Normal salivary gland ultrasonography could rule out the diagnosis of Sjogren's syndrome in anti-SSA-negative patients with sicca syndrome

Objective To evaluate the relevance of salivary gland ultrasound (SGUS) and its place in the diagnostic algorithm in patients referred with dry syndrome (DS) for a suspicion of Sjogren's syndrome (SS). Methods We included all patients assessed at our dedicated DS clinic from June 2015 to September 2019 for which a SGUS has been carried out. Images were read blindly and the worst salivary gland was scored according to OMERACT classification. Clinical features, disease activity and treatments were collected. Results 337 patients were seen from June 2015 to September 2019. 269 patients underwent SGUS. 77 patients were diagnosed with SS and 192 did not meet the ACR/EULAR criteria for SS: non-Sjogren's DS (NSDS). Of these 192 patients, 60 had another possible cause of DS, and 132 patients were diagnosed with SAPS (sicca, asthenia, polyalgia syndrome). SGUS abnormalities were significantly higher in patients with SS versus NSDS: 51% vs 8% for a score >= 2 (p<0.0001), and 43% vs 3% for a score >= 3 (p<0.0001). SGUS score >= 2 had a specificity (Sp) of 91%, sensitivity (Se) of 57%, positive predictive value (PPV) of 72% and negative predictive value (NPV) of 82% for SS diagnosis. SGUS's characteristics in SSA-negative patients were similar to the whole population (Se=42%, Sp=91%, PPV=42%, NPV=92%). The high specificity and NPV in this population could avoid labial salivary gland biopsy (LSGB) in SSA-negative patients with normal SGUS (186 patients, 69%). Conclusion SGUS is useful for SS diagnosis. If anti-SSA antibodies are negative and SGUS score <2, the diagnosis of SS is very improbable and LSGB could be avoided.