Purpose
To evaluate clinical features, management, and prognosis of biopsy-proven orbital and adnexal sarcoidosis.
Design
Observational, retrospective case series.
Methods
Clinical features and treatment methods were evaluated retrospectively in 30 consecutive patients managed in 1 center. Data from 11 patients without systemic disease at presentation were analyzed for their impact on occurrence of systemic sarcoidosis using univariate and multivariate regression models.
Results
Orbital and adnexal involvement included the lacrimal gland in 63% (19/30), the eyelid in 17% (5/30), the orbit in 13% (4/30), and the lacrimal sac in 7% (2/30). Thirty-seven percent (11/30) had known systemic disease. Of 63% (19/30) of cases with only orbital disease, systemic disease initially was found in 27% (8/30), subsequently in 7% (2/30), and never in 30% (9/30). On computed tomography scans, the lesions were well circumscribed in 85% (25/30) and diffuse in 15% (5/30). Management included systemic steroids in 50% (15/30) of cases with symptomatic disease, followed by excision in 27% (8/30) of cases and observation in 23% (7/30) of cases with asymptomatic disease. After a mean follow-up of 44 months, all cases showed regression or remained stable, but 7% (1/30) of cases treated with systemic steroids demonstrated recurrence. Using Kaplan-Meier estimates, systemic sarcoidosis developed in 8% at 5 years. In univariate or multivariate analysis, no clinical feature was found to be significantly predictive of systemic sarcoidosis.
Conclusions
In our series, in patients with orbital and adnexal sarcoidosis alone, systemic disease developed in 8% by 5 years. Systemic steroids were effective for patients with symptomatic disease.
Sarcoidosis is a multisystem disorder characterized histopathologically by noncaseating granulomas. It affects both sexes equally, and although it can present at any age, it usually develops before 50 years of age. In the United States, the annual incidence of sarcoidosis is estimated to be 36 in 100 000 for black persons and 11 in 100 000 for white persons, with 3.8-fold greater risk of developing sarcoidosis in black persons. Sarcoidosis most commonly involves the lungs and the mediastinal lymph nodes (86% to 92%), the eye and ocular adnexa (10% to 50%), the peripheral lymph nodes (33%), the skin (10% to 40%), the central nervous system (10%), and the heart (5%).
Sarcoidosis can affect any part of the eye, orbit, and adnexal structures. In a review of 43 patients with ophthalmic sarcoidosis, Smith and Foster reported that chronic or acute anterior uveitis was the most common manifestation (73%), followed by vitreitis (62%), fundus and choroidal lesions (34%), and ocular adnexal and orbit lesions (10%). In the literature, orbital and adnexal manifestation of sarcoidosis usually has been reported as case reports, rather than as case series. In a multicenter study, Prabhakaran and associates reviewed 26 patients with orbital and adnexal sarcoidosis, evaluating the clinical features, the management, and the response to treatment. It affected the lacrimal gland most frequently (42%), followed by the orbit (39%), the eyelid (12%), and the lacrimal sac (8%). Management included oral corticosteroids, surgical debulking, and careful follow-up. Most patients (85%) experienced complete resolution with treatment. The occurrence of, or risk factors for developing, systemic sarcoidosis in the patients with only orbital and adnexal sarcoidosis was not discussed.
In this study, we retrospectively evaluated the clinical features, imaging studies, management, and prognosis of biopsy-proven orbital and adnexal sarcoidosis in 30 patients. We statistically analyzed the occurrence of, and the risk factors for developing, systemic sarcoidosis in the patients with only orbital and adnexal sarcoidosis.
Methods
We performed a retrospective review of the medical records of all patients with histopathologically confirmed orbital and adnexal sarcoidosis managed by the senior author (M.D.C.) at the Department of Ophthalmology, Henry Ford Hospital, Detroit, Michigan, between January 2000 and March 2008. All patients were diagnosed by a biopsy of their orbital or adnexal mass.
We assessed information concerning both the patients’ ophthalmic and systemic sarcoidosis. Patient age at the time of ocular diagnosis, sex, race, and ocular symptoms were noted. The ophthalmic data collected included: affected eye(s), best-corrected Snellen visual acuity, pupils, visual fields, intraocular pressures, slit-lamp and ophthalmoscopic examination results, and eyelid and orbital findings. We evaluated any intraocular disease: uveitis and anterior and posterior segment lesions. We evaluated the orbital signs: proptosis, restriction of extraocular movements, edema or erythema of the conjunctival or eyelid or orbital, and any palpable masses in the lacrimal gland or orbit. In the orbit and adnexa, we evaluated the anatomic structure involved: eyelid, conjunctiva, lacrimal gland, and extraocular muscle(s). Restriction of extraocular motility was judged as mild (less than 20% ), moderate (20% to 50%), or severe (more than 50%). We evaluated the disease epicenter: by anteroposterior location (anterior, middle, or posterior orbit), by orbital quadrant (superior, inferior, nasal, temporal, diffuse), and by relation to the muscle cone (intraconal, extraconal, or both). We evaluated the treatment options: observation, excisional biopsy, and corticosteroid therapy. Orbital computed tomography and magnetic resonance imaging scans were analyzed. When indicated, the orbital lobe of the lacrimal gland was biopsied using an upper lid crease incision.
We obtained systemic data from the patients and from their physicians. We evaluated the organs involved and whether systemic sarcoidosis was present or absent at the time of diagnosis of the orbital and adnexal disease, and if, and when, systemic disease was diagnosed later. We collected follow-up information: visual acuity at last visit, response to therapy (regression, stable, recurrence, no response), and the patient status (with systemic sarcoidosis, without systemic sarcoidosis.
Statistical Analyses
We analyzed our data using the SPSS software version 13.0 (SPSS, Inc, Chicago, Illinois, USA). We used Kaplan-Meier survival estimates to analyze the development of systemic sarcoidosis as a function of the time from the initial presentation to the Department of Ophthalmology at the Henry Ford Hospital. Kaplan-Meier survival estimates are the most suitable test for survival analysis. If these events are dependent, Kaplan-Meier survival estimates are biased. This bias arises because the Kaplan-Meier survival estimates assumes that all events are independent.
Only clinical data were analyzed with regard to the subsequent development of systemic sarcoidosis in a patient with orbital and adnexal sarcoidosis. The effect of each individual demographic datum (age, sex, laterality at presentation), each clinical variable (presence of symptoms, visual acuity, restriction of extraocular movements, presence of periocular edema and erythema, conjunctival chemosis, dilated episcleral veins, lagophthalmos, papilledema, choroidal folds, venous dilatation, palpable mass, relative afferent pupillary defect, proptosis, involved orbital location, and magnetic resonance imaging and computed tomography features), each management option (observation, surgical excision, systemic sarcoidosis and immunosuppressive therapy), and the prognosis (development of recurrence) on the development of systemic sarcoidosis was assessed by a series of univariate logistic regression analyses. The continuous variables, such as age (< 60 years vs ≥ 60 years) and visual acuity (< 20/100 vs > 20/100), were turned into discrete variables, and all variables were analyzed as discrete categories. The variables that were significant on a univariate level ( P < .05) were entered into a multivariate analysis using a forward, stepwise approach (likelihood ratio). For variables that showed a high degree of correlation, only 1 variable from the set of associated variables was entered at a time in subsequent multivariate models. A final multivariate model fitted variables identified as significant predictors ( P < .05) in the stepwise model and variables considered clinically important for systemic sarcoidosis.
Results
Between January 2000 and March 2008, 379 sarcoidosis patients were managed in the Department of Ophthalmology, Henry Ford Hospital, Detroit, Michigan. Thirty of them (8%) had orbital and adnexal sarcoidosis. The mean age at presentation was 44 years (median, 43 years; range, 23 to 67 years). Ninety percent (27/30) were women and 10% (3/30) were men. Eighty-seven percent (26/30) were black and 13% (4/30) were white. The population of the city of Detroit is largely black.
At presentation, the mean duration of symptoms was 7 months (median, 3 months; range, 0 to 60 months). The most common ocular symptom at presentation was eyelid swelling in 63% (19/30), followed by a palpable eyelid mass in 23% (7/30), tearing in 7% (2/30), discomfort in 3% (1/30), and ptosis in 3% (1/30). The best-corrected visual acuity was between 20/20 and 20/40 in 97% (29/30) and 20/60 in 3% (1/30). Orbital and adnexal involvement was unilateral in half of the cases and bilateral in the other half. Unilateral disease became bilateral disease 3% (1/30) after 48 months of follow-up. The most common clinical feature at presentation was edema in 70% (21/30; upper eyelid in 53% [16/30], lower eyelid in 17% [5/30]), followed by erythema in 37% (11/30; upper eyelid in 20% [6/30], lower eyelid in 13% [4/30], and both eyelids in 3% [1/30]), palpable mass in 27% (8/30), periocular tenderness in 13% (4/30), inferior dystopia of the globe in 20% (6/30), superior dystopia of the globe in 3% (1/30), chemosis in 3% (1/30), restricted extraocular motility in 3% (1/30), and uveitis in 3% (1/30; Figures 1 and 2 ). None of the patients had fundus or optic nerve involvement. Anatomic location of involvement and imaging features are shown in Tables 1 and 2 , respectively.
| No. of Eyes (%) | |
|---|---|
| Anteroposterior location of orbital and adnexal sarcoidosis | |
| Anterior orbit | 29 (97%) |
| Midorbit | 1 (3%) |
| Posterior orbit | 0 (0%) |
| Radial location of orbital and adnexal sarcoidosis | |
| Superior | 20 (67%) |
| Inferior | 4 (13%) |
| Nasal | 2 (7%) |
| Diffuse | 4 (13%) |
| Conal location of orbital and adnexal sarcoidosis | |
| Extraconal | 29 (97%) |
| Intraconal | 1 (3%) |
| Muscle involvement | |
| Superior rectus | 1 (3%) |
| Involved orbital and adnexal organs | |
| Lacrimal gland | 19 (63%) |
| Eyelid | 5 (17%) |
| Orbit | 4 (13%) |
| Lacrimal sac | 2 (7%) |
| No. of Eyes (%) | |
|---|---|
| Configuration | |
| Diffuse | 5 (15%) |
| Circumscribed | 25 (85%) |
| Consistency | |
| Solid | 29 (97%) |
| Cystic | 1 (1%) |
| Shape | |
| Oblong | 17 (57%) |
| Round | 8 (27%) |
| Diffuse | 5 (15%) |
| Enhancement with contrast agent | 29 (97%) |
| Bony changes | 0 (0%) |
The management and prognosis are presented in Table 3 . Overall, the orbital and adnexal sarcoidosis showed regression in 70% (21/30) of patients and remained stable in 27% (8/30) after a mean follow-up of 44 months (median, 25 months; range, 2 to 155 months). Two intralesional triamcinolone injections (40 mg/mL) were given, 2 months apart, in 1 patient with a right orbital floor mass, and the lesion resolved. In 1 patient (3%), a recurrence developed 47 months after systemic steroid therapy, and the patient was given another course of systemic steroid therapy, with resolution of the disease.
