作者
Masaomi Tajimi,Tomoko Nii,Nobuyuki Takahashi,Toshinori Yamamoto,Shinich Koizumi
摘要
Context: Use of proton pump inhibitor (PPI), a potent gastric acid suppressant, may affect mineral absorption from gastrointestinal tract.Hypomagnesaemia in the setting of PPI therapy has been documented in case reports.The objective of this study was to examine the association between serum magnesium level as well as hypomagnesaemia and PPI use.Method: A retrospective data analysis of hospitalized adults aged 50 years or older (N= 1109) during 2008, including PPI users (N=381) and PPI non-users (N=728), was conducted.Among all study patients, 280 patients (25%) had magnesium levels measured during admissions, and the first laboratory value was used for data analysis.Hypomagnesaemia (N=51) is defined as magnesium levels less than 1.7 mg/dL.Multiple linear and logistic regression analyses were used to assess the association between PPI use and serum magnesium level.Results: PPI users (mean±SD: age 71.3±12.5 years; 61% female) had a mean magnesium level 1.89±0.34mg/dL, and PPI non-users (mean age 72.0±13.0years; 61% female) had the mean level 1.96±0.32mg/dL; p=.075.The linear regression model revealed that PPI use was associated with lower serum magnesium levels (coefficient β=-0.09;p=0.029;R 2 = 0.103) after adjusting for age; serum potassium, creatinine, bicarbonate, calcium, and albumin levels; sex; history of diabetes, congestive heart failure, and coronary artery disease; and uses of the following drugs: atypical antipsychotics, diuretics, supplementation of potassium, magnesium, and calcium, antihistamines, antidepressants, inhaled corticosteroids and β2 agonist bronchodilators, laxatives, and narcotics.PPI therapy was associated with an increased risk of hypomagnesaemia occurrence (crude odds ratio (OR) =2.48 [95% confidence interval (CI): 1.34-4.57]),and the association remained significant after adjusting for the above confounders (adjusted OR=2.78 [95% CI: 1.34-5.75]).CONCLUSION: PPI use was significantly associated with lower serum magnesium levels and hypomagnesaemia occurrences from a population of hospitalized adult patients in a community.Further studies are needed to confirm our findings.