DS is the most common chromosomal disorder causing mild to moderate intellectual disability. DS can be due to Trisomy 21, translocation of chromosome 21 or mosaicism1. Although ocular manifestations have been described in patients with DS,17–20 uveitis has not been characterized in this patient population to our knowledge.
In this case series of subjects with DS and uveitis, the mean onset of uveitis occurred in the third decade of life and 7 out of 8 subjects were diagnosed with idiopathic uveitis. The majority of subjects (6 of 8) had anterior involvement with 5 also having an intermediate component. One subject presented with posterior uveitis in the form of multifocal choroiditis (not associated with anterior chamber cell or vitreous cell). One subject had retinal vasculitis and CME with no documentation of anterior or intermediate uveitis. This subject was followed by a general ophthalmologist for a year prior to referral to a uveitis specialist and given the lack of detailed early records, we cannot exclude prior anterior or intermediate involvement. Since two of the described subjects did not have a documented leukocytic infiltrate, since at least one subject had ellipsoid zone loss detected on OCT, and since autoantibodies seem likely to play prominent role in this disease (see discussion below), the term autoimmune eye disease might be more broadly appropriate than the term, uveitis. We prefer to use the term, uveitis, if a leukocytic infiltrate is present in the anterior chamber or the vitreous humor. A few diseases like acute zonal occult outer retinopathy (also known as AZOOR) are often described as a uveitis despite the absence of a leukocytic infiltrate. In contrast, an autoimmune retinopathy is usually associated with a relative lack of infiltrating leukocytes and often ellipsoid zone loss. The ocular inflammatory disease associated with DS has aspects of uveitis and autoimmune retinopathy, in keeping with the observation that both anti-retinal antibodies and an autoimmune diathesis are characteristic of patients with DS. All the subjects in this study had bilateral involvement at the time of uveitis diagnosis. Cataract development (7 of 8 subjects) and ocular hypertension/glaucoma (7 of 8 subjects) were also common features in this subject group.
The bilaterality of the anterior uveitis presentations in this cohort is an interesting finding given that unilateral presentations are more common in the general population. A prospective study by McCannel et al., showed that 9.8% of anterior uveitis cases evaluated by community-based practices were bilateral compared to 32.6% in university referral practices23. Using the more conservative measure of bilaterality from university-based practice, the likelihood for 8 consecutive subjects all to be bilateral is (0.326)8 or p < 0.001. Similarly, the frequency of glaucoma, cataract, relatively similar age of onset, and severity prompting immunosuppression support the concept that indeed this is a distinct entity.
As noted in the introduction, patients with DS have higher rates of autoimmune conditions including diabetes mellitus3–5, celiac disease6,7, thyroid dysfunction9 and inflammatory arthritis10–12. A possible explanation for these increased rates may be due to a higher incidence of immune system dysregulation. Huggard et al., showed that children with DS have increased levels of inflammatory and anti-inflammatory cytokines when compared to age-matched controls; these cytokines include interleukin (IL)-1β, IL-2, IL-6, IL-10, IL-1 receptor antagonist (RA), erythropoietin (EPO), vascular endothelial growth factor (VEGF) and granulocyte-macrophage colony-stimulating factor (GM-CSF)15,24,25. Tumor necrosis factor (TNF)-α, and interferon (IFN)-γ have also been shown to be elevated in patients with DS15,26. Another study showed that patients with DS had lower levels of autoimmune regulator protein, AIRE, a transcription factor located on chromosome 21, resulting in reduced expression of peripheral antigens in the thymus. This phenomenon potentially leads to a failure of central tolerance and a subsequent autoimmune predisposition14. In fact, AIRE deficient mice are a well characterized model of uveitis due to an autoimmune response to interphotoreceptor retinoid binding protein (IRBP)27. Malle and colleagues found that activation of the phosphorylated STAT-3-IL-6 pathway was frequent in patients with DS 15. Our report is a retrospective study on a small number of subjects. As such, we did not attempt to compare therapeutic interventions. However, after the preparation of this manuscript, we treated one DS subject with tocilizumab, a solute IL-6 receptor. She developed macular edema following cataract surgery and responded extremely well to the inhibition of IL-6.
AIRE deficiency in humans results in a syndrome called APS-1, autoimmune poly-glandular syndrome. In a series of 91 patients from Finland with APS-1, 6 had iridocyclitis and 2 had retinal dystrophy13. Some of our subjects could be incorrectly diagnosed with a retinal dystrophy due to irregularity in the ellipsoid zone of the retina on OCT (optical coherence tomography) imaging and a relative lack of a cellular infiltrate in the uveal tract.
Patients with DS are considered a vulnerable population such that medical research involving patients with DS should adhere to strict regulation. This appropriate ethical standard made it more difficult to test for anti-retinal antibodies in subjects who resided outside of the Pacific Northwest. We had access to only 3 sera, each of which tested positive. About 10 percent of healthy adults do have detectable anti-retinal antibodies 21,22. To have 3 consecutive subjects test positive coincidentally would be expected at a rate of (0.1)3 or one in a thousand. Subjects in this study were not tested for antibodies to ATP1B2, a retinal antigen and the target of an immune response in patients with DS as reported elsewhere15. ATP1B2 is a membrane bound protein that anchors a protein called retinoschisin.16 Antibodies to ATP1B2 are not detectable with the methodology used by the OHSU Ocular Immunology Laboratory.
A limitation of our study is that the association between uveitis and DS could result from chance since DS is common among chromosomal abnormalities. However, the predisposition to autoimmune conditions in patients with DS and the similarity in uveitis presentations seen in this case series, with regard to age of onset, chronicity, complications such as cataract and glaucoma, bilateral anterior and intermediate findings, and severity such that immunosuppressive therapy was common all suggest a real association between DS and uveitis. This association is made more plausible by the known deficiency of the AIRE protein, the detection of anti-retinal AAbs in the 3 consecutive subjects who were tested, and the autoimmune diathesis characteristic of patients with DS15.
Our series should not be considered an epidemiologic study. There are multiple reasons why an ophthalmologist might prefer to not participate in a survey study such as ours. Portland, OR, USA is a metropolitan area of about 2 million people. In 2010, the population prevalence of DS in the US was estimated to be 6.7 for every 10 000 people28, meaning that about 1 300 people should be living with DS in Portland metropolitan area. The US prevalence of uveitis is about 1.2 per 1 00029. This would mean that one or two individuals with DS in the Portland area would be expected to have uveitis. And this is potentially an overestimate since many of the DS patients would be too young for the typical onset of uveitis. Our two participating practices in Portland evaluate only a small fraction of the population in Portland, which has four large, competitive health care systems. Thus, our experience in Portland where four subjects were identified leads us to believe that uveitis is not just characteristic in presentation in DS; it is also probably increased.
In summary, although autoimmune diseases are associated with DS, the autoimmune eye disease has not been previously characterized. The recognition of this association has implications regarding pathogenesis and predicting prognosis. We hypothesize that the recognition of this association will lead to improvements in therapy based on understanding the immunologic abnormalities associated with DS.