A bidirectional Mendelian randomization study about the role of morning plasma cortisol in attention deficit hyperactivity disorder

Abstract

Context: Cortisol, a hormone regulated by the hypothalamic-pituitary-adrenal (HPA) axis, has been linked to attention deficit hyperactivity disorder (ADHD). The nature of the relationship between cortisol and ADHD, and whether it is causal or explained by reverse causality, remains a matter of debate.

Objective: This study aims to evaluate the bidirectional causal relationship between morning plasma cortisol levels and ADHD.

Methods: This study used a bidirectional 2-sample Mendelian randomization (MR) design to analyze the association between morning plasma cortisol levels and ADHD using genetic information from the authoritative Psychiatric Genomics Collaboration (PGC) database (n = 55,347) and the ADHD Working Group of the CORtisol NETwork (CORNET) Consortium (n = 12,597). MR analyses were employed: inverse variance weighting (IVW), MR-Egger regression, and weighted medians. OR values and 95% CI were used to evaluate whether there was a causal association between morning plasma cortisol levels on ADHD and ADHD on morning plasma cortisol levels. The Egger-intercept method was employed to test for level pleiotropy. Sensitivity analysis was performed using the "leave-one-out" method, MR pleiotropy residual sum, and MR pleiotropy residual sum and outlier (MR-PRESSO).

Results: Findings from bidirectional MR demonstrated that lower morning plasma cortisol levels were associated with ADHD (ADHD-cortisol OR = 0.857; 95% CI, 0.755-0.974; P = 0.018), suggesting there is a reverse causal relationship between cortisol and ADHD. However, morning plasma cortisol levels were not found to have a causal effect on the risk of ADHD (OR = 1.006; 95% CI, 0.909-1.113; P = 0.907), despite the lack of genetic evidence. The MR-Egger method revealed intercepts close to zero, indicating that the selected instrumental variables had no horizontal multiplicity. The "leave-one-out" sensitivity analysis revealed stable results, with no instrumental variables significantly affecting the results. Heterogeneity tests were insignificant, and MR-PRESSO did not detect any significant outliers. The selected single-nucleotide polymorphisms (SNPs) F were all >10, indicating no weak instrumental variables. Thus, the overall MR analysis results were reliable.

Conclusion: The study findings suggest a reverse causal relationship between morning plasma cortisol levels and ADHD, with low cortisol levels associated with ADHD. No genetic evidence was found to support a causal relationship between morning plasma cortisol levels and the risk of ADHD. These results suggest that ADHD may lead to a significant reduction in morning plasma cortisol secretion.

Keywords: Mendelian randomization; attention deficit hyperactivity disorder; cortisol; hypothalamic-pituitary-adrenal; single-nucleotide polymorphism.