Performance of the 2012 Systemic Lupus International Collaborating Clinics and the 1997 American College of Rheumatology classification criteria for systemic lupus erythematosus in a real-life scenario.

OBJECTIVE: To evaluate the performance of the 2012 Systemic Lupus International Collaborating Clinics (SLICC) criteria in classifying systemic lupus erythematosus (SLE) in an uncontrolled real-life scenario.
METHODS: Chart review study was performed in which each criterion from the 1997 American College of Rheumatology (ACR) and the 2012 SLICC criteria to classify SLE was applied to patients from an outpatient rheumatology clinic. The clinical diagnosis was used as the gold standard.
RESULTS: The sensitivity and specificity of the 2012 SLICC criteria were 92% and 99%, respectively, compared with the 1997 ACR criteria, which were 97% and 99%, respectively. The 2012 SLICC criteria were similar to the 1997 ACR criteria in terms of positive (98.9% versus 99%) and negative (92.5% versus 97.1%) predictive values as well as positive (92 versus 97) and negative (0.08 versus 0.03) likelihood ratios. A concordance of 0.96 (95% confidence interval [95% CI] 0.92–1.00) was observed between clinical diagnosis and the 1997 ACR criteria, while the concordance was 0.91 (95% CI 0.85–0.97) for the 2012 SLICC criteria. Seven SLE patients classified by the 1997 ACR criteria did not meet the 2012 SLICC criteria because of either the new definition for lymphopenia (2 patients) or the presence of isolated cutaneous involvement (5 patients), while 2 SLE patients who were classified by the 2012 SLICC criteria did not meet the 1997 ACR criteria because of either the presence of erosive arthritis or biopsy-proven nephritis with circulating antinuclear antibodies.
CONCLUSION: Overall, the 1997 ACR and the 2012 SLICC criteria are similar to classify SLE in an uncontrolled real-life scenario, although several new items contained in the 2012 SLICC criteria could represent a step forward for research purposes in selected clinical settings.