“The cooperation of the ophthalmologists, neurologists and neurosurgeons is the only way to solve the complex cases none could approach separately.”

MD. Tatiana Roşca

Poster presented at: -

Download Poster (0Mb)

Special medical article:
Current Concepts in the Diagnosis, Pathogenesis and Management of Anterior Ischemic Optic Neuropathy
Neil R. Miller, MD FACS - Profesor de Oftalmologie, Neurologie şi Neurochirurgie, Frank B. Walsh Profesor of Neuro-Oftalmologie, Wilmer Eye Institute, Johns Hopkins Hospital,Baltimore, MD USA

Our deep gratitude to the author for his prompt persmission to use his excellent work on our site.


Anterior Ischemic Optic Neuropathy

Overall, the second most common optic nerve-related cause of permanent visual loss in adults after glaucoma.


Types of Anterior Ischemic Optic Neuropathy:


- Nonarteritic

- Perioperative

- Arteritic



Nonarteritic AION

  • Usually in patients over age 55
  • Men and women equally affected
  • Most have underlying systemic vascular disease (but may be undiagnosed at time of onset)
  • Occurs in 3-10 per 100,000


Nonarteritic AION

  • Usually unilateral
  • Eye pain rare; pain on eye movement very rare
  • Represents about 90% of all cases of nonarteritic ION (rest are retrobulbar)
  • Visual acuity loss variable: 20/20 to HM or worse
  • Color vision usually mirrors acuity
  • Visual field usually shows altitudinal or arcuate defect
  • RAPD always present if unilateral and no abnormality in opposite eye
  • Disc usually hyperemic
  • Peripapillary flame-shaped hemorrhages often present
  • Opposite optic disc small or normal-sized with little or no cup
  • Underlying systemic vasculopathy usually present but may not be known at time of visual loss
  • Hypertension
  • Hypercholesterolemia
  • Diabetes mellitus
  • Smoking a risk factor? [1]


Risk Factors for Most Cases

A normal or congenitally small disc with little or no cup PLUS Vascular compromise



  • Ischemia at level of prelaminar/laminar portion of optic nerve supplied by circle of Zinn-Haller via short posterior ciliary arteries
  • No evidence of SPCA occlusion [2]
  • Blood flow velocities of PCA and CRA are reduced in patients with acute NAION compared with controls [3]


Nonarteritic AION - Other settings:

  • Hypotension (nonsurgical)
  • Anemia (nonsurgical)
  • Obstructive sleep apnea
  • Hyperhomocysteinemia
  • Erectile dysfunction drugs
  • Amiodarone


Nonsurgical Hypotension and NAION:

  • Patients often awaken in the AM with visual loss (significance debated—SS Hayreh vs NJ Newman)
  • Cases, including challenge cases, occur after ingestion of erectile dysfunction drugs that lower BP
  • Cases reported in acutely hypotensive patients on or after dialysis (all with chronic anemia)
  • Some have improved with restoration of BP (medicolegal implications!)


Nonsurgical Anemia and NAION:

  • Several cases reported (without hypotension)
  • Patients on dialysis
  • Patients with epistaxis
  • Patients with uterine bleeding
  • Some patients improve following rapid restoration of hematocrit
  • Prognosis with improved hematocrit used as basis of medicolegal cases in patients with perioperative ION (similar to hypotensive issue)


Sleep Apnea Syndrome (SAS) and NAION

  • Relationship suspected because many patients with NAION note decreased vision upon awakening
  • Characterized by repetitive upper airway obstruction during sleep lasting 10 seconds to 2 minutes, inducing hypoxia and sleep deprivation
  • SAS present in 71-89% of patients with NAION [4]
  • Mojon et al.
    - In the 45-64 year-old age group, 50% with NAION had SAS vs 10% in a control group
    - In patients >64 years old with SAS, 89% NAION vs 24% in control group
  • Palombi et al.
    SAS 1.5-2x more likely to be present than other risk factors (eg, diabetes, hypertension)
  • Li et al.
    - Case-control study indicates that validated questionnaire can be used to identify at-risk patients if appropriate cut-off points are used


NAION and Sleep Apnea Syndrome


Should either use a validated questionnaire or question patient and spouse/significant other re: snoring loudly, stopping breathing during night sleep, awakening suddenly gasping for breath, perspiring during night sleep, and snoring/breathing worse when sleeping on the back and/or after alcohol use

  • If positive response, sleep studies indicated
  • If sleep studies positive, CPAP should be started


Hyperhomocysteinemia and NAION

  • Hyperhomocysteinemia considered an important risk factor for arterial disease
  • Hyperhomocysteinemia a strong risk factor for stroke, myocardial infarction in patients younger than 50
  • Percentage of patients with NAION who have hyperhomocysteinemia varies from 0% (Biousse et al.) to 45% (Pianka et al.)
  • Recommend homocysteine serum assay in all patients with NAION, especially
    - Young patients without apparent risk factors
    - Older patients with cardiac disease?
  • Cost of assay is about $24.00


NAION and Coagulopathies

  • Rare and usually in patients younger than 60 yrs old
    - Increased Factor VIII
    - Presence of antiphospholipid antibodies
    - Activated Protein C resistance [5]
  • May be appropriate to perform a coagulopathy evaluation in young individuals with NAION and no obvious risk factors


Erectile Dysfunction Drugs and NAION

  • Drugs to treat erectile dysfunction
    Viagra (Sildenafil)
    Cialis (Tadalafil)
    Levitra (Vardenafil)


Can Erectile Dysfunction Drugs Cause NAION?

  • Yes
    ED drugs are vasodilators and thus may cause systemic hypotension (biological plausibility)
    Challenge cases occur, so some association exists
  • No
    Less than 100 cases of AION reported compared with number of prescriptions written (50 million). [6]
    Retrospective cohort study of 4 million male VA patients prescribed PDE-5 inhibitors showed no difference to published rates of NAION. [7]
  • Who knows? [8]


Erectile Dysfunction Drugs and NAION

  • Most (but not all!) cases are anterior and have disc-at-risk (screening useful?)
  • Any male patient who develops apparent NAION should be asked about use of EDDs (in private!)
  • Any patient who asks about use should have careful disc assessment and be told about potential risk.



  • Developed in the 1960s for treatment of angina but now used as primary drug for atrial fibrillation in patients with left ventricular dysfunction and CHF [9]
  • Prolongs duration of action potential and refractory period in cardiac-conducting tissue
  • Long and variable half-life (26-107 days to reduce plasma concentration by 50%)
  • Ocular symptoms occur in 1-11% of patients [10]
  • Most frequent complaint is seeing blue-green rings or halos around lights


Amiodarone and Optic Neuropathy

  • 1982: E Chew et al. mentioned a case of what was thought to be nonarteritic AION in a patient taking amiodarone--no details given
  • 1987: JW Gittinger, GK Asdourian reported two cases of "papillopathy associated with amiodarone"
    - One unilateral; one bilateral
  • 1987: LA Feiner et al. Mayo Clin Proc 62:702
    - 447 patients taking amiodarone
    - 13 patients with optic neuropathy (2.9%)
    - Ages 52-73
    - Daily dose 200-600 mg for 1-72 months
    - Visual loss ranged from 20/20 to 20/200
    - Most had optic disc swelling similar to that seen in AION
    - Authors calculated that incidence of AION in patients over 50 during preceding 10 years was 0.3% compared with 1.8% for age-matched patients taking amiodarone (6x risk)
  • 1990 – present: at least 100 case reports, case series, and pathology reports published linking amiodarone to optic neuropathy
  • 1997: package insert modified to list optic neuropathy as an adverse event


Amiodarone and Optic Neuropathy

  • Amiodarone ON
    Onset: Insidious (months)
    Nature: Bilateral, simult
    Acuity: 20/20-20/200
    Res ODS: Months
    Onset: Rapid (days)
    Nature: Unilateral
    Acuity: 20/20-NLP
    Res ODS: Weeks


Amiodarone and Visual Loss

  • 1669 subjects received closed-label amiodarone (mean dose = 3.7mg/kg/day) (n = 837) or placebo (n = 832)
  • Followed for 27 months or until death (median F/U of survivors = 45.5 months)
  • End point was removal from study because of complaints of bilateral visual loss
  • No subject was removed from study because of bilateral visual loss
  • Calculated possible annual incidence rate of bilateral visual loss from amiodarone was 0.13%
  • Conclusion: At commonly used clinical doses, bilateral visual loss from amiodarone occurs infrequently if at all


Amiodarone and Optic Neuropathy
Absence of proof is not proof of absence

  • What We Think We Know About Amiodarone and Vision
    If it causes an optic neuropathy, the visual loss in most cases is not particularly severe.
  • Amiodarone and Optic Neuropathy
    Despite contentions by Johnson et al. (J Natl Med Assoc 96:1477, 2004), according to WHO definitions of causality assessment of suspected adverse reactions, amiodarone should be considered a POSSIBLE cause:
      "A clinical event ... occurring within a reasonable time sequence from administration of the drug but one that could also be explained by concurrent disease or the presence of other drugs or chemicals. Information on drug withdrawal may be lacking or clear."


Amiodarone and Optic Neuropathy:
What is Your Responsibility?

  • Any patient taking amiodarone who develops an optic neuropathy should be told of the possible association
  • Patient’s cardiologist and/or PCP should be contacted in any case of optic neuropathy (anterior or retrobulbar, unilateral or bilateral) and told of possible association
  • Decision to stop or continue drug should be made by cardiologist and patient.[11]


Nonarteritic AION: Prognosis

  • About 40% improve spontaneously (IONDT)
  • Visual acuity more likely to improve than visual field
  • Disc becomes pale but not cupped


Nonarteritic AION: Systemic Prognosis

  • Patients at risk for subsequent cerebrovascular and cardiovascular events (eg, TIA, Stroke, MI)
  • Increased mortality
  • No satisfactory treatment
    - Dilantin (?), systemic steroids (but see SS Hayreh), surgery (ONSF), ASA of no benefit
    - Intravitreal injections (?steroids, ?bevacizumab—S Kelman, NANOS & AAO, 2009)
  • Small risk for recurrent NAION in same eye (3-5%?)
  • Variable risk (15-25%?) for involvement of other eye
    - Factors include age, vascular disease, visual acuity
    - No prophylactic therapy (reduce or eliminate risk factors, ?ASA)


Incipient Nonarteritic AION

In incipient cases, 25% become symptomatic and 20% resolve and then recur symptomatically. [12]


Perioperative AION

Post-Cataract Surgery

Post-Cardiac Surgery

Post-Prone Position Back Surgery


CD Townes et al.

  • Complications of cataract surgery [13]
  • First description in the American literature?
  • Described four cases of "optic neuritis" following CE


AB Reese, RD Carroll

  • Optic neuritis following cataract extraction [14]
  • Described 17 eyes of 17 patients with post-CE "optic neuritis"
  • All patients underwent uncomplicated cataract extraction (CE) with normal post-op VA and IOP
  • No consistently associated anesthetic techniques
  • Visual loss occurred 6-12 weeks post-op
  • Developed central scotoma and optic disc swelling (anterior optic neuropathy)
  • Variable visual outcomes (mean VA ~ 20/50)
  • 10/17 patients underwent subsequent CE in the fellow eye
  • 3/10 developed similar "neuritis"


Post-CE Optic Neuropathy
DD Michaels, GS Zugsmith
Optic neuropathy following cataract extraction [15]

  • Two types of Post-CE optic neuropathy
  • Immediate: Associated with elevated intraoperative IOP
  • Delayed: Resembles ischemic optic neuropathy


FD Carroll
Optic nerve complications of cataract extraction [16]

Published 9 additional cases, 5 of whom also developed optic neuropathy after CE in the fellow eye.

Between his two series, 8 of 17 patients (about 50%) with post-CE optic neuropathy developed a similar condition in the fellow eye after CE.


TJ McCulley şi al.

Incidence of nonarteritic anterior ischemic optic neuropathy associated with cataract extraction. [17]

Retrospective analysis of records at Bascom Palmer for diagnosis of NAION within 1 year following CE.
3 patients diagnosed with NAION out of 5787 CE = 1/2000
Previously reported incidence for spontaneous NAION = 2.3-10.2/100,000 annually. Thus, incidence of Post-CE ON 5X higher than typical NAION.


TJ McCulley şi al.

Nonarteritic anterior ischemic optic neuropathy and surgery of the anterior segment: temporal relationship analysis. [18]

Hypothesis: If unrelated to CE, so-called post-CE ON should be randomly distributed throughout the year following CE

Retrospective cohort study: Outcome was time between CE and NAION diagnosis over 1-year period.


Is Post-CE ON "Ischemic"?

Risk of Nonarteritic Anterior Ischemic Optic Neuropathy After Cataract Extraction in the Fellow Eye of Patients with Prior Unilateral Nonarteritic Anterior Ischemic Optic Neuropathy. [19]

Retrospective cohort study at BPEI
- 325 patients with unilateral spontaneous NAION
- 17 underwent CE in fellow eye
- 9/17 (53%) developed AON in fellow eye
- 6 of 9 cases (67%) within 6 months (Median, 2.8 mo)
- 59/308 (19%) in control group (ie, no CE) developed NAION
Risk of NAION after CE in fellow eye = 3.6x expected.


Post-Cataract Optic Neuropathy

  1. An optic neuropathy resembling NAION occurs after uncomplicated CE at a rate much higher than would be expected.
    50/100,000 compared with 2-10/100,000 annually in similar population.
  2. There are two types of post-CE ON: Immediate and delayed.
    Immediate: May be associated with elevated IOP.
    Delayed: Factors unclear but may be similar to those influencing spontaneous NAION.
  3. IONDT demonstrated that patients with unilateral NAION have ~ 15-20% 5-year risk of NAION in the fellow eye: CE increases this risk 2-8 times
  4. Patients who have undergone CE and developed post-CE ON should be told of increased risk if CE done in fellow eye (50%).
  5. Patients who have experienced spontaneous NAION in one eye should be told of increased risk if CE done in fellow eye.
  6. Control of IOP during surgery may reduce risk of immediate post-CE ON.
  7. There is no way to prevent delayed post-CE ON?


Perioperative AION

Occurs most often after cardiac surgery with CPB.

After cardiac bypass surgery.
0.06% = 6 in 10,000 (Mayo Clinic)
0.11% = 1.1 in 1000 (JHH)

Also occurs after back surgery but most cases are POSTERIOR (retrobulbar).

One case of NAION reported after rotator cuff surgery. [20]

One case of NAION after repair of leg fracture.


Perioperative AION

Factors leading to visual loss

Significant blood loss (absolute and relative drop in Hgb)

Hypotension (low MAP)

Rotator case had no anemia, only hypotension with 42% drop in MAP for 80 minutes!

Leg fracture case was in long-standing diabetic, chronic hypertensive; 2 hr anesthesia; no blood loss; 50% drop in MAP for 40 minutes!

Cardiopulmonary bypass factors (coagulation issues?)


Perioperative AION

May be unilateral or bilateral.

When bilateral, usually simultaneous but may be sequential. [21]

Vision usually very poor.

Improvement rare and usually incomplete.

No proven treatment but anemia and hypotension should be corrected asap for both patient care and for medicolegal reasons. Direct evidence: [22]. Indirect evidence: [23]


Arteritic AION

Usually associated with giant cell (temporal) arteritis, but other vasculitides possible (eg, PAN).

No eye pain, but headache, scalp tenderness, jaw pain, ear pain, may be present (must ask directly).

May have had transient visual loss preceding.

Visual acuity loss usually profound (HM or worse).

Marked loss of visual field.

RAPD always present (unless other eye involved).

Classic features.

Disc usually shows pallid swelling (indicating infarct).

Soft exudate(s) may be present.



  • Careful history: Must ask about nonvisual symptoms!
  • Examination
  • Laboratory studies
  • Erythrocyte sedimentation rate and C-reactive protein most important
  • Platelet count?


Erythrocyte Sedimentation Rate. Most often done by Westergren method (more accurate at high levels)

Normal values:
Top normal for men: Age/2 Top normal for women: (Age + 10)/2


C-Reactive Protein

Acute phase reactant present in serum

Concentration rises after almost any tissue injury, thus often elevated in temporal arteritis

May be elevated in normals


Platelet Count

Thrombocytosis present in 44% of GCA pts

Related to severity of inflammation

May be risk factor in subsequent occlusive events

Steroid therapy reduces PLT level [24]



  • The ESR is 95% sensitive for the diagnosis of temporal arteritis
  • The CRP is 97.6% sensitive for the diagnosis of temporal arteritis
  • Combination of ESR and CRP is 98%+ sensitive for diagnosis of temporal arteritis
  • Platelet count most important of all?
    No (SS Hayreh)
    Yes (LA Levin)



  • If temporal arteritis suspected
    - Obtain emergency ESR and CRP (?platelet count)
  • If ESR or CRP elevated
    Steroids are the ONLY PROVEN TREATMENT
  • Give one dose of IV solumedrol 1 gm stat.
  • Begin solumedrol 1 gram q24 hrs (to prevent visual loss in other eye and perhaps to regain vision in affected eye)
  • Obtain TA biopsy (unilateral vs bilateral) asap.
  • If both ESR and CRP normal (3%), must decide if clinical presentation is sufficient to warrant temporary treatment and biopsy.


Temporal Artery Biopsy

  • Follow Sutton’s law: Perform biopsy on side of discomfort in area of discomfort.
  • Although skip areas exist, they usually are only a mm or two.
  • Performing a 4 mm biopsy offers a 99% probability of avoiding a skip area.
  • Performing a bilateral temporal artery biopsy increases the yield from 1-11%.
  • Mayo Clinic: 11%, U of Iowa: 7%, JHH: 3%, Wills: 1%
  • Consider prepping patient for bilateral and performing frozen sections on first side.
  • If frozen sections negative on first side, do other side.


Why do a TA biopsy in a straight-forward case of GCA?

  • Medical and medicolegal issues.
  • Side effects of systemic corticosteroids may be significant--need tissue diagnosis.
  • Tendency to stop medication because of side effects--diagnosis questioned.



  • If ESR and/or CRP elevated and unilateral TA biopsy negative.
  • Consider contralateral TA biopsy (increases yield 1-11%).
  • Consider alternative diagnoses (esp malignancy).


Alternatives to Steroids

  • Many drugs tried; mostly case reports and a few series
  • No drugs clearly beneficial as REPLACEMENT
  • Candidate drugs:
    1982: Ciclofosfamide (CR)
    1983: Dapsone (CR)
    1986: Azathioprine (CCT)
    1989: Methotrexate (CR)
    2006: Infliximab (CR)
    2007: Adalimumab (CR)


"The best treatment for those cases of giant cell arteritis which are poorly controlled by conventional corticosteroid therapy, or in which this therapy may be poorly tolerated, has yet to be discovered." (S De Vita et al. J Intern Med 232:373-375, 1992)


GCA Management

OK, can we at least reduce steroids or get the patient off steroids faster if we ADD another drug?

Drugs suggested:

  • Infliximab
  • Methotrexate
  • Heparin


Additions to Steroids - Infliximab

  • An IgG antibody against TNF-α.
  • Approved for use in Crohn disease and rheumatoid arthritis.
  • One randomized trial.
  • 44 pts randomized to steroids + infliximab or steroids and placebo. No benefit noted. [25]



  • Meta-analysis of three controlled clinical trials comparing steroids and methotrexate with steroids and placebo in 161 patients.
  • Conclusion: Adjunctive treatment with MTX lowers the risk of relapse and reduces side-effects of steroids. [26]



  • Theoretically helpful in cases with thrombocytosis and increased anti-cardiolipin antibodies.
  • Three case reports. [27]


Arteritic AION


Once clinical diagnosis of temporal arteritis is considered, must obtain ESR and CRP and act on the results!




Corticosteroids are the only proven treatment.


Initial treatment for patients with proven or suspected arteritic AION (or PION) should be 1 gm methylprednisolone IV stat (repeat every 24 hours for at least 3 days).


In cases of suspected GCA, the default should be a unilateral or bilateral TA bx (after Rx begun).


Methotrexate may be an appropriate addition to steroids.


Heparin may be an appropriate addition to steroids in selected cases in which there is thrombocytosis.



Where Do We Go From Here?

An Animal Model of AION


Rodent model of NAION established

SL Bernstein et al. IOVS 44:1052, 2003 (rat)

N Goldenberg-Cohen et al. IOVS 46:2716, 2005 (mouse)

C Zhang et al. Brain Res Jan 14, 2009 [Epub ahead of print]

Primate model of NAION established

CS Chen et al. IOVS 49:2985-2992, 2008

SL Bernstein et al. J Neuro-Ophthalmol 28:79-81, 2008

SL Bernstein, NR Miller. IOVS 2009 Sept 3 [Epub]

Primate Model

Rodent optic disc and vasculature differ markedly from human

No (or poorly developed) lamina cribrosa



Nonhuman primates have discs at risk!



[1] Lyme Disease - Author: Gerald W Zaidman, MD, Professor of Clinical Ophthalmology, New York Medical College; Chief of Cornea Service, Acting Director, Department of Ophthalmology, Westchester Medical Center Updated: Jul 25, 2008

[2] John O Meyerhoff, MD, Assistant Professor, Department of Internal Medicine, Johns Hopkins University School of Medicine; Clinical Scholar in Rheumatology, Department of Medicine, Sinai Hospital of Baltimore, Updated: Jul 24, 2009

[3] Augusto A Miravalle, MD, Fellow, Department of Neurology, Beth Israel Deaconess Medical Center, Harvard Medical School Coauthor(s): R Philip Kinkel, MD, FAAN, Associate Professor of Neurology, Harvard Medical School; Director, Multiple Sclerosis Center, Beth Israel Deaconess Medical Center; Consultant Neurologist, Children's Hospital of Boston Updated: Jul 23, 2009

[4] Gerald W Zaidman, MD, Professor of Clinical Ophthalmology, New York Medical College; Chief of Cornea Service, Acting Director, Department of Ophthalmology, Westchester Medical Center Updated: Jul 25, 2008

[5] Petri M. Epidemiology of the antiphospholipid antibody syndrome. J Autoimmun 2000;15:145-51

[6] Price BE, Rauch J, Shia MA, et al. Anti-phospholipid autoantibodies bind to apoptotic, but not viable, thymocytes in a ß2-glycoprotein I-dependent manner. J Immunol 1996;157:2201-8

[7] Levine JS, Subang R, Koh JS, Rauch J. Induction of anti-phospholipid autoantibodies by beta2-glycoprotein I bound to apoptotic thymocytes. J Autoimmun 1998;11:413-24[erratum, J Autoimmun 1999;12:143.

[8] Price BE, Rauch J, Shia MA, et al. Anti-phospholipid autoantibodies bind to apoptotic, but not viable, thymocytes in a ß2-glycoprotein I-dependent manner. J Immunol 1996;157:2201-8.

[9] E-Medicine: Lyme Disease John O Meyerhoff, MDJohn O Meyerhoff, 2009


Last update: 13.01.2017