Background The true impact of biologic agents in rheumatoid arthritis (RA) is difficult to assess without knowledge on the long-term disease course and outcome in the time preceding the introduction of these agents.
Objectives The present study intends to report the course and outcome of a cohort of patients with early RA included during the 1990ies and followed for up to 8 years.
Methods 839 patients from the BARFOT early RA cohort were included between the years 1992 and 1999 with the intention to be followed up for 8 years. Of these, 789, 824, 839, 742 and 640 attended the follow-up visits at 6 months, 1, 2, 5, and 8 years.
The Disease Activity Score (DAS28) <2.6 remission criterion was used. Remission was assessed at all follow-up visits (at 6 months, 1, 2, 5. and 8 years). Physical function was measured by the Stanford Health Assessment Questionnaire (HAQ). Radiographs were scored by the Sharp van der Heijde method and expressed as total score (SHS).
Results 33% of the patients had no episode with remission at any of the visits - the persistent disease (PD) group. The rest of the patients had one or more occasion of remission - the non-persistent disease (Non-PD) group.
In comparison with the Non-PD group, the PD group experienced a more severe disease with significantly more pain, more tender joints and more impairment of daily life function (HAQ) at all follow-up visits. Similarly, at all visits, SHS was significantly higher in the PD group.
Initially, DMARD and glucocorticoid treatment was similar in the two groups but at the 2 and 5 year visits, combination DMARD therapy was more prescribed in the PD group than in the Non-PD group (overall p=0.022 and 0.004, respectively). Biologics were subsequently added to a number of patients, most frequently to patients with PD (non significant). With time, glucocorticoids were given to significantly more patients with PD.
Univariate analyses of baseline clinical and demographic variables showed that the PD group had higher baseline mean DAS28, VAS PatGA (patient's global assessment), number of tender joints and ESR as well as VAS pain and HAQ.
PD was observed in 38% of the women vs 24% of the men, p=0.001, in 42% of current vs 30% of previous/never smokers, p=0.011 and in 38% of patients with disease duration at inclusion >6months vs 29% with shorter duration, p=0.027. Multivariate analyses confirmed these three variables as independent predictors of PD. Anti- CCP positivity was not associated with disease persistency, p=0.48.
Conclusions Many patients suffered a relentless disease course over 8 years with only minor improvement from disease onset. Women were more likely than men to develop an unfavorable disease course while baseline aCCP status was not predictive of disease persistency. Further identification of predictors for PD may guide initial treatment and prevent severe disease.
Disclosure of Interest : None declared