Diabetes is known to increase the risk of kidney stones, but the influence of antidiabetic drugs on this risk remains uncertain. Genetic instruments for antidiabetic drugs were identified as variants, which were associated with both the expression of genes encoding target proteins of drugs and glycated hemoglobin level (HbA1c). Here, we investigated the effect of antidiabetic drugs on kidney stones in a mendelian randomization (MR) framework, and further explore the potential effect on CaOx stone rat models induced by glyoxylic acid. Genetically proxied thiazolidinediones (PPARG agonists) significantly reduced the risk of kidney stones (OR=0.42; P=0.004) per 1-SD decrement in HbA1c, while no significant association was noted in sulfonylureas, SGLT2 inhibitors, or GLP-1 analogues. Other antidiabetic drugs were not analyzed due to unclear pharmacological targets or no identified instruments. Additionally, PPARG agonists pioglitazone ameliorated CaOx nephrocalcinosis in glyoxylic acid-induced rats. The summary-data-based MR (SMR) results showed that PPARG mRNA expression in blood or kidney was not associated with kidney stone risk, and thus we performed mediation MR of PPARG agonists, circulating metabolites, and kidney stones. Among 249 circulating metabolites, we identified an indirect effect of PPARG agonists on kidney stones through increasing phospholipids to total lipids ratio in very large VLDL, with a mediated proportion of 6.87% (P=0.018). Our study provided evidence that PPARG agonists reduced the risk of kidney stones partially via regulating lipid metabolism, and PPARG agonists may be a promising study subject in clinical studies for the prevention of kidney stones.
Keywords: Antidiabetic drugs; Kidney stones; Mendelian randomization; PPARG agonists.
Copyright © 2024. Published by Elsevier B.V.