Link between Cortisol, Diabetes, and Cardiovascular Disease Questioned

A Mendelian randomization (MR) study found no evidence that genetically predicted cortisol levels play a role in cardiovascular or type 2 diabetes (T2D) risk. These findings were published in BMC Medicine.

Data from three genome-wide association studies (GWAS; CORNET: n=12,597; Shin: n=7824; Long: n=2049) recruited participants of European descent who were examined to identify single nucleotide polymorphisms (SNPs) that independently predicted cortisol. Genetic predictions of cortisol were associated with risk for ischemic heart disease (IHD), ischemic stroke, and T2D among seven repositories of genome-wide data. The relationship between cortisol and diseases were analyzed using four methods: inverse variance weighting (IVW), weighted median (WM), an MR-Egger approach, and MR Pleiotropy RE-Sidual Sum and Outlier (MR-PRESSO).

Cortisol levels were associated with 547 SNPs overall by the CORNET, Shin GWAS, and Long GWAS studies. .After removing correlated SNPs or those not reaching a sufficient level of significance, the final analysis included 29 SNPs (CORNET: n=6; Shin: n=5; Long: n=17). One SNP was removed from the Long study after further analysis.

Stratified by GWAS SNPs and data sources, there was no evidence that genetically predicted cortisol associated with IHD (P range, 0.18-0.95) or T2D (P range, 0.07-0.999) risk.

No evidence supported the relationship between SNPs from CORNET (P range, 0.46-0.84) or Long (P range, 0.47-0.90) and ischemic stroke risk. There was some evidence from the IVW (odds ratio [OR], 0.39; 95% CI, 0.24-0.64; P =.0002), WM (OR, 0.48; 95% CI, 0.25-0.93; P =.03), and MR-PRESSO (OR, 0.39; 95% CI, 0.21-0.73; P =.01) analyses that the 5 SNPs from the Shin study were significant predictors of risk. However, the sensitivity analysis did not support the relationship between genetically predicted cortisol with ischemic stroke risk (Cochran’s Q, 3.15; P =.53).

Genetically predicted cortisol was also not found to be associated with body mass index (BMI),  waist-to-hip ratio, glycated hemoglobin (HbA_1C),  glucose, insulin, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, or triglycerides (all P ³.10). Genetically predicted cortisol was associated with systolic (Cochran’s Q, 21.11; P =.001) and diastolic (Cochran’s Q, 12.87; P =.02) blood pressure, however, horizontal pleiotropy could not be ruled out for systolic blood pressure (I², 71.5%; P =.001) and was trending toward significance for diastolic blood pressure (I², 71.5%; P =.07).

In a reverse sensitivity analysis, the investigators used SNPs which predicted IHD, ischemic stroke, and T2D to assess whether they affected cortisol. No evidence was found to support an association.

This study could have included bias, as functionally relevant SNPs for cortisol are not well-established,  and only 1 of the considered 29 SNPs reached genome-wide significance (P <5´10^-6).

“Our study found no evidence that cortisol causes IHD, ischemic stroke, T2D, or CVD risk factors, but cannot exclude a small effect,” the researchers concluded. “This study also found no evidence that IHD, ischemic stroke, and T2D affect cortisol. The role of cortisol-related pathway to CVD is cast into doubt.”

Reference

Kwok MK, Kawachi I, Rehkopf D, Schooling CM. The role of cortisol in ischemic heart disease, ischemic stroke, type 2 diabetes, and cardiovascular disease risk factors: a bidirectional Mendelian randomization study. BMC Med. 2020;18(1):363. doi:10.1186/s12916-020-01831-3