Is there treatment for nonarteritic anterior ischemic optic neuropathy

Abstract

Purpose of review 

Nonarteritic anterior ischemic optic neuropathy (NAION) is the most common cause of an acute optic neuropathy over the age 50 with an annual incidence of 2–10/100 000. Most patients are left with a permanent decrease in visual acuity and visual field loss. No approved treatment has conclusively reversed the process or prevented a second event that typically involves the previously unaffected eye. Many medical and surgical treatments have been proposed with conflicting results. The goal of this review is to present current data in order to permit clinicians and patients to make an educated decision about treatment.

Recent findings 

Recently, there has been a flurry of case reports, small clinical trials and testing in animal models of NAION for various treatments for NAION and this review attempts to present the data concisely with the authors’ opinions about the reliability of the data.

Summary 

To date, there is no class I evidence of benefit for the treatment of NAION; however, the aphorism attributed to Carl Sagan, PhD aptly applies: ‘Absence of evidence is not evidence of absence’.

INTRODUCTION

There is no conclusive beneficial treatment for nonarteritic anterior ischemic optic neuropathy (NAION) at this time. Many small trials and several uncontrolled large retrospective chart reviews indicate positive and negative results, often for the same treatment. None is robust. Before considering treatments, it is worthwhile to review the natural history, risk factors and causative factors proposed for NAION.

Box 1: no caption available

Natural history, risk and causative factors

Natural history of nonarteritic anterior ischemic optic neuropathy

Confounding any assessment of the benefit of treatment for NAION is that natural history studies indicate up to 43% of people will show spontaneous improvement of visual acuity of at least three Snellen lines in NAION patients within 6 months ^[1]. Following an uncontrolled study of optic nerve sheath decompression that showed visual benefit ^[2], a controlled trial finally showed that this procedure was actually harmful to vision ^[1,3][1,3]. A long-term large observational study showed 41% with visual acuity of poorer than 20 of 70 improved by 6 months. Two years later, 18% of this cohort had worsened ^[4]. Other studies have shown spontaneous recovery of at least three Snellen lines in 13–43%, especially among patients younger than 50 years ^[5].

Pathogenetic theories

The pathogenesis of NAION remains unsettled. It is said to result from inadequate perfusion of the optic nerve head, but why poor perfusion occurs remains uncertain ^[6]. Optic nerve fluorescein angiography in NAION patients shows delayed filling of the prelaminar optic disc with normal choroidal filling, suggesting that NAION is due to occlusion of the short posterior ciliary paraoptic end-arterioles/capillaries, not the more proximal watershed between the ophthalmic artery and proximal posterior ciliary arteries (PCAs). The distal paraoptic PCAs have vascular domains that respect the horizontal meridian, which coincides with altitudinal visual field loss common in NAION (inferior visual field loss most common), whereas the proximal PCAs do not respect the horizontal meridian ^[7,8][7,8].

Risk factors and other putative causative factors

The physical finding most commonly seen in patients with NAION is a small or absent optic nerve head physiologic cup, indicating a congenitally small optic disc, or a ‘disc at risk’. Patients with small optic discs may have congenital constriction of the PCAs ^[9]. Small physiologic cups are present in more than 90% of NAION cases. After age 45–50, and perhaps in combination with one or more vascular risk factors (hypertension, dyslipidemia, anemia, sleep apnea syndrome, diabetes and tobacco exposure), further impairment of oxygenation to the optic nerve head occurs, leading to NAION.

An alternative pathogenetic theory suggests that vitreous traction on the optic disc during posterior vitreous separation causes NAION. In two studies, patients with new onset NAION showed a higher proportion of vitreopapillary traction than age-matched controls (40 vs. 10%) on spectral domain optical coherence tomography (SD-OCT) ^{[10,11][10,11]}.

Diabetes is associated with a higher incidence of NAION. Out of a Medicare population of over 25,000 and an equal number of nondiabetic patients matched for age and vascular risk factors, the 8-year incidence of NAION was 0.7% in diabetic patients as compared with 0.5% in nondiabetic patients, a 40% increase ^[12].

Sleep apnea syndrome (SAS) has been associated with a higher incidence of NAION. SAS was documented in 12 (70%) of 17 consecutive NAION patients as compared with 18% in age-matched and sex-matched controls ^[13]. Proposed mechanisms of action include transient increased blood pressure and intracranial pressure and transient hypoxia leading to optic nerve head ischemia in patients with small optic disc physiologic cups.

Nocturnal hypotension has been suggested as another risk factor ^[14]. However, nocturnal hypotension was not found in a higher number of NAION patients compared with a control group matched for age and vascular risk factors ^[15].

Venous insufficiency due to occlusion of retrolaminar venules draining the optic nerve arterioles is postulated to explain the hyperemic optic disc edema and peripapillary hemorrhages of NAION, as contrasted to the pallid edema attributed to arterial occlusion in arteritic AION ^[16].

Proposed treatments

Optic nerve sheath decompression

Optic nerve sheath decompression was theorized to treat a compartment syndrome caused by tight optic discs with edema, which might lead to compression of neighboring PCAs and propagation of ischemia. In the Ischemic Optic Neuropathy Decompression Trial (IONDT), 127 patients were randomized to optic nerve sheath decompression surgery and compared with 131 patients treated with daily aspirin. The treatment group fared worse than the untreated group and the study was terminated early ^[1,17][1,17].

Radial optic neurotomy

Radial optic neurotomy (RON) involves two incisions at the nasal optic disc margin using a 23-gauge blade to a depth of 1 cm in order to surgically open the scleral canal and theoretically reduce a compartment syndrome at the optic nerve head. The first reported use of RON was for the treatment of central retinal vein occlusion in 11 consecutive patients ^[18]. An 82-year-old woman was diagnosed with NAION in the setting of long-standing optic disc drusen and progressive visual acuity and visual field loss over 5 weeks in her left eye. The patient underwent RON after failing 1 g/day intravenous (IV) methylprednisolone for 3 days with oral steroid taper for 11 days and daily aspirin. Two days postoperatively, visual acuity (VA) improved from finger counting to 20/30 and visual field significantly improved, remaining stable at 18 months ^[19]. As impressive as this result sounds, it has not been replicated and is confounded by the known natural history of improvement in vision in NAION.

Extracorporeal counterpulsation

Extracorporeal counterpulsation uses a series of cuffs wrapped around the lower extremities that compress the legs during diastole to improve venous return and increase diastolic pressure in patients with severe heart failure. The treatment was shown to increase ophthalmic artery velocity in elderly patients with atherosclerosis ^[20]. Sixteen patients with NAION were treated for 1 h/day on 12 consecutive weekdays within 18 days of symptom onset (mean 12 days). Median visual acuity was 20 of 120 before treatment and 20 of 50 after completing the trial (P = 0.003). Improvement of at least three Snellen lines occurred in 62% of patients. Humphrey visual field (HVF) mean deviation showed borderline statistically significant improvement (mean deviation pretreatment −15.9 and −15.1 posttreatment, P = 0.049) ^[21]. No control group was included.

Oral corticosteroids

An open label patient choice study included 312 patients who elected treatment with oral prednisone 80 mg/day for 14 days initiated within 14 days of vision loss, as compared with 301 patients who elected no treatment. Patients were followed for a mean 3.8 years. At 6 months, 70% of patients treated with prednisone and whose entry VA was poorer than 20 of 70 showed at least three Snellen lines improvement vs. 40% in the untreated group (P = 0.001). Goldmann visual field testing improved in 40% in the treated group as against 25% in the untreated group (P = 0.005). Patients with mild visual acuity loss (20/40–20/60) showed nonsignificant improvement (P = 0.9). The treated patients with mild field loss showed greater improvement (32%) than did the untreated patients (14%) (P = 0.03) ^[22]. A review of this study pointed out that confounding factors included the lack of randomization, blinding and the fact that the untreated group was older and more often had vascular risk factors than did the treated group ^[23].

In another study, et al.^[24▪] used the same prednisone dose schedule in 10 treated patients randomized against 27 untreated patients. At 6 months, there was no significant improvement in visual acuity (P = 0.28), HVF mean deviation (P = 0.21) or retinal nerve fiber layer thickness on spectral domain ocular coherence tomography (SD-OCT). One of the 10 treated patients suffered substantial corticosteroid side-effects, requiring early termination of treatment. More concerning was the fact that two of the 10 treated patients developed second eye NAION, one at 2 months and the other at 3 months, whereas none of the 27 control patients developed NAION. (Natural history studies indicate a 15% risk of second eye NAION within 5 years.) This study was not powered to determine whether oral steroids increase the risk of second eye NAION.

Intravitreal corticosteroids

Several small or uncontrolled trials of intravitreal steroids have shown mixed results in NAION. Kaderli et al.^[25] injected four patients and six untreated controls and showed improved mean visual acuity but not visual field in patients treated with 4 mg triamcinolone injected intravitreally. Sohn et al.^[26] reported one patient who showed improved visual acuity from 20/400 to 20/50 at 3 months.

Intravenous corticosteroids

Kinori et al.^[27▪] retrospectively compared two nonrandomized groups of 23 patients, one group treated with intravenous methylprednisolone 250 mg q6 h for 3 days followed by an oral prednisone taper for 11 days initiated within 14 days of symptom onset, the other group including patients who refused corticosteroid treatment or had contraindications to their use. The baseline mean visual acuity was 20 of 70 in both groups. The mean deviation on HVF testing at baseline was worse in the treated group (P = 0.007). At 6 months, mean visual acuity was 20 of 80 in the treated group and 20 of 53 in the untreated group (P = 0.3). The visual field mean deviation was equal in the two groups (P = 0.9). The authors concluded that high-dose intravenous treatment within 14 days of symptom onset of NAION had provided no statistical benefit.

Posterior sub-Tenon steroids

A randomized, double-blinded study of 40 NAION patients (20 treated and 20 controls) treated within 7 days with a single 40 mg posterior sub-Tenon injection of methylprednisolone reported at least three Snellen lines of visual acuity improvement in half of the treated group and in none of the control group at 8 weeks, a finding that did not reach statistical significance (P = 0.053). Visual field testing was not done ^[28].

Aspirin

In the IONDT, aspirin did not show a statistically significant improvement in visual function and did not reduce the risk of second eye NAION developing within 5 years ^[1,3][1,3].

A retrospective analysis of 153 NAION patients treated with aspirin after diagnosis and 278 not treated with aspirin showed that at 2 years, second eye NAION was 7% in aspirin group as compared with 15% in the untreated group. At 5 years, the incidence was 17% in the aspirin-treated group and 20% in the aspirin-untreated group ^[29]. The authors concluded that aspirin does not significantly reduce the risk of second eye NAION.

A retrospective case−control study compared 23 NAION patients taking various doses of aspirin before and after diagnosis of NAION with 55 NAION patients not taking aspirin before or after diagnosis. This study showed no difference in visual acuity or visual field mean deviation at diagnosis or follow-up ^{[30,31][30,31]}.

Dabigatran

Dabigatran is a direct thrombin inhibitor approved for the prevention of emboli due to nonvalvular atrial fibrillation and deep vein thrombosis. A single case report presented a 46-year-old female who developed NAION immediately following uncomplicated pars plana vitrectomy, laser treatment and gas tamponade for a macula-off retinal detachment of unknown duration. The patient's visual acuity improved from hand motion to 20 of 400 following treatment with prednisone 100 mg/day for 3 days, followed by a 12-day taper, and aspirin 100 mg/day. Dabigatran 110 mg twice daily was added at day 24. The following day, visual acuity improved to 20 of 200, and 25 days later, visual acuity was 20 of 50 and static perimetry was also better ^[32]. Because the patient in that study had developed perioperative NAION, results may not apply to NAION unassociated with preceding surgery.

Pentoxyfylline

Pentoxyfylline is a phosphodiesterase inhibitor that improves red blood cell deformability and reduces blood viscosity. It has been shown to increase optic nerve head perfusion in animal models of NAION. Pentoxyfylline was orally administered to 235 NAION patients who were compared with 221 NAION controls. Median time to development of second eye NAION was 7 months in the pentoxyfylline-treated group and 20 months in the control group. But during the observation period, 10% of the treated group developed second eye NAION as against only 3% in the control group. There was no benefit to vision in the treated group ^[33].

Caspase-2 inhibitor

Compound QPI-1007 is a synthetic siRNA that temporarily inhibits expression of caspase-2, which has been shown to reduce retinal ganglion cell apoptosis in rodent and nonhuman primate models of NAION. A phase I trial (ClinicalTrials. gov Identifier NCT01064505) included 30 patients with NAION with visual acuity of less than 20 of 40 who were injected with this compound intravitreally within 28 days of symptom onset. Half the patients had an improvement of at least three Snellen lines which was maintained at 6 months, a course comparable to that of natural history studies. No patients worsened at least three lines ^[34]. Quark Pharmaceuticals plans a phase II/III study.

Memantine

Memantine is an N-methyl-D-aspartate receptor antagonist approved for the treatment of dementia. A nonrandomized uncontrolled trial of 22 patients with NAION occurring less than 8 weeks earlier received 5 mg/day orally for 1 week and then 10 mg/day for 2 weeks. Only one patient had a disc at risk for NAION. At 3 months, mean visual acuity improved from 20/400 to 20/125 (P = 0.001). The average mean deviation improved from −19 to −16 (P = 0.02) ^[35]. The number of patients who improved was not reported, therefore, these results cannot be compared with those of natural history studies.

Minocycline

A rat model of NAION was used to test intraperitoneal injection with minocycline 22 mg/kg/day beginning 3 days before NAION and continued for 30 days. At 30 days, there was 10% loss of retinal ganglion cells in the eight treated rats compared with a 48% loss in the eight untreated rats (P < 0.05) ^[36].

Topical brimonidine

A randomized, double blinded, placebo-controlled trial involving 36 NAION patients above 40 years of age treated with topical brimonidine within 7 days of symptom-onset showed no statistically significant difference in visual acuity. There was a trend toward improved mean deviation ^[37].

Dalfampridine

An ongoing phase IV placebo-controlled trial of dalfampridine, a potassium channel blocker that increases nerve conduction in demyelinated nerve fibers, follows a placebo-controlled cross-over trial that showed slight visual acuity benefit (ClinicalTrials. gov Identifier NCT01975324.)

Sunitnib

Sunitinib (Sutent), a tyrosine kinase inhibitor approved for the treatment of renal cell and gastrointestinal tract carcinomas, showed higher retinal ganglion survival in a mouse model of NAION (58 vs. 37%, P < 0.05), possibly by reducing apoptosis ^[38]. No human trials have started to date.

Intravitreal anti-vascular endothelial growth factor agents

Several uncontrolled trials using either bevacizumab or ranibizumab intravitreally have documented faster resolution of optic disc edema after injection than have natural history studies. One uncontrolled study of 11 patients treated with bevacizumab within 14 days of symptom onset showed at least three Snellen lines improvement in visual acuity in five (45%) patients but no improvement in visual field mean deviation, similar to natural history studies ^[39]. Rootman et al.^[40] studied 17 NAION patients treated with intravitreal bevacizumab 1.25 mg injected within 15 days of symptom onset (mean 8 days) compared with eight untreated NAION controls. The mean deviation primary endpoint was equal (P = 0.9). The visual acuity secondary endpoint was also not significantly different (P = 0.6). The mean nerve fiber layer loss was also not statistically significant (P = 0.1). Of concern was that two treated patients (12%) developed same eye NAION at day 4 and at 14 months, compared with none in the control group, suggesting that bevacizumab may contribute to causing NAION. A retrospective study of ranibizumab in six patients treated within 15 days of onset (mean 8 days) showed an average improvement of 2 Snellen lines of visual acuity ^[41].

Intravitreal erythropoietin

Erythropoietin is a cytokine hormone that has shown neuroprotective and neuroregenerative properties in a rat model of traumatic optic neuropathy ^[42]. In a prospective human case series of 31 patients treated intravitreally within 30 days of onset of NAION (mean 11 days), 17 patients (55%) improved at least three Snellen lines of visual acuity (P < 0.001) at the 6-month examination. At 3 months, 61% had improved. The mean deviation did not improve significantly from a baseline of −19.6 to −18.6 (P = 0.6) ^[43].

CONCLUSION

It is reasonable to extrapolate data from cardiovascular and cerebrovascular studies showing a reduction in events with aspirin and risk factor modification, but there are no class I data for NAION. There is class I evidence that optic nerve fenestration surgery is ineffective and potentially harmful. The efficacy of steroids, administered by oral, intravenous, intravitreal or sub-Tenons routes, remains uncertain and several studies indicate a possible increased risk of second eye NAION.

ACKNOWLEDGMENTS

None.

Financial support and sponsorship

None.

Conflicts of interest

There are no conflicts of interest.

REFERENCES AND RECOMMENDED READING

Papers of particular interest, published within the annual period of review, have been highlighted as:

• ▪ of special interest
• ▪▪ of outstanding interest