116 patients were enrolled. We excluded 7 patients (4 that had suspected B.burgdorferi infection prior to arthritis, 1 boy with hondromalatiae patelae, 1 girl with parapatellar plica and 1 boy, who two years after arthritis developed ulcerous collitis. For further analysis the cohort included 109 patients, 69 (63.3%) were female. Study group characteristics are presented in Table 1. Triamcinolone hexacetonide was used in all patients. The average age at the time of first IAC was 8.0 years (1.2–18.3). All patients were white. Regarding the ILAR classification, 73% of patients had persistent oligoarticular JIA, 11% had extended oligoarticular JIA, 5% had psoriatic arthritis and 10% had enthesitis related arthritis.

Table 1 Characteristics of groups

After the first IAC 38.5% (42/109) did not require any further therapy and 14.7% (16/109) only required additional IAC. Systemic therapy was needed during the follow up in 45.9% (50/109) of patients, with 49 receiving methotrexate (MTX) and 1 receiving sulfasalazine (SSZ). Biologic therapy was introduced in 22.0% (24/109) of patients. At the last follow up visit 88.9% (97/109) had inactive disease. The patients were followed for the mean time 4.3 years (7 months – 8.2 years). Trajectory of treatment after first IAC is shown in Fig. 1. Regarding immunoserology 38.0% (41/108; one patient had missing data) were ANA positive, 14.0% (15/107; 2 patients had missing data) were HLA B27 positive, all patients were RF negative. In 76.1% (83/109) of children one joint was injected, in 22.0% (24/109) two joints and in 2 patients 3 joints were injected at the time of first IAC.

Fig. 1

The course of disease and therapy in study group and subgroups. A. The course of disease and therapy in the whole study group after first intra-articular corticosteroid injection. B. Disease course and therapy after first intra-articular corticosteroid injection. Patients are divided based on presence of ANA. C. Disease course and therapy after first intra-articular corticosteroid injection. Patients are divided based on the presence of HLA B27 antigen

The ILAR categories

The ILAR category was significantly associated with the required therapy (p = 0.005). Results are shown in Fig. 2. No further therapy was required in 45% of patients with persistant oligoarticular JIA, in no patient with extended oligoarticular JIA, in 40% of patients with psoriatic arthritis and 27.3% of patients with enthesitis related arthritis (ERA). Further local therapy was required in 20% of patients with persistant oligoarticular JIA, in no patient with extended oligoarticular JIA, in 20% of patients with psoriatic arthritis and in no patients with ERA. Systemic therapy was required in 35% of patients with persistant oligoarticular JIA, in all patients with extended oligoarticular JIA, in 40% of patients with psoriatic arthritis and in 72.3% patients with ERA. Biologic therapy was required in 15% of patients with persistant oligoarticular JIA, in 58% patients with extended oligoarticular JIA, in 20% of patients with psoriatic arthritis and in 36.4% patients with ERA.

Fig. 2

Therapy regarding the ILAR category. ILAR category was significantly associated with therapy requirement using two-way ANOVA (p = 0.005). Numbers of patients are shown in each cell

No further therapy required

In this group of 42 children the mean age was 9.2 (1.2–18) years, which was older than in the group that later required systemic therapy but the difference was not statistically significant (p > 0,05). 26.8% (11/41, one patient missing data) were ANA positive and 7.5% (3/40, 2 patients missing data) were HLA B27 positive. In 83.3% (35/42) only one joint was involved, in 14.2% (6/42) two joints were involved and one patient (2.3%, 1/42) had 3 joints injected. They were followed by mean of 3.3 years (7 months – 7.8 years). All patients were in remission off therapy at the last follow-up visit.

Only additional IAC required

In this group mean age was 6.9 years (1.5–13.5), 29.5% (5/17) were ANA positive, 17.6% (3/17) were HLA B27 positive. The number of affected joints was one in 70.6% (12/17) and 2 in 39.4% (5/17) of patients. In average, patients required additional IAC after 1.9 years (4 months – 6 years). They were followed by mean of 4.7 years (1.4–6.7). All patients were in remission off therapy at the last follow-up visit.

Disease modifying anti-rheumatic drugs (DMARDs)

Systemic therapy was required in 45.9% (50/109) of patients, with 49 receiving MTX and 1 receiving SSZ. We only included patients receiving MTX in further analyses and among those the average age at disease onset was 7.1 (1.5–18.3) years, 48.9% (24/49) were ANA positive, 18.4% (9/49) were HLA B27 positive. Only one joint was injected at first IAC in 73.5% (36/49), two joints in 24.4% (12/49) and three joints in 2% (1/49). Number of injected joints was not associated with requirement for systemic therapy with MTX (p = 0.5). Five patients (10.2%) developed uveitis in the course of the disease. MTX was introduced median 7.1 months (mean 14.2 months; 1 month − 6.1 years) after first IAC. The mean follow up was 5.1 years (1.4–8.1). During the follow-up 53.1% (26/49) had methotrexate as the only systemic therapy. At the last follow-up visit 73.5% (36/49) had inactive disease and 44% (16/36) of them were off therapy.

Biologic therapy

22% (24/109, 23 receiving MTX, 1 receiving SSZ) were eventually treated with biologic therapy. The mean age in this group was 7.7 years (1.8–18), 54.2% (13/24) were ANA positive, 25% (6/24) were HLA B27 positive. At the time of first IAC one joint was involved in 54.2% (13/24), two joints in 41.7% (10/24) and 3 joints in 4.2% (1/24) of patients. The number of injected joints was significantly associated to the requirement for biologic therapy using Fisher exact test (p = 0.006), presented in Fig. 3. They were followed for 5.2 years in average (1.4–8.1) and the mean time to biologic therapy was 2.2 years (3 months – 4.6 years). At the last follow up visit 75% (18/24) had inactive disease, 11.1% (2/18) of them were off therapy.

Fig. 3

Association of the number of injected joints at first intra-articular corticosteroid injection with the requirement for biologic therapy. Association is statistically significant using Fisher exact test

ANA positivity

Of 38% of patients who were ANA positive, 73% (30/41) required further therapy, 12.2% (5/41) required only additional IAC, 61% (25/41) required MTX and 31.7% (13/41) required additional biological therapy. Using the log-rank test (Mantel-Cox) of survival analysis ANA positivity was associated with the need for systemic therapy (P = 0.049, chi square 3.89). Shown in Figs. 1B and 4A.

Fig. 4

Survival functions from Kaplan Meier curves, showing difference in time to DMARD after first intra-articular corticosteroid injection in selected subgroups. A. Survival functions from Kaplan Meier curves, showing difference in time up to the DMARD after first intra-articular corticosteroid injection in ANA positive and ANA negative patients. B. Survival function from Kaplan Meier curves, showing difference in time up to the DMARD after first intra-articular corticosteroid injection in patients with and without HLA B27 antigen. C. Survival function from Kaplan Meier curves, showing difference in time up to the DMARD after first intra-articular corticosteroid injection in patients, that were older than 8 years at disease onset and were ANA and HLA B27 negative in comparison to all other patients

HLA B27 antigen

Of the 14% of patients who were HLA B27 positive, 80% (12/15) required further therapy, 73.3% required systemic therapy (10 patients MTX, 1 patient SSZ) and 40% (6/15) required additional biological therapy. Using the log-rank test (Mantel-Cox) of survival analysis, HLA B27 antigen was associated with the need for systemic therapy (P = 0.050, chi square 3.85). Shown in Figs. 1C and 4B.

In the subgroup of patients, that were ANA negative, HLA B27 negative and older than 8 years at the time of first IAC, only 25.8% (8/31) required systemic and only 9.6% (3/31) required biologic therapy. Using the log-rank test (Mantel-Cox) of survival analysis these patients had less chance to need systemic therapy in the course of their disease (p = 0.050, chi square 3.77). Shown in Fig. 4C.

Uveitis

6% (7/109) of patients developed uveitis in the course of disease, 57.1% (4/7) of these patients were ANA positive, 42.3% (3/7) of these patients were HLA B27 positive, one patient with uveitis was ANA and HLA B27 positive. Using the log-rank test (Mantel-Cox) of survival analysis ANA positivity was not significantly associated with uveitis (p = 0.38, Chi square 0.75). Uveitis was first noted average 25.4 months (2.5–78 months) after IAC. All but one patients with uveitis required systemic therapy.

Trauma

Four patients had minor trauma in the history before onset of arthritis. Imaging did not show any traumatic injury to the anatomical structures. Three of these did not require any further therapy and one required biologic therapy.

Orthopedic conditions

One patient had in the follow-up hondromalatia patelae and meniscus pathology and required orthopedic intervention. The patient had this pathology after being treated with MTX and biologic drug for extended oligoarthritis already for 3.5 years. One patient had non ossifying fibroma on a site distant from the site of arthritis.

Other accompanying disorders

Two patients had thyroid disease, one patient had vitiligo, there were no other autoimmune diseases. One patient was treated for Hodgkin lymphoma and finished treatment 2 years before arthritis onset. One patient had persistent thrombocytopenia, one patient had phenilketonuria. One patient had epilepsy, one patient was autistic. Two patients were treated for suspected infection with Borrelia burgdorferi prior to IAC. Both were treated with antibiotic before the negative serology excluded infection.