Resolution of disc oedema in nonarteritic anterior ischaemic optic neuropathy versus optic neuritis: a retrospective, cross-sectional study Authors: Jong Won Moon, Sook Hyun Yoon, Geun Woo Lee, Donghun Lee Published: 2026 ABSTRACT BackgroundThe common causes of unilateral disc oedema include nonarteritic anterior ischaemic optic neuropathy (NAION) and optic neuritis (ON). As prolonged disc oedema may affect visual function, early differentiation is crucial. Therefore, this study compared the resolution time and rate of disc oedema between NAION and ON cases and identified correlated clinical factors.MethodsIn this retrospective, cross-sectional comparison study, unilateral NAION or ON cases, in which disc oedema was confirmed through fundus photography, were reviewed. Optical coherence tomography data, such as structural changes in the peripapillary retinal nerve fibre layer thickness (pRNFLT), ganglion cell layer thickness at two time points of the first diagnosis, and stable states with subsided disc oedema, were analysed. The severity, resolution rate, and duration of disc oedema were compared between the two groups.ResultsThis study included 33 eyes each with NAION and ON (mean age: NAION, 69.1 ± 11.8 years; ON, 61.7 ± 15.7 years). No significant differences were observed in sex, visual acuity at diagnosis, colour vision, or visual field defect severity. Mean periods of disc oedema were 1.5 ± 0.6 months in NAION and 1.3 ± 0.7 months in ON cases. At diagnosis, the mean pRNFLT in NAION cases (215.7 ± 57.6 μm) was not different from that in ON cases (199.1 ± 60.4 μm); the resolution rate of disc oedema was not significantly different between the two diseases.ConclusionThe resolution rate of disc oedema—which did not differ significantly—cannot be used to distinguish between NAION and ON. In both diseases, greater initial disc oedema was associated with a longer duration of oedema, but the rate of oedema resolution was faster, resulting in a narrower gap in total resolution time between mild and severe cases.Trial registrationClinical trial number: not applicable. This retrospective study was approved by the Daegu Catholic University Hospital Institutional Review Board (IRB) for Human Studies (IRB number: 2025-03-012). All procedures adhered to the tenets of the Declaration of Helsinki. Informed consent was obtained verbally from all participants, with their guardians and residents as witnesses. Background The common causes of unilateral disc oedema include nonarteritic anterior ischaemic optic neuropathy (NAION) and optic neuritis (ON) [1–3]. NAION is caused by ischaemia and ON is caused by inflammation, and the two diseases differ in clinical features (symptom onset and presence of ocular pain) and pattern of visual field defects [1]. However, it is often difficult to distinguish between the two diseases based on the morphology of optic nerve oedema in the early phase (Fig. 1). Therefore, several studies have reported different findings between the two diseases on various ocular examinations such as funduscopy, optical coherence tomography (OCT) angiography, and orbital magnetic resonance imaging (MRI) [1, 3–5]. Fig. 1Changes in disc oedema and visual field defect in nonarteritic anterior ischaemic optic neuropathy (NAION; Figs A–F) and optic neuritis (ON; Figs G–L). In the early stages of both diseases, optic disc oedema (white arrow) (A, C: NAION; G, I: ON) and diffuse visual field defects (B: NAION; H: ON) are observed. Optic disc oedema on funduscopy (D) and OCT (F) gradually resolved but an inferior altitudinal field defect remained (E) in NAION. In contrast, the visual field (K) completely recovered in the chronic stage as optic disc oedema (J, L) recovers in ON Changes in disc oedema and visual field defect in nonarteritic anterior ischaemic optic neuropathy (NAION; Figs A–F) and optic neuritis (ON; Figs G–L). In the early stages of both diseases, optic disc oedema (white arrow) (A, C: NAION; G, I: ON) and diffuse visual field defects (B: NAION; H: ON) are observed. Optic disc oedema on funduscopy (D) and OCT (F) gradually resolved but an inferior altitudinal field defect remained (E) in NAION. In contrast, the visual field (K) completely recovered in the chronic stage as optic disc oedema (J, L) recovers in ON Adensina et al. [5] demonstrated that diffusion MRI showed focal diffusion restriction at the optic disc in NAION and could be used to differentiate acute NAION from ON. Using OCT angiography, Fard et al. [6] found that the whole peripapillary region vascularity on optic nerve head imaging in NAION cases was significantly lower than that in ON or papilledema cases. They reported the usefulness of OCT in distinguishing NAION from other types of disc oedema, mentioning a significantly decreased whole-image capillary density in NAION cases. Arnold and Hepler [7] found that fluorescein angiography of NAION showed a significant delay in the onset and time to completion of prelaminar optic disc filling. Additionally, studies have focused on differences in the optic disc morphology between these two diseases [2, 8]. As the visual prognosis and treatment of the two diseases are different, early differentiation is important for treating and counselling patients [4]. Prolonged disc oedema may induce optic atrophy and affect visual function [1]. However, to the best of our knowledge, the duration and resolution of swollen discs in these two diseases have not been previously compared. Therefore, this study aimed to compare the resolution rate and duration of disc oedema between NAION and ON patients and to determine the factors affecting the subsidence of disc oedema. Materials and methods This retrospective study was approved by the Daegu Catholic University Hospital Institutional Review Board (IRB) for Human Studies (IRB number: 2025-03-012). All procedures adhered to the tenets of the Declaration of Helsinki. Informed consent was obtained verbally from all participants, with their guardians and residents as witnesses. The medical records of patients with unilateral NAION or ON confirmed using funduscopy between January 2018 and July 2023 were retrospectively reviewed. The study was conducted on patients who visited the clinic within 2 weeks of symptom onset and were followed up for at least 2 months after diagnosis. The authors had access to identifiable patient information during data collection but ensured de-identification for analysis. Diagnostic criteria for NAION and ON The diagnostic criterion for NAION was painless sudden onset visual loss or field defect [1]. On ocular examination, a relatively afferent pupillary defect was detected. Disc oedema was observed on fundus examination and correlating field defects on the visual field test. Blood sampling and brain imaging were performed. All patients with disc oedema underwent orbital MRI, and T1 fat-suppressed images confirmed the absence of contrast enhancement along the optic nerve in NAION cases. Other causes, such as infection, inflammation, demyelination, mass lesions, and brain parenchymal infarction, were excluded. Patients with arterial anterior ischaemic optic neuropathy were excluded. The diagnostic criteria for ON were based on the 1991 Optic Neuritis Treatment Trial guidelines [9]. Acute clinical symptoms, including visual disturbances, colour vision deficiency, and field defects, were consistent with unilateral ON. Relative afferent pupillary and visual field defects were observed in affected eyes. Fundus examination revealed unilateral disc oedema. All patients suspected to have ON underwent orbital MRI using T1 fat-suppressed images to detect enhancement of the involved optic disc or nerve [10]. To exclude factors that may have affected optic disc morphology, patients with diabetic retinopathy; ischaemic retinal vessel occlusion; and other optic neuropathies, including glaucoma, toxic optic neuropathy, and papilledema, were excluded from the study. Other cases, including thyroid eye disease, ocular trauma, and history of orbital surgery, were also excluded. Patients with retrobulbar ON, in which disc oedema was not detected, were excluded from the study. Ocular examination Ophthalmological examinations included measurement of best-corrected visual acuity (BCVA) and intraocular pressure using a non-contact tonometer, manifest refractive error measurement using autorefractive keratometry, slit-lamp examination, spectral-domain OCT (Cirrus HD-OCT Carl Zeiss Meditec, Inc., Dublin, CA), fundus photography using a digital fundus camera (CF-60DSi; Canon, Tokyo, Japan), and visual field test (VFT) using static perimetry (Humphrey Field Analyzer SITA-Standard 24 − 2, Humphrey System, Dublin, CA). The refractive error results were converted to spherical equivalents, and BCVA was converted to the logarithm of minimal angle resolution (logMAR) for statistical analysis. Patients diagnosed with NAION or ON underwent fundus photography two or three times within the first week after diagnosis. Follow-up examinations were conducted at 2-week intervals for the first 2 months and, subsequently, at monthly intervals for 3 months. The time point for the stable state with resolution of optic disc oedema was defined as the absence of clinically visible disc swelling on fundus examination, together with the affected-eye mean peripapillary retinal nerve fibre layer thickness (pRNFLT) returning to within 10% of the fellow-eye mean pRNFLT. An experienced technician performed the OCT measurements. Optic disc and retina data were also obtained. The optic disc structure index included pRNFLT, rim area, disc area, and cup disc (CD) ratio, and the retinal structure index included ganglion cell/inner plexiform layer (GC/IPL) thickness. The pRNFLT measurements were assessed using a built-in software and obtained using the Cirrus HD-OCT Optic Disc Cube 200 × 200 protocol, which covered the 6 × 6-mm2 area centred on the optic disc. The 512 × 128-scan Cirrus HD-OCT macular cube protocol, centred on the fovea, was used for GC/IPL segmentation and measurements [11]. The macular cube protocol provided the average, minimum, and regional GC/IPL thickness values in six wedge-shaped sectors. The minimum GC/IPL thickness was defined as the lowest GC/IPL thickness over a single meridian crossing the annulus. Among them, we evaluated the average and minimum GC/IPL thickness values to examine the changes in macular structure. Patients with a high refractive error ≥ ± 6 diopters and images with artefacts and signal strength < 6 were excluded. In addition, all OCT scans were reviewed for obvious segmentation failures (e.g., misidentification of layer boundaries or gross decentration), and scans with evident segmentation errors were excluded from analysis. Because of the calculation inaccuracy of the CD ratio of the swollen optic disc in the acute stage, the value of the CD ratio at diagnosis was used along with that of each patient’s contralateral eye. Changes in optic disc oedema and visual function, including BCVA, index of the visual field test, and colour vision test, of the two diseases were compared. Analysis of optic disc oedema Data at two time points of diagnosis and the stable state of disc oedema subsidence were analysed. Disc oedema was analysed using four parameters: (1) pRNFLT at diagnosis, (2) severity of disc oedema, (3) duration of disc oedema, and (4) rate of disc oedema resolution. pRNFLT at diagnosis (µm): This value reflects the amount of disc oedema in the initial phase. It was automatically measured using a built-in software program on the OCT.Severity of disc oedema (µm): This value is calculated as the difference between pRNFLT at diagnosis and the stable phase when the oedema has subsided. This value reflects the true severity of disc oedema by adjusting the value of the intact pRNFLT, which varied for each patient.Duration of disc oedema (months): This value was determined as the period in which disc oedema was observed on fundus photography.Rate of oedema resolution (µm/day): This value was calculated as the severity of disc oedema divided by duration of disc oedema. pRNFLT at diagnosis (µm): This value reflects the amount of disc oedema in the initial phase. It was automatically measured using a built-in software program on the OCT. pRNFLT at diagnosis (µm): This value reflects the amount of disc oedema in the initial phase. It was automatically measured using a built-in software program on the OCT. Severity of disc oedema (µm): This value is calculated as the difference between pRNFLT at diagnosis and the stable phase when the oedema has subsided. This value reflects the true severity of disc oedema by adjusting the value of the intact pRNFLT, which varied for each patient. Severity of disc oedema (µm): This value is calculated as the difference between pRNFLT at diagnosis and the stable phase when the oedema has subsided. This value reflects the true severity of disc oedema by adjusting the value of the intact pRNFLT, which varied for each patient. Duration of disc oedema (months): This value was determined as the period in which disc oedema was observed on fundus photography. Duration of disc oedema (months): This value was determined as the period in which disc oedema was observed on fundus photography. Rate of oedema resolution (µm/day): This value was calculated as the severity of disc oedema divided by duration of disc oedema. Rate of oedema resolution (µm/day): This value was calculated as the severity of disc oedema divided by duration of disc oedema. Statistical analysis Data were analysed using the Statistical Package for the Social Sciences software (version 22.0; IBM Corp., Armonk, NY). Statistical significance was set at P < 0.05. Changes in the OCT parameters between the two time points were analysed using paired t-tests. Clinical factors between the NAION and ON groups were compared using the independent t-test and chi-square test. Pearson’s correlation analysis was performed to identify clinical factors that correlated with the severity and changes in disc oedema. Results A total of 66 patients were enrolled in this study (33 with NAION and 33 with ON). Patient demographics are presented in Table 1. The mean age at diagnosis, sex, duration of symptom onset, mean follow-up duration, and intraocular and spherical refractive errors were not significantly different between the two groups. Table 1Baseline characteristics in patients with NAION and ONParameterNAIONONP valuesNumber of eyes3333Laterality (OD: OS)18 (54.5%):15 (45.5%)13 (39.4%):20 (60.6%)0.218aSex (M: F)16 (48.5%):17 (51.5%)18 (54.5%):15 (45.5%)0.806aAge at diagnosis (years)69.1 ± 11.8 (Range, 45–92)61.7 ± 15.7 (Range, 36–83)0.051bDuration on symptom onset (days)8.9 ± 6.6 (Range, 1–14)7.4 ± 3.5 (Range, 3–14)0.241bDuration of follow up period (months)15.3 ± 11.7 (Range,2.1–42.4)14.1 ± 15.2 (Range, 2.0-48.9)0.720bPresence of ocular moving pain (n)028< 0.0001aIOP (mmHg)12.6 ± 2.9 (Range, 8–20)13.0 ± 2.1 (Range, 9–17)0.256bRefractive errors (SE, D)-0.1 ± 1.4 (Range, -4.0-+2.5)-0.8 ± 1.6 (Range, -6.0-+1.3)0.062bachi square test. bt test. The values are expressed as mean ± standard deviation (%)NAION, non-arteritic anterior ischemic optic neuropathy; ON, optic neuritis; M, male; F, female; OD, oculus dexter; OS, oculus sinister; IOP, intraocular pressure; SE, spherical equivalent; D, diopter Table 1Baseline characteristics in patients with NAION and ONParameterNAIONONP valuesNumber of eyes3333Laterality (OD: OS)18 (54.5%):15 (45.5%)13 (39.4%):20 (60.6%)0.218aSex (M: F)16 (48.5%):17 (51.5%)18 (54.5%):15 (45.5%)0.806aAge at diagnosis (years)69.1 ± 11.8 (Range, 45–92)61.7 ± 15.7 (Range, 36–83)0.051bDuration on symptom onset (days)8.9 ± 6.6 (Range, 1–14)7.4 ± 3.5 (Range, 3–14)0.241bDuration of follow up period (months)15.3 ± 11.7 (Range,2.1–42.4)14.1 ± 15.2 (Range, 2.0-48.9)0.720bPresence of ocular moving pain (n)028< 0.0001aIOP (mmHg)12.6 ± 2.9 (Range, 8–20)13.0 ± 2.1 (Range, 9–17)0.256bRefractive errors (SE, D)-0.1 ± 1.4 (Range, -4.0-+2.5)-0.8 ± 1.6 (Range, -6.0-+1.3)0.062bachi square test. bt test. The values are expressed as mean ± standard deviation (%)NAION, non-arteritic anterior ischemic optic neuropathy; ON, optic neuritis; M, male; F, female; OD, oculus dexter; OS, oculus sinister; IOP, intraocular pressure; SE, spherical equivalent; D, diopter Baseline characteristics in patients with NAION and ON achi square test. bt test. The values are expressed as mean ± standard deviation (%) NAION, non-arteritic anterior ischemic optic neuropathy; ON, optic neuritis; M, male; F, female; OD, oculus dexter; OS, oculus sinister; IOP, intraocular pressure; SE, spherical equivalent; D, diopter Changes in visual function and optic disc structure from diagnosis to the stable phase of the two diseases are shown in Table 2. At diagnosis, the pattern standard deviation of VFT in NAION was worse than that in ON (P = 0.009). On OCT, rim and disc areas were larger in NAION than in ON (P = 0.018 and P = 0.039, respectively). The CD ratio in the contralateral eye at diagnosis was smaller in NAION than in ON (0.3 ± 0.2 vs. 0.4 ± 0.3; P < 0.001). Regarding changes in visual function and disc oedema in patients with NAION, visual acuity and colour perception did not significantly improve during the stable phase (P = 0.805 and P = 0.08, respectively), but visual field defects improved significantly (P = 0.008, mean deviation; P = 0.006, pattern standard deviation; P = 0.005, visual field index). On OCT, GC/IPL during the stable phase appeared thinner than that at diagnosis in patients with NAION (P < 0.001, mean; P = 0.003, minimum). In ON, all visual function at diagnosis significantly improved during the stable phase. Visual function parameters, including visual acuity, colour perception, and visual field defects, significantly improved during the observation period. As disc oedema subsided, GC/IPL thickness values were lower at the stable phase than at diagnosis (P < 0.001, mean; P = 0.001, minimum). In the stable phase of the two diseases, visual acuity and visual field index are significantly worse in NAION than in ON. In structural comparison, the CD ratio in affected eyes was significantly lower in NAION (P = 0.012) and the minimum GC/IPL thickness was less in NAION than in ON (58.7 ± 13.3 vs. 69.9 ± 10.5 μm; P < 0.001). Changes in mean pRNFLT at diagnosis and at the stable phase in NAION and ON were shown in Fig. 2. Both groups showed a significant reduction in pRNFLT over time. However, there was no statistically significant difference between NAION and ON in pRNFLT values at either diagnostic or stable phase. Table 2Changes of visual function and optic disc structure from diagnosis to stable phase between NAION and ONParameterNAIONONP value1 vs. 3P value1 vs. 2P value3 vs. 4P value2 vs. 4At diagnosis1At stable phase2At diagnosis3At stable phase4Visual acuity (logMAR)0.7 ± 0.70.6 ± 0.70.7 ± 0.50.2 ± 0.30.9150.805< 0.0010.001Index of visual field test MD (dB)-14.7 ± 7.3-11.4 ± 7.0-13.8 ± 11.4-5.3 ± 7.20.7070.008< 0.0010.002 PSD9.4 ± 3.68.8 ± 3.36.7 ± 4.33.9 ± 3.10.0090.0060.003< 0.001 VFI (%)58.0 ± 24.269.7 ± 22.260.0 ± 33.687.3 ± 21.40.7920.005< 0.0010.003No. of correct of color vision test9.6 ± 5.711.1 ± 5.310.0 ± 5.813.0 ± 4.20.7770.080.0050.198Index of OCT parameter Mean pRNFLT (µm)215.7 ± 57.692.7 ± 12.4199.2 ± 60.497.2 ± 10.30.258< 0.001< 0.0010.112 Rim area (mm2)2.4 ± 0.61.5 ± 0.42.1 ± 4.91.4 ± 0.30.018< 0.001< 0.0010.209 Disc area (mm2)2.5 ± 0.61.8 ± 0.32.2 ± 0.52.5 ± 3.30.039< 0.0010.5780.229 Cup disc ratio0.3 ± 0.2*0.3 ± 0.30.4 ± 0.3*0.4 ± 0.3< 0.0010.4620.0680.012 Mean GC/IPL thickness (µm)78.7 ± 12.771.9 ± 9.783.4 ± 8.275.6 ± 8.60.079< 0.001< 0.0010.103 Mean minimum GC/IPL thickness (µm)68.6 ± 18.958.7 ± 13.377.2 ± 9.969.9 ± 10.50.0230.0030.001< 0.001The values are expressed as mean ± standard deviation*values in the contralateral eye in the same patient was used due to inaccuracy of the CD ratio value of the swollen optic disc in acute stage.NAION, nonarteritic anterior ischemic optic neuropathy; ON, optic neuritis; logMAR, logarithm of the minimum angle of resolution; MD, mean deviation; PSD, pattern standard deviation; VFI, visual field index; OCT, Optical coherence tomography; pRNFLT, peripapillary retinal nerve fiber layer thickness; GC/IPL, ganglion cell/inner plexiform layer Table 2Changes of visual function and optic disc structure from diagnosis to stable phase between NAION and ONParameterNAIONONP value1 vs. 3P value1 vs. 2P value3 vs. 4P value2 vs. 4At diagnosis1At stable phase2At diagnosis3At stable phase4Visual acuity (logMAR)0.7 ± 0.70.6 ± 0.70.7 ± 0.50.2 ± 0.30.9150.805< 0.0010.001Index of visual field test MD (dB)-14.7 ± 7.3-11.4 ± 7.0-13.8 ± 11.4-5.3 ± 7.20.7070.008< 0.0010.002 PSD9.4 ± 3.68.8 ± 3.36.7 ± 4.33.9 ± 3.10.0090.0060.003< 0.001 VFI (%)58.0 ± 24.269.7 ± 22.260.0 ± 33.687.3 ± 21.40.7920.005< 0.0010.003No. of correct of color vision test9.6 ± 5.711.1 ± 5.310.0 ± 5.813.0 ± 4.20.7770.080.0050.198Index of OCT parameter Mean pRNFLT (µm)215.7 ± 57.692.7 ± 12.4199.2 ± 60.497.2 ± 10.30.258< 0.001< 0.0010.112 Rim area (mm2)2.4 ± 0.61.5 ± 0.42.1 ± 4.91.4 ± 0.30.018< 0.001< 0.0010.209 Disc area (mm2)2.5 ± 0.61.8 ± 0.32.2 ± 0.52.5 ± 3.30.039< 0.0010.5780.229 Cup disc ratio0.3 ± 0.2*0.3 ± 0.30.4 ± 0.3*0.4 ± 0.3< 0.0010.4620.0680.012 Mean GC/IPL thickness (µm)78.7 ± 12.771.9 ± 9.783.4 ± 8.275.6 ± 8.60.079< 0.001< 0.0010.103 Mean minimum GC/IPL thickness (µm)68.6 ± 18.958.7 ± 13.377.2 ± 9.969.9 ± 10.50.0230.0030.001< 0.001The values are expressed as mean ± standard deviation*values in the contralateral eye in the same patient was used due to inaccuracy of the CD ratio value of the swollen optic disc in acute stage.NAION, nonarteritic anterior ischemic optic neuropathy; ON, optic neuritis; logMAR, logarithm of the minimum angle of resolution; MD, mean deviation; PSD, pattern standard deviation; VFI, visual field index; OCT, Optical coherence tomography; pRNFLT, peripapillary retinal nerve fiber layer thickness; GC/IPL, ganglion cell/inner plexiform layer Changes of visual function and optic disc structure from diagnosis to stable phase between NAION and ON The values are expressed as mean ± standard deviation *values in the contralateral eye in the same patient was used due to inaccuracy of the CD ratio value of the swollen optic disc in acute stage.NAION, nonarteritic anterior ischemic optic neuropathy; ON, optic neuritis; logMAR, logarithm of the minimum angle of resolution; MD, mean deviation; PSD, pattern standard deviation; VFI, visual field index; OCT, Optical coherence tomography; pRNFLT, peripapillary retinal nerve fiber layer thickness; GC/IPL, ganglion cell/inner plexiform layer Fig. 2Changes in peripapillary retinal nerve fiber layer thickness (pRNFLT) across clinical time points in non-arteritic anterior ischemic optic neuropathy (NAION) and optic neuritis (ON). Significant reduction in pRNFLT was observed in both NAION and ON, with no intergroup difference at either time point Changes in peripapillary retinal nerve fiber layer thickness (pRNFLT) across clinical time points in non-arteritic anterior ischemic optic neuropathy (NAION) and optic neuritis (ON). Significant reduction in pRNFLT was observed in both NAION and ON, with no intergroup difference at either time point A comparison of the four parameters of disc oedema between the NAION and ON groups are presented in Table 3. The pRNFLT measurement was insignificantly higher in NAION than in ON (215.7 ± 57.6 vs. 199.1 ± 60.4 μm; P = 0.258). The severity and duration of disc oedema were not significantly different between the groups (P = 0.146 and P = 0.161, respectively). The resolution rate of disc oedema was 2.93 ± 1.52 μm/day in NAION and 2.91 ± 1.55 μm/day in ON (P = 0.959). In summary, severity, duration of disc oedema persistence, and the rate of oedema resolution are not significantly different between the two diseases. Table 3Analysis of optic disc oedema between NAION and ONParameterNAIONONP valuesaMean pRNFLT at diagnosis (µm)215.7 ± 57.6 (Range, 125–356)199.1 ± 60.4 (Range, 120–348)0.258Severity of disc oedema (µm)123.1 ± 56.4 (Range, 33–246)101.9 ± 60.1 (Range, 18–249)0.146Duration of disc oedema (months)1.5 ± 0.6 (Range, 0.7–3.1)1.3 ± 0.7 (Range, 0.2–2.5)0.161Rate of oedema resolution (µm/day)2.9 ± 1.5 (Range, 0.5–6.9)2.9 ± 1.5 (Range, 0.7–7.6)0.959aIndependent t test was used.NAION, non-arteritic anterior ischemic optic neuropathy; ON, optic neuritis; pRNFLT, peripapillary retinal nerve fiber layer thickness.The values are expressed as mean ± standard deviation. Table 3Analysis of optic disc oedema between NAION and ONParameterNAIONONP valuesaMean pRNFLT at diagnosis (µm)215.7 ± 57.6 (Range, 125–356)199.1 ± 60.4 (Range, 120–348)0.258Severity of disc oedema (µm)123.1 ± 56.4 (Range, 33–246)101.9 ± 60.1 (Range, 18–249)0.146Duration of disc oedema (months)1.5 ± 0.6 (Range, 0.7–3.1)1.3 ± 0.7 (Range, 0.2–2.5)0.161Rate of oedema resolution (µm/day)2.9 ± 1.5 (Range, 0.5–6.9)2.9 ± 1.5 (Range, 0.7–7.6)0.959aIndependent t test was used.NAION, non-arteritic anterior ischemic optic neuropathy; ON, optic neuritis; pRNFLT, peripapillary retinal nerve fiber layer thickness.The values are expressed as mean ± standard deviation. Analysis of optic disc oedema between NAION and ON aIndependent t test was used.NAION, non-arteritic anterior ischemic optic neuropathy; ON, optic neuritis; pRNFLT, peripapillary retinal nerve fiber layer thickness.The values are expressed as mean ± standard deviation. Correlation with severity of disc oedema and duration of disc oedema in both diseases were evaluated (Figs. 3 and 4). A significant positive correlation was found between the mean pRNFLT at diagnosis and duration of disc oedema (r = 0.343, P = 0.05; Fig. 3A) and resolution rate of disc oedema (r = 0.612, P < 0.001; Fig. 3B). Furthermore, greater disc oedema corresponded to longer oedema duration (r = 0.389, P = 0.03; Fig. 3C) and faster rate of oedema resolution (r = 0.579, P < 0.001; Fig. 3D). In ON, a similar positive correlation was observed between the severity, duration, and resolution rate of disc oedema (Fig. 4). A significant positive correlation was observed between the mean pRNFLT at diagnosis and the duration of disc oedema (r = 0.512, P = 0.002; Fig. 4A) and resolution rate of disc oedema (r = 0.356, P = 0.04; Fig. 4B). Additionally, greater disc oedema corresponded to longer oedema duration (r = 0.490, P = 0.003; Fig. 4C) and faster rate of oedema resolution (r = 0.40, P = 0.02; Fig. 4D). Fig. 3Correlation with severity of disc oedema and duration of disc oedema in nonarteritic anterior ischaemic optic neuropathy. A significant positive correlation was found between the mean peripapillary retinal nerve fibre layer thickness at diagnosis and the duration of disc oedema (A) and the rate of disc oedema resolution (B). The greater the disc oedema, the longer the oedema duration (C) but the faster the resolution rate (D). Pearson’s correlation analysis was used Correlation with severity of disc oedema and duration of disc oedema in nonarteritic anterior ischaemic optic neuropathy. A significant positive correlation was found between the mean peripapillary retinal nerve fibre layer thickness at diagnosis and the duration of disc oedema (A) and the rate of disc oedema resolution (B). The greater the disc oedema, the longer the oedema duration (C) but the faster the resolution rate (D). Pearson’s correlation analysis was used Fig. 4Correlation with severity of disc oedema and duration of disc oedema in optic neuritis. Similar to the results from NAION, a significant positive correlation was observed between the mean peripapillary retinal nerve fibre layer thickness at diagnosis and the duration of disc oedema (A) and the rate of disc oedema resolution (B). The greater the disc oedema, the longer the oedema duration (C) but the faster the resolution rate (D). Pearson’s correlation analysis was used Correlation with severity of disc oedema and duration of disc oedema in optic neuritis. Similar to the results from NAION, a significant positive correlation was observed between the mean peripapillary retinal nerve fibre layer thickness at diagnosis and the duration of disc oedema (A) and the rate of disc oedema resolution (B). The greater the disc oedema, the longer the oedema duration (C) but the faster the resolution rate (D). Pearson’s correlation analysis was used Discussion In this study, we compared characteristics that could help differentiate between NAION and ON based on disc oedema changes observed on fundus photography and OCT and investigated the prognosis of visual function in both of them. The duration and severity of disc oedema did not differ significantly between NAION and ON. Although optic disc oedema is usually observed in both of them and the clinical profiles of these entities overlap [3], their progression is different. The visual prognosis in ON is typically favorable; however, visual loss or visual field defects in NAION are often irreversible, and effective treatments are not yet known [12]. Therefore, differentiating these two diseases is important for determining a patient’s prognosis and treatment plan. Previous studies have focused on the usefulness of optic disc morphology for differentiating between the two diseases [2, 8, 13], but there has been some debate. Warner et al. [13] reported that the appearance of the optic disc helps distinguish between two diseases; altitudinal oedema, pallor, arterial attenuation, and haemorrhage are commonly found in NAION. In contrast, Kim and Kim [2] found that disc morphology did not differ between the two diseases; rather, cotton-wool spots, focal telangiectasia, and venous congestion on the retina were more common in NAION than in ON. Rath et al. [8] studied the morphologic features of optic disc atrophy in the chronic stage and reported that disc pallor in NAION is more severe than that in ON. Regarding changes in visual function and optic disc structure from diagnosis to resolution of disc oedema, visual acuity significantly improved only in ON from 0.7 ± 0.5 to 0.2 ± 0.3 logMAR; visual acuity, 0.7 ± 0.7 logMAR in NAION did not change. On VFT, The sequelae of visual dysfunction were more profound in NAION, and this result corresponds with Christoph Kernstock’s report [14] that functional deficits are present directly after the onset of NAION and do not change significantly further. Additionally, the results of the smaller CD ratio in NAION were similar to those in Raed Behbehani’s reports [15], wherein the mean CD ratio in NAION was 0.3 ± 0.2. This indicates that crowded discs are an anatomical risk factor for NAION [16]. Chronic mechanical obstruction may lead to vascular insufficiency resulting in optic disc ischaemia. Regarding the mean minimum GC/IPL thickness, the values were significantly lower in the NAION group compared to the ON group in both the acute and stable phases. This may reflect multiple factors. The average age at diagnosis was numerically higher in the NAION group, and age-related thinning of GC/IPL may have partially influenced the lower values. Additionally, systemic vascular risk factors such as diabetes and hypertension, which are more prevalent in NAION, may have contributed to the thinner GC/IPL observed. We have now included this interpretation in the Discussion section to clarify the potential reasons behind the lower GC/IPL thickness observed in the NAION group. Regarding changes in GC/IPL thickness, the mean and minimum GC/IPL were significantly thinner in both diseases after the disc oedema subsided. Several studies have reported a significant decrease in GC/IPL thickness in NAION [17–19] and ON [20]. Similar to permanent structural loss of ganglion cells commonly occurring by 1–2 months [20], GC/IPL thinning occurs before pRNFLT thinning. However, GC/IPL measurements obtained during the acute phase of marked optic nerve head (ONH) swelling should be interpreted with caution, as segmentation artefacts and reduced measurement reliability may occur even when image quality criteria are met. Therefore, the GC/IPL findings in the present study should be regarded as supportive structural information, with greater emphasis placed on measurements obtained at the stable phase after disc oedema has subsided. Disc oedema, regardless of its aetiology—ischaemia or inflammation— leads to constriction of the optic disc inlet and inhibition of axoplasmic flow, ultimately inducing ganglion cell death before optic nerve atrophy [21]. Nevertheless, the exact magnitude of acute-to-stable GC/IPL change may be influenced by measurement limitations in the presence of marked disc oedema. The severity of GC/IPL loss did not differ between the NAION and ON groups—a result that corresponds with Erlich-Malona et al.’s [1]. report that progression and severity of ganglion cell complex and RNFL loss did not differ significantly in two diseases. In recent years, OCT angiography has been increasingly employed for the identification and quantification of ocular structures, thereby supporting the diagnosis and prognosis of eye diseases, including macular and optic nerve disorders [6, 22, 23]. Fard et al. [6] noted that whole peripapillary and nasal sextant values on optic nerve head imaging were lower in NAION eyes than in ON eyes. Additionally, Karrabi et al. [22] demonstrated that the reduced peripapillary vascularity in patients with NAION was attributable to axoplasm blockage, whereas the superficial vessel density reduction was attributable to thinning of the underlying RNFL layers. Consistent with previous findings, the thinning of the nerve layer observed in our data during the stable phase in both diseases was attributable to the irreversible damage to the RNFL and GC/IPL caused by acute axonal injury. Furthermore, a subsequent reduction in blood flow is anticipated. However, a limitation of our study is that we were unable to compare OCT angiography results between the two groups. Further research using OCT angiography to investigate longitudinal changes in RNFLT and peripapillary vascularity in NAION and ON would be valuable. The initial pRNFLT and duration, severity, and rate of disc oedema resolution were not significantly different between NAION and ON, implying that the severity and duration of disc oedema are difficult to use to distinguish NAION from ON—this corresponds with Kim and Kim’s report that the severity of hyperaemia and the pattern of disc oedema are not different between the two diseases. In this study, the initial pRNFLT was 215.7 ± 57.6 μm in NAION and 199.15 ± 60.4 μm in ON. The duration of oedema was 1.5 ± 0.6 months in NAION and 1.3 ± 0.7 months in ON, similar to the result in a study by Inés Contreras et al. [24] on changes in pRNFLT in NAION wherein RNFLT in the acute phase was 200.9 μm and decreased to values within normal limits at the 1.5-month follow-up visit. In previous studies of pRNFLT in ON, values in the acute phase varied from 100.7 to 164 μm [11]. This is thought to be because it varies depending on the cause of ON, such as demyelination or multiple sclerosis. Interestingly, a common pattern was observed in both diseases: greater initial disc oedema was associated with a longer overall duration of oedema, but also with a faster rate of oedema resolution. In other words, although oedema in these cases persisted for a longer period before reaching a stable state, the structural change per unit time during the resolution phase was more rapid. This suggests that disc oedema may reflect diffuse axonal injury, where more extensive nerve damage leads to more severe oedema but also triggers faster resolution. However, the total time to structural stabilization to normalization remained longer in cases with severe initial oedema. Notably, the difference in total duration between mild and severe oedema cases was less than expected due to this accelerated resolution rate. Further studies including longitudinal analysis of optic atrophy severity are meaningful for investigating the long-term anatomical changes in disc oedema. This study aimed to determine whether the degree of optic disc oedema and the duration of oedema resolution could help differentiate between NAION and ON. Contrary to our assumptions, duration and severity of disc oedema did not differ significantly between the two diseases and thus cannot help to distinguish them. However, we found that the greater the initial disc oedema, the longer the oedema duration. Additionally, upon comparing the prognosis of the two diseases, we found that while visual acuity, colour vision, and visual field defects all showed improvement in patients with ON, only a modest improvement in visual field defects was observed in patients with NAION during the stable phase. These results may be valuable in informing patient prognosis and could contribute to more effective clinical management. This study had some limitations. First, we did not perform subtype analysis according to antibodies. ON classification based on antibodies, including myelin oligodendrocyte glycoprotein (MOG) antibody and aquaporin-4 antibody, has been established, and the prognosis and clinical features reportedly vary across different subtypes [25]. Authors were unable to analyse changes in optic disc oedema according to antibody subtype owing to the lack of antibody testing for all patients. Therefore, this study might have included a mix of idiopathic ON, MOG antibody–associated disease, and neuromyelitis optica spectrum disorder. These subtypes can differ significantly in clinical presentation, prognosis, and treatment response. Accordingly, the heterogeneity of the ON group should be taken into consideration when interpreting the results, and future studies incorporating antibody testing will be necessary to better characterize these subgroups. Second, although the age difference between the NAION and ON groups did not reach statistical significance, the NAION group had a numerically higher mean age. Given that retinal layer thickness may physiologically decrease with age [26, 27], this discrepancy could have influenced some of the structural measurements. However, due to the distinct age profiles inherent to each disease, matching the groups by age was clinically challenging. Moreover, the average age of both groups fell within the sixth decade of life, and the changes in disc oedema were observed over a relatively short period of several months. Therefore, the impact of this age difference on our primary findings is likely to be limited. Third, some studies have shown that vascular changes on OCT angiography have a strong functional correlation with NAION [28, 29]. However, our study did not include OCT angiography data. OCT angiography-based longitudinal assessment of RNFLT and peripapillary circulation in patients with NAION and ON would be a meaningful direction for future research. Conclusion In conclusion, in both diseases, the greater the initial disc edema, the longer the edema duration but the faster the rate of edema resolution. As the visual prognosis and treatment of the two diseases differ, early differentiation is crucial for appropriate patient management and counselling. Our findings regarding the correlation between oedema severity, duration, and resolution rate may aid in anticipating disease course and guiding follow-up strategies. Further studies, including a comparison of OCT angiography data between patients with NAION and ON, are warranted.