SAT0362 Performance of the 2015 acr/eular classification criteria for gout in korean patients presenting with acute arthritis

Background A definite diagnosis of gout requires the detection of monosodium urate (MSU) crystals in synovial fluid or tissue. However, the procedure involved is invasive and often not available in acute or primary care settings. Therefore, several alternate sets of classification criteria for gout have been developed. Objectives This study was undertaken to assess the performance of the 2015 American College of rheumatology (ACR)/European League Against Rheumatism (EULAR) classification criteria for gout and to compare its performance to previous sets of criteria including the New York, American Rheumatism Association (ARA) 1977 and Mexico criteria, in Korean patients presenting with acute arthritis. Methods Patients with acute arthritis who attended rheumatology clinics and underwent arthrocentesis were enrolled in the study, during February 2017 and December 2017. The gout group included patients with MSU crystals in synovial fluid or tissue, as assessed by an experienced rheumatologist or pathologist. Since the presence of MSU crystals alone is sufficient to fulfil the criterion for gout according to the 2015 ACR/EULAR classification, New York, ARA 1977 and Mexico criteria, this criterion is considered to have a 100% sensitivity in the gout group by our definition. Therefore, we excluded positive MSU crystal results in these four criteria sets to avoid a case of circular reasoning. Results A total of 116 gout patients (as determined by the presence of MSU crystals) and 71 non-gout patients participated in the study. The number of male patients in the gout group (94.8%) was higher than that in non-gout group (15.1%). The mean (±SD) age of the patients was 58.4 (±15.0) and 64.5 (±13.3) years in the gout and non-gout group, respectively. The sensitivity, specificity, positive predictive value and negative predictive value of the criteria sets were 87.1%, 100%, 100% and 73.2% respectively (area under the curve, AUC 0.975) for the 2015 ACR/EULAR classification criteria, 71.5%, 97.2%, 97.6% and 67.6% (AUC 0.960) for the clinical-parameters-only version of the 2015 ACR/EULAR classification criteria, 68.1%, 81.7%, 85.9% and 61.0% (AUC 0.784) for the New York criteria, 77.6%, 81.7%, 87.4% and 69.0% (AUC 0.878) for the ARA 1977 criteria, and 87.9%, 63.4%, 79.7% and 76.3% (AUC 0.878) for the Mexico criteria. To determine the performance of the existing criteria sets for gout without relying on the detection of MSU crystals, we also calculated the performance of the criteria assuming an absence of both positive and negative crystal test results. By this parameter, the sensitivity of the 2015 ACR/EULAR classification criteria was 77.6%, specificity was 84.8%, positive predictive value was 89.1%, and negative predictive value was 69.8% (AUC 0.918). The performance of the New York, ARA 1977 and Mexico criteria did not change. Conclusions The performance of the 2015 ACR/EULAR classification criteria for gout was better than that for the New York, ARA 1977 and Mexico criteria in Korean patients presenting with acute arthritis. When an MSU crystal identification test is not available, 2015 ACR/EULAR classification criteria for gout could be a useful option in the diagnosis of gout in Korean patients presenting with acute arthritis. Disclosure of Interest None declared