Diabetes and impaired glycemic control as risk factors for stroke outcome - mendelian randomization

Diabetes and impaired glycemic control are risk factors for stroke. It is unclear whether diabetes and impaired glycemic control lead to worse functional outcome after stroke.

Aims

We will use genetic variants identified in genome-wide association studies that influence susceptibility to diabetes, hemoglobin A1 c levels (not the hemoglobin influencing SNPs), fasting glucose, fasting insulin and BMI. Only SNPS that are not in linkage disequilibrium will be used. Variants that influence glycemic control mediated through BMI will be removed from analyses of T2D, fasting insulin and HbA1c (e.g. FTO locus).
The GISCOME cohort will be used to analyze functional outcome as determined using the modified Rankin scale. The primary analysis will focus on good functional outcome (modified Rankin Scale 0-2). Secondary analysis will focus on excellent functional outcome (mRS 0-1).
We will perform SNP risk factor associations (eg SNP-diabetes) and SNP stroke functional outcome associations. We will first use an inverse-variance weighted method and conduct additional complementary analysis using the weighted median and MR-Egger regression methods and the MR-PRESSO to control for biases due to pleiotropy.
A Bonferroni correction will be performed.