Optical coherence tomography angiography in acute unilateral nonarteritic anterior ischemic optic neuropathy: A comparison with the fellow eye and with eyes with papilledema.

Purpose: The purpose of this study is to detect the optic nerve head (ONH) and peripapillary perfusion in eyes with acute nonarteritic anterior ischemic optic neuropathy (NAION) compared to the fellow normal eyes using optical coherence tomography angiography (OCTA) and to compare with nonischemic disc edema (papilledema).
Methods: Retrospective analysis of patients with unilateral NAION who underwent OCTA was performed. All patients underwent comprehensive ocular examination including visual field testing. ONH was imaged using 6 mm × 6 mm scan by Topcon DRI Triton® OCT system. Vascularity loss was analyzed using ImageJ software in diseased eyes in comparison to normal fellow eyes and eyes with papilledema.
Results: Twenty-one patients (15 males, 6 females) with unilateral NAION and 9 patients (18 eyes) with papilledema were included in the study. In eyes with NAION, two distinct patterns of loss of vasculature were noted - (a) diffuse loss of microvasculature cuff and vascular network around the optic disc in all the patients (100%) and (b) additional area of sectoral loss of vasculature extending from the disc in 12 of the 21 eyes (57.14%). All 18 eyes with papilledema showed loss of the microvasculature cuff; however, none showed the focal pattern of vascular defect. The mean area of the peripapillary vascular zone in eyes with NAION was significantly lesser than that in normals. Of the 12 eyes with NAION with focal loss of vasculature, 11 correlated with visual field defects (91.6%).
Conclusion: Deficient peripapillary choroidal vasculature is present in NAION and has a different pattern than in nonischemic disc edema and can cause corresponding visual field deficits.

Nonarteritic anterior ischemic optic neuropathy (NAION) is the most common optic neuropathy (other than glaucoma) beyond the age of 50 years. The pathogenesis of NAION is related to vascular dysfunction and is thought to be a result of the occlusion of the short posterior ciliary arteries.[1] Fundus fluorescein angiography (FFA) studies performed earlier have shown the peripapillary choroidal circulation to be largely intact in these patients;[2] however, recent studies on choroidal architecture have shown differences in the choroidal thickness in patients with NAION as compared to normal controls, with both thinning and pachychoroid being reported,[34] suggesting that choroidal vasculature may have a role in the pathogenesis of this entity.

While FFA can provide information about the superficial vasculature of the optic nerve head (ONH), it does not allow sufficient visualization of the deeper vessels. Optical coherence tomography angiography (OCTA) is a relatively new, noninvasive investigative modality, which allows visualization of ONH vasculature at multiple levels. There have been a few studies describing vascular changes in and around ONH following NAION using OCTA,[567] the study groups were not homogeneous in these studies , and also, a comparison with disc edema due to a nonischemic cause is still lacking.

The present study aims to study the vasculature of the optic nerve and the peripapillary choroidal vasculature in acute unilateral NAION and to compare it with fellow normal eyes and other patients with papilledema.

Methods

This retrospective observational study was done in a period of 4 months, from September 2016 to December 2016, at a tertiary eye care institute in India. The research protocols were approved by the Institutional Review Board. A written consent was obtained from all the participants, and the study adhered to the tenets of the Declaration of Helsinki.

Patients with sudden drop in visual acuity in one eye with a clinical diagnosis of acute NAION, with normal fellow eyes (except for the presence of the typical “disc at risk”), were enrolled in the study. Any patient with a previous history of NAION in the same or other eye or with any other ocular or neurological pathology which could contribute to visual field defects in either eye was excluded from the study. Patients with a diagnosis of any other optic neuropathy were also excluded from the study. The unaffected normal eyes were used as controls, and the angiography findings were compared to those seen in patients with papilledema. The papilledema group consisted of patients with bilateral optic disc edema due to raised intracranial pressure, with 20/20 vision, normal color vision, and no visual field defects other than an enlarged blind spot.

A swept source-optical coherence tomography-Topcon DRI OCT Triton plus (Topcon Corporation, Tokyo, Japan) coupled with the noninvasive OCT angiography technology (SS-OCT Angio™) was used to obtain the images. The Topcon SS-OCT uses a tunable laser as a light source to provide a 1050 nm-centered wavelength. The device reaches a scanning speed of 100,000 A-scans per second. This device allows deeper penetration into the choroid and hence better visualization of its microvasculature.

After dilating the pupils (with 1% tropicamide and 2.5% phenylephrine), participants were seated in front of the OCT scanner, and their heads were stabilized with the aid of both a supporting chin rest and a forehead rest. Participants were directed to focus their gaze on the internal fixation target, and a real-time en face view was used by the operator to visualize the imaging area on the fundus. 6.0 mm × 6.0 mm scans with a resolution of 320 × 320 were obtained at a speed of 100,000 A-scan/s. To accustom the patient to the process, the normal fellow eye with better vision was scanned first followed by the involved eye. Repeated scans were taken till a good quality image was obtained for both the eyes. Two such set of images were obtained and were analyzed.

OCTA image analysis: all the images were analyzed by a single investigator (PPC). Images of poor quality with signal strength <40 or with residual motion artifacts such as discontinuous vessels or disc boundary were excluded from further analysis. ImageJ software was used to analyze the filtered images. Each OCT angiography report comprised four 6 mm × 6 mm OCTA images of the optic disc and the surrounding peripapillary area for both the involved and the fellow normal eye in patients with NAION and in both the eyes in patients with papilledema. Cross sections at the peripapillary choroidal level were obtained in every angiogram report in which the peripapillary microvasculature was seen clearly. A qualitative analysis for any focal/complete drop or loss of microvasculature around the disc in each sector was carried out in comparison to the normal fellow eye. The filtered images were analyzed using ImageJ software. The software was used to standardize images, and then, a manual analysis of the vascular zone was done. To analyze the peripapillary vascular zone, a circle was drawn outlining the disc and a second circle was drawn at the edge of the visible peripapillary zone (which is seen as a whitish vascular area around the disc). A difference between the areas of the two circular zones was considered to be the area of the peripapillary vascular zone [Fig. 1a and b]. This area was calculated separately for the eyes with NAION, fellow normal eyes, and those with papilledema. Any focal defects were also manually delineated, and the area was calculated using ImageJ software (version 1.51n).

Figure 1

Images of optical coherence tomography angiography used for ImageJ analysis in an eye with nonarteritic anterior ischemic optic neuropathy (a) and in the normal fellow eye (b) showing a circle drawn around the disc and then a second circle drawn to delineate the peripapillary vascular zone

All patients underwent Humphrey visual field (HVF) testing for both the eyes. The patients with visual acuity worse than 20/200 in the involved eye could not be reliably tested for visual fields and hence were not included for the analysis. A Humphrey Field Analyzer II (Carl Zeiss Meditec, Inc) was used with the system set for 30-2 threshold test, size III white stimulus, and standard Swedish interactive threshold algorithm (SITA algorithm).

The fields were analyzed by a second investigator (UG), blinded to the clinical findings, to avoid bias in view of possible correlation of the OCT and HVF findings. The reliability of the fields was judged as per the standard available guidelines. Each field printout was read to look for the pattern of field defects such as altitudinal defects, typically seen in NAION. Fields with generalized depression with field loss in all the quadrants and no specific pattern were labeled as advanced field loss.

Statistical analysis

The basic statistical analysis was done using Microsoft SPSS software. Chi-square test was used to determine statistical significance, and P < 0.05 was considered statistically significant.

Results

Twenty-one patients (15 males, 6 females) with unilateral NAION with normal fellow eyes were included as per the inclusion criteria. Mean age of the study population was 53.13 (±10.9) years. All the patients complained of sudden, monocular vision loss, and the mean duration of presentation was 15.4 (±11.7) days. Seventeen eyes showed diffuse, total disc edema and four had sectoral disc edema. Twelve eyes had altitudinal visual field defect (ten inferior altitudinal field defects and two superior altitudinal defects) and eight had generalized field loss, which did not fit into a specific pattern. One patient with diffuse disc edema showed only few depressed points inferiorly not enough to classify it as an inferior altitudinal defect.

Eighteen eyes (of nine patients) with papilledema were included for comparison. All these patients were clinically diagnosed to have papilledema. These patients underwent neuroimaging – MRI (magnetic resonance imaging) of the brain with an MR venography study. After ruling out any intracranial space-occupying lesions, a lumbar puncture was done to note the opening pressure of the cerebrospinal fluid (CSF) and also routine CSF analysis was carried out. The mean CSF opening pressure was found to be 33.2 cm of water, thus confirming the diagnosis of idiopathic intracranial hypertension. OCTA scan was performed before any treatment was started.

The OCT angiography scans were analyzed with specific attention to the peripapillary microvascular network at the choroidal level. In the normal eyes, a dense peripapillary microvasculature was noted in all the layers of the angiogram scan. Moreover, a microvasculature “cuff” was evident at the choroid level [Fig. 2a and d]. When compared to the normal eye, vascularity loss was noted in the involved eye in all the patients (100%). Moreover, the “cuff” showed definite thinning as compared to the fellow normal eye. We observed two distinct patterns of loss or distortion of vasculature – one comprised a diffuse loss of the microvasculature cuff and a loss of vascular network seen around the optic disc [Fig. 2b and e], which was noted in the involved eye in all the patients (100%). The second was an additional area of sectoral loss of vasculature extending from the disc [Fig. 2c and f], which was noted in 12 of the 21 (57.14%) involved eyes and in none of the normal eyes.

Figure 2

Disc photograph (a) and optical coherence tomography angiography (d) of the normal, uninvolved eye. Disc photograph (b) and optical coherence tomography angiography (e) of an eye with nonarteritic anterior ischemic optic neuropathy showing diffuse loss of the microvasculature cuff (yellow arrows) Disc photograph (c) and optical coherence tomography angiography (f) of an eye with nonarteritic anterior ischemic optic neuropathy showing focal loss of peripapillary vasculature (yellow arrows)

All 18 eyes with papilledema showed loss of the microvasculature cuff as seen in the eyes with NAION; however, none of the eyes showed the focal pattern of vascular defect [Fig. 3a–d].

Figure 3

Disc photographs (a and c) and optical coherence tomography angiography (b and d) of a patient with papilledema showing diffuse 360° distortion and loss of the peripapillary microvasculature cuff

We correlated the field defects in patients with NAION with the pattern of microvasculature loss as seen on OCT angiography. We found that 11 of the 21 eyes (52.38%) showed a correlation between the area of loss of choroidal vasculature and the corresponding field defect. All of these showed focal loss of choroidal vasculature as described above, i.e., of the 12 eyes which showed a focal loss of vasculature, 11 correlated with the visual field (91.6%), for example, if a focal loss of vasculature was seen in the superior peripapillary region, a corresponding inferior altitudinal defect was noted [Fig. 4].

Figure 4

Disc photograph (a) of an eye with nonarteritic anterior ischemic optic neuropathy and optical coherence tomography angiography (b) showing superior focal loss of microvasculature (yellow arrows) and a corresponding inferior visual field defect (c)

We calculated the mean area of the vascular zone in each group separately, i.e., those eyes with NAION, the fellow normal eyes, and those with papilledema. The mean area of the peripapillary vascular zone in normal eyes was 5350 units, in eyes with NAION was 3715 units, and in papilledema was 4124 units. The area of the peripapillary vascular zone was significantly lesser in the eyes with NAION as compared to normals (P = 0.02) [Fig. 1a and b]; however, there was no statistical difference between normals and those with papilledema although the area was lesser in those eyes with papilledema.

Discussion

ONH perfusion in NAION has been studied previously using FFA, and a delay in the prelaminar optic disc filling as seen on FFA was noted; however, no delay in peripapillary choroidal filling was seen.[2] Recent studies have shown a decrease in the peripapillary blood flow in these cases. One of the earliest studies on OCTA in NAION[5] included patients with acute NAION presenting within 1 week on onset of visual loss and reported that while the normal (other, unaffected) eye showed a dense microvasculature around the optic disc, the eyes with NAION showed a reduction in the peripapillary flow densities, with the temporal peripapillary sector being affected the most. Similarly, we also found a loss of the distinct microvasculature cuff in all the eyes with NAION as compared with the fellow eye. The area of the peripapillary vascular zone was significantly reduced in eyes with NAION as compared to those with controls.

Another study, which included eyes with acute and old NAION, also reported hypoperfusion at the level of the retinal peripapillary capillaries and the peripapillary choriocapillaris, with a greater degree of hypoperfusion seen at the choroidal level.[6]

A decrease in the peripapillary vessel density at the corresponding location of the visual field defects in patients with NAION was also seen in another study by Hata et al. However, the duration of the disease is not clarified in the published report.[7] When this was analyzed using ImageJ, the area of the peripapillary vascular zone was significantly lower in eyes with NAION as compared to the fellow normal eyes.

In addition, we found a second pattern of loss of microvasculature, described as focal deficits seen at the choroidal level, extending from ONH. These may reflect areas of capillary dropout at the choroidal level and were seen in 12 of the 21 (57.14%) eyes with NAION.

Visual field assessment was done in all patients, and 20/21 eyes (95.2%) showed clear field defects and 1 eye showed few inferiorly depressed points. Of these 20 eyes, 8 eyes showed generalized depression and no specific pattern of field defect, whereas the other 12 showed a clear altitudinal defect. A clear correlation between the visual field loss and the loss of vasculature could not be found in those eyes that showed only loss of the peripapillary vasculature cuff but lacked a focal deficit. A correlation was found in 11 of the 21 eyes (52.38%), which is similar to that found in a study by Wright Mayes et al., where OCTA areas of peripapillary choriocapillaris hypoperfusion corresponded to visual field deficits in 60% of the eyes studied.[6]

However, when we separately analyzed those eyes with focal areas of hypoperfusion as seen on OCTA, 11 of the 12 eyes (91.6%) that showed focal loss were found to have a corresponding visual field defect, suggesting that this pattern likely reflects true hypoperfusion and maybe responsible for the functional deficit seen in these patients.

This study included only cases with acute NAION, with disc edema being present at the time of evaluation, and hence, we considered the possibility that pattern of loss of microvasculature seen on OCTA maybe, in part, a result of the disc edema itself which could cause mechanical compression of the blood vessels or affect the acquisition of the scan. A similar concern was expressed in a previous study as well.[5] Hence, a comparison with eyes with disc edema of a nonischemic etiology was done. In eyes with papilledema, also we found that the microvasculature cuff appeared to be lost. ImageJ analysis showed a lesser area of the peripapillary vascular zone in eyes with papilledema as compared to normal eyes; however, this difference was not found to be statistically significant. None of these eyes had any vision loss, color vision loss, or a field defect other than an enlargement of the blind spot, suggesting that there was no permanent optic nerve damage. Hence, the appearance of the loss of the cuff in these eyes may reflect the distortion of the peripapillary vasculature network as a result of the physical effect of the disc edema itself and may not indicate true ischemia. However, none of the eyes with papilledema showed the focal pattern of loss of vasculature or hypoperfusion as was seen in 57.14% of eyes with NAION.

Limitations of this study include its small sample size and lack of quantitative parameters for assessment of flow. We included only patients with acute NAION and hence did not assess the retinal nerve fibre layer (RNFL) thickness or the thickness of the ganglion cell-inner plexiform layer complex and the correlation with the areas of hypoperfusion or the field defects. Longitudinal studies looking at the temporal changes in the vasculature and further correlation with functional parameters may add more information about the natural history of the disease. The fellow eye was used for comparison, and there is a possibility that minor vasculature changes may occur in these eyes also, hence giving rise to errors. Similar studies using age-matched normals maybe more accurate in this regard. Furthermore, 9 of the 21 eyes with NAION did not show focal deficits, and we are unable to explain this finding completely. It is unclear exactly what characteristic or stage of the disease maybe responsible for the choroidal defects seen and would require additional studies. It may be possible that OCTA was unable to delineate the flow in these areas due to very slow flow.

Conclusion

This study shows that OCTA shows definite hypoperfusion of the peripapillary choroidal vasculature in patients with NAION. However, loss or distortion of peripapillary vasculature, maybe seen in eyes with nonischemic disc edema such as papilledema also. A focal pattern of loss of microvasculature, such as that seen in eyes with NAION, is more likely to represent true hypoperfusion and corresponds well with the visual field defects seen in these patients.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.