Sympathetic ophthalmia (SO) is a rare intraocular inflammation that presents as a bilateral diffuse granulomatous uveitis following penetrating trauma or ocular surgery involving one eye. Following trauma to one globe (the exciting eye), intraocular inflammation develops in the fellow eye (the sympathizing eye) after a variable period of time ranging from a few days to decades. Although the pathophysiology of this disease is not clearly understood, an autoimmune process against peptides of melanocytes has been proposed. In 1830 Mackenzie gave a detailed clinical description of SO, and Fuchs established the pathological features of this disease in 1905.
SO is a potentially devastating disease with many exacerbations, and long-term follow-up is essential. A high index of suspicion, early diagnosis, advances in surgical procedures, and the use of immunomodulatory agents have improved the visual outcome of SO.
Symptoms and signs
SO can present with diverse clinical presentations, and any bilateral uveitis following ocular surgery or trauma to an eye should alert the ophthalmologist to the possibility of this entity.
The onset of inflammation in the sympathizing eye has been reported to appear any time between 5 days and 66 years after the initial trauma. However, 80% of patients manifest with symptoms and signs of intraocular inflammation within 3 months, and 90% show evidence of inflammation within 1 year of the time of the initial insult.
Patients usually present with mild ocular pain, photophobia, epiphora, and blurring of vision. Their near vision may sometimes be compromised. The clinical signs may vary from a mild anterior uveitis to severe granulomatous panuveitis associated with moderate to severe vitritis.
Posterior-segment findings include papillitis, generalized retinal edema, and diffuse choroiditis. Small yellow-white lesions may be seen in the mid periphery of the retina and these are recognized as Dalen–Fuchs nodules ( Box 81.1 ). The fundus examination might also show evidence of multiple choroidal granulomas and exudative retinal detachment. Although patients with SO develop signs of panuveitis, they may initially present with clinical features of posterior uveitis in the sympathizing eye.
Exudative retinal detachment
Inadequately treated SO will run a complicated course of chronic recurrent uveitis. This, in turn, could cause secondary glaucoma, cataract, choroidal neovascularization, subretinal fibrosis, chorioretinal and optic atropy, and finally phthisis bulbi.
Duke-Elder and Perkins state that the first reference in the literature to the concept of SO is from Agathias in the anthology compiled from Constantius Cephalis in 1000 ad . Though many references to this disease entity were reported during the seventeenth and eighteenth centuries, the first comprehensive description of SO was written by William Mackenzie in 1830. He described many cases of SO and concluded with a complete discussion of this entity. Prichard, in 1851, was the first to practice enucleation of the injured eye as a therapeutic measure to save the opposite eye. But Critchett, in 1863, showed that enucleation was ineffective once the inflammation develops in the sympathizing eye. Knowledge of SO further increased after Schirmer’s (1905) critical survey and Fuchs’ (1905) classical histological studies of this disease.
There is a disparity in the reported incidence of SO, and the reliability of the reported figures is questionable because the suspected diagnosis was confirmed histopathologically in less than one-third of cases and 15% of the pathologically diagnosed cases were not identified clinically. Liddy and Stuart estimated the incidence of SO as 0.19% following penetrating injuries and 0.007% following ocular surgery. In 2000 Kilmartin et al estimated an incidence of 0.03/100 000 for the general population of the UK.
Although SO was once considered a disappearing disease, recent incidence figures, particularly for postsurgical cases, suggest an increasing trend. Advances in surgical training and the use of microsurgical instrumentation have enabled better management of traumatized eyes that would other wise have been enucleated in the past. In 1982, Gass reported an incidence of 0.06% after vitrectomy and 0.01% incidence when vitrectomy was the only operative procedure causing the penetrating wound.
Chan et al, in their retrospective study from 1982 to 1992, reported that 28% of patients with SO developed it following intraocular surgical procedures. Kilmartin et al reported in 2000 that ocular surgery, especially retinal surgery, accounted for 56% of all cases of SO. A report by Su and Chee stated that the proportion of SO caused by ocular surgery is 70%. All of these studies indicate that ocular surgery as a precipitating factor is gaining importance.
Previous studies indicate that SO is more prevalent in males and in children. But recent reports show no gender predominance and a smaller number of cases occurring in children. Elderly patients appear to be at an increased risk, probably because of the increased frequency in ocular surgery performed in this age group.
The clinical diagnosis of SO is based on history and clinical examination. There are no specific laboratory studies to establish the diagnosis. However, fluorescein angiography (FA) and indocyanine green (ICG) angiography are helpful in supporting the diagnosis.
The characteristic features of SO as seen on FA are multiple tiny foci of leakage at the level of retinal pigment epithelium (RPE) with late coalescence if there are areas of exudative detachment. Another, less common, angiographic appearance is similar to that seen in acute posterior multifocal placoid pigment epitheliopathy. Here the lesions appear as early focal obscuration of background choroidal fluorescence with late staining.
Since the disease predominantly involves the choroid, use of ICG could help support the diagnosis of SO. Two ICG patterns have been observed. In the first type, the hypofluorescent dark dot appearing during the intermediate phase persisted throughout the late phase; in the second type, the dots faded away during late phase. The first pattern was thought to represent chorioretinal atropic areas, and the second was thought to correspond to active choroidal space-occupying lesions. Hence, some believe that ICG provides additional information about choroidal involvement and subsequent evolution of the lesion. However, the enucleated SO globes reveal no histopathologic evidence of chorioretinal atropy. Moreover, the retina is typically spared from the inflammatory cell infiltration.
Ultrasound scans help to establish choroidal thickening, predominantly in the posterior choroid, and optical coherence tomography is believed to aid in monitoring the therapeutic response in patients with shallow retinal detachment.
When considering the differential diagnosis of SO, it is important to rule out intraocular infections that could cause severe endophthalmitis. Posttraumatic iridocyclitis may also cause an inflammatory reaction, but neither endophthalmitis nor iridocyclitis involves the fellow eye.
The differential diagnosis also includes any other cause of granulomatous uveitis, especially phacoanaphylaxis, Vogt–Koyanagi–Harada disease (VKH), sarcoidosis, or posterior scleritis ( Box 81.2 ).
Uveal lymphoid infiltration
Phacoanaphylaxis is a chronic granulomatous inflammation that occurs following traumatic or surgical lens capsule disruption and could closely simulate the clinical signs of SO. It typically manifests as a granulomatous anterior uveitis. The choroid, retina, and optic nerve are not usually involved in the disease process. Studies have shown that phacoanaphylaxis can coexist with SO (4–25%), and even though phacoanaphylaxis is normally a unilateral inflammation, bilateral occurrence has been reported in the absence of SO changes. In such cases, the posterior choroid is not thickened with inflammatory cellular infiltration, as seen in SO, and thus, no evidence of a thickened choroid is shown by ultrasonography. In bilateral phacoanaphylaxis, the first eye involved is usually quiet by the time the inflammation begins in the second eye. In contrast, in SO the exciting eye is usually severely inflamed at the time the sympathizing eye becomes involved.
VKH disease is a bilateral, diffuse, granulomatous uveitis with clinical, histopathological, immunohistochemical, and FA manifestations that are strikingly similar to those of SO. These associations suggest that the two conditions may involve closely related antigens and similar immune mechanisms. Shindo et al reported similar genetic backgrounds in patients with SO and VKH. Moorthy et al showed virtually identical histopathological changes in SO and VKH, including preservation of the choriocapillaris in acute VKH cases. Both entities have been associated with headache, tinnitus, alopecia, poliosis, and vitiligo. In most instances, a history of penetrating trauma would be the most helpful factor in making the diagnosis of SO.
Ocular sarcoidosis is another disease with a clinical resemblance to SO. However, systemic manifestations of sarcoidosis, if present, might help in the diagnosis. In the absence of such manifestations, the characteristic findings of SO seen in FA and ultrasound scan would help differentiate these two entities.
Both posterior scleritis and SO may present with exudative retinal detachment and disc edema. However, SO is typically a bilateral disease, whereas posterior scleritis most commonly presents as a unilateral inflammation. Although ultrasound scans may reveal diffuse choroidal thickening and exudative retinal detachment in both of these entities, choroidal thickening in posterior scleritis shows high internal reflectivity with evidence of retrobulbar edema.
The successful control of inflammation in patients with SO depends on early and aggressive treatment with large doses of corticosteroids until the inflammation is resolved. Before the use of corticosteroids, the prognosis for SO was considered very poor. Makley and Azar reported in 1978 that patients treated with corticosteroids had a favorable visual outcome, with 64% attaining a final visual acuity of 20/60 or better. Lubin et al in 1980 supported these data and demonstrated that corticosteroid therapy changed both the character and the severity of the inflammatory process. In 1983, Reynard et al stated that corticosteroid therapy prevented severe visual loss.
Some patients may be refractory to corticosteroid therapy or may experience unacceptable side-effects during treatment. Such patients have been shown to benefit from steroid-sparing immunosuppressive drug therapy, including methotrexate, cyclophosphamide, ciclosporin A, chlorambucil, and mycophenolate mofetil. Enucleation of the injured globe within 2 weeks of penetrating trauma is generally believed to prevent the development of SO. But with the advances in surgical techniques, many eyes once considered nonviable may have a fair prognosis and the injured eye might eventually have the better vision. Hence, the decision to perform primary enucleation should be made cautiously. Enucleation should be considered in those cases where the traumatized eye has no light perception, has a total afferent pupillary defect, and is severely disorganized, making it impossible to repair surgically. In such situations it is important to discuss with the patient the possibility of developing SO in the nontraumatized eye, and enucleation should be carried out within 2 weeks of trauma.
The advantage of enucleating the exciting eye once the disease has started in the sympathizing eye is a controversial issue. Review of the literature shows conflicting results with regard to the benefits of enucleation after the onset of inflammation. It is therefore important to try saving the injured eye, especially if any potential for useful vision exists in the exciting eye.
Prognosis and prevention
Before the use of corticosteroids, the visual prognosis of SO was generally considered to be poor. The aggressive use of corticosteroids in combination with immunosuppressive therapy has improved the visual prognosis. However, the relapsing nature of the disease demands careful long-term follow-up to prevent serious complications associated with recurrences.
It is interesting to analyze a few studies which stated prognostic factors based on histopathological findings. Lubin et al reported a direct correlation between the severity of inflammation in the exciting eye and the final visual acuity of the sympathizing eye. However, Winter and Reynard et al observed no direct correlation.
Makley and Azar reported on the long-term follow-up of patients with SO in 1978. In their series, they observed that relapses and complications such as secondary glaucoma, cataract, exudative retinal detachment, and choroidal scarring were common among patients with SO. Many patients treated with steroids retained a favorable visual acuity. The duration of such corticosteroid therapy was variable and ranged from a few months to 6 years.
The only suggested prevention for SO is early enucleation of the injured globe. In 2006, Su and Chee stated that industrial safety laws that mandate the use of personal protective equipment while working would reduce the incidence of work-related trauma and help reduce the incidence of SO. Advances in surgical techniques have enabled ophthalmologists to operate successfully on severely traumatized eyes and achieve good wound closure of penetrating eye injuries. This would, in turn, prevent the escape of uveal antigen to the regional lymphatic system, thereby reducing the incidence of SO. It is also important to remember the association of SO with intraocular surgical procedures, especially in patients with a previous history of penetrating trauma or repeated surgery. Such patients should be closely monitored, and appropriate levels of anti-inflammatory drug therapy should be introduced with any early evidence of inflammation.