Can the diagnosis of primary Sjögrens syndrome be simplified?

Subjective sicca symptoms of the mouth (xerostomia) (12%) and eyes (keratoconjuctivitis sicca) (20%) are commonly reported in Iceland, especially among women. Sometimes symptoms are associated with autoimmune disorders. Most patients do not, however, fulfil European-US criteria for the diagnosis of primary (pSS) or secondary (sSS) Sjögren's Syndrome. Investigations necessary to confirm a diagnosis of pSS are time-consuming and costly. Objectives: This retrospective study aimed to determine which of the numerous criteria measured were best associated with the final diagnosis of pSS. Methods: Extended data from a consecutive series of 37 referred female patients with presumed pSS were examined for evaluation of data on autoimmune status, salivary flow, ophthalmic investigation, salivary gland scintigraphy and labial gland biopsy. Results: 12 (32%) patients met the EU-US-criteria for pSS, 7 (19%) had sSS. Significant differences between pSS and non-pSS were found for: SSA, SSB, ANA, unstimulated salivary flow, Schirmer and Rose Bengal tests, scintigraphy, and labial gland biopsy (p<0.001). No significant differences were found for either RF or stimulated salivary flow. High sensitivity (0.92) was found for ANA. SSA and SSB had high specificity as did ANA and Schirmer tests taken together. As regards scintigraphy, parotid uptake of 99mTcO4 15 min. post injection, and uptake ratio (UR) gave a high specificity of 0.80 and 0.79 respectively. UR and stimulated salivary flow taken together had a specificity of 0.94. Biopsy had a sensitivity of 0.50 but a specificity of 0.80. Conclusions: If both ANA and Schirmer are negative, pSS is unlikely. If both SSB and Schirmer are positive pSS is likely, also if both stimulated salivary flow and UR are positive. Positive SSA gives rather high suspicion of pSS. While no single test confirms a diagnosis of pSS, a structured approach to investigations could increase diagnostic reliablility without undue costs and patient inconvenience.