Impact of patient-reported flares on radiographic progression and functional impairment in patients with rheumatoid arthritis: a cohort study based on the AMBRA trial.

ABSTRACT

OBJECTIVE: To investigate the impact of patient-reported flares on radiographic damage and disability in rheumatoid arthritis (RA). METHOD: Patients with low-active (Disease Activity Score based on 28-joint count with C-reactive protein < 3.2) RA were followed for 2 years. Based on annual questionnaires about incidence of flares, three 'flare phenotypes' were distinguished no flares (NF), transient flares (TF), and a mixed category reporting persistent joint complaints (PJC) in at least one year. Baseline and 2 year radiographs of hands and feet were evaluated according to the Sharp/van der Heijde method. Major outcomes were change from baseline in Total Sharp Score (ΔTSS) and functional impairment, expressed by the Health Assessment Questionnaire (ΔHAQ). Their association with flare phenotype was analysed by logistic regression. RESULTS: The study included 268 RA patients (70% female; 73% immunoglobulin M rheumatoid factor positive), with a median age (interquartile range) of 63 (55-70) years, and 7 (4-13) years' disease duration. Flares were recalled as NF (n = 77), TF (n = 141), and PJC (n = 50). ΔTSS > 0 was observed in 35%, 37%, and 46%, respectively (p = 0.42), but statistically significantly (p = 0.01) more patients progressed in the TF (10%) and PJC (14%) compared to NF (0%), based on the smallest detectable change (> 4.4 ΔTSS unit). ΔHAQ above the minimal clinically important difference (> 0.22) was seen in 13% (NF), 21% (TF), and 40% (PJC) (p = 0.0015), with PJC being associated with statistically significant impairment in function (odds ratio 4.47, 95% confidence interval 1.87-10.69) compared to NF. CONCLUSION: In RA patients with low disease activity, the incidence of radiographic progression and functional impairment was higher in patients with flares and persistent complaints, compared to those without flares.