Predictors of severe radiographic progression in patients with early rheumatoid arthritis: A Prospective observational cohort study.

Aim: To identify predictors of severe radiographic progression in patients with early rheumatoid arthritis (ERA).

Methods: A total of 374 patients with ERA were selected from a Korean prospective cohort. Based on their annual Sharp/Van der Heijde modified score changes (ΔSHS/year), patients were classified into severe and no progression groups. Predictors of severe progression were evaluated using a multivariable logistic regression.

Results: After a mean follow-up duration of 4.2 years, the median (interquartile range) ΔSHS/year were 6.3 (4.4-10.2) and 0 (0-0) in the severe and no progression groups, respectively. Multivariable regression model revealed that Health Assessment Questionnaire (HAQ) score (odds ratio [OR] = 2.17), anticyclic citrullinated peptide antibody (OR = 3.44), body mass index (BMI; OR = 0.88), 6-month cumulative erythrocyte sedimentation rate (OR = 1.01) and baseline SHS (OR = 1.07) were independent predictors of severe progression. A model incorporating all five predictors satisfactorily predicted severe progression, with an area under the curve of 0.80. Baseline SHS was the predictor with the highest contribution to the predictive power of the final model (38%).

Conclusions: Our predictive model composed of five clinical predictors showed high discriminative ability between severe and no radiographic progression in patients with ERA. Among them, baseline SHS was the strongest predictor. Also, low BMI in Korean patients with ERA have a high risk of severe radiographic progression, as has previously been found for Caucasians.