Prompt identification of patients with rheumatoid arthritis (RA), ideally within a window of opportunity of approximately 12 weeks, increases potential for antirheumatic treatments to dampen the inflammatory process in a… Click to show full abstract
Prompt identification of patients with rheumatoid arthritis (RA), ideally within a window of opportunity of approximately 12 weeks, increases potential for antirheumatic treatments to dampen the inflammatory process in a milder and more reversible stage of the disease, thus enabling more favourable outcomes. Over the past 20 years, strategies aimed at reducing delays in RA referral and treatment have included the widespread diffusion of dedicated early arthritis clinics (EACs), as well as the development of more sensitive classification criteria. Still, the percentage of patients seen within the window of opportunity apparently remains low, and the new RA criteria, heavily weighted on autoantibodies, may have further hindered the recognition and treatment of seronegative patients. Here, we analysed changes in the diagnostic delay and clinical presentation of patients with RA admitted to the EAC of the Division of Rheumatology of the San Matteo University Hospital, Pavia, Italy, from its institution in 2005 to 2017. Referral criteria to the EAC have remained stable over the years and include ≥3 swollen joints (SJs), or in case of <3 SJs, a positive squeeze test or morning stiffness >30 min. From all patients with early arthritis (N=1553), we selected 668 patients fulfilling at enrolment at least the 1987 American College of Rheumatology (ACR) criteria for RA before December 2010 (n=345, 88.4% also fulfilling the 2010 criteria) and at least the 2010 ACR/European Alliance of Associations for Rheumatology criteria after January 2011 (n=323, 63.5% also fulfilling the 1987 criteria). In line with published studies and with the prognostic value of the 2010 criteria, application of the two sets of criteria was used as reference for RA. Time from first selfreported joint symptom to referral was compared across different time periods: (1) 2005–2007, (2) 2008–2010, (3) 2011–2013 and (4) 2014– 2017. Clinical characteristics were collected according to standardised assessments. Data were analysed in the total population and after stratification for autoantibody status (doublenegative for rheumatoid factor (RF) and anticitrullinated protein antibodies (ACPA) vs RFpositive and/or ACPApositive). In all, delay in the referral of patients classified as RA collectively increased from a mean (SD) of 20.8 (20.5) weeks before 2010 to 24.4 (20.6) weeks thereafter (p=0.02) (online supplemental table S1), and the proportion of patients identified within 12 weeks nonsignificantly decreased from 39.3% to 35.3%. Still, patients presented with progressively milder inflammatory markers despite unchanged joint tenderness and patientreported outcomes (PROs) (online supplemental table S1). Trends were however remarkably different in different autoantibody subgroups. In RF/ACPApositive patients, a stable proportion of 41%–44% were referred within 12 weeks (figure 1A), with only marginal differences of around ±10% in relation to disease activity (online supplemental figure S1A,B). At presentation, patients had less SJs and lower C reactive protein (CRP) levels; the mean decrease of SJs and CRP from 2005–2007 to 2014–2017 were −5.6 and −1.1 mg/dL, respectively (figure 1B and online supplemental table S2). The improvement in PROs was smaller but still significant over time (figure 1C and online supplemental table S2). In contrast, in autoantibodynegative patients, the proportion of patients identified within 12 weeks progressively decreased from 37.9% to 25.6% (p=0.08) (figure 1D). Of note, the reduction in the rate of early referral after 2010 was prominent in patients classified as RA solely Letter
               
Click one of the above tabs to view related content.