In order to evaluate their sensitivity and specificity, parotid sialography (PS) and labial salivary gland biopsy were performed in 150 putative patients with Sjögren's syndrome (SS) (74 with SS alone and 76 with SS associated with other well-defined disorders). Although PS and LSGB strongly correlated to each other (Kendall's Tau 0.47, z = 7.78, p less than 0.0001), PS confirmed the diagnosis of SS in a significantly higher prevalence of patients with respect to LSGB (63% vs 37%, chi 2 = 18.25, p less than 0.001). When the sensitivity and specificity of PS and LSGB were calculated, PS showed a better sensitivity, and an only slightly lower specificity than LSGB (86% vs 70% and 78% vs 86% respectively). In conclusion, our study indicates that PS has been unduly overlooked in the past decades as a diagnostic tool for the salivary component of SS, and should be reconsidered as a potentially useful and relatively non-invasive procedure in the approach of patients with this disease.