THE PAROTID GLAND IN PRIMARY SJÖGREN SYNDROME: ASSOCIATION OF ULTRASOUND, HISTOPATHOLOGY AND SALIVA PRODUCTION IN THE DIAGNOSTICWORKUP

Objectives: The parotid glands are commonly involved in primary Sjögren syndrome (pSS). The aim of this study was to assess the diagnostic accuracy of ultrasound of the parotid glands (PSGUS) compared with parotid histopathology and parotid saliva production. Study Design: We included consecutive patients suspected to have pSS. All patients underwent a full diagnostic workup according to American College of Rheumatology– European League Against Rheumatism (ACR-EULAR) criteria, including PSGUS, parotid gland biopsy, and collection of stimulated parotid saliva. For PSGUS, the average score of hypoechogenic areas in both parotid glands was applied (range 0-3). On hematoxylin and eosin (H&E)–stained sections from parotid gland biopsies, focus score, presence of lymphoepithelial lesions (LELs), and germinal centers (GCs) were assessed. The area of lymphocytic infiltrate was calculated digitally on CD45-stained sections. The relative increase of IgG expressing plasma cells (≥30%) was evaluated on sections stained for immunoglobulin A (IgA) and IgG. The PSGUS score was associated with the focus score, percentage of infiltrate, and saliva flow and compared with plasma cell shift, LELs, and GCs by calculating the percentage of absolute agreement, sensitivity, and specificity. Results: Of the 111 included patients, 53 fulfilled the ACREULAR classification criteria for pSS. The PSGUS score showed moderate correlation with the focus score (ρ = 0.494; P < .001) and percentage of lymphocytic infiltrate (ρ = 0.575; P < .001). There was a moderate to good absolute agreement between the PSGUS and focus scores (78.5%), plasma cell shift (79.8%), LELs (81.4%), and GCs (82.7%). “Presence of hypoechogenic areas” was not very sensitive to predict focus score (69.2%), plasma cell shift (45.8%), LELs (61.5%), or GCs (34.6%). Interestingly, almost all patients with less than 25% presence of hypoechogenic areas in glandular parenchyma were also negative for GCs (98.7%). A substantial number of these patients did not have a positive focus score (81.4%), plasma cell shift (90.7%), or LELs (87.8%). There was a fair reversed correlation between PSGUS and stimulated parotid saliva flow (ρ = −0.259; P = .07). Conclusions: This is the first study comparing the diagnostic accuracy of PSGUS with histopathology and salivary secretion in detail. PSGUS and histopathology have a stronger association compared with PSGUS and parotid secretion. The specificity of PSGUS increases when the results are compared with plasma cell shift, LELs, and GCs, rather than with the focus score.