Predict rheumatoid arthritis conversion from undifferentiated arthritis with dynamic contrast-enhanced MRI and laboratory indexes.

Objectives: To investigate the clinical value of dynamic contrast-enhanced MRI (DCE-MRI) and laboratory indexes in predicting conversion from undifferentiated arthritis (UA) to rheumatoid arthritis (RA).

Methods: A total 81 DMARD-naive UA patients were studied. 37 cases were ultimately diagnosed as RA, 32 cases were diagnosed as other types of arthritis, and the remaining cases were still UA during the 1-year follow-up. The DCE-MRI and laboratory measures were fed into a logistic regression analysis.

Results: Wash-in rate and anti-cyclic citrullinated peptide (anti-CCP) antibody served as the final variables into the regression equation (p<0.05). The area under the ROC curve of wash-in rate was 0.966. With optimal cut-off point 29.84 s-1, wash-in rate achieved a sensitivity of 94.6% and specificity of 88.6% for predicting RA conversion from UA; anti-CCP antibody positive achieved a sensitivity of 37.8% and specificity of 90.9%. The combination of wash-in rate and anti-CCP antibody positive improved specificity (100%) but not sensitivity (27.3%).

Conclusions: The conversion from UA to RA is highly predictable. The wash-in rate of DCE-MRI can be used as an important biomarker to predict UA progression.