Sympathetic Ophthalmia

Sympathetic Ophthalmia (SO) is a rare, but serious, bilateral uveitis that occurs after either eye surgery or penetrating or perforating eye trauma.^1–11 The eye with a history of injury or surgery is referred to as the “inciting” or “exciting” eye, whereas the contralateral eye is said to be “sympathizing”. This distinction may be unclear, however, in patients who have undergone or experienced bilateral procedures or injury. Moreover, SO has been reported following intravitreal injection,¹² after infectious^13,14 or chemical keratitis,¹⁵ and in association with non-penetrating procedures, including irradiation for ocular melanoma^16–20 and laser- and cryo-cyclodestructive procedures to lower intraocular pressure.^21–26 The time from surgery or trauma to onset of SO varies from days to decades, with the vast majority of cases occurring within 12 months. The protective role of primary enucleation or evisceration is unproven and such procedures are generally discouraged unless surgical reconstruction of the eye is deemed impossible at the time of injury.^27,28 Castiblanco and Adelman⁷ reviewed the world literature on SO through 2008 – a total of 86 patients, 60 of whom had a description of clinical findings at presentation and 46 of whom had prior surgery. The most commonly reported clinical features at presentation included the presence of active inflammation (83.3%), nummular, depigmented, chorioretinitis spots known as Dalen-Fuchs nodules (33.3%), optic disc swelling (20.0%), serous retinal detachment (SRD; 11.7%), and choroiditis (8.3%). Anterior uveitis (55.0%) was slightly more common than vitritis (47.0%), with many eyes having both. Cataract surgery and vitrectomy were the most frequently performed antecedent surgical procedures at 18 (30%) each, but given that cataract surgery is performed vastly more often than vitrectomy, posterior segment surgery would appear to confer the greatest risk, with an SO event rate estimated by Kilmartin et al.¹⁰ to be 1 in 800 vitrectomies. While the total sample size was small, results from a retrospective study by Rishi et al.²⁹ suggested that SO following both trauma and therapeutic vitrectomy (n = 7) tended to occur sooner (1.5 vs 8.0 months) and was more likely to be accompanied by SRD at presentation (100% vs 30%) than eyes that developed SO with no history of trauma (n = 10). Although the pathogenesis of SO is incompletely understood, most evidence supports some variant of Elschnig’s original, 1910 suggestion³⁰ that the immune system is somehow sensitized to self when otherwise sequestered ocular antigens are presented to the extraocular immune system,¹ a notion supported by an association between SO and both specific class II human leukocyte antigens³¹ and interleukin-10 gene polymorphisms.³² Histological examination of eyes with SO is said to show granulomatous inflammation with relative sparing of the choriocapillaris, although recent studies suggest that both a mixed inflammatory infiltrate and involvement of the choriocapillaris are common.³³ Treatment of SO involves initial control of the uveitis with systemic and regional corticosteroids in virtually all patients, followed by long-term use of corticosteroid-sparing immunosuppressive agents in the vast majority.^2,34 A comprehensive review³⁵ and two original articles^36,37 in this issue of Ocular Immunology & Inflammation address important aspects of the pathogenesis, diagnosis, and management of SO.

Mahajan et al.³⁵ systematically reviewed the use of multimodal imaging in SO. The authors reminded us of the importance of history supported by suggestive posterior segment findings, particularly SRD in the acute phase and Dalen-Fuchs nodules both early and late. Fluorescein angiography (FA) is most relevant when such changes are present, particularly early in the course of disease when multiple and progressive pinpoint leaks through the retinal pigment epithelium (RPE) lead to late filling of the detachment spaces. Mild to moderate late leakage from the optic disc is also common. Irregular filling of the choroid may be present, and in some patients may be dramatic and resemble acute posterior multifocal placoid pigment epitheliopathy, with early hypofluorescence followed by late hyperfluorescence of involved areas. Over time, FA shows the location and extent of chorioretinal scarring and, when present, can reflect or reveal the presence of subretinal fibrosis and/or choroidal neovascularization. These acute and chronic FA findings are indistinguishable from those seen in many patients with Vogt–Koyanagi–Harada (VKH) disease and, with a history of penetrating trauma or surgery, are virtually pathognomonic of SO. Choroidal imaging is largely supportive in the acute phase, although assumes increasing importance to monitor disease activity and response to therapy once the RPE leaks and SRDs have resolved. When active, indocyanine green angiography shows multiple hypocyanescent spots believed to represent focal, choroidal, inflammatory infiltrates. Spectral domain-optical coherence tomography (SD-OCT), including enhanced depth imaging of the choroid, provides what is perhaps the most diagnostically and therapeutically relevant information after FA, including the presence of shallow SRDs not appreciated clinically or on FA, and infiltrative thickening of the choroid producing both undulations in the normally curvilinear and concave RPE line and loss of the vascular lacunae normally present in Sattler’s and Haller’s layers. Acutely, Dalen-Fuchs nodules imaged on SD-OCT appear as small, irregular, hyperreflective pigment epithelial detachments, often with disruption of the overlying RPE and extension of the hyperreflective, presumed inflammatory, material into the outer retina. More chronically they appear as focal chorioretinal scars, with circumscribed loss or disruption of the outer retina and inner choroid. B-scan ultrasonography can also be used to reveal thickening of the posterior eye wall, as well as medium-to-large SRDs, particularly when direct visualization of the posterior segment is limited, as is often the case in the inciting eye. While non-invasive, fundus autofluorescence is generally less useful in that it tends to reflect the location and extent of RPE disruption seen on FA. The authors concluded that FA and SD-OCT provide the most valuable diagnostic information early and that SD-OCT alone can often be used to monitor for disease activity.

Payal and Foster³⁶ studied responses to therapy and outcomes in 19 patients with SO seen over an 8-year period (2005–2013) at a uveitis referral center in Boston, USA. Patient age ranged from 16 to 95 years (median 58 years). All patients were followed up for at least 24 months and roughly half were treated for the entire period of the study (median follow-up, 7.1 years). A total of 10 patients (52.6%) were female. Most patients (11; 57.9%) developed SO following non-surgical trauma. Of the six patients who developed SO following surgery, four (66.6%) had prior retinal detachment repair. While the authors described their general approach to the management of severe uveitis of the sort seen in SO, and stated that inflammation was ultimately controlled in all subjects, specifics regarding initial and subsequent treatments used in the 19 patients were not provided. Tabular summaries of agents used at last visit for those unable to discontinue all medications suggested judicious use of regional and systemic corticosteroids, together with both conventional and biologic immunosuppressive agents. Of note, while long-term drug-free remission has been reported previously following treatment with chlorambucil, and the authors themselves describe a similarly treated patient who went into prolonged remission following 10 months of treatment with this agent, they claim to be the first to report sustained drug-free remission in two patients following 29 and 36 months of therapy with conventional immunosuppressive agents. Two patients in the cohort received chlorambucil for 10 and 24 months, respectively, but without experiencing remission. The potential dangers of alkylating agent immunosuppressive therapy³⁸ were highlighted by the death of a chlorambucil-treated patient following the development of acute myeloid leukemia. Best-corrected vision at last visit was 20/40 or better in 11 patients (57.9%) and 20/200 or worse in 3 (15.8%), figures generally in line with previous clinic-based cohorts.⁷ The authors concluded that SO can be controlled in virtually all patients with an aggressive, step-wise approach to corticosteroid-sparing immunosuppressive therapy, and that prolonged drug-free remission can be achieved in a minority of patients.

Goudot et al.³⁷ studied nine patients with SO seen over a 9-year period (2007–2015) at a uveitis referral center in Paris, France, for evidence at presentation of lymphocytic meningitis, which they defined as more than 10 leucocytes/mm³ in the cerebral spinal fluid (CSF). Pleocytosis was identified in 5 of 9 subjects (55.6%) – two of whom had symptoms of meningeal irritation (headaches). Other extraocular symptoms at presentation included hypoacusis in three subjects, and vitiligo, vertigo, and tinnitus in one patient each. Demographic and clinical features of the cohort were otherwise similar to those reported in earlier series.⁷ The authors cited previous reports of both CSF pleocytosis and meningeal symptoms in patients with SO and reminded us that such findings may be more common than generally appreciated. They also pointed out that the presence or absence of CSF pleocytosis and meningeal symptoms should not be used to distinguish between SO from VKH disease. Further work remains to be done to determine whether CSF lymphocyte immunotyping and/or cytokine profile analyses might be clinically useful.

Together, these studies highlight important issues related to the diagnosis and management of SO. Particularly noteworthy are the utility of multimodal imaging to diagnosis and management of the condition, and the power of prompt and aggressive corticosteroid followed by corticosteroid-sparing immunosuppressive therapy to both control inflammation and limit vision loss.

DECLARATION OF INTEREST

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.

FUNDING

Supported in part by The Pacific Vision Foundation (ETC) and The San Francisco Retina Foundation (ETC).

Additional information

Funding

Supported in part by The Pacific Vision Foundation (ETC) and The San Francisco Retina Foundation (ETC).