Management of contemporary early undifferentiated arthritis: data on EULAR’s recommendation on the risk of persistent disease

Early treatment initiation is crucial to improve long-term outcomes in rheumatoid arthritis (RA). This may also apply to undifferentiated arthritis (UA), patients at high risk of persistent arthritis/RA. Therefore, the EULAR recommendations for early arthritis recommend assessing the following risk factors for disease persistency in early UA: number of swollen joints, acute phase reactants (ie, C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR)), rheumatoid-factor (RF), anti-citrullinated protein antibodies (ACPA) and imaging findings/erosions.^{1 2} This recommendation was based on markers identified as predictive in a systematic literature review.² Importantly however, prognostic research in UA is based on an outdated definition of UA: not meeting 1987-RA-classification criteria and having no alternative diagnosis (‘conventional UA’). A proportion of these patients with conventional UA meet the 2010-RA-criteria and is currently considered to have RA.^3–5 Contemporary UA, in contrast, is defined as neither meeting the 1987-RA-criteria nor the 2010-RA-criteria and having no other clinical diagnosis. In addition, for some of the recommended risk factors data were lacking in the systematic literature review (polyarthritis) or it was concluded that adequately designed studies were lacking (CRP and ESR).² Since predictors may be disease stage or population dependent, and because predictors for persistent disease identified in conventional UA may not be applicable to contemporary UA, we conducted a large cohort study in contemporary UA to assess risk factors for persistent disease mentioned in the current EULAR recommendation. Conventional UA was studied for comparison.

In short, 710 patients with contemporary UA, not fulfilling 1987-RA-criteria or 2010-RA-criteria and having no alternative diagnosis, consecutively included in the Leiden Early Arthritis Clinic cohort between 2006–2019, when disease-modifying anti rheumatic drug (DMARD) start in UA was recommended, were studied (online supplemental figure 1). The cohort is described in detail elsewhere.⁶ At inclusion swollen joint counts and laboratory procedures were performed, including: ACPA (EliA CPP (anti-CCP2), Phadia, the Netherlands, elevated ≥10 U/mL)), IgM-RF (in-house ELISA, elevated ≥5.0 IU/mL), CRP (elevated ≥10 mg/L) and ESR. Patients were assessed after 4 months, 12 months and yearly thereafter. The outcome was sustained remission, thus absence of disease persistence, defined as sustained absence of clinical synovitis without DMARDs (including corticosteroids) for at least 1 year and the entire follow-up. The cumulative incidence was visualised using Kaplan-Meier. Cox regression analysis was used to test risk factors. Radiographic erosions at baseline were rare (1.8%) and not included in analyses. Within the same inclusion period, 1004 patients with conventional UA were included (not fulfilling the 1987-RA-criteria and having no other diagnosis); analyses were also performed in this population.

Patients with contemporary UA presented with a median of two swollen joints and were mostly ACPA-negative (online supplemental table 1 for baseline characteristics). The median follow-up was 6 years (IQR 3–9 years). DMARDs were started by 48% of the patients. Sustained remission after a median of 1.5 years (IQR 1–3) was achieved by 60% of the patients with UA, after which they were followed for another 5.5 years (IQR 3–8) without recurrence of arthritis, demonstrating the sustained absence of disease. Univariable analyses showed that CRP, ESR, ACPA and RF were associated with time to sustained remission (HR 0.77 (95% CI: 0.62 to 0.95), HR 0.79 (95% CI: 0.64 to 0.97), HR 0.18 (95% CI: 0.08 to 0.44) and HR 0.49 (95% CI: 0.31 to 0.77), respectively), while polyarthritis (HR 0.76 (95% CI: 0.57 to 1.01)) was not statistically significant. In multivariable Cox regression ACPA (HR 0.095 (95% CI: 0.03 to 0.32); figure 1A) and CRP (HR 0.67 (95% CI: 0.50 to 0.91); figure 1B) remained significantly associated (online supplemental table 2). Assessing the number of these remaining two risk-factors (0, 1, 2 factors; figure 1C) showed that patients with UA without any of these two factors (67% of patients with UA) achieved sustained remission in 77%. Patients with two risk-factors in contrast, were rare (2%) and had persistent disease in 56%.

Figure 1.

Cumulative incidence of sustained remission in contemporary undifferentiated arthritis for presence of ACPA positivity (A), elevated CRP (B), 0, 1 and 2 of these factors (C). (1A) Of all the patients with contemporary UA, 5% was APCA positive (≥10 U/mL); (1B) in total 31% of the patients with contemporary UA had an elevated CRP (≥10 mg/L); (1C) of all the patients with contemporary UA 67% had no risk factors, 31% had one risk factor and only 2% had two risk factors. ACPA, anti-citrullinated protein antibodies; CRP, C-reactive protein; UA, undifferentiated arthritis.

For comparison, patients with conventional UA were more often ACPA positive (online supplemental table 1 for baseline characteristics). Multivariable Cox regression analysis in these patients with UA revealed that ACPA, RF, CRP and polyarthritis were associated with sustained remission (online supplemental table 2).

Concluding, the population with contemporary UA is different from conventional UA and risk factors for disease persistence are partly dissimilar. ACPA and CRP remain to be predictive in contemporary UA. Other factors included in the current EULAR recommendation are uninformative (RF, ESR and polyarthritis) or rare (erosions). Further prognostic studies in contemporary UA are warranted, after which risk factors recommended in future EULAR recommendations may require revision.