Document resource
Background To evaluate the classification accuracy of ultra-high frequency ultrasound (UHFUS) of labial salivary glands (LSGs) using both Grey Scale (GS) and colour Doppler (CD). We also investigated its role as an alternative to the highest-weighted classification criteria, its classification performance in combination with SSA status, its contribution to patient stratification and age-related classification variability.Methods A total of 270 patients with suspected Sjögren’s disease (SjD) (136 SjD, 134 Sicca) underwent UHFUS. In addition to OMERACT GS (modified to LSG), CD and composite GS+CD Scores, distinct GS morphological lesions were assessed. Diagnostic accuracy parameters were calculated, and additional analyses included correlation, logistic regression and cluster analysis.Results The optimal classification cut-off was GS ≥2, yielding 66.2% sensitivity and 84.3% specificity. GS Score 3 was highly specific (98.5%), while very hypoechoic areas were exclusively observed in SjD. CD ≥2 showed accuracy comparable to GS, and GS+CD ≥3 enhanced accuracy. Incorporating GS into the American College of Rheumatology/European League Against Rheumatism criteria in place of biopsy achieved excellent accuracy (area under the curve 0.945). Using GS, GS+CD and SSA status, more than 75% of patients could be classified with an accuracy of 94%. Cluster analysis suggests different phenotypes: in Sicca, GS positivity indicated non-immune inflammatory alterations or sialoadenosis; in SjD, GS negativity defined low-grade inflammation. Classification accuracy of GS varied with age, whereas CD remained stable.Conclusions UHFUS of LSGs is a promising diagnostic tool in SjD when biopsy is not feasible. This study provides novel insights into GS lesions, Doppler assessment, phenotypes and age-related changes of ultrasound.