Science from our Sisters

Model Created for Predicting the Development of RA from UA

A point-system model that predicts progression of undifferentiated arthritis (UA) to rheumatoid arthritis (RA) using easily measured clinical variables was published in Arthritis & Rheumatism (2007;56(2):433-440).

When patients initially visit a physician because of joint pain, their arthritis is often in the early stages of disease and cannot yet be classified as RA. Of all patients with UA, in fact, only one-third progress to RA.

Early methotrexate treatment can slow progression of RA and prevent joint damage in this patient group. But another 40% to 50% of UA patients eventually go into spontaneous remission, and thus don’t need to be treated with methotrexate, which has side effects including liver toxicity.

“This is a dilemma if you’re trying to achieve the best individual treatment decisions,” says lead author Annette H. M. van der Helm-van Mil, MD, PhD, of Leiden University Medical Center in the Netherlands. “You don’t want to over-treat, but the patients who go on to RA will benefit from early treatment.”

The new model is based on data from 570 patients diagnosed with UA between 1993 and 2005. At the initial examination, patients were asked about pain symptoms and severity of any morning stiffness. Blood samples and hand and foot x-rays were taken. After one year, patients were tested again to see if they had progressed to RA.

The researchers performed a multivariate analysis to see how well each of nine clinical variables—gender, age, symptom localization, morning stiffness, tender joint count, swollen joint count, C-reactive protein level, rheumatoid factor positivity, and presence of anti–cyclic citrullinated peptide (anti-CCP) antibodies—predicted RA progression.

The researchers then created a scale ranging from 0 to 14, with different point values assigned for each variable according to how significant it was to predicting RA progression. The most heavily weighted variables are gender, presence of anti-CCP antibodies, and morning stiffness severity.

The scale had strong predictive ability: 91% of patients who scored a 6 or less on the scale did not progress to RA; 84% of patients that scored an 8 or greater did progress to RA. Clinicians could thus be relatively confident in prescribing medication to a patient scoring above an 8.

But the biggest advantage of such a scale, Dr. van der Helm-van Mil says, is that it “will make it easier for patients to be part of the discussion when the rheumatologist is making a treatment decision.” The researchers are now validating their model in various patient groups throughout Europe.