Abstract
Background
Numerous studies have reported the close relationship between branched-chain amino acids (BCAA) and diabetic nephropathy (DN). Nevertheless, whether there is a genetically causal association between them remains profoundly elusive.
Methods
A two-sample Mendelian randomization (MR) analysis was performed using the large genome-wide association studies of the European population. The causal association was primarily evaluated by the inverse variance weighted (IVW) method. In addition, MR-Egger regression, weighted median, simple mode, and MR-weighted mode were also conducted as the supplemented methods. For sensitivity, Cochrane’s Q test, MR-Egger regression, and MR-PRESSO were employed to evaluate the heterogeneity and pleiotropy, respectively.
Results
According to the IVW method, no significant causal effect was measured between three BCAA and DN (valine: OR: 1.202, 95% CI: 0.714–2.023, P = 0.488; isoleucine: OR: 0.878, 95% CI: 0.400–1.924, P = 0.744; leucine: OR: 1.395, 95% CI: 0.686–2.839, P = 0.358; total BCAA: OR: 1.374, 95% CI: 0.703–2.685, P = 0.352). For reverse MR analysis, DN as an exposure factor also had no causal effect on BCAA (valine: OR: 1.004, 95% CI: 0.994–1.014, P = 0.412; isoleucine: OR: 0.999, 95% CI: 0.990–1.009, P = 0.910; leucine: OR: 1.001, 95% CI: 0.992–1.011, P = 0.802; total BCAA: OR: 1.002, 95% CI: 0.993–1.012, P = 0.628).
Conclusion
Our results first demonstrated no significant causal association between BCAA and DN at the genetic level.