1. Top of page
  2. Abstract
  7. References


Esomeprazole is a new proton pump inhibitor, which has been compared to omeprazole for the treatment of reflux oesophagitis in clinical trials.


To compare the effectiveness of esomeprazole with the recommended dose of proton pump inhibitors in the healing of reflux oesophagitis, using omeprazole as a common comparator.


Systematic review of randomized controlled trials. Extraction and re-analysis of data to provide ‘intention-to-treat’ results. Meta-analysis using a Fixed Effects model.


A meta-analysis of healing rates of esomeprazole 40 mg compared to omeprazole 20 mg gave the following results: at 4 weeks (relative risk 1.14; 95% CI: 1.10, 1.18) and 8 weeks (RR 1.08; 95%CI: 1.05, 1.10). Other proton pump inhibitors compared to omeprazole 20 mg are as follows: lansoprazole 30 mg at 4 weeks (RR 1.02; 95%CI: 0.97, 1.08) and 8 weeks (RR 1.01; 95%CI: 0.97, 1.06); pantoprazole 40 mg at 4 weeks (RR 0.99; 95%CI: 0.91, 1.07) and 8 weeks (RR 0.98; 95%CI: 0.93, 1.04); rabeprazole 20 mg at 4 weeks (RR 1.00; 95%CI: 0.87, 1.14) and 8 weeks (RR 0.98; 95%CI: 0.91, 1.05).


Esomeprazole has demonstrated higher healing rates than omeprazole at 4 and 8 weeks. Other proton pump inhibitors (lansoprazole, pantoprazole and rabeprazole) have not shown higher healing rates when compared with omeprazole.


  1. Top of page
  2. Abstract
  7. References

Gastro-oesophageal reflux disease (GERD) affects up to 10% of the adult population daily, usually in the form of heartburn.1 This figure is based only on reported cases and probably underestimates the true prevalence of the disease. Proton pump inhibitors are very effective for healing reflux oesophagitis and have been shown to be superior to H2-receptor antagonists,2 both in terms of healing erosions/lesions and in relieving associated symptoms.3 This is likely to result from a greater suppression of 24 h intragastric acidity, as there is a significant correlation between the degree of acid suppression and oesophageal healing rates.4 Currently, five proton pump inhibitors are licensed in the UK for the treatment of reflux oesophagitis: omeprazole 20 mg, lansoprazole 30 mg, pantoprazole 40 mg, rabeprazole 20 mg and the new proton pump inhibitor, esomeprazole 40 mg, but few direct comparisons of their clinical effectiveness are available.

Esomeprazole is the S-isomer of omeprazole. Pharmacokinetic and pharmacodynamic studies have shown that esomeprazole is subject to less first-pass hepatic metabolism and lower plasma clearance than omeprazole. The area under the plasma concentration–time curve (AUC) has been shown to be 5-fold higher for esomeprazole 40 mg compared with omeprazole 20 mg.5 This results in greater and more consistent acid suppression.5, 6

Previously there have been several attempts to summarize the effect of treatments for reflux oesophagitis by systematic review and meta-analysis.2, 3, 7 However the focus of these previous works has been either broad, encompassing proton pump inhibitors and H2-receptor antagonists, or narrow, comparing only two proton pump inhibitors. This current work is the first systematic review that seeks to establish if there is a significant difference in efficacy between UK licensed doses of proton pump inhibitors, including the new proton pump inhibitor, esomeprazole. Omeprazole was chosen as the common comparator because it is the most common proton pump inhibitor that is used in clinical trials with other proton pump inhibitors and provides the largest pool of comparable trials for review. A strict end-point of endoscopic healing of oesophagitis at 4 or 8 weeks was employed for efficacy comparison in preference to symptom assessment, because the latter is not assessed consistently from study to study while the former can be objectively evaluated.


  1. Top of page
  2. Abstract
  7. References


EMBASE, Medline, BIOSIS and AstraZeneca’s internal database were searched for abstracts and papers. These searches were completed in December 2000.

The exact search strategy had to be modified to comply with the functionality of the individual databases but followed the following format: ((acute or heal* or short-term) AND (terms = reflux oesophagitis or terms = GERD) AND (((terms = 20 mg AND terms = clinical trial rabeprazole) OR (terms = 40 mg AND terms = clinical trial pantoprazole) OR (terms = clinical trial lansoprazole AND terms = 30 mg)) AND (terms = clinical trial omeprazole AND (terms = (20 mg or 40 mg))))) AND (random* AND terms = randomized).


The criteria for selection of trials for inclusion in the review were a study design that was a randomized controlled trial, direct comparison of omeprazole 20 mg with any UK licensed healing dose of proton pump inhibitor, which contained endoscopic healing data at 4 weeks and/or 8 weeks.

Quality assessment

The methodological quality of trials was assessed using the criteria devised by Jadad et al.8 The elements appraised include: study design, level of blinding, method of randomization and proportion of patients lost to follow up. However, because quality scoring is open to interpretation and this review was envisaged to be as all encompassing as possible, we performed a sensitivity analysis using all papers found.

Data extraction

Four- and 8-week healing data were taken from qualifying trials and recalculated if not presented in an intention-to-treat format. For the purposes of this review, we defined ITT as, ‘patients being analysed in the treatment arm that they entered at randomization, regardless of whether they drop-out, receive the incorrect treatment or withdraw before completion of the trial’. If a different definition of ITT was used in the trial being analysed then the results were recalculated where possible. Patients randomized to a treatment arm but for whom there were no 4 week or 8 week healing data were considered to be unhealed and were incorportated into the analysis accordingly.

Quantitative data synthesis

The primary method of calculating the summary effect estimate was by using a Fixed Effects model using the Mantel–Haenszel method.9 This was calculated as relative risk (RR) (proportion of healed patients with the comparator proton pump inhibitor divided by the proportion of healed patients with omeprazole 20 mg) with 95% confidence intervals (CI), where a confidence interval that includes the value 1.00 indicates no statistically significant difference in treatment effects. A Random Effects model using the DerSimonian and Laird method10 was used for supplementary analysis. All analyses were carried out using RevMan v4.1.11


  1. Top of page
  2. Abstract
  7. References

Trial flow

The implementation of the search strategy produced the results given in Figure 1. The systematic review was trying to capture the maximum number of qualified trials available for analysis, thus unpublished and abstract data were included if they met the quality criteria. A full list of the papers outlined in Figure 1 is available from the authors on request.


Figure 1.  Results of searches of EMBASE, Medline, BIOSIS and AstraZeneca’s internal database for direct comparisons between omeprazole 20 mg and standard-dose proton pump inhibitors.

Download figure to PowerPoint

Study characteristics

Of the 18 trials identified, one was unpublished, five were abstracts and 12 were full papers (Table 1). Only 12 of the 18 papers obtained a quality score ≥3. A sensitivity analysis was carried out to examine the effect of including the excluded trials.

Table 1.   Individual study characteristics Thumbnail image of

Quantitative data synthesis

Primary analysis:

The analysis comparing esomeprazole 40 mg with omeprazole 20 mg shows a significant difference in favour of esomeprazole at 4 weeks (RR 1.14; 95%CI: 1.10, 1.18) and 8 weeks (RR 1.08; 95%CI: 1.05, 1.10) (Table 2).

Table 2.   Relative risk (95% confidence interval) of endoscopic healing rates at 4 and 8 weeks comparing esomeprazole 40 mg with omeprazole 20 mg Thumbnail image of

When comparing lansoprazole 30 mg with omeprazole 20 mg, there is no significant difference in healing rates at 4 weeks (RR 1.02; 95%CI: 0.97, 1.08) and 8 weeks (RR 1.01; 95%CI: 0.97, 1.06) (Table 3). Similarly, there is no significant difference with pantoprazole 40 mg or rabeprazole 20 mg when compared with omeprazole 20 mg (Tables 4 and 5, respectively).

Table 3.   Relative risk (95% confidence interval) of endoscopic healing rates at 4 and 8 weeks comparing lansoprazole 30 mg with omeprazole 20 mg Thumbnail image of
Table 4.   Relative risk (95% confidence interval) of endoscopic healing rates at 4 and 8 weeks comparing pantoprazole 40 mg with omeprazole 20 mg Thumbnail image of
Table 5.   Relative risk (95% confidence interval) of endoscopic healing rates at 4 and 8 weeks comparing rabeprazole 20 mg with omeprazole 20 mg Thumbnail image of

A summary of the individual analyses can be seen in Figure 2 (4 weeks) and Figure 3 (8 weeks).


Figure 2.  Relative risk (with 95% confidence intervals) of endoscopic healing rates at 4 weeks of all standard-dose proton pump inhibitors compared with omeprazole 20 mg.

Download figure to PowerPoint


Figure 3.  Relative risk (with 95% confidence intervals) of endoscopic healing rates at 8 weeks of all standard-dose proton pump inhibitors compared with omeprazole 20 mg.

Download figure to PowerPoint

As a cumulative analysis, the data for lansoprazole, pantoprazole and rabeprazole were pooled and compared with omeprazole. Again this showed no significant difference in treatments at 4 weeks (RR 1.01; 95%CI: 0.97, 1.06) and 8 weeks (RR 1.00; 95%CI: 0.97, 1.03).


A chi-squared test30 was carried out to investigate possible heterogeneity in the individual analyses. For the comparison of esomeprazole with omeprazole the test indicates significant heterogeneity. On closer examination of the individual trials included in the analysis there are no obvious confounding factors. The designs of the trials, the patient populations and the severity of disease were very similar.

Using a Random Effects model for the meta-analysis results in a slightly different result at 4 weeks (RR 1.13; 95%CI: 1.04, 1.22) and 8 weeks (RR 1.07; 95%CI: 1.01, 1.13), which still favours esomeprazole.

There is no evidence of heterogeneity in any of the other comparisons and the use of a Random Effects model makes no significant difference to the summary estimates already described.

Sensitivity analysis:

As the quality scoring of papers can be quite subjective a sensitivity analysis was carried out on each of the proton pump inhibitor comparisons using all 18 trials found by the search strategy (Table 6).

Table 6.    Summary of meta-analyses without quality score (fixed effects) Thumbnail image of

The effect of using all 18 trials does not make a significant difference to any of the comparisons carried out in the primary analysis.


  1. Top of page
  2. Abstract
  7. References

In this systematic review esomeprazole 40 mg is the only proton pump inhibitor that provides higher endoscopic healing rates at 4 and 8 weeks than omeprazole 20 mg. Oesophageal healing is highly influenced by intragastric pH and the superior healing efficacy of esomeprazole may be explained in terms of its favourable pharmacodynamics. Esomeprazole suppresses acid to a level that retains the intragastric pH above 4 for, on average, 6 h more per 24 h than omeprazole.5 In addition to reduced acid secretion, this also results in a decreased volume of gastric contents and a consequent decrease in the volume of the gastro-oesophageal refluxate.4, 31 Many studies have compared drug effects on intragastric pH or suppression of gastric acid output, and these are generally conducted in healthy volunteers rather than GERD patients.32 Consequently, they have used surrogate endpoints rather than oesophageal healing or symptomatic improvement, which are more clinically relevant. However, esomeprazole has been shown to produce faster and more sustained heartburn resolution than omeprazole.13, 14 This is probably the most important indicator of treatment success from the patient’s point of view. In this study, oesophageal healing was used in preference to symptom relief, because symptom assessment varies between studies from degrees of symptom improvement to numbers of symptom-free days, etc. More comparative studies using standardized symptom assessments are required for all proton pump inhibitors, given that most cases of GERD are diagnosed on the basis of symptom assessment without endoscopy, and are managed accordingly.33 Standardization is also desirable for pharmacoeconomic evaluation.34

There are currently no direct comparisons of lansoprazole, pantoprazole and rabeprazole with esomeprazole in clinical trials. The results of a recent acid suppression study35 shows, however, that esomeprazole maintains intragastric pH above 4 for significantly longer than lansoprazole, pantoprazole or rabeprazole, which may translate into higher healing rates in reflux oesophagitis. While this study cannot replace direct comparisons in clinical trials, it does add supportive data to the conclusions drawn from the current work.

A recent meta-analysis looking at lansoprazole and omeprazole7 shows there is no significant difference in healing rates between these two proton pump inhibitors at 4 and 8 weeks. While this previous work used a different approach to the analysis carried out here the conclusion is the same.

There are no published systematic reviews of pantoprazole or rabeprazole, however none of the limited number of trials has shown any significant difference from omeprazole at 4 and 8 weeks.

A definitive cause could not be found for the heterogeneity in the meta-analysis of esomeprazole compared with omeprazole. A detailed study of the three trials isolates the heterogeneity to just one of them,12 but an explanation of why the results from this trial are different remains elusive. It is possible that it is just random error rather than a systematic error in the trial design since it is very similar to the other two trials. However, the effect of excluding this trial from the analysis would be to increase the difference between esomeprazole and omeprazole.

In conclusion this systematic review reveals that esomeprazole is the only proton pump inhibitor that has shown significantly greater healing than omeprazole in the treatment of acute reflux oesophagitis.


  1. Top of page
  2. Abstract
  7. References
  • 1
    Wienbeck M & Barnert J. Epidemiology of reflux disease and reflux oesophagitis. Scand J Gastroenterol 1989; 24: 713.
  • 2
    Moore RA & Phillips C. Reflux oesophagitis: quantitative systematic review of the evidence of effectiveness of proton pump inhibitors and histamine antagonists. Bandolier 1999;
  • 3
    Chiba N, De Gara CJ, Wilkinson JM, Hunt RH. Speed of healing and symptom relief in grade II to IV gastroesophageal reflux disease: a meta-analysis. Gastroenterology 1997; 112: 1798810.
  • 4
    Bell NJ, Burget D, Howden CW, et al. Appropriate acid suppression for the management of gastro-oesophageal reflux disease. Digestion 1992; 51(Suppl. 1): 5967.
  • 5
    Lind T, Rydberg L, Kylebäck A, et al.Esomeprazole provides improved acid control vs. omeprazole in patients with symptoms of gastro-oesophageal reflux disease. Aliment Pharmacol Ther 2000; 14: 8617.
  • 6
    Junghard O, Hassan-Alin M, Hasselgren G. The effect of AUC and Cmax of esomeprazole on acid secretion and intragastric pH. Gastroenterology 2000; 118: A64A64(Abstract).
  • 7
    Sharma VK, Leontiadis GI, Howden CW. Meta-analysis of randomized controlled trials comparing standard clinical doses of omeprazole and lansoprazole in erosive oesophagitis. Aliment Pharmacol Ther 2001; 15: 22731.
  • 8
    Jadad A, Moore RA, Carroll D, et al.Assessing the quality of reports of randomised controlled trials: is blinding necessary? Control Clin Trials 1996; 17: 112.DOI: 10.1016/0197-2456(95)00134-4
  • 9
    Mantel N & Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 1959; 22: 71948.
  • 10
    DerSimonian R & Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: 17788.
  • 11
    Review Manager (RevMan) [Computer program] Version 4.1 for Windows. Oxford, UK: The Cochrane Collaboration, 2000: CD-ROM and Internet.
  • 12
    AstraZeneca. Study B2 and Supplementary Information for Kahrilas et al. (ref. 14) Data on file (NEX/056/FEB2001).
  • 13
    Richter JE, Kahrilas PJ, Johanson J, et al.Efficacy and safety of esomeprazole compared with omeprazole in GERD patients with erosive esophagitis: a randomized controlled trial. Am J Gastroenterol 2001; 96: 65665.
    Direct Link:
  • 14
    Kahrilas PJ, Falk GW, Johnson DA, et al.Esomeprazole improves healing and symptom resolution as compared with omeprazole in reflux oesophagitis patients: a randomized controlled trial. Aliment Pharmacol Ther 2000; 337: 124958.
  • 15
    Dekkers CPM, Beker JA, Thjodleifsson B, et al.Double blind, placebo controlled comparison of rabeprazole 20 mg versus omeprazole 20 mg in the treatment of ersoive or ulcerative GORD. Aliment Pharmacol Ther 1999; 13: 4957.DOI: 10.1046/j.1365-2036.1999.00438.x
  • 16
    Delchier J, Cohen G, Humphries TJ. Rabeprazole is comparable to omeprazole in the healing of erosive GERD and provides more rapid heartburn relief. Gastroenterology 1999; 116: A146A146(Abstract).
  • 17
    Hatlebakk JG, Berstad A, Carling L, et al.Lansoprazole versus omeprazole in short-term treatment of reflux oesophagitis. Scan J Gastroenterol 1999; 28: 2248.
  • 18
    Vcev A, Stimac D, Vceva A, et al.Pantoprazole versus omeprazole in the treatment of reflux esophagitis. Acta Med Croat 1999; 53: 7982.
  • 19
    Pilotto A, Franceschi M, Leandro G, et al.Comparison of omeprazole, lansoprazole and pantoprazole in the treatment of elderly patients with esophagitis. Ital J Gastroenterol Hepatol 1998; 30(Suppl. 2): 142142(Abstract).
  • 20
    Vicari F, Belin J, Marek L. Pantoprazole 40 mg versus omeprazole 20 mg in the treatment of reflux esophagitis. Gastroenterology 1998; 114: A324A324(Abstract).
  • 21
    Vcev A, Stimac D, Vceva A, et al.Lansoprazole versus omeprazole in the treatment of reflux oesophagitis. Acta Med Croat 1997; 51: 1714.
  • 22
    Castell DO, Richter JE, Robinson M. Efficacy and safety of lansoprazole in the treatment of reflux oesophagitis. Am J Gastroenterol 1996; 91: 174957.
  • 23
    Cordova-Villalobas JA, Ramirez-Barba EJ, Ramirez-Covarrubias JC. Reflux oesophagitis: a comparative study of lansoprazole and omeprazole as short-term therapy. Rev Gastroenterol Mex 1996; 61: 3069.
  • 24
    Hotz J, Gangl A, Heinzerling H. Pantoprazole versus omeprazole in acute reflux esophagitis. Gastroenterology 1996; 110: A136A136(Abstract).
  • 25
    Mee AS & Rowley JL, Lansoprazole Clinical Research Group. Rapid symptom relief in reflux oesophagitis: a comparison of lansoprazole and omeprazole. Aliment Pharmacol Ther 1996; 10: 75763.
  • 26
    Corinaldesi R, Valentini M, Belaiche J, et al.Pantoprazole and omeprazole in the treatment of reflux esophagitis. Aliment Pharmacol Ther 1995; 9: 66771.
  • 27
    Mossner J, Holscher AH, Herz R, Schneider A. A double blind study of pantoprazole and omeprazole in the treatment of reflux esophagitis. Aliment Pharmacol Ther 1995; 9: 3216.
  • 28
    Petite JP, Group Multicentrique. Efficacté compareé du lansoprazole et de l’oméprazole dans le traitement de l’oesophagite peptique. Med Chir Dig 1995; 24: 2914.
  • 29
    Corallo J, Vicari F, Forestier S, Joubert M. Lansoprazole in the acute treatment of reflux esophagitis. Gastroenterology 1993; 104: A58A58(Abstract).
  • 30
    Pearson K. On a criterion that a given system of deviations from the probable in the case of a correlated system of variable is such that it can reasonably be supposed to have arisen in random sampling. Philosophical Magazine 1900; 50: 15775.
  • 31
    Lind T, Cederberg C, Ekenved G, et al.Effect of omeprazole – a gastric proton pump inhibitor – on pentagastrin stimulated acid secretion in man. Gut 1983; 24: 2706.
  • 32
    Fass R, Murthy U, Hayden CW, et al.Omeprazole 40 mg once a day is equally effective as lansoprazole 30 mg twice a day in symptom control of patients with gastro-oesophageal reflux disease (GERD) who are resistant to conventional-dose lansoprazole therapy – a prospective, randomized, multi-centre study. Aliment Pharmacol Ther 2000; 14: 1595603.
  • 33
    Kahrilas PJ. Strategies for medical management of reflux disease. Baillieres Best Pract Res Clin Gastroenterol 2000; 14: 77591.
  • 34
    Goeree R, O'Brien B, Hunt R, et al.Economic evaluation of long-term management strategies for erosive oesophagitis. Pharmacoeconomics 1999; 16: 67997.
  • 35
    Röhss K, Wilder-Smith C, Claar-Nilsson C, et al.Esomeprazole 40 mg provides more effective acid control than standard doses of all other proton pump inhibitors. Gastroenterology 2001; 120: A419A419(Abstract).