Diagnostic performance of tear function tests in Sjogren's syndrome patients

Abstract

To evaluate the diagnostic performance of the tests included in primary Sjogren's syndrome (SS-I) diagnostic criteria (Schirmer I, break-up time, vital dye staining) and to compare them with other examinations related to the ocular surface status. Clinical and cytological data were collected from 177 patients (62 SS-1, 56 non-SS autoimmune diseases, 59 Sicca syndrome). Tear tests included: a validated questionnaire on symptoms, Schirmer I, Jones test, Ferning test, BUT, corneal aesthesiometry, tear clearance test, lissamine green staining, impression conjunctival cytology. Data were statistically evaluated and sensitivity, specificity, likelihood ratio (LR+), receiver-operating characteristics (ROC) curves were calculated for each test. Data showed a poor diagnostic performance of Schirmer test I (sensitivity 0.42; specificity 0.76; LR+1.75) and BUT (sensitivity 0.92; specificity 0.17; LR+1.11) (area under the curve in ROC analysis <0.58). Validated subjective symptoms questionnaire (sensitivity 0.89; specificity 0.72; LR+3.18), Jones test (sensitivity 0.60; specificity 0.88; LR+5), corneal aesthesiometry (sensitivity 0.80; specificity 0.67; LR+2.42), and tear clearance test (sensitivity 0.63; specificity 0.84; LR+3.93), all exhibited a high diagnostic performance (area under the curve in the ROC analysis always >0.70). Lissamine green staining exhibited the best performance (sensitivity 0.63; specificity 0.89; LR+5.72) but the result could be distorted by an incorporation bias. Our data suggest to implement the items for ocular signs and symptoms contained in many SS-I diagnostic criteria with the use of a validated questionnaire, performance of Jones test, corneal aesthesiometry measurement, and tear clearance rate evaluation.