PPI-Related Hypomagnesemia: Putting it in Perspective

David A. Johnson, MD


March 07, 2011

This feature requires the newest version of Flash. You can download it here.

Hello. I'm Dr. David Johnson, Professor of Medicine and Chief of Gastroenterology at Eastern Virginia Medical School in Norfolk, Virginia. Today is March 2nd and the US Food and Drug Administration (FDA) has put out an advisory warning that patients on proton pump inhibitors (PPIs) may be at risk for hypomagnesemia. Now before you jump to call all your patients on proton pump inhibitors and tell them to have their magnesium levels checked, let me give you some perspective on these data and some guidance as to how I am handling this in my practice.

Let's start with the background. This alert was based on a number of reports over the course of the last several years, during which patients have presented with relatively profound hypomagnesemia, prompting hospitalization. Patients can present with profound muscle weakness, twitching, and cardiac dysrhythmias that may prompt the diagnosis of hypomagnesemia. In patients who were hospitalized, about 25% did not respond even to magnesium repletion until the proton pump inhibitor was stopped, and then they reabsorbed normally and their magnesium levels normalized. On occasion, these patients were rechallenged and the same event occurred.

Why does this occur with proton pump inhibitors? Magnesium, first of all, is basically passively transported and absorbed in the small bowel. This is where the bulk of magnesium transport occurs. To a smaller degree, active transport mechanisms in the small bowel are now recognized, including transient receptor potential melastatin 6 (TRPM6) and TRPM7. These have been described as active transport pathways for magnesium. We don't know what degree of magnesium is transported this way, but we do know that patients who have a homozygous mutation of these pathways actually [have] significant hypomagnesemia. This accounts for at least some of the patients who have normal calcium absorption.

How this interacts with proton pump inhibitors is unknown. We don't know if PPIs affect active or passive transport of magnesium, or whether in patients prone to have some TRPM deficiency, PPI use may trigger hypomagnesemia. The FDA alert put us on notice that this can potentially occur, albeit rarely. In the absence of mechanistic studies, we don't really know why hypomagnesemia would occur with PPI use.

The FDA's advice is if you consider starting a patient on a proton pump inhibitor, before doing so, the patient's magnesium level should be checked (especially if long-term use of PPIs is anticipated).

Before we jump to that, and I'll leave the FDA's advice to your own judgment, the way I look at this is that in patients who have other reasonable reasons to have hypomagnesemia (eg, malabsorption syndromes in the small bowel, renal losses, or if they're on some medication such as a diuretic that may predispose them to have an ongoing magnesium loss), there may be reason to check magnesium level when you're having a follow-up [appointment] with these patients. I'm not sure that I'm ready to start checking magnesium levels in every patient going on a proton pump inhibitor, even in that population, but it's probably not unreasonable.

Magnesium level deficiency or hypomagnesemia may present with some unusual circumstances, such as profound cardiac dysrhythmias or seizures. Sometimes simple things like muscle twitching can be a protean manifestation. Again, magnesium deficiency is something to consider in patients on proton pump inhibitors who present with unusual symptoms and who may have some magnesium loss risk.

Putting this in perspective, I would at least have this in the back of your mind when considering the population of patients on PPIs. It's extremely rare, as are many other side effects such as interstitial nephritis that we see with proton pump inhibitors. The mechanism of hypomagnesemia is not well understood. Every time we prescribe a PPI we should certainly consider: Does this patient really need this medication? If they do, appropriate use of a proton pump inhibitor certainly should not be preempted by concern for hypomagnesemia. I'll leave this to your best clinical judgment. I look forward to seeing you again soon. Thanks.