内科

目的采用两样本孟德尔随机化（MR）方法，探讨血清镁浓度与高血压之间的因果关系。方法基于GWAS数据库汇总数据，以血清镁浓度为暴露因素、高血压为结局变量，筛选满足MR核心假设的单核苷酸多态性（SNP）作为工具变量。采用逆方差加权（IVW）法、加权中位数估计（WME）法、MR-Egger回归法、加权模式法和简单模式法5种MR分析方法进行因果效应估计，并以IVW法作为主要分析方法。采用MR-Egger截距检验评估水平多效性，采用留一法进行敏感性分析。结果共筛选出4个SNP作为工具变量纳入分析。IVW法（主要分析方法）结果显示，血清镁浓度与高血压之间无统计学意义上的因果关系（OR=1.110，95%CI：0.825～1.493，P=0.491）；MR-Egger回归法、简单模式法及加权模式法的估计方向与IVW法相同（均P>0.05）。WME法的分析结果显示二者之间存在统计学关联（OR=1.135，95%CI：1.014～1.270，P=0.028）。在敏感性分析中，MR-Egger截距检验未发现存在水平多效性的统计学证据（截距=-0.002，P=0.467）；留一法分析显示，依次剔除任一SNP后，IVW法的主要估计结果保持稳定（95%CI均跨越无效线）。结论本研究主要分析方法（IVW法）未发现血清镁浓度与高血压的因果关联，但WME法提示存在潜在关联。由于不同方法结果存在异质性且工具变量数量有限，当前证据不足以支持二者存在明确的因果关联，未来需更大样本量验证。

Objective To investigate the causal relationship between serum magnesium concentration and hypertension with a two-sample Mendelian randomization (MR) approach. Methods Based on summary-level data from GWAS, serum magnesium concentration was used as the exposure factor and hypertension as the outcome variable, and single nucleotide polymorphisms (SNPs) satisfying the core assumptions of MR were selected as instrumental variables. Five MR analysis methods—inverse-variance weighted (IVW), weighted median estimator (WME), MR-Egger regression, weighted mode, and simple mode—were employed to estimate causal effects, with the IVW method serving as the primary analysis. The MR-Egger intercept test was used to assess horizontal pleiotropy, and leave-one-out method was performed for sensitivity analysis. Results A total of 4 SNPs were selected as instrumental variables and included in the analysis. Result of the IVW method (primary analysis) showed no statistically significant causal relationship between serum magnesium concentration and hypertension (OR=1.110, 95%CI: 0.825-1.493, P=0.491); the estimated directions of the MR-Egger regression, simple mode, and weighted mode methods were consistent with that of the IVW method (all P>0.05). The WME analysis results showed a statistically significant association between the two (OR=1.135, 95%CI: 1.014-1.270, P=0.028). In the sensitivity analysis, the MR-Egger intercept test provided no statistical evidence of horizontal pleiotropy (intercept=-0.002, P=0.467); the leave-one-out analysis demonstrated that the main estimates from the IVW method remained stable after sequentially removing each SNP individually (all 95%CIs crossed the null line). Conclusion The primary analysis of this study (IVW method) found no causal association between serum magnesium concentration and hypertension, whereas the WME method suggested a potential causal association. Given the heterogeneity of results across different methods and the limited number of instrumental variables, the current evidence is insufficient to support a definite causal association between the two; therefore, further studies with larger sample sizes are warranted for validation in the future.