The role of tender and swollen joints for the assessment of inflammation in PsA using ultrasound

Objective To evaluate tender joints (TJ) and swollen joints (SJ) for the assessment of ultrasound (US) defined inflammation in PsA. Methods Eighty-three PsA patients underwent clinical and US examinations at two scheduled study visits 12 months apart. Tenderness and swelling were assessed at 68 and 66 joints, respectively, and US examinations were conducted at all 68 joints. At patient level, associations with clinical composites and US scores were performed using correlations and by analysing patients with predominantly tender (pTender) or swollen joints (pSwollen). At joint level, a Power Doppler (PD) value >= 1 was defined as active synovitis. A generalized linear mixed model was created to assess the predictive value of TJ and SJ for active synovitis after 12 months. Results SJC showed better correlations with GS/PD scores (r = 0.37/0.47) than with TJC (PD: r = 0.33), while TJC correlated better with patient reported outcomes (PROMs) like patient global assessment (TJC: r = 0.57; SJC r = 0.39). Patients with pTender showed poorer results for PROMs and disease activity scores than patients with pSwollen, but not for laboratory or US markers of inflammation. Swollen joints showed active synovitis in 35% of cases, while only 16% of tender joints were active according to US. Swelling at baseline better predicted active synovitis at the same joint after 12 months [odds ratio (OR) 6.33, P < 0.001] as compared with tenderness (OR 3.58, P < 0.001). Conclusions SJ are more closely linked with US signs of inflammation as compared with TJ in PsA. Joint swelling is a better predictor for signs of US inflammation than tenderness after one year of follow-up.