POS0163 PERFORMANCE OF THE SALIVARY GLAND ULTRASONOGRAPHY OMERACT SCORE IN THE ACR-EULAR CLASSIFICATION CRITERIA FOR SJÖGREN’S SYNDROME

Abstract

BackgroundUltrasonography is a tool to assess the involvement of major salivary glands in patients with Sjögren’s syndrome (SS). Recently, the Outcome Measures in Rheumatology (OMERACT) scoring system has been developed for salivary gland ultrasonography (SGUS) in SS.[1]Incorporation of this scoring system in the 2016 American College of Rheumatology (ACR)-European League Against Rheumatism (EULAR) classification criteria might be of aid in the classification of SS.ObjectivesTo investigate if addition of the OMERACT SGUS score influences the performance of the 2016 ACR-EULAR classification criteria for SS in daily clinical practice.MethodsThis retrospective, cross-sectional study included consecutive patients of ≥18 years who visited the University Medical Centre Groningen (UMCG) between January 2017 and December 2021 for a diagnostic trajectory to determine whether they had SS. All items of the 2016 ACR-EULAR criteria were collected. Ultrasonographic images were assessed by using the Hocevar score. A transformation table was made by two experienced ultrasonographers (AS and KD) to transform the Hocevar score to the OMERACT SGUS score (range 0-12). Clinical diagnosis by the treating physician was used as gold standard. Receiver operating characteristic (ROC) analysis with area under the curve (AUC) was used to analyse the accuracy of the SGUS OMERACT score, the current ACR-EULAR classification criteria and the ACR-EULAR criteria with addition of the SGUS OMERACT score as well as to explore the optimal cut-off point to predict clinical diagnosis.ResultsOf the 419 included patients, mean age was 53±14 years, 366 (87%) were female and 164 (39%) were diagnosed with SS. Median (IQR) of the SGUS OMERACT score was 7 (4-10) in the SS group versus 1 (0-2) in the non-SS group (Table 1). The SGUS OMERACT score predicted the clinical diagnosis of SS with good accuracy, with an AUC of 0.868 (95% CI 0.830-0.906). The optimal cut-off point of the total OMERACT score to predict clinical diagnosis was ≥5, with a sensitivity of 68.9% and a specificity of 91.4%. An OMERACT score of ≥2 in at least 1 gland showed a sensitivity of 75.6% and a specificity of 85.5%.The current ACR-EULAR classification criteria showed excellent accuracy to predict clinical diagnosis, with an AUC of 0.956 (0.937-0.975). When incorporating the SGUS OMERACT score (defined as total score ≥5) as additional item in the ACR/EULAR criteria, the AUC was 0.968 (0.953-0.983). When incorporating the SGUS OMERACT score (defined as ≥2 points in 1 gland) as an additional item, the AUC was 0.967 (0.951-0.982).With the original ACR/EULAR cut-off score of ≥4 for classification as SS, the sensitivity was 92.7% and the specificity 87.5% for the current ACR-EULAR classification criteria, 93.3% and 87.1% for the ACR-EULAR criteria with addition of SGUS OMERACT positivity based on total score ≥5, and 93.9% and 85.5% for the ACR-EULAR criteria with addition of SGUS OMERACT positivity based on ≥2 points in 1 gland.ConclusionIncorporating SGUS in the criteria extends the diagnostics options in patients suspected with SS. The accuracy of the ACR-EULAR classification criteria in predicting the clinical diagnosis of SS remained high after incorporating the SGUS OMERACT score.Reference[1]Finzel S et al. Rheumatol 2020. Doi: 10.1093/rheumatology/keaa471.Table 1.Characteristics of study populationCharacteristicsSS (n = 164)No SS (n = 255)P-valueAge (years)53 ± 1453 ± 140.777Sex (female)143/164 (87.2)223/255 (87.5)0.939FS ≥1 in salivary gland biopsy107/133 (80.5)22/158 (13.9)<0.001Anti-SSA positive131/164 (79.9)32/253 (12.6)<0.001Ocular staining score ≥5*34/164 (20.7)15/255 (5.88)<0.001Schirmer ≤5 mm*109/158 (69.0)128/242 (52.9)0.001Unstimulated whole saliva ≤0.1 ml/min91/162 (56.2)90/253 (35.6)<0.001Total OMERACT score (0-12)7 [4-10]1 [0-2]<0.001Total OMERACT score ≥5 in 4 glands113/164 (68.9)22/255 (8.63)<0.001OMERACT score ≥2 in at least 1 gland124/164 (75.6)37/255 (14.5)<0.0012.Data presented as mean ± SD, median [IQR] or n (%)3.* Scored in at least 1 eyeAcknowledgementsSeveral authors of this publication are member of the European Reference Network ReCONNET.Disclosure of InterestsDagmar Rebel: None declared, Liseth de Wolff: None declared, Konstantina Delli: None declared, Alja J. Stel: None declared, Helene Kokol: None declared, Arjan Vissink Grant/research support from: A. Vissink has received unrestricted grants from Bristol-Myers Squibb and Roche., Frans G.M. Kroese Speakers bureau: F.G.M Kroese has worked as a speaker for Bristol-Myers Squibb, Roche and Janssen-Cilag., Consultant of: F.G.M Kroese has worked as a consultant for Bristol-Myers Squibb., Grant/research support from: F.G.M Kroese has received unrestricted grants from Bristol-Myers Squibb., Suzanne Arends Consultant of: Suzanne Arends is consultant for ArgenX and Novartis., Hendrika Bootsma Speakers bureau: H. Bootsma has worked as a speaker for Bristol-Myers Squibb and Novartis, Consultant of: H. Bootsma has worked as a consultant for Bristol-Myers Squibb, Roche, Novartis, Union Chimique Belge and ArgenX., Grant/research support from: H. Bootsma has received unrestricted grants from Bristol-Myers Squibb, Roche and AstraZeneca.