JUVENILE IDIOPATHIC ARTHRITIS: SYNOVIAL PREDICTORS OF OUTCOME

Background: Juvenile idiopathic arthritis runs an unpredictable course. Children
with oligoarticular disease may respond well to intra-articular steroids or have
recurrent episodes of synovitis. Children with polyarticular disease may respond to
Methotrexate or require steroids and/or biologics. Currently there are no reliable
predictors of outcome in early disease, which leads to sub-optimal treatment.
Methods: We are currently undertaking a five-year prospective study of children
with newly diagnosed and untreated JIA. All have knee involvement. At outset,
we obtain synovial biopsies (N ¼ 5) under ultrasound guidance. We then record
detailed clinical, functional, radiological and laboratory data every 3/12 for two
years. We are correlating outcomes at one and two years with synovial findings
recorded at outset. Here we report on those children whom to date we have
followed for one year.
Results: Of the first 30 children, 18 had oligoarticular disease. Their disease
activity score fell from a mean of 3.54 to 1.12 at one year. Ten had a single episode
of swelling of the index joint while eight had 2 or more episodes.
Twelve had polyarticular disease. All but two were improved at one year.
The mean ESR fell from 44 to 13, and CRP from 33 to 5. Haemoglobin rose from
a mean of 10.6 mg/dl to 12.5 mg/dl. The CHAQ improved in all but two.
We found significantly more synovial pathology in the poly compared with oligo
patients, with mean vessel score 6.9 vs. 2.6 (p < 0.05) and mean B-cell score
1.7 vs. 1.0 (p < 0.05).
On average, the oligo and polyarticular groups shared a similar degree of
synovial hyperplasia and a comparable macrophage distribution. The CD3þ cells
(mean 1.7 (0.8–2.7)) were predominately CD4þ (mean 1.7 (0.7–2.6)) and
distributed mainly within the sub-lining layer (SLL).
However, we observed significant differences between patients within clinical
subgroups.
Conclusions: Almost 50% of oligoarticular patients had recurrent knee swelling,
despite treatment. And while the majority of poly JIA responded to Methotrexate,
a significant minority still required steroids or were already on biologics at one year.
Infiltrates varied widely between patients; and we predict that synovial pathology,
particularly vascularity and B-cell infiltrates, will correlate with a poor outcome and
inadequate response to treatment.
We will therefore analyse whether synovial findings can predict outcome for
each patient.