Purpose
To examine systemic associations of sarcoid uveitis and association with uveitis clinical phenotype and ethnicity.
Design
Retrospective cross-sectional study.
Subjects
A total of 362 subjects with definite or presumed sarcoid uveitis from Moorfields Eye Hospital, Royal Victorian Eye and Ear, and Auckland District Health Board.
Methods
Data were collected from the review of clinical notes, imaging, and investigations. Sarcoidosis was diagnosed in accordance with the International Workshop on Ocular Sarcoidosis guidelines.
Main Outcome Measure
Diagnosis of associated systemic disease secondary to sarcoidosis.
Results
A total of 362 subjects with sarcoid uveitis were identified. Median age was 46 years, and 226 (62.4%) were female. Granulomatous anterior uveitis (47.8%), intermediate uveitis with snowballs (46.4%), and multifocal choroiditis (43.1%) were the most frequent clinical presentations, and disease was bilateral in 313 (86.5%). Periphlebitis was observed in 21.0%, and solitary optic nerve or choroidal granuloma in 11.3%. Lung parenchymal disease was diagnosed in 200 subjects (55.2%), cutaneous sarcoid in 98 (27.1%), sarcoid arthritis in 57 (15.7%), liver involvement in 21 (5.8%), neurosarcoid in 49 (13.5%), and cardiac sarcoid in 16 subjects (4.4%). Subjects with cardiac sarcoid were less likely to have granulomatous anterior uveitis ( P = .017). Caucasian subjects were older at presentation (48 vs 41 years; P = .009), had less granulomatous anterior uveitis (26.4% vs 51.7%; P < .001), and were less likely to present with cutaneous involvement (23.1% vs 35.4%; P = .040).
Conclusions
Ophthalmologists need to be aware of the systemic associations of sarcoid uveitis, in particular potentially life-threatening complications such as cardiac sarcoidosis. Differences observed in uveitis phenotype and between ethnicities require further investigation.
S arcoidosis is an idiopathic multisystem disease that may occur at any age (peak incidence 20-39 years) and in all ethnic groups. Lungs are the most commonly affected organ, but extrapulmonary disease is common, and may include cutaneous, cardiac, liver, and joint disease. , Ocular involvement occurs in 20%-40% of subjects, with dry eye and uveitis the most frequent presentation.
Substantial heterogeneity exists in the clinical presentation and systemic associations of sarcoidosis. Elderly onset sarcoidosis is more common in women and has a greater frequency of extrapulmonary manifestations. Rates of sarcoidosis are higher in northern European and African American subjects, and extrapulmonary disease has been reported to be more frequent, and more severe, in African Americans, and in New Zealand in Maori and Pacific Islanders. ,
The aim of this study was to examine systemic associations of sarcoid uveitis and to explore correlation with clinical uveitis phenotype and ethnicity.
METHODS
The medical and ophthalmology records of subjects with sarcoid uveitis attending Moorfields Eye Hospital (Lightman clinic) between 2011 and 2013, Royal Victorian Eye and Ear Hospital (RVEEH) from 1990 to 2014, and Auckland District Health Board Department of Ophthalmology from 2007 to 2019 were reviewed. Institutional review board approval was gained at each center (Auckland ethics approval NTX/12/EXP/085, Moorfields ethics approval ROAD16039, RVEEH approval 15/1215HS). Time to systemic diagnosis has previously been published for RVEEH subjects.
Diagnosis of ocular sarcoidosis was defined according to the criteria established by the First International Workshop on Ocular Sarcoidosis (IWOS), and ocular involvement was characterized according to the Standardization of Uveitis Nomenclature Working Group. Subjects were included if they had definite or presumed sarcoid uveitis diagnosed by the IWOS criteria. Those subjects with probable or possible sarcoid uveitis were excluded, as were subjects who did not fulfill the IWOS criteria for sarcoid uveitis.
All subjects were referred to a respiratory physician with systemic examination including detailed history, clinical examination, high-resolution computed tomography (CT) chest, and electrocardiogram (ECG) ± cardiac echocardiography (ECHO). Further investigations were performed as dictated by history, clinical examination, and investigations (tailored approach). Pulmonary involvement was defined as evidence of parenchymal disease on chest radiograph or high-resolution computerized tomography chest. Bi-hilar lymphadenopathy alone was not considered to be pulmonary sarcoid. Skin involvement was defined as biopsy-proven cutaneous sarcoidosis (excluding erythema nodosum). Neurologic involvement and cardiac involvement were defined as diagnosis by a neurologist or cardiologist, respectively. Hepatic involvement was defined as biopsy positive hepatic sarcoidosis. Elevated liver enzymes alone were not considered sufficient to be recorded as hepatic sarcoidosis. Joint involvement was defined as diagnosis of sarcoid arthritis by a rheumatologist.
Statistical Analysis
Data were entered into an Excel spreadsheet and analyzed in Stata version 15. Categorical data are recorded as number (%) and continuous data as median (interquartile range [IQR]). Categorical variables were compared with a χ 2 test and continuous with a Mann-Whitney U test. All tests were 2-tailed, and a P value of ≤.05 was considered significant.
RESULTS
A total of 362 subjects were included in the study. Of them, 226 were female (62.4%), and the median age at presentation with uveitis was 46 years (IQR, 35-57). Age distribution was unimodal for both male and female subjects. Only 110 subjects (30.3%) had a pre-existing diagnosis of systemic sarcoidosis when they presented with uveitis; the remainder had either simultaneous diagnosis of sarcoidosis or were diagnosed with systemic sarcoidosis subsequent to uveitis.
Clinical presentation with sarcoidosis is given in Table 1 according to the IWOS clinical diagnostic criteria. The most frequently positive signs seen were granulomatous anterior uveitis in 173 subjects (47.8%), intermediate uveitis with snowballs in 168 subjects (46.4%), and multifocal choroiditis in 156 subjects (43.1%), and disease was bilateral in 313 (86.5%).
| Sign | Number | Percentage |
|---|---|---|
| Sign 1. Granulomatous anterior uveitis | 173 | 47.8 |
| Sign 3. Intermediate uveitis with snowballs | 168 | 46.4 |
| Sign 4. Multifocal choroiditis | 156 | 43.1 |
| Sign 5. Periphlebitis | 76 | 21.0 |
| Sign 6. Isolated choroidal or optic nerve granuloma | 41 | 11.3 |
| Sign 7. Bilateral disease | 313 | 86.5 |
Systemic manifestations of sarcoid uveitis were observed in 279 subjects (77.1%). No association was observed between the number of clinical signs of sarcoid uveitis positive and systemic manifestations of sarcoidosis ( P = .273) ( Table 2 ).
| Number of Positive Signs of Sarcoid Uveitis | Number of Subjects | Systemic Sarcoidosis, n (%) |
|---|---|---|
| 0 | 9 | 7 (77.8) |
| 1 | 60 | 47 (78.3) |
| 2 | 113 | 89 (78.8) |
| 3 | 102 | 71 (69.6) |
| 4 | 55 | 44 (80.0) |
| 5 | 23 | 21 (91.3) |
| 6 | 0 | 0 |
Lung parenchymal disease was diagnosed in 200 subjects (55.5%). Median age at presentation was 45 years (IQR, 35-56), and 120 (60.0%) were female. Subjects with lung disease were more likely to have a diagnosis of sarcoidosis before uveitis (40.5% vs 17.9%; P < .001). There was no difference in clinical manifestations of sarcoid uveitis in those with lung parenchymal disease [ Table 3 ].
| Number | Sign 1 (173) | Sign 3 (168) | Sign 4 (156) | Sign 5 (76) | Sign 6 (41) | Sign 7 (313) | |
|---|---|---|---|---|---|---|---|
| Lung | 200 (55.2) | 95 (54.9) | 84 (50.0) | 80 (51.3) | 36 (47.4) | 22 (53.7) | 172 (55.0) |
| Skin | 98 (27.1) | 50 (28.9) | 43 (25.6) | 42 (26.9) | 19 (25.0) | 9 (22.0) | 86 (27.5) |
| Joint | 57 (15.7) | 24 (13.9) | 33 (19.6) | 27 (17.3) | 12 (15.8) | 4 (9.8) | 48 (15.3) |
| Liver | 21 (5.8) | 11 (6.4) | 10 (6.0) | 10 (6.4) | 4 (5.3) | 0 (0.0) | 16 (5.1) |
| Neurosarcoid | 49 (13.5) | 27 (15.6) | 24 (14.3) | 21 (13.5) | 13 (17.1) | 7 (17.1) | 44 (14.1) |
| Cardiac sarcoid | 16 (4.4) | 3 (1.7) * | 4 (2.4) | 5 (3.2) | 3 (3.9) | 3 (7.3) | 15 (4.8) |
Cutaneous sarcoidosis was diagnosed in 98 subjects (27.1%). Median age at diagnosis was 47 years (IQR, 36-57), and 69 (70.4%) were female. Only 33 subjects with cutaneous involvement had a prior diagnosis of sarcoidosis; the remaining 65 initially presented to ophthalmology. Subjects with cutaneous sarcoid were more likely to have isolated anterior uveitis (34.7% vs 23.5%; P = .032), but otherwise there was no difference in uveitis presentations [ Table 3 ].
Joint disease was diagnosed in 57 subjects (15.7%), of whom 38 (66.7%) were female. Subjects with joint disease were slightly younger with a median age of 41 (IQR, 35-50), but this did not reach statistical significance ( P = .085). Pre-existing diagnosis of sarcoidosis was present in 23 (40.4%), and the remaining 34 (59.6%) presented initially to ophthalmology. There was no difference in clinical manifestations of sarcoid uveitis in those with joint disease [ Table 3 ].
Liver involvement was diagnosed in 21 subjects (5.8%). Median age at diagnosis was 39 years (IQR, 32-50), and 12 (57.1%) were female. Pre-existing diagnosis of sarcoidosis was present in 4 subjects 19.0%), and the remaining 17 (81.0%) initially presented to ophthalmology. There was no difference in clinical manifestations of sarcoid uveitis in those with liver disease [ Table 3 ].
Neurosarcoid was reported in 49 subjects (13.5%). Median age was 39 years (IQR, 32-54), and 32 subjects were female (65.3%). Although most subjects with neurosarcoid initially presented to ophthalmology (51.0%), subjects with neurosarcoid were more likely to have a pre-existing sarcoid diagnosis at presentation to ophthalmology than the rest of the cohort (49.0% vs 27.5%; P = .002). There was no difference in the frequency of other systemic manifestations of sarcoidosis in those with neurosarcoid [ Table 3 ].
Cardiac sarcoidosis was diagnosed in 16 subjects (4.4%) [ Table 4 ]. Median age at presentation was 50 years (IQR, 45.5-59), and 12 (75.0%) were female. A total of 9 subjects had pre-existing sarcoid diagnosis at presentation, and 7 (43.8%) presented initially to ophthalmology. Subjects with cardiac sarcoidosis were more likely to have liver sarcoid (18.8% vs 5.6%; P = .023), and there was a trend toward more joint disease, but this did not reach significance (31.3% vs 15.0%; P = .082). Features of ocular sarcoidosis are reported in Table 3 . Subjects with cardiac sarcoidosis were less likely to have granulomatous anterior uveitis (Sign 1) ( P = .017). Features of ocular sarcoidosis in those with other organ system involvement are reported in Table 3 .
