There is an identified disparity between the efficacy of tumour necrosis factor (TNF) inhibition in the treatment of uveitis in paediatric onset inflammatory arthritis and that of joint disease.^1–,⁵

We retrospectively reviewed the case notes of six patients with aggressive, refractory joint and ocular paediatric-onset disease treated with infliximab in a multidisciplinary clinic. Our report uses the Standardised Uveitis Nomenclature (SUN) grading system⁶ for uveitis and steroid dose as an outcome measure. All patients received weekly infliximab infusions at 0, 2, 6 and 8 weeks.

Patients were maintained on low dose immunosuppression with methotrexate while receiving treatment. Five of six patients had previously been treated with another anti-TNF agent (three with adalumimab and two with etancercept).

At the time of commencing treatment, patients were aged 8–18 (median 14) years. Six patients were treated with infliximab. Two patients were treated for 6 months, two for 9 months, one for 12 months and 1 for 15 months, with four of six requiring a dose increase to 6 mg/kg to obtain adequate control. Infliximab was stopped at 15 months in one patient owing to a satisfactory response and at 1 year owing to treatment failure.

Drug induced remission on infliximab occurred in three (50%) patients, with improvement of ocular inflammation in two other patients; complete resolution of joint involvement occurred in five of six patients. In this cohort, patients were able to reduce steroid requirement from an average monthly dose of 250 mg in the year before infliximab treatment and 120 mg per month while receiving infliximab. There was a reduction in the daily oral prednisolone dose to 5 mg in three patients as a result of treatment.

Finally, biological therapy was associated with gain in vision in four of six patients, where there was at least a halving of the visual angle in one eye (approximating to a three-line improvement in the Snellen acuity). Use of infliximab also suppressed inflammatory activity to permit intraocular surgery in three patients without the need for high dose steroids.

While receiving infliximab, one child developed new psoriasis, and in others psoriasis failed to improve. Infliximab was well tolerated with no serious adverse event.

Our data confirm, in part, other reports of successful anti-TNF treatment suppressing joint disease more effectively than uveitis⁷; we, in addition, observed that infliximab was well tolerated and successfully suppressed ocular inflammation.⁸ Of note, we could not prove that failure of a previous biological agent predicts failure with infliximab.

Outside a clinical trial, the clinical setting can make outcomes difficult to interpret. For example, patient D had worsening joint and ocular inflammation after 1 year of treatment with infliximab. This patient had stopped methotrexate of her own accord for approximately 6 weeks. Whether this is treatment failure of infliximab is arguable, given that there had been good control of disease until methotrexate was stopped.

The use of a threshold steroid dose—for example, 5 mg/day⁶—has limitations in clinical practice for children. A number of patients already had osteoporosis or delayed growth where further oral steroid treatment was relatively contraindicated or weight gain undesirable. We observed a reduction in the average monthly dose of prednisolone, and additional intravenous or oral high dose steroid treatment was not required. This would support a genuine steroid-sparing role of infliximab therapy.