Background A certain proportion of patients with rheumatoid arthritis fail to respond to multiple DMARD therapies. Agreed measures of treatment response include joint involvement, acute phase reactants and patient and physician global assessments. Scores such as DAS28, ACR or EULAR response cover only certain dimensions of the disease whereas response as assessed by the rheumatologist may comprise further aspects under surveillance.
Objectives To analyse which clinical parameters are associated with the rheumatologists' assessment of response.
Methods 662 patients with early rheumatoid arthritis (symptom duration 12±7 weeks) were followed for 2 years. Response to therapy was measured using standard scores as well as the rheumatologist's global assessment (excellent/good/modest/poor). Outcomes were controlled for sex, age, serology, joint counts, patient-reported outcomes and physician-reported chronic conditions at baseline as well as use of therapies at 2 years and were bi- and multivariately associated with the response variable using ANOVA and logistic regression models. Patients reported functional capacity (FFbH, range 0-100% of full function), impact on quality of life according to the Rheumatoid Arthritis Impact of Disease Score (RAID) and Profile of Fatigue and Discomfort (PROFAD).
Results 106 patients (16%) were assessed by the rheumatologists as having inadequate response (IR= moderate or poor). IR was associated with a significantly higher DAS28 at 2 years and a higher use of therapies. Patients with IR additionally reported significantly more frequent morning stiffness and rated the RAID, PROFAD and FFbH dimensions poorer than responders. IR was not associated with baseline DAS28 and comorbidities but with patient-reported baseline FFbH, PROFAD and RAID scores. Multivariate analyses of baseline parameters revealed younger age (OR 1.3 per 10 years) and impaired FFbH (OR 1.2 per 10% function loss) as significant predictors of IR.
Conclusions The rheumatologist's rating of IR was highly associated with clinical outcome at two years reflected by significantly higher DAS28 and intensive need of therapies as well as inferior patient-reported outcomes. Early functional impairment seems to be predictive of IR.
Disclosure of Interest K. Albrecht Grant/research support: The observational cohort was funded by an unconditional grant from Pfizer., M. Chubrieva: None declared, J. Callhoff: None declared, E. Edelmann: None declared, M. Schneider: None declared, A. Zink: None declared, G. Westhoff: None declared