Comparable clinical efficacy and tolerability of 20 mg pantoprazole and 20 mg omeprazole in patients with grade I reflux oesophagitis


Dr K. D. Bardhan, Rotherham District General Hospital, Moorgate Road, Rotherham, South Yorkshire S60 2UD, UK. E-mail:



Several clinical trials have shown that pantoprazole (40 mg) and omeprazole (40 or 20 mg) have similar efficacy and safety in the treatment of grade II–IV reflux oesophagitis (Savary–Miller classification).


To compare the efficacy and safety of once-daily doses of pantoprazole (20 mg) and omeprazole (20 mg) with respect to symptom relief and healing of patients with grade I reflux oesophagitis.


Patients with endoscopically established grade I reflux oesophagitis (non-confluent, patchy red lesions with/without white fibrin coating) were enrolled into this randomized, open, parallel-group, multicentre study. A total of 328 patients (n=166 in the pantoprazole group, n=162 in the omeprazole group) were recruited in 23 centres. Patients received 4 weeks of treatment. If the reflux oesophagitis was not completely healed, the treatment was extended to 8 weeks.


After 2 and 4 weeks of treatment with either pantoprazole or omeprazole, the rate of symptom relief was similar (70% vs. 79% and 77% vs. 84%, respectively). High healing rates were observed after 4 and 8 weeks (pantoprazole: 84% and 90%, respectively; omeprazole: 89% and 95%, respectively). Both treatments were well tolerated. The most frequently reported adverse events on pantoprazole and omeprazole, respectively, were nausea (8% vs. 7%), diarrhoea (5% vs. 6%) and headache (6% vs. 3%).


After 4 and 8 weeks of treatment with pantoprazole (20 mg) or omeprazole (20 mg), patients with mild gastro-oesophageal reflux disease (grade I) showed comparably high rates of symptom relief and healing. Both treatments were safe and well tolerated.


The proton pump inhibitor pantoprazole has linear pharmacokinetics and a high and constant bioavailability of 80% which results in powerful acid inhibition, making it useful for the treatment of gastro-oesophageal reflux disease (GERD).1–3

Dose-finding studies and comparative trials have demonstrated that pantoprazole, 40 mg, is the optimal dose for the treatment of moderate to severe reflux oesophagitis and 20 mg for mild disease.4–6 Our aim was to compare a once-daily dose of pantoprazole, 20 mg, with omeprazole, 20 mg, in the treatment of grade I reflux oesophagitis as defined by Savary and Miller.7 Because omeprazole, 10 mg, does not consistently provide adequate control of acid secretion and reflux symptoms, a comparative dosage of 20 mg was chosen.8–11 An even higher dose of 40 mg omeprazole offers only marginal benefit over 20 mg, and hence 20 mg omeprazole has previously been recommended as the standard dose for the treatment of symptomatic reflux oesophagitis.12 In this study, patients received pantoprazole and omeprazole at the optimal dose or standard dose, respectively.


The study was conducted in accordance with the current version of the Declaration of Helsinki. Approvals from the appropriate Ethics Committees/Institutional Review Boards were obtained in each centre. Written informed consent to participate in the study was obtained prior to enrolment by the physicians from all patients. Out-patients fulfilling the inclusion criteria were enrolled and randomized (see Figure 1 for inclusion criteria).

Figure 1.

 Patients with reflux oesophagitis (grade I, classified according to Savary–Miller7) and their progress through the study. The definition of protocol violators is given in the Methods section.

Major exclusion criteria were: patients with grade II, III or IV reflux oesophagitis, pregnant and nursing women or women of child bearing age who had not used effective methods of contraception for at least 3 months prior to the start of the study, patients with malignant or any other severe diseases, patients with clinically relevant deviations from the normal range in laboratory values as assessed by the investigator and patients with a history of Zollinger–Ellison syndrome, pyloric stenosis, peptic ulcer or ulcer complications and previous surgery of the oesophagus and/or gastrointestinal tract (except for appendectomy, cholecystectomy and polypectomy). Patients were excluded if they had received antacids, or drugs which might potentially interact with either of the study drugs, had taken substituted benzimidazoles or H2-blockers for more than 3 days in the 20 days prior to the start of the study, had participated in a clinical study within the previous month, had a proven or suspected sensitivity to either of the study drugs or their related compounds, or were prone to allergic drug reactions. Regular intake of glucocorticosteroids or NSAIDs, simultaneous intake of drugs whose absorption was pH-dependent (such as ketoconazole), or a history of drug or alcohol abuse led to exclusion. Medication that might stimulate reflux, e.g. calcium antagonists, spasmolytics, nitrates, phenothiazines and theophylline preparations, were not allowed.


This was a randomized, open, parallel-group, multicentre study. Patients fulfilling the inclusion criteria were randomized according to a computer-generated randomization list to receive either once-daily 20 mg pantoprazole (gastro-resistant tablets) or 20 mg omeprazole (hard gelatine capsules). The study medication was taken orally in the morning before breakfast for 4 weeks. If healing was not complete, the treatment was extended to 8 weeks.

Conduct of the study

The study period for each patient was 4 weeks. Within this period, a maximum of three visits was scheduled: at the start of the study and after 2 and 4 weeks of treatment. An experienced endoscopist performed two endoscopies prior to study entry and after 4 weeks in order to determine the stage of GERD according to the Savary–Miller classification.7 Each endoscopy had to be performed by the same endoscopist, when feasible. Healing was defined as complete epithelialization of the lesions, so that the oesophagus appeared normal upon endoscopic examination. In the case of persisting lesions, the treatment was continued and an endoscopy was again performed after 8 weeks at an additional visit. If the reflux oesophagitis was still not healed after 8 weeks, the patient was classified as a non-responder. At the initial visit, standard laboratory investigations were performed, the Helicobacter pylori status was determined by the 13C-urea breath test, and demographic data, medical history, clinical symptoms and concomitant medication were documented. At each follow-up visit, compliance was checked and changes in concomitant medication and possible adverse events were documented. Adverse events were rated by the investigator as not related, unlikely, likely or definitely related to the medication.

Symptoms were assessed prior to the start of the study and after 2 and 4 weeks. The clinical assessment was performed by the investigator, without prior knowledge of the endoscopy findings. After 8 weeks, another symptom assessment was performed only for patients in whom treatment had to be continued. The patient was questioned about the main symptoms (acid eructation, heartburn and pain on swallowing) over the previous 3 days, as well as the other symptoms of reflux oesophagitis (epigastric pain, retrosternal pain, retrosternal tightness, eructation of air, nausea, vomiting and retching). The severity of the symptoms was classed by the investigator as follows: none, mild (symptoms were hardly perceived, only slight impairment of general well being), moderate (clearly noticeable symptoms, but tolerable without immediate relief) or severe (overwhelming discomfort, urgent desire for immediate relief).


The healing rates after 4 weeks were analysed using the Cochran-Mantel–Haenszel method (SAS procedure FREQ, Option/CMH, SAS version 6.11).13 The null hypothesis of equal treatment effects was tested vs. the alternative of different treatment effects. Including 150 patients per group, the power to prove a difference between treatments was 90% if there was a true odds ratio of at least 2.4, e.g. 83% vs. 67%. This calculation is based on an analysis by Fisher’s exact test, two-sided, with a level of significance of 5%.

The time until healing (‘4 weeks’, ‘8 weeks’, ‘not healed’) was compared by means of the U-test. To compare the difference between the treatment groups in the rates of freedom from symptoms, the odds ratios and confidence intervals were calculated in a similar manner to that used for healing rates.

All patients who took the study medication at least once were included in the intention-to-treat population. Patients were excluded from the per protocol analysis if they prematurely terminated the study for reasons that were not related to the study medication, or if they violated the study protocol (such as missed study visits). However, patients who were classified as dropouts were included in the per protocol analysis. Dropouts were defined as patients who discontinued the study prematurely due to: (i) an adverse event, rated by the investigator, as ‘likely’ or ‘definitely’ related to the study medication; or (ii) who showed deterioration in the gastrointestinal disease (i.e. lack of efficacy).

Safety analyses were performed on the basis of the intention-to-treat population and efficacy on the per protocol population.


In total, 328 patients were enrolled into the study and randomized (n=166 in the pantoprazole group, n=162 in the omeprazole group) (Figure 1). The study was conducted in the UK (n=16 centres), Republic of Ireland (n=3 centres) (both countries together, 193 patients) and South Africa (n=4 centres, 135 patients). Patients in the two treatment groups had comparable baseline entry characteristics, except for the fact that there were more male patients in the omeprazole group than in the pantoprazole group (Table 1).

Table 1.   Baseline characteristics of patients (intention-to-treat collective) Thumbnail image of

The per protocol population consisted of 264 patients (n=133 in the pantoprazole group, n=131 in the omeprazole group). No substantial differences were found in the incidence of acid eructation, heartburn and pain on swallowing at baseline (U-test, two-sided, alpha 0.05).

Symptom relief

After 2 and 4 weeks of treatment, the rates of symptom relief of all main symptoms were similar for pantoprazole (70% and 77%) and omeprazole (79% and 84%). Relief of acid eructation, heartburn and pain on swallowing was similar in the two treatment groups at 2 and 4 weeks, irrespective of the severity at baseline (Figure 2A–C). The same applied to the result with severe symptoms at entry (data not shown).

Figure 2.

 Symptom relief rates (per protocol analysis) of reflux oesophagitis patients (grade I) with either pantoprazole, 20 mg, or omeprazole, 20 mg, after 2 and 4 weeks for the main symptoms: (A) acid eructation; (B) heartburn; and (C) pain on swallowing.

A higher proportion with mild symptoms at entry had relief compared with patients with severe symptoms, and this was similar for both treatments.

The freedom from other symptoms of reflux oesophagitis (i.e. epigastric and retrosternal pain) was also similar in the pantoprazole (56% and 58%) and omeprazole (50% and 60%) groups after 2 and 4 weeks, respectively. Thus, the treatments were equivalent for symptom relief.

Healing rates

The healing rates (Figure 3) were high and similar in both treatment groups at 4 and 8 weeks. Statistically, neither treatment was judged to be superior to the other.

Figure 3.

 Healing rates of reflux oesophagitis patients (grade I) with either pantoprazole, 20 mg, or omeprazole, 20 mg, after 4 and 8 weeks of treatment; 95% confidence intervals and odds ratios are shown in the table. Superiority of one treatment was concluded if the 95% confidence interval for the odds ratio did not include the value 1.0.

Healing rates were similar for both treatments in patients with mild to moderate symptoms or with severe symptoms at entry (Table 2 reflects severe symptomatology).

Table 2.   Healing rates after 4 weeks in patients with severe symptoms at baseline (per protocol collective) Thumbnail image of

Helicobacter pylori infection was present in 112 of 264 patients (40%). Those infected had a cumulative healing rate of 106 of 112 patients (95%) at 8 weeks (both treatment groups combined) compared with 136 of 150 patients (91%) who were not infected (not significant).


Both therapies were well tolerated. In the pantoprazole group, 95 patients (57%) experienced adverse events compared with 80 patients (50%) on omeprazole. The majority of adverse events were mild, but in 10% (pantoprazole) and 13% (omeprazole) they were severe. No adverse event was classed as ‘definitely’ related to treatment; in 22% (pantoprazole) and 13% (omeprazole), the relation was judged to be ‘likely’. Most events were felt by the investigators to be ‘unrelated’ or ‘unlikely to be related’ to the study drug.

The most common adverse events (other than symptoms of the underlying disorder such as eructation, retrosternal pain) on pantoprazole and omeprazole were, respectively, nausea (8% vs. 7%), diarrhoea (5% vs. 6%) and headache (6% vs. 3%). Two patients experienced serious adverse events, which were not thought to be related to the study medication: one patient in the pantoprazole group suffered a pneumothorax which led to hospitalization; another patient in the omeprazole group had gallstones which caused persistent disability.


Pantoprazole, 20 mg daily, proved to be as effective as omeprazole, 20 mg daily (the full healing dose), in relieving reflux symptoms and in healing grade I oesophagitis. Table 3 compares the results obtained in this study with other studies where the same dosage has been used: the results are very similar. In general, about 80% become symptom-free by 4 weeks, but the majority on pantoprazole, 20 mg, by 2 weeks. A higher proportion of patients with mild symptoms at entry become asymptomatic but, even in those severely symptomatic at the start, about 80% become symptom-free at 4 weeks. Again, healing is 80% or greater at 4 weeks and unrelated to initial symptom severity.

Table 3.   Comparison of symptom relief and healing rates (%) on pantoprazole, 20 mg, and omeprazole, 10 mg, 20 mg, in patients with mild gastro-oesophageal reflux disease Thumbnail image of

The safety profiles of the two treatments were very similar. Both treatments were well tolerated; most adverse events were probably not related to the study drugs.

Proton pump inhibitors reduce acid less effectively in the absence of Helicobacter pylori,19, 20 and so healing rates would be expected to be lower amongst the uninfected patients. In our study, however, the difference was small and not significant.

In conclusion, after 4 and 8 weeks of therapy with pantoprazole, 20 mg, or omeprazole, 20 mg, patients with mild GERD showed comparably high rates of symptom relief and healing, and treatment was safe.


The investigators are very grateful to Dr Ursula H. Hole and Dr Kathy B. Thomas (Byk Gulden, Konstanz, Germany) for helpful suggestions during the preparation and editing of the manuscript, and to Byk Gulden (Germany) for supporting the study.

The following investigators participated in the study.

Great Britain

Dr A. Sensier (Adan House Medical Centre, Spennymoor).

Dr A. Heatherington (The Health Centre, Bishop Auckland).

Dr M. K. Kamdar (The Surgery, Canvey Island).

Dr S. Stock (Bishop Auckland Hospital, Bishop Auckland).

Dr C. Babbs (Royal Oldham Hospital, Oldham).

Dr G. Tildersley (Hartlepool and Peterlee Hospitals, Hartlepool).

Dr R. M. Patel (The Surgery, Benfleet).

Dr R. W. Chapman (John Radcliffe Hospital, Oxford).

Dr B. T. Cooper (City Hospital, Birmingham).

Dr M. J. Lancaster-Smith (Queen Mary’s Hospital, Sidcup).

Dr J. de Caestecker (Glenfield General Hospital, Leicester).

Dr J. Dillon (Ninewells Hospital, Dundee).

Dr D. Lynch (Blackburn Royal Infirmary, Blackburn).

Dr P. Sahay (Pontefract General Infirmary, Pontefract).

Dr G. Swift (Llandough Hospital, Penarth).

Dr D. Keating (Elm Lane Surgery, Sheffield).

Republic of Ireland

Dr F. E. Murray (Beaumont Hospital, Dublin).

Dr C. A. O’Morain (Meath Hospital, Dublin).

South Africa

Dr J. H. van Zyl (Universitas Hospital, Bloemfoentein).

Dr P. J. Honiball (Pretoria).

Dr F. P. Eloff (Sunnyside).