Article Text

Download PDFPDF

Toxicity of long-term use of proton pump inhibitors in children
  1. Pauline De Bruyne1,
  2. Shinya Ito2,3
  1. 1 Department of Paediatrics and Medical Genetics, Ghent University, Gent, Belgium
  2. 2 Division of Clinical Pharmacology and Toxicology, The Hospital for Sick Children, Toronto, Ontario, Canada
  3. 3 Department of Paediatrics and Department of Pharmacology and Toxicology, University of Toronto, Toronto, Ontario, Canada
  1. Correspondence to Dr Pauline De Bruyne, Department of Paediatrics and Medical genetics, Ghent University, De Pintelaan 185, Ghent 9000, Belgium; Pauline.DeBruyne{at}


Proton pump inhibitor (PPI) use is becoming increasingly common. Although the toxicity profiles of PPIs are not well understood particularly in children, PPIs have been associated with increased risks of gastrointestinal and respiratory tract infection, vitamin B12 deficiency, hypomagnesaemia, bone fractures, and rebound hyperacidity after discontinuation. Prescribers should take into account that PPI uses pose toxicity risks, which remain to be fully characterised in infants and children.

  • gastroesophageal reflux
  • proton pump inhibitors
  • safety
  • adverse events
  • children

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


During the last three decades, prescription of proton pump inhibitors (PPIs) has increased markedly.1–3 PPIs have shown to be the most effective drugs in children (older than 1 year) as well as in adults with gastro-oesophageal reflux disease (GERD). PPIs are also used in combination with antibiotics for Helicobacter pylori eradication.1 4–7 However, PPIs have not clearly shown efficacy for treatment of gastro-oesophageal reflux in neonates and infants. Overall, it has been widely recommended that PPI treatment in this age group should be reserved for cases with evidence of pathological exposure to acid reflux episodes and/or oesophagitis.7 Prescription of PPIs has dramatically expanded in this population, despite these clear recommendations.2 7–10

Together with the rise in the prescriptions of PPIs for infants and children and the reports about the lack of evidence for efficacy in infants, there have been increasing safety concerns about PPI use.7 8 Evidence of potential adverse effects currently challenges the previously assumed safety of these gastric acid inhibitors. To update our understanding on the PPI use in children (aged 0–17 years), we reviewed the relevant literature, more specifically on the safety of these drugs in paediatric GERD.

Toxicity profiles of PPI in children

The safety profile of PPIs is generally better described in adults than in children, and is especially less characterised in infants.4 11 A recent review by Cohen et al 12 reported adverse effects in 34% of children treated with PPIs, including most commonly headaches, nausea, diarrhoea and constipation. Additionally, prolonged use of PPIs has raised concerns about hypochlorhydria and hypergastrinaemia, increased risk of infections, vitamin and mineral deficiencies, adverse bone health, food allergy and drug interactions. These concerns have been raised in adults, but paediatric studies are limited.1 7 11 The specific safety concerns of prolonged PPI therapy and the respective evidence summary are listed in table 1. Each of the main adverse effects is discussed below.

Table 1

Concerns about long-term therapy with proton pump inhibitors (PPIs) in children


Infection in neonatal intensive care

There is increasing evidence that gastric acid suppression may place susceptible infants and children, particularly those with indwelling catheters or defective immune systems, at risk for nosocomial infections.12 Safety studies on the use of histamine-2 receptor antagonists (H2RAs) particularly in neonatal intensive care with hospital-acquired pathogens raised the following concerns: these gastric acid inhibitors in neonatal intensive care have been identified as a risk factor for necrotising enterocolitis,13 sepsis and meningitis,14 and bloodstream infections with Candida species.15 16 The pathogenic mechanism of these infections linked to gastric acid suppression is thought to be due to a reduction of the gastric acid barrier with increased gastrointestinal tract colonisation as a negative result.7 PPIs are thought to lead to these nosocomial infections via a similar pathogenic mechanism of gastric acid suppression and consequent gastrointestinal colonisations. To date, however, there are no studies with PPIs.

Gastrointestinal infection

Although pathogenic mechanisms have not been clearly defined, multiple theories are proposed for PPI-associated gastrointestinal infections. A low pH environment of the gastric lumen is an innate barrier against pathogenic organisms. PPI-induced gastric acid reduction may increase the risk for gastrointestinal infections by causing alterations in the gastrointestinal microbiome, which has been shown both in children and in adults.17–19 Additionally, elevated pH in the stomach reduces gastric mucus viscosity, compromises leucocyte activity and enhances bacterial invasion.

A variety of studies (as outlined in table 2) conducted in adults have shown a potential association between PPI treatment and increased risk of Clostridium difficile infections (CDIs) and other enteric infections, including Salmonella and Campylobacter infections.7 20–22 To date, four studies17 23–25 have investigated this association in paediatric patients. Canani et al 23 found in their prospective, open-label, controlled trial with 186 children that those treated with gastric acid inhibitors (H2RAs and PPIs) showed a higher rate of acute gastroenteritis than healthy controls. However, effects of confounders including antibiotic use and history of hospitalisation could not be ruled out. Moreover, the increased risk of acute gastroenteritis in this study of Canani et al was identified in the follow-up period without PPI treatment, which could weaken the case for an association. The authors conclude that this effect seems to be prolonged even after the end of the treatment. The retrospective, population-based, nested case–control study by Freedberg et al 17 described an increased risk of CDI with acid suppression medication in an outpatient setting. The case–control study of 68 children with CDI by Turco et al 24 also showed a significantly higher rate of PPI use in those cases with CDI. However, the fact that the group of CDI cases in that study had a higher proportion of children with inflammatory bowel disease than the control group is considered a major limitation.8 The study by Brown et al 25 did not show an increased CDI risk in children receiving PPIs, whereas a 4.5 times greater CDI risk was shown in children receiving H2RAs when compared with those not receiving acid-suppressive therapy. The risk of gastrointestinal infection associated with PPI therapy in children remains to be clearly defined.

Table 2

Study overview of proton pump inhibitors and gastrointestinal and respiratory infections

Respiratory infection

Concerns have been raised about a potential association between PPI use and respiratory tract infections. Biological plausibility of this association exists, because PPIs decrease gastric acid production and can increase bacterial colonisation in the stomach, which enhances invasion of pathogenic organisms from the gastrointestinal tract into the upper and lower respiratory tract, causing respiratory infection.

There are systematic reviews and meta-analyses evaluating PPI use as a risk factor for respiratory infection in adults, but results are conflicting. In one study, an elevated risk of pneumonia was suggested.26 Similarly, Johnstone et al in 2010 and Lambert et al in 2015 found a statistically significant association between community-acquired pneumonia and short-term PPI use.27 28 In contrast, prior to these systematic reviews, Sultan et al 29 in 2008 reported no significant association between PPI and respiratory infections.

The evidence on a possible link between PPI use and respiratory infections in children is as vague as the evidence in adults. Only one prospective study by Canani et al specifically evaluated community-acquired pneumonia as an outcome, showing that the children treated with gastric acid inhibition (either by ranitidine or omeprazole) were more likely to develop community-acquired pneumonia, compared with healthy children.23 This effect was identified in the follow-up period without PPI treatment, which could weaken the case for an association. Other studies30–32 reported upper and lower respiratory tract infections as adverse events during PPI therapy. However, limitations exist in these studies, including confounding factors (such as history of hospitalisation, antibiotic use and comorbidities).

Hypergastrinaemia, mucosal changes and rebound acid hypersecretion

Increasing gastric pH leads to hypergastrinaemia, which has growth-promoting effects on several epithelial cell types. Prolonged PPI use in adults is associated with parietal and enterochromaffin-like cell hyperplasia.7 33 34 Similar PPI-associated changes in the gastric mucosa have been found in children.35

The clinical relevance of rebound acid hypersecretion, defined as an increase in gastric acid secretion after PPI withdrawal, has been debated in adult literature.36 According to the proposed pathophysiological mechanism, hypergastrinaemia and growth of histamine-releasing enterochromaffin-like cells lead to an increased acid secretory capacity after PPI discontinuation.36 It has been reported in two studies that gastric acid rebound hypersecretion after PPI discontinuation causes reflux-like symptoms post-treatment in asymptomatic volunteers.37 38 However, the significance of this in patient populations is not clear. Studies in patients with reflux disease found no proof of symptomatic rebound acid hypersecretion, but these studies had methodological weaknesses (ie, no direct measurement of gastric acid and difficulty in distinguishing symptoms related to the underlying acid-related disorder from de novo symptoms).36 Information on rebound acid hypersecretion in children is even more limited. A randomised controlled trial by Winter et al 31 explored this issue in children, suggesting that rebound hyperacidity is not a clinically relevant problem in children who received PPIs for a short period (1–3 months).

Vitamin and mineral deficiencies

Vitamin B12 deficiency

PPIs may interfere with dietary protein-bound vitamin B12 absorption via the mechanism of acid-activated proteolytic digestion of dietary protein-bound vitamin B12 in the stomach.7 39 However, evidence of an effect of long-term PPI use in adults on vitamin B12 has shown conflicting results.7 40 41 Studies particularly evaluating this phenomenon in the paediatric population could not be found.


Recently, there were several reported cases of PPI-associated hypomagnesaemia in adults. Reported symptoms were ECG changes and neuroexcitability including seizures,7 but the mechanism of PPI-induced hypomagnesaemia is unknown. In contrast, no paediatric case of hypomagnesaemia has been reported. However, it is prudent to clinically monitor hypomagnesaemia-associated signs and symptoms particularly in children with long-term PPI treatment and those with risk factors for hypomagnesaemia.8

Adverse bone health/bone fractures

PPIs may compromise bone health. First, calcium absorption may be reduced in the presence of PPI since calcium ionisation necessary for intestinal absorption requires an acidic gastric environment. Second, function of osteoclasts may be inhibited by PPIs, decreasing bone resorption and influencing bone remodelling. These notions are the base for the concerns that PPIs impair bone health, increasing bone fractures, particularly in those with high-dose therapy and/or long-term treatment.42 It is important to note that most studies have been performed in adult patients with risk factors for osteoporosis and bone fractures, and that results are not consistent. Information on such a risk in the paediatric population is scarce. In their population-based, case–control study, Freedberg et al assessed PPI use and fractures in children (younger than 18 years of age) and young adults (between 18 and 29 years of age). This study has shown a higher risk of fractures in young adults (adjusted OR of 1.39) but not in children.43

Food allergy

Similar to the findings in mice, Untersmayr and Jensen-Jarolim44 observed in an observational cohort study that antiulcer treatment in humans, including H2RAs and PPIs, is associated with the development of food-specific IgE.44 One may speculate that dietary peptides remain undigested due to high gastric pH during PPI therapy, which are then presented as allergens to the immune system.7 45 Since then, no human study has been reported and the clinical significance remains to be elucidated.

Acute interstitial nephritis, acute kidney injury and the risk of chronic kidney disease

PPI use has been implicated in kidney damage such as acute kidney injury and acute interstitial nephritis.46–50 Recent studies have shown that PPI use is associated with an increased risk of chronic kidney disease.51 52 In a prospective cohort study, Lazarus et al 51 demonstrated a higher risk of incident chronic kidney disease in those receiving PPI. Similarly, Xie et al 52 showed in their large cohort study that adult patients on PPIs had a higher risk of renal dysfunction and incident chronic kidney disease than those with H2RAs. No paediatric studies have been published.

Drug interactions

PPIs can both induce and inhibit some drug-metabolising enzymes and as a consequence have the potential to provoke drug–drug interactions (DDIs) for selected pharmacological agents (eg, phenytoin, diazepam, warfarin). On the other hand, metabolism of PPIs may be altered by drugs that interfere with PPI-metabolising enzymes. Although DDIs involving PPIs through inhibition of cytochrome P450 (CYP) enzyme have been studied in adults, clinical significance has been recognised in only a few drug interactions.8 53–55

In adults, the DDI between PPIs and an antiplatelet drug, clopidogrel, through inhibition of CYP2C19 has been a focus of intense attention because reduction in CYP2C19-mediated clopidogrel conversion to its active metabolite may weaken its therapeutic effects in preventing cardiovascular events. However, a large, randomised, placebo-controlled trial in adults did not support this notion.56 One study in children suggested the occurrence of this interaction in a relatively small number of children; the Platelet Inhibition in Children on Clopidogrel trial reported that coadministration of PPI and clopidogrel may lead to reduced platelet inhibition in children with cardiac disease.8 57


Safety profiles of paediatric PPI use have yet to be fully examined, but current evidence suggests that PPIs are not free of adverse drug reactions. Rational prescribing requires deeper understanding of drug toxicity. It is our responsibility to use the scientific evidence both on efficacy and safety.58 59 Given the uncertainty in PPI safety in paediatric population, indications of PPIs should be limited to infants and children with GERD or gastric bleeding, which should be clearly distinguished from non-pathological gastrointestinal reflux. If PPIs are to be prescribed, it is important to share the current understanding on adverse effects of PPIs with patients and their family, so that an effective follow-up plan can be implemented.


View Abstract


  • Funding The clinical research of PDB is supported by the Agency for Innovation by Science and Technology in Flanders (IWT) through the SAFE-PEDRUG Project (IWT/SBO 130033).

  • Competing interests None declared.

  • Provenance and peer review Commissioned; externally peer reviewed.