Key points are not available for this paper at this time.
Abstract Background: Higher magnesium intake was linked to a lower risk of hepatocellular carcinoma (HCC). However, the relationship between blood magnesium level and HCC has not been fully characterized, especially among patients with liver cirrhosis who are at a higher risk for HCC. Methods: In the Mass General Brigham Biobank, we developed a new prospective cohort of 1,430 patients with liver cirrhosis without liver cancer history using the validated International Classification of Diseases codes. We used Cox proportional hazards models to generate hazard ratios (HRs) with 95% confidence intervals (CI) for incident HCC and used generalized estimating equations to compare changes in liver biomarkers according to baseline blood magnesium, adjusting for age, sex, race, lifestyles, body mass index, type 2 diabetes, model for end-stage liver disease score, and hepatitis infection. Results: During a median follow-up period of 4.26 years, 109 patients developed HCC. Magnesium deficiency (1.70 mg/dL; N = 158) was associated with a higher risk of HCC (HR = 1.93; 95% CI, 1.12–3.30) compared with magnesium sufficiency (≥1.70 mg/dL; N = 1282). This association remained robust in the 1-year lag analysis (HR = 2.18; 95% CI, 1.11–4.28) and in sensitivity analysis excluding patients with alcoholic liver disease (HR = 2.41; 95% CI, 1.23–4.74). Magnesium in the lowest quartile was associated with a faster increase in alanine transaminase (β = 4.35; 95% CI, 1.06–7.63), aspartate aminotransferase (β = 6.46; 95% CI, 0.28–12.6), direct bilirubin (β = 0.18; 95% CI, 0.01–0.35), and total bilirubin (β = 0.21; 95% CI, 0.03–0.39), compared with the highest quartile. Conclusions: Lower blood magnesium level is associated with higher HCC risk and unfavorable liver biomarker changes. Impact: If confirmed, our findings may potentially enable better identification of high-risk patients for HCC and inform better management strategies for liver cirrhosis.
Zhang et al. (Mon,) studied this question.