THU0332 Predictors of poor growth in children with early juvenile idiopathic arthritis: Results from the childhood arthritis prospective study (CAPS)

Background Juvenile Idiopathic Arthritis (JIA) has been reported to be associated with generalised growth retardation in active disease. Objectives Our aim was to describe growth patterns at two years following first presentation to paediatric rheumatology in an established inception cohort of children with JIA. Methods The Childhood Arthritis Prospective Study (CAPS), established in 2001, is an incident cohort of children with JIA. Data collected at first presentation and then annually include demographics, height, weight, disease duration, active and restricted joint counts, physician global assessment (PGA), parental general score of well-being (PGE), Childhood Health Assessment Questionnaire (CHAQ), pain score (100mm visual analogue score), ILAR classification and medication including Disease Modifying Anti-Rheumatic Drugs (DMARDS). Heights at each visit were standardised using WHO data to produce z-scores. Short stature was defined as height z-score≤-2. Baseline factors associated with z-score at two years were modelled using multivariate linear regression. The model was adjusted additionally for medical treatment, defined as oral/IV steroids ever (yes/no), methotrexate ever (yes/no) and biologics ever (yes/no). Results A total of 445 children were included (66% female, median age at baseline 7.5 yrs, median disease duration at presentation 5.7 months). Fifty six percent received DMARDs, 40.5% oral/IV steroids and 11% biologics during first 2 years after presentation. Mean z-score at presentation was -0.003 (95% CI -0.75, 0.78) but dropped to -0.38 (95% CI -1.11, 0.41) by two years. The lowest height z-scores at 2 years were seen in those children with systemic arthritis (N=15, median at baseline: 0.04; median at 2 years: -0.5), RF(–) polyarthritis (N=14, median at baseline: 0.28; median at 2 years: -0.6) and psoriatic arthritis (N=28, median at baseline: -0.2; median at 2 years: -0.8). Twenty six children (5.8%) had short stature at baseline increasing to 34 (7.7%) at year 2. Longer disease duration at presentation was independently associated with lower z-scores at 2 years (Coef: -0.011 per month; 95% CI: -0.019, -0.002, p=0.014) Higher baseline CHAQ was also found to be independently predictive of lower z-scores at 2 years (Coef: -0.33 per unit CHAQ; 95% CI: -0.60, -0.063, p=0.016). Baseline active joints count, PGA, PGE and pain score were not predictive of height at 2 years (Table 1). Conclusions The majority of children presenting with JIA since