The pathogenesis of nonarteritic anterior ischemic optic neuropathy (NAION) remains unknown, but is likely, in part, due to circulatory insufficiency within the optic nerve head (ONH) (1). The short posterior ciliary arteries are the main source of blood supply to the ONH, and they also supply peripapillary choroid which nourishes the prelaminar region of the ONH (2–4). Therefore, measurement of the peripapillary choroidal thickness (pCT) could be helpful to understand the pathophysiology of NAION.
Before enhanced depth imaging (EDI) technology was developed, no imaging modality could gather accurate in vivo measurements of CT. Results using this technique to evaluate pCT (5–7) and macular CT (8,9) in NAION eyes have been controversial. Some authors concluded that pCT (5,7) and macular CT (8) were thicker in NAION eyes and unaffected fellow eyes than control eyes. By contrast, authors found that eyes affected by NAION and contralateral unaffected eyes showed significantly thinner pCT (6) and subfoveal CT (8) compared with disease-free control eyes. In all of these studies, CT data were obtained by manual segmentation.
A new generation of high-penetration optical coherence tomography (OCT) devices, named as swept-source OCT (SS-OCT), has the potential to improve the assessment of the choroid and incorporates automated choroidal segmentation software which may be useful in reducing potential intraoperator and interoperator variability and allowing highly reliable, faster, and more objective evaluation of choroid (10,11). There are reports assessing pCT by SS-OCT in glaucoma, myopia, and healthy eyes (12–14), but not in NAION eyes.
Smaller optic disc area has been considered a predisposing risk factor for NAION, but data regarding this issue are also conflicting. Some authors found that the disc area was significantly smaller in the eyes of patients with NAION than in control eyes (5,15,16), while others did not confirm this finding (7,17). The Spectralis OCT Glaucoma Module Premium (GMP) Edition provides a new, objective method of ONH analysis using Bruch's membrane opening (BMO) as the anatomical border of the rim which represents a measurable aperture at the level of the ONH through which retinal ganglion cell axons can pass (18,19).
The aim of our study was to evaluate and compare peripapillary and macular CT using automated choroidal segmentation software as well to quantify and compare Bruch membrane opening-minimum rim width (BMO-MRW), retinal nerve fiber layer (RNFL) thickness, and optic disc area among patients with NAION, the contralateral unaffected eyes, and healthy control eyes.
This observational cross-sectional study analyzed a group of patients with NAION evaluated in 2 neuro-ophthalmology units from January 2013 to March 2016 and a group of healthy subjects' age matched (control group). The study followed the principles of the Declaration of Helsinki and informed consent for the research was obtained from all participants.
Patients who experienced NAION at least 6 months before the beginning of the study were considered for inclusion. Diagnosis of NAION was based on sudden loss of visual acuity (VA); relative afferent pupillary defect; optic disc edema on fundus ophthalmoscopy at onset; visual field defects consistent with NAION; erythrocyte sedimentation rate and C-reactive protein levels with normal values, with no signs or symptoms suggestive of giant cell arteritis; and resolution of disc edema in 2 months. Exclusion criteria were a refractive error greater than 5.0 diopters (D) of spherical equivalent or 3.0 D of astigmatism in either eye, media opacities that would preclude OCT scanning, glaucoma, coexistence of ophthalmic or neurologic disease, or other retinal pathology, or previous ophthalmic surgery (other than uneventful cataract extraction). SS-OCT scans with a scan image quality < 35 were excluded.
All participants underwent an ophthalmologic evaluation including assessment of best-corrected VA, pupillary testing, anterior and posterior segment biomicroscopy, and tonometry (Goldmann tonometer; Haag-Streit AG, Koeniz, Switzerland). Refractometry was performed using an automated refractometer (Nidek ARK-1; Nidek Co. LTD., Aichi, Japan) and axial length was measured using IOL Master (Carl Zeiss Meditec, Dublin, CA). Perimetry was performed with the standard Swedish Interactive Thresholding Algorithm (SITA) using the 24-2 pattern on the Humphrey Field Analyzer (Carl Zeiss Meditec, Dublin, CA).
Optical Coherence Tomography Imaging
All scans were performed by a single operator. The choroidal layer was imaged by SS-OCT (DRI Triton, Topcon, Japan) and automatically segmented. At least 2 scans were taken, and the image with the best quality was chosen.
The pCT was assessed from a circumferential 3.4-mm diameter section centered at the center of the optic nerve disc. CT was measured from the hyperreflective outer border of the retinal pigment epithelium to the choroidal–scleral interface at the fovea and 1.7 mm superior, temporal, inferior, and nasal to the ONH center (Fig. 1). The mean pCT from these 4 measurements was calculated. Macular CT was plotted with ETDRS grid that included 9 fields separated from the 3 concentric circles (diameter: 1, 3 and 6 mm) by the ±45° line. This was positioned in relation to the center of the fovea.
All enrolled eyes were imaged with the new GMP Edition provided by Spectralis 6.0c version (Heidelberg Engineering GmBH, Heidelberg, Germany) that include 24 radial and 3 circular scans. The mean corneal radius (C-curve value) affects the absolute measurements; therefore, it was introduced for all participants in the eye data box before starting any scanning. The examination ring was automatically placed using 2 fixed anatomical landmarks: the center of the fovea and the center of the BMO, it then creates a fovea–BMO center axis. The BMO-MRW parameter quantified neuroretinal rim tissue perpendicular to the orientation of the axons and takes into account the varying trajectory of nerve fibers entering the ONH at all points of measurement. RNFL thickness measurements of each individual eye were normalized for anatomic orientation of the fovea-to-BMO center axis to ensure accurate and consistent positioning of the RNFL thickness profile across eyes. We registered the figures provided by the inner circle scan (3.5 mm). After image acquisition, the BMO segmentation was reviewed and confirmed by 1 trained examiner (GR). All acquired spectral-domain OCT (SD-OCT) datasets had a quality score (Q) above 25.
Statistical analysis was performed using SPSS software version 17 (SPSS, Inc, Chicago, IL). The Shapiro–Wilk test was used to determine the normal distribution of the data. To assess the intrasession repeatability using the automated choroidal segmentation software of SS-OCT, images from the same eye of 15 subjects were obtained 3 times at an interval of 5 minutes. Three images from the same eyes were each measured by segmentation software to assess the repeatability of pCT and macular measurements and the intraclass correlation coefficients (ICCs) were calculated. X2, analysis of variance (ANOVA) (post hoc comparisons were performed using Bonferroni test), and Student t test, were used to detect the significance of any difference between the groups. The association between the CT and age, axial length, RNFL, and BMO-MRW were analyzed using Pearson correlation test. A P < 0.05 was considered statistically significant.
Among 34 patients with NAION initially enrolled, 26 were eventually included in the study. Eight patients were excluded (age-related macular degeneration n = 5; poor choroidal segmentation n = 3). Three of 26 patients had bilateral NAION. Two unaffected fellow eyes had to be excluded (peripapillary choroidal nevus n = 1; poor choroidal segmentation n = 1). Therefore, the data of 29 NAION eyes, 21 unaffected fellow eyes, and 29 healthy eyes were analyzed. The mean time lapsed from the acute onset of NAION was 44.3 months (SD: 41.50; range: 10–146).
Table E1 provides demographic and clinical characteristics of both groups (See Supplementary Digital Content 1, Table E1, http://links.lww.com/WNO/A245). No statistically significant differences were observed between groups in terms of sex, axial length, spherical equivalent, intraocular pressure (IOP), and eye. Visual field defects were classified according to Papchenko et al (20) as inferior altitudinal defects (7 eyes, 24.1%), superior altitudinal defects (1 eye, 3.4%), and diffuse defects (22 eyes, 72.41%).
The mean average pCT at all 4 locations in the 3 groups are shown in Table E2 (See Supplementary Digital Content 2, Table E2, http://links.lww.com/WNO/A246). Mean pCT in the NAION eyes, unaffected fellow eyes, and the control group was 130.51 ± 72 μm 149.61 ± 75.74 μm, and 103.66 ± 36.75 μm, respectively (ANOVA, P = 0.04).
The average pCT and all regional values were significantly greater in the unaffected NAION eyes when compared with the control eyes (P ≤ 0.02). Although pCTs in the NAION eyes were thinner compared with unaffected fellow eyes, and thicker compared with control group, differences were not significant. Peripapillary choroid at the superior quadrant was significantly thicker compared with the inferior quadrant in the NAION eyes, fellow eyes, and control eyes (146.3 ± 82.8 μm vs 111.5 ± 70.6 μm, 166.3 ± 82.7 μm vs 123.9 ± 74.6 μm, and 119.0 ± 42.7 μm vs 82.9 ± 32.8 μm, respectively) (P = 0.00). This asymmetry pattern was also present in the 7 eyes having an inferior altitudinal visual field defect (150.5 ± 67.98 μm vs 105.0 ± 43.03 μm, P = 0,00) and in the eyes showing diffuse visual field defects (n = 21) (147.5 ± 89.95 μm vs 115.0 ± 79.71 μm, P = 0.00).
The macular CT values in the 3 groups are shown in Table E3 (See Supplementary Digital Content 3, Table E3, http://links.lww.com/WNO/A247). We found significant differences among groups in all measurements (ANOVA, P ≤ 0.03) except for superior and temporal sectors (ANOVA, P ≥ 0.06). The mean macular CT and all regional values were significantly greater in the unaffected NAION eyes when compared with the control eyes (P ≤ 0.04) except for the superior sectors (P ≥ 0.09). All macular CT values were greater in the NAION eyes in comparison with the control group but difference was only significant in foveal CT, mean macular CT, and inner inferior and inner nasal sectors (P ≤ 0.05). Differences in CT measurements between NAION and unaffected fellow eyes were not significant.
A significant inverse association was found between the mean patient age and mean macular CT (−0.354, P = 0.01), but not between the mean age and mean pCT (−0.196, P = 0.08). A significant association was found between BMO disc area and mean macular thickness (0.24, P = 0.03). We did not found any association between pCT and macular CT with axial length (−0.063, P = 0.58; 0.22, P = 0.84, respectively); average RNFL thickness (−0.04, P = 0.729; −0.013, P = 0.913, respectively); or BMO-MRW thickness (−0.194, P = 0.09; −0.114, P = 0.32 respectively). No correlation was found between IOP and pCT (0.091, P = 0.40) and macular CT (−0.049, P = 0.653). No correlation was found between the mean time lapsed from the onset and pCT and macular CT (0.25, P = 0.31; 0.10, P = 0.67, respectively).
ICCs for pCT at the 4 quadrants and for macular CT at 9 fields ICC were excellent (≥0.98). No difference in the optic disc size yield by GMP between patients with NAION, unaffected fellow eyes, and the control group was found (1.73 ± 0.34, 1.78 ± 0.38, 1.74 ± 0.30 mm2 respectively) (ANOVA, P = 0.84).
Average and all sectorial RNFL and BMO-MRW measurements are shown in Table E4 (See Supplementary Digital Content 4, Table E4, http://links.lww.com/WNO/A248). All values were significantly thinned in NAION eyes compared with the unaffected fellow eyes and control eyes (P ≤ 0.00). Average and sectorial BMO-MRW were significantly thicker in contralateral unaffected eyes than in control group (P ≤ 0.02) except for the nasal sector (P = 0.09).
To the best of our knowledge, this is the first study using automated choroidal segmentation by Triton SS-OCT to measure CT in NAION eyes and unaffected fellow eyes and compared to healthy eyes. Automated choroidal segmentation software allows highly reliable, faster, and more objective evaluation of choroid (10,11) compared with manual techniques (5–9).
In our study, all pCT measurements and all macular CT (except at the superior sector) were significantly thicker in fellow uninvolved eyes compared with control eyes. Although all CT measurements were also thicker in NAION eyes compared with control eyes, the differences were not significant. Our results regarding pCT support the findings reported by Fard et al and Nagia et al (5,7), who found a thickening of the peripapillary choroid not only in NAION eyes but also in the unaffected fellow eyes of NAION patients, suggesting that a thicker choroid was not the effect, but the cause of NAION.
The intrasession repeatability of PCT measurements at 4 quadrants was excellent (ICC ≥0.98). These results are in line with a better reproducibility comparing SD-OCT and SS-OCT to measure CT (10).
Regarding the profile of the pCT, our results confirmed the asymmetrical distribution of pCT being the inferior peripapillary choroid significantly thinner than the superior choroid in all groups (P = 0.00). These observations are in line with previous studies using EDI-OCT in normal and NAION eyes which consistently have shown the inferior region to be significantly thinner than other regions (5,6,21). In addition, we analyzed the subgroup of eyes having inferior altitudinal visual field defects (n = 7) to evaluate whether the pCT changed after the NAION episode in the corresponding areas. However, superior peripapillary choroid continued being significantly thicker compared with the inferior choroid (150.5 ± 67.98 vs 105.0 ± 43.03, P = 0.00), supporting the possibility that a thicker choroid was not the consequence of, but a risk factor for, NAION. Longitudinal prospective studies are necessary to clarify whether pCT changes in NAION are primary or secondary.
Regarding macular CT, we found significant differences between NAION eyes, unaffected fellow eyes, and control eyes by automated segmentation (ANOVA, P ≤ 0.03). Schuster et al (8) and Fard et al (5) found a thicker macular thickness in NAION eyes and their contralateral unaffected eyes compared with healthy controls, whereas Dias-Santos et al (9) found a thinner macular thickness in chronic NAION eyes compared with controls.
Table E5 summarizes the pCT and macular CT measurements reported in the literature (See Supplementary Digital Content 5, Table E5, http://links.lww.com/WNO/A249). All our CT measurements are thinner than those previously reported. Manual segmentation and the device used could explain differences among studies. It is well known that values obtained by SS-OCT and SD-OCT cannot be directly compared (10,13,22). In fact, Tan et al (23) found that CT measurements by automated segmentations using DRI OCT-1 and manual segmentations by Spectralis OCT may differ by more than 50 μm. Other authors found a trend of CT measurements to be slightly thicker on SD-OCT vs SS-OCT but with limited clinical relevance (10,12).
In line with previous studies (22), we found a significant inverse association between age and macular CT (−0.354, P = 0.01). No association was found between the mean age and mean pCT (−0.196, P = 0.08).
Regarding disc area size as a risk factor for NAION, there are conflicting results (5,7,15–17). This study compared disc area in patients with NAION using SD-OCT and BMO-MRW. BMO represents a measurable aperture at the level of the ONH through which retinal ganglion cell axons pass (18,19). We did not find any difference in optic disc size between patients with NAION, unaffected fellow eyes, and the control group (ANOVA, P = 0.84). Our results are in line with those published by Nagia et al (7) who found no statistical differences in BMO area between groups. However, Fard et al (5) found that the optic disc area in NAION eyes was significantly smaller (1.86 ± 0.30 mm2) than normal control disc area (2.23 ± 0.44 mm2), but disc area was obtained by marking the optic disc contour at the inner border of the scleral ring on infrared OCT ONH images. Unlike clinical, photographic, or scanning laser tomographic evaluation, BMO-based rim measurement takes into account the clinically invisible termination of Bruch membrane and yields more objective values.
According to previous reports, average and all measurements of sectorial peripapillary RNFL thickness were significantly thinned in NAION eyes compared with fellow and control eyes (See Supplementary Digital Content 4, Table E4, http://links.lww.com/WNO/A248) (5,6). Similarly, all BMO-MRW measurements were significantly thinned in NAION eyes compared with unaffected and control eyes (See Supplementary Digital Content 4, Table E4, http://links.lww.com/WNO/A248). Remarkably, average and sectorial BMO-MRW were significantly thicker in unaffected contralateral eyes than in control eyes (P ≤ 0.02) except for the nasal sector (P = 0.09) (See Supplementary Digital Content 4, Table E4, http://links.lww.com/WNO/A248). Data regarding BMO-MRW thickness values from NAION eyes has not been reported previously.
BMO-MRW parameter is a more accurate reflection of axons passing through the ONH opening than peripapillary RNFL thickness because it is obtained perpendicular to the axis of the neural tissue and it takes into account the variable trajectory of axons over the point of measurement (18). There were no statistically significant differences in BMO area among groups, so a narrower scleral opening leading to an impaired axoplasmic flow with increased choroidal and BMO-MRW thicknesses might be considered a risk factor.
There are limitations of our study. First, our sample size was small but our findings achieved statistical significance. Second, diurnal fluctuations in CT were not taken into account (24,25). Third, CT does not reflect choroidal perfusion and, therefore, data about choroidal blood flow cannot be derived from our study.
STATEMENT OF AUTHORSHIP
Category 1: a. Conception and design: G. Rebolleda, F. J. Muñoz-Negrete; b. Acquisition of data: G. Rebolleda, F. J. Muñoz-Negrete, V. De Juan; c. Analysis and interpretation of data: G. Rebolleda, F. J. Muñoz-Negrete, V. De Juan, S. Noval. Category 2: a. Drafting the manuscript: G. Rebolleda, F. J. Muñoz-Negrete, V. De Juan, S. Noval; b. Revising it for intellectual content: G. Rebolleda, F. J. Muñoz-Negrete, V. De Juan, S. Noval. Category 3: a. Final approval of the completed manuscript: G. Rebolleda, F. J. Muñoz-Negrete, V. De Juan, S. Noval.
1. Miller NR, Arnold AC. Current concepts in the diagnosis, pathogenesis and management of nonarteritic anterior ischaemic optic neuropathy. Eye (Lond). 2015;29:65–79.
2. Hayreh SS. Blood supply of the optic nerve head and its role in optic atrophy, glaucoma, and oedema of the optic disc. Br J Ophthalmol. 1969;53:721–748.
3. Hayreh SS; The 1994 Von Sallman Lecture. The optic nerve head circulation in health and disease. Exp Eye Res. 1955;61:259–272.
4. 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.
5. Fard MA, Abdi P, Kasaei A, Soltani Mogaddam R, Afzali M, Moghimi S. Peripapillary choroidal thickness in nonarteritic anterior ischemic optic neuropathy. Invest Ophthalmol Vis Sci. 2015;56:3027–3033.
6. García-Basterra I, Lahrach I, Morillo Sánchez MJ, Kamal-Salah R, Ríus-Díaz F, Dawid Milner MS, García-Campos JM. Analysis of peripapillary choroidal thickness in non-arteritic anterior ischaemic optic neuropathy. Br J Ophthalmol. 2015;100:891–896.
7. Nagia L, Huisingh C, Johnstone J, Kline LB, Clark M, Girard MJ, Mari JM, Girkin CA. Peripapillary pachychoroid in nonarteritic anterior ischemic optic neuropathy. Invest Ophthalmol Vis Sci. 2016;57:4679–4685.
8. Schuster AK, Steinmetz P, Forster TM, Schlichtenbrede FC, Harder BC, Jonas JB. Choroidal thickness in nonarteritic anterior ischemic optic neuropathy. Am J Ophthalmol. 2014;158:1342–1347.
9. Dias-Santos A, Ferreira J. Choroidal thickness in nonarteritic anterior ischaemic optic neuropathy: a study with optical coherence tomography. Neuroophthalmology. 2014;38:173–179.
10. Copete S, Flores-Moreno I, Montero JA, Duker JS, Ruiz-Moreno JM. Direct comparison of spectral-domain and swept-source OCT in the measurement of choroidal thickness in normal eyes. Br J Ophthalmol. 2014;98:334–338.
11. Mansouri K, Medeiros FA, Tatham AJ, Marchase N, Weinreb RN. Evaluation of retinal and choroidal thickness by swept-source optical coherence tomography: repeatability and assessment of artifacts. Am J Ophthalmol. 2014;157:1022–1032.
12. Philip AM, Gerendas BS, Zhang L, Faatz H, Podkowinski D, Bogunovic H, Abramoff MD, Hagmann M, Leitner R, Simader C, Sonka M, Waldstein SM, Schmidt-Erfurth U. Choroidal thickness maps from spectral domain and swept source optical coherence tomography: algorithmic versus ground truth annotation. Br J Ophthalmol. 2016;100:1372–1376.
13. Song YJ, Kim YK, Jeoung JW, Park KH. Assessment of open-angle glaucoma peripapillary and macular choroidal thickness using swept-source optical coherence tomography (SS-OCT). PLoS One. 2016;16:e0157333.
14. Gupta P, Jing T, Marziliano P, Baskaran M, Cheung GC, Lamoreux EL, Cheung GY, Wong TY, Aung T, Cheng CY. Peripapillary choroidal thickness assessed using automated choroidal segmentation software in an Asian population. Br J Ophthalmol. 2015;99:920–926.
15. Mansour AM, Shoch D, Logani S. Optic disk size in ischemic optic neuropathy. Am J Ophthalmol. 1988;106:587–589.
16. Jonas JB, Gusek GC, Naumann GO. Anterior ischemic optic neuropathy: nonarteritic form in small and giant cell arteritis in normal sized optic discs. Int Ophthalmol. 1988;12:119–125.
17. Contreras I, Rebolleda G, Noval S, Mun[Combining Tilde]oz-Negrete FJ. Optic disc evaluation by optical coherence tomography in nonarteritic anterior ischemic optic neuropathy. Invest Ophthalmol Vis Sci. 2007;48:4087–4092.
18. Reis AS, O'Leary N, Yang H, Sharpe GP, Nicolela MT, Burgoyne CF, Chauhan BC. Influence of clinically invisible, but optical coherence tomography detected, optic disc margin anatomy on neuroretinal rim evaluation. Invest Ophthalmol Vis Sci. 2012;53:1852–1860.
19. Chauhan BC, Danthurebandara VM, Sharpe GP, Demirel S, Girkin CA, Mardin CY, Scheuerle AF, Burgoyne CF. Bruch's membrane opening-minimum rim width and retinal nerve fibre layer thickness in a normal white population. A multi-centre study. Ophthalmology. 2015;122:1786–1794.
20. Papchenko T, Grainger BT, Savino PJ, Gamble GD, Danesh-Meyer HV. Macular thickness predictive of visual field sensitivity in ischaemic optic neuropathy. Acta Ophthalmol. 2012;90:e463–e469.
21. Tanabe H, Ito Y, Terasaki H. Choroid is thinner in inferior region of optic disks of normal eyes. Retina. 2012;32:134–139.
22. Rhodes LA, Huisingh C, Johnstone J, Fazio MA, Smith B, Wang L, Clark M, Downs JC, Owsley C, Girard MJ, Mari JM, Girkin CA. Peripapillary choroidal thickness variation with age and race in normal eyes. Invest Ophthalmol Vis Sci. 2015;24:1872–1879.
23. Tan CS, Cheong KX, Lim LW, Sadda SR. Comparison of macular choroidal thicknesses from swept source and spectral domain optical coherence tomography. Br J Ophthalmol. 2015;100:995–999.
24. Tan CS, Ouyang Y, Ruiz H, Sadda SR. Diurnal variation of choroidal thickness in normal, healthy subjects measured by spectral domain optical coherence tomography. Invest Ophthalmol Vis Sci. 2010;53:261–266.
25. Chakraborty R, Read SA, Collins MJ. Diurnal variations in axial length, choroidal thickness, intraocular pressure, and ocular biometrics. Invest Ophthalmol Vis Sci. 2011;52:5121–5129.