University of Brescia Sotto il Monte Giovanni XXIII, Bergamo, Italy
Disclosure(s): No financial relationships with ineligible companies to disclose
Background/Purpose: To assess the 2012 SLICC and 2019 EULAR/ACR classification criteria performance as diagnostic criteria in two cohorts of ANA-positive patients clinically suspected of having CTDs. Methods: An observational study of 286 ANA-positive, treatment-naïve patients with CTD-related symptoms (≤1 year) was conducted. The 2012 SLICC and 2019 EULAR/ACR criteria were assessed at enrollment (T0) and during a 3-year follow-up. T1 was defined as the visit when patients met classification criteria or their last follow-up. Anonymized charts from T0 and T1 were reviewed by four independent consultants to assess diagnoses and treatment decisions. Results: At T1, consultants diagnosed SLE in 29 patients (10.14%). The 2012 SLICC criteria were met in 38 patients (13.29%) and the 2019 EULAR/ACR in 39 (13.64%). 6 patients (2.1%) were diagnosed with SLE despite not meeting either classification criteria, while 18 participants (6.29%) met at least one classification criterion but were not diagnosed with SLE by a consultant. Among these 18 patients, consultants recommended hydroxychloroquine monotherapy in 7 cases, immunosuppressive therapy alone in 6, a combination of the two in 1, and no treatment in 4.
Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy at T1 of the 2019 EULAR/ACR and 2012 SLICC SLE criteria are reported in table 1. Conclusion: In a diagnostic setting, the criteria are less effective than in a classification context and should therefore be applied with caution. They are more useful for ruling out SLE rather than confirming it, given their moderate sensitivity and PPV, but high specificity and NPV.
