Outcome in Juvenile Idiopathic Arthritis

A Population-Based Study From Sweden

Elisabet Berthold; Bengt Månsson; Robin Kahn


Arthritis Res Ther. 2019;21(218) 

In This Article


A total of 489 individuals with the ICD codes for JIA were found. Of these patients, 307 were confirmed to have JIA after a review of the charts. However, 56 patients were excluded, as their JIA diagnosis was later than the 16th birthday or their diagnosis was confirmed outside the study area. A total of 251 patients are thus remaining in the study cohort. Other reasons for exclusion were medical records not found (n = 5), other rheumatic diseases (n = 19), and non-rheumatic condition misdiagnosed as JIA (n = 158) (flow chart of the case collection procedure is enclosed as Additional file 1: Figure S1).

Diagnosis Distribution

Oligoarthritis is the largest subgroup (44.7%): persistent oligoarthritis 33.5% and extended oligoarthritis 11.2%, followed by uJIA (16.3%), RF− (13.9%), ERA (8.8%), RF+ (6.8%), JPsA (6.8%), and sJIA is the smallest subgroup (2.8%) (Table 1).

Incidence Rate

The mean annual incidence rate for JIA was estimated to be 12.8 (95% confidence interval 11.3–14.5) per 100,000 children < 16 years. In females, the mean annual incidence rate was 17.5 (15.0–20.4) per 100,000, and in males, the corresponding number was 8.3 (6.7–10.3) per 100,000.

When studying the age-specific annual incidence rates, the peak is at 2 years, 36/100,000. This peak is consistent in the female group, but the incidence peak among males is not until 12 years (Figure 1a).

Figure 1.

Mean annual incidence rate. a The bar chart shows the age-specific mean annual incidence rate divided by gender, presented per 100,000 children. The line shows the age-specific incidence rate of the total cohort. bh The line charts visualize the age-specific annual incidence rates per 100,000 children in the diagnostic subgroups. b Enthesitis-related arthritis (ERA). c Oligoarthritis. d RF-negative polyarthritis (RF−). e RF-positive polyarthritis (RF+). f Juvenile psoriatic arthritis (JPsA). g Systemic juvenile idiopathic arthritis (sJIA). h Undifferentiated juvenile idiopathic arthritis (uJIA)

We further investigated the incidence rate for the different subgroups. The oligoarticular subgroup has the highest incident rate in younger children, whereas the incidence rates for ERA and RF+ peak in the older age groups. The subgroups of uJIA and RF− have a more evenly distributed age-related incidence rate (Figure 1b).


In the total cohort, 2/3 of the children are female and it is only in the ERA group where males predominate. The median age at diagnosis in the cohort is 7.3 years; the highest median age of 12.7 is in the RF+ group and 10.7 years in the ERA group. The children are followed up for a median of 8 years. The median time between symptom debut and specialist diagnosis is 5 months; in the JPsA group, this duration is as long as 14 months.

Half of the children in the population are ANA-positive, and among them, 59.8% have oligoarthritis. RF is present in 10.8% of the population on at least one occasion; 63.0% of these patients have a polyarticular JIA; 14.7% are carriers of HLAB27; 48.6% of them have ERA (Table 1).

Pharmacological Treatment

The distribution of pharmacological treatment is presented in Table 2. The category "no treatment" refers to a year with no pharmacological treatment, i.e., a patient has discontinued all treatment during one calendar year. Only 43.3% of the patients met this criterion at some point during the follow-up. Almost all patients (98%) are at some point during their disease course prescribed non-steroid anti-inflammatory drugs (NSAID). Local, intra-articular steroid injections are also often used (78.9%) and are, except in the systemic group, relatively equally often used in the different subgroups. Methotrexate is the most common disease-modifying anti-rheumatic drug (DMARD) prescribed (60.6%). It is used by all children with RF+ disease and is a more common treatment option in all the polyarticular groups. Tumor necrosis factor alpha (TNFα) inhibitor is used as a treatment in 23.9% of the children, predominantly with polyarticular disease and in 83.3% in combination with methotrexate. Other DMARDs, conventional as well as biologic, have also been used as a treatment. However, all patients with other biological DMARDs have also tried at least one TNFα inhibitor. No cases of active tuberculosis were found after the initiation of biological DMARDs (a more detailed table with treatment options is enclosed as Additional file 2: Table S1).


In the entire cohort of children with JIA, consisting of all subgroups in the total follow-up period, 40.0% of the years were with inactive disease (defined as no arthritis or uveitis), 54.8% were active due to arthritis with or without uveitis, and 5.2% were active because of uveitis only. The median follow-up time was 8.0 years. In the subgroups, the percentages of inactive disease presents as follows: ERA 38.4%, oligoarthritis 42.5% (with extended oligoarthritis 33.3% and persistent oligoarthritis 46.5%), RF− 37.3%, RF+ 25.9%, JPsA 33.3%, sJIA 64.0%, and uJIA 43.5%. 28.8% of the inactive years were without treatment (percentage presented as gray bars) (Figure 2). One patient that was lost to follow-up was later found out to have died.

Figure 2.

Inactive disease. Inactive disease was defined as a year without arthritis or uveitis. The bars represent the years with inactive disease presented as the percentage of the total follow-up time (years) in every subgroup. The light gray areas represent the years with inactive disease without any pharmacological treatment, and the striped areas represent the years with inactive disease on medication

Uveitis was seen in 27 (10.8%) of the children, 8.0% had chronic uveitis, and 4.0% had acute uveitis (3 individuals have had both manifestations). Fourteen of the children have had uveitis in their debut year (10 chronic). The median debut age of chronic uveitis is 5.5 years (range 0–16 years). There are no cases of uveitis in the RF+, JPsA, or sJIA groups (Table 3). The risk of chronic uveitis is 10.0% at 12 years of follow-up using Kaplan-Meier survival analysis (Figure 3a).

Figure 3.

Chronic uveitis and joint corrective orthopedic surgery survival analysis. Survival curve according to Kaplan-Meier analysis. a First time chronic uveitis is present in 50% of the children the first year of disease but may occur throughout the entire follow-up time. Children with acute uveitis are not included in the analysis. b Joint corrective orthopedic surgery occurs throughout the entire follow-up period in JIA according to Kaplan-Meier analysis. At the end of the study period, 9.2% had been treated with joint corrective surgery

In total, 23 (9.2%) individuals have been treated with joint corrective orthopedic surgery, 8 of them with multiple procedures (3 with RF+ JIA, 2 with oligoarticular disease, 1 with RF− JIA, 1 with JPsA, and 1 with uJIA) (Table 3). However, only 11 individuals (4.4%) have undergone large orthopedic surgery (arthroplasty, osteotomies, or arthrodesis). The procedures were 17 synovectomies (5 with diagnostic purpose), 7 arthrodeses, 6 osteotomies, 4 medial knee epiphysiodeses, 1 arthrolysis, 1 arthroplasty (hip prosthesis), 1 volar tenosynovectomy, and 1 finger tendon transposition. The need for orthopedic surgery is the highest (23.5%) in the group with RF+ JIA. The risk for joint corrective surgery is 17.9% at 12 years of follow-up using Kaplan-Meier survival analysis (Figure 3b). The risk for a serious orthopedic procedure is 9.4% at 12 years of follow-up.