High dose versus standard dose proton pump inhibitor for short term management of erosive reflux oesophagitis

  • Protocol
  • Intervention

Authors


Abstract

This is the protocol for a review and there is no abstract. The objectives are as follows:

To assess the effectiveness of high dose PPIs versus standard dose PPIs in healing oesophagitis and curing reflux symptoms in patients with erosive reflux oesophagitis. To also compare the adverse effects of high dose PPIs versus standard dose PPIs.

Background

Description of the condition

Gastro-oesophageal reflux disease (GORD) develops when the reflux of gastric contents into the oesophagus results in troublesome symptoms or complications (Rosemurgy 2011; Vakil 2006). The refluxate contains varying amounts of acid, pepsin, bile salts and trypsin. The injury caused by these agents outside the stomach may manifest as a wide range of symptoms at oesophageal and at extra-oesophageal sites. The classical symptoms of GORD are heartburn, regurgitation and chest pain. Patients with GORD who have endoscopic evidence of erosions or ulcerations in the lower end of the oesophagus are considered to have erosive oesophagitis. GORD is one of the most common medical conditions with an estimated prevalence of 10% to 20% in the western world (Dent 2005; Peery 2012). It is the most frequent gastrointestinal outpatient diagnosis in the USA and the most common indication for upper endoscopy (Peery 2012). It is associated with reduced quality of life, reduced work productivity, and places a huge financial burden on society (Dent 2005).

Description of the intervention

Proton pump inhibitors (PPIs) are effective in suppressing gastric acid production and are the cornerstone of GORD therapy (Hrelja 2011). PPIs are effective in healing oesophagitis, improving symptoms and preventing recurrent oesophageal strictures. They are generally well-tolerated with a convenient once a day dosing schedule. In patients with erosive oesophagitis, it is a common practice to double the dose of PPI with the objective of improving the healing and producing faster relief of symptoms (Meineche-Schmidt 2004). This practice is, however, based on limited evidence. Randomized controlled trials (RCTs) which have been conducted to compare the efficacy of high dose PPI with the standard dose have shown varying efficacy. The overall efficacy of such an approach needs to be systematically assessed. Moreover, the side effects of high dose PPI are likely to be higher than with standard dose PPI.

How the intervention might work

The pharmacokinetic and pharmacodynamic profiles of currently available PPIs are suboptimal, leading to lack of sustained acid suppression that results in inadequate symptom relief as well as incomplete mucosal healing. The traditional PPIs (omeprazole, lansoprazole, pantoprazole and rabeprazole) block the action of proton pumps resulting in reduction in acid secretion but have a short half life of 1 to 1.5 hours. However, proton pumps are continuously regenerated thereby restoring the capacity to secrete acid once the serum levels of the drug starts dropping. Doubling the dose of PPI provides more sustained serum levels possibly resulting in better acid suppression and translation into better healing and symptom relief.

Why it is important to do this review

Currently only the standard doses of PPIs are licensed for use in the short term treatment of reflux oesophagitis, based on the available evidence. Current guidelines by the American Gastroenterology Association recommend using a standard dose of PPI for the short term management of GORD (Kahrilas 2008). However, they opined that the data supporting the use of PPIs in doses higher than the standard dose are weak and thus have not explicitly recommended their use. The efficacy of such an approach needs to be systematically assessed given the increasing body of evidence generated in recent large multicentre trials. A Cochrane review, published in 2007, partially addressed this issue (Moayyedi 2007). However, since its publication there have been newer RCTs that have addressed the issue of the efficacy of high dose versus standard dose PPI in oesophagitis. Also, newer PPIs, mainly in the form of enantiomers of the traditional PPIs, have been added to the armamentarium of drugs available for treatment. Enantiomers of traditional PPIs have been developed with the intention to provide better stereoselective pharmacokinetic properties resulting in higher bioavailability, a more homogenous response to therapy, and better efficacy with equal safety (Abel 2010; Zhou 2008). Esomeprazole, an enantiomer of omeprazole; dexlansoprazole, an enantiomer of lansoprazole; dexrabeprazole, an enantiomer of rabeprazole; and S-pantoprazole, an enantiomer of pantoprazole have been introduced into the market. Recent RCTs have evaluated the performance of these agents in comparison to their primary molecules. There have been previous systematic reviews and narrative reviews which have focused on either esomeprazole versus other agents, dexlansoprazole versus other agents, or the role of high dose PPI in a severe erosive oesophagitis subset and have made comparisons between various drug classes (Edwards 2001; Edwards 2006; Gralnek 2006; Klok 2003; Kushner 2011). Thus a systematic, comprehensive update is necessary in order to inform decision makers about the best available evidence for management of this condition.

Moreover, the FDA has recently issued warnings on the use of PPIs as recent studies have found that their long term use is associated with serious side effects in the form of increased risk of fractures, low magnesium levels, pneumonia and C difficile infection (Eom 2011; Moayyedi 2012). Also, high dose therapy has to be cost-effective in order to justify the expenditure incurred by government or the patient. Thus a pragmatic balance is necessary in assessing the risks versus benefits in prescribing high dose PPI therapy in the short term management of reflux oesophagitis.

Objectives

To assess the effectiveness of high dose PPIs versus standard dose PPIs in healing oesophagitis and curing reflux symptoms in patients with erosive reflux oesophagitis. To also compare the adverse effects of high dose PPIs versus standard dose PPIs.

Methods

Criteria for considering studies for this review

Types of studies

Randomized controlled clinical trials (RCTs) comparing the efficacy of high dose PPI versus standard dose PPI in the management of erosive reflux oesophagitis.

Types of participants

Studies evaluating the role of PPIs in adult patients (age ≥ 18 years) with endoscopically proven erosive reflux oesophagitis.

Types of interventions

PPIs which will be considered are: lansoprazole, omeprazole, pantoprazole, rabeprazole, esomeprazole, dexlansoprazole, dexrabeprazole and S-pantoprazole.

High dose PPI will be defined as at least double the standard dose.

Standard dose PPI will be defined as the dose recommended in the drug formulary (lansoprazole 30 mg, omeprazole 20 mg, pantoprazole 40 mg, rabeprazole 20 mg per day). For esomeprazole the standard dose will be defined as 20 mg per day as it is an enantiomer of omeprazole whose standard dose is 20 mg a day. For dexlansoprazole the standard dose will be defined as 30 mg per day as it is an enantiomer of lansoprazole whose standard dose is 30 mg a day. For dexrabeprazole the standard dose will be defined as 20 mg per day as it is an enantiomer of rabeprazole whose standard dose is 20 mg a day. For S-pantoprazole the standard dose will be defined as 40 mg per day as it is an enantiomer of pantoprazole whose standard dose is 40 mg a day. 

Types of outcome measures

Trials will be included if they have reported evidence of assessment of oesophagitis or reflux symptoms, or both, at between two to 12 weeks.  For purposes of analysis, the latest time point will be taken within this duration.

Primary outcomes
  1. The proportion of patients with persistent oesophagitis.

  2. The proportion of patients with any persisting reflux oesophagitis symptoms. Reflux symptoms will be defined as heartburn, regurgitation, or both of these features (global symptom outcome). When more than one of these symptoms are listed as an outcome, each will be analysed separately.

Secondary outcomes
  1. The proportion of patients with any improvement in reflux symptoms.

  2. Number and nature of adverse events.

Search methods for identification of studies

Electronic searches

We will conduct a literature search to identify all published and unpublished RCTs. The literature search will identify potential studies in all languages. We will translate the non-English language papers and fully assess them for potential inclusion in the review as necessary. The search strategies are given in the appendices. We will use the following electronic search databases for identifying potential studies:

Searching other resources

One review author will handsearch the published abstracts from the conference proceedings in Digestive Disease Week (published in Gastroenterology and Gastrointestinal Endoscopy) and the United European Gastroenterology Week (published in Gut) from 1 January 2007 to the present date. We will contact the authors of trial reports published only as abstracts. We will also handsearch references cited in studies found by the above search to identify further relevant trials. To find ongoing or recently finished trials, www.clinicaltrials.gov will be searched and the authors of these trials will be contacted for information.

Data collection and analysis

Selection of studies

After one author has screened the titles using broad criteria, two authors will independently examine the studies for eligibility. We will assess each trial for potential duplicate publication. We will resolve differences by discussion and consensus with a third review author. The same two review authors will retrieve and review the complete reports of all the selected articles. The review authors will initially pilot the method of selection for 10 studies.

Data extraction and management

Two review authors will independently record the study and patient characteristics in a pre-designed pro forma (Appendix 4). All data transcription will be checked for accuracy. Any disagreement will be resolved by consensus with a third review author.

Assessment of risk of bias in included studies

Two review authors will independently assess the methodological quality of the included studies using the GRADE approach. We will resolve disagreement by discussion and consensus with a third review author. The original authors may be contacted for further clarification as necessary. The following criteria to assess study quality will be used.

Random sequence generation

Generation of the allocation sequence, to assess selection bias, will be considered to be: adequate if the allocation sequence was generated by a computer, random number table, or drawing of lots, tossing of a coin, shuffling of cards or throwing dice if a person who was not otherwise involved in the recruitment of participants performed the procedure; unclear if the trial was described as randomized but the method used for generation of the allocation sequence was not described; and inadequate if a system involved using non-random or pseudo-random methods such as dates, names or admittance numbers.

Allocation concealment

Allocation concealment, to assess selection bias, will be considered to be: adequate if the allocation of patients involved a central independent unit, on-site locked computer or opaque sealed envelopes; unclear if the trial was described as randomized but the method used to conceal the allocation was not described; and inadequate if the allocation sequence was known to the investigators who assigned participants.

Blinding of participants and personnel

Blinding of participants and personnel, to assess performance bias, will be considered to be: adequate if the trial was described as blind to participants or assessors and the method of blinding was described; unclear if the trial was described as double-blind, but the method of blinding was not described; and inadequate if the trial was not blinded.

Blinding of outcome assessment

Blinding of outcome assessors, to assess detection bias, will be considered to be: adequate if the trial was described as blind to participants or assessors and the method of blinding was described; unclear if the trial was described as double-blind, but the method of blinding was not described; and inadequate if the trial was not blinded.

Incomplete outcome data

Study completion and follow-up, to assess attrition bias, will be considered to be: adequate if the numbers and reasons for dropouts or withdrawals and losses to follow-up in all intervention groups were described, or if it was specified that there were no dropouts or withdrawals; unclear if the report gave the impression that there had been no dropouts or withdrawals and no losses to follow-up but this was not specifically stated; and inadequate if the number or reasons for dropouts and withdrawals as well as losses to follow-up were not described.

Selective outcome reporting

Selective reporting bias will be considered to be: absent if the outcomes in all intervention groups and reasons for exclusions were described, or if it was specified that there were no missing outcome data; unclear if the report gave the impression that there had been no missing outcome data, but this was not specifically stated; and present if the outcomes were not described, or some of the outcome data were missing.

Other bias

Other risk of bias will be considered to be present if there are other threats to validity such as baseline imbalances between randomized groups, early stopping of the trial, the primary end point in the study is convoluted and is not intuitive or standard. We will report any other important concerns about bias identified in the studies.

Measures of treatment effect

We will express dichotomous outcomes as risk ratios (RR) of persistent oesophagitis or symptoms with 95% confidence intervals (CI) and express continuous outcomes as mean differences (MD) with 95% CIs. We will not exclude from the meta-analysis trials with zero events in both arms. We will calculate the numbers of patients needed to treat (or harm) by 1/(risk difference), expressed with 95% CI. For outcomes for which data are not reported or are reported in a different format, we will contact authors for clarification. We will use the intention-to-treat (ITT) sample (all randomised patients) in the analysis.

Unit of analysis issues

We will use the patient as the unit of analysis as we do not anticipate any variations on the standard parallel group, randomised trial design.

Dealing with missing data

We will contact authors for any missing data from the included studies. The primary analysis will be an ITT analysis which will include all randomised patients in the denominator. The missing patients will be assumed to have had the worst case scenario in both the active and comparator arms.

Assessment of heterogeneity

We will assess heterogeneity with the Chi² test (P < 0.15 for significant heterogeneity) and I² statistic (> 25% for heterogeneity) using a random-effects model. If significant heterogeneity exists, we will investigate possible explanations by subgroup analysis as described in the protocol below.

Assessment of reporting biases

We will estimate publication bias by examining the relationship between the treatment effects and the standard errors of the estimates using a funnel plot.

Data synthesis

We will conduct meta-analysis to assess the overall outcome in the high dose arm versus the standard dose arm with respect to the following outcomes:

  1. proportion of patients with persistent oesophagitis on endoscopy,

  2. proportion of patients with persistent symptoms of reflux oesophagitis,

  3. proportion of patients with any improvement in reflux symptoms,

  4. total number of adverse events, and

  5. number of individual adverse events.

We will calculate the RR of the incidence of endoscopic healing as the primary outcome. We will calculate the RRs of other dichotomous secondary outcomes including symptom relief and adverse events. We will use a random-effects model and the fixed effect model as a sensitivity analysis. Where significant heterogeneity is detected (Chi² test P < 0.15 or I² statistic > 25%), we will investigate possible explanations by assessing the importance of the predefined factors listed in the protocol. We will use the Cochrane Review Manager 5.1 software (RevMan 2011) to carry out the analysis.

Unweighted inter-rater agreement between two review authors for evaluation of the studies for eligibility and rating risk of bias will be assessed using kappa statistics.

Subgroup analysis and investigation of heterogeneity

The following will be prospectively evaluated as a cause for any heterogeneity in the study outcomes.

  1. Dose of drug.

  2. Drug class (traditional PPI versus enantiomers).

  3. Drug formulation (e.g. sustained release versus standard release).

  4. Duration of treatment.

  5. Timing of outcome assessment, < 4 weeks or ≥ 4 weeks.

  6. Severity of reflux oesophagitis included. The influence of the proportions of mild, moderate and severe reflux oesophagitis in each study on the outcome will be assessed. Mild or moderate is defined as Savary-Miller grade 1 to 2 or Los Angeles grade A to B, whilst severe is equivalent to Savary-Miller grade 3 to 5 or Los Angeles grade C to D (Lundell 1999; Savary 1978).

Sensitivity analysis

We will perform an a priori sensitivity analysis to assess the robustness of our conclusions. This will involve ITT versus per protocol (PP) analysis; and meta-analysis modelling (fixed-effect versus random-effects).

Appendices

Appendix 1. CENTRAL search strategy

  1. exp gastroesophageal reflux/

  2. (gastroesophageal adj3 reflux).tw.

  3. (gastro adj3 esophageal adj3 reflux).tw.

  4. (gastrooesophageal adj3 reflux).tw.

  5. (gastro adj3 oesophageal adj3 reflux).tw.

  6. (GERD).tw.

  7. (GORD).tw.

  8. (esophageal or oesophageal or esophagus or oesophagus).tw.

  9. (acid adj3 reflux).tw.

  10. 8 and 9

  11. exp heartburn/

  12. heartburn.tw.

  13. esophagitis/ or esophagitis, peptic/

  14. (esophagitis or oesophagitis).tw.

  15. or/1-7,10-14

  16. *proton pump inhibitor/

  17. (proton adj2 pump adj2 inhibitor$).tw.

  18. (PPI or PPIs).tw.

  19. *Omeprazole/

  20. omeprazole.tw.

  21. pantoprazole.tw.

  22. (lansoprazole or lanzoprazole).tw.

  23. rabeprazole.tw.

  24. esomeprazole.tw.

  25. dexrabeprazole.tw

  26. dexlansoprazole.tw.

  27. S-pantoprazole.tw.

  28. or/16-27

  29. 15 and 28

Appendix 2. MEDLINE search strategy

  1. randomized controlled trial.pt.

  2. controlled clinical trial.pt.

  3. randomized.ab.

  4. placebo.ab.

  5. clinical trials as topic.sh.

  6. randomly.ab.

  7. trial.ti.

  8. or/1-7

  9. exp animals/ not humans.sh.

  10. 8 not 9

  11. exp gastroesophageal reflux/

  12. (gastroesophageal adj3 reflux).tw.

  13. (gastro adj3 esophageal adj3 reflux).tw.

  14. (gastrooesophageal adj3 reflux).tw.

  15. (gastro adj3 oesophageal adj3 reflux).tw.

  16. (gord).tw.

  17. gerd.tw.

  18. (acid adj3 reflux).tw.

  19. exp heartburn/

  20. esophagitis/ or esophagitis, peptic/

  21. (esophagitis or oesophagitis).tw.

  22. heartburn.tw.

  23. or/11-22

  24. *proton pump inhibitor/

  25. (proton adj2 pump adj2 inhibitor$).tw.

  26. (PPI or PPIs).tw.

  27. *Omeprazole/

  28. omeprazole.tw.

  29. pantoprazole.tw.

  30. (lansoprazole or lanzoprazole).tw.

  31. rabeprazole.tw.

  32. esomeprazole.tw.

  33. dexrabeprazole.tw

  34. dexlansoprazole.tw.

  35. S-pantoprazole.tw.

  36. or/24-35

  37. 10 and 23 and 36

  38. limit 37 to ed=19800101-20120131

Appendix 3. EMBASE search strategy

  1. random:.tw. or placebo:.mp. or double-blind:.tw.

  2. exp gastroesophageal reflux/

  3. (gastroesophageal adj3 reflux).tw.

  4. (gastro adj3 esophageal adj3 reflux).tw.

  5. (gastrooesophageal adj3 reflux).tw.

  6. (gastro adj3 oesophageal adj3 reflux).tw.

  7. (GORD).tw.

  8. (GORD).tw.

  9. (esophageal or oesophageal or esophagus or oesophagus).tw.

  10. (acid adj3 reflux).tw.

  11. 9 and 10

  12. heartburn/

  13. heartburn.tw.

  14. esophagitis/ or reflux esophagitis/ or ulcerative esophagitis/

  15. (esophagitis or oesophagitis).tw.

  16. or/2-8,11-15

  17. *proton pump inhibitor/

  18. (proton adj2 pump adj2 inhibitor$).tw.

  19. (PPI or PPIs).tw.

  20. *omeprazole/

  21. omeprazole.tw.

  22. *pantoprazole/

  23. pantoprazole.tw.

  24. *lansoprazole/

  25. (lansoprazole or lanzoprazole).tw.

  26. *rabeprazole/

  27. rabeprazole.tw.

  28. *esomeprazole/

  29. esomeprazole.tw.

  30. dexrabeprazole.tw

  31. dexlansoprazole.tw.

  32. S-pantoprazole.tw.

  33. or/17-32

  34. 1 and 16 and 33

Appendix 4. Data extraction

The predefined pro forma containing a structured format to extract the following information in Excel spreadsheet. 

  • Setting (single or multi-centre)

  • Setting: primary or secondary care

  • Country of origin;

  • Enrolment period;

  • Year of publication, format (abstract or journal article);

  • Study design;

  • Inclusion and exclusion criteria used;

  • Mean age of patient included in the study

  • Dose of the drug used

  • Dosing pattern (single or double)

  • Nature of the drug class  (e.g. omeprazole/ lansoprazole/ pantoprazole/ rabeprazole/ esomeprazole/ dexlansoprazole/ S-pantoprazole and dexrabeprazole; traditional PPI or enantiomers)

  • Duration of treatment

  • Baseline comparability between treatment groups

  • Treatments compared and number of patients in each arm

  • Drop-outs reported and their reasons

  • Grading system for reflux oesophagitis (e.g. Savary-Miller, Los Angeles)

  • Severity of reflux included. The influence of the proportions of mild/ moderate/ severe reflux oesophagitis in each study on the outcome will be assessed. Mild/moderate is defined as Savary-Miller grade 1-2 or Los Angeles grade A-B, whilst severe is equivalent to Savary-Miller grade 3-5 or Los Angeles grade C-D.

  • GORD questionnaires: validated or not

  • Adverse events: the total number reported

  • Nature of adverse reactions

  • Study quality (generation of allocation sequence, allocation concealment, blinding, incomplete outcome data, selective reporting, other bias).

Contributions of authors

  • Draft the protocol: all authors

  • Develop search strategy: Dr Usha Dutta, Racquel Simpson

  • Search for trials: Dr Usha Dutta, Racquel Simpson

  • Select trials to include: Dr Usha Dutta, Dr Yuhong Yuan (Dr Paul Moayyedi as arbiter)

  • Extract data from the trials: Dr Usha Dutta, Dr Yuhong Yuan

  • Enter data into RevMan: Dr Usha Dutta

  • Carry out the analysis: Dr Usha Dutta, Dr G Leontiadis

  • Interpret the analysis: all authors

  • Draft the final review: all authors

Declarations of interest

Paul Moayyedi has received speaker's fees and acted as an independent medical advisor for AstraZeneca and Johnson & Johnson. His chair is partly funded by an unrestricted donation by AstraZeneca to McMaster University. Dr G Leontiadis has received research funding from Astra Zeneca. The remaining authors declare no conflict of interest.

Sources of support

Internal sources

  • McMaster University, Canada.

External sources

  • No sources of support supplied

Notes

This protocol has been split from a previous version (Moayyedi 2007).

Ancillary