Objectives: to verify the sensitivity and specificity of the criteria for systemic lupus erythematosus, proposed by the Systemic Lupus International Collaborating Clinics (SLICC) and compare it to the ACR lupus criteria, in a pediatric population. Patients and methods: this is an observational cohort study, with a descriptive analysis of data from a Pediatric Rheumatology center, including 23 patients with Juvenile Systemic Lupus Erythematosus (jSLE) and a control group of 24 patients with Juvenile Idiopathic Arthritis (JIA), both groups recently diagnosed and virgin of treatment. Information on signs and symptoms was obtained on the diagnostic consult, and the ACR and SLICC criteria were applied to both groups. Statistical analysis on descriptive data was performed, presenting them in absolute and relative frequency and calculating sensitivity and specificity for each set of criteria. Results: by comparing the ACR and SLICC criteria, we obtained higher sensitivity and accuracy using the SLICC criteria (100% and 97.9%, respectively) and equal specificity. Individually, the positive ANA criterion had 100% sensitivity but only 58.3% specificity in both classifications. The other criteria showed low sensitivity and high specificity when individually analyzed; renal disorder, leukopenia or lymphopenia, positive anti-DNA antibody and low complement level were the only criteria with sensitivity above 50%. Arthritis was the least specific criterion. Conclusion: our results were similar to previous studies with both children and adults, and classification criteria should be used with caution. The SLICC criteria showed high sensitivity and specificity for the classification of jSLE.