Background/Purpose: To evaluate the feasibility and performance of the AECG and ACR Classification Criteria for Sjögren’s syndrome (SS) in patients with systemic autoimmune diseases. 

Methods: 350 patients with primary SS (n=50), systemic lupus erythematosus (n=100), rheumatoid arthritis (n=100), or scleroderma (n=100) were randomly selected from our patients’ registry. Each patient was clinically diagnosed as probable/definitive SS or non-SS by two rheumatologists following a standardized evaluation including clinical symptoms and manifestations, confirmatory tests (fluorescein staining test, non-stimulated whole salivary flow, Schirmer-I test) autoantibodies (antinuclear antibodies, anti-Ro/SSA, anti-La/SSB, rheumatoid factor), lip biopsy, and medical chart review. Using the clinical diagnosis as gold standard, the degree of agreement with each criteria set, and between both criteria sets was estimated. We estimated the sensitivity, specificity, positive predictive value, and negative predictive value with 95% CI. We used the kappa statistic.

Results: 154 (44%) patients were diagnosed with SS. The AECG criteria were incomplete in 36 (10.3%) and the ACR criteria in 96 (27.4%), P<0.001. Nevertheless, their ability in classifying patients was almost identical, sensitivity 61.6 vs. 62.3, specificity 94.3 vs. 91.3, respectively. Either criteria were met by 123 (80%); 95 (61.7%) met AECG and 96 (62.3%) ACR criteria, but only 68 (44.2%) patients met both sets. The concordance rate between clinical diagnosis and AECG or ACR criteria was moderate, k statistic 0.58 and 0.55, respectively. Among 99 patients with definitive SS the sensitivity was 83.3 vs. 77.7, and specificity 90.8 vs. 85.6, respectively. A discrepancy between clinical diagnosis and criteria was seen in 59 (17%) patients. Patients classified by the AECG criteria were older, had more sicca symptoms, parotid enlargement, positive Schirmer-I test, and lower NSWSF rate; whereas those classified through the ACR criteria had more often keratoconjunctivitis sicca, focal sialadenitis, rheumatoid factor, and the combination of rheumatoid factor plus antinuclear antibodies >1:320.

Conclusion:

The feasibility applying the AECG is superior to the ACR criteria; however the performance of both sets was similar among patients with systemic autoimmune diseases. Nevertheless, a subset of patients still is missed by both criteria sets.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/utility-of-the-american-european-consensus-group-and-american-college-of-rheumatology-classification-criteria-for-sjogrens-syndrome-in-patients-with-systemic-autoimmune-diseases-in-the/