Validation of the ACR/EULAR classification criteria for systemic sclerosis in patients with early scleroderma

The aim of this study was to validate the 2013 ACR/EULAR classification criteria for systemic sclerosis (SSc) in patients with SSc, including patients with early SSc. Fifty-six consecutive patients with early SSc (2001 LeRoy and Medsger criteria), 122 patients with established SSc (1980 ACR classification criteria), and 141 patients with SSc-like disorders were included in this cross-sectional study. The diagnostic performance of the 2013 ACR/EULAR criteria was compared with the 1980 ACR criteria in several subsets of patients. The performance of individual variables was also obtained. Receiver operating characteristic (ROC) curves and optimal cut-off values were computed. The sensitivity and specificity in the whole cohort of 178 SSc patients were 77.6 and 98.5%, respectively, using the 2013 ACR/EULAR criteria and 68.5 and 100%, respectively, using the 1980 ACR criteria. Twenty-eight percent of the patients with early SSc met the 2013 ACR/EULAR criteria. Among the patients with early SSc, 53% of those who had Raynaud's phenomenon, abnormal capillaroscopy and positive SSc-related antibodies met the 2013 ACR/EULAR criteria. The area under the ROC curve was 0.975 (95% confidence interval 0.962-0.987). The best cut-off value for the total score was >= 8 (sensitivity 82%; specificity 97.9%). The individual variables with the highest specificity values were proximal skin thickening, sclerodactyly (specificity 100%), telangiectasia and SSc-related antibodies (specificity 98.6%). Raynaud's phenomenon had the best sensitivity (99.4%) but had low specificity (4.2%). In conclusion, the 2013 ACR/EULAR classification criteria showed high accuracy and increased sensitivity in the classification of patients with early SSc.