol 2022; 42: 328-333 17. 18. 19. 20. retinal nerve fiber layer defects in eyes with glaucoma. PLoS One. 2015;10:1–17. Abegg M, Dysli M, Wolf S, Kowal J, Dufour P, Zinkernagel M. Microcystic macular edema: retrograde maculopathy caused by optic neuropathy. Ophthalmology. 2014;121:142–149. Bottin C, Grieve K, Rossant F, Pedinielli A, Mrejen S, Paques M. Directional variability of fundus reflectance in acute macular neuroretinopathy: evidence for a contribution of the stiles-crawford effect. Retin Cases Brief Rep. 2018;12(suppl 1):S19–S24. Rossi EA, Granger CE, Sharma R, Yang Q, Saito K, Schwarz C, Walters S, Nozato K, Zhang J, Kawakami T, Fischer W, Latchney LR, Hunter JJ, Chung MM, Williams DR. Imaging individual neurons in the retinal ganglion cell layer of the living eye. Proc Natl Acad Sci U S A. 2017;114:586–591. Kisimbi J, Shalchi Z, Mahroo OA, Mhina C, Sanyiwa AJ, Mabey D, Mohamed M, Plant GT. Macular spectral domain optical coherence tomography findings in Tanzanian endemic optic neuropathy. Brain. 2013;136(11):3418–3426. 333 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.