THU0259 ULTRASOUND OF MAJOR SALIVARY GLANDS IDENTIFIES PATIENTS WITH MORE COMPLEX SALIVARY GLAND HISTOPATHOLOGY BUT NOT MILDER LESIONS

Abstract

Background: Ultrasonography of major salivary glands (SGUS) has been proposed as a feasible and reliable tool for the assessment of parotid and submandibular glands inflammation and damage in Sjogren’s syndrome (pSS). However, the relationship between major SGUS and labial salivary gland (LSG) histopathology is still unclear. Objectives: to investigate the correlation between SGUS abnormalities and LSG histopathology in patients newly diagnosed patients with pSS Methods: Consecutive patients suspected of having pSS were enrolled in this study. All the patients underwent a complete diagnostic work up. The echostructure of each gland on B-mode images was graded on a 5-point scale (0–4), and a SGUS score ≥2 was defined as pathological. Hypo-anechoic areas in the glands were defined as isolated ( 50%). Hyperechoic bands present in more than 50% of the parenchyma were also recorded. An expert pathologist analyzed all the LGS samples assessing focus score (FS), number of foci and presence of ectopic germinal centers (GCs). Results: We included 115 pts (105 F: 10 M, mean age (SD) 54.8 (12.5) years). Out of them 51 were diagnosed with pSS and 64 represented the no-SS sicca controls. A SGUS score ≥2 identified pSS patients with a sensitivity of 62.7% and a specificity of 96.7%. SGUS score significantly correlated with FS (r=0.457**, p=0.000), number of foci (r=0.359**, p=0.000) and number of ectopic GCs (r=0.505**, p=0.000). However, the overall concordance between SGUS and LSG histopathology was moderate (Cohen’s kappa =0.574). Thus, we specifically focused on 19/51 pSS “discordant” cases who presented normal SGUS findings but focal lymphocytic sialadenitis in their LSG biopsy. These “discordant” patients tended to present a lower FS (1.7±0.5 vs 2.6±1.5, p=0.04) and a significantly lower number of foci in their biopsies (2.7±1.6 vs 4.6 ±2.9, p=0.01) when compared to patients with both abnormal SGUS findings and LGS focal lymphocytic sialadenitis. Conclusion: SGUS and LSG histopathology showed only a moderate correlation in patients with pSS. SGUS may be useful to identify pSS patients with a more severe inflammatory LSG infiltrates, ultimately contributing to promote non-invasive phenotyping of multiple subsets in pSS. Disclosure of Interests: Chiara Baldini: None declared, Francesco Ferro: None declared, Nicoletta Luciano: None declared, Gianmaria Governato: None declared, Marta Mosca Paid instructor for: GlaxoSmithKline, Lilly, UCB, Stefano Bombardieri: None declared, Valentina Donati: None declared