Proton pump inhibitors: too much of a good thing? (Prescribing Practice Review)

Published in MedicineWise News

Date published: About this date

Clinical content may change after this date. This information is not intended as a substitute for medical advice from a qualified health professional. Health professionals should rely on their own expertise and enquiries when providing medical advice or treatment.

Key messages

  • Establish whether ongoing proton pump inhibitor (PPI) therapy is necessary in each patient.
  • Decrease PPI use to low doses or intermittent, symptom-driven therapy once symptoms of gastro-oesophageal reflux disease (GORD) are controlled.
  • All PPIs are very effective in controlling GORD symptoms and are clinically equivalent in most patients.
  • Consider testing for and treating Helicobacter pylori (H. pylori) in people with uninvestigated dyspepsia or who are using PPIs long term.

PPI prescribing is growing

PPIs are effective, well tolerated drugs for relieving symptoms that can be debilitating and concerning for patients. A marked increase in PPI prescribing followed the removal of the PBS authority listing in 2001 and the number of PPI prescriptions written continues to grow (Figure 1).1 Is the growth in PPI prescribing justified by improved outcomes for patients?

 
Prescribing of PPIs 2000 to 2003

Figure 1: Prescribing of PPIsA

A. Source: HIC PBS item reports.1 Number of prescriptions refers to the volume of services processed by HIC. Year refers to year that service was processed by HIC, not the date of prescribing or the date of supply.

Before starting a PPI…

Is investigation needed?

Anyone with alarm symptoms (such as difficulty or pain on swallowing, unexplained weight loss, evidence of GI bleeding, recurrent vomiting or upper abdominal mass) should be referred for investigation.2

There is currently debate over the appropriate age threshold for early endoscopy in people with dyspepsia, with investigation variously advised for people aged over 45 or 55 years.3,4 The Gastroenterological Society of Australia (GESA) suggests that people with mild, typical reflux symptoms and no alarm symptoms be given a trial of therapy without investigation.2

Are drugs causing or exacerbating symptoms?

Where appropriate, stop, replace or adjust drug regimens that may be causing symptoms, such as nonsteroidal anti-inflammatory drugs (NSAIDs), calcium channel blockers, bisphosphonates, nitrates and theophylline.3

Suggest lifestyle changes

Before seeking treatment, people will often have identified foods or activities that exacerbate their symptoms. Reinforce that continuing to avoid these while taking a PPI will help to prevent breakthrough symptoms.

Losing weight, stopping smoking, healthy eating and moderating alcohol intake can also be suggested; although evidence for these measures in improving symptoms is lacking, all have general health benefits and may be effective in individual patients.

Establish a treatment plan

Initiate treatment with a PPI based on an explicit goal, such as control of reflux symptoms or ulcer healing.

Review the need for ongoing therapy in every patient

Plan to review the success of initial treatment with a view to reducing or ceasing PPIs as appropriate

The fact that approximately twice as many prescriptions are written for continuation than for initiation of PPIs underlines the importance of establishing the need for ongoing therapy.5

In any patient presenting for a repeat prescription, consider whether ongoing treatment with a PPI is warranted. If a PPI has been initiated during hospitalisation, review the need for it after discharge.

Managing GORD

Use a step-down approach

The step-down approach has gained popularity because it rapidly achieves the goals of initial therapy and minimises the need for repeat consultations.2 For long-term management, reduce or cease use of PPIs where possible; ongoing daily standarddose PPIs are often unnecessary.

Initiate with a PPI to aid diagnosis, control symptoms, reassure patients, heal oesophagitis2

Diagnosis of GORD is usually based on the presence of heartburn or acid regurgitation as the predominant symptom. Patient understanding of the term ‘heartburn’ is variable; asking about ‘a burning feeling rising up from the stomach or lower chest towards the neck’ may identify GORD more accurately.2

A clear response to PPI therapy supplements a symptom-based diagnosis and can help to reassure patients that their symptoms are not the result of serious underlying disease. Endoscopy has a limited role in routine diagnosis of GORD but is indicated if the diagnosis is unclear, symptoms are suggestive of severe or complicated oesophagitis, or alarm symptoms are present.2

Initiate with a trial of a PPI at a standard dose for 4 weeks. Those with insufficient response should receive a further 4 weeks’ treatment.6 If this fails to control symptoms, doubling the PPI dose may be effective; consider seeking specialist advice.2

Following PPI therapy, the absence of symptoms is related to healing of oesophagitis; endoscopy to confirm healing is usually unnecessary in uncomplicated GORD.3

Decrease PPI use once GORD symptoms are controlled

The goals of long-term treatment for GORD are to maintain symptom control and prevent complications, while minimising costs.2 After a successful initial course of treatment, try reducing PPI use while monitoring symptoms. Needs for ongoing maintenance therapy vary widely; aim for the lowest dose that maintains symptom control.

Options for step-down

Low-dose PPIs

Taken daily, a low-dose PPI maintains endoscopic remission and symptom control in a substantial proportion of people with uncomplicated healed oesophagitis.7–11 Currently, less than 10% of all PPI prescriptions are for lower strengths.1

Intermittent, symptom-driven use

Patients are advised to take a PPI on days when symptoms occur. Although intermittent use allows symptoms to recur, the vast majority of patients in clinical trials have been willing to continue with this strategy after 6 months.12–14

Ceasing PPIs

Manage symptoms with lifestyle changes, antacids and histamine-2 receptor (H2) antagonists if needed. Some patients with milder disease will not relapse when PPI treatment is withdrawn. GESA endorses a trial of treatment withdrawal to identify these patients; those who relapse can be treated with a repeat course of the initially successful therapy, then treatment stepped down to the lowest dose that maintains symptom control.2

Exceptions to step-down

People known to have severe oesophagitis will relapse unless they continue to take PPIs daily.2,6 Those with complications such as strictures, scleroderma or Barrett’s oesophagus also require daily PPIs at standard or higher doses.6

Consider step-up for mild GORD symptoms

In people with mild or intermittent symptoms, consider a step-up approach: initiate with lifestyle measures and antacids, then switch to an H2 antagonist, then a PPI, if further symptom control is required.

Managing patient expectations

Explain the treatment plan

Patients can be reluctant to reduce their PPI dose when they have experienced profound symptom relief with drug therapy. Explaining that PPI treatment is directed at controlling symptoms, rather than curing the condition, may increase acceptance of suggested changes to treatment.

Patients’ concerns about the safety of long-term use of medicines may lead them to take PPIs intermittently15; reinforce that on-demand use of PPIs is appropriate for many patients.

Choosing a PPI

All PPIs are very effective and clinically equivalent in most patients

PPIs are superior to H2 antagonists for healing oesophagitis16 and resolving symptoms in short-term empirical treatment.17

Studies have found no significant differences in clinical efficacy in oesophagitis between most PPIs.16 Some efficacy differences have been reported for esomeprazole, and there has been large uptake of this drug since its PBS-listing in August 2002.1 Does esomeprazole provide a clinically significant benefit over other PPIs for people with GORD?

Esomeprazole is the s-enantiomer of omeprazole

Omeprazole is a racemate; that is, a mixture of equal amounts of two enantiomers, r- and s-omeprazole. Enantiomers are isomers that are mirror images of one another. Esomeprazole and r-omeprazole have the same inhibitory effect at the proton pump18, but are metabolised differently: after equal milligram doses, esomeprazole reaches much higher plasma concentrations than omeprazole.19

Clinical studies have used higher doses of esomeprazole than of comparator PPIs…

Omeprazole 20 mg has been compared to esomeprazole 20 mg or 40 mg. Thus, inequivalent doses of esomeprazole and other PPIs have been used to assess comparative clinical efficacy.

Although higher doses of esomeprazole have been used, studies indicate that few additional people will benefit from using it instead of another PPI.20–23

…but have found that esomeprazole provides a limited additional benefit

The clinical advantage of esomeprazole over other PPIs is limited to a relatively small benefit in people with (either active or healed) erosive oesophagitis, who make up less than half of all patients with reflux symptoms.2 In comparative clinical trials in erosive oesophagitis, esomeprazole 40 mg, omeprazole 20 mg and lansoprazole 30 mg have all produced 8-week healing rates of over 80%; differences between treatment groups have ranged from 4–10%.B20–22 For maintenance treatment in healed oesophagitis, 11 people need to be treated for 6 months with esomeprazole 20 mg instead of lansoprazole 15 mg to prevent one additional relapse.B,23

B. Note that standard and low doses of esomeprazole are more expensive than corresponding doses of omeprazole and lansoprazole.24

Uninvestigated dyspepsia

Symptoms do not reliably predict diagnosis in uninvestigated dyspepsia

Heartburn and acid regurgitation as predominant symptoms are relatively specific indicators of the presence of GORD; however, symptoms do not reliably predict other diagnoses such as peptic ulcer disease or non-ulcer dyspepsia.

Management options in uninvestigated dyspepsia include prompt endoscopy, empirical acid suppression therapy or the H. pylori test-and-treat approach.

Empirical therapy or test-and-treat is generally preferred. Endoscopy is usually reserved for those at risk of serious pathology (that is, over 45 yearsC or with alarm symptoms) and those whose symptoms persist after initial therapy.

Consider test-and-treat in uninvestigated dyspepsia

Test-and-treat refers to a strategy in which patients presenting with uninvestigated dyspepsia (excluding those with indications for prompt endoscopy, with suspected GORD or who are NSAID users) receive a non-invasive test for H. pylori (such as the urea breath test, faecal antigen test or serology). Those who test positive receive eradication therapy and those who test negative receive a short course of a PPI. People whose symptoms persist after confirmed H. pylori eradication or an adequate trial of a PPI can be referred to a specialist.

A recent study found that the test-and-treat approach reduced symptoms and rates of referral for endoscopy more than empirical acid suppression.25 However, evidence for this approach in primary care is still limited: much of the current evidence comes from studies considering the efficacy of eradication therapy in H. pylori-infected subjects in secondary care settings.26 Nevertheless, many guidelines now recommend this approach.4,27,28 GESA suggests that it is reasonable to consider eradicating H. pylori in dyspeptic patients younger than 50 years without further investigations if there are no alarm features.29

C. Some guidelines suggest that a non-invasive H. pylori test-and-treat approach may be as appropriate as early endoscopy in patients aged over 55 years who are not taking NSAIDs and do not have alarm symptoms.4

Non-ulcer dyspepsia

What is non-ulcer dyspepsia?

Non-ulcer or functional dyspepsia describes people who have had investigations to rule out structural or biochemical causes for their dyspeptic symptoms.

Drugs are generally not effective in non-ulcer dyspepsia

PPIs, H2 antagonists, H. pylori eradication and motility stimulants have all been evaluated in non-ulcer dyspepsia; each is effective only in small numbers of patients.30,31 Non-ulcer dyspepsia is thought to encompass a range of underlying causes, including abnormal gastrointestinal motility, acid sensitivity and H. pylori infection; this may explain the lack of a single effective therapy.

Explanation and reassurance are key aspects of management

Reassure patients that although symptoms are understandably troubling, they are part of a common condition and are not due to serious underlying disease. Explain that treatments are not usually very effective, although a small number of people may be helped by some medicines.

Consider test-and-treat

Eradicating H. pylori in infected people with non-ulcer dyspepsia improves or eliminates symptoms in a small proportion: 15 people must be treated for one to benefit.31 Although the effect size is small, eradication can be considered because it eliminates the need for ongoing therapy in responders. However, Australian information on the cost-effectiveness of this approach is not available.

Consider a short course of an H2 antagonist or PPI in symptomatic patients in whom H. pylori has been excluded or eradicated

Short courses of H2 antagonists or PPIs improve symptoms in 10–20% of people with non-ulcer dyspepsia.30 There is currently no evidence that PPIs are more effective than H2 antagonists in non-ulcer dyspepsia30; use an H2 antagonist for 4 weeks first because it is less expensive, but consider switching to a PPI for a further 4 weeks if the patient fails to respond.3 Encourage intermittent, short courses of treatment as needed.

Role of motility stimulants

Cisapride is the best-studied motility stimulant for non-ulcer dyspepsia30 but its use has been restricted due to the potential for serious cardiac arrhythmias. Evidence for metoclopramide and domperidone is insufficient to support their use in this indication.

Eradicating H. pylori in long-term PPI users

Benefits are currently uncertain

It has been proposed that long-term acid suppression in the presence of H. pylori infection accelerates the development of atrophic gastritis, which may lead to gastric cancer. At present, evidence for PPIs accelerating H. pylori gastritis is conflicting.28 In the absence of definitive evidence, some guidelines suggest testing for and eradicating H. pylori in people on long-term PPIs to reduce the potential associated risks.28

Where people on long-term PPIs have underlying peptic ulcer disease, subsequent cure of the ulcer following H. pylori eradication should allow PPIs to be discontinued.28

Table 1: Standard and low doses of PPIs6

PPI  

Standard doseD Low doseD
esomeprazole 40 mg daily 20 mg daily
lansoprazole 30 mg daily 15 mg daily
omeprazole  20 mg daily 10 mg daily
pantoprazole   40 mg daily  20 mg daily
rabeprazole 20 mg daily 10 mg daily
D. Standard dose refers to the dose usually recommended for initial therapy in reflux oesophagitis. Low dose refers to the lower dose recommended for maintenance therapy.
References
  1. Health Insurance Commission PBS statistics. (www.hic.gov.au/providers/health_statistics/statistical_reporting/pbs.htm, accessed 19 February 2004.)
  2. Gastroenterological Society of Australia. Gastro-oesophageal reflux disease in adults: guidelines for clinicians. 3rd edition, 2001.(www.gesa.org.au/members_guidelines/goreflux/01.htm, accessed 19 February 2004.)
  3. Australian Medicines Handbook 2004.
  4. Scottish Collegiate Guidelines Network. Guideline 68. Dyspepsia: a national clinical guideline, March 2003. (www.sign.ac.uk/pdf/sign68.pdf, accessed 19 February 2004).
  5. Data provided by Health Communication Network from the General Practice Research Network.
  6. Therapeutic Guidelines: Gastrointestinal. Version 3, 2002.
  7. Birbara C et al. Eur J Gastroenterol Hepatol 2000;12:889–97.
  8. Plein K et al. Eur J Gastroenterol Hepatol 2000;12:425–32.
  9. Escourrou J et al. Aliment Pharmacol Ther 1999;13:1481–91.
  10. Thjodleifsson B et al. Dig Dis Sci 2000;45:845–53.
  11. Robinson M et al. Ann Intern Med 1996;124:859–67.
  12. Lind T et al. Aliment Pharmacol Ther 1999;13:907–14.
  13. Talley N et al. Eur J Gastroenterol Hepatol 2002;14:857–63.
  14. Talley N et al. Aliment Pharmacol Ther 2001;15:347–54.
  15. Hungin A et al. Br J Gen Pract 1999;49:463–4.
  16. Clin Evid, Issue 10, 2003.
  17. van Pinxteren B et al. Short-term treatment with proton pump inhibitors, H2 receptor antagonists and prokinetics for gastro-oesophageal reflux disease-like symptoms and endoscopy negative reflux disease. (Cochrane Review). In: The Cochrane Library, Issue 1, 2004. Chichester, UK: John Wiley and Sons, Ltd.
  18. Lindberg P et al. Aliment Pharmacol Ther 2003;17:481–8.
  19. Andersson T et al. Aliment Pharmacol Ther 2001;15:1563–9.
  20. Castell DO et al. Am J Gastroenterol 2002;97:575–83.
  21. Richter JE et al. Am J Gastroenterol 2001;96:656–65.
  22. Kahrilas PJ et al. Aliment Pharmacol Ther 2000;14:1249–58.
  23. Lauritsen K et al. Aliment Pharmacol Ther 2003;17:333–41.
  24. Schedule of Pharmaceutical Benefits, 1 February 2004.
  25. Manes G et al. BMJ 2003;326:1118.
  26. Delaney BC et al. Initial management strategies for dyspepsia (Cochrane Review). In: The Cochrane Library, Issue 1, 2004. Chichester, UK: John Wiley and Sons, Ltd.
  27. PRODIGY Guidance–Dyspepsia–symptoms. Last revised April 2002. (www.prodigy.nhs.uk/guidance.asp?gt=Dyspepsia%20-%20symptoms, accessed 19 February 2004.)
  28. Malfertheiner P et al. Aliment Pharmacol Ther 2002;16:167–80.
  29. Gastroenterological Society of Australia. Helicobacter pylori: Guidelines for Healthcare Providers. (www.gesa.org.au/members_guidelines/helicobacter/ index.htm, accessed 9 March 2004.)
  30. Moayyedi P et al. Pharmacological interventions for non-ulcer dyspepsia (Cochrane Review). In: The Cochrane Library, Issue 1, 2004. Chichester, UK: John Wiley and Sons, Ltd.
  31. Moayyedi P et al. Eradication of Helicobacter pylori for non-ulcer dyspepsia (Cochrane Review). In: The Cochrane Library, Issue 1, 2004. Chichester, UK: John Wiley and Sons, Ltd.