european crystal network workshop

    Performance of the 2015 ACR-EULAR classification criteria for gout in a primary care population


    Janssens H., Fransen J., Janssen M., Neogi T., Schumacher H.R., Jansen T., Dalbeth N., Taylor W.J.

    Department of Primary and Community Care Radboud University Medical Centre Nijmegen, the Netherlands


    Objective: To test the performance of the 2015 ACR-EULAR gout classification criteria against presence of synovial fluid MSU crystals in a primary care population of patients presenting with monoarthritis.

    Methods: The criteria were applied to prospectively collected data from consecutive patients with monoarthritis presenting to family practitioners. Diagnostic imaging items were scored 0, as they were not available in the used dataset that was originally aimed to develop a diagnostic rule for easy use in daily primary care practice without hospital facilities. All patients underwent arthrocentesis and synovial fluid microscopy to define true gout status. Test characteristics sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and calibration plots were calculated for full and clinical-only versions of the criteria, and compared to the original characteristics of the diagnostic rule.

    Results: Of 381 patients enrolled into the study, 216 (57%) had MSU crystal proven gout. The 2015 ACR-EULAR gout classification criteria had high specificity and PPV (98% and 98%) but less satisfactory sensitivity and NPV (68% and 70%). The clinical only version had lower specificity and PPV (84% and 84%) but these values were still higher than from the data-derived diagnostic rule at the respective threshold scores of ≥8 and ≤ 4 (71% and 80%). Sensitivity and NPV were highest for the diagnostic rule (99% and 97%).

    Conclusions: In patients presenting monoarthritis in primary care the 2015 ACR-EULAR gout classification performs well in identifying patients with gout for the purpose of enrolling into clinical trials, but may lack sufficient sensitivity for outcomes research or epidemiological research. Further prospective studies are required to confirm these findings.