Article Text

OP0003 Are patient global and fatigue predictive of structural outcomes, 3 years later, in patients in remission in early arthritis? Results from the french espoir cohort
  1. L. Gossec1,
  2. C. Gaujoux-Viala2,
  3. A.-C. Rat3,
  4. B. Fautrel2,
  5. A. Saraux4,
  6. F. Guillemin3,
  7. S. Paternotte1,
  8. M. Dougados1
  1. 1Rheumatology B Department, Cochin Hospital, Paris Descartes University
  2. 2Rheumatology Department, Pitié Salpétrière Hospital, Paris 6 University, Paris
  3. 3Lorraine university, Nancy hospital, Nancy
  4. 4Rheumatology Department, Cavale Blanche hospital, Brest, France


Background The new ACR/EULAR remission criteria were developed to identify patients with good outcomes, among early arthritis (EA) patients [1]. It has been hypothesized that patient-reported symptoms and in particular patient global, is often a limiting factor to reaching ACR/EULAR remission [2]. Furthermore, fatigue is an important symptom for patients, but is it predictive?

Objectives To determine if patient global and fatigue over the first year of EA disease course, play a significant role in comparison to other ACR/EULAR remission components, in predicting a hard outcome, structural progression, over 3 years.

Methods Study design: the ESPOIR observational study. Patients: EA (at least 2 swollen joints for <6 months). Definitions of remission: (a) ACR/EULAR Boolean remission at both the 6 month and 12 month visits (i.e., tender and swollen joint counts ≤1, CRP≤1 mg/dl, and patient global ≤1/10), (b) near-remission at the same timepoints, defined as remission for 3 of the 4 criteria (all criteria except patient global), and (c) fatigue-remission at the same points, defined as near-remission AND fatigue, ≤1/10 on a visual analog scale. Outcome: change in total Sharp-van der Heijde score between baseline and at 3 years. Analysis: kappa agreement statistics between the definitions of remission, and predictive capacities of the 3 remissions by multiple linear regression, adjusted on baseline radiographic score. Models were compared by various characteristics: R2 (should be maximal), C(p) (should be closest to 2 in this case), AIC and BIC (should be minimal). Furthermore, a stepwise selection was performed to explain the outcome, including near remission, patient global and fatigue. There was no imputation of missing data.

Results Of the 776 patients, 57 patients (7.4%) were in ACR/EULAR remission both at 6 and 12 months, whereas 145 patients (18.7%) reached near-remission and only 24 patients (3.1%) reached fatigue-remission. Agreement between ACR/EULAR remission and the other definitions was moderate: kappa, 0.51 (95% CI, 0.43-0.60) and 0.39 (0.26-0.53) respectively. Prediction of radiographic progression was similar in all models, but appeared maximal in the models with near-remission and ACR/EULAR remission (table). In stepwise selection only near-remission was predictive.

Table 1. Models for prediction of change in radiographic score at 3 yrs, adjusted on baseline radiographic score

Conclusions Near-remission is much more frequent than ACR/EULAR remission in EA. It appears from this analysis that near-remission (not taking into account patient global) predicts radiographic progression over 3 years in EA, as well as ACR/EULAR remission. Near-remission may be a valid and even a sufficient predictive outcome in EA.

  1. Felson DT et al. Ann Rheum Dis 2011; 70(3):404-13.

  2. Vermeer M et al. Arthr Rheum 2011, ACR Congress, abstract 2459

Disclosure of Interest None Declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.