Search

Electroretinographic studies were carried out in ten patients suffering from unilateral occlusion of the central retinal artery by using twin light flashes (interval of 500 msec), the intensity of which varied within a range of 5 log units. The pathologic conditions in the affected eyes caused a characteristic flattening of the intensity-response curves predominantly in the scotopic part of the curve. The shape changes of the intensity-response curves due to the relative refractory period following the first retinal responses are of a different nature. The procedure enables the quantitative analysis of the wide range of retinal intensity functions.

It is the purpose of this paper to present the result of the renal arteriosclerosis in the aged. Special references have been made on interrelation between the arterial measurement of renal biopsy materials and age, blood pressure, minimal pressure of central retinal artery and groups of several diseases. Methods. Eleven cases of coronary sclerosis, nine cases of cerebrovascular sclerosis (group A and B, by classification of research group for arteriosclerosis in education ministry), twenty-three cases of essential hypertension (by WHO's criteria), fifteen cases of chronic glomerulonephritis, eleven cases of diabetes mellitus and eight aged patients without above mentioned disease as the control were studied by percutaneous renal biopsy. The areas of arterial branches of a given PAS stained kidney section were projected on the projection screen (Nicon) and measured directly by planimeter. Narrowing index of each vessel was calculated as quotient of area of arterial space by area of arterial wall. In this paper less than 0.5 of narrowing index represent definite narrowing of the vessel. The branches of renal artery were divided into two groups according to the diameter arterioles with the diameter below 50 microns, and small arteries with the diameter from 50 to 200 microns. Result. The narrowing indexes of arterioles were not correlated with the age. Those of small arteries were proportional to the age. In age group of from 40 to 49, from 50 to 59, from 60 to 69 and over 70, 80.6%, 81.6%, 83.6% and 91.8% of small arteries showed less than 0.5 of narrowing indexes respectively. There were no correlation finded between narrowing indexes of arterioles and systolic or diastolic blood pressure. In the systolic hypertensive group 90.5% and in the non hypertensive group 79.3% of small arteries showed less than 0.5 of narrowing index. In the diastolic hypertensive group 89.0% and in the non hypertensive group 74.5% of small arteries showed less than 0.5 of narrowing index. In the cases showing higher minimum pressure of central retinal artery the narrowing indexes of arterioles inclined to be less than 0.5. But no definite relation was found between minimum pressure of central retinal artery and narrowing index of small arteries.

WHEN a coincidence is duplicated, one wonders whether the sum total of the circumstances do not have a causal relation rather than constitute an unrelated, chance happening. REPORT OF CASES Case 1. —On April 1, 1947, S. L., a man in his late fifties, was operated on for carcinoma of the head of the pancreas. Cyclopropane-oxygen-ether anesthesia was administered from 8:05 to 12:30 a.m., and cholecystojejunostomy was performed. The patient was in shock but recovered well. At 5 p.m. the same day he complained of blurred vision in the right eye. On April 2 ophthalmologic examination disclosed an abrasion of the right cornea, extending over half the surface. Examination of the fundus disclosed closure of the central retinal artery with a typical cherry red spot in the macula. The cornea healed; the vision never returned, and the impression was that of an unusual coincidence of abrasion of the cornea and

Patients with central retinal artery occlusion studied by fluorescein fundus angiography revealed remarkable slowing of dye flow. The single instance of total obstruction showed retrograde venous filling from disc capillaries. "Segmentation" of the stagnant blood imparts an easily detectable granular appearance to the fluorescent column, caused by absorption of fluorescence by the red blood cells. Dye flowed swiftly through the unobstructed cilioretinal arteries and in one instance retrograde filling of a branch of a central retinal artery was demonstrated. Capillary supply to the disc may be demonstrated by fluorescein contrast when not appreciable on direct examination. Accurate timing of the sequence of dye filling should be useful in the evaluation of some of the many therapies recommended in acute central retinal artery occlusion.

The causes of atrophy of the optic nerve are many and diverse. They include such unusual conditions as profuse hemorrhage and thallium poisoning, as well as more common causes, such as chronic glaucoma and cerebral tumor. In many cases the cause is a primary vascular lesion, angiospastic in nature, which occurs with acute retrobulbar neuritis, tobacco amblyopia and closure of the central retinal artery. The present article will not be concerned with atrophy of the optic nerve due to any of the aforementioned factors, but will be limited to syphilitic atrophy of the optic nerve. It is possible for tobacco, or glaucoma, or a cerebral tumor to cause optic nerve atrophy in a patient who has syphilis. However, with the help of the history, the ophthalmologic examination, the neurologic examination, the Wassermann test, the roentgenogram of the skull and a study of the visual fields, it is usually possible to

described during the injection of the contrast medium (iodopyracet [diodrast®], 35%). This is associated with filling of the external carotid system. Pain in and behind the eye is associated with filling of the internal carotid system. In all probability, the pain is due to irritation of sensory nerves in the blood vessels. Flashes of light have been described, but in the experience of my colleagues and myself that symptom has not been complained of. In occasional instances there is homolateral dilatation of the pupils for several hours. According to Schurr, there is momentary narrowing of the retinal vessels during the injection. This narrowing, he states, is followed by widening. We have not observed this. Petechial hemor¬ rhages occur in the skin of the face, in the conjunctivas, and in the retina. Falls and his colleagues studied 80 consecutive cases for the occurrence of this complication and found a high incidence. An allergic reaction to the contrast media (iodopyracet) was seen in one of our cases ; injection of thorotrast® did not produce any untoward reaction. Changes in visual acuity have been reported by Curtis and by Falls and associates. We have had homolateral loss of vision in a single case about four hours after the injection was made; this case was similar to that reported by Curtis. Also, we have observed a unique case. After the injection of iodopyracet into the right carotid artery, digital compression of the left carotid artery produced imme¬ diate total bilateral loss of vision. After stellate blocks had been performed, and within a half-hour, there was return of bare light perception in each eye. Twelve hours later the right half-fields began to function, and within 36 hours both visual fields were normal and vision was normal for each eye. At no time was there an ophthalmoscopic appearance of obstruction of the central retinal arteries. Thrombosis of the internal cartoid artery may account for blindness of the homolateral eye. The mechanism undoubtedly is thrombosis extending from the ophthalmic to the central retinal artery. In other instances transitory spells of blindness precede the development of the hemiplegia, and then the affected eye ceases to give further trouble. It would seem that in such cases the collateral circulation from the external carotid system to the ophthalmic artery is increased by the obstruction to more posteriorly situated vessels. In still other cases there are loss of vision, which is incomplete, and partial optic nerve atrophy. In these instances, in all probability, the arteries providing nutrition to the optic nerve are obstructed. We have seen retinal hemorrhages infrequently ; this is surprising, since they are not uncommon in cases of obstruction of the central retinal artery. Homonymous hemianopsia occurs more frequently than unilateral blindness. In four cases under our observation in such long-standing cases there was a

Eye ground findings recorded by the criteria of Keith-Wagener or Scheie in essential hypertension and cerebral arteriosclerosis, have proven to be of much value clinically. However, these observation are mostly morphological and do not express were hemodynamic changes. Studies in our department have been centered around the vessel reaction of verious organs caused by induced hypoxemia and the findings concerning the coronary, cerebral, hepatic, and renal circulations have been reported previously. Recently, the determination of retinal blood flow has been established by ophthalmologists and this prompts us to study the hemodynamics of the retinal circulation during the resting state and during induced hypoxemia with the purpose of investigating the pathophysiology of the retinal circulation and the possibility of advancing a step towards the diagnosis of retinal vessel function. Methods and Materials A) Methods used : Values concerning retinal hemodynamics during the resting state were calculated by measuring the brachial and retinal blood pressures and eye ground photography. Immediately after those Procedures a mixture of 10% O2 and 90% N2 was administered for 20 minutes and the above procedures were repeated to determine the retinal hemodynamics during the induced hypoxemia. B) Calculations i) Retinal blood flow was calculation by the Hagen-Poiseuille law. ii) Comparison of central retinal artery and brachial artery blood pressures. iii) Systolic blood pressure ratio iv) Diastolic blood pressure ratio v) Pulse pressure ratio vi) Further relative differential pressure ratio

The orbital vessels in rhesus monkeys were studied. The arterial supply is by the ophthalmic artery, a branch of the internal carotid artery in the cranial cavity. The ophthalmic artery gives multiple branches throughout its entire course: from the intracranial part, a prominent ophthalmic communicating artery; from the intracanalicular part, one or more branches; from the intraorbital part, the central retinal artery, lateral posterior ciliary artery, lacrimal artery, branch to superior rectus, medial muscular artery in common with medial posterior ciliary arteries, branch to medial rectus, ethmoidal artery, branch to levator palpebrae superioris, medial palpebral artery, diploic branch, and terminal branches (supraorbital, supratrochlear and, rarely, dorsal nasal), generally in that order from the proximal to the distal part of the artery. Their origin, course and supply are discussed. The arterial pattern of rhesus monkeys is compared with that of human beings. The venous drainage is by superior and inferior ophthalmic veins which join together near the superior orbital fissure to open ultimately in the cavernous sinus. Tributaries of each vein are described with particular reference to the ocular veins.

From the present study we may conclude that a central optic nerve vascular system exists always. An individualized arteriole is less frequent. Penetrating without joining the leptomeningeal network, it is extremely rare. The poor contribution of the central retinal artery to the nutrition of the optic nerve is proved not only by the anatomical investigation, but also by clinical facts, hydrodynamic relations, selective arteriolar injections and histopathological studies.

The classic ophthalmoscopic picture of total central retinal artery occlusion is well known, as is the histopathologic picture of old postocclusive change in the retina and optic nerve. The etiology of central retinal artery occlusion is varied, and is thought to include such diverse mechanisms as atheroma formation, 1 emboli of atheromatous 2 and non-atheromatous nature, 3 spasm of the central retinal artery, dissecting aneurysm, 4 migraine, 5 and foreign body obstruction. 6 The purpose of this paper is to report a case of central retinal artery occlusion in which the patient died 68 hours following his occlusion. Both eyes were subsequently studied histologically. Because postocclusive scarring had not had time to develop, the mechanism by which arterial occlusion had occurred could readily be seen. The associated retinal changes are also reported. Report of Case The patient, a 49-year-old white male physician with a long history of moderate hypertension with

The importance of intravascular pressure as a factor in both the pathogenesis and the progression of diabetic retinopathy has been the subject of considerable controversy ever since Hirschberg categorized the specific diabetic retinal changes in 1890.1 All previous investigators have studied the systemic blood pressure in diabetics and attempted to correlate the severity of the retinal changes with the presence or absence of hypertension. Initially it was stated that severe diabetic retinopathy did not exist without associated systemic hypertension.2,3However, it was not long until many cases of retinopathy in patients with normal or even lowered blood pressures were documented, and the concept of hypertension as a factor in diabetic retinopathy was ignored. One of the deficiencies in these studies was that only systemic blood pressure was measured, whereas it is the pressure in the central retinal artery that might more logically influence the diabetic retinopathy. The present

Embolism of the central retinal artery or of one of its branches is a rather frequent occurrence; embolism involving several branches of the arteria ophthalmica, however, seems to be rare enough to justify the reporting of a new case. REPORT OF CASE History .—A white man, aged 31, an embroidery worker, eight years previous to examination had had gonorrhea of four months' duration without further complications and without other venereal disease. For several years following this, he had repeated attacks of an eruption of the skin chiefly localized at the outer sides of the legs. He did not have any history of disease of the lungs, and the family history in this respect was negative. He had been married for three years ; his wife had never been pregnant ; she was sickly. He had not suffered any serious injuries, and did not drink or smoke to any great extent. On

This paper is the summation of the study and correlation of many cases of sudden blindness caused by the abrupt closure of the central retinal artery, or a branch of the same. The conclusions have not been collected haphazardly but are presented after repeated examinations of the photographs and clinical histories. Textbooks, those proverbial exponents of repetitions, ancient and modern, rarely guide the student through the maze of generalizations to the correct diagnosis of changes in the retinal vessels. This is not said to disparage, for the compilation of any good textbook is an engaging, time-consuming task, but, rather, to ask the reader to start with an open mind for the reception of a new clinical differentiation between embolism of the central retinal artery and endarteritis. However, before the clearly defined groups are separated, it is essential to remember that there are cases of combined lesions, as, for instance, those

It has long been suspected that the optic atrophy and the associated visual field de­ fects which occur in established cases of chronic simple (open-angle) glaucoma are directly related to the rate of blood flow in the optic nerve. In 1939 to 1942, while studying the cor­ relation between visual field loss and systolic and diastolic blood pressure, as measured in the central retinal artery on the optic disc by the Bailliart and Kukan ophthalmodynamometers in a variety of conditions, I made the clinical observation that those patients with glaucoma who exhibited the most ex­ tensive and rapid loss of visual field were the ones with generalized arteriosclerosis, low brachial and retinal arterial blood pres­ sure, and increased intraocular pressure, sometimes of only moderate degree. Con­ versely, a relatively slow rate of visual field loss could be expected in an individual with increased intraocular pressure if the sys­ temic and retinal arterial blood pressure was elevated and arteriosclerotic changes were minimal in degree.

The inhalation of 100% oxygen at high atmospheric pressures is associated with a great increase in oxygen carrying capacity of the blood and produces a significant prolongation of visual function after occlusion of the retinal circulation by ophthalmodynamometric pressure on the eye. 1,2 It is not known, however, whether such hyperbaric oxygenation prevents or reverses the effects of retinal arterial occlusive disease. A significant reproducible improvement has been observed following inhalation of 100% oxygen at normal atmospheric pressure in three of four patients with this illness. 3 The present report describes the clinical and ophthalmologic findings following hyperbaric oxygenation in three patients with retinal arterial occlusive disease, caused by embolism or thrombosis of the central retinal artery or its branches. Report of Cases Case 1. —A 51-year-old white male textile worker entered the Durham Veterans Administration Hospital with the complaint of numbness of the left hand and transient

static endophthalmitis, is now extremely rare. Findings on physical examination include ciliary injection, exudate in the anterior chamber, swollen muddy iris with posterior synechiae, vitreous filled with exudate, fundus not visible, and blindness; pathologic findings include edema, thickening, and infiltration by white blood cells of the ciliary body, iris, choroid, and retina. 7 It is felt that ocular involvement is either secondary to the septicemia and extension from the vessels of the retina and choroid or rarely due to direct extension from the inflamed meninges along and around the optic nerve and central retinal artery. Endophthalmitis, panophthalmitis, suppurative iridochoroiditis, or metastatic ophthalmia occurs most frequently 3 days after the onset of iIlness but has been reported as early as 24 hours, s Meningococci are rarely isolated from the vitreous. The outcome varies from complete recovery to damage requiring enucleation. Endophthalmitis has been associated with meningococcal meningitis more frequently than with any other disease and has been seen most frequently in children who have recovered from the meningitis.

Optic nerve ultrastructure was studied in nine humans and 11 monkeys. Unmyelinated nerve fibers, belonging to the peripheral nervous system, lie in the adventitia of the central retinal artery, but all axons of the optic nerve itself are myelinated. The ultrastructure of axons and glial cells is the same as elsewhere in the central nervous system. Astrocyte cell bodies and processes insinuate themselves among the nerve fibers and nerve fiber bundles. An astroglial layer lines all the neuroectodermal surfaces that abut upon connective tissue, such as the pia mater, the adventitia of the central retinal vessels, and the septa. Blood vessels are confined to the connective tissue of the septa and the pia mater. The vessels are mainly capillaries with nonfenestrated endothelium. Adjacent endothelial cells seem to be held together by zonulae occludentes (tight junctions).

THE duct of Cloquet is a tubular area which traverses the eyeball in its anterior-posterior axis from the optic disk to the posterior pole of the lens. It develops at the expense of the central or original vitreous, which is progressively absorbed from the third month of intra-uterine life, during the formation of the definitive vitreous. In the human fqetus the duct of Cloquet exists from the sixth month and generally disappears shortly after birth. It contains the hyaloid artery, the ramifications of which form a vascular sheath embracing the posterior lens capsule. In young children, and even in adults, remnants either of the duct of Cloquet or of the hyaloid artery can sometimes be seen. The duct of Cloquet may persist in the shape of a membrane, which is inserted into some part of the optic disk, covers the emergence of the vessels and ends in the vitreous, either in the form of a point or club, or in the form of membranous expansions. Some types of hyaloid remnants are easily recognized. The following are some examples: (1) A white tubercle, bright and quite small found on the emergence of the central retinal artery like a bud on the bough of a fruit-tree. (2) An arterial handle leaving the optic disk and projecting into the vitreous. (3) A mobile cord or dark'filament, arising from the optic disk and ending by a free extremity in the vitreous. These prepapillary formations should not be mistaken for hyaline drusen at the optic disk or for the epipapillary membrane. The latter covers the whole of the optic disk and is attached to the vessels where they emerge. It may be opaque or transparent and somewhat ill-defined, and may encroach slightly upon the neighbouring retina. Very exceptionally one can find prepapillary cystic formations, probably related to remnants of the hyaloid artery. The cases of Durst (1909), Risley (1896), Koller (1900), Paton (1910), Levitt and Lloyd (1939), Yudkin (1926), Hilgartner (1940), Levine (1932), Purtscher (1936) and Merigot de Treigny and Subileau (1927) are

While all types of aneurysm of the larger branches of the central retinal artery are extremely rare, the occurrence of an uncomplicated, asymptomatic arterial aneurysm of congenital origin is unique. It is felt worth while, therefore, to report a case of this exceptional lesion and to review the related literature. REPORT OF A CASE A youth aged 19 was seen in the course of a routine examination of aviation cadets. The patient had never had trouble with his eyes, and there was no family history of ocular or cerebral disease. Vision in each eye was 20/15 (6/5). Ocular examination revealed nothing remarkable except for the fundus of the left eye. This presented a most unusual picture (figure). The optic disk was almost entirely obscured by a serpentine aneurysmal dilatation of the inferior temporal artery, which occupied the temporal seven eighths of the nerve head. The aneurysm consisted of sausage-like, saccular

Introduction The architecture and cellular structure of the retinal vessels of the human eye have been described previously. 1-5 In the present investigations flat preparations of the eyes of human fetuses and prematurely born infants were made. Development of the retinal vessels in the human fetus eye have been reported by Mann 6 and Michaelson. 1 Both state that the central retinal artery is recognizable at the 4th month. At this stage a small bulbous swelling appears on the trunk of the hyaloid artery where it passes through the disk. This enlarges and forms small buds which are the beginning of the main branches of the central retinal artery. Other arterial twigs arise from these and ramify in the center of the disk. By the 4th month they have vascularized a small area of the nerve fiber layer around the disk. Then they spread peripherally and at the same time

Case 1—A 58-year-old white man was first seen two months prior to admission with a one-week his­ tory of tearing in the left eye. Two days after the onset of symptoms, the eye became red and painful. The pain and tearing subsided spontaneously in two days but the redness continued. Examination one week after the beginning of symptoms revealed a normal right eye with visual acuity correctable to 20/20. The left eye was correctable to 20/50. There was a 2-freticulate injection of the conjunctiva, the cornea was steamy, and the anterior chamber was of average depth. The pupil was reactive but the iris stroma was covered with small, newly formed blood vessels. The left lens was clear and the optic nerve re­ vealed a slight pallor. A small flame-shaped hemor­ rhage was present at 2 o'clock on the disk border. There was no other evidence of retinopathy. Gonioscopy revealed intermittent trabecular synechiae with 4 + neovascularization of the angle. By applanation tonotnetry the intraocular pres­ sure measured 12 mm Hg in the right eye and 18 mm Hg in the left eye. The right coefficient of out­ flow was not determined because of poor fixation. The left coefficient of outflow was 0.04 mmVmin/mm Hg. The central visual fields were normal to 2/1000 white. The left eye was treated with topical dexamethasone and remained comfortable. The central visual acuity ranged between 20/50 and 20/100 and the intraocular pressure varied between 18 and 28 mm Hg, as determined by applanation tonometry. The clinical impression at that time was that this represented an occlusion of the long posterior cili­ ary arteries and an impending central retinal artery

THE pathway of the centrifugal (efferent, antidromic) nerves has been studied histologically in the optic nerves, chiasm, and optic tracts of a patient who had both eyes enucleated 50 years before. Centrifugal fibres in the optic nerve of animals have been known to exist for a long time (Cajal, 1894; Dogiel, 1895), and this fact is supported by recent physiological observations (Granit, 1955; Dodt, 1956). There can no longer be much doubt that a great number of centrifugal nerves also exists in the human optic nerve. Numerous remaining nerve fibres in two atrophic nerve stumps eleven and sixteen years after enucleation were the first evidence (Wolter and Liss, 1956). Subsequently, the terminal branches of centrifugal nerves were demonstrated in the normal human retina and optic nerve (Liss and Wolter, 1956; Wolter, 1957a) and it was concluded that at least some of the centrifugal nerves supply blood vessels. The observation of interrupted neurites in the human nerve fibre layer with terminal swellings pointing away from the optic disc was further evidence of the presence of centrifugal fibres (Wolter, 1956). Attempts at regeneration of centrifugal fibres were observed in a child's optic nerve stump eleven days after enucleation (Wolter, 1960) and later the reactions of these fibres could be studied four days after enucleation (Pfister and Wolter, 1963) in another case. A peculiar proliferation (hyper-regeneration) of centrifugal nerves was observed around blood vessels and micro-aneurysms in advanced diabetic retinopathy (Woltei, 1961). Very recently the remaining centrifugal nerves in the human optic disc were examined histologically ten days after complete occlusion of the central retinal artery that caused necrosis of the ganglion cells and all centripetal (afferent) retinal nerves (Wolter, n.d.). This latter study allowed for the conclusions that there are several morphologically different types of centrifugal nerves and that about 10 per cent. of the nerves in the normal human optic disc are centrifugal in nature.

In a woman of sixty-seven years, who had general arteriosclerosis, had had several paralytic strokes, and died a little later from coronary thrombosis, sudden blindness occurred in association with the ophthalmoscopic picture of obstruction of the central artery of the retina, together with the striking appearance of fragmentation of the retinal blood stream, as illustrated in the color drawings by the author which form the frontispiece to this issue.

Interruption of in vivo blood circulation in the eyes of cats and rabbits by means of external pressure for 1 hr decreases the adenosine triphosphate content of the retina by approximately 40%. Interruption of the blood supply to the retina only, in cats by clamping off of the optic nerve and with it the central retinal artery for 1 hr, does not cause a significant difference between treated and control eyes in regard to adenosine triphosphate content.

This report is concerned with a type of retinopathy which affects young or middle-aged adults, chiefly men, showing no generalized hypertension. It is seen almost exclusively when it affects the macular region and is characterized by symptoms and changes in the fundus which we believe can best be explained as the result of spastic contraction of the smaller retinal arterioles or capillaries supplying this region. A search of the literature reveals little in the English language concerning the condition which we wish to consider. Temporary closure of the central retinal artery or one of its main branches, presumably due to spasm, is well known. It may affect young persons without other signs of vascular disease. A history of partial or complete unilateral blindness with complete recovery, which can hardly be explained on any other basis, is not uncommon. A few such cases have been seen while the artery was closed

The conceptions of chorea now held by neurologists are briefly reviewed. Among the more important ocular complications described in the literature are mentioned; obstruction of the central retinal artery or its branches, sometimes followed by optic atrophy, optic neuritis, disease of the oculomotor nuclei and resultant weakness or palsies. Three cases, coming under the latter heading, are reported. Read before the Ophthalmological Section of the Baltimore City Medical Society, January 22, 1925.

Summary A case is presented of the traumatic orbital apex syndrome following reduction of a fractured zygoma. The intra-orbital haemorrhage was so severe that it led to complete loss of vision. Immediate decompression of the orbit via the maxillary antrum and intra-arterial infusion of the central retinal artery with Papavarine, achieved by cannulating the supra-orbital artery, was completely successful. This complication is said to occur only after delayed reduction of zygomatic fracture. Evidence is produced to indicate it can occur at any time. The method of supra-orbital artery cannulation for the treatment of occlusion of central retinal artery has not been previously described.

A case ending in recovery is reported. Recent views of the mechanism of such obstruction are contrasted with that suggested by Graefe and perpetuated in the literature. A table is given of the reported cases of cure or partial cure with a summary of the methods of treatment employed. Condensed from Membership Thesis submitted to the American Ophthalmological Society.

A case of bilateral central retinal artery and vein occlusion is described, its onset immediately consequent upon a tooth extraction. The author offers the most likely etiologic possibilities but draws no definitive conclusions. From the Neurological and Ophthalmological Services of the Montefiore Hospital.

✓ The unusual complication of sudden ipsilateral loss of vision following alcohol block of the infraorbital nerve for tic douloureux is described and the possible physiopathology discussed. The author believes that spasm of the central retinal artery was the cause and that repeated injections may have been a predisposing factor. Review of other cases reported shows that the loss of vision has been attributed to embolization of injected agents via collateral blood vessels.