Serum magnesium and proton-pump inhibitors use: a cross-sectional study

Magnésio sérico e uso de inibidores de bomba de prótons: estudo transversal

ORIGINAL ARTICLE

Serum magnesium and proton-pump inhibitors use: a cross-sectional study^*

Magnésio sérico e uso de inibidores de bomba de prótons: estudo transversal

Gustavo Adolpho Moreira Faulhaber^I,II; Bruna Maria Ascoli^I; Adriano Lubini^II; Márcio Mossmann^II; Gabriela Rossi^II; Guilherme Geib^II; Tânia Weber Furlanetto^I,II

^IPostgraduate Program in Medicine, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, RS, Brazil

^IIInternal Medicine Division, Hospital de Clínicas de Porto Alegre, UFRGS, Porto Alegre, RS, Brazil

^{Corresponding author}

ABSTRACT

OBJECTIVE: The aim of this study was to evaluate the association of serum magnesium levels with proton pump inhibitors (PPIs) use and other factors.

METHODS: This was a cross-sectional study of 151 patients admitted with acute diseases in the Internal Medicine Division of the Hospital de Clinicas de Porto Alegre, after the exclusion of conditions that are commonly associated with hypomagnesemia: diarrhea; vomiting; chronic alcohol use; severely uncompensated diabetes mellitus; and chronic use of laxatives, diuretics or other drugs causing magnesium deficiency.

RESULTS: All patients had normal serum magnesium levels. Serum albumin and creatinine levels were positively associated with serum magnesium levels, after adjusting for confounders. There was no difference between mean serum magnesium levels of PPI users and non-users, nor between men and women; there was also no correlation among age, serum phosphorus, and potassium levels with serum magnesium levels. Limitations of this study include the absence of an instrument for measuring adherence to PPI use and the sample size.

CONCLUSION: The association of PPI use and hypomagnesemia is uncommon. Congenital defects in the metabolism of magnesium may be responsible for hypomagnesemia in some patients using this drug class.

Keywords: Magnesium; Proton pump inhibitors

RESUMO

OBJETIVO: O objetivo desse estudo foi verificar a associação do nível sérico do magnésio com o uso de inibidores de bomba de prótons (IBP) e outros fatores.

MÉTODOS: Realizou-se estudo transversal com 151 pacientes admitidos com doenças agudas no serviço de medicina interna do Hospital de Clínicas de Porto Alegre. Foram excluídos aqueles pacientes com condições usualmente relacionadas à hipomagnesemia: diarréia; vômitos; diabéticos agudamente descompensados; uso crônico de laxantes, álcool, diuréticos ou outros fármacos relacionados.

RESULTADOS: Todos os pacientes apresentaram níveis normais de magnésio. Albumina e creatinina sérica se associaram positivamente com os níveis de magnésio sérico, após ajuste para fatores confundidores. Não houve diferença no nível sérico de magnésio em usuários ou não-usuários de IBP ou entre homens e mulheres. Não houve correlação com idade, nível sérico de fósforo e potássio. As principais limitações desse estudo foram a ausência de instrumento para medir a adesão aos IBPs e o tamanho da amostra.

CONCLUSÃO: A associação do uso de IBP e hipomagnesemia é rara. Defeitos congênitos no metabolismo do magnésio devem ser responsáveis pelo surgimento de hipomagnesemia em usuários de dessa classe de fármacos.

Palavras-chave: Magnésio; Inibidores da bomba de prótons

Introduction

The most potent drugs available to reduce gastric acid secretion are the proton pump inhibitors (PPIs).¹ PPIs selectively and irreversibly inhibit the gastric H ⁺/K⁺ ATPase. As this is the final step of acid secretion, this drug class has proved to be more effective than other anti-secretory drugs.2 These drugs are widely used, and they are generally safe and effective. However, there are recent reports associating PPI use and magnesium deficiency.^3-10

Magnesium is essential for the function of more than 300 cellular enzymes. Its deficiency can be accompanied by hypocalcemia, hypokalemia, and serious neuromuscular and cardiovascular problems.¹¹

On March 2, 2011, the Food and Drug Administration (FDA) issued a Drug Safety Communication informing the public that prescription PPI drugs may cause low serum magnesium levels if taken for prolonged periods of time.¹² Nevertheless, there is little knowledge about this association. Thus, the aim of this study was to evaluate the prevalence of hypomagnesemia in patients with no other factors commonly associated to low serum magnesium, admitted to the emergency room of the Internal Medicine Division of the Hospital de Clinicas de Porto Alegre (HCPA), a tertiary care facility in Southern Brazil. Factors associated to serum magnesium levels were also studied.

Methods

A cross-sectional study was designed. The study was approved by the local ethics committee, and patients were included after informed consent. Blood and urine samples were collected in the morning, after a fast of at least four hours. Data such as age, gender, use of PPI, and use of other medications were obtained from medical charts. Patients with diarrhea or vomiting, chronic alcohol use, and severe uncompensated diabetes mellitus were excluded, as well as patients with chronic use of laxatives, diuretics, and other drugs that could cause magnesium deficiency, such as amphotericin B, aminoglycosides, and cyclosporine.

Serum and urine were kept at -70 ºC until biochemical measurements in the same assay run. Serum and urinary magnesium levels were measured by a method based on reaction with blue xilidil. The magnesium ions react with blue xilidil in an alkaline medium to form a water soluble purple/red complex. The increase in absorbance of the blue xilidil at 505/694 nm is proportional to the concentration of magnesium in the sample. The measurement was performed on the device Advia^® 1800, with normal range for serum magnesium: 1.3-2.7 mg/dL.

Serum calcium, phosphorus, potassium, creatinine, albumin, and urine creatinine levels were measured according to the HCPA routine.

Factors possibly associated with serum magnesium levels were evaluated through Student's t-test or the Mann-Whitney test, and Pearson's and Spearman's correlation coefficients, when indicated. Results were considered as statistically significant when p < 0.05. Multiple linear regression was used to isolate possible confounding variables. All analyses were performed using the Statistical Package for Social Sciences (SPSS), version 17.0.

Results

From September, 2010 to June, 2011 approximately 800 patients were admitted to the emergency room. Of these, around 635 patients had exclusion criteria, 14 patients refused to participate, and 151 patients were included in the study. The baseline data are described in Table 1. Their ages ranged from 17 to 94 years.

All patients had normal serum magnesium levels. Omeprazole was the only PPI used in all patients treated with this drug class, andwas administered as a single dose 30 minutes before breakfast. 42 patients received 20mg and 14 received 40 mg. The reasons for PPI use included chronic gastritis with H. pylori positive culture, dyspeptic symptoms, gastroesophageal reflux disease, peptic duodenal ulcer, dyspepsia, gastritis, epigastric pain, and esophageal hiatus hernia.

There was no difference (p = 0.59) between the mean serum magnesium levels of men (1.98 ± 0.27) and women (2.00 ± 0.19). There was no correlation between age and serum phosphorus or potassium levels with serum magnesium levels. Serum albumin, creatinine, and calcium levels were positively correlated with serum magnesium levels, as shown in Fig. 1. After multiple linear regression, only serum albumin and creatinine levels were independently associated with serum magnesium levels (Table 2).

Discussion

In this study, no patient had hypomagnesemia, although 56 were using PPIs. The prevalence of hypomagnesemia was lower than previously described in hospitalized patients, ranging from 12% to 50%.¹³ This may be due to the exclusion criteria of this study, which withdrew patients with conditions or using drugs commonly associated to hypomagnesemia, such as chronic diarrhea, vomiting, use of diuretics, gentamicin, cisplatin, ciclosporin, and others.

Data from this study suggest that the association between PPI use and hypomagnesemia is uncommon. It is possible that hypomagnesemia occurs in patients with genetic susceptibility, which may become clinically evident with PPI use. Several inherited diseases were described as causing low serum magnesium. Hypomagnesemia with secondary hypocalcemia is caused by a defect in TRPM6 channel, present in the intestine and renal tubules. The loss of function of TRPM6 leads to a reduction in intestinal absorption of magnesium, also accompanied by decreased renal reabsorption of this cation.^14,15 There are other hereditary causes of hypomagnesemia that lead to renal losses, such as familial hypomagnesemia with hypercalciuria and nephrocalcinosis, where a mutation in the gene encoding claudin-16, a tight junction protein present in the kidney responsible for the paracellular transport of magnesium, decreases magnesium reabsorption. Gitelman's syndrome also affects renal transport, and is characterized by hypomagnesemia and hipokalemia. Bartter's syndrome may have mutations of various proteins, and all these mutations affect the transport of magnesium through the thick ascending loop of Henle.¹⁶

The mechanism of hypomagnesemia associated with PPI use is unknown; however, there is a decrease in the intestinal absorption of magnesium and the role of the kidney it is not well established.⁹ As an increase in urinary excretion of magnesium before correction of hypomagnesemia has been observed in one patient, increased urinary loss cannot be excluded.³ Although two patients described by Cundy et al. presented increased urinary magnesium excretion only when serum magnesium levels were normal, after intravenous magnesium infusion, it could not be excluded that other drugs affecting renal magnesium handling could have contributed to their results.⁹

Serum albumin and magnesium levels were positively correlated, as expected, since 30% of magnesium circulates bound to this protein.¹¹ For other ions such as calcium, the calculation of total serum concentration is performed considering the albumin levels,¹⁷ but this correction is not usually recommended for magnesium. The positive correlation between albumin and magnesium suggests that this correction should also be made for this ion in order to obtain more reliable serum values.

There was also a positive correlation between serum creatinine and magnesium levels. Serum magnesium increases in the presence of kidney damage, reflected by increased blood creatinine and decreased glomerular filtration rate. This is due to loss of the ability of the kidneys to properly excrete magnesium. This correlation was expected, since it is also true for other electrolytes.

Limitations of this study include the lack of an instrument to measure adherence to the use of PPI, and the sample size. No patient had low levels of magnesium, so a larger sample is needed to evaluate the prevalence of this adverse effect.

Conclusion

PPI use did not affect mean serum magnesium levels in this study. The association between PPI use and hypomagnesemia is uncommon.

Financial support

The study received grants from the Fundo de Incentivo à Pesquisa do Hospital de Clínicas de Porto Alegre).

Conflicts of interest

The authors declare no conflicts of interest.

Received 2 October 2012

Accepted 2 December 2012

Available online 13 May 2013

• 1. Ali T, Roberts DN, Tierney WM. Long-term safety concerns with proton pump inhibitors. American Journal of Medicine. 2009;122:896-903.
• 2. Shi S, Klotz U. Proton pump inhibitors: an update of their clinical use and pharmacokinetics. European Journal of Clinical Pharmacology. 2008;64:935-51.
• 3. Furlanetto TW, Faulhaber GA. Hypomagnesemia and proton pump inhibitors: below the tip of the iceberg. Archives of Internal Medicine. 2011;171:1391-2.
• 4. Hoorn EJ, van der Hoek J, Man RA, Kuipers EJ, Bolwerk C, Zietse R. A case series of proton pump inhibitor-induced hypomagnesemia. American Journal of Kidney Diseases. 2010;56:112-6.
• 5. Regolisti G, Cabassi A, Parenti E, Maggiore U, Fiaccadori E. Severe hypomagnesemia during long-term treatment with a proton pump inhibitor. American Journal of Kidney Diseases. 2010;56:168-74.
• 6. Broeren Maarten AC, Geerdink Engelein AM, Vader Huib L. Hypomagnesemia induced by several proton-pump inhibitor. Annals of Internal Medicine. 2009;151:755-6.
• 7. Kuipers MT, Thang HD, Arntzenius AB. Hypomagnesemia due to use of proton pump inhibitors. Netherlands Journal of Medicine. 2009;67:169-72.
• 8. Shabajee N, Lamb EJ, Sturgess I, Sumathipala RW. Omeprazol and refractory hypomagnesemia. BMJ. 2008;337:a425.
• 9. Cundy T, Dissanayake A. Severe hypomagnesemia in long-term use of proton-pump inhibitors. Clinical Endocrinology. 2008;69:338-41.
• 10. Agus ZS. Hypomagnesemia. Journal of the American Society of Nephrology. 1999;10:1616-22.
• 11. Food and Drugs Administration. FDA drug safety communication: low magnesium levels can be associated with long-term use of proton pump inhibitor drugs (PPIs) [cited 2011 mar 2]. Available from: http://www.fda.gov
• 12. Wong ET, Rude RK, Singer FR, Shaw Jr ST. A high prevalence of hypomagnesemia and hypermagnesemia in hospitalized patients. American Journal of Clinical Pathology. 1983;79:348-52.
• 13. Ayuk J, Gittoes NJ. How should hypomagnesaemia be investigated and treated? Clinical Endocrinology. 2011;75:743-6.
• 14. Schlingmann KP, Sassen MC, Weber S, Pechmann U, Kusch K, Pelken L, et al. Novel TRPM6 mutations in 21 families with primary hypomagnesemia and secondary hypocalcemia. Journal of the American Society of Nephrology. 2005;16:3061-9.
• 15. Naderi AS, Reilly Jr RF. Hereditary etiologies of hypomagnesemia. Nat Clin Pract Nephrol. 2008;4:80-9.
• 16. Thode J, Juul-Jorgensen B, Bhatia HM, Bartles PD, Fogh-Andersen N, Siggaard-Andersen O. Ionized calcium, total calcium and albumin corrected calcium in the serum in 1213 patients with suspected calcium metabolic diseases. A prospective multicenter study. Ugeskrift for Laeger. 1989;151:2423-5.
• 17. Hsu CY, Chertow GM. Elevations of serum phosphorus and potassium in mild to moderate chronic renal insufficiency. Nephrology, Dialysis, Transplantation. 2002;17:1419-25.

*

Study conducted at Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil.

Publication Dates

• Publication in this collection
  05 July 2013
• Date of issue
  June 2013

History

• Received
  02 Oct 2012
• Accepted
  02 Dec 2012