2034: What Is the Best Initial Screening for Suspected Sjögren's Disease?

Katja Perdan Pirkmajer, Jelka Kramarič, matija Tomšič and Alojzija Hocevar, University Medical Centre Ljubljana, Ljubljana, Slovenia

Background/Purpose: During the COVID-19 pandemic the routine diagnostic procedure in Sjögren's disease (SS) was highly hampered.

The aim of our study was to determine in a cohort of well-defined sicca patients the best screening tool for SS, obviating, when possible, the initial exposure to potentially infectious body fluids (eg. saliva, tears) and/or invasive procedures (lip biopsy) We also evaluated the performance of the newly proposed 5-item Sjögren's Syndrome Screening Questionnaire (SSSQ) derived from SICCA study data, that was shown to better differentiate SS from non-SS sicca patients compared to conventional sicca questionnaire¹.

Methods: Our cohort consisted of 282 subjects (109 SS and 173 non-SS sicca patients) who completed a standardized diagnostic procedure for SS based on the ACR/EULAR 2016 classification criteria² (standard sicca questions, Schirmer test, Rose Bengal test, unstimulated salivary flow test, immunoserological tests (anti-SSA) and lip biopsy) between Jan/2016 to Dec/2018. In addition, major salivary gland ultrasound (SGUS) was performed in each subject at the time of diagnostic procedure³. Afterwards all subjects were contacted by phone to fulfil the new SSSQ. Descriptive statistics and ROC curve analysis, calculating area under the curve (AUC) were used to analyse data.

Results: The results of standard sicca questions, SSSQ, Schirmer test, Rose Bengal test, unstimulated salivary flow test, SGUS, anti-SSA and lip biopsy were positive in 90.4%, 34.0%, 50.4%, 30.1%, 63.8%, 19.1%, 27.7% and 33.3%, respectively. The Table 1 shows the distribution of positive tests according to final diagnosis (SS vs. non-SS). The new SSSQ had a sensitivity of 41.3% and a specificity of 70.5% for SS (in comparison, the traditional sicca questions were 85.3% sensitive but only 6.4% specific for SS). Table 2 and Figure 1 show the performance of combinations of various tests that were easily available in the daily clinical practice for supporting SS diagnosis.

Conclusion: Our data suggest, that screening with a combination of anti-SSA and SGUS provides the highest AUC and specificity for SS, while a combination of SSSQ, anti-SSA and Schirmer tear test yielded the best sensitivity among the compared combinations.

References: Yu K, et al. J Clin Rheumatol 2021; doi: 10.1097/RHU.0000000000001760b
Shiboski CH, et al. Arthritis Rheumatol 2017;69:35–45. 10.1002/art.39859
H. Hocevar A, et al. Rheumatology 2005; 10.1093/rheumatology/keh588

Table 1. Distribution of positive tests based on final diagnosis

Table 2. The performance of different combinations tests in SS screening.
Legend: PLR Positive Likelihood Ratio; NLR Negative Likelihood Ratio; PPV Positive Predictive Value ; NPV Positive Predictive Value; AUC area under the ROC curve;


Figure 1. ROC curve analysis of different test combinations for diagnosing SS.
Disclosures: K. Perdan Pirkmajer, None; J. Kramarič, None; m. Tomšič, None; A. Hocevar, None.