2012-03-12 17:26:10 UTC

Long-Term PPI Therapy Risks: “Keep Calm and Carry On”

Aug. 2, 2011


Colin W. Howden, MD, AGAF

Professor of Medicine, Division of Gastroenterology, Northwestern University Feinberg School of Medicine, Chicago, IL


I stop PPI therapy far more often than I initiate it, since many of the patients I see have been started on a PPI inappropriately for problems that are not acid-mediated. Regrettably, such patients are often given the somewhat derogatory label of “PPI failure” and referred for further evaluation. In many instances, the most appropriate initial course of action is to stop the PPI.

However, I strongly believe that long-term PPI therapy is essential for many other patients, including those with erosive esophagitis and those at risk of NSAID-related upper GI tract bleeding. PPIs are efficacious and safe for these intended uses. Obviously, PPIs are not devoid of side effects (after all, no drug is). The most typical side effects with PPIs are generally mild and often self-limiting. I rarely encounter PPI-related headache, diarrhea or abdominal pain, but would generally approach them with dose reduction or, perhaps, a switch to an alternative PPI (assuming the patient still required PPI therapy)

Concerns have been raised about more serious potential consequences of long-term PPI use, including community-acquired pneumonia, bacterial enteric infections, fractures, malabsorption of cobalamin, and rare, idiosyncratic reactions. These have all been extensively reviewed elsewhere.1, 2 In view of space limitations here, I cannot discuss each of these. However, absolute risks have generally been of small magnitude, and the evidence to support causality has often been of low quality and based on retrospective studies or case reports. Furthermore, there has been evident channeling bias in many studies in which PPI use may simply have acted as a surrogate for poor general health.3

Many patients and physicians have been alarmed about reports of potentially serious consequences of long-term PPI therapy. There are at least anecdotal reports of some patients stopping PPI therapy for unwarranted concerns, and thereby being put at risk of NSAID-related upper GI bleeding or recurrence of esophagitis. Many of us will have received inquiries about the level of risk and the best course of action to follow. My own approach has usually been based on the wartime adage from the U.K., as suggested by the title (figure 1 on Page 6). If a patient has a good reason to be on a PPI, he/she should be on a PPI at the lowest effective dose. Patients who are taking a PPI unnecessarily should stop it. There is nothing iconoclastic about these recommendations; in medical school, we were all taught these basic prescribing principles for drugs in general.

Recently, the FDA has required PPI manufacturers to amend their labels to indicate that long-term use, especially in high dose, may be associated with an increased risk of fractures. Based largely on the results of in vitro studies, the FDA has counseled against the use of particular PPIs by patients taking clopidogrel. The FDA has also expressed concern about hypomagnesemia, and has suggested that prescribers should consider checking serum magnesium levels in patients about to start a PPI and in particular patients during PPI therapy. I will briefly review these three topics that are likely to be of the most concern to prescribers.

PPIs and fractures: a tough break?

There are conflicting studies in the literature regarding a possible association between PPI use and fracture risk. However, we may now safely assume that a “tipping point” has been reached in the fracture farrago, since we have a well-conducted meta-analysis of observational studies.4 The authors found a pooled odds ratio for hip fracture in current PPI users of 1.25 (95 percent confidence interval 1.14 to 1.37). This, in itself, does not prove causality. Furthermore, they found no evidence of a duration-response relationship; long-term PPI use appeared to carry no greater risk of fracture than shortterm use. In my opinion, there may be a small absolute risk of fracture in PPI users, but there are also many other potential confounding factors. Assuming that the risk is real, we should acknowledge that we have no clear understanding of its mechanism. Furthermore, we have no evidence on which to base recommendations on supra-normal calcium supplementation or additional bone density monitoring. For now, if there is a valid reason for PPI use, see figure 1.

PPI-clopidogrel interaction: the beat goes on …

The main results of the Clopidogrel and the Optimization of Gastrointestinal Events trial5 will already be well known to this readership; omeprazole co-therapy was not associated with an increased risk of adverse cardiovascular events in clopidogrel users, but was associated with a significant reduction in adverse GI events. In the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel &emdash Thrombolysis in Myocardial Infarction 38,6 PPI users taking clopidogrel had no excess of cardiovascular events, there were no obvious differences among the various PPIs taken, and there was a statistically significant reduction in all-cause mortality. (Regarding that last finding, see table 3 of the study.) U.S. consensus recommendations endorsed by the professional societies of our cardiology brethren7 support PPI co-therapy for patients with additional risk factors for upper GI bleeding. Furthermore, they did not find substantive differences among the PPIs. Therefore, I continue to believe that a patient who is taking clopidogrel and aspirin for a valid indication should also receive a PPI if there are other risk factors for upper GI bleeding. PPI co-therapy is not a panacea, but it does reduce GI risk. However, current FDA recommendations are to avoid omeprazole (and esomeprazole) in patients on clopidogrel. This is still a “watch this space” issue; for now, see figure 1.

Et in magnesia ego?

The FDA has been made aware of sporadic cases of hypomagnesemia in patients who had been on a PPI for at least a year &emdash and sometimes for much longer. Patients presented with alarming features, including syncope, seizures and cardiac arrhythmias. Conveniently, published case reports (some of which are also in the FDA’s collection) have recently been summarized elsewhere.8 Most PPIs have been implicated, suggesting a class effect. Magnesium levels returned to normal after PPI withdrawal, but decreased again in some patients following re-treatment with the original or an alternative PPI. Excessive urinary magnesium losses were not apparent. Hypomagnesemia was often accompanied by hypocalcemia and hypokalemia. The FDA has warned that hypomagnesemia may be asymptomatic and can, therefore, go unrecognized in some patients. This may pose a threat for the elderly, those on diuretic therapy and those with heart disease. Unfortunately, we have no real understanding of the frequency of the problem. However, given the millions of patient-years of PPI use, we can conclude that clinically apparent hypomagnesemia is extremely rare. Physicians must decide on whether or not to monitor serum magnesium levels in their patients. At present, I feel that there is insufficient evidence to recommend this routinely. However, checking serum magnesium would not be unreasonable in certain long-term PPI users, including the elderly, those with heart disease and those on a diuretic. We need to learn much more about this issue, including a possible mechanism, its true frequency and whether supplementation is necessary in those (presumably rare) patients with a subnormal magnesium level and a genuine requirement for PPI therapy. Prescribers need to be aware of this issue since patients may ask about it. In the longer term, the sound advice in figure 1 may yet turn out to be appropriate.


PPIs should be stopped whenever they are being used inappropriately. Generally, this is easily accomplished and I continue to be unimpressed by the hyperbole surrounding the issue of “rebound hypersecretion&lrquo; after PPI withdrawal.9 However, I feel strongly that the benefits of PPI therapy for appropriate clinical indications greatly outweigh any risks for most patients. Therefore, when a PPI is genuinely indicated, please “keep calm and carry on.”

1. Nealis TB, Howden C W. Is there a dark side to long-term proton pump inhibitor therapy? American Journal of Therapeutics 2008; 15: 536 – 542.

2. Parikh N, Howden CW. Safety of drugs used in acid-related disorders and functional gastrointestinal disorders. Gastroenterology Clinics of North America 2010; 39: 529 – 542.

3. Howden CW. PPIs and clopidogrel: The band plays on. American Journal of Gastroenterology 2010; 105: 2438 – 2439.

4. Ngamruengphong S, Leontiadis GI, Radhi S, Dentino A, Nugent K. Proton pump inhibitors and risk of fractures: a systematic review and meta-analysis of observational studies. American Journal of Gastroenterology 2011; in press.

5. Bhatt DL, Cryer BL, Contant CF, et al. Clopidogrel with or without omeprazole in coronary artery disease. New England Journal of Medicine 2010; 363: 1909 – 1917.

6. O’Donoghue ML, Braunwald E, Antman EM, et al. Pharmacodynamic effect and clinical efficacy of clopidogrel and prasugrel with or without a proton-pump inhibitor: an analysis of two randomised trials. Lancet 2009; 374: 989 – 997.

7. Abraham NS, Hlatky MA, Antman EM, et al. ACCF/ACG/AHA 2010 expert consensus document on the concomitant use of proton pump inhibitors and thienopyridines: A focused update of the ACCF/ACG/AHA expert consensus document on reducing the gastrointestinal risks of antiplatelet therapy and NSAID use. American Journal of Gastroenterology 2010; 105: 2533 – 2549.

8. Cundy T, Mackay J. Proton pump inhibitors and severe hypomagnesemia. Current Opinion in Gastroenterology 2011; 27: 180 – 185.

9. Howden CW, Kahrilas PJ. Just how “difficult” is it to withdraw PPI treatment? American Journal of Gastroenterology 2010; 105: 1538 – 1540.

More on Diarrhea

Course for APPs benefits all experience levels

April 16, 2018

AGA's popular course for advanced practice providers will take place Aug. 10 through 12 in Chicago, Illinois.

APPs: Enhance your delivery of patient care

March 28, 2018

AGA's popular course for advanced practice providers will be held this summer in Chicago.

The Role of Anchoring in Working Up Diarrhea: A Practical Teaching Case

Feb. 1, 2018

A 48-year-old woman with a family history of Crohn’s disease in her mother, sought gastroenterology consultation for a 2-year history of diarrhea described as multiple bowel movements per day with urgency.