To identify risk factors for a severe uveitis course among children with noninfectious uveitis.
Retrospective cohort study.
This was a retrospective analysis of a prospectively collected database. Records of 94 children with uveitis were reviewed at enrollment and every 3–6 months (2011–2015). Severe uveitis was defined as a history of ocular complications or a visual acuity (VA) of ≤20/200. Children were compared by disease, VA, complications, and race. Regression models were used to model risk factors for severe disease. When examining race, we focused on non-Hispanic African-American and non-Hispanic white children only.
Of 85 children with uveitis and complete ocular examinations, 27 (32%) had a history of a VA of ≤20/200. A subanalysis of non-Hispanic African-American and white children showed an increased prevalence of VA ≤20/200 in non-Hispanic African-Americans (18/25; 72% vs 4/43; 9%). Non-Hispanic African-Americans were more likely to be diagnosed at an older age ( P = .030) and to have intermediate uveitis ( P = .026), bilateral disease ( P = .032), a history of VA ≤20/50 ( P = .002), VA ≤20/200 ( P < .001), and a higher rate of complications ( P < .001). On multivariable analysis, non-Hispanic African-American race was a significant predictor of blindness (OR = 31.6, 95% CI 5.9–168.5, P < .001), after controlling for uveitis duration. Non-Hispanic African-Americans also developed 2.2 times more unique complications per year of disease than non-Hispanic whites when controlling for uveitis type and duration.
There appear to be racial differences in the outcomes of children with uveitis. Non-Hispanic African-American children with non–juvenile idiopathic arthritis–associated uveitis may have worse visual outcomes with increased vision loss and ocular complications. These findings highlight the need for future studies in minority populations.
Pediatric noninfectious uveitis can lead to severe ocular complications and permanent vision loss. Uveitis can occur as a primary disease without systemic associations (idiopathic uveitis) but can also be related to other diseases such as juvenile idiopathic arthritis (JIA), sarcoidosis, and Behcet disease.
In North America, pediatric uveitis is most commonly associated with JIA, affecting 10%–20% of children with inflammatory arthritis. Three percent to 66% develop vision loss and ocular complications related to risk factors such as male sex, short duration between arthritis and uveitis diagnoses, a uveitis diagnosis prior to arthritis diagnosis, young age at uveitis onset, and a poor initial ocular examination (ie, anterior chamber cell score ≥1+, initial visual acuity [VA] of 20/200 or worse, or presence of complications). However, few studies have examined risk factors in a prospective cohort of children. Early identification and treatment of those at greatest risk for severe disease may prevent devastating ocular sequelae.
Race plays a significant role in various autoimmune conditions such as JIA, systemic lupus erythematosus, and sarcoidosis. In JIA, children of European descent predominantly develop oligoarticular disease, whereas non-European children develop polyarticular rheumatoid factor (RF)-positive JIA with worse arthritis outcomes. Since JIA subtypes differ in their association with uveitis, and race predisposes one to different JIA categories, race may influence uveitis risk. Less extensive investigation has been conducted in racial association with pediatric uveitis.
The primary objective of this study is to identify risk factors for a severe disease course among children with noninfectious uveitis, defined as a history of ocular complications or a VA of 20/200 or worse.
Data were collected retrospectively and prospectively. Some children in our cohort were enrolled after their initial uveitis diagnosis, requiring retrospective data collection. However, after enrollment, all data were collected prospectively, including that of children who were enrolled close to their time of uveitis diagnosis. Patients were managed in the Division of Pediatric Rheumatology of the Emory Children’s Center. This study, including medical chart data collection from pediatric rheumatologists and ophthalmologists, was approved by the Institutional Review Board of Emory University and conformed to the requirements of the US Health Insurance Portability and Accountability Act. Informed consent was obtained from the parent/legal guardian, and assent was obtained from the children if applicable. Research adhered to the tenets of the Declaration of Helsinki.
Eligible children presenting with a diagnosis of noninfectious uveitis at less than 18 years of age were enrolled from the pediatric rheumatology outpatient clinics at Emory Children’s Center from November 30, 2011 to January 8, 2015. They were followed prospectively from time of enrollment and returned for their usual follow-up appointments every 3–6 months. Inclusion criteria included a diagnosis of noninfectious uveitis and English speaking. Exclusion criteria included refusal to participate.
We reviewed rheumatology and ophthalmology medical records from time of diagnosis to time of presentation to our clinics and then every 3–6 months during the usual pediatric rheumatology follow-up visit until March 1, 2015. Data collected at the baseline visit included date of birth, sex, self-described race and ethnicity, JIA category established by the International League of Associations for Rheumatology (ILAR) classification in children with arthritis, uveitis characteristics (onset date; diagnosis date; laterality; location; ocular complications such as cataracts, glaucoma, synechiae, band keratopathy, ocular hypertension, cystoid macular edema, and amblyopia; and ophthalmic surgeries), ocular examinations (best-corrected VA, intraocular pressure, and anterior chamber cell score), laboratory values (antinuclear antibody [ANA], RF, erythrocyte sedimentation rate [ESR], HLA-B27, and angiotensin-converting enzyme [ACE]), and both past and current medications. Race was designated as American Indian or Alaskan Native, Asian, black or African-American, Native Hawaiian or other Pacific Islander, and white by self-report. Ethnicity was also self-reported as Hispanic or non-Hispanic. Since children without a diagnosis of JIA do not commonly undergo regular screening examinations, an accurate determination of disease duration is challenging. In this study, disease duration refers to time from uveitis diagnosis to analysis. Follow-up data on arthritis, uveitis, laboratory values, and medications were collected at 3- to 6-month intervals (time of last study visit to current study visit) and during uveitis flares. Data from the ocular examination were recorded from every ophthalmology visit with varied intervals depending on disease activity. Not all patients had complete data from their initial ocular examination or subsequent visits, as some patients had chronic uveitis prior to enrollment and records were unavailable.
Statistical analyses were conducted using SAS 9.3 for Windows (SAS Institute, Inc, Cary, North Carolina, USA). Statistical significance was assessed at the 0.05 level unless otherwise noted. Data are summarized using means and standard deviations, medians, and interquartile ranges (25th–75th percentile) or counts and percentages, when appropriate. Duration of uveitis was calculated as time between dates of uveitis diagnosis to last study visit. For statistical purposes, we converted Snellen VA to logMAR VA. We compared the following: (1) children with JIA-associated uveitis to those with other forms of uveitis; (2) children with a history of a VA of 20/200 or worse to those with a VA better than 20/200; (3) children with and without a history of ocular complications; and (4) children of non-Hispanic African-American and non-Hispanic white race, using χ 2 tests and 2-sample t tests. In instances of small expected cell counts (<5), exact χ 2 tests were used. When continuous data were skewed or non-normal, a nonparametric test (Mann-Whitney U or Kolmogorov-Smirnov test) was used in place of the 2-sample t test.
Multivariable logistic regression was used to simultaneously model multiple risk factors for development of complicated disease (ie, blindness or complications). Candidate predictors identified in the univariate analyses ( P < .1) were entered into the model using backwards elimination. Variables were systematically removed until all variables in the model were significant at the 0.05 level or until a significant reduction in model fit was observed. In addition, other variables that were demonstrated to be associated with the risk of poor outcome and race (eg, duration of disease defined as time from disease diagnosis to analysis) were included in the models. In addition, to compare the number of ocular complications among non-Hispanic African-American and non-Hispanic white children, multivariable negative binomial models were used to model the number of unique complications developed while adjusting for disease duration. To further examine the role of non-Hispanic African-American race with poor visual outcomes in a more homogenous cohort of children with uveitis, a subanalysis was performed in non-Hispanic white and non-Hispanic African-American children with only JIA-associated uveitis or chronic anterior uveitis.
Demographics and Clinical Characteristics of Cohort
Demographic and clinical characteristics of 94 children with uveitis are presented in Table 1 . Of these, 52 (55%) also had JIA, 30 (32%) had idiopathic uveitis, 6 (6%) were HLA-B27 positive, 2 (2%) had sarcoidosis, and 4 (4%) had other types of uveitis associated with vitritis, nodular scleritis, pars planitis, and sympathethic ophthalmia. There were 57 white (61%) and 28 African-American children (30%). Eleven children (12%) were Hispanic. The median (25th–75th) age at uveitis diagnosis was 6.3 years (4.0–10.6). Many had anterior uveitis (n = 65; 74%) with bilateral involvement (n = 62; 70%). Most common ocular complications included synechiae (n = 41, 44%) and cataracts (n = 38, 40%).
|Characteristics||Overall N = 94||JIA-Associated Uveitis |
N = 52 (55.3%)
|Other Types of Uveitis |
N = 42 (44.7%)
|Age at most recent visit, median (25th–75th)||11.8 (8.5–14.9)||13.3 (8.4–14.6)||13.3 (9.2–15.0)||.225|
|Sex, female||66 (70.2%)||40 (76.9%)||26 (61.9%)||.173|
|White||57 (60.6%)||38 (73.1%)||19 (45.2%)||.003 b|
|African-American||28 (29.8%)||8 (15.4%)||20 (47.6%)|
|Other||9 (9.6%)||6 (11.5%)||3 (7.1%)|
|Hispanic||13 (13.8%)||10 (19.2%)||3 (7.1%)||.184|
|Private||54 (57.5%)||31 (59.6%)||23 (54.8%)||.863|
|Medicaid||37 (39.4%)||19 (36.5%)||18 (42.9%)|
|None||3 (3.2%)||2 (3.9%)||1 (2.4%)|
|Age at uveitis diagnosis (y), median (25th–75th)||6.3 (4.0–10.6)||4.8 (3.5–10.4)||8.7 (5.3–10.7)||.006 b|
|Location of disease a (N = 88)|
|Anterior||65 (73.9%)||38 (80.9%)||27 (65.9%)||.110|
|Intermediate||5 (5.7%)||0 (0.0%)||5 (12.2%)||.014 b|
|Posterior/panuveitis||4 (4.6%)||0 (0.0%)||4 (9.8%)||.029 b|
|Unknown||14 (15.9%)||9 (18.0%)||5 (12.2%)||.373|
|Bilateral disease a (N = 89)||62 (69.7%)||34 (72.3%)||28 (66.7%)||.562|
|Worst ocular examination a (N = 85)|
|Slit-lamp examination (cells)|
|0 & 0.5+||19 (22.4%)||11 (23.9%)||8 (20.5%)||.797|
|1+ and worse||66 (77.6%)||35 (76.1%)||31 (79.5%)|
|Intraocular pressure, median (25th–75th)||20.0 (17.0–26.0)||19.0 (17.0–25.0)||21.0 (17.0–28.0)||.282|
|Visual acuity 20/50 or worse||49 (57.7%)||21 (44.7%)||28 (73.7%)||.009 b|
|Visual acuity 20/200 or worse||27 (31.8%)||9 (19.2%)||18 (47.4%)||.009 b|
|Type of complications|
|Cataracts||38 (40.4%)||16 (30.8%)||22 (52.4%)||.038 b|
|Glaucoma/ocular hypertension||16 (17.0%)||9 (17.3%)||7 (16.7%)||1.000|
|Synechiae||41 (43.6%)||16 (30.8%)||25 (59.5%)||.007 b|
|Band keratopathy||28 (29.8%)||13 (25.0%)||15 (35.7%)||.364|
|Cystoid macular edema||22 (23.4%)||8 (15.4%)||14 (33.3%)||.052|
|Initial ocular examination a|
|≥1+ cells at first examination (N = 46)||29 (63.0%)||16 (61.5%)||13 (65.0%)||1.000|
|Intraocular pressure >21 (N = 25)||2 (8.0%)||1 (7.7%)||1 (8.3%)||1.000|
|LogMAR visual acuity (N = 42), median (25th–75th)||0.18 (0.00–0.40)||0.18 (0.00–0.30)||0.18 (0.10–0.48)||.622|
|Laboratory values a|
|ANA positive (N = 92)||37 (40.2%)||28 (54.9%)||9 (22.0%)||.003 b|
|Earliest ESR (N = 89) median (25th–75th)||11.0 (6.0–29.0)||22.0 (8.0–38.0)||9.0 (3.0–14.0)||<.001 b|
|HLA-B27 positive (N = 62)||16 (25.8%)||6 (18.8%)||10 (33.3%)||.250|
|Medication use a (N = 93)|
|Methotrexate oral||57 (61.3%)||36 (70.6%)||21 (50.0%)||.055|
|Methotrexate subcutaneous||65 (69.9%)||38 (74.5%)||27 (64.3%)||.365|
|Infliximab||27 (29.0%)||17 (33.3%)||10 (23.8%)||.364|
|Adalimumab||17 (18.3%)||12 (23.5%)||5 (11.9%)||.184|