Chapter 179 – Sympathetic Uveitis
GEORGE E. MARAK JR.
• A bilateral uveitis that develops after penetrating injury to one eye; believed to be an autoimmune disease.
• Bilateral anterior or posterior uveitis of variable severity with a uniform choroiditis.
• Histopathologically, predominantly consists of mononuclear and epithelioid cells.
• Dalen–Fuchs spots.
• Dysacousis and tinnitus.
• Alopecia, poliosis, vitiligo.
Sympathetic uveitis (sympathetic ophthalmia) was a fairly common and dreaded disease during the nineteenth century. Many cases of bilateral blindness associated with injury and inflammation were diagnosed as sympathetic uveitis. However, during the 20th century, with the development of accurate clinical and pathological definitions, it became evident that this is a relatively rare disease. MacKenzie gave a detailed clinical description of sympathetic uveitis in 1865. Fuchs established the pathological definition in 1905.
EPIDEMIOLOGY AND PATHOGENESIS
No reliable figures exist for the incidence of sympathetic uveitis because of the unreliability of diagnosis, inherent sampling errors, and medicolegal concerns. Sympathetic uveitis occurs in fewer than 1 in 100 nonsurgical penetrating wounds and in fewer than 1 in 10,000 surgical penetrating wounds. The incidence is higher in men than women, but the difference results from the higher incidence of accidental penetrating wounds in men. The frequency of occurrence is higher in children under 10 years of age because of the higher frequency of injuries; it also increases in those over the age of 60 years because of the increase in eye surgery. There may be a genetic propensity to develop sympathetic uveitis after a penetrating wound, because there is an increased incidence of the same human lymphocyte antigen types in both oriental and occidental populations. No other reliable demographic data exist.
Sympathetic uveitis is believed to represent an autoimmune response to a component or components of the retina, retinal pigment epithelium, or choroid (e.g., retinal “S” antigen tyrosinase-related protein). 
Experimental models similar to those for sympathetic uveitis have been described but, like experimental models of multiple sclerosis, these do not reproduce exactly the clinical disease. Lymphocytes with both helper (CD4+ and CD4- ) and suppressor (CD8+ and CD8- ) cell surface markers have been found in sympathetic uveitis. Cell-mediated immune responses to retinal antigens also have been found in patients who have sympathetic uveitis. The specific antigen or antigens involved in sympathetic uveitis have not been identified. A penetrating wound appears to be essential for the development of sympathetic uveitis. Additional insults, such as vitrectomy, and laser or radiation therapy, may increase the incidence. Several studies show a disproportionate incidence of sympathetic ophthalmia after primary or secondary vitrectomy. Although there are periodic reports of sympathetic uveitis developing in the absence of a penetrating histopathological wound, all affected eyes that have been subjected to a careful and complete serial sectioning demonstrate scleral penetrations.
Presumably, a penetrating wound with uveal prolapse permits tolerated ocular antigens to reach dendritic cells, or the so-called professional antigen-presenting cell, outside of the eye when there is an inflammatory stimulus. Because the antigen-presenting cells of the eye appear to be functionally suppressed in situ, these antigens normally would produce an inactivation signal.
The anterior chamber–associated immune deviation (ACAID) phenomenon has been offered as an explanation for the role of the penetrating wound in sympathetic uveitis. This is a curious laboratory phenomenon, wherein systemic tolerance to an antigen can be produced by injecting large amounts of the antigen into the anterior chamber of the eye. A large antigenic load in the absence of antigen-presenting cells is well known to be tolerogenic. When dendritic cells are added to the anterior chamber inoculum, this becomes an immunogenic stimulus. Because the large antigenic loads required to induce the ACAID phenomenon cannot occur under any conceivable clinical conditions, this curious laboratory phenomenon has limited value in explaining the role of the penetrating wound in the development of sympathetic uveitis. Recent transgenic studies indicate that endogenous ocular immunoregulation differs considerably from this artificial laboratory phenomenon.
Sympathetic uveitis and phacoanaphylactic endophthalmitis frequently were observed to occur together. It was suggested that common pathogenic features may exist. More recent studies, however, show that the concurrent development of these two inflammations is now relatively infrequent. This change is presumably because of improvements in the management of penetrating wounds. Sympathetic uveitis is believed to be an autoimmune response to a retinal, retinal pigment epithelial, or choroidal antigen that involves a cell-mediated immune response. Phacoanaphylactic endophthalmitis is clearly an immune complex reaction to a lens protein. The basis for the association of these two diseases is akin to the correlation between the price of rum and preachers’ salaries—it is a fortuitous consequence that the development of both diseases is related to the penetrating wound.
Sympathetic uveitis may appear with varying degrees of severity, from an anterior uveitis or peripapillary choroiditis to a severe
Figure 179-1 Dalen–Fuchs spot in a patient who has sympathetic ophthalmia.
Figure 179-2 Retinal and macular scarring in sympathetic ophthalmia.
panuveitis with mutton-fat keratic precipitates. Focal, elevated choroid infiltrates are common in the midperiphery. These Dalen–Fuchs spots ( Fig. 179-1 ) represent the Dalen–Fuchs nodules seen on histopathological analysis. Inflammation of the optic nerve, or papillitis, is a common feature of sympathetic uveitis and provides a useful means by which to follow the progress of the disease. Exudative retinal detachment may occur in the more severe cases. Chorioretinal scarring frequently is observed clinically when the inflammation abates ( Fig. 179-2 ). Severe scarring throughout the retina and in the posterior pole and macula may be observed in 25–30% of the cases. The peripheral retina may take on a moth-eaten appearance. The macular scars may be responsible for considerable visual loss. A “sunset glow” fundus appearance, similar to that seen in Harada’s syndrome, may be observed.
Sympathetic uveitis is not easily diagnosed. Only 20% of clinically suspected cases are confirmed histopathologically. In one series, one third of the cases were not diagnosed within 1 year of the time of onset of the symptoms. The time of onset is not reliably known, but clinical diagnosis is made within 3 months of the injury to the exciting eye in most cases. The earliest cases occur within 9–10 days of injury. Some cases are diagnosed many
Figure 179-3 Fluorescein angiography in sympathetic ophthalmia. Note the multiple hyperfluorescent dots.
years after injury. Unilateral sympathetic uveitis, based on histopathology of the enucleated eye and absence of clinical inflammation in the sympathizing eye, has been reported. A likely explanation for late-onset or unilateral sympathetic uveitis is that of a mild, overlooked, prior inflammatory episode. One patient followed by the author had a single, brief episode of anterior uveitis in the sympathizing eye. This lasted a few days and cleared spontaneously. The injured eye was enucleated, 1 month after injury, for severe pain and loss of vision. The unexpected histopathological diagnosis was typical sympathetic uveitis. On rare occasions, spontaneous clearance of inflammation in the sympathizing eye may occur.
The history of a penetrating wound and bilateral anterior and posterior uveitis serves as the basis for diagnosis. Along with typical ocular and associated features, fluorescein angiography and ultrasonography may assist in diagnosis. Serological tests and HLA typing are not helpful. The characteristic fluorescein findings consist of multiple, persistent foci of leakage from which the dye may spread ( Fig. 179-3 ). Coalescent pools of dye are seen in areas of exudative retinal detachment. Occasionally, hypofluorescent foci (which stain later) are seen early in the angiogram. Ultrasonography may demonstrate choroidal thickening when the fundus is obscured by severe inflammation, which may be particularly helpful in the differentiation of bilateral phacoanaphylactic endophthalmitis and sympathetic uveitis.
Differential diagnosis includes Harada’s syndrome, bilateral phacoanaphylactic endophthalmitis, multifocal choroiditis, and other forms of panuveitis or posterior uveitis. A history of penetrating injury helps to differentiate sympathetic uveitis from Harada’s syndrome. Bilateral phacoanaphylaxis manifests primarily as severe anterior uveitis. Ultrasonography can differentiate this entity from sympathetic uveitis, because the latter shows marked thickening of the choroid in the posterior segment.
Sympathetic uveitis may be accompanied by headache, pleocytosis of the spinal fluid, dysacousis, tinnitus, alopecia, poliosis, and vitiligo. The systemic associations are similar to those seen in Harada’s syndrome. The principal differentiating feature is the history of a penetrating wound.
Some controversy exists concerning the incidence of systemic associative features and histopathological differences between sympathetic uveitis and Harada’s disease. No evidence is known
Figure 179-4 Diffuse choroidal inflammation in sympathetic ophthalmia. In this case the retina is detached and is not shown in the illustration.
for a difference in the ocular manifestations, systemic signs, or histopathological characteristics between sympathetic uveitis and Harada’s syndrome. 
The typical histological picture of sympathetic uveitis is a uniform infiltration of the choroid with mononuclear and epithelioid cells ( Fig. 179-4 ). The inflammation has been characterized as non-necrotizing, sparing the choriocapillaris, and not extending into the retina. Cellular collections termed Dalen–Fuchs nodules project from the retinal pigment epithelium and are the pathological features that correspond to the spots observed with the ophthalmoscope. The collections of cells contain modified retinal pigment epithelial cells, histiocytes, and lymphocytes. Epithelioid and giant cells that contain pigment are seen in the choroid. Some evidence exists that necrosis of melanocytes occurs and that phagocytic cells are present, but the latter appear very different to other melanophages. Considerable variations from the typical picture are seen,  such as frequent retinal detachments (50%) and retinal perivasculitis (55%). Obliteration of the choriocapillaris (40%), retinal extension of the inflammation (18%), plasma cell infiltration (60%), eosinophilia (34%), and optic nerve involvement (25%) also are significant variations from the classic histopathology of sympathetic uveitis. With increasing severity of inflammation, there are increased numbers of inflammatory cells in the normal drainage channels, such as the scleral canals and meningeal sheaths of the optic nerves. Uveal pigmentation is related to the severity of the choroidal inflammation.  Because the pathologist studies only a few sections of the eye, the incidence of chorioretinal scarring probably is underestimated, because this is seen clinically in 25% of the cases.
Sympathetic uveitis may be prevented by enucleating the injured eye within 2 weeks of injury. Enucleation after that time is not preventive. However, sympathetic uveitis may develop after the injured eye has been removed, so every attempt should be made to save an injured eye if a reasonable expectation exists for useful vision. Careful clinical and ultrasonographic monitoring and possibly an exploratory vitrectomy may save some potentially useful eyes. Those eyes that, in the surgeon’s opinion, have little hope of useful vision should be enucleated with the patient’s consent. Because evisceration does not protect against sympathetic uveitis, it should not be the procedure of choice. Prophylactic corticosteroids do not prevent the development of sympathetic uveitis. Large, immunosuppressive doses of corticosteroids introduce an unacceptable risk of infection in a recently traumatized eye.
It is not justifiable to remove a potentially functional injured eye in established cases of sympathetic uveitis, for the injured eye may ultimately have the better vision.
Two reports suggest that early enucleation of the exciting eye improves the vision in the sympathizing eye. Both of these reports have statistical problems which vitiate their conclusions. In one report, the patients who had early enucleation also had better vision in the sympathizing eye prior to enucleation. They were compared with patients who had worse vision in the sympathizing eye prior to later enucleation. It, therefore, is not surprising that those who began with better vision before enucleation had better vision after enucleation. The second report is more difficult to evaluate, because the number of cases changes from table to table, new supporting data were found after publication, and regressions were calculated from nonparametric enumeration data. Also, the authors of this report added 9 to 62 and gave a total of 73, so it is difficult to justify sacrifice of a patient’s eye on the basis of their statistics.  If sympathetic uveitis is an autoimmune disease, it is as reasonable to remove an eye as it is to remove half of the brain in a patient with multiple sclerosis to improve the outcome.
The objective of anti-inflammatory treatment is to suppress completely the inflammation as soon as possible and to continue treatment for an extended period. Topical corticosteroids may control very mild cases, but generally large doses (1.0–1.5?mg/kg) of prednisone may be required to suppress the inflammation initially. After control has been achieved, the corticosteroid dosage may be tapered and alternate-day therapy started. Treatment should be continued for several months. Treatment for 3–6 months after the inflammation has cleared on alternate-day therapy of 10–20?mg of prednisone is desirable.
For those cases in which corticosteroids either do not control the inflammation or produce unacceptable side effects, reduced dosages may be administered with other immunosuppressive agents, such as methotrexate, chlorambucil, azathioprine, or cyclosporine.
COURSE AND OUTCOME
Early, aggressive treatment improves the visual outcome, but one third of the patients in a recent series had visual acuity of worse than 20/200 (6/60). Relapses occur in more than one half the patients and may be delayed for several years. Long-term follow-up is necessary for all patients, including those who have been free of inflammation for several years.
Common complications include band keratopathy, cataract, glaucoma, macular edema, and scarring, retinal detachments (both exudative and rhegmatogenous), hypotony, and phthisis bulbi.
Cataract surgery involves no unusual risks when performed during remission. No reliable data exist for the complications of glaucoma and retinal surgery in patients with sympathetic uveitis.
Although rare, sympathetic uveitis is a serious disease that results in blindness in a significant number of patients. The importance of early, aggressive treatment and regular follow-up cannot be overemphasized.
1. Marak GE. Recent advances in sympathetic ophthalmia. Surv Ophthalmol. 1979; 24:141–56.
2. Yamaki K, Gocho K, Hayakawa K, et al. Tyrosine family proteins are antigens specific to Vogt–Koyanagi–Harada disease. J Immunol. 2000;165:7323–9.
3. Kilmartin DJ, Dick D, Forrester JV. Br J Ophthalmol. 2000;84:259–63.
4. Wacker WB, Rao NA, Marak GE. Experimental sympathetic ophthalmia. In: Silverstein A, O’Connor G, eds. Immunology and immunopathology of the eye. New York: Masson & Cie; 1979;135–7.
5. Chan CC, Benezra D, Rodrigues MM, et al. Immunochemistry and electron microscopy of choroidal infiltrates and Dalen–Fuchs nodules in sympathetic ophthalmia. Ophthalmology. 1985;92:580–90.
6. Marak GE, Aye MS, Alepa EP. Cellular hypersensitivity in penetrating eye injuries. Invest Ophthalmol. 1973;12:380–2.
7. Lewis ML, Gass JDM, Spencer WH. Sympathetic uveitis after trauma and vitrectomy. Arch Ophthalmol. 1978;96:263–7.
8. Stafford WR. Sympathetic ophthalmia, report of a case with onset 18 years after contusion with unsuspected scleral rupture. Surv Ophthalmol. 1965;10:232–7.
9. Matzinger P. Tolerance, danger and the extended family. Ann Rev Immunol. 1994; 12:991–1045.
10. Marak GE. Phacoanaphylactic endophthalmitis. Surv Ophthalmol. 1992;4:129–35.
11. Greyerson DC, Dorr C. Spontaneous induction of immunoregulation by an endogenous retinal antigen. Invest Ophthalmol Vision Sci. 2002;43:2984–91.
12. Chan CC, Roberge EG, Whitcop SM, Nussenblatt RB. 32 cases of sympathetic ophthalmia. Arch Ophthalmol. 1995;113:597–600.
13. Kayazama F. A case of sympathetic uveitis. Ann Ophthalmol. 1980;12:1106–8.
14. Kumagi N, Shinda Y, Yamamoto T, et al. Clinical studies on sympathetic ophthalmia. In: Dernouchamps JP, Verougstraetec C, Caspers-Velu C, Tassignon MJ, eds. Recent advances in uveitis. New York: Kugler; 1992:199–200.
15. Rao NA, Marak GE. Sympathetic ophthalmia simulating Vogt–Koyanagi–Harada’s disease: clinicopathologic study of four cases. Jpn J Ophthalmol. 1983;27:506–11.
16. Inomata H. Necrotic changes of choroidal melanocytes in sympathetic ophthalmia. Arch Ophthalmol. 1988;106:239–42.
17. Croxato JO, Rao NA, McLean IW, Marak GE. Atypical histopathologic features in sympathetic ophthalmia. Int Ophthalmol. 1981;4:129–35.
18. Lubin JR, Albert DM, Weinstein M. Sixty-five years of sympathetic ophthalmia. Ophthalmology. 1980;87:109–21.
19. Marak GE, Font RL, Zimmerman LE. Histopathologic variations related to race in sympathetic ophthalmia. Am J Ophthalmol. 1974;78:935–8.
20. Marak GE, Ikui H. Pigment associated histopathological variations in sympathetic ophthalmia. Br J Ophthalmol. 1980;64:220–2.
21. Winter EC. Sympathetic ophthalmia: a clinical and pathological study of the visual result. Am J Ophthalmol. 1955;39:340–7.
22. Reynard M, Riffenburg RS, Maes EF. Effect of corticosteroid treatment and enucleation on the visual prognosis of sympathetic ophthalmia. Am J Ophthalmol. 1983;96:290–4.
23. Lubin JR, Albert DM, Weinstein M. Letter to the editor. Ophthalmology. 1982;89: 1291–2.
24. Marak GE. Sympathetic ophthalmia. In: Fraunfelder FT, Roy FH, eds. Current ocular therapy. Philadelphia: WB Saunders; 1995:454–5.
25. Jennings T, Tessler H. Twenty cases of sympathetic ophthalmia. Br J Ophthalmol. 1989;73:140–3.
26. Makey TA, Azar A. Sympathetic ophthalmia. Arch Ophthalmol. 1978;96:257–62.