Mediterranean diet and risk of Sjögren's syndrome.

Aleksander Machowicz, Isaac Hall, Paola de Pablo, Saaeha Rauz, Andrea Richards, Jon Higham, Ana Poveda-Gallego, Fumiaki Imamura, Simon J Bowman, Francesca Barone, Benjamin A Fisher

Non-genetic risk factors for Sjögren's syndrome (SS) are poorly understood. Adherence to a Mediterranean diet has been associated with reduction in other autoimmune diseases. We examined the association of Mediterranean diet with SS. New patients attending a single centre warranting investigation for primary SS (pSS) were recruited into the Optimising Assessment in Sjögren's Syndrome cohort established in Birmingham, UK (2014-2018). Participants were classified into pSS and non-SS sicca, considered as cases and non-cases, respectively, and asked to complete an optional food frequency questionnaire on their diet before onset of symptoms. A semi-quantitative Mediterranean diet score (MDS) was calculated (possible range=0 to 18). Using multivariate logistic regression, corrected for energy intake, body-mass index, sex, age, symptom duration, and smoking status, we examined the association of MDS with SS. Dietary data were available for 133/243 (55%) eligible patients (n=82 pSS and n=51 sicca). In the adjusted model, a higher total MDS (mean ± SD, 9.41±2.31 points) was associated with lower odds of pSS (OR 0.81, 95% CI 0.66-0.99; p=0.038) per one unit of MDS. Among MDS components, the strongest association was seen with fish with OR 0.44 (95% CI 0.24-0.83; p=0.01) in the comparison between <1 portion/week and 1 to 2.5 portions/week. Higher galactose, vitamin A-retinol-equivalents and vitamin C showed associations with lower odds of pSS in multivariate analysis, where the association of vitamin C was attenuated when adjusted for MDS. When adjusted for potential confounders, adherence to the Mediterranean diet was associated with lower likelihood of having pSS.