STOCKHOLM — The search continues for sensitive and specific classification criteria for psoriatic arthritis.
At an international conference on psoriasis and psoriatic arthritis, the latest results were presented from the classification criteria for psoriatic arthritis (CASPAR) group, an international team of leading psoriatic arthritis researchers. The group has initiated a prospective study to evaluate existing diagnostic criteria as well as to derive new, more accurate criteria, said Philip S. Helliwell, Ph.D., of the rheumatology and rehabilitation research unit at the University of Leeds (England). The group's efforts are modeled on those of the OMERACT (Outcome Measures in Rheumatology) core set for rheumatoid arthritis.
Although clinical and radiologic evidence supports psoriatic arthritis as a separate disease, there's no consensus on diagnostic criteria. Such standardization would enhance research efforts by making patient comparisons easier. Although there are several sets of classification criteria for diagnosing psoriatic arthritis, only one was derived statistically from patient data, Dr. Helliwell said.
A diagnosis according to CASPAR criteria requires established inflammatory articular disease and a score of at least 3 points from the following features: current psoriasis (2 points), history of psoriasis but no evidence of psoriasis (1 point), family history of psoriasis but no evidence of psoriasis (1 point), dactylitis (1 point), juxtaarticular new bone formation (1 point), negative rheumatoid factor (1 point), and nail dystrophy (1 point). These criteria were specific (0.99) and fairly specific (0.91) for the diagnosis of psoriatic arthritis.
The criteria were derived using data collected prospectively from 588 patients with psoriatic arthritis and from 536 controls with other inflammatory arthritis diagnoses at 30 clinics in 13 countries. Of the controls, 71% had rheumatoid arthritis, 14% had ankylosing spondylitis, 7% had undifferentiated arthritis, 3% had connective tissue disorders, and 5% had other diseases.
The researchers collected data on more than 100 clinical and historical features. They also performed x-rays of the spine, hands, and feet. Samples were analyzed for rheumatoid factor, human leukocyte antigen, and anti-cyclic citrullinated peptide antibody.
For the first iteration of the criteria, the researchers performed a classification and regression tree analysis of existing criteria. The presence of two findings—a history of psoriasis and current psoriasis—was 97% sensitive and 96% specific. “It's very hard to beat that,” Dr. Helliwell noted.
By multivariate logistical regression analysis, the top predictive features were negative rheumatoid arthritis factor, current dactylitis, a history of dactylitis, and a history of psoriasis. The results of those two analyses were combined to produce the CASPAR criteria.
Until now, the diagnosis of psoriatic arthritis has been widely based on the Moll-Wright criteria developed in 1973 (Semin. Arthritis Rheum. 1973;3:55–78). These criteria require an inflammatory arthritis (peripheral arthritis and/or sacroiliitis or spondylitis), the presence of psoriasis, and the absence of serologic tests for rheumatoid factor.
The Moll-Wright criteria are considered simple and sensitive; however, they are not very specific, suggesting that some seronegative rheumatoid arthritis patients with coincidental psoriasis are mistakenly classified with psoriatic arthritis, Dr. Helliwell said.
“Clearly we need criteria to help us to distinguish this group that may be confounding because of the seronegative rheumatoid factor and coincidental psoriasis,” he said. Another group that is hard to diagnose comprises those who meet all other criteria for psoriatic arthritis but who have not yet developed psoriasis.