Association between CKD and Kidney Stones

We have thoroughly reviewed Zhou et al.'s article titled “Association between Kidney Stones and CKD: A Bidirectional Mendelian Randomization Study,” published in JASN and wish to extend our sincere appreciation.1 Through the application of two-sample Mendelian randomization (MR), genetic correlation analysis, and observational studies, the authors have effectively examined the causal relationships, comorbid characteristics, and underlying biologic mechanisms linking kidney stones and CKD.1 Their rigorous and insightful methodologies provide valuable perspectives for clinicians managing patients with these conditions and offer new directions for future etiologic research and clinical interventions. Despite the significant findings presented in this study, we believe that there are areas where the effect of the study can be further enhanced. Below, we provide suggestions for improvement to offer readers a more comprehensive understanding of the study. First, although the authors reviewed relevant single-nucleotide polymorphism (SNP) loci and excluded rs77924615 on the basis of its association with the uromodulin-protein disulfide isomerase-like thioredoxin locus and CKD risk during the MR analysis of kidney stones (UK Biobank as exposure/FinnGen as outcome) on CKD risk—demonstrating a certain level of understanding of the no-pleiotropy assumption within the core MR framework—the evaluation of all SNP-disease associations remains incomplete. Using the LDstrait tool to thoroughly assess SNP associations with other traits (R2: 0.1 within a ±500,000 base pair window), we identified that certain instrumental variables, including rs10051765, rs2079742, and rs77924615, are associated with both CKD and kidney stones.2 However, we observed that the authors did not fully exclude SNPs with potential pleiotropic effects in their bidirectional MR analyses investigating the causal relationship between CKD and kidney stones. This limitation may lead to violations of the core assumption of MR (the no-pleiotropy assumption), thereby raising concerns about the reliability of the results. This observation aligns with the study's sensitivity analysis results, which highlight the robustness of the findings and the consistency of conclusions across different methodological approaches. To further enhance the reliability of the results, we recommend the following: (1) perform a comprehensive review of all SNP disease associations to eliminate potential confounders and (2) incorporate alternative sensitivity analysis methods, such as mixture of inverse-variance weighted and egger regression and other MR approaches, to validate the robustness of the findings.3 These additional steps will help ensure that the results are robust and free from potential biases introduced by pleiotropy. Second, the authors used multivariable Mendelian randomization (MVMR) by including cardiometabolic diseases (hypertension, diabetes, and obesity) as covariates to investigate the bidirectional causal relationship between CKD and kidney stones. However, the high genetic correlation among these cardiometabolic conditions may introduce multicollinearity, potentially compromising the robustness of the estimates.4 To mitigate this issue, we suggest adopting MVMR-constrained maximum likelihood and MVMR-Lasso methods.5 These approaches can help filter out redundant correlations and reduce the instability of regression coefficients, thereby ensuring more robust and reliable results. Finally, we commend the authors for their valuable contributions to this field. With further research, future studies are expected to provide deeper insights. We are honored to comment on this important work and anticipate more high-quality research outcomes in this area. In summary, the study by Zhou et al. is significant in elucidating the causal relationship between kidney stones and CKD. By refining SNP association assessments and incorporating advanced multivariable analytical methods, their findings will gain greater robustness and persuasiveness.