To the Editor: We read with great interest the article by Sakaguchi et al.1 on a cohort study of 142,555 hemodialysis patients investigating the relationship of serum magnesium (Mg) with mortality risk as well as the comment by Courivaud and Davenport.2 The authors observed a J-shaped association of Mg serum levels and mortality. They argue that the increasing risk of cardiovascular disease–associated mortality in the highest Mg sextile might be attributable to some extent to parathyroid hormone (PTH) oversuppression. Courivaud and Davenport2 therefore advise caution, as oversuppressing PTH by mild hypermagnesemia may lead to adynamic bone disease. Recently, we could demonstrate in vitro that PTH suppression by Mg is mainly observed when moderately low calcium (Ca) concentrations are present.3 Mean Ca levels of the Japanese patients were 9.3±0.9 mg/dl, and hence well in the normal range. Further, clinical data of the CALcium acetate MAGnesium carbonate (CALMAG) study and its post hoc evaluation4 revealed that mean serum Mg levels of 3.11 mg/dl (1.279 mmol/l) in the CaAc/MgCO3 group were not paralleled by changes of PTH and/or bone turnover markers. The serum Mg levels found in that study were equivalent to the levels at the lower end of the highest sextile in the Japanese cohort. Even though we fully agree with the authors that prospective studies on Mg supplementation for investigating its influence on bone are highly needed to fully answer this question, we feel that a general relationship of increased serum Mg levels and adynamic bone disease is not supported by existing studies.