PPIs and magnesium: can omeprazole or pantoprazole lower magnesium levels?

Yes, a drop in blood magnesium has been observed in some people who take a PPI for a long time, and medicines agencies have added it to the leaflets^1,2. Severe forms remain uncommon, and their exact frequency is not known².

The first cases were described in 2006 in a letter to the New England Journal of Medicine, which reported hypoparathyroidism linked to very low magnesium in patients taking PPIs³. On 2 March 2011, the US FDA issued a safety communication after reviewing 38 cases from its pharmacovigilance database and 23 published cases. It added hypomagnesaemia to the warnings of all prescription PPIs¹. The agency itself states that this disorder is probably underdiagnosed, which makes it impossible to estimate its frequency from reports alone.

In Europe, the UK medicines regulator concluded in April 2012, based on cases published or reported to European authorities, that PPIs may cause hypomagnesaemia². The summaries of product characteristics published by the French medicines agency (ANSM), such as that of Mopral (prescription omeprazole), now contain a paragraph devoted to this risk⁴. It applies to the whole class: omeprazole, esomeprazole, pantoprazole, lansoprazole and rabeprazole.

Three meta-analyses conclude that there is a statistical association between PPI use and hypomagnesaemia, with a risk 1.4 to 1.8 times higher depending on the study. All are based on observational studies, which show an association without proving cause.

In 2014, Park and colleagues pooled nine studies and 115,455 patients. They found an odds ratio of 1.78 (95% confidence interval: 1.08 to 2.92). The median proportion of people with hypomagnesaemia reached 27.1% among PPI users and 18.4% among non-users. The authors nonetheless report very high heterogeneity between the studies (I² of 98%), which prevents them from reaching a definitive conclusion⁵.

In 2015, Cheungpasitporn and colleagues pooled nine studies covering 109,798 patients. The overall relative risk was 1.43 (1.08 to 1.88). It rose to 1.63 (1.14 to 2.23) when only the highest-quality studies were kept⁶. In 2019, a third meta-analysis of sixteen studies and 131,507 patients found hypomagnesaemia in 19.4% of users compared with 13.5% of non-users. The adjusted odds ratio was 1.71 (1.33 to 2.19). Taking high doses was associated with a higher risk than low doses, with an odds ratio of 2.13⁷.

A large study carried out in intensive care on 11,490 admissions adds an important nuance. The association appeared only in patients who were also taking a diuretic, with an odds ratio of 1.54. In the others, PPI use was not linked to magnesium levels⁸. A 2016 review points out that no controlled prospective study can yet establish causality⁹.

The main hypothesis concerns active intestinal absorption of magnesium, which depends on the acidity of the environment. By reducing acid secretion, a PPI could slow down this absorption pathway⁹.

The gut absorbs magnesium through two routes. The first is passive: magnesium passes between the cells of the intestinal wall, following a concentration gradient. The second is active: further along the gut, in the caecum and colon, specialised channels called TRPM6 and TRPM7 bring magnesium into the intestinal cells. TRPM6 activity increases when the environment is more acidic. The authors of the review therefore suggest that a less acidic intestinal pH on PPIs reduces the activity of this channel and lowers absorption⁹.

The same authors consider that this mechanism probably explains only part of the picture, within a more complex set of factors involving pharmacology and a genetic predisposition⁹.

In most reported cases, hypomagnesaemia appeared after more than a year of treatment, and sometimes as early as three months^1,2. The French leaflet uses these same two benchmarks⁴.

The FDA considers the risk very low for over-the-counter PPIs taken as the US label directs, that is, fourteen-day courses no more than three times a year¹. The UK regulator reminds users that a PPI bought without a prescription should not be taken for more than four weeks without medical advice².

Three situations recur in agency documents and in studies. The first concerns the duration of treatment. The second relates to dose, since high doses are associated with a higher risk⁷. The third is combination with medicines that also lower magnesium, first and foremost diuretics, or whose toxicity increases when magnesium falls, such as digoxin^1,8.

Hypomagnesaemia may cause no symptoms at all, or develop insidiously and go unnoticed^1,2. Only a blood test can confirm it.

The first signs described are non-specific: loss of appetite, nausea, vomiting, fatigue and weakness. As the deficiency worsens, tingling, numbness, muscle contractions and cramps may appear, followed by heart rhythm disturbances or seizures¹⁰. The leaflets add tetany, dizziness and delirium. Severe hypomagnesaemia may be accompanied by low blood calcium or potassium^4,10.

The FDA advises seeking medical attention promptly in the event of palpitations, an abnormal heart rhythm, muscle spasms, tremors or seizures during PPI treatment¹. These signs have many other possible causes, which is why a measurement is more useful than a guess.

The FDA, the UK regulator and the French leaflet suggest considering a blood magnesium test before starting prolonged treatment, then repeating it periodically^1,2,4. They focus primarily on people who also take digoxin or a diuretic.

It is therefore not a routine check for everyone, but a targeted test that the doctor decides on according to the planned duration and other treatments. Mentioning that you have been taking a PPI for more than a year, including one bought without a prescription, helps raise this question¹.

When hypomagnesaemia is confirmed, treatment generally relies on magnesium intake. In around a quarter of the cases reviewed by the FDA, supplementation alone was not enough and the PPI had to be stopped¹. The UK regulator and the French leaflet state that in most cases hypomagnesaemia improved after supplementation and stopping the PPI^2,4. A treatment is never stopped on your own initiative: the FDA reminds patients of this, and any decision is made with the doctor or pharmacist¹.

The documents consulted, the French leaflet, the FDA and the NIH, describe no interaction through which a magnesium supplement would interfere with the action of a PPI^1,4,10. The relationship described goes the other way: PPIs are associated with lower magnesium.

Searches for “omeprazole magnesium” or “esomeprazole magnesium” often cause confusion. In these names, the word magnesium refers to the salt in which the active molecule is manufactured. The medicine is not a magnesium supplement and does not replace magnesium intake.

How far apart to take them depends mainly on other medicines. Magnesium forms insoluble complexes with antibiotics of the tetracycline and quinolone families: these antibiotics should be taken at least two hours before or four to six hours after a supplement containing it. It also reduces the absorption of bisphosphonates used for osteoporosis, which must be separated from it by at least two hours¹⁰.

Taking magnesium does not protect against established hypomagnesaemia, which only a blood test can detect. Before adding a supplement on top of a PPI, it is worth talking to the doctor or pharmacist, particularly in the case of kidney disease, a diuretic or digoxin.