662 ท AMERICAN JOURNAL OF OPHTHALMOLOGY Fig. 11 (Goldberg). Schematic representation of several sea-fans (Stage III); i.e., right eye at 6:30, 8:30-10, 11, and 11:45 o'clock. The large arborizing sea-fan at 8:30-10 o'clock had spontaneously bled (dotted area) into the posterior pole (Stage IV). MARCH, 197 is likely that more cases with Stage III fund are included in this series than would be in cluded in a random sample of SC patients. I: remains to be seen what percentage of unse lected SC patients reach the various stages and substages of the classification proposed herein. The sea-fan lesion is an immediately recognizable fundus abnormality in SC disease Although not pathognomonic, it is highly characteristic of one stage (Stage III) in the development of proliferative retinopathy in SC disease. In perspective, then, this lesion represents one rather specific variety of proliferative retinopathy. I have also observed Stage III lesions (although less often) in other varieties of sickle hemoglobinopathies, including sickle thalassemia (Sthal), sickle cell disease (SS), and even in cases of sickle cell trait (SA). The precise factors leading to the development of the sea-fan and other varieties of Stage III lesions are not well understood. The earliest clinically observed fundus abnormality in SC disease is peripheral arterio · VOL. 71, NO. 3 PROLIFERATIVE SICKLE RETINOPATHY lar occlusion with essentially complete absence of retinal vascular perfusion in involved areas of the peripheral retina. While ischemia and hypoxia may thus exist in the inner layers of the retina, currently no specific information is available concerning ox"genation of the outer retinal layers in SC .isease. Obstructed arterioles enlarge and apparently anastomose with adjacent venules at the interface of ischemic and nonischemic portions of the retina. This presumably sets the stage for neovascular proliferation on the retinal surface. Possibly in response to some chemotactic or other stimulus, and seeming in an attempt to revascularize the ischemic retina, vasoproliferative lesions sprout anteriorly from these anastomoses. Uncommonly neovascular growth may occur directly from obstructed arterioles. With time, recurrent localized vitreous hemorrhages occur, which, upon organization, result in white fibrous components in the proliferating tissue. How often-if ever-fibrous proliferations occur in the absence of pre-existing vitreous hemorrhage in PSR is currently unkown. It appears, however, that most of these fibrotic lesions do arise only after antecedent hemorrhage. These lesions leave the surface of the retina and project into the vitreous at varying intervals, possibly as a result of processes of vitreous contraction similar to those observed in proliferative diabetic retinopathy. At any time in the development of neovascularization in PSR, massive vitreous hemorrhage, and, ultimately, retinal detachment may occur. Such detachments are associated with vitreous hemorrhage, contraction, and band formations, and are also usually associated with retinal tears. By defining substages according to Table 1, a useful separation of Stages I, II, III, and IV (Table 3) is obtained. However the four eyes with retinal detachments in this series (Stage V, Tables 2 and 3) were all classified substage 4, because all had retinal de 663 tachments of 91 degrees or more in extent. Although it may further complicate the classification, re-definition of Stage V's substages may prove to be necessary after further experience with SC detachments is gained. For example, one could define the substages of Stage V as: up to 90 degrees (1), 91-180 degrees (2), 181-270 degrees (3), and 271-360 degrees (4). Then, the four eyes with retinal detachment are separated into Stage V-2 (L.S., left eye), V-3 (T.B.), and V-4 (J.B. and right eye of L.S.), rather than including each in V-4 according to the current definitions of Table 1. Until additional cases are detected and evaluated, however, the cause of simplicity is served by defining the substages of all five major stages in the same way. Comparative study of proliferative retinopathies associated with other diseases (retrolental fibroplasia, diabetes mellitus, carotid hypotension, Takayasu's disease is of value in underscoring the apparent importance of ischemia in the development of PSR. In retrolental fibroplasia, for example, Patz and others have shown that the primary effect of hyperoxia in the immature retina is vasoconstriction and ischemia. In addition, hyperoxia has a direct cytotoxic, destructive effect on the vascular endothelium, with resulting obliteration of an affected vessel's lumen. After restoration of normal oxygen levels, the secondary response of vasoproliferation into ischemic areas of retina occurs. The peripheral and temporal predilection for this proliferative retinopathy, as well as the apparent purposeful attempt at revascularization of an ischemic retinal area, obviously resemble Stage III of PSR. In diabetes mellitus, Davis and associates' have suggested that ischemia is the underlying pathogenetic factor responsible for all stages of diabetic retinopathy, but especially for the proliferative phase. Taylor and Dobrees have shown that proliferative diabetic retinopathy preferentially affects the retinal quadrants in a pattern similar to that of pro 664 AMERICAN JOURNAL OF OPHTHALMOLOGY liferative sickle retinopathy-i.e., superotemporal, inferotemporal, superonasal, inferonasal. Knox9,10 has also reviewed reports that retinal neovascularization associated with atheromatous or surgically induced carotid insufficiency may be due primarily to ischemia of the retina. Similar evidence from fluorescein angiography is available in the case of the proliferative retinopathy of Takayasu's (pulseless) disease. 11-13 Charache and Conley have indicated that all events in sickle cell disease can be related to the sickling phenomenon, which in turn, may be considered a response to ischemia and hypoxia. MARCH, 197 ent that vaso-occlusive phenomena are the Although the high frequency of Stage I lesions in this series of patients with PSR underscores the importance of retinal ischemia in the development of subsequent neovascularization, the exact pathogenesis of The ocular fundi of 24 selected patients peripheral arteriolar occlusion in SC disease with sickle cell-hemoglobin C disease were has not yet been delineated. It is possible that studied by indirect ophthalmoscopy and the small caliber of involved peripheral vesfluorescein angiography. Based on these sels prevents passage of clumps of sickled erythrocytes. It is also possible that the oxy-proliferative sickle retinopathy (PSR) was studies, a new, quantitative classification of gen tension in the temporal retina falls to a level that is low enough to induce local sickling and thrombosis. The classification proposed in this paper reflects both the severity of the retinopathic process as well as the presumed natural course of events in its development. Since PSR can be quantitated relatively easily by means of this classification, it is now possi ble to describe the natural course of this disease in rather precise terms. It should also be possible to describe the effect of various types of prophylactic and therapeutic interventions in similar terms.15,16 In addition, use of this classification makes data on PSR from different institutions suitable for valid comparison. A previously offered classification of sickle retinopathy1 appears inadequate in view of the new findings from fluorescein angiography and indirect ophthalmoscopy. With the use of these techniques, it is appar devised, which reflects both increasing severity of the retinopathy as well as its natur. progression. The first sign of PSR is peripheral arteriolar occlusions (Stage I); 91% of eyes were affected. Peripheral arteriolar-venular anastomoses (Stage II) were observed in 85% of (Stage III), which develop from the A-V eyes. Neovascular and fibrous proliferations anastomoses, occurred in 81% of eyes. Seafan lesions are part of this stage. Vitreous hemorrhages (Stage IV) from the neovascu30% of eyes, and retinal detachment (Stage lar proliferations occurred in approximately V) was observed in 8% of eyes. Pathogenetic parallels appear to exist be tween PSR and other vasoproliferative disorders, such as diabetes mellitus, hypotensive vascular diseases, and retrolental fibroplasia. ACKNOWLEDGMENTS Hemoglobin analyses were carried out by the Hematology Division, Department of Medicine, VOL. 71, NO. 3 PROLIFERATIVE SICKLE RETINOPATHY Johns Hopkins Hospital. Photographic assistance was provided by Terry George and Norbert Jedock. REFERENCES 1. Welch, R. B, and Goldberg, M. F.: Sickle-cell emoglobin and its relation to fundus abnormality. ch. Ophth. 75:353, 1966. 2. Hannan, J. F.: Vitreous hemorrhages associited with sickle-cell hemoglobin C disease. Am. J. Ophth. 42:707, 1956. 3. Ryan, S. J., and Goldberg, M. F.: Anterior segment ischemia following scleral buckling in sickle-cell hemoglobinopathy. Submitted for publicaion. 4. Newcomb, N., Goldberg, M. F., and Welch, R. B.: Intravenous fluorescein photography of the ocuar fundus in sickle-cell anemia. Med. Biol. Illus. (London) 17:95, 1967. 5. Hyvärinen, L., Maumenee, A. E., George, T., and Weinstein, G.: Fluorescein angiography of the choriocapillaris. Am. J. Ophth. 67:653, 1969. 6. Patz, A.: Retrolental fibroplasia. Surv. Ophth. 14:1, 1969. 7. Davis, M. D., Myers, F. L., Engerman, R. L., Devenecia, G., and Magli, Y. L.: Clinical observations concerning the pathogenesis of diabetic reti1opathy. In Goldberg, M. F., and Fine, S. L. (eds.): Symposium on the Treatment of Diabetic Retinopathy. Washington, D. C., USPHS Publication, No. 1890, 1969. 665 8. Taylor E., and Dobree, J. H.: Proliferative diabetic retinopathy. Site and size of initial lesions. Brit. J. Ophth. 54:11, 1970. 9. Knox, D. L.: Ischemic ocular inflammation. Am. J. Ophth. 60:995, 1965. 10. Knox, D. L. Ocular aspects of cervical vascular disease. Surv. Ophth. 13: 245, 1969. 11. Kojima, K., and Awaya, S.: Ocular fundus of pulscless disease. Jap. J. Clin. Ophth. 18: frontispiece, 1964. 12. Kojima, D., Niimi, K., and Awaya, S.: The initial symptoms of pulseless disease. Jap. J. Clin. Ophth. 19: 1439, 1965. 13. Shikano, S., and Shimizo, K.: Atlas of fluorescence fundus angiography. Philadelphia, W. B. Saunders, 1968, p. 58. 14. Charache, S., and Conley, C. L.: Rate of sickling of red cells during deoxygenation of blood from persons with various sickling disorders. Blood 24:25, 1964. 15. Goldberg, M. F.: Natural history of untreated proliferative sickle retinopathy. Arch. Ophth. In press. 16. Goldberg, M. F.: Treatment of proliferative sickle retinopathy. Tr. Am. Acad. Ophth. Otol. In press. 17. Lieb, W. A., Geeraets, W. J., and Guerry, III, D.: Sickle-cell retinopathy. Acta Ophth. 58 (Suppl.): 1, 1959. 18. Goodman, G., von Sallmann, L., and Holland, M. G. Ocular manifestations of sickle-cell disease. Arch. Ophth. 58:655, 1957. Reprinted from AMERICAN JOURNAL OF OPHTHALMOLOGY |