Causal associations between autoimmune diseases and cognitive impairment: A Mendelian randomization study.

Epidemiological evidence suggests that there is an association between systemic lupus erythematosus (SLE), type 1 diabetes (T1D), ankylosing spondylitis (AS), celiac disease (CD), and cognitive impairment (CI). However, the causal relationship between SLE, T1D, AS, CD, and CI remains unclear. The causal effects of the instrumental variables were analyzed using the random-effects inverse variance weighted method. Horizontal pleiotropy was examined by sensitivity analyses applying the weighted median method and the Mendelian randomization-Egger method. In order to avoid bias resulting from single-nucleotide polymorphisms, a leave-one-out analysis was employed. Our Mendelian randomization study identified the causality between SLE, CD, and declining cognitive performance (OR = 1.01, 95% CI: 1.001-1.030, P = .04; OR = 1.02, 95% CI: 1.02-1.04, P = .01), and the causality between CD and cognitive function (OR = 1.03, 95% CI: 1.01-1.05, P = .01). There was no evidence of a causative link between T1D, AS, and cognitive performance (OR = 1.00, 95% CI: 1.00-1.01, P = .23; OR = 1.01, 95% CI: 0.98-1.05, P = .46). And no causal relationship was found between SLE, T1D, AS and cognitive function (OR = 0.99, 95% CI: 0.96-1.02, P = .38; OR = 1.01, 95% CI: 0.97-1.05, P = .59; OR = 1.01, 95% CI: 0.98-1.01, P = .81). Our findings support causal relationships between SLE, CD, and declining cognitive performance, and between CD and cognitive function in European populations. No causal association was discovered between T1D, AS, and declining cognitive performance, or between SLE, T1D, AS and cognitive function.