A bidirectional Mendelian randomization study of spleen volume and Crohn disease.

Abstract

In observational studies, there has been an association found between spleen volume and Crohn disease. We conducted a two-way, two-sample Mendelian randomization analysis to determine whether these associations have a causal relationship. Single nucleotide polymorphisms (P < 5 × 10-8) were used as instrumental variables for spleen volume and Crohn disease. Estimates of the genetic associations between spleen volume and Crohn disease were obtained from the Integrative Epidemiology Unit, European Bioinformatics Institute, UK Biobank, and FinnGen databases. Analysis was performed using MR-Egger regression, weighted median estimator, inverse variance weighted, simple model, and weighted model. Genetically predicted spleen volume was found to be associated with Crohn disease. In the IEU database, the odds ratios (ORs) for Crohn disease caused by spleen volume were 1.237 (95% CI, 1.056-1.417, P = .021), and the ORs for spleen volume caused by Crohn disease were 1.015 (95% CI, 0.985-1.044; P = .049). In the EBI database, the ORs for Crohn disease caused by spleen volume were 1.292 (95% CI, 1.120-1.463, P = .003), and the ORs for spleen volume caused by Crohn disease were 1.026 (95% CI, 1.005-1.046; P = .013). Results from the UKB and FinnGen databases showed no causal relationship between the two. The summary results showed that Crohn disease caused an increase in spleen volume, with ORs of 1.009 (95% CI, 1.000-1.018; P = .047). This study provides evidence for a mutual causal relationship between spleen volume and an increased risk of Crohn disease.