Rheumatoid factor is the major predictor of increasing severity of radiographic erosions in rheumatoid arthritis - Results from the Norfolk Arthritis Register study, a large inception cohort

Objective. To identify the relative contributions of clinical and laboratory variables, determined at baseline, in predicting the deterioration of radiographic damage 5 years after presentation in patients with inflammatory polyarthritis. Methods. Data from 439 subjects who sought primary care for inflammatory polyarthritis were analyzed. All subjects had paired radiographs, of which the first was obtained within 24 months of presentation and the second at 5 years after presentation. The contribution of baseline clinical and laboratory variables in predicting the degree of radiologic severity as judged by the Larsen score was assessed at both time points. Additionally, the role of these factors in predicting change after adjustment for baseline severity was also measured. Results. By 5 years, 49% of subjects had evidence of erosions. The median Larsen score on the first film was 2 (interquartile range [IQR] 0-10) and the median score on the followup film was 7 (IQR 1-25). These corresponded to a median deterioration of 3 (IQR 0-14) in all subjects, whereas those subjects with evidence of erosions at first film showed a median deterioration of 15 (IQR 6-29) on followup. The rheumatoid factor (RF) status, C-reactive protein levels, the presence of nodules, and number of swollen joints at baseline were all predictive of radiographic severity at first film. Not surprisingly, the baseline radiographic score was a predictor of severity of deterioration. However, after adjusting for baseline severity, a high titer of RF (>1:160) was also an independent predictor of deterioration over 5 years: individuals with an initial RF at that level had a progression in their Larsen score that was 2.3 times (95% confidence interval 1.7-3.2) higher than that in the RF-negative individuals. Apart from this, only age had an independent effect, after adjusting for baseline severity, in predicting increasing radiographic joint damage. Conclusion. High-titer RF is an important variable in predicting continuing severity of radiographic damage during the first 5 years after presentation with inflammatory polyarthritis.