Genetically predicted plasma cortisol and common chronic diseases: A Mendelian randomization study

Abstract

AbstractObjectiveCushing's syndrome is characterized by hypercortisolaemia and is frequently accompanied by comorbidities such as type 2 diabetes, hypertension, osteoporosis, depression and schizophrenia. It is unclear whether moderate but lifelong hypercortisolaemia is causally associated with these diseases in the general population. We aimed to address this research gap using a Mendelian randomization approach.MethodsWe used three cortisol‐associated genetic variants in the SERPINA6/SERPINA1 region as genetic instruments in a two‐sample, inverse‐variance‐weighted Mendelian randomization analysis. We obtained summary‐level statistics for cortisol and disease outcomes from publicly available genetic consortia, and meta‐analysed them as appropriate. We conducted a multivariable Mendelian randomization analysis to assess potential mediating effects.ResultsA 1 standard deviation higher genetically predicted plasma cortisol was associated with greater odds of hypertension (odds ratio: 1.12; 95% confidence interval [CI]: 1.05–1.18) as well as higher systolic blood pressure (mean difference [MD]: 0.03 SD change; 95% CI: 0.01–0.05) and diastolic blood pressure (MD: 0.03 SD change; 95% CI: 0.01–0.04). There was no evidence of association with type 2 diabetes, osteoporosis, depression and schizophrenia. The association with hypertension was attenuated upon adjustment for waist circumference, suggesting potential mediation through central obesity.ConclusionThere is strong evidence for a causal association between plasma cortisol and greater risk for hypertension, potentially mediated by obesity.