Course, complications, and outcome of juvenile arthritis related uveitis

Transcription

1 Major Articles Course, complications, and outcome of juvenile arthritis related uveitis Kourosh Sabri, MB, ChB, FRCOphth, a Rotraud K. Saurenmann, MD, b,a Earl D. Silverman, MD, FRCPC, b,c and Alex V. Levin, MD, MHSc, FRCSC a,c PURPOSE To describe the clinical features of uveitis in patients with juvenile idiopathic arthritis ( JIA). METHODS Retrospective chart review of a subset of 1,081 consecutive JIA patients who were younger than 18 years of age and had uveitis, with a minimum of 1-year follow-up at a single center. RESULTS One hundred forty-two patients (13.1%) developed uveitis after a mean follow-up of 6.3 years (range, ). Uveitis types were chronic anterior (97/142, 68.3%), acute anterior (23/142, 16.2%), recurrent anterior (17/142, 12%), and panuveitis (5/142, 3.5%). Uveitic complications were observed in 37.3% of cases (53/142) and 32.5% of eyes (74/228). When we compared uveitic eyes with complications to uveitic eyes without complications, we found the following significant differences: time interval from diagnosis of JIA to diagnosis of uveitis was shorter (mean, 1.3 years vs. 2.2 years; p 0.003) and use of oral prednisone was greater (59.1% vs 15.6%; p ) in the eyes with complications. Twenty-one children (21/142, 14.8%) with uveitis underwent a total of 62 ocular surgeries. Good visual acuity (20/40 or better) was found in 90.8% of eyes (159/175) and in both eyes of 87% of cases (94/108), impaired visual acuity in 6 eyes of 4 cases (6/175, 3.4%), and blindness in 10 eyes of 10 cases (10/175, 5.7%). Only 2 patients had reduced visual acuity in both eyes. Surgery was the single most important risk factor for reduced visual acuity at the last follow-up ( p ). CONCLUSIONS Most uveitic eyes with JIA achieved good visual outcome despite uveitic complications. ( J AAPOS 2008;12: ) U veitis in children can be classified as primary or secondary. One of the most frequently found causes of secondary uveitis is juvenile idiopathic arthritis ( JIA). It is currently suggested by many authors that the criteria defined by the International League of Associations for Rheumatology 1,2 should be used in the classification of JIA. By using this classification, clinicians can divide JIA into major subtypes based on the number of See editorial on page 537. Author affiliations: a Department of Ophthalmology and Vision Science, b Division of Rheumatology, and c Department of Paediatrics, The Hospital for Sick Children, University of Toronto, Toronto, Canada Presented in part at the 33rd Annual Meeting of the American Association for Pediatric Ophthalmology and Strabismus, Seattle, April 11-15, This study was supported by the Swiss Arthritis Society, Stiftung Sanitas (Switzerland), and Brandan s Eye Research Fund. The first two authors contributed equally to the research and manuscript. This research was conducted in the Departments of Paediatrics and Ophthalmology & Vision Science at The Hospital for Sick Children in Toronto. Submitted August 10, Revision accepted March 17, Published online September 16, Reprint requests: Dr. Alex V. Levin, Department of Ophthalmology, The Hospital for Sick Children, 555 University Avenue, Toronto M5G 1X8 Canada ( alex. levin@sickkids.ca). Copyright 2008 by the American Association for Pediatric Ophthalmology and Strabismus /2008/$ doi: /j.jaapos joints involved in the first 6 months (oligoarticular or polyarticular) and the presence or absence of fever (systemic JIA). Further subclassification would then occur within both the polyarticular (depending on the presence or absence of rheumatoid factor) and oligoarticular subtypes (depending on extension of the arthritis to more than 4 joints after the initial 6 months, all rheumatoid factor negative). In addition, patients with psoriasis have a separate subclassification as do patients with HLA-B27-associated arthritis (enthesitis-related arthritis). Uveitis is the most common extra-articular involvement found in patients with JIA and can be acute anterior, recurrent anterior, chronic anterior, or anterior uveitis with vitritis. 3 Chronic anterior uveitis is the most common type of uveitis found in some JIA cohorts. 4 The most important complication of JIA-associated uveitis is visual loss with significant visual impairment in 3% to 66% of JIA patients who develop uveitis. 4,5-14 There are mixed reports in the literature regarding the incidence of JIA associated uveitis; the authors of some reports indicate a reduction 15 whereas others suggest no change. 16 We studied the course of uveitis in the 142 of 1,081 JIA patients who developed intraocular inflammation. We have previously described the prevalence and risk factors 539

2 540 Sabri et al Volume 12 Number 6 / December 2008 Table 1. Demographic data on 142 patients who developed uveitis with and without ocular complications Without ocular complications* With ocular complications* Total cases, n (%) 89 (62.7) 53 (37.3) Gender, n (%) Female (n 113) 74 (65.5) 39 (34.5) Male (n 29) 15 (51.7) 14 (48.3) JIA subtype, n (%) Persistent oligoarticular (n 57) 37 (64.9) 20 (35.1) Extended oligoarticular (n 30) 17 (56.7) 13 (43.3) Polyarticular (n 32) 21 (65.6) 11 (34.4) Psoriatic (n 12) 7 (58.3) 5 (41.7) Enthesitis related (n 9) 5 (55.5) 4 (44.5) Systemic (n 1) 1 (100) 0 Other (n 1) 1 (100) 0 Age in years at diagnosis of JIA (mean, median, range) 4.20, 3.10, , 3.10, Age in years at diagnosis of uveitis (mean, median, range) 6.40, 5.30, , 4.80, Interval in years from diagnosis of JIA to diagnosis of uveitis (mean, median, range) 2.20, 1.30, , 0.58, *Complications: posterior synechiae, cataract, band keratopathy, glaucoma, or macular edema. Sign in range indicates uveitis presenting before JIA. p for development of uveitis in this cohort. 17 We now describe the features and complications of their uveitis. Subjects and Methods Patients A retrospective chart review at the Hospital for Sick Children in Canada revealed 1,081 children with a confirmed diagnosis of JIA made before June 30, 2002, all of whom had ocular examination. In addition, telephone and questionnaire surveys were used to obtain the necessary clinical information. Detailed explanation of the methodology regarding selection of the patient cohort and data collection has been previously published. 17 The study was approved by the Research Ethics Board of the Hospital for Sick Children (Canada), and informed consent was obtained for all patients prior to enrollment. The results of these investigations have been previously reported in part. 17 As detailed in our previous report, 17 we categorized our patients in accordance with the types of uveitis as classified by the International Uveitis Study Group Recommendations for the evaluation of intraocular inflammatory disease 3 : 1. Acute anterior uveitis: anterior uveitis with sudden onset and 3 months duration. 2. Recurrent anterior uveitis: uveitis of 3 months duration but with recurrent episodes with intervals between episodes of 3 months duration without medication. 3. Chronic anterior uveitis: uveitis of 3 months duration. 4. Anterior uveitis with vitritis: does not include the spillover vitritis observed with anterior uveitis or the snowbanking seen with intermediate uveitis. As in our previous work, 17 we used the final best-corrected visual acuity ( VA) to define 3 categories of visual outcome: 1) good: VA of 20/40 or better; 2) impaired: VA between 20/50 and 20/100 (inclusive); and 3) legal blindness: VA 20/200 or worse. As in our previous report 17 statistical analyses were performed with the JMP IN 5.1 program (SAS Institute, Cary, NC). The 2 test, one-way analysis of variance, and Cox proportional hazards regression analysis were used to test for differences between groups. Results As noted in our previous report, 17 after a mean follow-up time of 6.9 years (range, ), uveitis developed in 13.1% of patients (142/1,081) and in 10.5% of eyes (228/ 2,162). The mean time from diagnosis of JIA to diagnosis of uveitis was 1.8 years (range, 4.2 months to 10 years), the mean age at diagnosis of uveitis was 6.2 years (range, years), and the mean ophthalmology follow-up duration postdiagnosis of uveitis was 6.3 years (range, years). Of the 1,081 JIA patients, 156 (14.4%) had a follow-up of less than 2 years subsequent to the diagnosis of JIA. Of these patients, 11 of 156 (7.1%) developed uveitis in that follow-up period. The majority of patients (87/142; 61.2%) who developed uveitis had oligoarticular JIA with either a persistent course (57/87) or an extended (30/87) course; second most common were patients with RF-negative polyarticular JIA (32/142; 22.5%), whereas uveitis was not seen in any patient with RF-positive polyarticular JIA and in only one patient with systemic JIA (Table 1). Approximately onethird of patients with uveitis (53/ % of cases; 74/ % eyes) developed secondary ocular complications defined as posterior synechiae, cataract, band keratopathy, glaucoma, or macular edema, the last-named as diagnosed by clinical examination and/or optical coherence tomography. The majority of patients who developed uveitis (79.6%) were female. Although male patients tended to have a greater rate of uveitis complications (48.3% vs. 33.5% in female patients), this difference was not statistically significant ( p 0.17). However, male patients were more likely to have earlier complications

3 Volume 12 Number 6 / December 2008 Sabri et al 541 Table 2. Uveitis complications in cohort of 142 children with uveitis Synechiae, n (%) Cataract, n (%) Band keratopathy, n (%) Glaucoma, n (%) Macular edema, n (%) Number of children (n 142) 31 (21.8) 33 (23.2) 20 (14.1) 24 (16.9) 8 (5.6) Number of eyes (n 228) 39 (17.1) 46 (20.2) 28 (12.3) 31 (13.6) 9 (3.9) Table 3. Ocular surgery on uveitic eyes Number of patients (n 21), n (%) Number of eyes (n 50), n (%) Interval from diagnosis of uveitis to surgery in years (mean, median, range) Visual outcome Good Impaired Blind Not known Cataract extraction 16 (76.2) 24 (48) 3.46, 1.76, Glaucoma surgery 13 (61.9) 16 (32) 5.62, 6.02, Vitrectomy 7 (33.3) 10 (20) 3.91, 0.75, than female patients ( p 0.008), including earlier synechiae formation ( p 0.03). Bilateral uveitis was more common than unilateral uveitis (86/142 children [60.6%] vs 56/142 children [39.4%]). 17 The most common type of uveitis was chronic anterior uveitis (97/142; 68.3%), followed by acute anterior uveitis (23/142; 16.2%), recurrent anterior uveitis (17/142; 12%), and panuveitis was rarely seen (5/142; 3.5%). 17 All 5 cases of panuveitis were seen in female patients: 2 unilateral, 1 severe bilateral, 1 with chronic anterior uveitis in the contralateral eye, and 1 with acute anterior uveitis in the contralateral eye. Table 2 details the complications observed in our cohort. The most common complication was cataract. This was true for both patients and eyes. There was a statistically significantly shorter time interval from diagnosis of JIA to first diagnosis of uveitis between the children with complications (mean, 1.30 years; range, 4.20 to years) versus the children without uveitic complications (mean, 2.20 years; range, 0.30 to 9.30 years; p 0.003; Table 1). Univariate analysis showed that time from diagnosis of arthritis to diagnosis of uveitis ( p 0.003), type of uveitis ( panuveitis had the worst prognosis; p ), and length of uveitis follow-up ( p 0.01) were all associated with the development of uveitic complications. Cox regression analysis confirmed that the type of uveitis ( p ) and the time interval from diagnosis of JIA to diagnosis of uveitis ( p 0.04) were statistically significantly associated with the development of uveitic complications, although there was a trend for increasing follow-up time to be associated with development of uveitic complications ( p 0.06). Twenty-one children (21/142; 14.8%) with uveitis underwent a total of 62 ocular surgical procedures as a result of uveitic complications (Table 3). In 12 cases, all surgical procedures were confined to one eye. Cataract extraction was the most common procedure as well as the most common first procedure, performed in 16/21 cases (76.2%). This was closely followed by glaucoma surgery, performed in 13/21 cases (61.9%). Surgery was the single most important risk factor for reduced visual acuity at the last follow-up ( p ; risk ratio 3,037, 295). However, the total number of surgeries, the types of surgery, and the time interval from uveitis onset to the first surgery were not statistically significant risk factors for a reduced visual acuity outcome. Good visual acuity was found in more than 90% of eyes (159/175 with best-corrected visual acuity available at last follow-up) and in both eyes of more than 80% of the cases (94/108; 87%). In contrast, impaired visual acuity was found in only 6 eyes of 4 different patients (6/175; 3.4%) and blindness in 10 eyes of 10 different patients (10/175; 5.7%). 17 Only 2 patients had reduced visual acuity in both eyes. One patient had bilateral visual impairment, whereas the other patient was visually impaired in one eye and legally blind in the other eye. Of the 175 eyes, 66 had uveitic complications, and their visual outcome was as follows: good, 50/66 (75.8%); impaired, 6/66 (9.1%); blind, 10/66 (15.2%). Among the 50 eyes with complications and good visual outcome, the most common complications were cataracts (28/50; 56%), glaucoma (23/50: 46%), and posterior synechiae (21/50; 42%). Among the 6 eyes with complications and impaired visual outcome, the most common complications were cataracts (5/6; 83%), band keratopathy (4/6; 66.7%), posterior synechiae (4/6; 66.7%), and glaucoma (2/6; 33.3%). Of 10 children, 10 eyes had a blind visual outcome. Of these children, 8 (80%) had oligoarticular JIA, and 5 of the eyes (50%) had panuveitis. Further details are reported elsewhere. 17 A more rapid progression to blindness was associated with increasing numbers of complications ( p ). Of the 10 eyes with a blind visual outcome, all had cataracts (100%) and 9 had posterior synechiae (90%). Multivariate analysis revealed that a blind visual outcome was associated with the following independent risk factors: cataract formation ( p 0.002), synechiae ( p 0.019), earlier age at diagnosis of uveitis ( p 0.01), type of uveitis ( p ), and history of ocular surgery ( p Of note, blindness occurred within 1 year of diagnosis of uveitis in 5/10 (50%) cases. Table 4 gives details of the JIA subtypes among the 175 eyes whose visual outcomes were known. Over 90% of eyes with polyarticular, oligoarticular-extended, and systemic JIA had a good visual outcome. Of the 10 eyes, 8

4 542 Sabri et al Volume 12 Number 6 / December 2008 Table 4. Visual outcome according to juvenile idiopathic arthritis subtype* Persistent oligoarticular (n 73), n (%) Extended oligoarticular (n 39), n (%) Polyarticular (n 40), n (%) Psoriatic (n 13), n (%) Enthesitis related (n 9), n (%) Systemic (n 1), n (%) Good 65 (89) 36 (92.3) 39 (97.5) 11 (84.6) 7 (77.8) 1 Impaired 2 (2.7) 1 (2.6) 0 1 (7.7) 2 (22.2) 0 Blind 6 (8.2) 2 (5.1) 1 (2.5) 1 (7.7) 0 0 *Data available for 175 of 228 eyes. Table 5. Visual outcome according to uveitis subtype* Anterior acute (n 24) Anterior chronic (n 125) Anterior recurrent (n 19) Panuveitis (n 7) Good 24 (100%) 115 (92%) 18 (94.7%) 2 (28.6%) Impaired 0 6 (4.8%) 0 0 Blind 0 4 (3.2%) 1 (5.3%) 5 (71.4%) *Data available for 175 of 228 eyes. (80%) with blindness were in patients with oligoarticular onset JIA. 17 Table 5 details the uveitis subtypes among the 175 eyes whose visual outcomes were known. While over 90% of eyes with anterior acute, anterior chronic and anterior recurrent uveitis had a good visual outcome; over 80% of eyes with panuveitis had a blind visual outcome (Cox regression, p ). Systemic corticosteroids were used in 39 patients (27.5%) and methotrexate in 63 patients (44.4%). Use of methotrexate was associated with a greater overall rate of: all complications ( p ), synechiae ( p 0.005), cataract ( p ), glaucoma ( p ), and blindness ( p 0.03). Similarly, systemic corticosteroid use was associated with a greater overall rate of: all complications, synechiae, keratopathy, cataract, macular edema, and glaucoma ( p for each of the complications) and blindness ( p ). When we compared the patients whose treatment with methotrexate was started before the diagnosis of uveitis with the patients in whom methotrexate was started after the diagnosis of uveitis, patients treated with methotrexate before the diagnosis of uveitis had less: cataracts ( p 0.04), glaucoma ( p 0.008), cataract extraction ( p 0.03), glaucoma surgeries ( p 0.03), and eye surgeries ( p 0.04); however, there was no statistically significant difference in the rate of all complications ( p 0.1), synechiae ( p 0.2), band keratopathy ( p 0.3), macular edema ( p 0.6), blindness ( p 0.4) or the requirement for systemic corticosteroids ( p 0.8). Logistic regression analysis revealed that only a reduction in the rate of glaucoma but not cataract extraction, glaucoma surgeries, cataracts or eye surgeries was associated with early methotrexate treatment. Among the 175 eyes with a known visual outcome, patients who received oral prednisone were more likely to develop complications than patients who did not received oral prednisone (39/66 [59.1%] of eyes with vs 17/109 or [15.6%] without uveitic complications; p ). Upon examining the use of topical mydriatic, we found that patients with complications were more likely to be treated with topical mydriatic later than those without complications (mean time to start of 0.79 years with a median of 0 years and range years vs a mean of 0.24 years with a median of 0 years and a range years; p 0.017). However, the difference in the interval between diagnosis of uveitis and the start of topical steroid therapy did not reach statistical significance for eyes with (mean, 0.16; median, 0; range 3 to 3.34 years) versus without (mean, 0.14; median, 0; range, 2.58 to 6.53 years) uveitic complications ( p 0.64). Discussion We report new data that characterize the uveitis of JIA. The overall prevalence of JIA associated uveitis in our study was 13.1%. As 14.4% of patients had less than 2 years of follow-up, only 7.1% of those had uveitis develop during the follow-up period, and the mean time to development of uveitis after JIA diagnosis was 1.8 years, then we might hypothesize that the actual rate of uveitis in our JIA cohort would have been higher if follow-up was longer for these patients. A recent report from the United Kingdom that also used the International League of Associations for Rheumatology classification criteria found a prevalence of JIA-associated uveitis of 11.4%. 13 Others have reported a prevalence ranging from 9.3% to 30.8%. 11,12,16 This large variation may be the result of differences regarding inclusion criteria, subtypes, and classification of juvenile arthritis, periods of follow-up, and systemic treatments used for the JIA prior to uveitis onset. As recently reported by our group on the same cohort, 17 the greatest rate of uveitis was found in patients with oligoarticular JIA (20.9%), followed by those with RFnegative polyarticular JIA (14%). Others have also found oligoarticular JIA and RF-negative polyarticular JIA to be associated with the highest prevalence of uveitis. 12 Sim et al 13 found uveitis to be more prevalent in extended oligoarticular JIA than persistent disease. Although at first glance our results supported their findings (30% in extended vs. 18% in persistent disease), analysis of oligoarticular subtype at the time of diagnosing uveitis showed that the onset of uveitis was more common in persistent disease compared with extended disease. 17 This association has not been reported by others to our knowledge. Our previous publication showed that ANA positivity, age 6 years at diagnosis of JIA, and JIA subtype were

5 Volume 12 Number 6 / December 2008 Sabri et al 543 Table 6. Ocular complications secondary to JIA-associated uveitis Study* Cohort size Cataract Band keratopathy Uveitic complications Ocular hypertension or glaucoma Posterior synechiae Cystoid macular edema Dana et al 4 43 cases 84% 70% 37% N/R 42% Chalom et al cases 14% 17% 11% 10% N/R Foster et al cases N/R N/R 42% N/R N/R Kotaniemi et al cases 21.2% 6.7% 7.7% N/R 7.7% Edelsten et al cases 14% N/R 11% N/R N/R Paroli et al cases 64.4% eyes 59.2% eyes 25% eyes N/R N/R Kump et al eyes 64% 46% 20% 58% 26% This study 228 eyes 142 cases 20.2% 23.2% N/R, result not reported. *Different studies have reported their results in terms of either cases or eyes. 12.3% 14.1% 13.6% 16.9% 17.1% 21.8% 3.9% 5.6% the significant independent risk factors for developing uveitis. 17 Among our uveitic children, a shorter interval between diagnosis of JIA and uveitis was a significant risk factor for developing uveitic complications (Table 1), with mean intervals of 2.2 years for patients without complications versus 1.3 years for patients with complications. A similar association has been reported by Chalom et al, 11 who found the interval from JIA to uveitis diagnosis to be 22 months for cases without complication versus 18 months for those with complications. This shorter time interval may represent a more aggressive form of uveitis, which is more likely to lead to complications. Comparing uveitic eyes with complication versus those without complication, we found no significant difference regarding time from uveitis diagnosis to start of topical steroid therapy. However, patients with complications were more likely to start topical mydriatic therapy later than those without complications. We cannot explain these findings, because a delay in starting mydriatic therapy would likely only affect the incidence of posterior synechiae and secondary pupillary block glaucoma a rare complication in our experience whereas a delay in beginning topical steroid therapy would be predicted to result in an increased complication rate, which was not the case. We did find the presence of synechiae to be a significant risk factor for visual impairment at the last follow-up. Although we have no evidence that posterior synechia cause poor vision, they likely reflect a more aggressive or chronic uveitis which in turn would raise the risk for an adverse outcome through other means such as cataract, glaucoma, or band keratopathy. The effect of mydriatic therapy on outcome requires further study. The significantly greater use of oral prednisone among uveitic children with versus without (59.1% vs 15.6%) ocular complications is most likely an indicator of the more severe disease process in the former group. The observed greater overall rate of uveitic complications among patients receiving methotrexate is most likely also an indicator of the more severe disease process which required use of this medication in this subgroup of the cohort. Cataracts were the most common complication found in our cohort (23.2% of patients, 20.2% of involved eyes) followed by posterior synechiae (21.8% of patients, 17.1% of eyes). These results are consistent with several 4,12,14,18,19 but not all reports: one study demonstrated that band keratopathy was more common (17% with band keratopathy vs 14% with cataracts). 11 Despite the relative consensus regarding the identity of the most common uveitic complications, there has been some variation regarding the reported rates (Table 6). 4,11,12,14,18-20 There are several reasons as to why there may be such large differences in the reported rates of secondary complications among JIA-associated uveitic children. Differences between studies are found regarding inclusion criteria for JIA, type of referral center where the study was conducted, whether the primary investigators were rheumatologists or ophthalmologists, length of follow-up, and treatment policies for uveitis and arthritis. For example, a study conducted at a tertiary center dealing with the more severe uveitic and arthritic cases may be expected to overestimate the complication rates. Furthermore, differences in the timing and actual treatment protocols for uveitis and arthritis may affect the observed complication rates. Longer periods of follow-up may be expected to demonstrate a greater complication rate. Despite a mean follow-up of 6.2 years after diagnosis of uveitis, our observed complication rates were lower than many other studies mentioned above. We believe that the lower complication rates in our study are caused by frequent use of systemic anti-tumor necrosis factor 21 agents and, as we speculated in our previous report, 17 the aggressive use of chronic high-dose topical steroids often at frequencies of every 1 to 2 hours with very slow weans over months if not years, judicious use of eye surgery (eg, no surgery for an asymptomatic monocular cataract in a child beyond the age of amblyogenesis), and close collaboration between the pediatric ophthalmologists and rheumatologists with a multidisciplinary approach to management. It would be

6 544 Sabri et al Volume 12 Number 6 / December 2008 Table 7. Vision outcome among children with JIA-associated uveitis Final visual acuity Study* Cohort size Mean follow-up in years 20/40 20/50 Legally blind Cassidy et al 6 38 cases (67 eyes) 4.8 N/R 37% cases 15% eyes Cabral et al eyes 9.4 N/R 15% 10% Chalom et al cases N/R N/R 11% 0 Kotaniemi et al cases % 2.9% 0 Edelsten et al eyes 4 years (median) N/R 6.5% 4% Sim et al cases (30 eyes) 10 90% eyes N/R 4.3% cases This study 175 eyes % 9.1% 5.7% 108 cases 98.1% 1.9% 0.9% N/R, result not reported. *Different studies have reported their results in terms of either cases or eyes. difficult to do a comparative study with another center to investigate whether these assumptions are correct. A review of the literature during the last 3 decades shows an overall improvement in the visual outcome of JIA-associated uveitis (Table 7). Despite the overall trend of reported improving visual outcome in JIA-associated uveitis, there are still some recent studies that place the visual loss much higher. Kump et al 14 carried out a chart review of 89 children with JIA-associated uveitis with a mean follow-up period of 2.96 years. The baseline visual acuities in this cohort were as follows: 20/40 or better in 64% eyes, 20/50 to 20/150 in 14% eyes, 20/200 or worse in 22% eyes. On final examination, they found 20/40 or better in 73% of eyes, 20/50 to 20/150 in 9% of eyes, and 20/200 or worse in 18% of eyes. Although there was an apparent improvement in visual acuity over time, this trend did not reach statistical significance. Once again, the same variables regarding study design, inclusion criteria, treatment protocols, and follow-up periods mentioned above may explain some of the variations in the reported visual outcomes of JIA-associated uveitic children. In our study, the uveitic cohort was followed for a longer period than in most other studies, and yet the visual outcomes were better than most other reports. We believe this to be related to our lower complication rates. In this large retrospective review of JIA-associated uveitis at a tertiary referral center, we have found the secondary complication rates to be relatively low and the overall final visual outcome very good. Regular screening, close collaboration between ophthalmologists and rheumatologists, early aggressive use of anti-tumor necrosis factor agents and steroids may have been important in achieving these results. Literature Search The literature search for this study was done using the following databases: MEDLINE 1950 to present and EMBASE 1988 to present. Search terms used included: juvenile arthritis, juvenile idiopathic arthritis, juvenile rheumatoid arthritis, oligoarticular, pauciarticular, and uveitis. References 1. Petty RE, Southwood TR, Baum J, Bhettay E, Glass DN, Manners P, et al. Revision of the proposed classification criteria for juvenile idiopathic arthritis: Durban, J Rheumatol 1998;25: Petty RE, Southwood TR, Manners P, Baum J, Glass DN, Goldenberg J, et al. International League of Associations for Rhematology classification of juvenile idiopathic arthritis. Second revision. Edmonton, J Rheumatol 2004;31: Jabs DA, Nussenblatt RB, Rosenbaum JT. Standardization of uveitis nomenclature for reporting clinical data: Results of the First International Workshop. Am J Ophthalmol 2005;140: Dana MR, Merayo-Lloves J, Schaumberg DA, Foster CS. Visual outcomes prognosticators in juvenile rheumatoid arthritis-associated uveitis. Ophthalmology 1997;104: Smiley WK. Iridocyclitis in Still s disease: Prognosis and steroid treatment. Trans Ophthalmol U K 1965;85: Cassidy JT, Sullivan DB, Petty RE. Clinical patterns of chronic iridocyclitis in children with juvenile rheumatoid arthritis. Arthritis Rheum 1977;20: Chylack LT Jr., Bienfang DC, Bellows AR, Stillman JS. Ocular manifestations of juvenile rheumatoid arthritis. Am J Ophthalmol 1975;79: Wolf MD, Lichter PR, Ragsdale CG. Prognostic factors in the uveitis of juvenile rheumatoid arthritis. Ophthalmology 1987;94: Kanski JJ. Juvenile arthritis and uveitis. Surv Ophthalmol 1990;34: Cabral DA, Petty RE, Malleson PN, Ensworth S, McCormick AQ, Shroeder ML. Visual prognosis in children with chronic anterior uveitis and arthritis. J Rheumatol 1994;21: Chalom EC, Goldsmith DP, Koehler MA, Bittar B, Rose CD, Ostrov BE, et al. Prevalence and outcome of uveitis in a regional cohort of patients with juvenile rheumatoid arthritis. J Rheumatol 1997;24: Kotaniemi K, Kautiainen H, Karma A, Aho K. Occurrence of uveitis in recently diagnosed juvenile chronic arthritis: A prospective study. Ophthalmology 2001;108: Sim KT, Venning HE, Barrett S, Gregson RM, Amoaku WM. Extended oligoarthritis and other risk factors for developing JIAassociated uveitis under ILAR classification and its implication for current screening guideline. Ocul Immunol Inflamm 2006;14: Kump LI, Castaneda RA, Androudi SN, Reed GF, Foster CS. Visual outcomes in children with juvenile idiopathic arthritis-associated uveitis. Ophthalmology 2006;113: Oren B, Sehgal A, Simon JW, Lee J, Blocker RJ, Biglan AW, et al. The prevalence of uveitis in juvenile rheumatoid arthritis. J AAPOS 2001; 5:2-4.

7 Volume 12 Number 6 / December 2008 Sabri et al Chia A, Franco Lee V, Graham EM, Edelsten C. Factors related to severe uveitis at diagnosis in children with juvenile idiopathic arthritis in a screening program. Am J Ophthalmol 2003;135: Saurenmann RK, Levin AV, Feldman BM, Rose JB, Laxer RM, Schneider R, Silverman ED. Prevalence, risk factors and outcome of uveitis in juvenile idiopathic arthritis a long-term follow up study. Arthritis Rheum 2007;56: Edelsten C, Lee V, Bentley CR, et al. An evaluation of baseline risk factors predicting severity in juvenile idiopathic arthritis associated uveitis and other chronic anterior uveitis in early childhood. Br J Ophthalmol 2002;86: Paroli MP, Speranza S, Marino M, Pirraglia MP, Pivetti-Pezzi P. Prognosis of juvenile rheumatoid arthritis-associated uveitis. Eur J Ophthalmol 2003;13: Foster CS, Havrlikova K, Baltatzis S, Christen WG, Merayo- Lloves J. Secondary glaucoma in patients with juvenile rheumatoid arthritis associated iridocyclitis. Acta Ophthalmol Scand 2000;78: Saurenmann RK, Levin AV, Rose JB, Parker S, Rabinovitch T, Tyrrell PN, Feldman BM, Laxer RM, Schneider R, Silverman ED. Tumour necrosis factor alpha inhibitors in the treatment of childhood uveitis. Rheumatology 2006;45: An Eye on the Arts The Arts on the Eye She confirmed what I already knew about denial. Vision loss is like falling from the throne. When people first suffer from macular degeneration or other impairments, the reaction is usually disbelief... I can see. they will insist. They put themselves in difficult situations, like taking the subway alone. Or they displace problems on something else, like claiming that a sign is in the wrong place or someone has moved a piece of equipment. Sooner or later, when this does not work, terror sets in. People become depressed, although they don t always realize it. Some withdraw and merely vegetate. Most become deeply angry at life, at people with full vision, at anyone trying to help. by Henry Grunwald (from Twilight: Losing Sight and Gaining Insight [Memoirs of the Editor of Time Inc., who had macular degereration])