Ocular manifestations have long been reported as common findings in SLE. Although not part of the ACR diagnostic criteria, it has highest possible effect on the disease activity according to SLEDAI. While the literature is conclusive that ocular involvement is common and merits attention, it is not known whether ocular involvement is present at the time of diagnosis or develops later in the disease process. Looking to answer this question, this study is the first report of the prevalence of retinopathy in SLE at the time of diagnosis.
The most important finding of this study is that 15.8% of newly-diagnosed SLE patients had evidence of retinopathy, as examined by an ophthalmologist. This is substantially higher than previous reports on ambulatory patients, but still lower than admitted patients with an active disease, in whom a prevalence of up to 50% has been reported [1,2,3,4, 6,7,8,9, 13, 17].
Based on isolated case reports, some authors have suggested that ocular manifestations might be a presenting symptom of SLE [1, 3, 5, 13, 17, 20]. As stated previously, the literature on the subject is limited to patients already being treated for SLE, and therefore this conclusion is not backed by evidence. We would argue to the contrary of this statement, as we did not find a single ophthalmic symptom in 114 patients who were newly-diagnosed with SLE. Periorbital and anterior-segment manifestations of SLE, especially dry eye syndrome, are commonly associated with secondary pathologies including Sjogren’s syndrome. Therefore, we did not expect to find a high prevalence of such involvement. However, we believe that 16% of asymptomatic patients having evidence of retinal involvement is an alarmingly-high rate for newly-diagnosed patients and warrants a paradigm shift from a reactionary approach to ocular involvement towards incorporating ophthalmic examination as a screening test for all patients who are diagnosed with SLE. To decrease the confounding effect, we excluded patients with underlying hypertension, diabetes mellitus, and coagulopathy (either primary or drug-induced) to have a clear picture of retinal involvement in SLE. Also, we took extra care to only include patients who had never received treatment for SLE to prevent misdiagnosing medication-related retinal toxicity. However, while the stringent inclusion criteria we employed in this study decreases the likelihood of a false-positive diagnosis, it also means that the prevalence of retinopathy in the everyday clinical setting would be higher than observed in this study.
The most common findings among patients with retinopathy were cotton-wool spots, retinal hemorrhage, and Roth spots, which is in line with previous studies [3, 4, 6, 8,9,10,11, 13]. All of these pathologies reflect vascular damage. We did not find any cases of retinal vein or artery occlusion or choroidopathy, which might be due to the fact that our patients were newly-diagnosed and these lesions might be indicative of a longer duration of the disease. Although patients with retinopathy tended to be younger, clinical findings and diagnostic criteria were not significantly different between groups. However, patients with retinopathy had significantly lower hemoglobin levels. Seth et al. reported similar findings, and also reported a higher incidence of autoimmune hemolytic anemia in SLE patients with retinopathy [7]. Similar to previous reports, we found that patients with retinopathy have lower C3 and C4 concentrations, and have higher levels of ANA and Anti-dsDNA, findings which indicate an antibody-mediated retinal damage pathogenesis [1,2,3,4,5,6, 8,9,10,11, 13, 17]. Urine casts were also significantly more prevalent in patients with retinopathy.
Retinopathy has been linked to a more severe disease and poor prognosis by several authors [1,2,3,4,5,6,7,8,9,10,11, 13, 17, 21]. It should be noted that older studies used legacy disease activity indices, which did not score ocular involvement as a marker of disease severity [2]. However, with SLEDAI as the gold-standard disease activity index, and the fact that ‘visual disturbance’ adds 8 points to the index, this comparison is not meaningful, as any group of patients with ocular involvement will have an 8-point advantage over patients without. Quite expectedly, we found that patients with retinopathy had a significantly higher SLEDAI. However, this comparison did not hold significance when 8 points were deducted from patients with retinopathy (P = 0.06). Although with this unvalidated method, we are not trying to imply that ocular involvement is not related to disease activity, we would like to emphasize that with the current methodology (SLEDAI), it might not be feasible to infer the correlation between retinopathy and disease activity. Retinopathy has also been linked to a poor prognosis, which we cannot comment on owing to our study design. However, we recommend a reappraisal of evidence surrounding these two concepts: the correlation between disease activity and retinopathy, and the effect retinopathy might have on the patient’s prognosis with current treatment paradigms.
We acknowledge limitations to our study, including those inherent to the study design. While we did not include patients undergoing treatment for SLE, we did not assess the effect of treatment on ocular involvement. This is the subject of a current study at our institution. Also, the prevalence of retinopathy among patients referred to a referral SLE clinic might be different than a more general setting, which we did not have a workaround for. On the other hand, excluding patients treated for SLE and those with hypertension and diabetes, and performing an a priori power analysis are among the strengths of this study.