Predicting Sjogren's syndrome in patients with recent-onset SLE

Methods. A total of 103 patients participating in a prospective cohort of recent-onset SLE were assessed for fulfilment of the American European Consensus Group criteria for SS using a three-phase approach: screening (European questionnaire, Schirmer-I test and wafer test), confirmation (fluorescein staining test, non-stimulated whole-salivary flow and anti-Ro/La antibodies) and lip biopsy. Anti-Ro/SSA and anti-La/SSB antibodies and RF were measured at entry into the cohort and at SS assessment. Results. Ninety-three females and 10 males were included. Mean age at lupus diagnosis was 25.9 +/- 8.9 years, and lupus duration at SS assessment was 30.9 +/- 9.1 years. SS was diagnosed in 19 (18.5%) patients, all female, and the patients were older at SLE diagnosis than patients without SS (30.8 +/- 9.3 vs 24 +/- 8.8 years, P = 0.004). Anti-Ro/SSA antibody was more common in SLE-SS patients (84% vs 55%, P = 0.02, LR + 1.53, 95% CI 1.14, 2.04). In the multivariate analysis, age epsilon 25 years and anti-Ro/SSA antibodies at SLE diagnosis were identified as predictors of SLE-SS, while the absence of anti-Ro/SSA, anti-La/SSB and RF seems to be protective (LR- 0.14, 95% CI 0.02, 0.95). Conclusion. The overlap of SLE and SS occurs in almost one-fifth of SLE patients and presents early during its evolution. SLE onset at age epsilon 25 years plus the presence of anti-Ro/SSA antibody at diagnosis are useful predictors, while the absence of anti-Ro/SSA, anti-La/SSB and RF identifies patients at lowest risk.