Is hypomagnesemia associated with using proton pump inhibitors?

Proton pump inhibitors (PPIs) have been very useful in various gastrointestinal diseases. However, hypomagnesemia was reported in patients taking PPIs, and cross-sectional or retrospective studies for a causal relationship between PPIs and hypomagnesemia were carried out. Until now, this question remains unresolved although a recent meta-analysis study concluded that PPI use may increase the risk of hypomagnesemia [1]. In the past issue of this Journal, Park et al [2] added a controversy to the association between hypomagnesemia and the PPI use. This study was interesting because it was performed on a relatively homogeneous population, who was undergoing percutaneous coronary intervention and whose basal serum magnesium level was normal. The authors made an effort to demonstrate the relationship through various subgroup analyses. However, in this study protocol, the definition of PPI group is not sufficient for detailed history about taking PPIs, as it was uncertain when PPI usage began. The authors may need to show more data regarding previous PPI usage, serum magnesium level before starting PPIs, and the proportion of patients with previous PPI usage before enrollment. Despite previous PPI usage, some patients in the PPI group may consistently maintain magnesium levels in normal range during the study period. Furthermore, in the previous reports [3,4], the association between PPI use and a low serum magnesium level was observed among population with higher prevalence of hypomagnesemia, ranging from 6% to 24%. The authors already pointed out a very low incidence of hypomagnesemia in the study population. Unlike calcium balance, magnesium is not readily mobilized from the bone stores. However, if negative magnesium balance continues, the bone stores help to