A Family History of Psoriasis in a First-degree Relative in Children with JIA: to Include or Exclude?

DISCUSSION

Our study confirms the observation that an FRP is the most common reason for classifying children as having UJIA. Disregarding this exclusion criterion allowed 67% of patients with UJIA to be classified in defined categories of JIA. Our study did not confirm our previous findings suggesting that classification of JPsA would not be affected by disregarding FRP as an inclusion criterion. In 11 of 77 (14%) of our patients with JPsA, an FRP was critical to that classification. Removing FRP as an inclusion criterion would result in some children currently classified as JPsA being reclassified as oligoarticular or polyarticular RF-negative JIA, although many had dactylitis, an uncommon feature for these 2 groups. The median age of onset (2.6 yrs) of the patients reclassified as polyarticular RF-negative is also younger than usual for that JIA category. This suggests there is value in preserving the inclusion criterion of an FRP in classifying JPsA⁴.

Skepticism remains as to whether an FH of psoriasis, unless diagnosed by a medical professional, is reliably reported, and has been a point of debate^4,10,11. Removing FRP as an exclusion criterion may allow for more appropriate classification of children who may otherwise have a typical course of oligo-, poly-, or systemic JIA¹².

Removing FRP as an inclusion criterion for JPsA posits whether arthritis combined with dactylitis or nail changes is significant enough for classification as JPsA¹³, as in the CASPAR criteria^5,14. Three of the 85 patients with UJIA met CASPAR criteria for PsA: 1 ERA patient with an FRP and 2 patients who fulfilled criteria for ERA and JPsA. The predictive value of dactylitis and nail pits for future development of full-blown JPsA is unknown, although in younger children these may be present before frank skin disease develops¹⁵. In the CASPAR criteria, they have equal weight as supporting features of PsA⁵. Psoriasis can precede or follow the onset of arthritis within 15 years¹⁶. It is also possible that the skin disease may be absent or delayed when being treated with antirheumatic drugs.

If one removes FRP as an exclusion criterion and maintains it as an inclusion criterion in JPsA, there would still be difficult cases to classify. For example, a child with 2 swollen joints, an FRP, and nail pits would fulfill criteria for both oligoJIA and JPsA and thus would be unclassified. A child with 2 swollen joints, dactylitis, and nail pits but no FRP would also be unclassified. Using a hierarchical system of classification as per Manners, et al¹¹ may help avoid these situations.

Removing FRP as an exclusion criterion may also result in some children initially categorized in a defined category of JIA being later reclassified as JPsA if they develop psoriasis. Evolution of disease is a reality in rheumatology. Nordal, et al⁷ showed that 10.8% of patients with JIA in a Nordic cohort changed ILAR categories when followed for at least 7 years after disease onset. The reasons for change were emerging clinical findings such as onset of enthesitis, sacroiliitis, acute uveitis, or psoriasis, or new information about these conditions in a first-degree relative of the patient. The percentage of children with JPsA increased from 1.4% to 3.2%⁷.

In our current study, 17 children were labeled as having UJIA because of a single positive RF test. ILAR criteria require 2 positive RF tests performed at least 3 months apart as an exclusion criterion in all subtypes of JIA with the exception of polyarticular RF-positive disease, where it is a major inclusion criterion¹. Without a second confirmatory RF test, it is impossible to know whether the test was falsely positive, as can occur in cases of concurrent or preceding infection¹⁷. In contrast, a single negative RF test as per ILAR criteria is interpreted as a true negative.

From their inception, it was anticipated that ILAR criteria would be modified based on new evidence. Proposed revisions include algorithmic exercises¹¹, categorization based on ANA status¹⁸, and age at disease onset¹⁹. Ideally, a revised classification should be shown to be substantially improved before it is adopted and should correlate with disease course, response to treatment, and biological homogeneity, e.g., cytokine profiles and genetic signatures²⁰. Our study suggests that eliminating FRP as an exclusion criterion but keeping it as an inclusion criterion in JPsA will reduce the number of unclassified subjects and simplify use of the ILAR criteria. Whether this would improve the overall classification remains unclear.