While episcleritis has been classically described as self-limited, scleritis may require systemic immunosuppression and be more commonly associated with systemic diseases and ocular complications.1 Prior studies are limited to retrospective case series from tertiary care settings, which may overestimate ocular complications and systemic disease associations due to referral bias.2,3 Therefore, there is a need for a population-based study describing the clinical characteristics of scleritis and episcleritis, including associated diagnoses, complications, and treatments. We used Kaiser Permanente Hawaii (KPH) as a source of data. This retrospective cohort study compares clinical characteristics of KPH members with scleritis and episcleritis, and elucidates practice patterns in a general population.

To that end, electronic medical records for KPH enrollees (n = 217,061) from January 1, 2006 to December 31, 2007 were searched for International Classification of Diseases, 9th ed., codes corresponding to ocular inflammation as previously described.4 Clinical diagnoses of scleritis and episcleritis were confirmed by chart review. Associated diagnoses, ocular complications, prescribed medications, and smoking status of scleritis and episcleritis patients were compared using Fisher's exact test. Patients were followed for 2 years after the study period for the development of associated diagnoses. A p value less than 0.05 was considered statistically significant. STATA 11.0 (StataCorp, TX) was used. Institutional review board/ethics committee approval was obtained.

Seventeen scleritis and 93 episcleritis cases were confirmed. Reasons for exclusion have been previously described.4 A total of 76.5% of scleritis cases and 60.2% of episcleritis cases were female. The median age at diagnosis was 56 years for scleritis (interquartile range [IQR]: 42–60) and 45 years for episcleritis (IQR: 32–54).

Clinical characteristics of scleritis and episcleritis varied. The most common associated diagnosis was rheumatoid arthritis, affecting 29.4% of scleritis patients and 0% of episcleritis patients (p < 0.001). Additionally, there was 1 scleritis patient with inflammatory polyarthropathy and 2 episcleritis patients with psoriasis. In the 2 years after the study period, 1 scleritis patient developed systemic lupus erythematosus and sarcoidosis and 1 episcleritis patient developed psoriasis.

Ocular complications and prescribed medications varied among scleritis and episcleritis patients. Ocular complications occurred in 29.4% of scleritis patients and 7.5% of episcleritis patients (p = 0.02, Table 1). Notably, anterior uveitis complicated 11.8% of scleritis cases, compared to 0% of episcleritis cases (p = 0.02). Additionally, scleritis patients were more likely to have been prescribed cycloplegic eyedrops (p = 0.02), systemic immunosuppressants (p < 0.001), and systemic corticosteroids (p = 0.03). Additionally, 23.5% of scleritis patients and 32.2% of episcleritis patients were smokers (p = 0.58).

Our population-based study confirms the burden of rheumatoid arthritis previously described in scleritis patients.2,3,5 Furthermore, our results suggest that prior studies overestimate systemic disease associations and ocular complications, especially in episcleritis, likely due to referral bias.2,3 For example, in contrast to a recent study from specialty eye care practices reporting that 15.3% of episcleritis patients suffered from a connective tissue or vasculitic disease, we found no such episcleritis patients.3 The discrepancy in these findings may be due to referral of complicated episcleritis cases to tertiary care centers.

Prescribed therapies offer insight into practice in a general population. The differential prescription of cycloplegic eyedrops to scleritis patients is likely related to the complication of anterior uveitis. As in tertiary care centers, patients with scleritis were more likely to be prescribed systemic immunosuppressants than episcleritis patients.2,3 We saw no statistically significant difference in the prescription of nonsteroidal anti-inflammatory drug (NSAID) or corticosteroid eyedrops; however, a study from a tertiary eye care center showed that 50% of episcleritis patients and 2.9% of scleritis patients were treated with topical corticosteroids.2 This may indicate a difference in beliefs about the utility of topical corticosteroids among various practitioners or a difference in disease severity.

There are limitations to acknowledge. While oral NSAIDs are a common treatment, our methods did not capture their use since they do not require a prescription. Furthermore, three factors may result in the underascertainment of cases: patients seeking care outside of Kaiser Permanente Hawaii, miscoding of diagnoses, and misdiagnosis of scleritis or episcleritis. However, the likelihood of each is low. First, less than 5% of Kaiser Permanente Hawaii patients have alternate insurance. Secondly, the initial screen based on ICD-9 codes was so broad that nearly 9 out of 10 cases initially identified were excluded after medical record review, reducing the likelihood of missed cases due to miscoding. Finally, misdiagnosis is unlikely given the clinical expertise of the ophthalmologists in the Kaiser Permanente Hawaii system, which includes comprehensive ophthalmologists and specialists in ocular inflammatory disease. Additionally, differences in the racial composition of Hawaii and the remainder of the United States raise questions of generalizability. Nevertheless, aspects of our data are consistent with studies from other populations (i.e. female predominance and age at presentation).2,3

In conclusion, our study provides population-based comparisons of the clinical characteristics of scleritis and episcleritis, which have been lacking, in a unique population. Furthermore, these results confirm interesting differences in systemic disease associations that have been observed in previous studies and suggest that prior studies overestimate systemic disease associations and ocular complications, especially in episcleritis, likely secondary to referral bias. By highlighting the high prevalence of the associated systemic diagnosis of rheumatoid arthritis among scleritis patients, even in a general population, our study lends credence to the need for a systemic evaluation of scleritis patients. Furthermore, the higher rate of complications noted in scleritis patients, compared to episcleritis patients, points to the need for aggressive systemic therapy in this group.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.