Causal analysis of serum polyunsaturated fatty acids with juvenile idiopathic arthritis and ocular comorbidity

To investigate the causal effects of plasma Polyunsaturated fatty acids (PUFAs) on the risk of juvenile idiopathic arthritis (JIA) and ocular comorbidity through Mendelian randomization (MR) analysis.

Genetic variants (formerly single nucleotide polymorphisms, SNPs) that are strongly associated with PUFAs levels (P < 5×10^-8) were selected as instrumental variables. Summary-level MR was performed with outcome estimates for JIA (n = 31,142) and JIA associated iridocyclitis (n = 94,197). The inverse variance-weighted (IVW) method was employed as the main approach to combine the estimation for each SNP. Two set of models with summary statistics were conducted and multiple sensitivity analyses were applied for testing of pleiotropic bias.

In model 1, genetically predicted n-6 PUFAs linoleic acid (LA) and arachidonic acid (AA) were associated with lower and higher risk of JIA associated iridocyclitis using IVW (OR_LA = 0.940, 95% CI: 0.895-0.988, P = 0.015; OR_AA = 1.053, 95% CI: 1.007-1.101, P = 0.024). No such association was observed between each plasma PUFAs and JIA susceptibility (P > 0.05). In further MR analysis, results from model 2 also showed a consistent trend. Besides, multiple sensitivity analyses revealed that there was no obvious evidence for unknown pleiotropy (P > 0.05).

Our MR study provides genetic evidence on the possible causality that plasma LA level might protect against JIA associated iridocyclitis, whereas AA was responsible for opposite effect.