Background: An ACR/EULAR task force released new criteria in 2013 to classify patients with systemic sclerosis (SSc). Objectives: This study evaluates the diagnostic performance of these criteria in a multidisciplinary care setting. Methods: Patients with an active follow-up in a Systemic Autoimmune Diseases Unit with a clinical diagnosis of SSc were matched by age and gender with consecutive patients referred to a capillaroscopy clinic. The classification criteria were tested on discrimination and diagnostic accuracy between both groups of patients defined as cases-SSc and controls. Receiver operating characteristic (ROC) curve and the area under the curve (AUC) was calculated for the global score to define the best cut off to classify the patients as having SSc. Results: A total of 130 patients with SSc and 130 matched-controls were included in this analysis, 90% women, with a mean age of 61.5. Main diagnosis for the control groups were primary Raynaud´s phenomenon (34.6%), undiferentiated connective tissue disease (13.1%), and mixed connective tissue disease (9.2%). The 92% and 8% of patients in the SSc-cases and control groups met the 2013 ACR/ EULAR SSc classification criteria respectively. Sensitivity and specificity of the criteria were 81.5% and 93.7%, respectively. The best cut offs for the total score were 8 and 9, and the AUC (95%CI) was 0.962 (0.939-0.985). The individual items with a better discriminatory capacity were abnormal capillaroscopy, telangiectasia and anticentromere antibody positivity. Table 1. Demographic data and ACR/ EULAR SSc classificacion criteria of SSC patients and controls. Cases-SSc, n= 130 Controls, n= 130 p Age, mean (SD) 62.4 (16.0) 61.3 (14.8) 0.58 Female, % 90.5 90.8 0.92 Disease duration from onset of symptoms, mean (SD) 7.5 (6.4) 7.8 (6.7) 0.73 % Patients with individual ACR/EULAR 2013 Criteria Items Skin thickening 15.8 0.7 <0.001 Sclerodactily 45.5 6.1 <0.001 Puffy fingers 27.7 1.5 <0.001 Digital tip ulcers 30.2 5.3 <0.001 Fingertip pitting scars 18.8 4.6 0.001 Telangiectasia 51.5 6.1 <0.001 Abnormal nailfold capillaries 79.4 0.8 <0.001 Pulmonary Arterial Hypertension 16.3 3.8 0.001 Pulmonary fibrosis 18.8 6.2 0.002 Raynaud´s phenomenon 90.7 66.9 <0.001 Anticentromere antibody 60.7 18.9 <0.001 Scl70 12.0 0 <0.001 Total Score 12.5 (4.8 ) 3.4 (2.9 ) <0.0001 Figure 1. ROC curve for global score of the ACR/EULAR2013 SSc classification criteria. Conclusion: The ACR/EULAR 2013 criteria showed good diagnostic properties in this cohort reflecting daily practice. Individual items showing the highest discriminatory capacity were abnormal capillaroscopy, telangiectasia and anticentromere antibody positivity. References: [1]van den Hoogen F, Khanna D, Fransen J, et al. 2013 classification criteria for systemic sclerosis: an American college of rheumatology/European league against rheumatism collaborative initiative. Annals of the Rheumatic Diseases 2013;72:1747-1755. Disclosure of Interests: None declared

Label Smoothing Auxiliary Classifier Generative Adversarial Network with Triplet Loss for SAR Ship Classification