Update in juvenile rheumatoid arthritis


Juvenile rheumatoid arthritis (JRA) is a relatively common chronic illness in children. The prevalence of JRA is about 1 per 1,000 children, and the annual incidence is about 1 per 10,000 children (1). JRA is defined by the presence of joint swelling or 2 of the following: joint tenderness, decreased range of motion (ROM), pain on ROM, or joint warmth for at least 6 weeks without another cause in children younger than 16 years of age (2).

We reviewed 5 recent, relevant articles regarding JRA. These articles help to clarify issues about the natural history of JRA, describe the potential important relationship between assessment of serial radiographs of clinically involved joints and measures of clinical outcomes, and add insight to the medical management of children with JRA.

Long-term outcome

Long-term outcome in patients with juvenile idiopathic arthritis. (Arthritis Rheum, 2002) (3).

In this descriptive case series of 2 populations of children with juvenile idiopathic arthritis (JIA) seen in Germany between 1978 and 1988, fewer than 10% of subjects were severely disabled or handicapped after more than 15 years of disease duration. Of the 260 eligible subjects (all older than 15 years of age by December 1999), 215 were studied by survey and by direct clinical examination. Assessment tools included the Health Assessment Questionnaire (HAQ), global assessments (of wellbeing, level of function and pain, etc), joint counts (including a 42-joint count), and the Steinbrocker functional classification. The American College of Rheumatology (ACR) criteria for clinical remission of rheumatoid arthritis (RA) were used to classify patients into complete remission categories (e.g., no active disease and no antirheumatic therapy) or partial remission (no active disease on antirheumatic therapy).

Median duration of symptoms was 4 months at initial evaluation and the median followup period was 16.5 years. Approximately 40% of the group had oligoarthritis, 14% had systemic-onset disease, and 33% had psoriatic, enthesitis-related, or other forms of arthritis. Median duration of disease was 4 months at initial evaluation, and the median duration of followup was 16.5 years. Approximately one-third of the oligoarticular-onset cases had polyarticular extension.

Approximately 15% of the patients developed uveitis after a median disease duration of 4 years. Half of the patients developed sequelae including reduced visual acuity and 2 patients (1% of total cohort) lost their eyesight. At last followup, 4 of these patients were still receiving treatment for uveitis. No deaths were reported in this cohort.

Three patients achieved a final height below the third percentile (2 patients with systemic onset arthritis, and 1 with polyarticular arthritis). Approximately 25% of the total cohort had limb length discrepancies averaging one centimeter, and 33% of these required shoe lifts. Micrognathia occurred in 22% of patients with polyarticular disease, in 13% of patients with systemic-onset disease, and in 8% of those with oligoarticular-onset disease.

Approximately half (45%) of the children had a surgical procedure sometime during the course of their disease, the majority of which were synovectomies. Only 5 patients underwent joint replacement surgery, after an average disease duration of 13 years.

At the last followup visit, 40% of the cohort had complete remission, and 5% had partial remission. Children with oligoarthritis had the best chance of complete or partial remission (59%), followed by those with systemic-onset disease (54%) and polyarthritis (41%). Approximately 40% of the cohort had swollen or tender joints at the final followup visit, and almost 60% had joints with limited ROM. Approximately 50% of the patients reported joint pain, and ∼33% reported morning stiffness at final followup visit. About 20% reported using a nonsteroidal antiinflammatory drug and about 25% reported taking a disease-modifying antirheumatic drug (DMARD) at their final visit.

Sixty-one percent of the cohort reported a HAQ score of zero, and 6.5% reported a HAQ score ≥1.0, which correlated well with the Steinbrocker functional classification and disease activity, with trends towards correlation with disease duration and polyarticular disease. Twenty-one patients were classified as Steinbrocker functional class III and 1 patient classified as class IV at last followup visit. There were no significant differences in educational level and employment/vocational status when compared with age-matched controls.

The data from this study can be applied to discussions with patients and families of children with a new diagnosis of JIA. Because JIA includes enthesitis-associated arthropathies, some of the relative percentages would be different for children with JRA. In this long-term study, meaningful clinical and functional outcomes were measured. Approximately 10% were classified as either Steinbrocker functional class III or IV, but no deaths occurred. Overall long-term outcomes of children with JIA from these 2 cohorts are generally favorable.

The early pattern of joint involvement predicts disease progression in children with oligoarticular (pauciarticular) juvenile rheumatoid arthritis. (Arthritis Rheum, 2002) (4).

In this descriptive case series of 205 children with pauciarticular JRA from 2 North American Centers, followed for at least 5 years, ∼40% had extension to >4 joints, and ∼50% of these patients had ≥10 joints involved. The medical records of 365 children diagnosed with oligoarticular JRA from 1965–1994 were reviewed. Two-hundred and five of these children met the ACR criteria for oligoarticular onset JRA, were followed up for at least 5 years, were not rheumatoid factor positive, and did not meet criteria for psoriatic or other seronegative spondylarthropathies during the first 6 months of their disease.

The median age at diagnosis was 4.9 years, ∼80% were female, and the median followup was 10.8 years. Approximately 40% had >4 joints involved after 6 months, and 18% had >10 joints involved. Nine percent of this cohort went on to develop a more clear-cut spondylarthropathy. The presence of symmetric joint involvement, ankle or wrist involvement, and an elevated erythrocyte sedimentation rate (ESR) significantly increased the likelihood of extension to at least 10 joints. These findings were also associated with a statistically significant increased likelihood of erosions, use of DMARDs, persistent disease activity, and other negative outcomes.

An important observation from this study is that almost 20% of children with rheumatoid factor negative JRA, who presented with oligoarticular disease, extended to at least 10 affected joints. Risk factors for this extension included wrist or ankle involvement, symmetric disease, and an elevated ESR. In addition, this group was more likely to have other adverse clinical and radiographic outcomes. Children who present with oligoarticular JRA, with these risk factors should be monitored closely for extension into polyarticular disease.

Prognosis of JRA

Prognostic factors for radiographic progression, radiographic damage, and disability in juvenile idiopathic arthritis. (Arthritis Rheum, 2003) (5).

In this prospective study of 94 children with various forms of JIA, early changes in hand/wrist radiographs were shown to correlate with long-term joint damage and physical disability.

All children seen in Pavia, Italy since 1986 who had polyarthritis with bilateral wrist involvement and had at least 3 sets of hand/wrist radiographs available for review by December 2001 were eligible for the study. Radiographs obtained at baseline, 1 year after diagnosis, and at final visit were required for this study. The children were seen every 6 months, and assessments included global assessments of disease activity (physician), global assessments of over-all wellbeing (parent), Childhood Health Assessment Questionnaire (C-HAQ), swollen joint counts, tender joint counts, and acute phase reactants. Radiographic outcomes were assessed by the Poznanski method (6) of relative carpal shortening.

Median age of the cohort was 4.6 years, and 60 of the 94 patients were male. Thirty-five patients had systemic-onset disease, 24 had extended oligoarthritis, and 30 had polyarticular presentation. The median duration of symptoms at diagnosis was 1.1 years, and the median duration of followup was 4.5 years. Almost all of the cohort received DMARDs, and 87 were treated with methotrexate.

The Poznanski scores declined (“progressed”) between the baseline and the 1-year hand/wrist radiographs, and between the 1-year and final hand/wrist radiographs in a statistically significant fashion. The degree of progression during the first year was predictive of the final C-HAQ score of disability and the final Poznanski score of long-term joint damage. Also, male sex correlated with these 2 adverse outcomes. Systemic-onset disease correlated with long-term joint damage, but not with the final C-HAQ score.

The data from this study suggests that measures of radiographic studies, such as radiographs of the wrist, may potentially be useful outcome measures, similar to their usefulness in RA. In adults with RA, radiographs of the small joints of the feet and hand/wrist are well-established tools for assessing important clinical outcomes. This study is the first to have longitudinal followup of a sizable cohort of children with polyarticular JIA and wrist involvement that were followed from diagnosis with sequential radiographs of the wrist. These radiographs were assessed with an established method (Poznanski), and radiographic changes over time correlated with important clinical outcomes. In our experience, a high percentage of children with polyarticular JRA have radiographic abnormalities of the hands/wrists at the time of diagnosis (7).

One of the limitations of this study is that the Poznanski method of assessing radiographic damage can only be applied when the carpal physes are open. In a younger cohort of children with polyarticular JIA such as in this study, this method would be widely applicable. However, this method would not be useful for older children and in those whose carpal growth plates have closed. An unusual aspect of this cohort is the high frequency of boys with JIA, which may influence the generalizability of these findings.

Data from this study and from others (8–10) should be important in developing methods to accurately describe clinically meaningful data for measuring radiographic outcomes in children with chronic arthritis.

Management of JRA

Efficacy of etanercept for the treatment of juvenile idiopathic arthritis according to the onset type. (Arthritis Rheum, 2003) (11).

In this prospective open-label study of 61 children with various forms of active JIA that was poorly responsive to methotrexate therapy, meaningful clinical improvement was seen in 73% of subjects by 3 months, but in only 39% after 12 months of etanercept therapy. Twelve of these subjects had stopped the etanercept secondary to adverse reactions, and children with systemic-onset disease were less likely to respond to etanercept than the other onset types.

French children with at least 5 swollen joints and at least 3 painful and restricted joints who were intolerant of or did not respond to a weekly dosage ≥0.4 mg/kg of methotrexate were eligible for this multicenter, open-label trial. The clinical assessments included swollen and tender joint counts, C-HAQ, and global assessments of disease activity (physician, parent) and pain (patient). Clinical improvement was as defined by Giannini et al (12).

Median age of the cohort was 12.2 years, with ∼80% being female. Twenty-two had systemic-onset disease, 24 had extended oligoarthritis, and 13 had polyarticular presentation. The median duration of disease at baseline was 6.6 years. All had previously received methotrexate and 80% were receiving systemic corticosteroids. At baseline, the mean C-HAQ score was 1.81 and the mean number of active joints was 16. The median duration of followup was 13 months.

A 30% improvement at 3 months was seen in 73% of these children, but by 12 months, only 39% were still improved. A 70% improvement at 3 months was noted in 38% of the children; by 12 months, this number declined to 26%. Children with systemic-onset JIA were much less likely to achieve a 30% improvement than either those with extended oligoarthritis or polyarticular arthritis (P > 0.003).

Data from this trial of etanercept in children with active JIA despite methotrexate therapy are different than data from an earlier trial conducted by Lovell et al (13). In that study, a response rate of 74% was observed at the end of the open-label portion of the study (3 months). Only one subject withdrew from this study secondary to adverse events.

There seem to be 2 major differences between these 2 trials of etanercept in children with chronic arthritis. One of these is that the total length of followup in the Quartier et al (11) study was 12 months compared with 7 months for the Lovell study. The second major difference is that the Quartier study had a much larger proportion of children with extended oligoarticular JIA in their cohort. The duration of disease at study entry, entry criteria, and measures of outcome were otherwise similar, although the Lovell study utilized a 3-month open label enrollment phase, and those who responded were then randomized to receive either etanercept or placebo in the control phase. In the Lovell study, all subjects who were receiving methotrexate prior to the study had it discontinued; in the Quartier study, 14 patients either continued receiving methotrexate or had it restarted.

In reviewing data from these studies, it appears that etanercept is effective, at least initially, in children with polyarticular forms of JIA, but the duration of its effectiveness is not clear. It may be that the polyarticular-onset type responds better than the other types. This would seem intuitive since polyarticular-onset JRA shares many features of rheumatoid arthritis.

Growth hormone improves height in patients with juvenile idiopathic arthritis: 4-year data of a controlled study. (J Pediatr, 2003) (14).

In this controlled trial of human growth hormone (hGH) in 38 prepubertal children receiving systemic corticosteroids for severe systemic onset or polyarticular JIA and growth retardation, the mean height was improved over the 4-year study period.

Children who had either systemic-onset JIA or polyarticular JIA and were either greater than 2 SD below the mean for their age in height or had a growth velocity of less than the 25th percentile for the preceding year were eligible for the study. All were prepubertal at the start of this study, all were receiving stable doses of systemic corticosteroids, and all were screened for primary growth hormone deficiency. Thirty-eight children participated, and were randomized to either 1 of 2 doses of weekly hGH or standard care (e.g., no placebo arm). Mean age was 10 years, mean age at diagnosis was 3.8 years, and mean duration of daily corticosteroid use was 3.9 years. Success was defined as either an increase in growth velocity of greater than 2 SDs compared with baseline or an increase of height of at least 1 SD by the end of the study. The treatment groups received either 0.2 mg/kg (6 patients) or 0.33 mg/kg (12 patients) per week of hGH.

There were no statistically significant differences in the baseline characteristics of either treatment group. Eight of the 18 children in the hGH groups achieved an increase of at least 1 SD in height as compared with 2 of 20 children in the non-hGH treated group. Further height loss occurred in 15 of 20 control subjects, but only 1 in the treatment arms. There was no significant difference in responses between the groups with the differing doses of hGH. Factors influencing the rate of growth velocity included mean ESR, C-reactive protein, and prednisolone doses (averages taken from all of the study visits).

The data from this study suggest that children with systemic-onset or polyarticular JIA who are receiving systemic corticosteroids may be able to improve deficiencies in their stature with the use of hGH. The degree to which stature can be improved is less certain and may be influenced by clinical factors. Because very short stature may impart potential devastating consequences, further study in this area is warranted.