Background Recent progress in its treatment has given remarkable benefits to rheumatoid arthritis (RA) patients. Given the fact that early initiation of therapy can ameliorate the disease course, it is important to diagnose RA patients as early as possible. In parallel with this purpose, new classification criteria were published in August 2010. Although the diagnostic performance of these new criteria compared with the ACR 1987 criteria were investigated in several cohorts from different populations, no such study was performed among patients from Turkey. Indeed, some forms of arthritis like familial Mediterranean fever (FMF) and Behçet’s disease (BD), are almost unique to our geographical region and these diseases were not represented in cohorts established in other countries.

Objectives To evaluate the sensitivity of ACR/EULAR 2010 criteria and ACR 1987 revised criteria in an early arthritis cohort from Turkey.

Methods From May 2009 to April 2010, patients who first visited university hospital outpatient clinic with arthritis and who does not fulfill the classification criteria of a specific diagnosis (n=90) were included to the study (the cohort was initiated before the announcement of ACR/EULAR 2010 criteria). Demographic, clinical and laboratory features of all enrolled patients were recorded. Attending physicians were asked to classify patients either as RA or undifferentiated arthritis. All patients were reassessed at 12th month and were re-classified. Proportions of patients fulfilling the aforementioned criteria were determined and compared with clinicians’ diagnoses.

Results At the beginning of the cohort, none of the patients were fulfilling the ACR-1987 revised criteria, while 16 (17.8%) of them were fulfilling the 2010 classification criteria for RA. On the other hand, 43 patients were diagnosed as RA by their attending physicians, which make the sensitivity of the new criteria as 37% at the beginning of the cohort. At the end of the 12-month follow-up period, rates of fulfilling old and new criteria were 31%, and 50%, respectively. The sensitivities of the old and new criteria were 58% and 94%, respectively. At the end of the follow-up, number of patients diagnosed as spondylarthritis, Behcet’s disease and Familial Mediterranean fever were 2, 2 and 3, respectively.

Conclusions Our results have confirmed the high sensitivity of new RA classification criteria for patients with early arthritis in a country where the diseases such as FMF and BD are prevalent. It is also of interest to note that, in patients diagnosed as RA by attending physicians at the time of enrollment, 39 out of 43 of them were re-classified as RA and all were fulfilling the 2010 classification criteria at the end of 12-month follow-up period.

Disclosure of Interest None Declared