Journal of Neuro-Ophthalmology, December 1985, Volume 5, Issue 4

Branch retinal artery occlusion in the Churg-Strauss syndrome.

Amaurosis fugax followed by retinal infarction occurred as an early manifestation of exacerbation in a 46-year-old woman with allergic angiitis and granulomatosis (Churg-Strauss syndrome). There was evidence of widespread intraluminal branch retinal artery occlusions without accompanying retinal vasculitis. Low molecular weight dextran and heparin, in addition to systemic steroid therapy, appeared to be beneficial. A review of the ophthalmic manifestations of the Churg-Strauss syndrome and of other hypereosinophilic conditions suggests that thromboembolism resulting from a hypercoagulable state may be as important as vasculitis in causing visual symptoms. Specific therapy attempting to alter blood coagulation and rheology may have an important role in these conditions.

J. Clin. Neuro-ophtludmol. 5: 229-237, 1985 © 1985 Raven Press, New York Branch Retinal Artery Occlusion in the .Churg-Strauss Syndrome UNDA RABINOWITZ DAGI, M.D. JON CURRIE, M.B., B.S., F.R.A.C.P. Abstract Amawosis fugax followed by retinal infarction oc­curred as an early manifestation of exacerbation in a 46-yeaJ'oOld woman with allergic angiitis and granulomatosis (Churg-Strauss syndrome). There was evidence of widespread intraluminal branch retinal artery occlusions without accompanying ret­inal vasculitis. Low molecular weight dextran and heparin, in addition to systemic steroid therapy, appeared to be beneficial. A review of the ophthalmic manifestations of the Churg-Strauss syndrome and of other hypereosinophilic condi­tions suggests that thromboembolism resulting from a hypercoagulable state may be as important as vasculitis in causing visual symptoms. Specific therapy attempting to alter blood coagulation and rheology may have an important role in these con­ditions. Allergic angiitis and granulomatosis (Churg­Strauss syndrome) forms part of the spectrum of disseminated systemic necrotizing vasculitis, of which classic pol~arteritis nodosa is the best recognized entity.1- With polyarteritis nodosa, Churg-Strauss syndrome shares the features of fibrinoid necrosis of small- and medium-sized muscular arteries, causing multisystem organ involvement.3- 6 However, in addition, Churg­Strauss syndrome is associated with an allergic diathesis, adult-onset asthma, fever, transient pulmonary infiltrations, high levels of periph­eral eosinophilia, and eosinophilic tissue infil­tration and granuloma formation, all of which are uncommon in classic polyarteritis nodo­sa. I -5.7 Histologically there is also a substantial degree of inflammatory involvement of small vessels such as capillaries and venules, which does not occur in polyarteritis nodosa.3,7 Tissue ischemia is the "universal common de­nominator" of the vasculitides and most atten­tion has been focused on vascular occlusion From the National Eye Institute, National Institutes of Health, Bethesda, Maryland. Write for reprints to: Jon Currie, Nationallnstilutes of Health, Building 10, Room 105227, Bethesda, MD 20892. December 1985 caused by acute inflammation or by sustained fibrotic narrowing of the vessel wall after acute inflammation has subsided.s This certainly ap­plies to the ocular manifestations of polyarteritis nodosa, for which reports of retinal, choroidal, and conjunctival ischemia invoke an underlying arteritis. 9- 12 Reports of ophthalmic complica­tions in Churg-Strauss syndrome are much rarer, but again they usually implicate the un­derlying vasculitis. 1j- 15 We present a case of Churg-Strauss syn­drome in which amaurosis fugax and eventual retinal infarction were clearly caused by branch retinal artery thromboembolism rather than by retinal artery vasculitis. From this case, together with a review of the few reports of ophthalmic complications in Churg-Strauss syndrome and a comparison with the reported ophthalmic manifestations of the hypereosinophilic syn­drome, we suggest that, in Churg-Strauss syn­drome, thromboembolism resulting from a hy­percoagulable state may be as important as vas­culitis in causing visual symptoms. This has important implications for specific therapy for such visual loss in the Churg-Strauss syn­drome. Case Report A 46-year-old Asian woman with Churg­Strauss syndrome was referred for neuro­ophthalmic evaluation following sudden visual loss in the right eye. The clinical diagnosis of Churg-Strauss syndrome had been made 12 months before on the basis of adult-onset asthma at age 40, intermittent petechial and purpuric vesicular rash, fevers, recurrent sinu­situs, and a mononeuritis multiplex involving both motor and sensory nerves. Laboratory studies had shown white blood cell counts ranging from 10,000 to 20,Ooo/mm3, an eosino­philia of 10-70%, platelet counts from 250,000 to 600,OOO/mm3, and sedimentation rates (Wes­tergren) as high as 140 mmlh. Biopsy material from skin, gastrocnemius muscle, and sural nerve had shown acute and chronic vasculitis with a prominent eosinophilic 229 Churg-Strauss Syndrome with Retinal Artery Occlusion component and focal areas of perivascular in­flammatory infiltration with eosinophils, plasma cells, and some small and large lympho­cytes (Fig. 1). While some arteries showed nar­rowing and occlusion from vasculitic changes in their walls, a few small arteries appeared to be occluded by intraluminal collections of eosino­phils and fibrin debris, without vasculitic changes (Fig. 2). Bronchial and nasal sinus biop­sies showed eosinophilic subepithelial inflam­mation and soft tissue infiltration with eosino­phils, plasma cells, and occasional lympho­cytes. Like the clinical and laboratory features, these histologic changes were consistent with the diagnosis of Churg-Strauss syndrome.3•7 The patient's clinical condition remained rel­atively stable, with maintenance prednisone therapy of alternating daily doses of 60 and 10 mg. Four months prior to her sudden visual loss in the right eye, she had suffered three epi­sodes, 3 h apart, of blurring of the visual field in her left eye. Each episode lasted only a few minutes, but a slight residual blurring of vision remained for about 1 week. On the day before her visual loss, the patient experienced three episodes of transient obscuration of vision in the right eye, one predominantly affecting the upper visual field, one predominantly the lower visual field, and the third involving the whole field of vision. These lasted from 5 to 30 min before resolving, but slight residual clouding of vision persisted. A further episode of amaurosis occurred the next day while she was under­going a medical examination. On funduscopy at that time, white embolic material was seen passing through the superior temporal retinal artery. Several minutes after this material had migrated distally and disappeared from view, the patient's vision returned to its baseline level. However, within several hours, she again noticed a dense central scotoma obscuring vi­sion in the right eye. This time the scotoma did not resolve but became more extensive during the next 2 h. Visual acuity was reduced to 201 200 in the right eye and remained 20/13 in the left eye, with a relative afferent pupillary defect on the right. Slit lamp examination showed no' evidence of inflammatory cells in the anterior chamber or vitreous. Intraocular pressures were normal, as were extraocular movements. Gold­mann perimetry demonstrated a large central scotoma in the right eye, extending superonas­ally (Fig. 3, top left) and no abnormalities in the left eye. In the right fundus there was white material occluding both the superior and infe­rior retinal artery bifurcations at their origins on the optic disc (Fig. 4a). Venous return through the inferior retinal vein was severly impeded at the arteriovenous crossing. The blood column in the inferotemporal arterial branches was seg­mented and stationary and further areas of white material could be seen occluding more distal arterial bifurcations in both the superior and inferior arterial arcades (Figs. 5 and 6a). A segmental area of retinal nerve fiber layer edema and opacification extended from the optic disc to the macula, with an incipient mac­ular cherry-red spot. However, there were no hemorrhages, exudates, or vessel changes sug­gesting an underlying retinal vasculitis, and the Figure 1. Biopsy from gastroc­nemius muscle showing vascu· lilic occlusion of a small vessel and perivascular infiltration, with prominent eosinophUic component. 230 Journal of Clinical Neuro-ophthalmology Dagi, Currie Figure 2. Biopsy from gastrocnemius muscle shOWing occlusion of a small ar­teriole by an intraluminal collection of eo­sinophils, without vasculitic changes. left fundus was normal. Laboratory studies showed a white blood cell count of 19,400/mm3 with 48% eosinophils, hemoglobin level of 11.4 gldl, platelet count of 450,OOO/mm3, and eryth­rocyte sedimentation rate (Westergren) of 45 mm/h. Bilateral carotid arteriogram and echo­cardiogram were normal. The clinical evidence for an embolic retinal vascular lesion was so strong that therapy was begun with low molecular weight dextran and heparin, in addition to increasing the predni­sone dosage to 100 mglday. Within 24 h, visual acuity in the right eye had improved to 20/50, with considerable decrease in the size of the central scotoma (Fig. 3, top right). The white material occluding the bifurcations of the su­perior and inferior retinal arteries was no longer present (Fig. 4b). There were still peripheral oc­clusions in the inferior temporal vessels, but distal blood flow had improved (Fig. 6b). FIuo- Pigure 3. Goldmann perimetry for the right eye showing initial dense central scotoma (top left), with continued resolution at 24 h (top right), 14 days (lower left), and 12 months (lower right) after ONet of visual loss. December 1985 231 Churg-Strauss Syndrome with Retinal Artery Occlusion (a) (b) Figure 4. a: Occlusions at the bifurcations of the superior and inferior retinal arteries (arrows) with impaired venous return at the inferior arteriovenous crossing (arrow). b: Resolution of the arterial obstruction within 24 h and improvement of inferior retinal vein flow. rescein angiogram showed focal arterial occlu­sions without flow, but no evidence of retinal vasculitis such as dye leakage from arteries, veins, or optic disc. By 36 h, there was resto­ration of flow in the inferior temporal arcade vessels (Fig. 6c), with only small areas of pe­ripheral focal obstruction where bifurcation oc­clusions had previously been. Retinal edema persisted in the perimacular area. Fluorescein 232 angiography showed only one area of focal dye leakage, at a site of previous occlusion in the inferior branch artery (Fig. 7a). Again there was no evidence of retinal vasculitis. This area of focal dye leakage was no longer present when the fluorescein angiogram was repeated 3 days later (Fig. 7b). The low molecular weight dextran and hep­arin were ceased after 5 days, at which time Journal of Clinical Neuro-ophthalmology visual acuity in the right eye was 20/25, but with perimacular retinal edema and a small central scotoma still resolving. Over the next 2 weeks, increasing cough and asthma, new ischemic skin lesions, and an increasing eosinophilia, platelet count, and erythrocyte sedimentation rate necessitated the addition of cyclophospha­mide to the patient's treatment regimen, with considerable clinical improvement. On review 12 months later, visual acuity in the right eye was 20/25, with a small residual central scotoma (Fig. 3, lower right) and an in­ferior maculopapular nerve fiber bundle defect as the only fundus abnormality. No further ep­isodes of transient amaurosis had occurred. Discussion There are very few reports of ophthalmic complications in the Churg-Strauss syndrome, and only occasionally do these involve the ret­inal vasculature. Cury et al. 13 reported a chronic bilateral uveoscleritis associated with progres­sive visual loss, bilateral disc edema, and optic nerve inflammation in a 67-year-old woman with Churg-Strauss syndrome. Histology avail­able for the most affected eye showed necro­tizing eosinophilic granulomatous inflammation involving sclera, ciliary body, iris, vitreous, optic nerve, and choroid. Retinal vessels were "almost entirely replaced by inflammatory reac- Dagi, Currie tion." Although transient vJsual obscurations occurred in this patient, they are more likely to have been related to her disc edema16 than to transient retinal artery ischemia. Meisler et al. 14 reported conjunctival inflam­mation and amyloidosis in a 32-year-old woman with Churg-Strauss syndrome. Again, his­tology showed a diffuse extravascular eosino­philic inflammatory infiltrate, this time in­volving the conjunctiva. No clinical evidence of retinal vascular inflammation was noted. Chumbley and associates2 make brief mention of marginal corneal ulceration in one of 30 cases with Churg-Strauss syndrome that they re­viewed. In 1983, Weinstein et aLI5 described neuro­ophthalmic complications in two cases of Churg-Strauss syndrome. One patient had a left fourth nerve palsy with an otherwise normal eye examination. Rackemann and Greenl7 had also previously described fourth nerve palsy in the Churg-Strauss syndrome. The second case reported by Weinstein et al. was that of a 61-year-old man with visual loss and ischemic optic neuropathy in the right eye and symptoms of amaurosis fugax together with scattered retinal infarcts in the left eye. Al­though a diagnosis of retinal vasculitis was made, in fact the retinal vessels in both eyes were normal, there was no evidence of intra­ocular inflammatory cells, and only a single ret- Fipre S. Multiple branch occlusions with white intraluminal material in the superior retinal arterial arcade at the lime of initial visual loss. December 1985 233 Fjgu~ 6. a: Intraluminal arterial occlusion in a distal in­ferotemporal retina1 artery at the lime of visual loss (ar­rows), with poor distal blood flow. b: Twenty-four hours later, some residual arterial occlusion remains (arrows), but distal flow has improved. c: Thirty-six hours later, resolution of arterial occlusion (arrows) and restoration of distal flow. (a) (el Churg-Strauss Syndrome with Retinal Artery Occlusion (b) inal hemorrhage was noted along the right su­perotemporal vascular arcade. Most recently, Ibanez Bermudez et at. 18 re­ported an episode of visual loss with "central retinal artery occlusion" in a 44-year-old woman with Churg-Strauss syndrome. Although the clinical details concerning this episode are ex­tremely sketchy, it was apparently reversible and occurred during a period of severe exacer­bation of the patient's underlying asthma. There is no suggestion that any retinal vascular abnormalities were present other than the cen­tral retinal artery occlusion. Thus, in only one reported case of Churg­Strauss syndrome is there convincing evidence of retinal vasculitis,13 and that occurred in the setting of chronic panocular inflammation, a sit- 234 uation quite different from that in our case and in those of Weinstein et al. IS and Ibanez Ber­mudez et al. I8 In our patient, at least one epi­sode of amaurosis fugax occurred while embolic material was observed passing through the ret­inal arteries. Retinal infarction was accompa­nied by branch retinal artery occlusions with ag­gregations of intraluminal white material at vessel bifurcations (Figs. 5 and 6a). These arte­rial occlusions cleared rapidly within 36 h and were not associated with any concomitant oph­thalmoscopic evidence of arterial or venous vas­culitis or with any evidence of inflammatory cells in the vitreous. Fluorescein angiography showed only one area of focal arterial leakage, present at the time and site of an arterial 0b­struction and resolving within 3 days of the res- Joumal of Clinical Neuro-op~tha1mology FiPft 7. &: Fluorescein angiogram, 36 h after visual loss, showing focal leakage of dye from the inferior tem­poral arterial bifurcation illustrated in Fig. 6 (arrow). b: Three days later. the area of focal leakage has resolved (arrow). la) Ib) olution of the arterial obstruction (Fig. 7a and b). Such focal arterial leakage has been de­scribed as a hallmark of embolic branch retinal artery occlusion.19 There was no evidence of the more widespread leakage from arteries, veins, or optic disc itself that has commonly been de­scribed in various forms of retinal vasculitis. 20-23 There are many reports of different forms of retinal vasculitis occuring in conjunction with a December 1985 wide variety of inflammatory and vasculitic dis­eases. 11.12.21,24-32 Recently, however, some at­tention has been focused on intraarterial throm­boembolism without vasculitis as a cause of ret­inal arter~ occlusion in some of these con­ditions. 33- 6 Mechanisms such as immune com­plex deposition, fibrin-platelet pluggin~ and hypercoagulability have been suggested. -38 A marked peripheral and tissue eosinophilia 235 Churg-Strauss Syndrome with Retinal Artery Occlusion is a feature that distinguishes the Churg­Strauss syndrome from other necrotizing vas­culitides. I - 4 In considering the possibility of nonvasculitic retinal thromboembolism in the Churg-Strauss syndrome, it is interesting to re­view the ophthalmic manifestations of the hy­pereosinophilic syndrome. Here, peripheral blood eosinophilia is accompanied by organ system involvement without evidence of a gen­eralized vasculitis. 39.40 Histologically, vessels often show occlusion by microthrombi together with perivascular eosinophilic infiltrates, without an accompanying vasculitis. 39.40 Evi­dence of choroidal and retinal vessel throm­boembolism is frequent (66 and 58%, respec­tively), with areas of choroidal and retinal ische­mia associated with both arterial occlusions and capillary dropout. 41 Gardner-Thorpe and colleagues41 noted occlusion of the right supe­rior retinal artery in one of four patients they described who had bilateral cerebral hemi­sphere damage from "disseminated eosino­philic collagen disease." Blacklock et al. 42 de­scribed recurrent episodes of transient retinal ischemia in an 18-year-old man with profound eosinophilia. On one occasion, "white emboli were observed in both inferior temporal retinal arteries." After therapy with heparin and leu­kapheresis, only one further episode of amau­rosis occurred. The common occurrence of nailbed splinter hemorrhages43 and evidence of arterial emboli throughout the body44 is also well described in the hypereosiniphilic syn­drome, an~ extensive peripheral thromboem­bolism may be a cause of death.45 Endocardial lesions are one source of emboli in the hypereosinophilic syndrome,43-4S and similar myocardial and endocardial inflamma­tion and fibrosis can occur in the Churg-Strauss syndrome.1 In our patient, however, echocar­diography was normal. Local arterial thrombus formation with peripheral embolus propagation may also arise in the hypereosinophilic syn­drome from direct endothelial cell damage pro­duced by Charcot- Leyden crystals,46 or fol­lowing activation of Ha~eman factor by eosin­ophilic cationic protein. 7 Our patient's retinal artery occlusion was an early manifestation of a period of increased disease activity. At the time of her visual loss, the peripheral eosinophil count was 9,300/mm3 and the platelet count was 450,OOO/mm3. Such eosinophilia and throm­bocytosis are both characteristic of the Churg­Strauss syndrome,S and both are risk factors for hypercoagulability and thromboembolism. Of particular interest in our patient is the fact that previous biopsies of skin, gastrocnemius muscle, and sural nerve had demonstrated some intraluminal vascular occlusions caused by collections of eosinophils and fibrin debris 236 (Fig. 2), in addition to the more frequently seen vasculitic occlusions of small arteries (Fig. 1). Chaine et al. 40 suggested that thromboembolic occlusions in the hypereosinophilic syndrome may be more widespread than currently sus­pected; this may also be the case in the Churg­Strauss syndrome and may be the basis for some of the complications in both syndromes. The presence of retinal artery emboli without vasculitis in our patient led us to treat her with low molecular weight dextran and heparin in an attempt to alter the blood rheology, in addition to increasing her regular anti-inflammatory therapy of steroids. We believe that this may have been helpful both in reducing her residual visual loss and in preventing additional epi­sodes of retinal embolization during a period of enhanced coagulability. This case emphasizes that the possibility of thromboembolism as well as vasculitis should be considered in the Churg-Strauss syndrome, particularly in patients presenting with symp­toms of visual loss, where direct observation of the retinal vessels is possible. References 1. Churg, J., and Strauss, L.: AUergic granuloma­tosis, allergic angiitis, and periarteritis nodosa. Am. J. Pathol. 27: 277-300, 1950. 2. Chumbley, L. c., Harrison, E. G., and De­Remee, R. A.: Allergic granulomatosis and an­giitis (Churg-Strauss Syndrome): report and analysis of 30 cases. Mayo Clin. Proc. 52: 477-484, 1977. 3. Fauci, A. S., Haynes, B. F., and Katz, P.: The spectrum of vasculitis: clinical, pathologic, im­munologic and therapeutic considerations. Ann. Intern. Med. 89: 660-667, 1978. 4. Fan, P. T., Davis, J. A., Somer, T., et al.: A clin­ical approach to systemic vasculitis. Semin. Ar­thritis Rheum. 9: 248-304, 1980. 5. Cupps, T. R., and Fauci, A. S.: The Vasculitides, vol. 21: Major Problems in Internal Medicine. W.B. Saunders, Philadelphia, 1981, pp. 6-50. 6. Zeek, P. M.: Periarteritis nodosa: a critical re­view. Am. J. Clin. Pathol. 22: 777-790, 1952. 7. Oonishi, T., and Ishiko, T.: Allergic granuloma­tosis and angiitis (Churg-Strauss). Acta. Pathol. lpn. 30: 335-345, 1980. 8. Moor~, P: M., and Cupps, T. R.: Neurological complications of vasculitis. Ann. Neurol. 14: 155­167, 1983. 9. Goldstein, 1., and Wexler, D.: The ocular pa­thology of periarteritis nodosa. Arch. Opht1ulimol. 2: 288-299, 1929. 10. Goldsmith, J.: Perarteritis nodosa with involve­ment of the choroidal and retinal arteries. Am. /. Ophthalmol. 29: 435-446, 1946. 11. ~?gan, D. G.: Corneosclerallesions in periarter­Ihs nodosa and Wegener's granulomatosis. Trans. Am. Ophtha/mol. Soc. 53: 321-344, 1955. 12. Sheehan, B., Harriman, D. G. F., and Bradshaw, J. P. P.: Polyarteritis nodosa with ophthalmic Journal of Clinical Neuro-ophthalmology and neurologic complications. Arc/I. Ophthalmol. 60: 537-547, 1958. 13. Cury, D., Breakey, A. 5., and Payne, B. F.: Al­lergic granulomatous angiitis associated with uveoscleritis and papilledema. Arch. Oplltllalmol. 55: 261-266, 1956. 14. Meisler, D. M., Stock, E. L., Wertz, R. D., et al.: Conjunctival inflammation and amyloidosis in al­lergic granulomatosis and angiitis (Churg­Strauss Syndrome). Am. /. Ophthalmol. 91: 216­219, 1981. 15. Weinstein, J. M., Chui, H., Lane,S., et al.: Churg-Strauss syndrome (allergic granulomatous angiitis). Arch. Ophthalmol. 101: 1217-1220, 1983. 16. Sadun, A. A., Currie, J. N., and Lessell, 5.: Tran­sient visual obscurations with elevated optic discs. Ann. Neurol. 16: 489-494, 1984. 17. Rackemann, F. M., and Green, J. E.: Periarteritis nodosa and asthma. Trans. Assoc. Am. Physicians 54: 112-118, 1939. 18. Ibanez Bermudez, F., Bonal-Pitz, P., Fernandez­Torres, c., et al.: A proposito de un caso de Churg Strauss con afeccion ocular. Med. Clin. (Barc.) 81: 769-771, 1983. 19. Muci-Mendoza, R., Arruga, J., Edward, W.O., et al.: Retinal fluorescein angiographic evidence for atheromatous microembolism: demonstration of ophthalmoscopically occult emboli and post­embolic endothelial damage after attacks of amaurosis fugax. Stroke 11: 154-158, 1980. 20. Gass, J. D. M.: Fluorescein angiography in en­dogenous intraocular inflammation. In Clinical Methods in Uveitis, S. B. Aronson, Ed. St. Louis, Mosby, 1968, pp. 202-229. 21. Cogan, D. G.: Retinal and papillary vasculitis. In Cant, J. S. (ed.): The William Mackenzie Cente­nary Symposium on the Ocular Circulation in Health and Disease. C. V. Mosby, St. Louis, 1%9, pp. 249-270. 22. Karel, I., Votockova, J., and Peleska, M.: Fluo­rescence angiography in unusual forms of idio­pathic retinal vasculitis. Ophtlullmologica 168: 446­461, 1974. 23. Lanham, J. G., Barrie, T., Kohner, E. M., et al.: SLE retinopathy: evaluation by fluorescein angi­ography. Ann. Rheum. Dis. 41: 473-478, 1982. 24. Coppeto, L., and Lessell, S.: Retinopathy in sys­temic lupus erythematosis. Arch. Ophthalmol. 95: 794-797, 1977. 25. Gold, D., Feiner, L., and Henkind, P.: Retinal arterial occlusive disease in systemic lupus ery­thematosis. Arch. Ophthalmol. 95: 1580-1585, 1977. 26. Donders, P. c.: Eales' disease. Doc. Ophtlullmol. U: 1-105, 1958. 27. Raab, E. L., Leopold, 1. H., and Hodes, H. L.: Retinopathy in Rocky Mountain spotted fever. Am. J. Ophthalmol. 68: 42-46, 1%9. 28. Oralesi, N., and Ricciardi, L.: Segmental retinal periarteritis. Am. J. Ophtha/mol. 72: 55-59, 1971. 29. Hayreh, S. S.: Optic disc vasculitis. Br. J. Ophthalmol. 56: 652-670, 1972. . 30. Coutu, R. E., Klein, M., Lessell, S., et al.: LIm­ited form of Wegener's granulomatosis: eye in­volvement as a major sign. j.A.M.A. 233, 868­871, 1975. December 1985 Dagi, Currie 31. Crouch, E. R., and Goldberg, M. F.: Retinal per­iarteritis secondary to syphilis. Arch. Ophthalmol. 93: 384-387, 1975. 32. Jampol, L. M., Isenberg, S. J., and Goldberg, M. F.: Occlusive retinal arteriolitis with neovas­cularization. Am. J. Ophthalmol. 81: 583-589, 1976. 33. Pfaffenbach, D. D., and Hollenhorst, R W.: Mi­croangiopathy of the retinal arterioles. j.A.M.A. 225: 480-483, 1973. 34. Susac, J. 0., Hardman, J. M., and Selhorst, J. B.: Microangiopathy of the brain and retina. Neu­rology 29: 313-316, 1979. 35. Schneiderman, J. H., Sharpe, J. A., and Sutton, D. M. c.: Cerebral and retinal vascular compli­cations of inflammatory bowel disease. Ann. Neurol. 5: 331-337, 1979. 36. Coppeto, J. R., Currie, J. N., Monteiro, M. L. R, et al.: A syndrome of arterial-occlusive retinop­athy and encephalopathy. Am. f. Ophthalmol. 98: 189-202, 1984. • 37. Von Sallmon, L., Myers, R E., Lerner, E. M., et al.: Vaso-occlusive retinopathy in experimental allergic encephalomyelitis. Arch. Ophtha/mol. 78: 112-120, 1967. 38. Aronson, A. J., Ordonez, N. G., Diddie, K. R, et al.: Immune complex deposition in the eye in systemic lupus erythematosus. Arch. Intern. Med. 139: 1312-1313, 1979. 39. Chusid, M. J.. Dale, D. c., West, B. C, et al.: The hypereosinophilic syndrome. Analysis of fourteen cases with review of the literature. Med­icine (Baltimore) 54: 1-27, 1975. 40. Chaine, G., Davies, J., Kohner, E. M. et al.: Ophthalmic abnormalities in hypereosinophilic syndrome. Ophthalmology 89: 1348-1356, 1982. 41. Gardner-Thorpe, C, Harriman, D. G. F., Par­sons, M., et aI.: Loffler's eosinophilic endocar­ditis with Balint's syndrome (optic ataxia and pa­ralysis of visual fixation). Q. 1- Med. 40: 249-260, 1971. 42. Blacklock, H. A., Cleland, J. F., Tan, P., et al.: The hypereosinophilic syndrome and leukapher­esis, letter. Ann. Intern. Med. 91: 650-651, 1979. 43. Brockington, I. F., and Olsen, E. G. J.: Loffler's endocarditis and Davies endomyocardial fibrosis. Am. Heart J. 85: 308-322, 1973. 44. Parillo, J. E., Borer, J. 5., Henry, W. L., et a\.: The cardiovascular manifestations of the hyper­eosinophilic syndrome. Prospective study of 26 patients, with review of the literature. Am. ]. Med. 67: 572-582, 1979. 45. Spry, C J. F., Davies, J., TaL P. C, et al.: Clin­ical features of 15 patients with the hypereosi­nophilic syndrome. Q. /. Med. 52: 1-22, 1983. 46. Dincsoy, H. P., Burton, T. J., and van der Bel­Kahn, J. M.: Circulating Charcot-Leyden crystals in the hypereosinophilic syndrome. Am. J. Clin. Patho/. 75: 236-243, 1981. 47. Venge, P., Dahl, R., Hallgren, R, et aI.: Cationic proteins of human eosinophils and their role in the inflammatory reaction. In The Eosinophil in Health and Disease, A. A. F. Mahmoud, Ed. Grune and Stratton, New York, 1980, pp. 131­144. 237