Abstract
Introduction: Anterior ischemic optic neuropathy (AION) can mimic glaucoma and consequently cause difficulties in differential diagnosis. The purpose of this paper was to summarize differences in diagnostic tests that can help perform a correct diagnosis. Methods: The search strategy was performed according to the PRISMA 2009 guidelines, and four databases were used: MEDLINE, Embase, Web of Science, and Cochrane. Totally, 772 references were eligible; 39 were included after screening with respect to inclusion criteria that included English language and published in the 20 years before search date. Results: Ninety percent (n = 35) of included studies used optical coherence tomography (OCT). Glaucomatous eyes had a significantly greater cup area, volume and depth, cup-to-disk ratio, a lower rim volume and area, and a thinner Bruch’s membrane opening-minimum rim width. Retinal nerve fiber layer (RNFL) thinning in glaucomatous eyes occurred primarily at the superotemporal, inferotemporal, and inferonasal sectors, while AION eyes demonstrated mostly superonasal thinning. Glaucoma eyes showed greater macular ganglion cell layer thickness, except at the inferotemporal sector. OCT angiography measurements demonstrated a significant decrease in superficial and deep macular vessel density (VD) in glaucoma compared to AION with similar degree of visual field damage; the parapapillary choroidal VD was spared in AION eyes compared to glaucomatous eyes. Conclusion: By use of OCT imaging, optic nerve head parameters seem most informative to distinguish between glaucoma and AION. Although both diseases affect the RNFL thickness, it seems to do so in different sectors. Differences in structure and vascularity of the macula can also help in making the differential diagnosis.
Introduction
Clinicians within ophthalmology are not infrequently confronted with optic neuropathies. Although different in etiology, in later stages these conditions can present similarly, complicating a clinical diagnosis. This review focuses on the added value of diagnostic testing when clinical findings do not allow differentiation between a glaucomatous optic neuropathy (GON) and ischemic optic neuropathy.
Glaucoma, the most prevalent optic neuropathy, is defined as a chronic progressive neuropathy of the optic nerve and is characterized by optic disk cupping and corresponding visual field (VF) loss [1]. An increased intraocular pressure (IOP) is the main risk factor for developing glaucoma, but it is not mandatory as a subgroup of glaucoma patients suffers from normal-tension glaucoma (NTG) [2‒4]. Other risk factors besides IOP might be more decisive in NTG, such as impaired ocular blood flow [2, 4‒8]. Diagnosis is made by a combination of medical history, eye examination, including IOP measurement, gonioscopy, and fundoscopy, and a measurement of optic nerve structure (e.g., retinal nerve fiber layer [RNFL] thickness via optical coherence tomography [OCT]) and function (VF testing) [9‒11].
Ischemic optic neuropathies can be broadly classified as anterior or posterior, with anterior ischemic optic neuropathy (AION) being by far the most prevalent. AION is a neuro-ophthalmologic condition that is caused by occlusion in one (or more) of the ciliary arteries in the optic nerve head (ONH), resulting in acute (monocular) vision loss, and is often accompanied by an altitudinal VF defect (VFD) [12‒14]. Although the exact etiopathogenesis of nonarteritic AION (NAION) is still unclear, risk factors such as an anatomically crowded ONH, nocturnal hypotension, local thrombosis, or atherosclerosis of the circulation may contribute to the hypoperfusion of the optic nerve [15]. Arteritic AION (AAION) is considerably more severe, with worse prognosis, and is secondary to giant cell arthritis [16]. It is often accompanied by jaw claudication, anorexia, elevated erythrocyte sedimentation rate or reactive protein C levels, and a positive temporal artery biopsy [16‒18].
In the acute phase, in most cases, NAION and AAION present differently from glaucoma. In AION, the optic disk is swollen, and splinter hemorrhages can be detected; optic disk pallor can develop in the following weeks/months [10, 19]. However, after a few months, the acute changes fade and an optic nerve affected by AION may start to look very similar to a glaucomatous optic nerve – excavated, atrophic, and pale – while associated with sometimes similar VFDs [20‒22]. The distinction between AAION and glaucoma can mostly be made on the basis of the sudden and often very profound decrease in vision in AAION, frequently accompanied by general symptoms.
The VF exam plays a crucial role in diagnosing AION. In AAION, VF defects depend on the severity of optic nerve damage, with defects generally being more extensive and visual acuity more profoundly decreased compared to NAION. In NAION, visual acuity is often spared, and patients may present with a variety of VF defects, including the typical altitudinal defects. This differs from glaucoma, which usually presents as a gradual loss of peripheral vision before advancing to affect central vision in more severe cases [23]. Further, differences on OCT angiography (OCT-A) can be found between NAION and AAION. More specifically, analysis of choroidal vessel density (VD) is known to be more severely diminished in AAION when correcting for age [24].
Another difference is the presence of a relative afferent pupillary defect (RAPD), which appears to be more often observed in AION than in glaucoma with comparable VF defects. This is due to the asymmetric presentation of an AION compared with the often symmetric VF defects in glaucoma [25].
However, with IOP values in the normal range in particular, the clinical presentation of NAION in the chronic phase and NTG can be very much alike [3]. This may be explained by a common background of arterial hypotension and defective vascular autoregulation. Moreover, ischemic optic neuropathy and GON can occur superimposed in the same patient [26‒31].
In cases where the distinction between NTG and NAION is unclear based on clinical findings or where there is a suspicion of NAION on top of primary open-angle glaucoma (POAG) or NTG, additional diagnostic tests such as OCT and OCT-A could help in making a correct diagnosis. The aim of this paper was to summarize the differences in ocular diagnostic tests between glaucoma and longstanding AION.
Methods
A literature search (shown in Fig. 1) was conducted using MEDLINE (PubMed), Embase, Web of Science, and the Cochrane Library by two separate authors. The goal was to identify studies that made a direct comparison between glaucoma and AION. As such, the concepts “glaucoma” and “AION” were adjusted for each database. Besides database searching, gray literature provided additional references (Fig. 1: “other sources”). A detailed overview of the search query for each database is provided in the online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000535568). This resulted in 769 references.
The inclusion criteria for this study comprised cross-sectional studies that compared AION and glaucoma patients through the usage of imaging techniques. References were included from the year 2000 onward (date of search: March 3, 2022).
Language restriction was done to include only papers written in English. Reviews, abstracts, animal studies, and in vitro studies were excluded. Title and abstract screening resulted in 321 references. Any discrepancies in the included articles between the authors were resolved through discussion with a third author present. Afterward, full-text screening resulted in 32 references. Since 90% of the studies used OCT, we have decided to focus on this technology and report the findings of the other technologies separately (VF perimetry, Heidelberg retinal tomography [HRT], color fundus photography, scanning laser polarimetry [GDx], and color Doppler imaging [CDI]). Moreover, we have opted to report the findings regarding OCT-A in a separate section. The following data were extracted: author, year, number of patients and eyes, mean age of patients, female/male ratios, types of OCT devices, types of glaucoma or AION, type of analyzed parameters.
Results
A total of 3,307 eyes from 32 studies were considered in this systematic review. Twenty-eight studies included NAION patients, 2 included both NAION and AAION patients, and 2 studies did not differentiate between N- or AAION.
The main outcomes of the included studies are summarized in Table 1 together with patient characteristics, study design, diagnostic modality, and analyzed parameters. The included subjects show heterogeneity in mean age. Most studies accounted for potential confounding due to age, gender, and cardiovascular risk factors by multivariate regression or exclusion.
Characteristics, study design, diagnostic modality, analyzed parameters, and results of included studies
. | Author, year . | NGON . | GON . | N eyes (M/F) . | Age, mean ± SD, years . | Study design (LOE) . | Diagnostic modality/device . | Parameters analyzed . | Results . |
---|---|---|---|---|---|---|---|---|---|
ONH | |||||||||
1 | Braga, 2019 [32] | NAION | OAG | 25 NAION | 64.23±10.5 y (42–91) | Retrospective, observational, cross-sectional study | Spectralis | BMO area | The MRW in OAG eyes was significantly reduced compared with that of NAION and healthy control eyes. There was no significant difference in pRNFL thickness between NAION, CON, and GON. In non-glaucomatous optic disk atrophy, the MRW is frequently preserved, despite the presence of diffuse pRNFL loss. The MRW/pRNFLT ratio can differentiate between NAION and OAG. |
CON | 25 CON | BMO-MRW | |||||||
25 GON | RNFL | ||||||||
25 CO | |||||||||
2 | Fard, 2016 [33] | NAION | POAG | 32 POAG | POAG 65.5 (13.1) | Cross-sectional | Spectralis | ONH | Eyes with POAG had greater ALD and thinner LC than control eyes and NAION eyes in all regions of the ONH (p < 0.001 for both). There was a marked PT thinning in POAG eyes compared to control and NAION eyes (p < 0.001). LCT and ALD of NAION eyes were not different from their fellow eyes and control eyes. Although prelaminar thickness was thinner in NAION eyes compared to their fellow eyes (p = 0.005), it was thicker than in control eyes (p < 0.001). |
30 NAION | AION 58.4 (10.5) | ||||||||
29 control | control 63.4 (7.9) | ||||||||
3 | Jurišić, 2917 [12] | NAION | POAG | 40 NAION | / | Prospective | HRT II | GCC | VF defects such as decreased retinal sensitivity prevailed in the eyes with POAG, whereas in the eyes with NAION they were mostly manifested as concentrically narrowed VF and quadrant excesses. Topographic ONH alterations, examined by HRT II, showed the same number of sectors to be affected in the eyes with POAG and NAION. A larger number of sectors in the upper part of ONH were affected in the eyes with NAION. Optic disk morphology differed significantly between the eyes with POAG and NAION by a higher rate of NRR thinning and higher mean cup depth in the POAG group. |
40 POAG | ONH | ||||||||
4 | Lee, 2019 [16] | NAION | NTG | 12 NAION (2/10) | NAION: 55.2 6 9.8 | Retrospective, cross-sectional | Spectralis | BMO, HRW, MRW, VRW, vertical/horizontal thicknesses at knee of curve at rising curvature of the cup wall | Within the ONH, BMO-MRW, BMO-HRW, horizontal width at the knee of curve, and central PT showed significantly larger values in NAION compared to NTG (p < 0.05). The difference of NRR thickness between NAION and NTG increased in a centripetal manner, being maximum at the knee of curve. The mean HV ratio was 1.63 in NAION, 0.83 in NTG, and 1.06 in controls (p < 0.001). OCT showed disproportionately less altered PT in NAION. |
12 NTG (8/4) | NTG: 61.0 6 6.6 | ||||||||
12 control (4/8) | Control: 58.1 6 7.9 | ||||||||
5 | Saito, 2006 [14] | NAION | OAG | 33 NAION (17/16) | NAION: 61.8±10.4 | Cross-sectional | HRT II | ONH | The cup area, C/D ratio, and mean cup depth were significantly smaller, and the cup shape measure more negative, in the NAION eyes than in the OAG eyes (p < 0.001), whereas rim area was significantly greater (p < 0.001). Multivariate analyses showed that none of disk area, rim area, and mean cup depth in the NAION eyes and only rim area (p = 0.029) in the OAG eyes was significantly associated with MD. Ellipse average of RNFL thickness significantly correlated with MD in the NAION eyes (p = 0.045) and in the OAG eyes (p = 0.022). |
33 OAG (18/15) | OAG: 61.8±10.3 | GDx VCC | |||||||
ONH and RNFL | |||||||||
1 | Yang, 2013 [11] | NAION | GON (not further specified) | 15 NAION | NAION: 57.5±9.10 | Cross-sectional | FD OCT | RNFL | Glaucomatous eyes had the largest cup area and cup volume, and the smallest rim area, rim volume, and disk volume (p < 0.05). NAION eyes had the smallest cup area and cup volume (p < 0.05), but their rim area, rim volume, and disk volume were comparable to those of control eyes (p > 0.05). The C/D ratio was increased in glaucomatous eyes but reduced in NAION eyes compared with control eyes. Glaucomatous eyes had the greatest loss of RNFL thickness in the TU, ST, and TL regions (p < 0.05), whereas NAION eyes had the smallest RNFL thickness in the SN and NU regions (p < 0.05). The AROCs of the temporal, superior, and inferior RNFL in glaucomatous eyes were greater compared with that of the disk area (p < 0.05). In addition, the AROCs of the temporal, superior, and inferior RNFL were higher compared with that of nasal RNFL (p < 0.05). The AROCs of all parameters for NAION were not significantly different, with the exception of superior, nasal superior, and inferior temporal RNFL (p < 0.05). In conclusion, FDOCT is able to detect quantitative differences. |
26 GON | GON: 50.7±19.3 | ONH | |||||||
30 control | Control: 51.4±16.1 | ||||||||
2 | Thitiwichienlert, 2015 [34] | AION (not further specified) | POAG | 2 AION | NGON | Cross-sectional, case-control | Spectralis EDI | CLT | LTs of GON and non-GON group eyes were thinner than those of control group eyes (p < 0.01); LTs of GON group eyes were thinner than those of non-GON group eyes (p = 0.01). LTs of severe GON subgroup eyes were thinner than those of moderate and mild GON subgroup eyes (p < 0.001; p = 0.024, respectively). LTs of severe non-GON subgroup eyes were thinner than those of mild non-GON subgroup eyes (p = 0.002). |
Compressive | NTG | 9 ON | GON | ||||||
2 compressive | |||||||||
ON traumatic | 2 LOHN | Control | |||||||
1 traumatic | |||||||||
9 POAG | |||||||||
9 NTG | |||||||||
30 control | |||||||||
3 | Danesh-Meyer, 2010 [35] | NAION | OAG | OAG (103 persons, 152 eyes) | OAG 69.3 (11.2 y) | Observational, cross-sectional | HFA | ONH | With similar damage, OAG eyes had larger, deeper cups; smaller rims; more cup volume; and less rim volume (all p ≤ 0.001). There were differences in disk topography between NAION and AAION, but they were not consistent for both measures of damage. Disk area and MD were also significantly associated with many HRT parameters. RNFL thickness was greater at the same MD for both AAION and NAION compared with OAG. |
AAION | NAION (53 persons, 57 eyes) | NAION 60.3 (12.0) | Stratus | RNFL | |||||
AAION (18 persons, 20 eyes) | AAION 73.0 (7.3) | HRT2 or HRT3 | RGC | ||||||
4 | Garas, 2011 [36] | NAION | POAG | POAG 4/256 | 64.3±10.5 (26–95) | Cohort | RTVue-100 | RNFL | During the screening of this Caucasian population with 3.5% prevalence of optic neuropathy due to glaucoma and NA-AION, the RTVue-100 OCT was found to be useful for screening, with both RNFL and GCC parameters providing high accuracy and PLR values. |
NTG | NTG 2/256 | ONH | |||||||
NAION 3/256 | GCC | ||||||||
5 | Han, 2015 [9] | NAION | OAG | 17 NAION (8/9) | NAION: 60.0±9.6 | Prospective | Cirrus | RNFL | The MD and pattern standard deviation were not significantly different between the groups. In the affected eye, although the disk area was similar between the two groups (2.00±0.32 and 1.99±0.33 mm2, p = 0.586), the rim area of the OAG group was smaller than that of the NAION group (1.26±0.56 and 0.61±0.15 mm2, respectively, p < 0.001). RNFL asymmetry was not different between the two groups (p = 0.265), but the inferior RNFL thickness of both the affected and unaffected eyes was less in the OAG group than in the NAION group. In the analysis of optic disk morphology, both affected and unaffected eyes showed significant differences between two groups. |
26 OAG (15/11) | OAG 64.0±11.2 | ONH | |||||||
6 | Kim, 2020 [37] | NAION | NTG | 48 NAION (28/20) | NAION: 60.8±9.1 | Retrospective | Spectralis | RNFL | RNFL thicknesses of the matched affected sectors did not differ between the NAION and NTG groups (p = 0.347). LCD and the LCCI were significantly larger in the NTG group than in the NAION and healthy control groups at all seven planes (p < 0.001 each), but were comparable in the NAION and healthy control groups. The LCCI was larger in the affected than in the unaffected sector of NTG eyes (p = 0.010) but did not differ in NAION eyes (p = 1,000). LCD did not differ between affected and unaffected sectors in either NAION (p = 0.600) or NTG (p = 0.098) eyes. |
48 NTG (25/23) | NTG: 60.3±9.2 | ONH (LCI) | |||||||
48 control (20/28) | Control: 61.7±10.0 | ||||||||
7 | Leaney, 2020 [18] | ION (not further specified) | NTG | 22 ION | / | Multicenter cohort | Spectralis | MRW-BMO | The 5-fold cross-validated AUC for glaucoma versus nonglaucoma from logistic regression models using MRW-BMO values from all sectors was 0.95 (95% confidence interval: 0.86–1.00). |
ON | 14 ON | RNFL | |||||||
Compressive | 8 compressive | ||||||||
Drusen inherited | 4 Drusen | ||||||||
Nutritional | 3 inherited | ||||||||
traumatic | 2 nutritional | ||||||||
1 traumatic | |||||||||
27 NTG | |||||||||
8 | Lee, 2016 [16] | NAION | NTG | 21 NAION (13/8) | NAION: 61.6±10.6 | Prospective cross-sectional | Spectralis EDI | RNFL | In the sector-matched comparison, LCD was largest in NTG patients, followed by NAION patients, while PT was thinner in NTG patients than in NAION patients (all p < 0.001). NAION patients had a comparable LCD and a thinner PT relative to normal controls (p = 0.170 and <0.001, respectively). |
42 NTG (16/26) | Control: 60.0±9.7 | ONH | |||||||
42 control (22/20) | NTG: 60.0±9.7 | ||||||||
9 | Rebolleda, 2019 [21] | NAION | POAG | 23 NAION (11/12) | NAION: 68.6 (10.3) | Observational, cross-sectional | Spectralis | BMO-MRW | BMO diameter was significantly larger in primary POAG eyes than in control eyes (p = 0.02). LC and disk cup were deeper in eyes with primary POAG than both control and NAION eyes (p < 0.001). PT thickness was significantly thinner in primary POAG eyes than in NAION eyes (p < 0.001). LC was shallower in both NAION and unaffected fellow eyes compared to healthy eyes (p < 0.001 and p = 0.04, respectively). No differences were found in the optic disk area. |
17 fellow eyes (9/8) | Fellow eye: 68.6 (10.6) | RNFL optic disk area | |||||||
25 POAG (14/11) | NTG: 72.3 (9.8) | ||||||||
23 control (10/13) | Control: 68.6 (10.7) | ||||||||
10 | Resch, 2018 [13] | NAION | POAG | 20 NAION (12/8) | NAION: 66.8±8.3 | Cohort | Cirrus and Spectralis | RNFL | BMO-based OCT measurements of the ONH together with RNFL thickness might be an objective way to distinguish POAG from NAION and that none of the two OCT instruments used performed better in this respect. |
20 POAG (9/11) | POAG: 71.2±6.0 | BMO-MRW NRR | |||||||
20 control (11/9) | Control: 65.5±8.1 | ||||||||
RNFL | |||||||||
1 | Fard, 2016 [20] | NAION | POAG | 42 NAION | NAION 58.02 (8.83) | Cross-sectional, single center | Spectralis | RNFL | There was no significant difference in peripapillary RNFL, total macula, and outer region GCIPL thicknesses between the affected eyes of the patients with NAION and glaucoma patients. However, the inner region GCIPL was significantly thinner in NAION eyes compared to POAG eyes. |
42 NAION fellow eyes | POAG 64.59 (9.79) | GCIPL | |||||||
32 OAG (moderate to severe) | Control: 64.82 (8.37) | ||||||||
22 control | |||||||||
2 | Heo, 2016 [38] | NAION | POAG | 35 NAION | NAION: 63.6±8.6 | Retrospective, cross-sectional | Cirrus | RNFL | RNFL thickness of the nasal quadrant and superonasal sector was thinner in NAION. RNFL thickness of the inferotemporal sector was thinner in POAG. |
70 POAG | POAG: 64.4±9.0 | GCC | |||||||
3 | Horowitz, 2010 [19] | NAION | OAG | 18 NAION (10/8) | NAION: 64±11 | Retrospective, cross-sectional | Stratus | RNFL | The mean RNFL thickness in the quadrants corresponding to the affected hemifield in the NAION and glaucomatous eyes was not significantly different (p > 0.9), but the values for both were decreased compared to the control eyes (p < 0.0001). The mean RNFL thickness in the quadrant corresponding to the unaffected hemifield was significantly lower in the glaucomatous eyes (73.8±20.04 μ) than in the NAION eyes (96.6±23.32 μ, p = 0.023), and in both study groups compared to the controls (117.2±13.44 μ, p < 0.0001 for glaucomatous vs. control eyes, and p < 0.025 for NAION vs. control eyes). Smax/Tavg and Imax/Tavg of the quadrant corresponding to the unaffected hemifield had the strongest power to differentiate the two diseases (an AUC of 0.92). |
29 OAG (12/17) | OAG: 73±8 | ||||||||
4 | Lee, 2019 [10] | Compressive | NTG | 18 compressive (9/9) | Compressive: 60.3±10 | Retrospective | Cirrus | RNFL mGCIPL | 88.9% NAION (8/9) has positive TRS compared to 91.0 (62/67) in NTG. The TRS and the decision tree-based model both showed good diagnostic efficacy for application to the HD-OCT mGCIPL thickness map. In the case of mGCIPL thinning, the TRS can be a simple but useful tool for discriminating between glaucomatous and nonglaucomatous structural change. |
CNS inflammation | 11 CNS inflammation (2/9) | CNS inflammation: 52.8±10.3 | |||||||
NAION | 9 NAION (5/4) | NAION: 56.2±9.9 | |||||||
ON other | 16 ON (4/12) | ON: 52.8±16.1 | |||||||
19 other NGON (7/12) | Other NGON: 48.3±18.6 | ||||||||
67 NTG (20/47) | NTG: 65.8±12.2 | ||||||||
5 | Lee, 2017 [39] | NAION | POAG | 35 NAION | NAION (63.6±8.6) | Retrospective, cross-sectional | Cirrus | RNFL mGCIPL | Mean RNFL thicknesses at the 1 and 2 o’clock (superonasal) positions were thinner in NAION than in POAG (both p < 0.05). Mean RNFL thickness at 7 o’clock (inferotemporal) was thinner in POAG than in NAION (p = 0.001). Although there was no significant difference between NAION and POAG in average GCIPL thickness, all of the sectoral GCIPL thicknesses were thinner in NAION (all p < 0.05), except in the inferior and inferotemporal sectors. The ranges of the clock-hour RNFL with damage greater than the average RNFL thickness reduction, versus fellow eyes and control eyes, were 7 h in NAION and 4 h in POAG. |
35 NAION unaffected fellow eyes | NAION fellow eye (63.6±8.6) | ||||||||
70 POAG | POAG (64.4±9.0) | ||||||||
70 control | Control (64.4±9.0) | ||||||||
6 | Suh, 2011 [40] | NAION | OAG | 22 NAION (15/7) | NAION: 60.0 (8.3) | Prospective, cross-sectional | Cirrus | RNFL NRR | A significant linear rim-RNFL correlation was observed in global and all clock-hour sectors, except the 3-, 4-, and 9-o’clock sectors, in OAG (0.045 < r2 < 0.64, p < 0.05, respectively). All eyes with NAION were outside the 95% PI of the rim-RNFL correlation of OAG in at least 1 clock-hour sector in terms of clock-hour parameters, as compared with 63.6% of eyes in terms of global parameter. All NAION eyes (n = 21) with 7- or 11-o’clock involvement had a rim-RNFL correlation outside the 95% PI of OAG for corresponding clock-hour sectors. |
113 OAG (69/44) | OAG: 58.3 (10.5) | HRT II | |||||||
OCT-A | |||||||||
1 | Kim, 2021 [37] | NAION | NTG | 38 NAION (15/23) | NAION: 61.6±8.8 | Prospective | Spectralis Topcon OCT-A | VD and RNFL | VDs in the PLT and LC were lower in NTG eyes than in both NAION and healthy eyes (p ≤ 0.008), and did not differ between the NAION and healthy eyes. VDs in the PR did not differ between the NTG and NAION eyes. In intersectoral comparisons, VDs in the PLT (p = 0.030) and LC (p = 0.028) were lower in the affected than in the unaffected sector of eyes with NTG, but the differences did not occur in eyes with NAION. VD in the PR was lower in the affected than in the unaffected sector in both NTG and NAION eyes (both p < 0.001). |
38 NTG (24/14) | NTG: 58.1±11.4 | ||||||||
38 control (18/20) | Control: 58.4±9.8 | ||||||||
2 | Lee, 2021 [41] | NAION | NTG | 19 NAION (12/7) | NAION: 63.3±9.9 | Cross-sectional | Cirrus | RNFL | MvD was present in both diseases, marginally more frequently in NAION eyes (19/19, 100.0%) than in OAG eyes (38/47, 80.6%, p = 0.050), without a discernible difference in appearance. NAION eyes also showed wider MvD and RNFL defects compared to OAG eyes (both p < 0.001). In topographical measurements, the distribution of MvD showed a strong correspondence to superimposition areas of βPPA and RNFL defects, more distinctly than to RNFL defects (all p < 0.001). The outline of superimposition area also remarkably resembled the MvD area. |
POAG | 38 NTG | OAG: 61.1±10.7 | Topcon OCT-A | ONH | |||||
9 POAG | Microvascular dropout | ||||||||
PPA | |||||||||
3 | Shin, 2021 [42] | NAION | NTG | 27 NAION (16/11) | NAION: 65.2±7.8 | Retrospective, cross-sectional | Cirrus | RNFL pVD | CMvD was observed in 15 eyes (55.6%) of the NAION group and 20 (74.1%) of the NTG group. The area and angular width of CMvD were significantly greater in eyes with NAION (0.278±0.172 mm2 and 86.5±42.3) than in those with NTG (0.138±0.068 mm2 and 35.1±16.2 , p = 0.002 and p < 0.001, respectively). CMvD in eyes with NAION were distributed in 120–250 and most frequently located at the temporal region, while CMvD in eyes with NTG showed double peaks at 220–280° and 110–140° and most frequently located at the inferotemporal region. The factors associated with the discrimination of NAION from NTG were greater area of CMvD (OR, 1.181; 95% CI, 1.021–1.366; p = 0.025) and location closer to the temporal region of the CMvD (OR, 0.904; 95% CI, 0.838–0.975; p = 0.009). |
27 NTG (10/17) | NTG: 61.7±11.1 | AngioVue | cVD | ||||||
27 control (11/16) | Control: 61.6±7.1 | PPA area | |||||||
4 | Fard, 2020 [43] | NAION | POAG | 37 NAION | AION: 55.46 (11.38) | Comparative, cross-sectional study | Spectralis | RNFL | In NAION and POAG eyes with similar RNFL loss and consequently similar peripapillary superficial vascular density dropout in RPC layer (both large and small vessels), PPCMv differed significantly. Although inner annulus PPCMv density was lower in POAG eyes compared with NAION, outer annulus PPCMv was not different between NAION and POAG groups. |
34 NAION unaffected fellow eyes | POAG: 60.38 (12.27) | AngioVue | PPCMv | ||||||
47 moderate and severe POAG eyes | Control 55.27 (15.69) | ||||||||
54 healthy control | |||||||||
5 | Liu, 2017 [44] | NAION | OAG | 10 NAION (4/6) | NAION: 59.90±10.70 | Retrospective, cross-sectional | Avanti | RNFL | There was statistically significant difference in peripapillary wiVD, cpVD, macular wiVD, and pfVD between the three groups (p < 0.05 for all). In comparison between OAG and NAION groups, the NAION group demonstrated marked decrease in average cpVD (p = 0.008) and in most sectors of cpVD except the inferior one, while the OAG group demonstrated significant decreased macular wiVD and pfVD (p = 0.03 and 0.003, respectively). Multivariate analysis indicated that average thickness of RNFL was the only predictor for peripapillary wiVD and cpVD (p = 0.005 for both). By contrast, thickness of ganglion cell complex was the only predictor for macular wiVD (p = 0.007). |
16 OAG (9/7) | OAG: 53.75±10.21 | AngioVue | GCC | ||||||
27 control (14/13) | Control: 55.74±14.51 | VD | |||||||
FD | |||||||||
6 | Mastropasqua, 2018 [45] | NAION | NTG | 22 NAION (10/12) | NAION: 68.1±4.3 | Observational, case-control, single-center study | AngioVue | WPD | The WPD was 0.41±0.04 in the NTG group (p < 0.0001 in comparison with healthy subjects and NAION patients), 0.46±0.04 in the NAION group (p < 0.0001 in comparison with the control group), and 0.56±0.03 in the control group. |
22 NTG (9/13) | NTG: 66.3±7.0 | HFA | CPD | The CPD was significantly reduced in both NTG (0.48±0.04, p < 0.0001) and NAION eyes (0.52±0.05, p < 0.0001), after comparison to control eyes (0.59±0.03). Moreover, the CPD was significantly lower in NTG than in NAION eyes (p = 0.006). | |||||
23 control (12/11) | Control: 63.9±7.0 | RNFL | |||||||
VF MD | |||||||||
7 | Hondur, 2021 [46] | NAION | POAG | 21 NAION (12/9) | NAION: 61±7 | Cross-sectional | Spectralis | pVD | The POAG eyes had lower peripapillary VDs in all areas compared with the NAION eyes, which was most marked in the inferior and nasal sectors (p = 0.005 for both). RNFL loss was similar between the 2 groups in all areas, except for a preserved thickness in the inferior sector in NAION eyes (p = 0.01). Peripapillary VD demonstrated stronger correlations with global RNFL thickness in the peripapillary region in the NAION eyes compared with that of the POAG eyes (r = 0.91, p < 0.00001; r = 0.42, p = 0.03, respectively). In multivariate analysis, the peripapillary VD correlated with age and RNFL thickness in the POAG eyes, while it correlated with SSI and RNFL thickness in the NAION eyes. |
26 POAG (13/13) | POAG: 62±11 | AngioVue | RNFL | ||||||
30 control (17/13) | Control: 64±9 | ||||||||
8 | Fard, 2020 [47] | NAION | POAG | 37 moderate and advanced POAG | AION: 57.4 (12.2) | Prospective, cross-sectional | Spectralis | GCC | In NAION and POAG with similar RNFL and macular damage, macular OCT-A shows less involvement of superficial and deep vascular plexus in NAION in contrast to POAG, which might show a primary vascular insult in addition to secondary vascular damage due to ganglion cell damage. |
19 NAION | POAG: 63.6±10.4 | AngioVue | RNFL | ||||||
40 control | Control: 56.7±14.7 | VD | |||||||
9 | Fard, 2018 [48] | NAION | POAG | 31 NAION | AION: 54.11 (11) | Comparative, cross-sectional study | Spectralis | Whole-image and whole-annulus PCDs | While the whole PCD values were not different in chronic NAION and POAG, the greater correlation of inferior PCD with corresponding RNFL sectors in POAG compared to NAION suggests greater susceptibility of the inferior RPC in the pathogenesis of POAG. |
31 unaffected fellow eyes | POAG: 60.2 (8) | AngioVue | |||||||
42 POAG (12 moderate, 30 severe) | Control: 58.4 (10.3) | ||||||||
77 control | |||||||||
10 | Chen, 2017 [49] | AAION | POAG | 1 AAION | 57.5±15.2 | Retrospective, observational, cross-sectional study | Cirrus | pVD | All optic neuropathies showed a decrease in peripapillary VD on OCT-A, regardless of the etiology of the optic neuropathy. |
NAION | 3 NAION | AngioVue | Reductions in peripapillary capillary flow density in many optic neuropathies, including glaucoma, are secondary to optic nerve (and RNFL) damage, rather than the primary cause of the optic neuropathy. | ||||||
ON CON | 3 ON | ||||||||
1 CON | |||||||||
TON | 1 TON | ||||||||
1 GON | |||||||||
Doppler imaging | |||||||||
1 | Kuerten, 2019 [50] | NAION | NTG | 41 NAION | NAION: 67.03±10.35 | Prospective, monocenter, observational study | Doppler imaging using a 7.5 Mhz linear phase-arrayed transducer (Siemens, Sonoline, Sienna) | PSV, EDV | No significant differences were recorded for either age or IOP in between the 2 groups. Systolic blood pressure was significantly higher in the NAION group, whereas no significant differences were recorded for the diastolic blood pressure. Only 3 CDI parameters were found to differ significantly. The PSV (p < 0.005) and EDV (p < 0.02) in the CRA were significantly higher in NTG patients. Furthermore, the RI in the OA was significantly higher in the NAION patients (p < 0.005). |
. | Author, year . | NGON . | GON . | N eyes (M/F) . | Age, mean ± SD, years . | Study design (LOE) . | Diagnostic modality/device . | Parameters analyzed . | Results . |
---|---|---|---|---|---|---|---|---|---|
ONH | |||||||||
1 | Braga, 2019 [32] | NAION | OAG | 25 NAION | 64.23±10.5 y (42–91) | Retrospective, observational, cross-sectional study | Spectralis | BMO area | The MRW in OAG eyes was significantly reduced compared with that of NAION and healthy control eyes. There was no significant difference in pRNFL thickness between NAION, CON, and GON. In non-glaucomatous optic disk atrophy, the MRW is frequently preserved, despite the presence of diffuse pRNFL loss. The MRW/pRNFLT ratio can differentiate between NAION and OAG. |
CON | 25 CON | BMO-MRW | |||||||
25 GON | RNFL | ||||||||
25 CO | |||||||||
2 | Fard, 2016 [33] | NAION | POAG | 32 POAG | POAG 65.5 (13.1) | Cross-sectional | Spectralis | ONH | Eyes with POAG had greater ALD and thinner LC than control eyes and NAION eyes in all regions of the ONH (p < 0.001 for both). There was a marked PT thinning in POAG eyes compared to control and NAION eyes (p < 0.001). LCT and ALD of NAION eyes were not different from their fellow eyes and control eyes. Although prelaminar thickness was thinner in NAION eyes compared to their fellow eyes (p = 0.005), it was thicker than in control eyes (p < 0.001). |
30 NAION | AION 58.4 (10.5) | ||||||||
29 control | control 63.4 (7.9) | ||||||||
3 | Jurišić, 2917 [12] | NAION | POAG | 40 NAION | / | Prospective | HRT II | GCC | VF defects such as decreased retinal sensitivity prevailed in the eyes with POAG, whereas in the eyes with NAION they were mostly manifested as concentrically narrowed VF and quadrant excesses. Topographic ONH alterations, examined by HRT II, showed the same number of sectors to be affected in the eyes with POAG and NAION. A larger number of sectors in the upper part of ONH were affected in the eyes with NAION. Optic disk morphology differed significantly between the eyes with POAG and NAION by a higher rate of NRR thinning and higher mean cup depth in the POAG group. |
40 POAG | ONH | ||||||||
4 | Lee, 2019 [16] | NAION | NTG | 12 NAION (2/10) | NAION: 55.2 6 9.8 | Retrospective, cross-sectional | Spectralis | BMO, HRW, MRW, VRW, vertical/horizontal thicknesses at knee of curve at rising curvature of the cup wall | Within the ONH, BMO-MRW, BMO-HRW, horizontal width at the knee of curve, and central PT showed significantly larger values in NAION compared to NTG (p < 0.05). The difference of NRR thickness between NAION and NTG increased in a centripetal manner, being maximum at the knee of curve. The mean HV ratio was 1.63 in NAION, 0.83 in NTG, and 1.06 in controls (p < 0.001). OCT showed disproportionately less altered PT in NAION. |
12 NTG (8/4) | NTG: 61.0 6 6.6 | ||||||||
12 control (4/8) | Control: 58.1 6 7.9 | ||||||||
5 | Saito, 2006 [14] | NAION | OAG | 33 NAION (17/16) | NAION: 61.8±10.4 | Cross-sectional | HRT II | ONH | The cup area, C/D ratio, and mean cup depth were significantly smaller, and the cup shape measure more negative, in the NAION eyes than in the OAG eyes (p < 0.001), whereas rim area was significantly greater (p < 0.001). Multivariate analyses showed that none of disk area, rim area, and mean cup depth in the NAION eyes and only rim area (p = 0.029) in the OAG eyes was significantly associated with MD. Ellipse average of RNFL thickness significantly correlated with MD in the NAION eyes (p = 0.045) and in the OAG eyes (p = 0.022). |
33 OAG (18/15) | OAG: 61.8±10.3 | GDx VCC | |||||||
ONH and RNFL | |||||||||
1 | Yang, 2013 [11] | NAION | GON (not further specified) | 15 NAION | NAION: 57.5±9.10 | Cross-sectional | FD OCT | RNFL | Glaucomatous eyes had the largest cup area and cup volume, and the smallest rim area, rim volume, and disk volume (p < 0.05). NAION eyes had the smallest cup area and cup volume (p < 0.05), but their rim area, rim volume, and disk volume were comparable to those of control eyes (p > 0.05). The C/D ratio was increased in glaucomatous eyes but reduced in NAION eyes compared with control eyes. Glaucomatous eyes had the greatest loss of RNFL thickness in the TU, ST, and TL regions (p < 0.05), whereas NAION eyes had the smallest RNFL thickness in the SN and NU regions (p < 0.05). The AROCs of the temporal, superior, and inferior RNFL in glaucomatous eyes were greater compared with that of the disk area (p < 0.05). In addition, the AROCs of the temporal, superior, and inferior RNFL were higher compared with that of nasal RNFL (p < 0.05). The AROCs of all parameters for NAION were not significantly different, with the exception of superior, nasal superior, and inferior temporal RNFL (p < 0.05). In conclusion, FDOCT is able to detect quantitative differences. |
26 GON | GON: 50.7±19.3 | ONH | |||||||
30 control | Control: 51.4±16.1 | ||||||||
2 | Thitiwichienlert, 2015 [34] | AION (not further specified) | POAG | 2 AION | NGON | Cross-sectional, case-control | Spectralis EDI | CLT | LTs of GON and non-GON group eyes were thinner than those of control group eyes (p < 0.01); LTs of GON group eyes were thinner than those of non-GON group eyes (p = 0.01). LTs of severe GON subgroup eyes were thinner than those of moderate and mild GON subgroup eyes (p < 0.001; p = 0.024, respectively). LTs of severe non-GON subgroup eyes were thinner than those of mild non-GON subgroup eyes (p = 0.002). |
Compressive | NTG | 9 ON | GON | ||||||
2 compressive | |||||||||
ON traumatic | 2 LOHN | Control | |||||||
1 traumatic | |||||||||
9 POAG | |||||||||
9 NTG | |||||||||
30 control | |||||||||
3 | Danesh-Meyer, 2010 [35] | NAION | OAG | OAG (103 persons, 152 eyes) | OAG 69.3 (11.2 y) | Observational, cross-sectional | HFA | ONH | With similar damage, OAG eyes had larger, deeper cups; smaller rims; more cup volume; and less rim volume (all p ≤ 0.001). There were differences in disk topography between NAION and AAION, but they were not consistent for both measures of damage. Disk area and MD were also significantly associated with many HRT parameters. RNFL thickness was greater at the same MD for both AAION and NAION compared with OAG. |
AAION | NAION (53 persons, 57 eyes) | NAION 60.3 (12.0) | Stratus | RNFL | |||||
AAION (18 persons, 20 eyes) | AAION 73.0 (7.3) | HRT2 or HRT3 | RGC | ||||||
4 | Garas, 2011 [36] | NAION | POAG | POAG 4/256 | 64.3±10.5 (26–95) | Cohort | RTVue-100 | RNFL | During the screening of this Caucasian population with 3.5% prevalence of optic neuropathy due to glaucoma and NA-AION, the RTVue-100 OCT was found to be useful for screening, with both RNFL and GCC parameters providing high accuracy and PLR values. |
NTG | NTG 2/256 | ONH | |||||||
NAION 3/256 | GCC | ||||||||
5 | Han, 2015 [9] | NAION | OAG | 17 NAION (8/9) | NAION: 60.0±9.6 | Prospective | Cirrus | RNFL | The MD and pattern standard deviation were not significantly different between the groups. In the affected eye, although the disk area was similar between the two groups (2.00±0.32 and 1.99±0.33 mm2, p = 0.586), the rim area of the OAG group was smaller than that of the NAION group (1.26±0.56 and 0.61±0.15 mm2, respectively, p < 0.001). RNFL asymmetry was not different between the two groups (p = 0.265), but the inferior RNFL thickness of both the affected and unaffected eyes was less in the OAG group than in the NAION group. In the analysis of optic disk morphology, both affected and unaffected eyes showed significant differences between two groups. |
26 OAG (15/11) | OAG 64.0±11.2 | ONH | |||||||
6 | Kim, 2020 [37] | NAION | NTG | 48 NAION (28/20) | NAION: 60.8±9.1 | Retrospective | Spectralis | RNFL | RNFL thicknesses of the matched affected sectors did not differ between the NAION and NTG groups (p = 0.347). LCD and the LCCI were significantly larger in the NTG group than in the NAION and healthy control groups at all seven planes (p < 0.001 each), but were comparable in the NAION and healthy control groups. The LCCI was larger in the affected than in the unaffected sector of NTG eyes (p = 0.010) but did not differ in NAION eyes (p = 1,000). LCD did not differ between affected and unaffected sectors in either NAION (p = 0.600) or NTG (p = 0.098) eyes. |
48 NTG (25/23) | NTG: 60.3±9.2 | ONH (LCI) | |||||||
48 control (20/28) | Control: 61.7±10.0 | ||||||||
7 | Leaney, 2020 [18] | ION (not further specified) | NTG | 22 ION | / | Multicenter cohort | Spectralis | MRW-BMO | The 5-fold cross-validated AUC for glaucoma versus nonglaucoma from logistic regression models using MRW-BMO values from all sectors was 0.95 (95% confidence interval: 0.86–1.00). |
ON | 14 ON | RNFL | |||||||
Compressive | 8 compressive | ||||||||
Drusen inherited | 4 Drusen | ||||||||
Nutritional | 3 inherited | ||||||||
traumatic | 2 nutritional | ||||||||
1 traumatic | |||||||||
27 NTG | |||||||||
8 | Lee, 2016 [16] | NAION | NTG | 21 NAION (13/8) | NAION: 61.6±10.6 | Prospective cross-sectional | Spectralis EDI | RNFL | In the sector-matched comparison, LCD was largest in NTG patients, followed by NAION patients, while PT was thinner in NTG patients than in NAION patients (all p < 0.001). NAION patients had a comparable LCD and a thinner PT relative to normal controls (p = 0.170 and <0.001, respectively). |
42 NTG (16/26) | Control: 60.0±9.7 | ONH | |||||||
42 control (22/20) | NTG: 60.0±9.7 | ||||||||
9 | Rebolleda, 2019 [21] | NAION | POAG | 23 NAION (11/12) | NAION: 68.6 (10.3) | Observational, cross-sectional | Spectralis | BMO-MRW | BMO diameter was significantly larger in primary POAG eyes than in control eyes (p = 0.02). LC and disk cup were deeper in eyes with primary POAG than both control and NAION eyes (p < 0.001). PT thickness was significantly thinner in primary POAG eyes than in NAION eyes (p < 0.001). LC was shallower in both NAION and unaffected fellow eyes compared to healthy eyes (p < 0.001 and p = 0.04, respectively). No differences were found in the optic disk area. |
17 fellow eyes (9/8) | Fellow eye: 68.6 (10.6) | RNFL optic disk area | |||||||
25 POAG (14/11) | NTG: 72.3 (9.8) | ||||||||
23 control (10/13) | Control: 68.6 (10.7) | ||||||||
10 | Resch, 2018 [13] | NAION | POAG | 20 NAION (12/8) | NAION: 66.8±8.3 | Cohort | Cirrus and Spectralis | RNFL | BMO-based OCT measurements of the ONH together with RNFL thickness might be an objective way to distinguish POAG from NAION and that none of the two OCT instruments used performed better in this respect. |
20 POAG (9/11) | POAG: 71.2±6.0 | BMO-MRW NRR | |||||||
20 control (11/9) | Control: 65.5±8.1 | ||||||||
RNFL | |||||||||
1 | Fard, 2016 [20] | NAION | POAG | 42 NAION | NAION 58.02 (8.83) | Cross-sectional, single center | Spectralis | RNFL | There was no significant difference in peripapillary RNFL, total macula, and outer region GCIPL thicknesses between the affected eyes of the patients with NAION and glaucoma patients. However, the inner region GCIPL was significantly thinner in NAION eyes compared to POAG eyes. |
42 NAION fellow eyes | POAG 64.59 (9.79) | GCIPL | |||||||
32 OAG (moderate to severe) | Control: 64.82 (8.37) | ||||||||
22 control | |||||||||
2 | Heo, 2016 [38] | NAION | POAG | 35 NAION | NAION: 63.6±8.6 | Retrospective, cross-sectional | Cirrus | RNFL | RNFL thickness of the nasal quadrant and superonasal sector was thinner in NAION. RNFL thickness of the inferotemporal sector was thinner in POAG. |
70 POAG | POAG: 64.4±9.0 | GCC | |||||||
3 | Horowitz, 2010 [19] | NAION | OAG | 18 NAION (10/8) | NAION: 64±11 | Retrospective, cross-sectional | Stratus | RNFL | The mean RNFL thickness in the quadrants corresponding to the affected hemifield in the NAION and glaucomatous eyes was not significantly different (p > 0.9), but the values for both were decreased compared to the control eyes (p < 0.0001). The mean RNFL thickness in the quadrant corresponding to the unaffected hemifield was significantly lower in the glaucomatous eyes (73.8±20.04 μ) than in the NAION eyes (96.6±23.32 μ, p = 0.023), and in both study groups compared to the controls (117.2±13.44 μ, p < 0.0001 for glaucomatous vs. control eyes, and p < 0.025 for NAION vs. control eyes). Smax/Tavg and Imax/Tavg of the quadrant corresponding to the unaffected hemifield had the strongest power to differentiate the two diseases (an AUC of 0.92). |
29 OAG (12/17) | OAG: 73±8 | ||||||||
4 | Lee, 2019 [10] | Compressive | NTG | 18 compressive (9/9) | Compressive: 60.3±10 | Retrospective | Cirrus | RNFL mGCIPL | 88.9% NAION (8/9) has positive TRS compared to 91.0 (62/67) in NTG. The TRS and the decision tree-based model both showed good diagnostic efficacy for application to the HD-OCT mGCIPL thickness map. In the case of mGCIPL thinning, the TRS can be a simple but useful tool for discriminating between glaucomatous and nonglaucomatous structural change. |
CNS inflammation | 11 CNS inflammation (2/9) | CNS inflammation: 52.8±10.3 | |||||||
NAION | 9 NAION (5/4) | NAION: 56.2±9.9 | |||||||
ON other | 16 ON (4/12) | ON: 52.8±16.1 | |||||||
19 other NGON (7/12) | Other NGON: 48.3±18.6 | ||||||||
67 NTG (20/47) | NTG: 65.8±12.2 | ||||||||
5 | Lee, 2017 [39] | NAION | POAG | 35 NAION | NAION (63.6±8.6) | Retrospective, cross-sectional | Cirrus | RNFL mGCIPL | Mean RNFL thicknesses at the 1 and 2 o’clock (superonasal) positions were thinner in NAION than in POAG (both p < 0.05). Mean RNFL thickness at 7 o’clock (inferotemporal) was thinner in POAG than in NAION (p = 0.001). Although there was no significant difference between NAION and POAG in average GCIPL thickness, all of the sectoral GCIPL thicknesses were thinner in NAION (all p < 0.05), except in the inferior and inferotemporal sectors. The ranges of the clock-hour RNFL with damage greater than the average RNFL thickness reduction, versus fellow eyes and control eyes, were 7 h in NAION and 4 h in POAG. |
35 NAION unaffected fellow eyes | NAION fellow eye (63.6±8.6) | ||||||||
70 POAG | POAG (64.4±9.0) | ||||||||
70 control | Control (64.4±9.0) | ||||||||
6 | Suh, 2011 [40] | NAION | OAG | 22 NAION (15/7) | NAION: 60.0 (8.3) | Prospective, cross-sectional | Cirrus | RNFL NRR | A significant linear rim-RNFL correlation was observed in global and all clock-hour sectors, except the 3-, 4-, and 9-o’clock sectors, in OAG (0.045 < r2 < 0.64, p < 0.05, respectively). All eyes with NAION were outside the 95% PI of the rim-RNFL correlation of OAG in at least 1 clock-hour sector in terms of clock-hour parameters, as compared with 63.6% of eyes in terms of global parameter. All NAION eyes (n = 21) with 7- or 11-o’clock involvement had a rim-RNFL correlation outside the 95% PI of OAG for corresponding clock-hour sectors. |
113 OAG (69/44) | OAG: 58.3 (10.5) | HRT II | |||||||
OCT-A | |||||||||
1 | Kim, 2021 [37] | NAION | NTG | 38 NAION (15/23) | NAION: 61.6±8.8 | Prospective | Spectralis Topcon OCT-A | VD and RNFL | VDs in the PLT and LC were lower in NTG eyes than in both NAION and healthy eyes (p ≤ 0.008), and did not differ between the NAION and healthy eyes. VDs in the PR did not differ between the NTG and NAION eyes. In intersectoral comparisons, VDs in the PLT (p = 0.030) and LC (p = 0.028) were lower in the affected than in the unaffected sector of eyes with NTG, but the differences did not occur in eyes with NAION. VD in the PR was lower in the affected than in the unaffected sector in both NTG and NAION eyes (both p < 0.001). |
38 NTG (24/14) | NTG: 58.1±11.4 | ||||||||
38 control (18/20) | Control: 58.4±9.8 | ||||||||
2 | Lee, 2021 [41] | NAION | NTG | 19 NAION (12/7) | NAION: 63.3±9.9 | Cross-sectional | Cirrus | RNFL | MvD was present in both diseases, marginally more frequently in NAION eyes (19/19, 100.0%) than in OAG eyes (38/47, 80.6%, p = 0.050), without a discernible difference in appearance. NAION eyes also showed wider MvD and RNFL defects compared to OAG eyes (both p < 0.001). In topographical measurements, the distribution of MvD showed a strong correspondence to superimposition areas of βPPA and RNFL defects, more distinctly than to RNFL defects (all p < 0.001). The outline of superimposition area also remarkably resembled the MvD area. |
POAG | 38 NTG | OAG: 61.1±10.7 | Topcon OCT-A | ONH | |||||
9 POAG | Microvascular dropout | ||||||||
PPA | |||||||||
3 | Shin, 2021 [42] | NAION | NTG | 27 NAION (16/11) | NAION: 65.2±7.8 | Retrospective, cross-sectional | Cirrus | RNFL pVD | CMvD was observed in 15 eyes (55.6%) of the NAION group and 20 (74.1%) of the NTG group. The area and angular width of CMvD were significantly greater in eyes with NAION (0.278±0.172 mm2 and 86.5±42.3) than in those with NTG (0.138±0.068 mm2 and 35.1±16.2 , p = 0.002 and p < 0.001, respectively). CMvD in eyes with NAION were distributed in 120–250 and most frequently located at the temporal region, while CMvD in eyes with NTG showed double peaks at 220–280° and 110–140° and most frequently located at the inferotemporal region. The factors associated with the discrimination of NAION from NTG were greater area of CMvD (OR, 1.181; 95% CI, 1.021–1.366; p = 0.025) and location closer to the temporal region of the CMvD (OR, 0.904; 95% CI, 0.838–0.975; p = 0.009). |
27 NTG (10/17) | NTG: 61.7±11.1 | AngioVue | cVD | ||||||
27 control (11/16) | Control: 61.6±7.1 | PPA area | |||||||
4 | Fard, 2020 [43] | NAION | POAG | 37 NAION | AION: 55.46 (11.38) | Comparative, cross-sectional study | Spectralis | RNFL | In NAION and POAG eyes with similar RNFL loss and consequently similar peripapillary superficial vascular density dropout in RPC layer (both large and small vessels), PPCMv differed significantly. Although inner annulus PPCMv density was lower in POAG eyes compared with NAION, outer annulus PPCMv was not different between NAION and POAG groups. |
34 NAION unaffected fellow eyes | POAG: 60.38 (12.27) | AngioVue | PPCMv | ||||||
47 moderate and severe POAG eyes | Control 55.27 (15.69) | ||||||||
54 healthy control | |||||||||
5 | Liu, 2017 [44] | NAION | OAG | 10 NAION (4/6) | NAION: 59.90±10.70 | Retrospective, cross-sectional | Avanti | RNFL | There was statistically significant difference in peripapillary wiVD, cpVD, macular wiVD, and pfVD between the three groups (p < 0.05 for all). In comparison between OAG and NAION groups, the NAION group demonstrated marked decrease in average cpVD (p = 0.008) and in most sectors of cpVD except the inferior one, while the OAG group demonstrated significant decreased macular wiVD and pfVD (p = 0.03 and 0.003, respectively). Multivariate analysis indicated that average thickness of RNFL was the only predictor for peripapillary wiVD and cpVD (p = 0.005 for both). By contrast, thickness of ganglion cell complex was the only predictor for macular wiVD (p = 0.007). |
16 OAG (9/7) | OAG: 53.75±10.21 | AngioVue | GCC | ||||||
27 control (14/13) | Control: 55.74±14.51 | VD | |||||||
FD | |||||||||
6 | Mastropasqua, 2018 [45] | NAION | NTG | 22 NAION (10/12) | NAION: 68.1±4.3 | Observational, case-control, single-center study | AngioVue | WPD | The WPD was 0.41±0.04 in the NTG group (p < 0.0001 in comparison with healthy subjects and NAION patients), 0.46±0.04 in the NAION group (p < 0.0001 in comparison with the control group), and 0.56±0.03 in the control group. |
22 NTG (9/13) | NTG: 66.3±7.0 | HFA | CPD | The CPD was significantly reduced in both NTG (0.48±0.04, p < 0.0001) and NAION eyes (0.52±0.05, p < 0.0001), after comparison to control eyes (0.59±0.03). Moreover, the CPD was significantly lower in NTG than in NAION eyes (p = 0.006). | |||||
23 control (12/11) | Control: 63.9±7.0 | RNFL | |||||||
VF MD | |||||||||
7 | Hondur, 2021 [46] | NAION | POAG | 21 NAION (12/9) | NAION: 61±7 | Cross-sectional | Spectralis | pVD | The POAG eyes had lower peripapillary VDs in all areas compared with the NAION eyes, which was most marked in the inferior and nasal sectors (p = 0.005 for both). RNFL loss was similar between the 2 groups in all areas, except for a preserved thickness in the inferior sector in NAION eyes (p = 0.01). Peripapillary VD demonstrated stronger correlations with global RNFL thickness in the peripapillary region in the NAION eyes compared with that of the POAG eyes (r = 0.91, p < 0.00001; r = 0.42, p = 0.03, respectively). In multivariate analysis, the peripapillary VD correlated with age and RNFL thickness in the POAG eyes, while it correlated with SSI and RNFL thickness in the NAION eyes. |
26 POAG (13/13) | POAG: 62±11 | AngioVue | RNFL | ||||||
30 control (17/13) | Control: 64±9 | ||||||||
8 | Fard, 2020 [47] | NAION | POAG | 37 moderate and advanced POAG | AION: 57.4 (12.2) | Prospective, cross-sectional | Spectralis | GCC | In NAION and POAG with similar RNFL and macular damage, macular OCT-A shows less involvement of superficial and deep vascular plexus in NAION in contrast to POAG, which might show a primary vascular insult in addition to secondary vascular damage due to ganglion cell damage. |
19 NAION | POAG: 63.6±10.4 | AngioVue | RNFL | ||||||
40 control | Control: 56.7±14.7 | VD | |||||||
9 | Fard, 2018 [48] | NAION | POAG | 31 NAION | AION: 54.11 (11) | Comparative, cross-sectional study | Spectralis | Whole-image and whole-annulus PCDs | While the whole PCD values were not different in chronic NAION and POAG, the greater correlation of inferior PCD with corresponding RNFL sectors in POAG compared to NAION suggests greater susceptibility of the inferior RPC in the pathogenesis of POAG. |
31 unaffected fellow eyes | POAG: 60.2 (8) | AngioVue | |||||||
42 POAG (12 moderate, 30 severe) | Control: 58.4 (10.3) | ||||||||
77 control | |||||||||
10 | Chen, 2017 [49] | AAION | POAG | 1 AAION | 57.5±15.2 | Retrospective, observational, cross-sectional study | Cirrus | pVD | All optic neuropathies showed a decrease in peripapillary VD on OCT-A, regardless of the etiology of the optic neuropathy. |
NAION | 3 NAION | AngioVue | Reductions in peripapillary capillary flow density in many optic neuropathies, including glaucoma, are secondary to optic nerve (and RNFL) damage, rather than the primary cause of the optic neuropathy. | ||||||
ON CON | 3 ON | ||||||||
1 CON | |||||||||
TON | 1 TON | ||||||||
1 GON | |||||||||
Doppler imaging | |||||||||
1 | Kuerten, 2019 [50] | NAION | NTG | 41 NAION | NAION: 67.03±10.35 | Prospective, monocenter, observational study | Doppler imaging using a 7.5 Mhz linear phase-arrayed transducer (Siemens, Sonoline, Sienna) | PSV, EDV | No significant differences were recorded for either age or IOP in between the 2 groups. Systolic blood pressure was significantly higher in the NAION group, whereas no significant differences were recorded for the diastolic blood pressure. Only 3 CDI parameters were found to differ significantly. The PSV (p < 0.005) and EDV (p < 0.02) in the CRA were significantly higher in NTG patients. Furthermore, the RI in the OA was significantly higher in the NAION patients (p < 0.005). |
LCT, lamina cribrosa thickness; RPC, radial peripapillary capillary; pfVD, perifoveal vessel density; CRA, central retinal artery; TU, temporal upper; ST, superior temporal; TL, temporal lower; SN, superior nasal; NU, nasal upper; AROCs, areas under the receiver operator characteristic curve; wiVD, whole-image vessel density; cpVD, circumpapillary vessel density; mGCIPL, macular ganglion cell-inner plexiform layer.
Optical Coherence Tomography
ONH Morphology
The ONH is generally defined as the structure in the posterior pole from which the ganglion cell axons exit, and the blood vessels enter and exit the eye. The measured parameters of the ONH are the dimensions of the optic disk and the cup (disk area, cup area, cup-to-disk ratio (C/D ratio), cup volume, rim volume, rim area, and cup depth). In the past, fundoscopy and fundus photography were the conventional methods for assessing these parameters. However, recent technological advancements have established confocal laser ophthalmoscopy (HRT) and OCT as the contemporary gold standard for objectively measuring the ONH and analyzing its progression.
In general, glaucomatous eyes had larger disks (disk area), with a larger cup area, higher C/D ratio, higher cup volume, lower (vertical) rim volume, smaller rim area, and a greater cup depth compared to AION [9, 11, 17, 21, 40]. Evaluation of the usefulness of the RTVue-100 OCT in screening for glaucoma and NAION found that the ONH parameters (cup area, C/D ratio, and rim area) showed a high specificity (89.0–97.6%) and accuracy (87.8–94.5%) but a positive likelihood ratio (PLR) of <8.3. Although this did not allow for differentiation between glaucoma and AION, it still demonstrates the potential of ONH morphology analysis [36].
Analysis of the deeper lying structures of the ONH (Fig. 2), such as the lamina cribrosa (LC) and Bruch’s membrane opening (BMO), best evaluated through OCT imaging, might prove valuable in the differential diagnosis between GON and AION. Research into the deeper optic disk parameters (Fig. 3 for overview of parameters) has shown that glaucomatous eyes had greater LC depth (LCD) than control and NAION eyes [16, 21, 33, 37]. In NAION eyes, most studies found a comparable [16, 33, 37] or even a shallower [21] LCD compared to healthy controls. The central LC thickness of glaucomatous eyes and AION eyes both show greater thinning compared to the control group, whereby the lowest values could be found in the glaucoma group [33, 34]. The LC curvature index (LCCI) is another parameter to analyze the LC morphology and, similarly to the LCD, was significantly greater in the glaucomatous group than in NAION and healthy control eyes [37]. The prelaminar tissue (PT), which is portion of the ONH located anterior to the LC, is comprised of bundles of retinal ganglion cell (RGC) axons, astrocytes, capillaries, and extracellular material. Several studies demonstrated a lower PT thickness in glaucomatous eyes compared to both NAION and controls [16, 21, 33]. Another study demonstrated that there is indeed sparing of the PT in NAION eyes when compared to NTG [51].
Downs, J. Crawford et al. “The Optic Nerve Head as a Biomechanical Structure” (2010).
Downs, J. Crawford et al. “The Optic Nerve Head as a Biomechanical Structure” (2010).
ONH parameters measured by OCT (B-scan images). a (1) ILM (green line); (2) ALCS (yellow dashed line); (3) BMO (blue dot); (4) BMO-MRW (red arrow) is the minimum distance from BMO to ILM; (5) PT thickness (white line) is measured from ILM to ALCS. b (6) BMO reference line (blue dashed line); (7) PLCS (orange dashed line); (8) LCT (black double-headed arrow) is the distance between ALCS and PLCS; (9) LCD (white arrow) is the maximum vertical distance from the BMO reference line to the ALCS. c Same B-scan as in (b): (10) LC curve depth (yellow double-headed arrow); (11) width of the ALCS reference line (red double-headed arrow); (12) LCCI is measured (as Kim et al. [37] 2020 proposed) by dividing the LC curve depth within the BMO by the width of the ALCS reference line and then multiplying the ratio by 100. ALCS, anterior lamina cribrosa surface; BMO, Bruch’s membrane opening; BMO-MRW, Bruch’s membrane opening-minimum rim width; ILM, internal limiting membrane; LC, lamina cribrosa; LCCI, lamina cribrosa curvature index; LCD, lamina cribrosa depth; LCT, lamina cribrosa thickness; PLCS, posterior lamina cribrosa surface; PT, prelaminar tissue.
ONH parameters measured by OCT (B-scan images). a (1) ILM (green line); (2) ALCS (yellow dashed line); (3) BMO (blue dot); (4) BMO-MRW (red arrow) is the minimum distance from BMO to ILM; (5) PT thickness (white line) is measured from ILM to ALCS. b (6) BMO reference line (blue dashed line); (7) PLCS (orange dashed line); (8) LCT (black double-headed arrow) is the distance between ALCS and PLCS; (9) LCD (white arrow) is the maximum vertical distance from the BMO reference line to the ALCS. c Same B-scan as in (b): (10) LC curve depth (yellow double-headed arrow); (11) width of the ALCS reference line (red double-headed arrow); (12) LCCI is measured (as Kim et al. [37] 2020 proposed) by dividing the LC curve depth within the BMO by the width of the ALCS reference line and then multiplying the ratio by 100. ALCS, anterior lamina cribrosa surface; BMO, Bruch’s membrane opening; BMO-MRW, Bruch’s membrane opening-minimum rim width; ILM, internal limiting membrane; LC, lamina cribrosa; LCCI, lamina cribrosa curvature index; LCD, lamina cribrosa depth; LCT, lamina cribrosa thickness; PLCS, posterior lamina cribrosa surface; PT, prelaminar tissue.
Another parameter that can be considered is the BMO-minimum rim width (BMO-MRW). The BMO is the anatomical border of the optic rim, through which RGC axons travel, whereas the BMO-MRW represents the neuroretinal rim (NRR) [13, 32]. Several findings indicate that NAION eyes had a thicker BMO-MRW in contrast to glaucomatous eyes which showed a greater thinning of the BMO-MRW [13, 18, 32, 51]. The lowest MRW values could be found in the temporal [13, 32] and inferotemporal sectors [18] in glaucomatous eyes. Consequently, the MRW/RNFL (including global, nasal, and temporal) ratios were higher in NAION eyes than in glaucomatous or healthy eyes [18, 32].
Retinal Nerve Fiber Layer
Thinning of the RNFL is observed in both glaucomatous and chronic AION eyes when compared to healthy control eyes [11, 13, 19, 20, 32‒34]. In early/preperimetric glaucoma, the RNFL thinning is observed in the superotemporal and inferotemporal region while moderate glaucoma also involved the temporal RNFL. Finally, in severe glaucoma, the nasal RNFL also shows thinning [9, 11, 38, 39]. In AION, on the other hand, the acute phase shows diffuse thickening of the RNFL which over the ensuing weeks makes room for the thinning of the RNFL in the affected segment of the optic disk [52].
Depending on the region that is being examined, subtle differences between glaucomatous and AION eyes are identified. Horowitz et al. [19] found a significantly higher RNFL thickness in the temporal sector in moderate to severe glaucomatous eyes compared to NAION eyes. Other studies reported a more diffuse thinning of the RNFL in glaucomatous eyes, whereas NAION eyes showed more focal thinning [10, 19], especially in the superonasal sector [11, 38, 39] and focal preservation of the RNFL in the inferior sector [46]. Subsequently, NAION eyes seem to have a tendency to cause inferior altitudinal field defects [9, 19]. Braga et al. [32] however did not find a significant difference in focal or global RNFL thinning between NAION and OAG which they attribute to not differentiating for severity of VFDs.
RNFL thickness parameters (average, superior, and inferior) measured by RTVue-100 OCT in screening for glaucoma and NAION showed a specificity of 99.2–99.6%, accuracy of 96.5–96.9%, and PLR of 27.3–54.7 but a sensitivity of 11.1–22.2%, whereas sectoral RNFL thickness showed a specificity of 89.0%, accuracy of 87.5%, PLR of 4.1, and sensitivity of 44.4% [36]. In addition, another study demonstrated that the average RNFL thickness in the quadrant corresponding to the unaffected hemifield was significantly lower in glaucomatous eyes compared to NAION eyes [9].
Combined Measurements
Suh et al. [40] investigated if comparison of the rim-RNFL correlation, described as the correlation between NRR area (measured by HRT II) of the ONH and peripapillary RNFL thickness (determined by Cirrus OCT), could differentiate between glaucoma and NAION. They demonstrated a significant linear rim-RNFL correlation in glaucomatous eyes at global and in all regions except for nasal (3 o’clock), inferonasal (4 o’clock), and temporal (9 o’clock) [40]. Furthermore, all NAION eyes were outside of the 95% prediction interval (PI) for the rim-RNFL correlation of OAG in at least 1 clock-hour sector, as compared to 63.6% of eyes in terms of global parameters. Also, all NAION eyes with inferotemporal (7 o’clock) or superotemporal (11 o’clock) involvement displayed a rim-RNFL correlation outside the 95% PI of glaucoma for corresponding sectors [40].
Macula
Fard et al. [20] assessed the inner and outer region of the macular ganglion cell plus inner plexiform layer (mGCIPL) and total macular thickness by OCT (Spectralis) in POAG and NAION eyes with matched mean deviation (MD) on VF testing. NAION eyes showed greater loss of parafoveal (inner region) GCIPL tissue thickness compared to POAG, after adjusting for age, sex, and MD of the VF [20]. Total macula and outer region GCIPL however did not significantly differ between affected NAION and glaucomatous eyes [20]. Lee et al. [10, 39] also examined differences in GCIPL by OCT (Cirrus). NAION and POAG eyes showed no significant differences in average GCIPL thickness, but all of the sectoral GCIPL thicknesses were thinner in NAION, except in the inferior and inferotemporal sectors [39]. The latter study matched the macular parameters with age and peripapillary RNFL compared to a more logically MD-matched MD on VF in the study from Fard et al. [20, 39]. This difference might explain for the difference in results between the two [10, 39].
Scanning of the macular ganglion cell complex (GCC) by RTVue-100 OCT in screening for glaucoma and NAION showed a specificity of 95.9%, accuracy of 94.0%, PLR of 10.7, and a sensitivity of 44.4% [36]. Furthermore, Lee et al. [10] examined the potential of the temporal raphe sign (TRS), described as positive in patients in whom there was a straight line longer than one-half of the length between the inner and outer annulus in the temporal elliptical area of the mGCIPL thickness, for discriminating GON from non-GON (NGON) in eyes with mGCIPL thinning. They designed a decision tree-based model that revealed two things: “(1) if the TRS is positive and the RAPD is absent, the case should be diagnosed as GON; (2) if the TRS is absent regardless of the presence or absence of the RAPD, or both the TRS and RAPD are present, the case should be diagnosed as NGON” [10]. Analysis of the NGON subgroups showed that 88.9% (8/9) of NAION patients presented with a positive TRS and thus would possibly be misdiagnosed as GON. However, because the researchers added RAPD to the decision tree, discrimination between GON and NAION was correctly done [10].
Optical Coherence Tomography Angiography
Measurement of VD was performed using OCT-A. Compared to control eyes, both AION and glaucomatous eyes show significant decrease in the peripapillary VD [43‒45, 48, 49]. However, this can be found in all optic neuropathies with RNFL thinning and VFD, regardless of the etiology of the optic neuropathy [49]. This is due to the peripapillary VD being strongly correlated with RNFL thickness and has shown similar measurements in both POAG and AION eyes [43, 48, 53].
However, other studies report statistically lower peripapillary VD in OAG when compared to NAION [45, 46] and one previous study showed a significant thinning in average peripapillary VD and most sectors of the VD in NAION eyes except for the inferior sector in comparison with OAG [44]. There is however evidence for significantly lower peripapillary VD in affected than in unaffected sectors in both NTG and NAION [53].
Moreover, NAION and POAG eyes with similar RNFL loss showed similar dropout in peripapillary VD yet showed a significant difference in choroidal microvasculature [43]. The parapapillary choroidal microvasculature (PPCMv) density in POAG patients had a significantly reduced PPCMv density in both inner and outer annuli (respectively 1 mm and 2 mm around optic disk edge) compared to the normal subjects and the NAION group [43]. Mean inner and outer annular region PPCMv densities in the NAION group did not differ from unaffected fellow eyes and the control group [43]. When comparing the peripapillary choroidal microvascular dropout (CMvD) between NAION and OAG [41] and NTG [42], the area and angular width of CMvD were significantly greater in NAION eyes [41, 42]. Furthermore, they showed that the area of CMvD in NAION was most frequently located at the temporal region [42].
When examining VD of the ONH itself, there was a significantly lower VD in the PT and LC in NTG eyes when compared to NAION and healthy controls [53]. When looking at the macular VD, POAG eyes showed a significant decrease in superficial macular VD compared to AION with similar VF damage [43, 44, 47]. Conversely, when looking at the parafoveal superficial VD, the lowest values could also be found in the glaucomatous group [43, 44, 47]. More specifically, Liu et al. [44] showed a significantly decreased macular whole-image VD and perifoveal VD in glaucoma eyes when compared to NAION eyes. These findings are supported by Fard et al., who presented a significantly lower macular superficial whole-image VD in POAG compared to NAION eyes, and sparing of the parafoveal superficial VD in NAION eyes in contrast to POAG [43, 47]. Deep macular vasculature examination was possible after projection artifact removal and showed that whole-image and parafoveal deep macular vessels in glaucomatous eyes were more affected than in NAION and control eyes [43, 47]. Especially temporal and inferior parafoveal deep vascularity zones were more susceptible to glaucomatous damage than observed in NAION eyes [43, 47].
Other Examination Modalities
Three studies used different diagnostic techniques. Two of them used VF perimetry, HRT, color fundus photography, and GDx [12, 14], and another study used CDI to examine hemodynamic parameters [50].
Two studies examined morphological ONH differences by HRT II and concluded that NAION eyes showed significantly lower cup area, C/D ratio, and mean cup depth, whereas NRR and rim area were significantly greater compared to OAG eyes [12, 14]. A larger number of sectors affected in the upper part of ONH in NAION than in POAG eyes were also identified. Additionally to OCT, Suh et al. [40] also used HRT to examine the ONH morphology and found analogous results: cup area, C/D ratio, cup volume, and cup depth were significantly greater and rim area and rim volume significantly lower in OAG than in NAION eyes, without significant differences in optic disk size (area). Danesh-Meyer et al. [17] made this comparison as well, but with an additional subdivision between AAION and NAION eyes. Cup area, C/D ratio, cup volume, and cup depth were significantly lower in NAION, followed by AAION, and the largest values were detected in glaucomatous eyes, while rim area was significantly smaller in OAG, followed by AAION and largest in NAION. Rim volume was significantly greater and disk area significantly smaller in NAION compared to OAG and AAION eyes [17].
Jurišić et al. [12] compared 40 randomly selected POAG patients with 40 NAION patients. They identified that the type of VFD (measured by Humphrey VF analyzer) in OAG eyes was predominantly generalized decreased retinal sensitivity in POAG, whereas the NAION group mostly showed concentrically narrowed VFs and quadrant irregularities. RNFL examination by GDx did not show significant differences between NAION and OAG, matched for age and VFD [14].
The third study used CDI to examine hemodynamic parameters between NTG and acute NAION and found that three parameters differed significantly: the peak systolic velocity (PSV), end-diastolic velocity (EDV), and resistive index (RI = (PSV-EDV)/PSV) [50]. The RI in the ophthalmic artery (OA) of NAION eyes was significantly higher than in NTG eyes (0.81 ± 0.06 vs. 0.77 ± 0.08; p < 0.005), and there was a significantly lower PSV and EDV in the central retinal artery in NAION eyes compared to NTG eyes (7.05 ± 1.92 vs. 8.27 ± 1.59; p < 0.005, and 1.99 ± 0.52 vs. 2.31 ± 0.73; p < 0.02, respectively) [50].
Discussion
The clinical differential diagnosis between a longstanding AION and a GON can be challenging. Moreover, a superimposed (subclinical) AION cannot always be excluded in glaucomatous neuropathy such as NTG with sudden deterioration of VF [27‒29]. This review highlights the characteristics in technical diagnostic tests which can aid in differentiating between a GON and ischemic optic neuropathy.
OCT allows a detailed assessment of the ONH and RNFL and has become the most widely used method for optic neuropathy diagnosis [32]. This appears to be valid regardless of the OCT device, given both Cirrus (Zeiss) and Spectralis (Heidelberg) OCT measurements of ONH parameters and RNFL thickness showed similar performance for discrimination between primary POAG and NAION [13]. However, OCT findings in glaucomatous-like optic neuropathies can still be inconclusive as to which diseases are responsible, so that clinical examination is still mandatory [54]. Therefore, OCT is not a substitute for clinical examination but an added value in discrimination between GON and NGON [54].
Optic Nerve Head
Clear differences in ONH parameters were identified through OCT (and HRT) imaging: glaucomatous eyes had a greater cup area, C/D ratio, cup volume, and cup depth and a lower (vertical) rim volume and rim area compared to AION eyes [9, 11, 12, 14, 17, 21, 40]. Danesh-Meyer et al. [17] presented a hypothesis that both OAG and AION result in damage to the prelaminar and laminar zone of the ONH but at a different degree of severity and with a different preference for damaging specific structures. The laminar region, through which the axon bundles pass, contains the supporting connective tissue of the lamina, and the greater cup depth and volume in OAG (at the same degree of RGC axon loss as in AION) imply that the supporting connective tissues of the lamina are retro-displaced and/or thinned much more in OAG than in AION [12, 17]. Furthermore, differences in structure of extracellular matrix that could affect laminar region stability were also cited as a possible explanation for the observed differences in predilection site of ONH damage between glaucoma and AION [12].
The exact reason why sparing of the rim volume and rim area occurs in AION was not identified [12]. However, relative preservation of the NRR in AION eyes (when compared to OAG at similar RNFL thinning or MD loss) may manifest as more noticeable optic disk pallor and RGC loss in AION eyes [17]. In addition, AAION eyes were found to have significantly higher cup volumes than NAION eyes and consequently mimic glaucoma even more [17]. This finding might be explained by differences in pathologic events, a different baseline optic disk structure and possibly also a different composition of laminar connective tissue including a greater resistance to deformation in NAION eyes which could explain the smaller disk diameter in NAION eyes [12, 17]. Also, it was reported that AAION is associated with an occlusion of the major arteries of the retina, choroid, and ciliary, whereas the vascular insult in NAION seemed to be limited to the ONH [17].
LCD appears to be mainly associated with IOP stress and consequently ONH deformation and displacement of the LC [16, 21, 33, 34]. Studies indeed demonstrated a thinner and more posteriorly located LC (greater LCD) in POAG eyes compared to control and NAION eyes in all ONH regions [21, 33]. In both POAG and NAION eyes, two forces that result in sclera and BMO expansion are proposed: one pushing the LC backward and compressing the PT, and another one related to the hoop stress (inner circumferential pressure supported by the peripapillary sclera) [21]. Mechanical stress thus plays an important role in the pathogenesis, but both diseases remain multifactorial [16, 34, 37]. Other factors include interindividual variability (e.g., the geometry of the eyeball), vascular factors, an abnormally low cerebrospinal fluid pressure in NTG eyes resulting in higher translaminar pressure gradient which could induce posterior LC displacement, a hypercompliant LC in NTG compared to NAION, inflammation, infiltration, toxicity, heredity factors, and a possibly innate deeper LC in NTG eyes compared to healthy and NAION eyes [16, 34, 37].
Another parameter to analyze LC morphology is the LCCI which could be given the preference over LC given that it could be affected by choroidal thickness [37]. Analogous to LCD, LCCI is significantly greater in NTG eyes than in NAION eyes [37].
Analysis of the PT, the anterior part of the LC, has shown significantly greater thickness in NAION compared to glaucomatous eyes. The explanation is fourfold.
- (1)
The chronic effect of an increased IOP on the PT in POAG eyes resulting in mechanical compression, which is absent in NAION eyes [21].
- (2)
ONH conformation in response to IOP fluctuations [21].
- (3)
Higher degree of reactive gliosis in NAION eyes and sparing of glial cells during an ischemic insult [10, 16].
- (4)
Innate and premorbid factors [10, 16].
These findings are also supported by evaluation of the microvasculature of the ONH, where a significantly lower VD was found in the PT and LC in NTG eyes compared to NAION eyes and healthy controls [53]. BMO-MRW was introduced since it was suggested that this parameter could provide a direct estimation of the degree of NRR changes and thus evaluate RNFL thickness [10, 13]. The BMO-MRW showed a high level of sensitivity and specificity to differentiate between GON and NGON and was therefore recommended to be used in routine examination [18].
Retinal Nerve Fiber Layer
The RNFL, being closely related to the MRW, does show significant differences in injury to ischemic or non-ischemic insult. Comparison of the MRW-RNFL correlation using a 95% PI was suggested by Suh et al. as a method to differentiate between glaucoma and NAION [40]. In a study with 22 NAION compared to 113 OAG eyes, all NAION eyes had an MRW-RNFL correlation outside of the 95% prediction [40]. Braga et al. [32] added to this by showing that an MRW/RNFL thickness ratio ≥3.80 had a sensitivity of 88% and a specificity of 88% for the diagnosis of NAION. However, Leaney et al. [18] noted that the lower prevalence of NGON creates more false-positive results in contrast to the higher prevalence of GON that creates more false-negative results. The significant differences in MRW/RNFL thickness ratio between glaucoma and AION do indicate that a relative preservation of the MRW in the presence of RNFL loss should highly suggest NGON [32].
Most studies indicated that RNFL thinning in glaucomatous eyes occurred primarily at the superotemporal and inferotemporal regions [9, 11, 38, 39]. NAION on the other hand mostly showed thinning at the superonasal region [11, 32, 38, 39]. As explained by Jurišić et al. [12], a large part of the vascular supply of the ONH passes vertically through the temporal part of the disk and is more sensitive to ischemia when perfusion pressure drops in POAG eyes. The altitudinal asymmetry can be explained according to the disease mechanism [39]. In NAION eyes, the vascular supply (circle of Zinn-Haller) derived from the short posterior ciliary arteries is divided into superior and inferior halves [10, 12, 39]. Consequently, the perfusion deficit is consistent with the observed altitudinal damage, either superior or inferior [10, 12, 39]. Note that ischemic events in NAION seem to occur more frequently in the superior region with consequently more inferior altitudinal field defects [9, 19, 39].
Macula
In general, significantly more (inner sectoral) GCIPL thinning in NAION compared to glaucomatous eyes was found, except in the inferotemporal sector [10, 20, 39]. In this sector, similar GCIPL thinning between glaucoma and AION was demonstrated and although not significant, slightly more thinning was noted in glaucomatous eyes [10, 20, 39]. This can be explained by the finding that papillomacular nerve fiber bundles are likely to be generally damaged in NAION eyes and therefore more general GCIPL thinning could be detected, whereas they are supposedly spared in glaucomatous eyes and only become affected at a later stage [20, 39]. However, the inferior region is often affected earlier and consequently similarly involved in both glaucomatous and AION eyes [20, 39]. In addition, the presented combination of TRS and RAPD in the decision tree-based model provides the possibility to discriminate between glaucoma and AION in eyes with GCIPL thinning [10]. This finding emphasizes the importance of also performing a complete neuro-ophthalmological examination, including color vision testing, pupillary light reflexes (swinging flashlight test), eye movement testing, and VF testing [3, 10].
Vasculature
In both OAG and AION, a decrease in peripapillary VD compared to control eyes was detected through OCT-A but the results of the comparisons between glaucomatous and AION eyes varied [44, 45, 47‒49, 53]. The decrease in VD can be due to ischemia. However, the reduced radial peripapillary capillary blood flow can be the result of autoregulation, whereby (chronic) optic neuropathy can lead to neural loss which in turn causes a reduction of metabolic need and leads to a decrease in blood flow [44, 45, 47‒49, 53]. Interestingly, ocular blood flow deficit may be more important in NTG than high-tension glaucoma since the vascular damage was found to be greater and related to a global alteration of the retinal and optic nerve flow in NTG [45]. Furthermore, Liu et al. [44] explained that RGCs undergo a stage of reversible dysfunction when exposed to a stressful environment. This period before autoregulatory mechanisms fail to sustain normal RGC function may be relatively longer in glaucoma than in AION [44]. Therefore, the impairment of vascular perfusion can be less severe in glaucomatous eyes; however, this also depends on time of inclusion in a study (time after insult) [44, 49]. Lastly, the decreased retinal VD could represent the closure or degeneration of capillaries that occurs along with the RNFL loss rather than primary reduction of retinal perfusion [44, 48].
In addition, the group of Fard et al. [43] identified greater PPCMv loss in POAG, indicating that while the superficial parapapillary vessel loss could be secondary to microvascular changes from RNFL loss, reduced parapapillary deep choroidal vasculature in POAG eyes could contribute primarily to axonal damage in patients with glaucoma. On the other hand, sparing of the PPCMv in NAION may indicate involvement of the short posterior ciliary vessels and little to no involvement of the choroidal vasculature. Moreover, Fard et al. suggested that a thick choroid can possibly predispose to NAION by increasing PT thickness that might be compressed by the deeper PPCMv and that thick PT may contain a component of crowded optic disks [43]. However, the authors highlight that OCT-A may lack sensitivity to detect possible PPCMv changes in NAION eyes, whereas the thick choroid may in turn affect OCT-A measurements and signals. This is because initially, en face imaging of the choriocapillaris was made, followed by calculation of the PPCMv using specific MATLAB software after shadow removal of the large retinal vessels [43].
Analysis of peripapillary CMvD between NAION and OAG [41] and NTG [42] revealed that the area and angular width of CMvD were significantly greater in NAION eyes [41, 42] and the area of CMvD in NAION was most frequently located at the temporal region [42]. A significant decrease in superficial macular VD in glaucoma compared to AION with similar degree of VFD was found [44, 47, 55]. Early macular damage appears to occur often in glaucomatous eyes, so that superficial macular VD thinning may already be more pronounced when compared to NAION eyes [44]. In contrast, the reason why superficial macular VD injury is less prominent in NAION eyes seems less understood, especially since NAION eyes show similar GCC thinning as glaucomatous eyes [44]. The authors hypothesized that in chronic NAION eyes, the inner retinal layer in the macular area could be mainly affected but the middle and outer retinal layers can remain stable and thereby maintain autoregulation of macular blood flow [44]. However, this does not clarify why there is greater macular VD damage in glaucoma than in NAION [44]. Fard et al. [47] further explained that at a given amount of GCC loss, glaucomatous eyes displayed more VD loss than NAION and suggested that therefore, reduced ocular blood flow could play a greater role than supposedly secondary microvascular changes following GCC loss.
Likewise, deep macular VD was significantly more affected in glaucomatous eyes compared to AION eyes [47, 55]. Fard et al. [47] explained that deep vessels represent anastomotic capillary networks supplied by vertically oriented interconnecting vessels derived from the superficial plexus. Because of the greater impairment of this superficial plexus in POAG eyes, compensatory mechanisms such as autoregulation could possibly fail to ensure sufficient blood flow in the deep vascular plexus.
Next, CDI examination by Kuerten et al. [50] suggested that the significantly higher RI in the OA in NAION patients compared to NTG may be a result of higher overall resistance in the peripheral vessels due to arterial hypertension. However, the authors point out that previously published literature [56, 57] reported ambiguous results. One study found no significant difference between the RI in the OA between patients with hypertension and control, nor a significant difference between the RI in the OA in hypertensive patients with retinopathy compared to those without retinopathy [56]. However, glaucomatous eyes were excluded from the study. The other study found a significantly higher RI in patients with hypertension compared to control and a reduction in the RI of the OA after treatment with blood pressure-lowering medication [57]. Note that AAION and untreated high-pressure glaucomatous eyes were excluded from this study. Because no direct comparison between glaucoma and AION was made in both studies, their results remain inconclusive for this review. Furthermore, the significant reduction of PSV and EDV in the central retinal artery in NAION patients may as well indicate an impaired volumetric blood flow because of swelling of the optic disk in acute NAION eyes [50].
Conclusion
In this review, we identified differences in ocular diagnostic tests that can help differentiate between glaucoma and AION. Table 2 provides an overview of the main conclusions. OCT imaging to assess the ONH morphology, RNFL, and macula seems essential, especially when IOP is at a normal range. Further research into ONH parameters such as LCD/LCCI and the MRW-RNFL ratio could lead to practical differentiation between glaucoma and AION. More recently and still actively being developed is the analysis of VD of both peripapillary and macula regions where clear differences can be seen between glaucoma (NTG more specifically) and NAION.
Key messages
Key messages . | |
---|---|
Measure structural parameters with OCT | |
ONH | |
Cup area, C/D ratio, cup volume, and cup depth | Glaucoma > AION |
Rim area and rim volume | Glaucoma < AION |
BMO-MRW and MRW/pRNFL ratios | Glaucoma < AION |
LCD, LCCI | Glaucoma > AION |
LCT, PT | Glaucoma < AION |
RNFL thickness | |
Superotemporal, inferotemporal | Glaucoma < AION |
Superonasal | Glaucoma > AION |
Macula | |
Overall mGCIPL | Glaucoma > AION |
Infero(temporal)* mGCIPL | Glaucoma < AION |
Measure vascular parameters with OCT-A | |
Superficial and deep macular VD | Glaucoma < AION |
Parapapillary choroidal VD | Glaucoma < AION |
Key messages . | |
---|---|
Measure structural parameters with OCT | |
ONH | |
Cup area, C/D ratio, cup volume, and cup depth | Glaucoma > AION |
Rim area and rim volume | Glaucoma < AION |
BMO-MRW and MRW/pRNFL ratios | Glaucoma < AION |
LCD, LCCI | Glaucoma > AION |
LCT, PT | Glaucoma < AION |
RNFL thickness | |
Superotemporal, inferotemporal | Glaucoma < AION |
Superonasal | Glaucoma > AION |
Macula | |
Overall mGCIPL | Glaucoma > AION |
Infero(temporal)* mGCIPL | Glaucoma < AION |
Measure vascular parameters with OCT-A | |
Superficial and deep macular VD | Glaucoma < AION |
Parapapillary choroidal VD | Glaucoma < AION |
OCT, optical coherence tomography; ONH, optic nerve head; BMO, Bruch’s membrane opening; BMO-MRW, Bruch’s membrane opening-minimum rim width; LCD, lamina cribrosa depth; LCCI, lamina cribrosa curvature index; LCT, lamina cribrosa thickness; PT, prelaminar tissue; RNFL, retinal nerve fiber layer; mGCIPL, macular ganglion cell-inner plexiform layer; OCT-A, OCT angiography; VD, vessel density.
Limitations
This review has several limitations. First, because of the restriction to the English literature, selection bias cannot be eliminated. Second, several factors were not considered when comparing the results, including the distinction between NTG and other types of glaucoma, as well as the possible influence of systemic medication use such as IOP-lowering medication or antihypertensives. Lastly, heterogeneity in study design can influence results of this review. Not all studies accounted for the difference in disease severity when comparing AION to glaucoma. In an ideal setup, all studies would have a VF (or MD)-matched comparison to more accurately compare both disease entities.
Statement of Ethics
An ethics statement is not applicable because this study is based exclusively on published literature.
Conflict of Interest Statement
All authors report no conflicts of interest.
Funding Sources
All authors have no funding to disclose.
Author Contributions
Lemmens, Breda, and Stalmans initiated and organized the study. Smeets and Margot performed literature review as stated in the methods section, adhering to the PRISMA guidelines. Margot wrote the first draft of the manuscript which was finished by Smeets. All the authors revised the manuscript and approved the final version of the manuscript.
Data Availability Statement
All data generated or analyzed during this study are included in this article and its online supplementary material files. Further inquiries can be directed to the corresponding author.