• Rheumatoid Arthritis
• Treatment
• Disease Activity

We thank Mankia et al1 for their interest in the European League Against Rheumatism (EULAR) definition of arthralgia suspicious for progression to rheumatoid arthritis (RA).2 The authors agree with the taskforce that derivation of criteria for imminent RA is an ambitious next step and that such criteria will probably consist of a combination of clinical, serological and imaging biomarkers.1 In this light we appreciate the work the authors have done to identify serological and imaging markers that are predictive in patients with anti-citrullinated protein antibody (ACPA)-positive arthralgia.

While studies on blood and imaging markers in arthralgia are relatively frequent, only few studies have addressed the symptoms and signs of the disease stage that may precede clinical arthritis. In addition, the clinical delineation of this preclinical stage, reflected by the intuitive contrast in the description ‘clinically suspect arthralgia (CSA)’, is difficult: there is not one key symptom. Still, in our experience rheumatologists are capable of identifying patients with arthralgia that may progress to RA based on their expertise and on (intuitive) pattern recognition. In an attempt to strip the term ‘suspect arthralgia’ of its connotation of subjectivity and to promote the inclusion of homogeneous groups of patients with arthralgia in future studies, the taskforce has agreed on a consensual definition of ‘arthralgia at risk for RA’. This definition is deliberately meant to be used in secondary care, for patients in whom imminent RA is considered a more likely explanation for the complaints than another disease, but who not (yet) have clinical arthritis.

Mankia et al rebut that in several settings patients with arthralgia are followed in primary care (too long) until clinical arthritis has become manifest. They propose to use the CSA definition as a referral tool in primary care.1 Referral tools share characteristics of screening tools, such as high sensitivity and lower specificity, that pose huge challenges: Unlike the specialist setting, most patients with musculoskeletal symptoms presenting in primary care will have other more trivial explanations for their complaints than (imminent) RA. Consequently, the prior risk of RA will be low, as will be the predictive value of a positive CSA definition. The impact on specialist care may be significant.

We reiterate that the EULAR definition was not developed for the primary care setting, nor was it designed as a diagnostic test. The EULAR definition of CSA was designed by rheumatologists, with their perception of imminent RA as a reference frame, and was tested in patients from secondary care. Its seven items should be assessed in patients presenting to the rheumatologist in whom the specialist does not find clinical arthritis but imminent RA is still considered a likely diagnosis.2 General practitioners often find it difficult to detect synovitis and to evaluate if imminent RA is more likely than other, trivial arthralgia's (instead, this uncertainty will often be the reason to refer to secondary care anyway). Therefore, the entry condition cannot be adequately evaluated in a primary care setting. Even so, there is a chance that the seven items will perform better than expected as a reference tool (either when applied in isolation or in combination with an additional test). Still to arrive at accurate referral criteria, these are ideally designed in primary care.