Peripheral Retinal Neovascularization in Rheumatic Fever James S. Kelley, MD, Harry G. Randall, MD \s=b\ During a two-year period, a 15-year\x=req-\ old boy had three major manifestations of rheumatic fever, ie, chorea, rash, and carditis. Following the acute carditis, peripheral retinal neovascularization appeared in the left eye. Associated transient hemiparesis suggests that emboli may account for the retinal vascular occlusion. Rheumatic fever should be considered in the medical survey of patients who have retinal new vessels of obscure origin. (Arch Ophthalmol 97:81-83, 1979) "Dheumatic fever is a multisystem disease that follows streptococcal throat infections. The major clinical manifestations are carditis, polyarth¬ ritis, chorea, erythema marginatum, and subcutaneous nodules. The minor clinical manifestations, as listed in the Jones criteria for diagnosis of rheu¬ matic fever,1 are arthralgia, fever, and a previous history of rheumatic dis¬ ease. Laboratory tests, such as the ESR, the C-reactive protein levels, and streptococcal antibody studies provide supporting evidence in the diagnosis but are not specific for rheu¬ matic fever. The patient described herein had an episode of chorea with rash and fever; then two years later, he had an episode of severe carditis with fever. During the recovery from the carditis, peripheral retinal neovas¬ cularization was discovered in the left eye. Accepted for publication March 8, 1978. From the Department of Ophthalmology, Greater Baltimore Medical Center. Reprint requests to 14 W Mount Vernon PI, Baltimore, MD 21204 (Dr Kelley). Peripheral neovascularization or retinal sea-fans can be associated with a number of systemic diseases. These include diabetes, the sickle cell hemoglobinopathies, polycythemia, macroglobinemia, leukemia, and sarcoido¬ sis.-' Local retinal vascular occlusions, central venous occlusion, and branch retinal vein occlusion may be asso¬ ciated with neovascularization. These local occlusive diseases may, in turn, be associated with systemic arterial disease or systemic hypertension. To our knowledge, this is the first report documenting peripheral retinal neo¬ vascularization in association with rheumatic fever. REPORT OF A CASE General History A 15-year-old boy was first admitted to the Greater Baltimore Medical Center in July 1973, with a ten-day history of abnor¬ mal, uncontrollable movements of the upper extremities. The first symptom had been difficulty in writing, followed by the onset of abrupt, involuntary movements that were most prominent in the arms but that also involved the lower extremities. Facial grimaces were noted, and speech became unintelligible and occurred in short bursts. The boy's family recognized in¬ creasing irritability and a depressed mood. Generalized weakness became evident. The past medical history was one of good health, aside from the usual childhood diseases and two minor injuries to the hand. There was no recent pharyngitis. The patient had had a sore throat eight months earlier, but no culture was taken. There were occasional episodes of hay fever. The family history revealed that the father had had rheumatic fever at the age of 12 years. The disease had required treat¬ ment for a number of years but left no sequels. The patient lived in a suburban area in a single family home and had only one sibling. Downloaded From: by a Tufts Univ. Hirsh Health Sciences Library User on 07/24/2018 Clinical History Results of initial physical examination were normal, except for the choreoathetoid movements. During the ensuing six days of hospitalization, however, the following additional findings were noted: A fever spiking to 39.4 °C developed during the second week and graduali}' subsided in a ten-day period. As the fever decreased, an irregular, slightly raised, blotchy rash was seen on the patient's shoulders, trunk, and groin. This also gradually disappeared with no specific treatment in a ten-day period. The chorea was treated with diazepam, 10 mg every four hours with good response. In the last week of hospitalization, general weakness was much improved, speech was clear, and the involuntary motion was limited to the left hand. There was no arthritis and no signs or symptoms of carditis, nor were there ocular signs or symptoms. Results of laboratory and radiologie tests mainly normal, showing no indication of disease other than rheumatic chorea. Chest film, skull films, and brain scans were normal. The hemoglobin level was 16.1 g/dl; the hematocrit, 46%; WBC count, 6,700/cu mm with 65% neutrophils, 20% lymphocytes, 4% monocytes, and 2% eosino¬ phils. The ESR was repeatedly between 7 and 10 mm. Also normal were the heterowere phil agglutination titer, lupus erythemato(LE) preparations, serum copper and serum ceruloplasmin levels, serum electro¬ lyte values, the radioimmunoassay, and the antistreptolysin O titer (120 Todd units). sus There was a mild, transient elevation in the total bilirubin level and in the SGOT level. The discharge diagnosis was Sydenham's chorea. The patient was given penicillin G and phénobarbital as the only outpatient therapy. The penicillin therapy was contin¬ ued without incident for almost three years. Further Incidence of Disorder In August 1976, while working in a coffee house at a beach resort, the patient artery through vein into neovascular frond. Fig 1.—Dilated tortuous vessel feeds from neighboring retinal Fig 2.—Neovascularization fills slowly. Aneurysmal dilation is bright. Small isolated segment of artery fills by retrograde flow. Fig 3.—Bright frond of new vessels is surrounded by broad area of retinal capillary closure. Fig 4.—In more proximal field there is obstruction of flow in both branch retinal artery and branch retinal vein. Vein wall stains with dye. noted a sore throat, weakness, nausea, shortness of breath, and ankle edema. The medical history was otherwise negative. The prophylactic antibiotic had not been used daily. The major physical findings were tachycardia (130 beats per minute), hepatosplenomegaly, and pitting ankle edema. Electrocardiogram showed atrial enlargement and ventricular hypertrophy with numerous premature beats. The cardiac enlargement was further docu¬ mented on chest film and echocardiography. No definite murmur wTas heard, but auscultation was difficult because of the rapid rate. Throat cultures were positive for a penicillin-resistant Staphylococcus. The ESR remained under 10 mm. The C-reactive protein was positive undiluted. The SGOT, SGPT, and serum lactic dehy¬ drogenase levels were greater than 400 units. Screening serology for virus infec- tion was negative. The patient was hospitalized for the following six weeks. The severity of the cardiomyopathy required transfer to an intensive cardiac care unit. Steroids were administered for systemic effect but yielded little improvement. The congestive heart failure gradually responded to sup¬ portive care. During the recovery, a right hemiparesis suddenly developed. Skull films, including computerized tomography, normal. There was no laboratory evidence for connective tissue disorders (negative antinuclear antibody titers and negative LE preparations). The neurologic deficit was ascribed to an embolus in the left middle cerebral artery. Anticoagulation with warfarin sodium (Coumadin) was begun. The weakness on the right side gradually resolved. During the fourth week of hospitalization, the patient complained were Downloaded From: by a Tufts Univ. Hirsh Health Sciences Library User on 07/24/2018 of floating spots and a hazy area in the lower visual field. Ocular examination at that time was normal, except for the fundus findings. The visual acuity was 6/6 in both eyes. The superior temporal fundus on the left showed small retinal and preretinal hemor¬ rhages. No neovascularization was visible. No specific ocular treatment was advised. The cardiac condition gradually respond¬ ed to the rest and supportive therapy. No further neurologic signs developed. Following discharge from this second hospitalization, the patient slowly regained strength and returned to full activity. The ocular symptoms of spots and a veil in the lower field of vision persisted. Peripheral Neovascularization Serial ocular examinations were entirely normal except for fluctuating amounts of hemorrhage in the upper outer quadrant of the left retina. Approximately eight months after the onset of visual symptoms, a definite frond of peripheral neovascular¬ ization was seen in this area of the fundus. Fluorescein angiography confirmed this finding. A tortuous arteriovenous connec¬ tion contributed to the earliest filling of the preretinal new vessels (Fig 1). An aneurysmal dilation stained brightly with dye as the frond filled. A small segment of an obstructed branch artery filled by retro¬ grade flow (Fig 2). The retinal capillary bed around the neovascularization did not fill with dye (Fig 3). Just proximal to the new vessels, the angiogram documented obstruction of a branch retinal vein and a branch artery. The vein wall stained with the dye (Fig 4). There were a few small hemorrhages obscuring the underlying fluorescence. These retinal abnormalities are currently being followed up with serial drawings and photographs. COMMENT Ocular complications of rheumatic fever are relatively rare. Smith4 surveyed a series of 200 rheumatic fever patients and found only 12 with ocular complications, two with Roth spots, three with conjunctival splinter hemorrhages, and four with various signs of cerebral embolism involving the eye. Other than the Roth spots, usually a late finding associated with subacute bacterial endocarditis, there was only one patient who had a fundus finding, ie, a chorioretinal scar. There are cases, however, in which the retinal circulation has been affected by rheumatic disease. Emboli from rheumatic valvular disease have been well documented. The calcific emboli occur late in the course,1 not in young people suffering the acute rheumatic fever. Earl and Swash4 noted that transient loss of vision may be reported in one third of adults who have a history of rheumatic cardiac disease. These episodes are presumed to be on an embolie basis. Of more pertinence to the present report are those cases of central reti¬ nal artery occlusion associated with Sydenham's chorea. Twelve cases have been reviewed by Ling and co-work¬ ers," who added another case to the literature. In most such cases, the artery occlusion occurred during or within a few months of the onset of the chorea. These were young patients who ranged from 8 to 26 years in age. Definite cardiac disease was found in only seven of the 12 cases. The authors noted a refractile plaque in a retinal vessel in their own case and postu¬ lated that the occlusions were of an embolie nature. A similar mechanism seems likely to account for the retinal abnormality in our patient. There were clinical signs of cerebral embolism. There was no evidence of other causes of occlu¬ sion, such as vasculitis. An embolie occlusion would account for the peripheral area of nonperfusion docu¬ mented on the fluorescein angiogram. This local retinal ischemia, in turn, may account for the later growth of neovascularization. This is consistent with the current hypothesis concern¬ ing the pathogenesis of retinal neo¬ vascularization. If the medical history had been negative, the present case would have been classified as Eale's disease, idiopathic retinal vasculitis. The list of conditions associated with secondary retinal neovascularization is growing rapidly. With a careful medical sur¬ vey, the cases of primary retinal neo¬ vascularization may become the ex¬ ception, rather than the rule. Takayashu's disease, other forms of occlusive arteritis, and retinopathy of prema¬ turity must be ruled out, in addition to the conditions previously cited. Incontinentia pigmenti, a subtle skin disease that occurs in females, may be associated with peripheral retinal neovascularization.4 Wise et als have reviewed the possible association be¬ tween Eale's disease and numerous less common systemic disorders. In most cases of peripheral retinal neovascularization, a diligent survey will reveal an underlying systemic condition. Rheumatic fever is of particular interest in this regard. Autopsy studies suggest that there is a notable amount of subclinical rheu¬ matic disease.4 Thus, the clinical histo¬ ry may be insufficient to rule out this entity. If the retinal vascular changes are seen late in the course of the systemic disease, even the antistreptolysin test may be normal. A careful clinical examination of the heart and a serology for a variety of streptococcal products (streptozyme test) is the best screening method. Rheumatic disor¬ ders may account for many cases of peripheral neovascularization current¬ ly classified as primary or idiopathic. References 1. Taranta A: Rheumatic fever, in Hollander JL, McCarty DJ (eds): Arthritis and Allied Conditions. Philadelphia, Lea & Febiger, 1972, pp 736-821. 2. Asdourian GK, Goldberg MF, Busse BJ: Peripheral retinal neovascularization in sarcoidosis. Arch Ophthalmol 93:787-791, 1975. 3. Smith JL: Ocular complications of rheumatic fever and rheumatoid arthritis. Am J Ophthal- mol 43:575-582, 1957. 4. Penner R, Font RL: Retinal embolism from calcified vegetations of aortic valve: Spontaneous complication of rheumatic heart disease. Arch Ophthalmol 81:565-568, 1969. 5. Earl CJ, Swash M: Transient visual obscurations in chronic rheumatic heart disease. Lancet 2:323-326, 1970. 6. Ling W, Ofteldal G, Simon T: Central retinal Downloaded From: by a Tufts Univ. Hirsh Health Sciences Library User on 07/24/2018 artery occlusion in Sydenham's chorea. Am J Dis Child 118:525-527, 1969. 7. Watzke RC, Stevens TS, Carney RG Jr: Retinal vascular changes of incontinentia pigmenti. Arch Ophthalmol 94:743-746, 1976. 8. Wise GN, Dollery CT, Henkind P: The Retinal Circulation. New York, Harper & Row Publishers, 1971, pp 377-381.