Paediatric uveitis is being recorded twice as often as a decade ago

Design

retrospective database incidence and prevalence study with Wilson score confidence intervals

Population

30,533 children aged 17 or under with non-infectious uveitis from 30,729,490 in the TriNetX US network, 2016 to 2025

Primary outcome

annual and cumulative incidence and prevalence, complications, systemic associations and treatment patterns

Effect

incidence 22 to 33 per 100,000 and prevalence 106 to 226 per 100,000 over the decade; glaucoma in 8.8%, blindness or low vision in 13.4% of posterior uveitis

A retrospective analysis of the TriNetX US Collaborative Network covered 30,729,490 children, of whom 30,533 (0.10%) had non-infectious uveitis recorded between 2016 and 2025. Median age was 11 years and 54.6% were male, which is worth noting against the usual teaching that juvenile idiopathic arthritis-associated uveitis is a girl's disease — it is, but the whole of paediatric uveitis is not.

Anterior uveitis dominated, 22,481 cases (73.6%). Annual incidence rose from 22 to 33 per 100,000 over the decade and recorded prevalence from 106 to 226 per 100,000 — a doubling. Glaucoma complicated 8.8% (95% CI 8.5 to 9.1) and cataract 6.9% (6.6 to 7.2). Posterior uveitis carried the highest rate of blindness or low vision at 13.4% (12.2 to 14.6). Prednisone (13.6%) and methotrexate (13.1%) were the commonest drugs, antimetabolites the commonest class (15.6%). Juvenile idiopathic arthritis was the most frequent systemic association at 10.4%, with tubulointerstitial nephritis and uveitis at 1.5%.

The numbers that change a clinic are the complication rates. Nearly one child in eleven develops glaucoma, and that is a screening obligation, not a possibility — intraocular pressure at every visit, in a child who is also on topical steroid, which raises the pressure independently. The concentration of vision loss in posterior disease argues for referral thresholds that differ by anatomical subtype rather than by symptom severity.

Whether uveitis is actually becoming more common cannot be answered here. A doubling of recorded prevalence over a decade in a database built from coding is at least as likely to reflect better detection, better coding and a growing network.

• Measure intraocular pressure at every visit in a child with uveitis, especially on topical steroid — glaucoma affected 8.8%
• Set referral urgency by anatomical subtype: posterior disease carried the highest rate of vision loss
• Screen for juvenile idiopathic arthritis and ask about renal symptoms; tubulointerstitial nephritis and uveitis was the second commonest association
• Document the Standardization of Uveitis Nomenclature subtype in the notes, because prognosis follows it
• Look for amblyopia in the younger child; it is the complication a pressure check will not find

The statistics, in plain English

With 30 million children in the denominator, the confidence intervals are narrow and the proportions are precise — but precision is not validity. These are coded diagnoses, not adjudicated ones, so a child coded with uveitis may not have had it, and a child with it may not be coded. The doubling of prevalence is the figure most vulnerable to that: coding practice and network composition both changed over the same decade, and neither has anything to do with disease biology. The complication proportions are more robust because they are internal comparisons within the coded cohort.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.