Flares in rheumatoid arthritis: do patient-reported swollen and tender joints match clinical and ultrasonography findings?

Objectives. To investigate how patient-reported flares in RA are related to clinical joint examination and inflammation detected by US. Methods. Eighty RA patients with DAS28-CRP <3.2 and no swollen joints at baseline were followed for 1 year. In case of patient-reported hand flare with swollen and tender joints (SJ and TJ, respectively), patients underwent clinical examination for SJ/TJ and US of bilateral wrists, MCP and PIP 1st-5th, six extensor tendon compartments and wrist flexor tendons for synovitis/tenosynovitis. Percentage agreement and kappa were calculated between patient-reported SJ and TJ, clinical examination for SJ/TJ and US findings indicative of inflammation. With US as reference, sensitivity, specificity, positive/negative predictive value and accuracy of patient-reported and clinically examined joints were determined. Results. Hand flare was reported by 36% (29/80) of patients. At time of flare, all clinical and ultrasonographic measures of disease activity deteriorated compared with baseline. Agreement between patient-reported SJ/TJ, clinically examined SJ/TJ and US was slight (kappa = 0.02-0.20). Patients and clinicians agreed in 79-93% of joints, more frequently on SJ than TJ. With US as reference, specificities were 86-100% and 88-100%, and sensitivities 12-34% and 4-32% for patient-reported SJ/TJ and clinically examined SJ/TJ, respectively. Conclusion. Over 12 months of follow-up, hand flare was reported by every third RA patient. Self-reported flares were associated with increased disease activity as determined by clinical examination and US. Patient-reported joint assessment may aid in capturing flares between routine clinical visits.