Article Text

Download PDFPDF

Long-term visual and systemic prognoses of 83 cases of biopsy-proven sarcoid uveitis
  1. Cédric Rochepeau1,2,
  2. Yvan Jamilloux2,3,
  3. Sebastien Kerever4,5,6,
  4. Camille Febvay1,2,
  5. Laurent Perard2,7,
  6. Christiane Broussolle2,3,
  7. Carole Burillon2,8,
  8. Laurent Kodjikian1,2,
  9. Pascal Seve2,3
  1. 1Department of Ophthalmology, Hôpital de la Croix-Rousse, Hospices Civils de Lyon, Lyon, France
  2. 2University Claude Bernard-Lyon 1, Villeurbanne, France
  3. 3Department of Internal Medicine, Hôpital de la Croix-Rousse, Hospices Civils de Lyon, Lyon, France
  4. 4Department of Anesthesiology and Critical Care, St Louis—Lariboisière University Hospital, AP-HP, University Denis Diderot—Paris VII, Paris, France
  5. 5ECSTRA Team, Epidemiology and Biostatistics Sorbonne Paris Cité Research Centre UMR 1153, Inserm, Paris, France
  6. 6University of Paris VII Denis Diderot, Paris, France
  7. 7Department of Internal Medicine, Hôpital Edouard Herriot, Hospices Civils de Lyon, Lyon, France
  8. 8Department of Ophthalmology, Hôpital Edouard Herriot, Hospices Civils de Lyon, Lyon, France
  1. Correspondence to Professor Pascal Seve, Département de Médecine Interne, Hôpital de la Croix-Rousse, 103 Grande Rue de la Croix-Rousse, Lyon 69004, France; pascal.seve{at}


Aims To determine the long-term visual and systemic outcomes of uveitis patients with biopsy-proven sarcoidosis.

Methods A retrospective study of biopsy-proven sarcoid uveitis, with a 3-year minimum follow-up, seen at Lyon University Hospital, between April 2004 and January 2016.

Results A total of 83 patients were included, with a median age at onset of 52 (37−62) years and an unbalanced gender ratio (women 77.1%). Thirty-one patients had original systemic sarcoidosis in addition to ocular localisation, whereas 52 initially presented with isolated sarcoid uveitis. Among the latter, 7.7% (n=4) developed an extraocular disease after a median follow-up duration of 60 (44−110) months. The systemic spread in these patients included cutaneous sarcoids (n=2), arthritis (n=1) and multiple mononeuritis (n=1). Complete visual recovery was obtained for 60.2% of all patients and 89.2% had retrieved best-corrected visual acuity (BCVA) >20/50 in both eyes. A unilateral loss of BCVA of worse than 20/200 was documented in two patients in the isolated sarcoid uveitis group. No patient suffered from bilateral severe visual impairment or blindness. Factors linked to a poor visual prognosis, defined by BCVA ≤20/50 in at least one eye, were chronic macular oedema (p=0.009) and persistent ocular inflammation (p=0.0005).

Conclusions In this large European series of biopsy-proven sarcoidosis to date, with a prolonged follow-up, sarcoid uveitis was suggestive of a favourable systemic and visual outcome. Clinically isolated uveitis that revealed sarcoidosis remained a strictly ocular condition in most cases.

  • Inflammation
  • Epidemiology

Statistics from


Sarcoidosis is a systemic inflammatory disorder of unknown aetiology that is characterised, pathologically, by a non-caseating epithelioid cell granuloma that primarily affects the lungs and lymphatics, and is clinically defined by various symptoms and outcomes.1 The wide spectrum of ocular manifestations is highly variable. Almost every part of the eye and other orbital structures can be affected. Ocular involvement may coexist with symptomatic or asymptomatic systemic features and may precede systemic involvement by several years. In retrospective series of histologically proven sarcoidosis, uveitis affected 20%–50% of patients, with 80% of cases being diagnosed within the first year after onset.2 Moreover, uveitis was the presenting symptom in 30% of cases.3 Few studies have described systemic long-term outcomes in patients with sarcoid uveitis.4 ,5 In a series of 75 patients with definite or presumed sarcoid uveitis that were followed for a median of 4 years, Edelsten et al showed that the disease had spread to other organs in 13 patients (17%), including eight patients with involvement of the central nervous system.

Most studies that have assessed the visual prognoses of patients with sarcoid uveitis date back >20 years or have only included a limited number of patients.6 ,7 In these studies, about 10% of patients with sarcoid uveitis had serious visual impairment with blindness in at least one eye. The main cause of visual loss is cystoid macular oedema.

The objectives of this study were (i) to examine the systemic prognoses of patients with isolated sarcoid uveitis at the onset of disease and (ii) to evaluate the long-term visual outcomes.

Patients and methods

Study population

The records of patients seen at two departments of Internal Medicine and two departments of Ophthalmology at Lyon University Hospitals between April 2004 and January 2016 were retrospectively reviewed.

All included patients were aged ≥18 years, had biopsy-proven sarcoid uveitis and had a ≥3-year follow-up. Patients with other granuloma-forming diseases (such as tuberculosis or fungal infections) were excluded, as were patients who had a positive treponemal antibody test.

Patients were divided into two groups defined at the onset of uveitis. The first group included those with clinical or biopsy evidence of extraocular sarcoidosis within the first year after uveitis onset. The second group included patients with isolated sarcoid uveitis. Asymptomatic changes (ie, serum testing or chest imaging) were not considered as extraocular involvements. The extraocular extent of the disease was defined as a new symptomatic sarcoidosis in an organ that was not involved at baseline and that had developed at ≥1 year after the disease was first diagnosed. This study was approved by the Institutional Review Board of the Hospices Civils de Lyon.

Clinical data

Demographic, clinical and ophthalmological data were collected and analysed by three investigators (CR, CF, PS) using a standardised form.

White patients were considered as Caucasian; all the others were considered as non-Caucasian.

Data from clinical examinations, dosages of serum ACE, lysozyme concentrations and imaging results (chest X-ray and chest CT) were gathered by internists in the corresponding internal medical departments. Serum ACE and lysozyme were considered positive if they were, respectively, ≥70 UI/L and ≥16.7 mg/L, according to the standards of our laboratory. A chest X-ray was defined as positive when it showed bilateral hilar lymphadenopathy, pulmonary granuloma or ground-glass parenchymal opacity. A chest CT was considered positive when it showed hilar or mediastinal lymph nodes with a short axis diameter of >10 mm, perilymphatic pulmonary nodules or other parenchymal lung abnormalities. All imaged findings were interpreted by radiologists and outside films were reviewed outside our institutions.

Ophthalmological data included assessment of best-corrected visual acuity (BCVA), intraocular pressure (IOP), a complete biomicroscopic examination (eyelids, conjunctiva, cornea, anterior chamber, iris, lens, vitreous and retina) and, optionally, optical coherence tomography, and fluorescein and indocyanine green angiography. These data were collected at presentation and during the follow-up by various ophthalmologists employed in the two University Hospital Departments. Uveitis was classified according to the Standardization of Uveitis Nomenclature criteria.8

Permanent changes in BCVA were defined as no improvement after >1 year associated with an absence of a reversible intraocular cause, such as a cataract or non-treated chronic macular oedema (CME). Assessment of BCVA at the last follow-up led to dividing patients into two groups: (i) those with a ‘poor visual prognosis’ corresponding to BCVA ≤20/50 and (ii) a ‘preserved visual prognosis’ equivalent to BCVA >20/50. Severe visual impairment (BCVA <20/200) and blindness (BCVA <20/400) were defined according to the 10th revision of the International Classification of Diseases.9 The analysis by patient rather than by his/her eyes did not distinguish between unilateral and bilateral visual loss. For patients with bilateral uveitis, the analysis was performed on the most visually impaired eye, and on the basis of clinical assessment and visual acuity at baseline and during the follow-up.

Statistical analyses

Data are described as frequencies and percentages for categorical variables and as medians (25th–75th percentile range) for quantitative variables. Categorical variables were compared using Fisher's exact test and quantitative variables using Wilcoxon's ranked-sum test. All tests were two sided and statistical significance was set at the p=0.05 level. All analyses were performed using R-software, V.3.1.1 (R Foundation for Statistical Computing, Vienna, Austria).


Baseline characteristics

A total of 83 patients with biopsy-proven sarcoid uveitis were studied: 77% were women (n=64). Overall median age was 52 (37−62) years. Twenty-seven per cent (n=22) of patients were non-Caucasian: 16.9% Maghrebin, 7.2% African and 2.4% Asian. The patients' general characteristics are shown in table 1. A familial history of sarcoidosis was found in 6% of cases (n=5).

Table 1

General characteristics of 83 patients with biopsy-proven sarcoid uveitis, n=83

Histological evidence was obtained by a biopsy in the following organs: 22.9% (n=19) the bronchi, 19.3% (n=16) the mediastinal lymph nodes, 21.7% (n=18) the labial minor salivary glands, 19.3% (n=16) the skin, 6% (n=5) from conjunctiva, 2.4% (n=2) from peripheral lymph nodes and 1.2% (n=1) from the kidney or stomach. Six per cent (n=5) of patients with normal bronchi and normal labial minor salivary gland biopsies underwent an endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA), which showed granuloma.

At the onset of disease, 52 patients had isolated sarcoid uveitis. In the others, extraocular involvement included symptomatic pulmonary sarcoidosis in 3.6% (n=3) patients and extrapulmonary localisations in 33.7% (n=28): that is, skin, n=21; arthritis, n=7; liver, n=2; cardiac, n=2; central nervous system, n=1; peripheral lymph nodes, n=1; spleen, n=1; kidney, n=1; muscle, n=1; and sicca syndrome, n=1.

Overall, 75% of patients (n=62) had uveitis, isolated or not, as a revealing symptom for sarcoidosis.

The median duration of the follow-up was 60 (44−110) months.

The demographic characteristics, serum markers, chest imaging and ophthalmological features did not differ between those with isolated sarcoid uveitis and those with systemic sarcoidosis.

Ophthalmological findings

Sixty-seven per cent (n=56) of patients had chronic uveitis and 76% (n=63) had bilateral involvement (table 2). Half of all cases had panuveitis (48%). Elevated IOP and peripheral multifocal choroiditis were found in, respectively, 12% (n=10) and 30% (n=24) of patients. During the follow-up, the most prevalent ophthalmological complication was CME, found in 45% (n=36) of cases. Other ophthalmological involvements consisted of seven cases of conjunctival nodes, two of scleritis and one case of episcleritis.

Table 2

Ophthalmological findings from 83 patients with biopsy-proven sarcoid uveitis, n=83

Extent of sarcoidosis

The extent of systemic symptomatic sarcoidosis among the 52 patients with isolated sarcoid uveitis at onset occurred in 8% (n=4) of patients: 4% (n=2) presented with skin sarcoids and 2% with arthritis (n=1), mononeuritis (n=1) and hypercalcaemia (n=1).

Among the 31 patients with extraocular involvement at the onset of sarcoid uveitis, 10% (n=3) had supplementary systemic changes during the follow-up. The involvement of other organs included pulmonary (n=1), skin (n=1), neurosarcoidosis (n=1), haematological (n=1) and testicular (n=1).

The median delay of extraocular extent did not significantly differ between the isolated sarcoid uveitis group and those with systemic sarcoidosis (38 vs 24 months; p=0.86)

Treatment and visual outcomes

Comparisons of treatments and visual prognoses between isolated sarcoid uveitis and systemic sarcoidosis at onset are shown in table 3.

Table 3

Comparison of treatments and visual outcomes between patients having isolated sarcoid uveitis and systemic sarcoidosis at onset

Seventy per cent of patients (n=57) required local treatment, whereas 74% (n=61) needed oral steroids. Forty-nine per cent (n=40) needed a combination of both local and oral steroids. Local injections (subconjunctival, subtenon or intravitreal) and methotrexate were used in 21% and 22% of patients, respectively. Other immunosuppressive drugs were given to 18% of patients (n=15) when there was an unfavourable outcome and/or a disabling side effect from steroids: hydroxychloroquine was given to 14% (n=12), azathioprine to 2% (n=2) and experimental LX-211 to 1% (n=1) of cases. Hydroxychloroquine was indicated for chronic skin involvement (n=8) and arthritis (n=4).

At the last follow-up, patients with symptomatic extraocular features at onset were significantly more likely to be treated with systemic therapy than those in the isolated ocular group (69% vs 31%, p=0.001). The median duration of systemic therapy was also significantly greater in patients with extraocular manifestations at onset (30 vs 14 months, p=0.01). Overall, oral-steroid dosage of >10 mg/day was needed for 16% of patients (n=13) to control the disease, with no significant difference between the two groups.

At the last follow-up, a poor visual prognosis was documented in nine patients, among which eight had isolated sarcoid uveitis and one had systemic sarcoidosis at onset. A preserved visual prognosis was found in 89% of patients (n=74), with no significant difference between those in the two groups. Sixty per cent of patients (n=50) obtained complete visual recovery (same BCVA at baseline as at the last follow-up). Unilateral severe visual impairment occurred in 2% (n=2) of patients, belonging to those with isolated sarcoid uveitis. No patient suffered from bilateral severe visual impairment or blindness.

Comparisons of characteristics, ophthalmological findings and treatments between patients with a poor visual prognosis and those with a preserved visual prognosis are summarised in table 4. There was no significant difference in visual prognosis when baseline variables were compared. The main factor significantly associated with decreased vision was CME: it was found in 89% of patients with BCVA ≤20/50 vs 39% of patients with BCVA >20/50 (p=0.009). Persistent ocular inflammation was also significantly more frequent in those with a poor visual prognosis than in those with a preserved visual prognosis (56% vs 5%, p=0.0005). Similarly, the use of periocular or intraocular injections (subconjunctival, subtenon, intravitreal) was more frequent in those with a poor visual prognosis than in those with a preserved visual prognosis (67% vs 15%, p=0.002). Those with a poor visual prognosis were also more frequently treated with systemic therapy at the last follow-up compared with those with a preserved visual prognosis (100% vs 38%, p=0.0004), particularly regarding oral steroids (100% vs 31%, p<0.0001).

Table 4

Comparison of characteristics and clinical findings between patients having a poor visual prognosis (BCVA≤20/50) and patients with a preserved visual prognosis (BCVA >20/50)


This study reports the systemic outcomes and visual prognoses from one of the largest series of patients with biopsy-proven sarcoidosis.4 ,5 ,10 Moreover, our cohort had a mean follow-up duration of >7 years, allowing better characterisation of the prognoses of patients that presented with sarcoid uveitis. Our results show that, when it was the presenting symptom of sarcoidosis, clinically isolated uveitis remained a strictly ocular condition over time. In addition, sarcoid uveitis was suggestive of a favourable visual and systemic prognosis.

To increase its relevance and favour comparisons with previous studies,5–7 ,11 histological proof and exclusion of other granulomatous diseases were required for inclusion. This step matches the first level of diagnosis outlined in the International Workshop on Sarcoidosis.12 Similar to previous series, we did not find any association between specific extraocular manifestations and uveitis.2 ,5 ,13 Compared with patients with a negative chest X-ray, CT or 18-fluorodeoxyglucose positron-emission tomography scan, patients with asymptomatic radiological features did not have different systemic outcomes.

Previous studies from the USA have reported a higher frequency of ocular sarcoidosis in African-Americans.14–16 We did not observe such a result. Nevertheless, in our study, Caucasians accounted for nearly 75% of the population, probably because of local ethnic demographic distribution.

Ophthalmological characteristics at onset were comparable to previously reported data, with a majority of cases of chronic, bilateral and granulomatous uveitis.2 However, we observed a predominance of panuveitis, whereas it was only reported in 9%−30% of cases in previous studies.17 Although a familial history of sarcoidosis varied throughout the reports, our results were similar to those reported in a cohort from the UK, where sarcoidosis was familial in 5.9% of cases.18

In contrast to Edelsten et al,5 who reported systemic dissemination in ∼17% of cases, we only observed disease spread in 8%. Moreover, in our study, dissemination involved various organs whereas Edelsten et al found neurological involvement was predominant (11.7%). However, regarding the systemic clinical course, patients with extraocular disease fared no differently to those with isolated ocular sarcoidosis, which is a similar finding to that of Edelsten et al.5

The visual prognoses were similar in our study compared with those in the literature. Sarcoid uveitis is associated with a favourable visual prognosis, with most patients experiencing mild or no visual impairment.10 ,11 Cystoid macular oedema has been reported as the main predictor for an unfavourable visual outcome.6 ,7 Persistent ocular inflammation has been also correlated with poor visual recovery. This was recently reported in the Multicenter Uveitis Steroid Treatment trial. In this study, the authors speculated that, in intermediate, posterior and panuveitic eyes, successful control of inflammation tended to be associated with visual improvement; in contrast, the persistence or incidence of vitreous haze was associated with a worsening condition.19 In contrast to Lobo et al,11 we did not identify additional factors associated with outcome, such as older age, African ethnicity or multifocal choroiditis.

Injections (subconjunctival, subtenon or intravitreal) were more frequently given to patients with a poor visual outcome. Because published data are lacking, we were not able to make comparisons concerning this parameter. Intravitreal injection of corticosteroids is a recent alternative to general therapy in sarcoid uveitis, and the indications will probably expand in future years.20 ,21 Our use of systemic steroids, in addition to local therapy, was similar to that used in other series.3 It was however noticeable in our study that immunosuppressive agents were given in slightly larger proportions than in most other reports.22

Our study has some limitations. First, it was carried out in departments within the same institution, which is a tertiary reference centre. Thus, a chronic course, posterior uveitis, panuveitis and systemic clinical involvement were probably overestimated as our patients mostly had (clinically) severe uveitis with posterior segment involvement and/or multisystemic involvement. Moreover, the visual prognoses may also have underestimated a significant proportion of cases of mild, (acute) anterior or clinically isolated uveitis that had a fast resolution and did not benefit from histological investigation and/or expertise in a medical tertiary centre.23 Second, as with all retrospective studies, complete data sets could not be obtained from all patients, particularly full descriptions of the initial ocular findings.

In conclusion, this study reports the prognoses from the largest European series of biopsy-proven sarcoid uveitis. Isolated sarcoid uveitis frequently remains a strictly ocular condition. Chronic CME and persistent inflammation are strong prognostic factors for a poor visual prognosis. Nevertheless, sarcoid uveitis is associated, overall, with a very favourable prognosis. Further studies are needed to better characterise the clinical course and treatment of sarcoid uveitis. As suggested by recent studies, biomolecular progress could help us understand the clinical heterogeneity and prognostic factors involved in visual outcomes, as well as guiding therapeutics.24 ,25



  • Contributors CR, CF, LK, LP, YJ, ChB, CaB and PS collected the data. CR, LK, SK, LP, YJ, ChB, CaB and PS performed the statistical analyses and interpreted the data. CR, LK, SK, LP, JI, YJ, ChB, CaB and PS drafted and revised the manuscript. All authors have read and approved the final version.

  • Competing interests None.

  • Ethics approval The institutional review board of Hospices Civils de Lyon gave its approval to review retrospectively the records of patients seen at two departments of Internal Medicine and two departments of ophthalmology at Lyon University Hospitals between April 2004 and January 2016.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • At a glance
    Keith Barton James Chodosh Jost Jonas