Observational studies have identified a possible link between thyroid function and diabetic microangiopathy, specifically in diabetic kidney disease (DKD) and diabetic retinopathy (DR). However, it is unclear whether this association reflects a causal relationship.
To assess the potential direct effect of thyroid characteristics on DKD and DR based on Mendelian randomization (MR).
We conducted an MR study using genetic variants as an instrument associated with thyroid function to examine the causal effects on DKD and DR. The study included the analysis of 4 exposure factors associated with thyroid hormone regulation and 5 outcomes. Genomewide significant variants were used as instruments for standardized freethyroxine (FT4) and thyroid-stimulating hormone (TSH) levels within the reference range, standardized free triiodothyronine (FT3):FT4 ratio, and standardized thyroid peroxidase antibody (TPOAB) levels. The primary outcomes were DKD and DR events, and secondary outcomes were estimated glomerular filtration rate (eGFR), urinary albumin-to-creatinine ratio (ACR) in diabetes, and proliferative diabetic retinopathy (PDR). Satisfying the 3 MR core assumptions, the inverse-variance weighted technique was used as the primary analysis, and sensitivity analysis was performed using MR-Egger, weighted median, and MR pleiotropy residual sum and outlier techniques.
All outcome and exposure instruments were selected from publicly available GWAS data conducted in European populations. In inverse-variance weighted random-effects MR, gene-based TSH with in the reference range was associated with DKD (OR 1.44; 95%CI 1.04, 2.41; P = 0.033) and eGFR (β: -0.031; 95%CI: -0.063, -0.001; P = 0.047). Gene-based increased FT3:FT4 ratio, decreased FT4 with in the reference range were associated with increased ACR with inverse-variance weighted random-effects β of 0.178 (95%CI: 0.004, 0.353; P = 0.046) and -0.078 (95%CI: -0.142, -0.014; P = 0.017), respectively, and robust to tests of horizontal pleiotropy. However, all thyroid hormone instruments were not associated with DR and PDR at the genetic level.
In diabetic patients, an elevated TSH within the reference range was linked to a greater risk of DKD and decreased eGFR. Similarly, decreased FT4 and an increased FT3:FT4 ratio within the reference range were associated with increased ACR in diabetic patients. However, gene-based thyroid hormones were not associated with DR, indicating a possible pathway involving the thyroid-islet-renal axis. However, larger population studies are needed to further validate this conclusion.