Clinical Review

Review of the Long-Term Effects of Proton Pump Inhibitors

Proton pump inhibitors are widely used, but up to 70% of patients take them without an appropriate indication and face risk of adverse effects associated with long-term use.

Author and Disclosure Information



Proton pump inhibitors (PPIs) are one of the most frequently used drug classes, given that they are readily accessible over-the-counter as well as via prescription. About 100 million PPI prescriptions dispensed annually.1 One study showed that between 25% and 70% of patients taking a PPI do not have an appropriate indication.2 It is common for patients to be prescribed a PPI during and following an inpatient hospital stay. Upon discharge, many of these patients continue taking PPIs without proper documentation or a planned termination date.

The human stomach uses 3 primary neurotransmitters that regulate gastric acid secretion: acetylcholine (ACh), histamine (H), and gastrin (G). The interactions between these neurotransmitters promote and inhibit hydrogen ion (H+) generation. Stimulation of their corresponding receptors draws H+ into parietal cells that line the stomach. Once in the cell, a H+-K+-ATPase (more commonly known as the proton pump) actively transports H+ into the lumen of the stomach. The H+ bind with chlorine ions to form hydrochloric acid, which increases stomach acidity.5 Histamine receptors were thought to be responsible for the greatest degree of stimulation. Hence, histamine type 2-receptor antagonists (H2RAs) became a novel means of therapy to reduce stomach acidity. While utilizing H2RAs was effective, it was theorized the downstream inhibition of the action of all 3 neurotransmitters would serve as a more successful therapy. Therefore, PPIs were developed to target the H+-K+-ATPase Over the past decade, many studies have evaluated the long-term PPI adverse effects (AEs). These include calcium and magnesium malabsorption, vitamin B12 deficiency, Clostridium difficile (C difficile) associated disease (CDAD), and community-acquired pneumonia (CAP). Within the past year, data have become available linking PPI use to dementia and chronic kidney disease (CKD).3,4 The following article reviews literature on the safety of long-term PPI use and proposes recommendations for proper use for their most common indications.


Calcium & Long-Term Fracture Risk

Calcium is an essential component in bone health and formation. In fact, 99% of all calcium found in the body is stored in bones.6 The primary source of calcium is through diet and oral supplements. After it is ingested, calcium is absorbed from the stomach into the blood in a pH dependent manner. If the pH of the stomach is too high (ie, too basic) calcium is not absorbed into blood and remains in the gastrointestinal (GI) tract for fecal excretion. Without sufficient calcium, the body’s osteoclasts and osteoblasts remain inactive, which hinders proper bone turnover.7

The decrease in acidity leads to calcium malabsorption and increases fracture risk long- term.8 Khalili and colleagues surveyed 80,000 postmenopausal women to measure the incidence of hip fracture in women taking PPIs. The study found that there was a 35% increase in risk of hip fracture among women who regularly used PPIs for at least 2 years (age-adjusted hazard ratio [HR] 1.35; 95% confidence interval [CI], 1.13 -1.62). Adjusted HRs for 4-year and 6- to 8-year use of a PPI was 1.42 (95% CI, 1.05-1.93) and 1.55 (95% CI, 1.03-2.32), respectively, indicating that the longer women were on PPI therapy, the higher the risk of hip fracture. The study also evaluated the time since stopping PPI and the risk of hip fracture. Women who stopped PPI use more than 2 years prior had a similar risk to that of women who never used a PPI, indicating that the effect was reversible.9


Magnesium is an important intracellular ion that has a number of key functions in metabolism and ion transport in the human body. Once ingested, magnesium is absorbed into the bloodstream from the small and large intestines via passive and active transport. Transient receptor potential melastatin 6 (TRPM6) is one of the essential proteins that serve as a transporter for magnesium.10 The high affinity for magnesium of these transporters allows them to maintain adequate levels of magnesium in the blood. In states of low magnesium (hypomagnesemia), the body is at risk for many AEs including seizures, arrhythmias, tetany, and hypotension.11

Proton pump inhibitors have been linked to hypomagnesemia, and recent evaluation has clarified a potential mechanism.12 TRPM6 activity is increased in an acidic environment. When a PPI increases the pH of the stomach, TRPM6 and magnesium levels decrease.12 Luk and colleagues identified 66,102 subjects experiencing AEs while taking a PPI. Hypomagnesemia had a prevalence rate of 1% in these patients. According to the researchers, PPIs were associated with hypomagnesemia and that pantoprazole had the highest incidence among all other PPIs studied (OR, 4.3; 95% CI, 3.3 – 5.7; P < .001).13

Vitamin B12

In recent years, vitamin B12 has been the subject of many studies. An area of concern is vitamin B12’s neurologic effect, as it has been successfully demonstrated that vitamin B12 is essential for proper cognitive function.14 Some data suggest that degeneration is present in parts of the spinal column in patients with cognitive decline or neurologic problems. These lesions are due to improper myelin formation and are specific to vitamin B12 deficiency.15 In 2013 the CDC published the Healthy Brain Initiative, which stated cognitive impairment can be caused by vitamin B12 deficiency.16

Similar to calcium, vitamin B12 needs an acidic environment to be digested and absorbed.17 Vitamin B12 is released from food proteins via gastric acid and pepsin. Once free, the vitamin B12 pairs with R-binders secreted in the stomach. Pancreatic enzymes then degrade this complex into a form that can be absorbed into circulation by the intestine. Given that PPIs reduce the acidity of the stomach, they also reduce the body’s ability to release vitamin B12 from food proteins and be paired with the R-binders.18

In 2013, Lam and colleagues evaluated the association between vitamin B12 deficiency and the use of PPIs and H2RAs. An extensive evaluation was performed on 25,956 patients with a diagnosis of vitamin B12 deficiency and 184,199 patients without. About 12% of patients with vitamin B12 deficiency had received more than a 2-year supply of a PPI, whereas only 7.2% of the patients without vitamin B12 deficiency received a 2-year supply of a PPI. Four point 3 percent of patients with vitamin B12 deficiency received more than a 2-year supply of an H2RA. Only 3.2% of patients without vitamin B12 deficiency received more than a 2-year supply of H2RA. The study concluded that a 2-year or greater history of PPI (OR, 1.65; 95% CI, 1.58-1.73) or H2RA (OR, 1.25; 95% CI, 1.17-1.34) use was associated with vitamin B12 deficiency.19

PPIs and Infections

Clostridium difficile-associated disease

Nationwide CDAD has become a prevalent infection nationwide. In 2011, C difficile caused nearly 500,000 infections and was associated with 29,000 deaths in the U.S.20 One study stated that C difficile is the third most common cause of infectious diarrhea in people aged >75 years.21

C difficile is part of the body’s normal flora in the large intestine. It grows and colonizes in an environment of low acidity. Therefore, in the stomach, where the pH is relatively low, C difficile is unable to colonize.22 When a PPI is introduced, the increased gastric pH increases the risk for CDAD.

Dial and colleagues conducted a multicenter case control study to determine whether gastric acid suppression increases the risk of CDAD. Compared with patients who did not take a gastric acid suppressant, those taking a PPI had a 2.9-fold increase in developing CDAD (95% CI, 2.4-3.4). Comparatively, H2RAs had a 2.0-fold increase for CDAD (95% CI, 1.6 to 2.7). These results correlated with the fact that PPIs have a greater impact on gastric pH than do H2RAs.23

Community-Acquired Pneumonia

Community-acquired pneumonia (CAP) has become a growing concern in the U.S. According to the Infectious Disease Society of America (IDSA) and American Thoracic clinical consensus guidelines, CAP remains one of the top reasons for hospitalizations in the U.S., and about 10% of patients admitted to the hospital for CAP end up in the intensive care unit (ICU).24 In the past, PPIs have been linked to patients’ predisposal for developingCAP.25 Although controversial, available evidence suggests a direct association. In 2008 Sarker and colleagues theorized a mechanism that the acid reduction of the gastric lumen allows for increased bacterial colonization in the upper part of the GI tract.26 Since the acidity of the stomach serves as a defense mechanism against many ingested bacteria, many pathogens will be able to survive in the more basic environment.25

Sarkar and colleagues went on to evaluate 80,000 cases over 15 years. The objective was to examine the association between PPI use and the date of diagnosis of the CAP infection, known as the index date. The study demonstrated that PPI use was not associated with increased CAP risk in the long-term (adjusted odds ratio (OR), 1.02; 95% CI, 0.97-1.08). The study did find a strong increase in the risk of CAP if a PPI was started within 2 days (adjusted OR, 6.53; 95% CI, 3.95-10.80), 7 days (adjusted OR, 3.79; 95% CI, 2.66-5.42), and 14 days (adjusted OR, 3.21; 95% CI, 2.46-4.18) of the index date.26

Four years later, de Jagar and colleagues examined the differences in microbial etiology in CAP patients with and without an active PPI. Over a 4-year study period, 463 individuals were selected with clinical suspicion of CAP. The microbial etiology could be determined in 70% of those patients. The remaining 30% were excluded due to an alternative diagnosis. One of the most likely pathogens to cause a CAP infection is Streptococcus pneumoniae (S pneumonia).27 Patients prescribed a PPI were significantly more likely to be infected with S pneumoniae than those not prescribed a PPI (28% vs 11%). The study concluded that the risk of S pneumoniae in patients taking a PPI was 2.23 times more likely (95% CI, 1.28-3.75).28


Recommended Reading

NIOSH Adds to Hazardous-Drugs List
Federal Practitioner

Related Articles