Epidemiological studies on magnesium intake and primary liver cancer (PLC) are scarce, and no prospective studies have examined the associations of magnesium intake with PLC incidence and mortality. We sought to clarify whether higher magnesium intake from diet and supplements was associated with lower risks of PLC incidence and mortality in the US population. Magnesium intake from diet and supplements was evaluated through a food frequency questionnaire in a cohort of 104,025 participants. Cox regression was employed to calculate hazard ratios for PLC incidence and competing risk regression was employed to calculate subdistribution hazard ratios for PLC mortality. Restricted cubic spline regression was employed to test nonlinearity. We documented 116 PLC cases during 1,193,513.5 person‐years of follow‐up and 100 PLC deaths during 1,198,021.3 person‐years of follow‐up. Total (diet + supplements) magnesium intake was found to be inversely associated with risks of PLC incidence (hazard ratio_{tertile 3 vs . 1}: 0.44; 95% confidence interval: 0.24, 0.80; p _trend = 0.0065) and mortality (subdistribution hazard ratio_{tertile 3 vs . 1}: 0.37; 95% confidence interval: 0.19, 0.71; p _trend = 0.0008). Similar results were obtained for dietary magnesium intake. Nonlinear inverse dose–response associations with PLC incidence and mortality were observed for both total and dietary magnesium intakes (all p _nonlinearity < 0.05). In summary, in the US population, a high magnesium intake is associated with decreased risks of PLC incidence and mortality in a nonlinear dose–response manner. These findings support that increasing the consumption of foods rich in magnesium may be beneficial in reducing PLC incidence and mortality.

中文翻译：

---

前列腺癌，肺癌，大肠癌和卵巢癌筛查试验中的镁摄入量和原发性肝癌的发病率和死亡率。

关于镁摄入量和原发性肝癌（PLC）的流行病学研究很少，并且没有前瞻性研究检查镁摄入量与PLC发生率和死亡率之间的关系。我们试图弄清饮食和补品中较高的镁摄入量是否与美国人群中PLC发生率和死亡率降低的风险有关。通过饮食频率问卷调查了104,025名参与者的饮食和补品中镁的摄入量。使用Cox回归计算PLC发生率的危险比，使用竞争风险回归计算PLC死亡率的子分布危险比。限制性三次样条回归用于测试非线性。我们记录了1,193,513.5人年的随访期间116例PLC病例，以及1,198,021例100例PLC死亡。随访3人年。发现总（饮食+补充剂）镁摄入量与PLC发生风险呈负相关（风险比）_{三分位3 VS。1}：0.44；95％置信区间：0.24，0.80; p_趋势= 0.0065）和死亡率（子分布危险比_{三分位3 VS 1。}：0.37; 95％置信区间：0.19，0.71; p_趋势= 0.0008）。饮食中镁的摄入量也获得了相似的结果。在总镁和膳食镁摄入量中均观察到非线性的剂量-反应与PLC发生率和死亡率的负相关（所有p_非线性） <0.05）。总之，在美国人群中，镁的摄入量高，以非线性的剂量反应方式降低了PLC发生率和死亡率的风险。这些发现支持增加含镁食品的摄入量可能有助于降低PLC的发病率和死亡率。