Scheie Eye Institute, Department of Ophthalmology, University of Pennsylvania, Philadelphia, Pennsylvania.
Address correspondence to Robert A. Stoltz, MD, PhD, Scheie Eye Institute, Department of Ophthalmology, University of Pennsylvania, 51 North 39th Street, Philadelphia, PA 19104; E-mail: firstname.lastname@example.org
A 76-year-old man developed a sudden painless superior field defect in the right eye, retinal whitening along the inferior temporal arcade, and fluorescein angiographic evidence of lobular choroidal non-perfusion. One week later, ophthalmoscopy revealed inferior optic nerve edema with splinter hemorrhages consistent with an anterior ischemic optic neuropathy (AION) and a new cholesterol plaque near the macula. There was no clinical, serologic, or pathologic evidence of giant cell arteritis. Carotid ultrasound revealed no evidence of significant stenosis but did show an echolucent soft plaque in the right carotid artery. Transthoracic echocardiography demonstrated normal left ventricular function with no source of emboli. The presumed cause of the clinical findings in this patient was embolism, a rarely reported cause of AION. An embolic origin may be considered in non-arteritic AION associated with choroidal non-perfusion.
Embolic disease is thought to be a rare cause of anterior ischemic optic neuropathy (AION), with only a few reported cases (1-3). Rather, non-arteritic AION is believed to occur secondary to transient hypoperfusion of vessels in the anterior part of the optic nerve (4), although thrombotic and anatomic vascular factors also may contribute. The presence of choroidal non-perfusion on fluorescein angiography is not a feature of non-arteritic AION (5,6).
We present a patient who had non-arteritic AION and fluorescein angiographic evidence of choroidal non-perfusion. A single cholesterol embolus was noted within the retinal vasculature on follow-up examination, suggesting that the AION may have been secondary to emboli.
A 76-year-old man with a history of hypertension and hypercholesterolemia had sudden painless loss of the superior field of vision in his right eye. The patient reported no headaches, diplopia, jaw claudication, scalp tenderness, polymyalgia rheumatica, or prior transient visual obscuration. Visual acuity was 20/30 in the right eye and 20/20 in the left eye. There was a mild right afferent pupillary defect (APD). Dilated ophthalmoscopic examination of the right eye revealed a small area of retinal whitening along the inferior temporal arcade (Fig. 1A). The optic disc appeared normal. Ophthalmoscopic examination of the left eye revealed a normal-appearing optic nerve with no evidence of disc crowding (Fig. 1B).
Fluorescein angiography showed lobular choroidal non-perfusion inferior to the optic disc at 24.2 seconds (Fig. 2), with patchy filling defects present at 733 seconds. The early-phase fluorescein angiogram demonstrated delayed filling of the optic disc inferiorly, adjacent to the choroidal filling delay (Fig. 2). Westergren erythrocyte sedimentation rate (ESR) was 10 mm/h; C-reactive protein was <0.7 mg/dl. A transthoracic echocardiogram revealed no source of emboli, with normal left ventricular function. Carotid ultrasound studies showed no evidence of significant stenosis, but an echolucent soft plaque was noted in the right carotid artery.
On follow-up examination one week later, the patient reported no change in symptoms. Visual acuity was 20/40, and a right afferent pupillary defect was observed. Goldmann visual field testing revealed a paracentral scotoma with a superior altitudinal field defect (Fig. 3). Ophthalmoscopy revealed inferior optic nerve edema with splinter hemorrhages as well as the previously observed area of retinal ischemia (Fig. 4). In addition, a cholesterol embolus was noted in a small arteriole branching off of the superior temporal arcade (Fig. 4). This embolus was not seen on review of fundus photographs taken at the initial presentation.
A temporal artery biopsy showed no histologic evidence of giant cell arteritis.
Six weeks later, visual acuity had improved to 20/20 in the right eye; a mild right APD was still present. Goldmann visual field testing revealed a small paracentral scotoma with some improvement in the superior altitudinal field defect (Fig. 5). Ophthalmoscopy showed no evidence of optic nerve swelling and showed resolution of the area of retinal whitening with nerve fiber layer dropout. The cholesterol plaque noted earlier was still present in the same position.
Our patient had an area of retinal ischemia, segmental optic disc edema, choroidal non-perfusion, and a Hollenhorst plaque. Although a link between retinal artery occlusion and emboli can be established through ophthalmoscopy, it is more difficult to establish a relation between AION and embolic disease because the short posterior ciliary arteries cannot be directly visualized. Horton (7) addressed this issue by reporting two cases of patients with cilioretinal artery occlusion secondary to emboli from the ipsilateral carotid artery. Because the cilioretinal artery is derived from the short posterior ciliary arteries, the presence of emboli within the cilioretinal artery suggests that embolic occlusion of the short posterior ciliary arteries could occur, giving rise to AION and choroidal non-perfusion.
Our case has clinical features that distinguish it from the presentations classically seen in non-arteritic and arteritic AION. The patient initially presented with an area of retinal ischemia and choroidal non-perfusion, suggesting that both the retinal and choroidal circulations were simultaneously affected. In addition, the delayed optic nerve head filling adjacent to the area of choroidal non-perfusion in the early-phase fluorescein angiogram suggests that optic nerve ischemia may have been present on initial presentation. Using fluorescein angiography, Arnold and Hepler (8) demonstrated a statistically significant delay in prelaminar optic nerve filling in patients with acute non-arteritic AION. The presence of ischemia in the retina, choroid, and optic nerve would be unusual for non-arteritic AION.
On one-week follow-up examination, optic nerve head edema developed with splinter hemorrhages, and a new cholesterol plaque was observed within a retinal arteriole. The low ESR and C-reactive protein levels, negative temporal artery biopsy, and stable clinical course without subsequent vision loss make giant cell arteritis an unlikely explanation for these findings.
Although the patient's carotid ultrasound examination failed to show significant stenosis, it did reveal the presence of an echolucent soft plaque in the ipsilateral carotid artery. There has been recent evidence associating echolucent carotid plaques with an increased risk of embolic ischemic cerebrovascular events, independent of the degree of stenosis and other cardiovascular risk factors (9,10). Other possible sources of emboli include the heart and the aorta. Although the transthoracic echocardiogram was negative, a transesophageal echocardiogram might better have ruled out cardiac or aortic sources of embolism.
There is little evidence to suggest embolism as the cause of AION. Although their patient was not examined in life, Lieberman et al (1) reported the only case of histologically-proven thromboembolic occlusion of a short posterior ciliary artery leading to infarction of the optic nerve head. Portnoy et al (2) reported a patient with evidence of AION who also displayed multiple cholesterol emboli within the retinal vasculature and choroidal non-perfusion. In that case, the evidence for embolic disease as a cause for the clinical event was only circumstantial. In addition, no ESR levels or temporal biopsy results were reported to definitively rule out giant cell arteritis as a cause. Tomsak et al (3) reported three cases that presented with AION and evidence of Hollenhorst plaques within the central retinal artery and its branches. Our case differs in that all three of those cases occurred in conjunction with invasive procedures (two bypass procedures and one cardiac catheterization) that may have iatrogenically dislodged the emboli. In addition, both bypass surgeries may have caused transient hypoperfusion of the optic nerve head leading to AION.
We believe that our case represents a unique presentation of embolization that led to retinal ischemia, AION, and choroidal non-perfusion. The presence of multifocal ischemic events, the appearance of a new Hollenhorst plaque during a follow-up examination, and the negative workup for giant cell arteritis distinguish our case from previously reported cases of embolic non-arteritic AION. An embolic origin should be considered in non-arteritic AION associated with choroidal perfusion abnormalities.
1. Lieberman MF, Shahi A, Green WR. Embolic ischemic optic neuropathy. Am J Ophthalmol 1978;86:206-10.
2. Portnoy SL, Beer PM, Packer AJ, et al. Embolic anterior ischemic optic neuropathy. J Clin Neuroophthalmol 1989;9:21-5.
3. Tomsak RL. Ischemic optic neuropathy associated with retinal embolism. Am J Ophthalmol 1985;99:590-2.
4. Hayreh SS. Anterior ischemic optic neuropathy: differentiation of arteritic from non-arteritic type and its management. Eye 1990;4:25-41.
5. Siatkowski RM, Gass JD, Glaser JS, et al. Fluorescein angiography in the diagnosis of giant cell arteritis. Am J Ophthalmol 1993;115:57-63.
6. Quillen DA, Cantore WA, Schwartz SR, et al. Choroidal nonperfusion in giant cell arteritis. Am J Ophthalmol 1993;116:171-5.
7. Horton JC. Embolic cilioretinal artery occlusion with atherosclerosis of the ipsilateral carotid artery. Retina 1995;15:441-4.
8. Arnold AC, Hepler RS. Fluorescein angiography in acute nonarteritic anterior ischemic optic neuropathy. Am J Ophthalmol 1994;117:222-30.
9. Mathiesen EB, Bonaa KH, Joakimsen O. Echolucent plaques are associated with high risk of ischemic cerebrovascular events in carotid stenosis: The Tromso Study. Circulation 2001;103:2171-5.
10. Gronholdt ML, Nordestgaard BG, Schroeder TV, et al. Ultrasonic echolucent carotid plaques predict future strokes. Circulation 2001;104:68-73.
© 2006 Lippincott Williams & Wilkins, Inc.