OBJECTIVE: Previous studies have suggested a link between inflammation and central sensitization of pain in patients with RA. We conducted a prospective cohort study to determine whether US Doppler (USD), temporal summation (TS) of pain-assessed at baseline-and the potential interaction between them could predict the treatment response in RA.

METHODS: Patients were recruited from Departments of Rheumatology in the Copenhagen area and from private clinics. RA patients, who were scheduled to initiate therapy with either a conventional synthetic DMARD (csDMARD; including DMARD-naïve patients) or a biologic DMARD (bDMARD) agent (including bDMARD switch), were included and examined before the start of treatment and after 4 months. During the 4 months, patients received routine care from their treating rheumatologist.

RESULTS: Multiple regression analysis was conducted, with change in DAS28 as the dependent variable. In unadjusted models, baseline USD score was significantly associated with DAS28 (β = 0.06; 95% CI: 0.02, 0.09; P < 0.001), whereas TS and their interaction were not. After adjusting for age, sex, disease duration, baseline DAS28 and whether patients initiated a csDMARD or a bDMARD, the USD score was still significantly associated with change in DAS28 (β = 0.032; 95% CI: 0.001, 0.064; P = 0.046), whereas TS remained insignificant.

CONCLUSION: USD assessed at baseline is valuable as a prognostic factor for change in disease activity in 'real-life' RA patients. We did not find evidence to suggest that TS or the interaction between USD and TS was prognostically important for this group of patients.