Article Text

SAT0011 What is Undifferentiated Arthritis (UA) in 2013, and can Progression to Rheumatoid Arthritis (RA) be Predicted?
  1. S. C. Horton1,2,
  2. R. J. Wakefield1,2,
  3. A. L. Tan1,2,
  4. J. E. Freeston1,2,
  5. M. H. Buch1,2,
  6. P. Emery1,2
  1. 1Division of Rheumatic and Musculoskeletal Disease, University of Leeds
  2. 2NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom


Background In applying 2010 criteria to patients with early inflammatory arthritis, the weighting of serology in the criteria is such that the majority of ACPA positive arthritis patients are classified as RA. A significant proportion of patients with UA have persistent disease1,2. Differences between ACPA negative(-) and ACPA positive(+) patients suggest their disease pathogenesis is divergent.

Objectives Compare UA and ACPA- RA, and determine clinical and/or ultrasound (US) features predictive of progression to RA.

Methods DMARD-naïve patients with UA or RA (2010 criteria) with outcomes available at 6 months were selected from the Leeds Early inflammatory arthritis cohort (recruitment since 2010). Steroid within the preceding 4 weeks was an exclusion. Patients with UA were categorised as palindromic, oligo- or polyarthritis according to intermittent or persistent joint tenderness and/or swelling and involvement of <4 or ≥4 joints respectively.

Results ACPA negativity was observed in 33/37 UA patients and 35/108 RA patients. Baseline characteristics are shown in the table: n(%) or median(IQR), unless otherwise stated. Symptom duration was longer in UA patients, potentially due to a trend towards lower disease activity and lower HAQ in this group.

At 6 months, 70% and 89% of patients with UA and ACPA- RA respectively were receiving any DMARD (p=0.08). There were no statistically significant differences (UA vs ACPA- RA) in; median(IQR) DAS44 1.7(1.3-2.4) vs 2.2(1.7-2.7), improvement in DAS44 from baseline 0.7(-0.1-1.3) vs 0.8(0.1-1.7) or HAQ 0.4(0.1-0.8) vs 0.8(0.3-1.0).

In patients with UA, 7(19%), 19(51%) and 11(30%) were categorised as palindromic, oligo- and polyarthritis respectively at baseline. Over 6 months, 5/19(26%) of patients with oligoarthritis progressed to polyarthritis, only 1 of these patients(5%) fulfilled 2010 criteria when applied cumulatively. Whereas 3/7(43%) palindromic and 4/11(36%) polyarthritis patients progressed to RA (p=0.02 for comparison between oligoarthritis and palindromic/polyarthritis combined). Of palindromic UA 2 patients had subclinical synovitis on US, both of whom progressed to RA (1 patient was ACPA+). Baseline US findings were not significantly associated with progression to RA or need for DMARD over 6 months amongst oligoarthritis or UA patients overall.

Conclusions Clinically important levels of disease activity and disability exist in patients with UA. Differences in clinical phenotype of patients with UA may be predictive of disease progression.


  1. Krabben A, et al. ARD 2012;71(2):238-41.

  2. van der Helm-van Mil A, Huizinga TW. ARD 2012;71:1596-8.

Disclosure of Interest S. Horton: None Declared, R. Wakefield: None Declared, A. L. Tan: None Declared, J. Freeston: None Declared, M. Buch Consultant for: Abbot, Pfizer, Roche, Chugai, BMS, UCB, P. Emery Consultant for: Abbot, Pfizer, Roche, Chugai, BMS, UCB, MSD

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.