Journal of Neuro-Ophthalmology, December 1998, Volume 18, Issue 4

Retinal Ischemia in Aortic Arch Atheromatous Disease

Retinal ischemia is often caused by emboli arising from the cardiac chambers or the common carotid artery bifurcation; the latter are often composed of cholesterol. However, in many patients no lesions are identified after evaluation of these sources of emboli. Two patients were observed who had retinal ischemia and emboli originating from aortic atheromatous plaques that were visualized by transesophageal echocardiography. Cardiac, carotid, and intracranial sources of emboli were excluded. The embolic nature of retinal ischemia was further corroborated by the presence of microembolic signals during transcranial Doppler insonation of the middle cerebral artery on the side ipsilateral to the symptomatic retina. In patients with Hollenhorst plaques the aortic arch can be a potential source of emboli. Transesophageal echocardiography should be considered in these patients when the initial evaluation does not identify a cardiac or carotid lesion.

Journal of Neiim- Oplilhalnwfogy 18( 4): 237- 241, 1998. © 1998 Lippincolt Williams & Wilkins. Philadelphia Retinal Ischemia in Aortic Arch Atheromatous Disease Jose G. Romano, M. D., Viken L. Babikian, M. D., Christine A. C. Wijman, M. D., and Thomas R. Hedges III, M. D. Retinal ischemia is often caused by emboli arising from the cardiac chambers or the common carotid artery bifurcation; the latter are often composed of cholesterol. However, in many patients no lesions are identified after evaluation of these sources of emboli. Two patients were observed who had retinal ischemia and emboli originating from aortic atheromatous plaques that were visualized by transesophageal echocardiog­raphy. Cardiac, carotid, and intracranial sources of emboli were excluded. The embolic nature of retinal ischemia was further corroborated by the presence of microembolic signals during transcranial Doppler insonalion of the middle cerebral artery on the side ipsilateral to the symptomatic retina. In patients with Hollenhorst plaques the aortic arch can be a potential source of emboli. Transesophageal echocardiography should be consid­ered in these patients when the initial evaluation does not iden­tify a cardiac or carotid lesion. Key Words: Aortic arch plaque- Retinal embolism- Retinal ischemia. A common cause of transient monocular blindness and retinal artery occlusion is retinal embolism ( 1). Emboli may be composed of calcific material ( 2), fibrin and platelets ( 3), or, more commonly, cholesterol ( Hollen­horst plaques) ( 2,4). The latter often arise from athero­matous plaques at the common carotid artery bifurcation ( 5). However, as many as 50% of asymptomatic patients with Hollenhorst plaques have no evidence of carotid disease ( 6), and in two thirds a source of emboli could not be found after carotid evaluation and precordial echocardiography ( 7). That the aortic arch can be the source of retinal embolism is not well documented ( 8,9). We observed two patients with retinal embolism origi­nating from the aortic arch in whom the diagnosis was established by means of ultrasound and magnetic reso­nance imaging ( MRI) tests. Manuscript received November 17, 1997; accepted June 16, 1998. From the Departments of Neurology, University of Miami, School of Medicine ( J. G. R.), Miami, Florida; and the Boston University School of Medicine ( V. L. B., C. A. C. W.) and the Departments of Neurology and Ophthalmology, Tufts University School of Medicine ( T. R. H.), Boston, Massachusetts, U. S. A. Address correspondence and reprint requests to V. L. Babikian, M. D., Department of Neurology, Boston University School of Medi­cine and Boston Veterans Administration Medical Center, 150 South Huntington Avenue, Boston, MA 02130, U. S. A. CASE REPORTS Case 1 A 91- year- old man suddenly lost vision in the lower half of his left visual field. He had no associated oph­thalmologic or neurologic symptoms. Medical history in­cluded diet- controlled diabetes mellitus and coronary ar­tery disease, for which he took aspirin. There was no history of headache, jaw claudication, neck or shoulder pain, previous transient visual loss, or cerebral ischemic events. On examination 2 days after onset of symptoms, visual acuity was 20/ 70 OD and 20/ 50 OS. There was an inferonasal quadrantic visual defect in the left eye. Reti­nal infarction in the distribution of the left superior reti­nal branch artery secondary to an embolus was noted ( Fig. 1 A). General, cardiac, and neurologic examinations were unrevealing. Erythrocyte sedimentation rate was 3 I mm/ hr. A ca­rotid duplex study showed minimal plaque formation at the origins of both internal carotid arteries, causing di­ameter reductions estimated at less than 30%. Brain MRI showed an old lacunar infarct in the left caudate nucleus. A magnetic resonance angiogram showed normal intra­cranial arteries and left common and internal carotid ar­teries. A plaque causing 50% to 70% luminal diameter reduction was detected in the right common carotid ar­tery. A transthoracic echocardiogram ( TTE) showed hy-pokinesis of the left ventricle but no embolic source. A transesophageal echocardiogram ( TEE) showed calcified plaques in the aortic arch and the ascending and descend­ing aorta ( Fig. IB). The largest in the arch was 8 mm thick. No mobile elements were seen. Diffuse left ventricular hypokinesis was again observed, and no in­tracardiac thrombus was found. A transcranial Doppler ultrasound ( TCD) study performed 7 days after onset of symptoms showed more than 60 microembolic signals in the left middle cerebral artery during a 30- minute recording ( Fig. 1C). No signals were detected on the right side. Microembolic signals persisted in the left middle cerebral artery and were again absent on the right during a repeat TCD study 6 days later. The laboratory's TCD methods have been described in a previous report ( 10). A diagnosis of retinal embolism and infarction was 237 238 J. G. ROMANO ET AL. made. The suspected source of emboli was an aortic arch plaque distal to the right and proximal to the left com­mon carotid artery ostia. Case 2 A 67- year- old man had a history of two attacks of transient loss of vision of the right eye. He described both as though a shade were being pulled over the eye. The attacks were 24 hours apart. The first lasted 3 hours and the second lasted 30 minutes. There were no other accompanying neurologic or cardiovascular symptoms. His medical history included hypertension, hyperlipid-emia, chronic renal insufficiency, and diabetes mellitus. Approximately 1 year previously, he had had transient symptoms of vertebrobasilar ischemia for which he was prescribed warfarin. International normalized ratio was 3.0 at the time of present symptoms. Physical examina­tion showed normal visual acuity, visual fields, and pu­pillary responses. Funduscopic examination of the right eye showed white material in the original segment of the inferior branch retinal artery and an embolus at the first bifurcation of the temporal branch artery ( Fig. 2A). In FIG. 1. Case 1. Funduscopic evaluation of the left eye ( A) shows an ischemic retinal whitening in the superior retina and an em­bolus in the superior branch retinal artery ( arrow). A visual fixa­tion artifact is to be noted ( curved arrow). A calcified plaque ( arrows) of the ascending aortic arch is seen on transesophageal echocardiogram ( B). Transcranial Doppler ultrasound monitoring ( C) shows a microembolic signal ( arrow) in the left middle cere­bral artery. addition, a small embolus was seen in the inferior branch of the retinal artery. Six days later, a new shiny retractile embolus was seen at the second bifurcation of the supe­rior temporal branch retinal artery ( Fig. 2B). On general examination, several toes on both feet appeared blue, and no pedal pulses were found. Cardiac and neurologic ex­aminations were both unremarkable. Erythrocyte sedimentation rate was 34 mm/ hour. A brain MRI showed extensive, chronic, periventricular ischemic changes and bilateral basal ganglia lacunes. Minimal plaque was seen by duplex ultrasound at both common carotid artery bifurcations. A TCD study and magnetic resonance angiography studies indicated mod­erate bilateral siphon stenoses. Transthoracic echocardi­ography showed left ventricular hypertrophy and mild dilatation of the aortic root and raised the possibility of debris in the aortic arch. Transesophageal echocardio­graphy showed a large plaque 10 mm in thickness in the ascending aorta; in addition, smaller plaque formation was noted in the arch and descending aorta ( Fig. 2C). A cardiac source of emboli was not found. A TCD study, performed 5 days after onset of symptoms, detected five ./ Neiim- Ophlhulmol, Vol. IS. No. 4, 1998 AORTIC ARCH ATHEROMATOUS DISEASE 239 FIG. 2. Case 2. Funduscopic evaluation of the right eye ( A) shows embolic material in several retinal arteries ( arrows). Six days later ( B), an additional shiny retractile embolus is seen ( arrow). Transesophageal echocardiogram shows a plaque in the ascending aorta ( C). A red col­ored microembolic signal is detected in the right middle cerebral artery during transcranial Doppler testing ( D). t& ff- :> o- 9| U M ; . • ? • * * D 1 1 •• : . : • • • • ' • . ' • . . • • • : m microembolic signals in the right middle cerebral artery during a 30- minute recording ( Fig. 2D). Microembolic signals persisted a month later. Further evaluation showed progressive renal dysfunc­tion. A diagnosis of cholesterol embolism from aortic arch plaque with involvement of the retinal, renal, and peripheral vascular circulations was made. DISCUSSION The two patients described sought medical attention because of symptoms of retinal ischemia. Transesoph­ageal echocardiograms showed thick, presumably athero­sclerotic, plaques in their aortic arches, and in both cases the embolic nature of retinal ischemia was corroborated by the observation of emboli in arterioles of the symp­tomatic retinas in addition to TCD microembolic signals along the distributions of the affected internal carotid arteries. These findings indicate that aortic arch plaques can be the source of embolism in the retina and can be diagnosed with ultrasound and MRI techniques. Although severe stenoses of the cervical internal ca­rotid artery are considered a common origin of retinal emboli, the latter can also originate from other sources. Previous reports have described external carotid artery disease ( 11), atrial myxomas ( 12), myocardial infarcts with mural thrombus formation ( 13), mitral valve pro­lapse ( 14), and prosthetic valves ( 15) as sources of em­bolism. Common carotid artery bifurcation and cardiac lesions were ruled out in our patients. In addition, the J Neuro- Oplilhiilnwl, Vol. IK. No. 4. I99K 240 J. G. ROMANO ET Ah. finding of microembolic signals in the middle cerebral arteries on the symptomatic sides ruled out the possibil­ity of embolism from ophthalmic artery lesions. Further­more, Hollenhorst plaques arise from aortic or carotid atherosclerotic plaques, eliminating the possibility of a cardiac source of emboli. The aortic arch has previously been suspected to be the source of retinal emboli in people with the disseminated cholesterol embolism syn­drome ( 8,9). Cerebral, cardiac, renal, mesenteric, and peripheral vascular embolism can occur in this context ( 8,9,16,17). Our second patient had clinical or laboratory evidence of retinal, cerebral, and systemic embolism and illustrates the syndrome. The findings in our first patient suggest, however, that retinal embolism from aortic arch plaques can occur in the absence of clinically evident cerebral or systemic embolization. To our knowledge, this has not been recognized in previous studies. Thus, the aortic arch may be the source of isolated retinal em­bolism and should be studied in patients with nonreveal-ing carotid and cardiac evaluations. Aortic arch atheromas have been increasingly recog­nized as a potential source for cerebral embolism ( 18). Atheromas more than 4 mm thick, particularly when ul­cerated or mobile, have been associated with cerebral infarction ( 19). They are also predictors of recurrent vas­cular events and are often found in patients with cerebral infarction of undetermined cause ( 20). Until recently, these lesions were missed during the regular evaluation of patients with retinal or cerebral ischemia because available diagnostic tests, such as aortic angiography, were not sensitive enough to detect them. With the de­velopment of TEE as a relatively safe ( 21) and effective method of studying the aortic arch ( 22), there has been a growing awareness of the importance of evaluating this arterial segment in patients with retinal or cerebral em­bolism in whom cardiac and carotid evaluations have been unrewarding. In our patients, TEE helped identify the source of emboli that was suspected from fundus-copic examinations and TCD studies. It is to be noted that technically adequate TTE studies did not permit sat­isfactory imaging of the aortic arches in either of our patients. This is expected, given the limitations of the technique, and is similar to previous experience with retinal embolism ( 23). The superiority of TEE over TTE in detecting cardioembolic sources has been well docu­mented, with TEE revealing an embolic source in as many as 50% of patients with a negative TTE ( 24,25). Regarding the aortic arch, TTE cannot reliably image this area and often misses atherosclerotic plaques found by TEE ( 22). Thus, a TEE study may be indicated in the patient with retinal ischemia after a nonrevealing work up. Transcranial Doppler ultrasound signals similar to the ones detected in our patients can correspond to in vivo microemboli composed of platelet fibrinogen, choles­terol, or gaseous material ( 10,26). In patients with com­mon carotid artery stenosis these signals have been as­sociated with cerebral ischemia, severe stenosis, and plaque ulceration ( 27- 30). In the cases presented above, their presence identifies the link between aortic arch plaques and retinal emboli. It also indicates that micro-embolism is not a phenomenon of short duration; rather, it may persist for at least a month after a symptomatic event. The finding is consistent with the high risk of recurrent events in patients with large aortic arch plaques. The aortic arch is a potential source of emboli in pa­tients with Hollenhorst plaques. These are associated with significant morbidity and mortality. Although the risk of stroke after transient monocular blindness of all causes is approximately 2% per year ( 31), patients with asymptomatic retinal cholesterol emboli have an annual stroke incidence of up to 8.5% ( 32). The mortality rate is also increased when compared with that in age- and sex-matched control subjects ( 33- 35) and is mostly caused by coronary artery disease and cerebrovascular disease ( 35). Early detection of aortic arch plaques may enable the initiation of appropriate therapy. The optimal treatment of patients with aortic arch atheromatous embolization remains to be defined. Al­though some early reports suggested that anticoagulation may potentiate cholesterol embolization by preventing adequate thrombosis over the atheromatous plaque ( 36- 38), others have reported the disappearance of mobile elements from aortic arch plaques in anticoagulated pa­tients ( 39- 42). This beneficial effect is most likely re­lated to prevention of thrombus formation over an ulcer­ated plaque. However, anticoagulation does not exert ab­solute protection against platelet- fibrin embolization. Laperche et al. ( 43) reported recurrent embolic events in 27% of 15 patients with aortic arch plaques, despite hep­arin therapy. There is some evidence of stabilization of these plaques and reduction of their emboligenic poten­tial with lipid- lowcring drugs ( 44). Successful aortic end-arterectomy ( 45) or aortic arch replacement ( 46) has also been reported for prevention of recurrent aortic embo­lization. In summary, aortic arch plaques can be a source of cholesterol embolism to the retina and the brain. A TEE study should be considered in patients with symptomatic retinal embolism whose initial evaluations do not iden­tify a carotid or cardiac source for embolism. Acknowledgment: The authors thank Mr. Lewis C. Bittle for his help with the review of echocardiography tapes. REFERENCES 1. Wray SH. Visual symptoms ( eye). In: Bogousslavsky J, Caplan L, eds. Stroke syndromes. Cambridge, UK: Cambridge University Press, 1995; 68- 79. 2. Amiga J, Sanders MD. Ophlhalmological findings in 70 patients with evidence of retinal embolism. Ophthalmology 1982; 89: 1336- 47. 3. Fisher CM. Observations of the fundus oculi in transient monoc­ular blindness. Neurology 1959; 9: 333- 47. 4. Hollenhorst RW. Significance of bright plaques in the retinal ar­terioles. JAMA 1961; 178: 123- 9. 5. Kollarits CR, Lubow M, Hissong SL. Retinal strokes: I. Incidence of carotid atheromata. JAMA 1972; 222: 1273- 5. J Neum- Ophllmlmol. Vol. IS, No. 4, 1998 AORTIC ARCH ATHEROMATOUS DISEASE 241 6. Bunt TJ. The clinical significance of the asymptomatic Hollenhorst plaque. ./ Vase Surg 1986; 4: 559- 62. 7. Smit RL, Baarsma GS, Koudstaal PJ. The source of embolism in amaurosis fugax and retinal artery occlusion, lnt Ophthalmol 1994; 18: 83- 6. 8. Darsee JR. Cholesterol embolism: the great masqueradcr. South Med J 1979; 72: 174- 80. 9. Coppeto JR, Lessell S, Lessell IM, Greco TP, Eisenberg MS. Dif­fuse disseminated atheroembolism: three cases with neuro-ophthalmic manifestations. Arch Ophthalmol 1984; 102: 225- 8. 10. Babikian VL, Rosales R, Pochay V. Composition of particles as­sociated with embolic signals on transcranial Doppler ultrasonog­raphy. J Stroke Cerebrovasc Dis 1994; 4: 86- 90. 11. Burnbaum MD, Selhorst JB, Harbison JW, Brush JJ. Amaurosis fugax from disease of the external carotid artery. Arch Neurol 1977; 34: 532- 5. 12. Jampol LM, Wong AS, Albert DM. Atrial myxoma and central retinal artery occlusion. Am J Ophthalmol 1973; 75: 242- 9. 13. Zimmerman LE. Embolism of central retinal artery. Arch Ophthal­mol 1965; 73: 822- 6. 14. Lesser RL, Heinemann M, Borkowski H, Cohen LS. Mitral valve prolapse and amaurosis fugax. ./ Clin Neuro- Ophthalmol 1981; 1: 153- 60. 15. Ide CH, Almond CH, Hart WM, Simmons EM, Wilson RJ. He­matogenous dissemination of microemboli. Eye findings in a pa­tient with Starr- Edwards aortic prosthesis. Arch Ophthalmol 1971; 85: 614- 7. 16. Case Records of the Massachusetts General Hospital: case 2- 1991. N Engl,/ Med 1991 ; 324: 113- 20. 17. Beal F, Williams RS, Richardson EP, Fisher CM. Cholesterol em­bolism as a cause of transient ischemic attacks and cerebral infarc­tion. Neurology 1981; 31: 860- 5. 18. Amarenco P, Duyckaerts C, Tzourio C, Henin D, Bousscr MG, Hauw JJ. The prevalence of ulcerated plaques in the aortic arch of patients with stroke. N Engl J Med 1992; 326: 221- 5. 19. Amarenco P, Cohen A, Tzourio C, et al. Atherosclerotic disease of the aortic arch and the risk of ischemic stroke. N Engl J Med 1994; 331: 1474- 9. 20. Briley DP, Lababidi Z, Bowdler AJ, Krishnamurthy M, Cansino SP, Touchon RC. Aortic atherosclerosis and stroke. J Stroke Cere­brovasc Dis 1997; 6: 125- 9. 21. Daniel WG, Erbel R, Kasper W, et al. Safety of transesophageal echocardiography: A multiccntcr study of 10,419 patients. Circu­lation 1991; 83: 817- 21. 22. Tunick PA, Kronzon I. Protruding atherosclerotic plaque in the aortic arch of patients with systemic embolization: a new finding seen by transesophageal echocardiography. Am Heart J 1990; 120: 658- 60. 23. Sharma S, Naqvi A, Sharma SM, Cruess AF, Brown GC, for the Retinal Emboli of Cardiac Origin Group. Transthoracic cchocar-diographic findings in patients with acute retinal arterial obstruc­tion. Arch Ophthalmol 1996; 114: 1189- 92. 24. Pop G, Sutherland GR, Koudstaal PJ, Sit TW, de Jong G, Roelandl JRTC. Transesophageal echocardiography in the detection of in­tracardiac embolic sources in patients with transient ischemic events. Stroke 1990; 21: 560- 5. 25. Pearson AC, Labovitz AJ, Tatineni S, Gomez, C. Superiority of transesophageal echocardiography in detecting cardiac source of embolism in patients with cerebral ischemia of uncertain etiology. J Am Coll Cardiol 1991; 17: 66- 72. 26. Markus H, Loh A, Brown MM. Computerized detection of cerebral emboli and discrimination from artifact using Doppler ultrasound. Stroke 1993; 24: 1667- 72. 27. Sitzer M, Muller W, Siebler M, et al. Plaque ulceration and lumen thrombus are the main sources of cerebral microemboli in high-grade internal carotid artery stenosis. Stroke 1995; 26: 1231- 3. 28. Siebler M, Nachtmann A, Silzer M, et al. Cerebral microcmbolism and the risk of ischemia in asymptomatic high grade internal ca­rotid artery stenosis. Stroke 1995; 26: 2184- 6. 29. Tegeler CH, Burke GL, Dallcy GM, Stump DA. Carotid emboli predict poor outcome in stroke ( abstract). Stroke I993; 24: I86. 30. Babikian VL, Wijman CAC, Hyde C, et al. Cerebral microembo-lism and early recurrent cerebral or retinal ischemic events. Stroke 1997; 28: 1314- 8. 31. Hurwitz BJ, Hcyman A, Wilkinson WE, Haynes CS, Ulley CM. Comparison of amaurosis fugax and transient cerebral ischemia: A prospective clinical and arleriographie study. Ann Neurol 1985; 18: 698- 704. 32. Bruno A, Jones WL, Austin JK, Carter S, Quails C. Vascular outcome in men with asymptomatic retinal cholesterol emboli: a cohort study. Ann Intern Med 1995; 122: 249- 53. 33. Savino PJ, Glaser JS, Cassady J. Retinal slrokc: is the patient at risk? Arch Ophthalmol 1977; 95: 1 185- 9. 34. Howard RS, Ross- Russell RW. Prognosis of patients with retinal embolism. ./ Neurol Neurosio- g Psychiatry I987; 50: l 142- 7. 35. Pfaffenbach DD, Hollenhorst RW. Morbidity and survivorship of patients with embolic cholesterol crystals in the ocular fundus. Am J Ophthalmol 1973; 75: 66- 72. 36. Moldveen- Geronimous J, Merrian JC. Cholesterol embolization: from pathological curiosity to clinical entity. Circulation 1967; 35: 946- 53. 37. Feder W, Auerbach R. Purple toes: an uncommon sequelae of the oral Coumadin drug therapy. Ann Intern Med 1961; 55: 91 1- 7. 38. Brims FJ, Scgcl DP, Aplcr S. Control of cholesterol embolization by discontinuation of anticoagulant therapy. Am J Med Sci 1978; 275: 105- 8. 39. Freedberg RS, Tunick PA, Culliford AT, Tatelbaum RJ, Kronzon I. Disappearance of a large intraaortic mass in a patient with prior systemic embolization. Am Heart .1 1993; 125: 1445- 7. 40. Bansal RC, Pauls GL, Shankcl SW. Blue digit syndrome: transe­sophageal echocardiographic identification of thoracic aortic plaque- related thrombi and successful outcome with warfarin. ./ Am Soc Echocardiogr 1993; 6: 319- 23. 41. Blackshear JC, Jahangir A, Oldenburg WA, Safford RF. Digital embolization from plaque- related thrombosis in the thoracic aorta: identification with transesophageal echocardiography and resolu­tion with warfarin therapy. Mayo Clin Proc 1993; 68: 268- 72. 42. Pasierski T, Jasek S, Firek B, Przybylski A, Szwed 1- 1, Sadowski Z. Resolution of an aortic mobile mass with anticoagulation without evidence of arterial embolism. Clin Cardiol 1996; 19: 151- 2. 43. Laperche T, Laurian C, Roudaut R, Steg PG, for the Filialc Echocardiographic de la Sociclc Francaise dc Cardiologie. Mobile thromboses of the aortic arch without aortic debris. Circulation 1997; 96: 288- 94. 44. Cabili S, Hochman I, Goor Y. Reversal of gangrenous lesions in the blue toe syndrome with lovastatin: a case report. Angiologv 1993; 44: 821- 5. 45. Culliford AT, Tunick PA, Katz ES, Kronzon I, Galloway AC, Ribokavc GH. Initial experience with removal of protruding atheroma from the aortic arch: diagnosis by transesophageal echo, operative technique, and follow- up ( abstract). ./ Am Coll Cardiol I993; 21: 342A. 46. Bojar RM, Payne DD, Murphy RE, et al. Surgical treatment of systemic atheroembolism from the thoracic aorta. Ann Thome Surg 1996; 61: 1389- 93. J Nciiro- Oplillialiiiol. Vol. IK, No. 4. 1998