AIMS: To investigate the relationship between glucose-lowering medications and cardio-renal-liver-metabolic (CRLM) health.
METHODS: Two-sample Mendelian randomization (MR) was applied to assess the causal relationships between seven classes of glucose-lowering drugs and CRLM health outcomes. Genetic proxies for drug exposure were identified as cis-acting single nucleotide polymorphisms in the drug target genes, linked to both gene expression levels and glycosylated hemoglobin (HbA1c) or body mass index (BMI). Validation included colocalization and linkage disequilibrium analyses. A two-step MR strategy was employed to explore potential metabolite mediators.
RESULTS: Sulfonylureas were associated with a reduced heart failure (HF) risk (odds ratio [OR] = 0.38 per standard deviation change in glucose-lowering drug target perturbation equivalent to 1 unit of HbA1c lowering, P = 2.46E-04) but increased aspartate aminotransferase levels (OR = 1.39, P = 9.50E-07). Sodium-glucose cotransporter-2 inhibitors reduced the risk of acute myocardial infarction (AMI) (OR = 0.37, P = 1.36E-05), venous thromboembolism (VTE) (OR = 0.48, P = 1.89E-04), microalbuminuria (MA) (OR = 0.49, P = 1.65E-06), and increased estimated glomerular filtration rate (eGFR) (OR = 1.04, P = 2.98E-05). They also showed a potential protective effect against general liver disorders (OR = 0.41, P = 4.79E-04), metabolic dysfunction-associated steatotic liver disease (MASLD) (OR = 0.25, P = 0.008), and metabolic syndrome (MetS) (OR = 0.63, P = 5.40E-04). Thiazolidinediones (TZDs) reduced the risk of AMI (OR = 0.44, P = 4.31E-38), hypertension (HT) (OR = 0.60, P = 1.40E-43), MASLD (OR = 0.53, P = 5.61E-05), and MetS (OR = 0.66, P = 1.00E-20) but increased the risk of VTE (OR = 1.40, P = 2.53E-07), atrial fibrillation (OR = 1.61, P = 1.43E-09), and alcoholic liver disease (OR = 3.00, P = 6.56E-11). TZDs also negatively affected eGFR (OR = 0.97, P = 4.68E-05) and increased blood urea nitrogen levels (OR = 1.03, P = 2.52E-09). Glucagon-like peptide-1 receptor agonists demonstrated significant benefits for chronic kidney disease, driven by reductions in both glucose levels (OR = 0.11, P = 0.016) and BMI (OR = 0.20, P = 0.012), with potential cardiometabolic benefits from BMI reduction. Some mediating roles of metabolites have been identified (e.g., SGLT2 inhibitors mediate effects on MA via isoleucine and TZDs mediate effects on MASLD through lipid metabolites).
CONCLUSIONS: Glucose-lowering medications exert significant and heterogeneous effects on CRLM health, highlighting the need for personalized treatments and further investigations into their mechanisms and long-term impacts on CRLM outcomes.
