MODULE V

OMEMAT INSTA 40mg SACHET

Factory: Plot#122, Block-B, Phase-v, Industrial Estate, Hattar, Pakistan
Head office: House No. 93, Street No. 03, Sector: 02, Airport Housing Society, Rawalpindi, Pakistan
------------------------------------------------------------------------------------------------------------------------------------------------------- -------------------------------------------------------

MODULE-5
CLINICAL DOCUMENTS

Page 1 of 2
 MODULE V
OMEMAT INSTA 40mg SACHET
Factory: Plot#122, Block-B, Phase-v, Industrial Estate, Hattar, Pakistan
Head office: House No. 93, Street No. 03, Sector: 02, Airport Housing Society, Rawalpindi, Pakistan
------------------------------------------------------------------------------------------------------------------------------------------------------- -------------------------------------------------------

MODULE 5.1- CLINICAL STUDY REPORTS

Section Documents
5.1 Tablet of contents
5.2 Clinical Study Reports
5.3 List of Key References

Page 2 of 2
 Alimentary Pharmacology and Therapeutics

Control of intra-oesophageal pH in patients with Barrett's

oesophagus on omeprazole-sodium bicarbonate therapy
L. B. Gerson*, S. Mitra*, W. F. Bleker† & P. Yeung†

*Division of Gastroenterology and SUMMARY

Hepatology, Stanford University
School of Medicine, Stanford, CA,
USA.
Background
†
Santarus, Inc, San Diego, CA, USA. Approximately 30–40% of patients with Barrett's oesophagus (BE) patients
manifest abnormal oesophageal pH profiles despite proton pump inhibitor
(PPI) therapy.
Correspondence to:
Prof. L. B. Gerson, Division of
Gastroenterology and Hepatology, Aim
450 Broadway Street, 4th Floor To determine control of oesophageal reflux using Bravo pH monitoring in
Pavilion C, MC: 6341, Redwood City, patients with BE on omeprazole-sodium bicarbonate oral suspension pow-
CA 94063, USA.
der (Ome-NaBic) 40 mg twice daily.
E-mail: [email protected]

Methods
Publication data Initial pH monitoring off PPI for 1 week was performed. All patients under-
Submitted 31 October 2011 went repeat pH testing on Ome-NaBic administered before breakfast and at
First decision 8 November 2011
bedtime after 21–28 days of therapy depending upon the prior PPI therapy.
Resubmitted 25 December 2011
Accepted 17 January 2012 The goal was to enroll 30 subjects, however, the trial was terminated prema-
EV Pub Online 22 February 2012 turely when the sponsor lost financing due to a change in business strategy.

Results
A total of 88 patients responded to the study invitation, 27 patients signed
informed consent, and 21 patients underwent pH testing of PPI. A total of
15 patients completed the protocol (13 men, 2 women). Demographic
information for patients completing at least one Bravo study included a
mean (±s.d.) age of 62 ± 9 years; body mass index 31 ± 8 kg/m2 (range 23
–48); mean BE length of 2.6 ± 2 cm; 9 (43%) patients had long segment
BE. All (100%) patients demonstrated normalisation of supine pH on both
days of Ome-NaBic therapy. One patient (6%) demonstrated abnormal
upright reflux on the second day of monitoring on Ome-NaBic therapy; all
the other patients demonstrated normal pH scores on therapy.

Conclusions
Administration of twice daily Ome-NaBic demonstrated control of nocturnal
oesophageal reflux in 100% of patients with Barrett's oesophagus, and complete
control of oesophageal pH during 97% of the 24-h recording periods.

Aliment Pharmacol Ther 2012; 35: 803–809

ª 2012 Blackwell Publishing Ltd 803

doi:10.1111/j.1365-2036.2012.05016.x
L. B. Gerson et al.

INTRODUCTION into the study, patients had to meet the following study
Barrett's oesophagus (BE), a change of the normal squa- entry criteria including: (i) Presence of at least 0.5 cm of
mous mucosa of the oesophagus to specialised intestinal BE on upper endoscopy with confirmation of the pres-
metaplasia, is detected in approximately 10% of patients ence of intestinal metaplasia on biopsy, (ii) Negative test-
with chronic gastro-oesophageal reflux disease (GERD).1 ing for Helicobacter pylori infection by serology, stool
The condition has also been described in a similar pro- antigen testing, or gastric biopsy. Patients with prior H.
portion of asymptomatic adults.2, 3 White men over the pylori infection were required to have documented eradi-
age of 50 are at the greatest risk for the development of cation by subsequent negative stool testing, breath testing
BE.4 When ambulatory oesophageal pH monitoring stud- or histology, (iii) Age  18 years, (iv) Ability to provide
ies are performed comparing GERD patients with and informed consent. Exclusion criteria included (i) history
without BE, the latter group are more likely to demon- of oesophageal, gastric, or duodenal surgery (including
strate more severe pathological supine reflux, abnormal anti-reflux surgery or endoscopic anti-reflux procedures),
oesophageal motility, reduced lower oesophageal sphinc- except for simple ulcer closure, (ii) Zollinger-Ellison syn-
ter pressures, and greater hiatal hernia size.5 drome, (iii) endoscopic evidence or suspicion of a patho-
A number of studies using 24-h ambulatory pH moni- logical or infiltrative process in the stomach, (iv) inability
toring have demonstrated that patients with BE are more to tolerate Bravo pH placement without sedation.
likely to manifest prolonged supine periods of gastroeso- All patients were required to undergo baseline ambula-
phageal reflux. This phenomenon has been reported to tory pH monitoring using the Bravo pH monitoring device
occur even when patients document that they are asymp- (Given Imaging, Yoqneam, Israel). Any baseline Bravo pH
tomatic on their current proton pump inhibitor therapy monitoring studies off of proton pump inhibitor therapy
for reflux. Published series of patients on omeprazole, within 8 weeks of enrollment were accepted. Only patients
esomeprazole, and other PPIs have demonstrated that 30– with an abnormal pH recording (i.e. percentage of time
40% of BE patients still manifest abnormal pH profiles over 24 h with oesophageal pH <4 of  5.3%) were
on PPI therapy.6, 7 allowed to continue in the trial and enter the Treatment
Omeprazole-sodium bicarbonate, (Ome-NaBic; Santa- Period. Screening was planned to be continued until 50
rus, Inc. San Diego, CA, USA) is a proton pump inhibi- patients had completed two Bravo pH studies according to
tor that combines an omeprazole compound without the protocol. Enrollment was anticipated to include more than
enteric coating and a sodium bicarbonate with a goal of 50 patients if some of the patients had inadequate Bravo
producing more immediate release of the PPI drug due tracings for technical reasons or early probe dislodgement.
to the effect of the bicarbonate in buffering gastric acid. The study period was terminated prematurely due to a
Since most BE patients demonstrate nocturnal gastric change in the business strategy of the sponsor.
acid breakthrough in the early morning hours,8 use of
Ome-NaBic before bedtime may prevent this phenomenon. Administration of omeprazole-sodium bicarbonate
The purpose of this prospective cohort study was to therapy
determine the effectiveness of Ome-NaBic therapy taken Following the initial Bravo pH study that was conducted
twice daily (b.d.) on oesophageal pH profiles measured off of the baseline PPI therapy for 1 week, all patients
during 48-h Bravo pH monitoring. Although prior small were administered Ome-NaBic 40 mg orally 1 h before
studies have demonstrated adequate suppression of breakfast and bedtime. Patients who had been using
oesophageal pH in GERD patients while using Ome-Na- omeprazole previously were required to take Ome-NaBic
Bic,9, 10 the efficacy of this compound in patients with for at least 3 weeks and then undergo repeat Bravo pH
BE has not been previously evaluated. testing on therapy. Patients on PPIs other than omepra-
zole were required to use 4 weeks of Ome-NaBic ther-
METHODS apy. Therefore, the second Bravo pH study on PPI
The Human Subjects Panel at Stanford University therapy occurred between days 22–29 of Ome-NaBic
approved the clinical protocol. Patients with a known therapy depending upon the prior PPI administered.
history of BE were sent letters of invitation to participate
in the study. In addition, local radio advertising was con- Bravo probe placement
ducted daily during 3-week periods. Interested patients To be enrolled into the trial, patients were required to
were required to contact the study staff to undergo an have a prior endoscopy documenting the location of the
initial interview to determine eligibility. To be entered oesophagogastric junction in centimetres from the

804 Aliment Pharmacol Ther 2012; 35: 803-809

ª 2012 Blackwell Publishing Ltd
 Omeprazole-sodium bicarbonate in Barrett's oesophagus

mouth, or a prior oesophageal manometry with measure- time). The questions ask the patient to rate the amount of
ment of the lower oesophageal sphincter (LES). For the interference with sleep, emotional functioning, food and
baseline Bravo pH study, patients were required to be off drink problems, and physical and social functioning
of proton pump inhibitor therapy for at least a week and related to heartburn problems.12 The mean score can be
H2-blockers for 5 days. calculated by dividing the total score by 25 or a series of
Placement of the Bravo pH probe has been recom- subscale scores can be calculated. A meaningful change is
mended to be 6 cm above the oesophagogastric junction considered to be a change by 1.5 points or more. The RDQ
as determined by endoscopy or 9 cm above the LES as is a five item survey that asks the patient to rate the fre-
determined by prior manometry.11 Given the variation quency and severity of GERD symptoms from 1 to 5 where
in lengths of the BE segments for patients enrolled in a score of 1 is equivalent to an asymptomatic state.13
the study from 0.5 to 8 cm, the decision was made to
place the Bravo pH probe 6 cm above the proximal Statistical analysis
extent of the gastric folds to be consistent among Differences in patient demographics, pH test parameters,
patients. Therefore for patients with BE segments over and questionnaire results were assessed via Chi-squared
6 cm in length, the Bravo pH probes were placed within tests when variables were categorical or Student's t-tests
the Barrett's segment. The second probe was placed in when variables were continuous. (SAS 9.2 for windows;
the same location as the first study. SAS Institute, Cary, NC, USA) For all statistical analyses,
Patients arrived to the Endoscopy unit in the fasting an a priori significance level of alpha = 0.05 was utilised.
state and received topical cetacaine spray prior to intro-
duction of the Bravo delivery system into the mouth. Sample size calculation
The delivery system was introduced orally with the Based on our prior studies in patients with BE on PPI
patient placed in the left lateral decubitus position in the therapy,5, 7, 14 approximately 50% of BE patients on tra-
Motility lab. With the delivery system placed into the ditional PPI therapy achieved suppression of oesophageal
oesophagus and positioned correctly, the external vac- pH on PPI therapy. We estimated that approximately
uum pump was activated to apply suction to the well of 70% would achieve normalisation on Ome-NaBic ther-
the pH capsule and adjacent oesophageal mucosa. Suc- apy. Using a two-tailed alpha of 0.05 and power of 90%,
cessful capture of oesophageal mucosa was assumed when the estimated sample size was 32 patients.
the vacuum gauge on the pump stabilised at a value over
500 mmHg for 10 s. The plastic safety guard on the han- RESULTS
dle was then removed and the activation button A total of 88 patients responded to the invitation for par-
depressed. This action activated release of a spring-loaded ticipation. Thirty-four patients with BE subsequently
stainless steel pin from the pH capsule, to secure the pH declined enrollment. Twenty-eight subjects were excluded
capsule to the oesophageal mucosa. The activation button for the following reasons: lack of BE after review of
on the handle was then twisted clockwise 90° and endoscopy and/or pathology reports (N = 22), presence
re-extended to release the pH capsule from the delivery of oesophageal cancer (N = 1), post-ablative therapy for
system. The delivery system was then removed from the BE (N = 2), presence of achalasia (N = 2), and Asian
patient. Verification of placement was not performed for background (N = 1). A total of 27 patients signed
any of the patients by means of x-ray or upper endoscopy. informed consent; 22 patients underwent the first Bravo
All patients signed informed consent for the Bravo pH study off of PPI therapy. Four patients did not return for
procedures and were provided with instructions regarding a second Bravo study, whereas two patients violated pro-
the symptom diaries during the study, tocol and stopped Ome-NaBic prior to the second pH
study. One patient had two failed Bravo studies likely due
Study questionnaires to underlying scleroderma. A total of 15 patients
All patients were administered the Quality of Life in completed the protocol (13 men, 2 women, Figure 1).
Reflux and Dyspepsia (QOLRAD) and Reflux Disease Demographic information for patients completing at
Questionnaire (RDQ) off PPI therapy and then after the least one Bravo study are shown in Table 1. The major-
course of Ome-NaBic therapy. The QOLRAD is a 25- ity of patients were obese Caucasian males over the age
item survey that asks patients to rate the amount of of 60 with at least one co-morbid condition. The mean
quality of life impairment from heartburn symptoms BE length was 2 cm (two patients had 0.5 cm segments)
using a scale from 1 (all of the time) to 7 (none of the whereas nine (43%) patients had long segment BE. The

Aliment Pharmacol Ther 2012; 35: 803-809 805

ª 2012 Blackwell Publishing Ltd
L. B. Gerson et al.

Patients responding to
advertisement
N = 88

Patients screened
N = 88

Patients not qualified Patients without BE

N = 28 N = 22

Patients signed Patients dropped out

consent N=5
N = 27

First bravo completed Patients dropped out

N = 22 N=4

Second bravo Protocol violation

completed N=2 Figure 1 | Study flow diagram.
N = 18
Fifteen patients completed all
required study interventions
and had complete data sets
Study completed for two 48-h Bravo pH
N = 15 monitoring periods.

majority (52%) of the patients used omeprazole to con- prior PPI administered as described in the methods.
trol GERD symptoms prior to study entry. Six (29%) of There was a statistically significant decrease in all param-
the subjects were using PPI twice daily, whereas one patient eters on Ome-NaBic for the total, upright, and supine
was only using H2-blocker therapy. (Table 1) Duration of time that the oesophageal pH was less than 4.0 as dem-
GERD symptoms was stated as ‘unknown’ in six patients, onstrated in Table 2. The mean (+s.d.) Demeester score
between 6 and 10 years in four patients, 10–20 years in six on Ome-NaBic was 2 ± 2 for the first day and 3 ± 6 on
patients, and over 20 years in six patients. the second day of monitoring. All (100%) patients dem-
Results from the Bravo pH monitoring studies are onstrated normalization of supine pH on both days of
shown in Table 2. All the patients had abnormal DeMe- Ome-NaBic therapy. One patient with 8 cm segment of
ester scores on the first day of pH monitoring with three BE demonstrated pathological upright GERD during the
patients manifesting only supine pathological reflux, eight second day of ambulatory pH monitoring on Ome-Na-
patients pathological upright GERD, and 10 patients with Bic with total time pH <4 of 7.2%. Figures 2 and 3 dem-
bipositional GERD. On the second day of pH monitor- onstrate the individual values per patient for upright and
ing, one patient had a normal pH profile, while five supine reflux activity off PPI therapy and while on Ome-
patients demonstrated isolated supine pathological reflux, NaBic twice daily.
four patients manifested solely upright GERD, and bipo- Impact of GERD symptoms on health-related quality
sitional GERD was present in 11 patients. of life was measured using the QOLRAD. The mean
A total of 15 patients completed the second Bravo pH QOLRAD score increased from 5 ± 1 to 163 ± 18 while
study on Ome-NaBic 40 mg twice daily with evaluable on Ome-NaBic therapy. GERD symptom severity was
data tracings. The second pH study occurred between measured using the RDQ and decreased significantly on
days 22–29 of Ome-NaBic therapy depending on the Ome-NaBic therapy. All patients tolerated Ome-NaBic

806 Aliment Pharmacol Ther 2012; 35: 803-809

ª 2012 Blackwell Publishing Ltd
 Omeprazole-sodium bicarbonate in Barrett's oesophagus

40
Table 1 | Demographic information for patients
Upright % pH < 4
completing first Bravo study (N = 21) 35
Supine % pH < 4
Variable 30
(Mean ± s.d.) Mean ± s.d. Range

% time pH < 4
25
Mean (±s.d.) age 61 ± 10 42–83
Body mass 30 ± 7 22–48 20
index (kg m2)
15
Male (%) 17 (81) –
Caucasian 20 (95%) – 10
Co-morbid conditions 18 (85%) –
Length BE (cm) 2.4 ± 1.9 0.5–8 5
PPI prior to study Number Dosages
Omeprazole 12 20 mg q.d.s. (7), 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
40 mg q.d.s. (1);
20 mg b.d. (4)
Lansoprazole 2 30 mg q.d.s. (1); Figure 2 | Graph demonstrating percentage time
30 mg b.d. (1) upright and supine pH <4% for each patient off of PPI
Pantoprazole 1 40 mg b.d.
therapy.
Rabeprazole 2 20 mg q.d.s.
Esomeprazole 2 40 mg q.d.s.
Dexlansoprazole 1 60 mg q.d.s.
Non-PPI: Famotidine 1 40 mg b.i.d. ical supine reflux in these patients was believed to be
associated with prolonged exposure of the oesophageal
PPI, proton pump inhibitor.
squamous mucosa to acid reflux leading to more severe
erosive oesophagitis and formation of columnar metapla-
well without adverse events. The majority (93%) of sia.
patients reported a bitter taste associated with the Ome- In contrast to the data from prior studies in BE
NaBic suspension formulation. patients on PPI therapy demonstrating abnormal pH
profiles in 30–50% of patients,5–7 our current study dem-
DISCUSSION onstrated adequate control of intra-oesophageal acidity
Prior studies in the literature demonstrated that patients in 97% of patients on Ome-NaBic twice daily and 100%
with BE were more likely to manifest abnormal oesopha- control of supine oesophageal pH. The data were partic-
geal pH profiles on PPI therapy, despite adequate control ularly impressive since this is the first study in the medi-
of reflux symptoms. The presence of persistent patholog- cal literature to use Bravo pH monitoring in this patient

Table 2 | Results from 48-h Bravo pH monitoring studies

Baseline off PPI (N = 21) On Ome-NaBic 40 mg b.d. (N = 15)*

Variable First day Second day First day Second day

Total time pH <4 (%) 14 ± 6 (5–29) 14 ± 7 (3–29) 0.4 ± 0.5 (0–1.2) 0.6 ± 1.8 (0–7.2)
Upright time pH <4 (%) 16 ± 6 (7–28) 15 ± 8 (0–30) 0.7 ± 0.7 (0–2) 1 ± 3 (0–10)
Supine time pH <4 (%) 10 ± 11 (0–34) 12 ± 10 (0–36) 0.06 ± 0.1 (0–0.5) 0.05 ± 0.2 (0–0.7)
Number of reflux episodes 81 ± 42 (12–176) 69 ± 46 (4–172) 9 ± 8 (0–22) 8 ± 18 (0–72)
DeMeester score 47 ± 22 (19–97) 46 ± 29 (1–103) 2 ± 2 (0–5) 3 ± 6 (0–24)
Number of 24 h periods with 0/21 (0%) 1/21 (5%) 15/15 (100%) 14/15 (93%)
normal oesophageal pH
Mean QOLRAD score 5 ± 1 (3.4–7) 163 ± 18 (127–175)
Mean RDQ heartburn score 10 ± 6 (0–18) 0.5 ± 1.5 (0–6)

* The second Bravo pH studies occurred between days 22–29 of therapy depending upon whether the patient was initially using
omeprazole vs. another PPI prior to study entry. One patient with 8 cm segment of BE demonstrated pathological upright GERD
during the second day of pH recording on Ome-NaBic with total time pH <4 of 7.2%. P-values for all variables <0.001 comparing
results on and off of Ome-NaBic therapy.

Aliment Pharmacol Ther 2012; 35: 803-809 807

ª 2012 Blackwell Publishing Ltd
L. B. Gerson et al.

2.0
Table 3 | Prior literature on intra-oesophageal pH
1.8 values in BE patients on PPI therapy
Upright % pH < 4
1.6 Supine % pH < 4
No (%)
1.4 patients with
abnormal
% time pH < 4

1.2 No. BE oesophageal

1.0 Author, Year patients PPI (no Pts) pH on PPI*

0.8 Katz (1998)19 12 Ome or Lan† 6 (50%)

Oatu-Lascar 30 Lan 12 (40%)
0.6
(1998)20
0.4 Fass (2000)21 25 Ome 6 (24%)
Yeh (2003)7 13 Eso 8 (62%)
0.2 Gerson 48 Ome (N = 13) 24 (50%)
0.0 (2004)5 Lan (N = 11)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Rab (N = 20)
Eso (N = 3)
Panto (N = 1)
Figure 3 | Graph demonstrating percentage time
Gerson 17 Ome (N = 9) 5 (56%) Ome
upright and supine pH <4% for each patient while on
(2005)14 Lan (N = 8) 5 (62%) Lan
Ome-NaBic therapy. Patients on prior omeprazole Rab (N = 17) 8 (47%) Rab
therapy underwent the second Bravo pH study after Spechler 31 Eso 40 mg t.i.d. 5 (16%)
3 weeks of Ome-NaBic therapy, while patients on other (2006)16 Eso 40 mg b.i.d. 7 (23%)
PPIs underwent repeat pH monitoring after 4 weeks of Eso 20 mg b.i.d. 6 (19%)
Ome-NaBic treatment. Current study 15 Ome-NaBic 1 (7%)

PPI, proton pump inhibitor.

cohort and collect data for 48-h time periods. The limi- * Defined as DeMeester score >14.7, % time oesophageal pH
tations of this study included the small sample size, a <4 greater than 5%, or supine time >3% during 24-h ambula-
tory oesophageal pH monitoring studies.
cohort of patients with primarily short-segment BE, and
† Number of patients taking Ome or Lan not stated in the
the fact that the study relied on comparison of intra-
text.
oesophageal pH levels on Ome-NaBic to other PPIs
based on historical controls.
Although this is the first study to examine oesopha- in BE patients. Approximately 40–62% of BE patients
geal pH profiles in BE patients on Ome-NaBic therapy, using PPI therapy q.d.s. or b.d. demonstrated abnormal
the prior literature has demonstrated excellent suppres- oesophageal pH profiles. In a 2006 study where esomep-
sion of daytime and nocturnal oesophageal pH on Ome- razole was prescribed up to three times daily, this value
NaBic in small cohorts of GERD patients who received fell to 16%.16 However, if we calculate that only 1/15
the drug once daily.15 In a prospective open-labelled (7%) of patients in our study demonstrated an abnormal
study of patients prescribed Ome-NaBic once daily either pH profile on Ome-NaBic b.d. therapy, this value
in the morning or at bedtime, four of five (80%) of the remains significantly improved compared to the prior
GERD subjects taking the medication in the morning published literature. Larger studies in BE patients are
normalised their oesophageal pH profiles on therapy, warranted to confirm these encouraging results.
and one (100%) subject administered the medication at The potential advantages of Ome-NaBic appear to be
bedtime demonstrated a normal 24-h oesophageal pH related to the ability to provide sustained control of in-
recording period.9 In a subsequent follow-up of this tragastric pH at steady state. Prior studies have demon-
study, six of seven (86%) subjects prescribed Ome-NaBic strated that administration of Ome-NaBic at bedtime
in the morning demonstrated normal pH profiles, and was effective for the control of nocturnal oesophageal
three of five (60%) subjects taking the medication at bed- pH and night-time GERD symptoms.17 In a randomised,
time demonstrated normal pH profiles.10 open-label, crossover trial, Ome-NaBic at bedtime
There have not been prior studies examining b.d. dos- offered significantly better control of nocturnal gastric
ing of Ome-NaBic on oesophageal pH profiles to date. pH compared with once or twice daily delayed-release
Table 3 highlights the prior literature with regard to pantoprazole tablets, in 36 patients with symptomatic
control of intra-oesophageal pH values on PPI therapy nocturnal GERD.18

808 Aliment Pharmacol Ther 2012; 35: 803-809

ª 2012 Blackwell Publishing Ltd
 Omeprazole-sodium bicarbonate in Barrett's oesophagus

In summary, despite the prior literature demonstrat- or not the usage of Ome-NaBic in patients with BE
ing significant nocturnal gastric acid breakthrough and might be associated with reduction of the risk of pro-
pathological supine oesophageal pH reflux in patients gression to dysplasia.
with BE, we demonstrated that the usage of Ome-NaBic
twice daily before breakfast and bedtime resulted in ACKNOWLEDGEMENTS
100% control of 48 h supine intra-oesophageal acid pro- Declaration of personal interests: Lauren B. Gerson has
files. These results are intriguing given the prior chal- served as a consultant to Takeda, Eisai and Santarus,
lenges with control of pH values in these patient and received research funding from Santarus, Inc. Shilpa
cohorts. It should be noted that our findings are preli- Mitra has no personal interests to declare. William Ble-
minary due to premature stopping of the study, and ker is an employee of Santarus. Philip Yeung is an
that furthermore comparative studies should be encour- employee of Santarus. Declaration of funding interests:
aged. Future studies are indicated to examine whether This study was funded in full by Santarus, Inc.

REFERENCES
1. Ward EM, Wolfsen HC, Achem SR, 8. Peghini PL, Katz PO, Bracy NA, Castell 16. Spechler SJ, Sharma P, Traxler B,
et al. Barrett's esophagus is common in DO. Nocturnal recovery of gastric acid Levine D, Falk GW. Gastric and
older men and women undergoing secretion with twice-daily dosing of esophageal pH in patients with Barrett's
screening colonoscopy regardless of proton pump inhibitors. Am J esophagus treated with three
reflux symptoms. Am J Gastroenterol Gastroenterol 1998; 93: 763–7. esomeprazole dosages: a randomized,
2006; 101: 12–7. 9. Francis D, Dabade T, Abdalla A. double-blind, crossover trial. Am J
2. Gerson LB, Shetler K, Triadafilopoulos Randomized open-label trial to assess Gastroenterol 2006; 101: 1964–71.
G. Prevalence of Barrett's esophagus in the impact of dosage timing of 17. Katz PO. Review article: putting
asymptomatic individuals. omeprazole/sodium bicarbonate immediate-release proton-pump
Gastroenterology 2002; 123: 461–7. (zegerid 40 mg) on healing of severe inhibitors into clinical practice –
3. Rex DK, Cummings OW, Shaw M, reflux esophagitis: preliminary results. improving nocturnal acid control and
et al. Screening for Barrett's esophagus Am J Gastroenterol 2009; 104: S8. avoiding the possible complications of
in colonoscopy patients with and 10. Romero Y, Francis D, Grothe R. excessive acid exposure. Aliment
without heartburn. Gastroenterology Randomized open-label trial to assess Pharmacol Ther 2005; 22(Suppl. 3): 31–
2003; 125: 1670–7. the impact of dosage timing of 8.
4. Rubenstein JH, Scheiman JM, Sadeghi omeprazole/sodium bicarbonate 18. Castell D, Bagin R, Goldlust B, Major J,
S, Whiteman D, Inadomi JM. (zegerid 40 mg) on healing of severe Hepburn B. Comparison of the effects
Esophageal adenocarcinoma incidence reflux esophagitis: a progress report. of immediate-release omeprazole
in individuals with gastroesophageal Am J Gastroenterol 2010; 105: S22. powder for oral suspension and
reflux: synthesis and estimates from 11. Pandolfino JE, Richter JE, Ours T, pantoprazole delayed-release tablets on
population studies. Am J Gastroenterol Guardino JM, Chapman J, Kahrilas PJ. nocturnal acid breakthrough in patients
2011; 106: 254–60. Ambulatory esophageal pH monitoring with symptomatic gastro-oesophageal
5. Gerson LB, Boparai V, Ullah N, using a wireless system. Am J reflux disease. Aliment Pharmacol Ther
Triadafilopoulos G. Oesophageal and Gastroenterol 2003; 98: 740–9. 2005; 21: 1467–74.
gastric pH profiles in patients with 12. Wiklund IK, Junghard O, Grace E, 19. Katz PO, Anderson C, Khoury R,
gastro-oesophageal reflux disease and et al. Quality of life in reflux and Castell DO. Gastro-oesophageal reflux
Barrett's oesophagus treated with dyspepsia patients. Psychometric associated with nocturnal gastric acid
proton pump inhibitors. Aliment documentation of a new disease-specific breakthrough on proton pump inhibitors.
Pharmacol Ther 2004; 20: 637–43. questionnaire (QOLRAD). Eur J Surg Aliment Pharmacol Ther 1998; 12: 1231–4.
6. Milkes D, Gerson LB, Triadafilopoulos Suppl 1998; 583: 41–9. 20. Ouatu-Lascar R, Triadafilopoulos G.
G. Complete elimination of reflux 13. Locke GR, Talley NJ, Weaver AL, Complete elimination of reflux symptoms
symptoms does not guarantee Zinsmeister AR. A new questionnaire does not guarantee normalization of
normalization of intraesophageal and for gastroesophageal reflux disease. intraesophageal acid reflux in patients with
intragastric pH in patients with Mayo Clin Proc 1994; 69: 539–47. Barrett's esophagus. Am J Gastroenterol
gastroesophageal reflux disease (GERD). 14. Gerson LB, Shetler K, Triadafilopoulos 1998; 93: 711–6.
Am J Gastroenterol 2004; 99: 991–6. G. Control of intra-oesophageal and intra- 21. Fass R, Sampliner RE, Malagon IB,
7. Yeh RW, Gerson LB, Triadafilopoulos gastric pH with proton pump inhibitors in et al. Failure of oesophageal acid
G. Efficacy of esomeprazole in patients with Barrett's oesophagus. Dig control in candidates for Barrett's
controlling reflux symptoms, Liver Dis 2005; 37: 651–8. oesophagus reversal on a very high
intraesophageal, and intragastric pH in 15. Lutsi B, Hirano I. Ambulatory pH dose of proton pump inhibitor.
patients with Barrett's esophagus. Dis monitoring: new advances and indications. Aliment Pharmacol Ther 2000; 14: 597
Esophagus 2003; 16: 193–8. Gastroenterol Hepatol 2006; 2: 835–42. –602.

Aliment Pharmacol Ther 2012; 35: 803-809 809

ª 2012 Blackwell Publishing Ltd
 113

Development of an oral formulation of

omeprazole
PILBRANT A and CEDERBERG C
Departments of Pharmaceutics and Medicine,
AB Hassle, Molndal, Sweden.

Pilbrant A,Cederberg C. Development of an oral formulation of omeprazole.

Scand J Gastroenterol 1985;2O(suppl 108):113-120.
Scand J Gastroenterol Downloaded from informahealthcare.com by York University Libraries on 08/04/13

Omeprazole has a low water solubility and is chemically labile in an acid environment. In
the formulation of an oral dosage form of omeprazole the possibilities of dissolution rate
limited absorption and preabsorption degradation must be kept in mind. A water suspen-
sion of omeprazole was tested in a pilot bioavailability study. The suspension was
given to six healthy, fasting volunteers on two occasions - together with sodium bicar-
bonate solution and together with the same volume of water. When the suspension was
given with water the bioavailability was reduced by about 50 Vo owing to preabsorption
degradation. In another bioavailability study the slowest of three granule formulations
with differing in vitro dissolution rates showed a reduced extent of absorption.
A controlled-release pellet formulation (enteric-coated) was formulated and tested in a
series of bioavailability studies. A single dose given with food resulted in a delayed
absorption and possibly lower bioavailability than under fasting conditions. When the
granules were given on an empty stomach before the morning meal the length of time
between dosage and meal was of no importance. Concomitant administration of a liquid
For personal use only.

antacid had no influence on the bioavailability of omeprazole.

Key-words: Bioavailability; controlled release; dosage form; enteric coating; omeprazole;
stability
Ake Pilbrant, Farm. Lic., Dept of Pharmaceutics, A B Hassle,
S-43183 Molndal. Sweden

Introduction than 10 minutes at pH-values below 4. At pH 6.5 the

half-life of degradation is 18 hours; at pH 11 about
Omeprazole (Figure 1) is a substituted benz- 300 days.
imidazole which selectively inhibits the proton
pump in the gastric mucosa (1, 2). Omeprazole is Preformulation studies have shown that moisture,
very slightly soluble in water, but is very soluble in solvents and acidic substances have a deleterious ef-
alkaline solutions as the negatively charged ion. It is fect on the stability of omeprazole and should be
an ampholyte with pKa-4 (pyridinium) and 8.8 avoided in pharmaceutical formulations.
(benzimidazole).

Omeprazole degrades very rapidly in water solu-

tions at low pH-values. Figure 2 shows a plot of the
logarithm of the observed rate constant for
degradation as a function of pH. In each experi-
ment, the initial, pseudo-first-order rate of
degradation was calculated from the amount of un- I
Figure 1. Omeprazole, H 168168, 5-methoxy-2-[[(4-
changed omeprazole in buffer (3). The methyl] sulfinyl] -1H-
methoxy-3,S-dimethyl-2-pyridinyl)
rate of degradation proceeds with a half-life of less benzimidazole.
 114
Scand J Gastroenterol Downloaded from informahealthcare.com by York University Libraries on 08/04/13
For personal use only.

The aim was to develop a stable, oral, phar- omeprazole - particle surface area larger than 2.5
maceutical formulation of omeprazole in doses of m2/g - was suspended in a 0.25 Yo water solution
20-60 mg, with acceptable bioavailability charac- of methylcellulose also containing sodium bicar-
teristics. bonate as p H buffer. The suspensions can be stored
at refrigerator temperature for a week, or stored
deep frozen for more than a year, and still retain
Pharmaceutical con- 99 To of their initial potency. To avoid acidic
siderations degradation of omeprazole, the suspensions must
For a substance which is very slightly soluble in be administered orally together with large amounts
water and which rapidly degrades in the acid en- of pH buffering substances.
vironment of the stomach, there is a limited number
of options as far as pharmaceutical development is Solid dosage forms
concerned.
There are two principle options for the formulation
of an oral, solid dosage form of omeprazole:
Solutions
0 A conventional oral dosage form from which
In animal experiments and in initial studies in man
omeprazole is released and absorbed rapidly
it is highly preferable to use water solutions of the
enough to avoid degradation in the stomach.
drug in order to avoid influences of the dosage form
on the pharmacokinetics and pharmacodynamics 0 An enteric-coated dosage form, which releases
of the drug. Omeprazole is, however, only soluble in omeprazole for absorption in the small
alkaline water solutions with physiologically unac- intestine.
ceptable, high pH-values. The first option was ruled out in a pilot
bioavailability study (see below), where it was
Suspensions shown that more than half of the omeprazole in a
In toxicological and phase I clinical studies, suspen- rapidly dissolving dosage form degrades in the
sions of omeprazole in water were used. Micronised stomach.
 115

An enteric-coated dosage form, which does not with food were found to stay in the stomach for
release the active ingredient for dissolution and ab- more than 14 hours (7). Enteric-coated aspirin
sorption until it has been transported down to the tablets showed a prolonged and erratic gastric emp-
neutral reacting part of the small intestine, offers tying, while enteric-coated granules of a size of
the best possibilities. The dosage form - a tablet, a about 1 mm dispensed in hard gelatine capsules
capsule or granules - is coated with a polymer, dispersed in the content of the stomach and
which is insoluble in acid media but soluble in gradually emptied in to the small intestine in a
neutral to alkaline media. Depending on the choice reproducible way (5).
of polymer and on the thickness of the coated layer,
In the pharmaceutical manufacture o f coated
the pH-solubility profile of the enteric-coating can
dosage forms, it is impossible to coat every single
be controlled.
unit in a batch perfectly. A small fraction of units
Scand J Gastroenterol Downloaded from informahealthcare.com by York University Libraries on 08/04/13

An enteric-coated dosage form of omeprazole must will have imperfect coating, or else be damaged
be perfectly coated and acid resistant, since, if any during further handling and transport. For a single
drug leaks out of the dosage form in the stomach, it unit, enteric-coated dosage form of omeprazole,
is almost immediately degraded. The same is the e.g, a tablet, there is always a small risk that all
case if an acidic medium can diffuse into the dosage omeprazole contained in the dose can be degraded
form through pin-holes or damage in the enteric- in the stomach. For a multiple unit, enteric-coated
coating. dosage form, e.g, enteric-coated granules dispensed
in a hard gelatine capsule, only the omeprazole con-
Solid particles of a size exceeding 2-4 mm, such as
tained in a few pellets out of a total of hundreds is
enteric-coated tablets or capsules, are known to re-
lost. Our efforts were, therefore, concentrated on
main in the stomach for a long time before they are
developing an enteric-coated granule formulation.
emptied into the small intestine (4-6). Non-
For personal use only.

disintegrating, coated tablets administered together

Formulation and in vitm
%
testing
In the formulation of a solid dosage form of
omeprazole, having a low water solubility of 0.1
mg/ml, there is always the risk of dissolution rate
limited absorption. Three spherical granule for-
mulations containing 10 '70 of omeprazole were
manufactured and classified. The fraction with a
diameter between 0.7 - 1.0 mm was used for further
studies.
The dissolution ratein vitro was determined using a
slightly modified beaker method according to Levy
and Hayes (8). 500 ml of deaerated phosphate buff-
er pH 6.5, ionic strength 0.1, was kept at +37"C
and stirred at a rate of 100 rpm. An amount of
granules corresponding to 10mg of omeprazole was
added and the amount dissolved determined from
0 10 20 30min
the continuous recording of the absorbance at 300
Figure 3. Dissolution of omeprazole from three granule
formulationsin vitro in phosphate buffer, pH= 6.5. The
nm in a spectrophotometer using 1 cm flow cells.
cumulative amount dissolved ("0) is plotted as a function The cumulative amount dissolved is plotted as a
of time. function of time in Figure 3. All three formulations
granules, batch H 370-9-1
A granules, batch H 370-8-1 released most of their content of omeprazole within
V granules, batch H 370-1-1 30 minutes.
 116

A pilot bioavailability study showed that the A suspension of micronised omeprazole, 60 mg, in
two faster dissolving granules were absorbed to the water, 50 ml, also containing 8 mmoles of sodium
same extent, while the extent of absorption of the bicarbonate (pH=9) was administered in the
more slowly dissolving granules was reduced. following way: In the morning, after fasting for at
least ten hours, the volunteers were given a solution
Rapidly dissolving, spherical granules containing
of 8 mmoles of sodium bicarbonate in 50 ml of
10 To of omeprazole were enteric-coated with ap-
water. Five minutes later the volunteers took the
proximately 15 To by weight of polymer. The
omeprazole suspension and rinsed it down with
coating was sprayed onto the granules from an
another 50 ml of sodium bicarbonate solution. 10,
organic solvent solution in a fluidised bed ap-
20, and 30 minutes later, a further 50 ml of sodium
paratus. After drying, the coated granules were
bicarbonate solution was taken. The amount of
analysed and dispensed in hard gelatine capsules.
Scand J Gastroenterol Downloaded from informahealthcare.com by York University Libraries on 08/04/13

sodium bicarbonate is sufficient to buffer the p H of

The granules were tested for acid resistance in vitro the gastric content to neutral values for at least 45
in the following way: An amount of granules cor- minutes.
responding to 10mg of omeprazole was dispersed in
In another experiment, a suspension of omeprazole
SO0 ml of 0.1 molar hydrochloric acid at a
in water (pH adjusted to 9 by sodium hydroxide)
temperature of +37 " C . Stirring was done with a
was administered according to the same protocol as
paddle at a rate of 100 rpm. After two hours the
above but with the sodium bicarbonate solutions
granules were removed from the vessel, rinsed with
replaced by water. Doses were given in random
water, dried and analysed for omeprazole by liquid
order with a week's interval between doses. Venous
chromatography. After two hours exposure to
blood was sampled frequently over a period of four
acid more than 85 To of the initial amount of
hours and blood plasma was analysed for
omeprazole was recovered. When tested for dissolu-
For personal use only.

omeprazole by liquid chromatography (9).

tion rate in vitro in a medium of p H 6.5, more than
90 To dissolved within 15 minutes. The results of the plasma analyses are shown in
Figure 4. The absorption of omeprazole proceeds
Omeprazole capsules have an acceptable storage
rapidly and peak plasma concentrations were
stability when stored in a proper package. The
stability characteristics of omeprazole - the
substance is sensitive to moisture - necessitate the *mOl I

presence of a desiccant in the package.

Bioavailability evaluation of
dosage forms
Omeprazole suspension given
with and without pH-buffers
The solubility and stability properties of
omeprazole prevent the use of water solutions as the
reference formulation in animal and human
studies. A rapidly dissolving suspension of
micronised omeprazole is the second best choice as
the reference formulation. However, since
omeprazole degrades rapidly in an acid environ-
ment, it is essential to know the magnitude of the 0 1 2 4 h

.
degradation occurring prior to the absorption of an Figure 4. plasma concentration of omeprazo~ein six
oral dose. A pilot bioavailability study was healthy, fasting volunteers, mean + or - SEM after oral
administration of omeprazole, 60 mg, given as:
therefore performed using six healthy, male buffered suspension
volunteers. A suspension without buffer
 117

reached after a mean of 13 minutes in both ex- From this experiment, it can be concluded that the
periments. The area under the plasma concentra- in vitro dissolution rate method used can dis-
tion time curve (AUC) was calculated by the criminate between acceptable and non-accept-
trapezoidal method up to four hours after ad- able batches. In order to be fully absorbed, the
ministration and extrapolated to infinity by in vitro dissolution rate should be as high as, or
dividing the last plasma concentration by the higher than, that of granules H 370-8-1.
negative slope of the terminal, linear part of the
log/linear plasma concentration time curve. Bioavailability of enteric-coated
When the omeprazole suspension was given granules
together with sodium bicarbonate buffer, the mean Six, healthy, male volunteers participated in a three-
AUC was 4.8 pmol x h/l (range 2.8 - 8.8). With- way, cross-over bioavailability study. They received,
Scand J Gastroenterol Downloaded from informahealthcare.com by York University Libraries on 08/04/13

out the buffer protection the AUC was reduced in random order, 60 mg single, oral doses of
to a mean of 2.1 pmol x h/l (44 Vo), indicating omeprazole either as a buffered suspension given
that more than half of the dose was lost due to together with sodium bicarbonate solution, or as
degradation in the acidic stomach. enteric-coated granules dispensed in hard gelatine
A straight-forward pharmacokinetic analysis of the capsules given together with 300 ml of water on an
data showed that the absorption of omeprazole was empty stomach or as enteric-coated granules in cap-
rapid and completed within the period during sules together with a meal. In each experiment,
which the stomach was neutral. The results clearly standardised meals were served 2.5 and 6 hours
show that a conventional, non-buffered, oral after administration of the dose. Venous blood was
dosage form of omeprazole will have a low systemic sampled frequently for four hours (suspension) or
bioavailability owing to preabsorption degrada- seven hours (granules). Blood plasma was analysed
For personal use only.

tion of omeprazole in the stomach. for omeprazole by liquid chromatography accord-

ing to Persson et a1 (9).
Bioavailability of granules - in-
The results of the plasma analyses are shown in
fluence of dissolution rate Figure 5 . The absorption of omeprazole after the
A pilot bioavailability study in six, healthy suspension was given was rapid, and peak plasma
volunteers was performed in order to clarify the in- concentrations were reached within 10 - 20
fluence of the dissolution rate on the absorption of minutes. After administration of the enteric-coated
omeprazole. Three granule formulations - the
dissolution curves are shown in Figure 3 - were
tested using buffered suspension as the reference
formulation. In order to avoid problems with the
degradation of omeprazole, doses of 60 mg were
given together with sodium bicarbonate, as describ-
ed above. Venous blood was frequently sampled and
blood plasma analysed for omeprazole (9). The
AUC for the two faster dissolving granules
(H 370-9-1 and H 370-8-1) relative to that of the
reference formulation was 95 and 92 Vo, respective-
ly. The granules with the lowest in vilro dissolu-
tion rate (H 370-1-1) were absorbed to a lower extent
and had a mean relative AUC of 73 Vo only. The
I-^_ ^^
-1

sorption. A enteric-coated granules with breakfast

 118

granules a certain time was required for gastric between dosing and food intake. To be able to
emptying and for dissolution of the enteric-coating answer the question, we performed a bioavail-
before absorption of omeprazole started. In most ability study comparing omeprazole enteric-
cases, gastric emptying occurred in connection with coated granules given 15 minutes before the
the meal served 2.5 hours after the dose. In one sub- breakfast with the same dose given 2 minutes
ject, when enteric-coated granules were given with before the meal. A buffered suspension was
food, gastric emptying of granules did not start un- again used as the reference formulation.
til in connection with the second meal, served six Twelve healthy volunteers participated in the
hours after the dose. study. The doses were given as described above.
Standardised meals were served after 2.5, 6, 10
The plasma concentration-time curves obtained
and 13 hours. Blood samples were collected
after administration of enteric-coated granules
Scand J Gastroenterol Downloaded from informahealthcare.com by York University Libraries on 08/04/13

over a period of 6 hours (suspension) and 24

were flat and broad, and peak plasma soncentra-
hours (granules).
tions were low. The total amount absorbed, as
reflected by the AUC, was, however, only de- Eleven of the subjects completed the study and are
creased by 14 070 when the granules were given on an included in the results. The resulting mean plasma
empty stomach in comparison with the buffered concentration time curves are shown in Figures 6
suspension. The corresponding figure for en- and 7. It is interesting to note that theabsorption of
reric-coated granules administered wirh a meal omeprazole from the enteric-coated granules in
is higher, but since absorption of omeprazole some subjects started as early as 30 minutes after
was not completed in all subjects when the ex- the dose and that most subjects had a second plas-
periment was terminated, the exact figure is un- ma concentration peak shortly after the second
known. Although this study is not fully con- meal served 2.5 hours after dose.
For personal use only.

clusive regarding the bioavailability of omeprazole The AUC, relative to that of the reference formula-
given with food, it is rxommended that ome-
tion, was very similar in the two experiments with
prazole should be taken in the morning before
breakfast. prnol I
7
The effect of omeprazole on gastric acid secretion is
long lasting (10). The effect is not a direct function
of blood concentration of omeprazole at any time, 6

but is rather a function of the total amount of

orneprazole reaching the general circulation, i.e., 5
directly proportional to the AUC ( 2 , 10). This
means that the same pharmacological effect is
achieved with dosage forms of omeprazole produc- 4
ing equal AUCs. The shapes of the plasma concen-
tration-time curves are of no importance.
3

Bioavailability of enteric-
coated granules administered 2

at different times before

break fast 1

When omeprazole enteric-coated granules are given

with a meal, the rate of absorption of omeprazole is 0
reduced. Patients are therefore recommended to i, i 4 6 h

take the dose on an empty stomach before the morn- Figure 6 . Plasma concentration of omeprazole, mean
ing meal. However, in practice it is
+ or - SEM, in eleven healthy volunteers given a 60 mg
single, oral dose of omeprazole as a buffered suspension
necessary to know what length of time is required under fasting conditions.
 119

pmolil
1.5

1
Scand J Gastroenterol Downloaded from informahealthcare.com by York University Libraries on 08/04/13

0.5

0
i
0 2 4 6 8 10 12 24 h
For personal use only.

Figure 7. Plasma concentration of omeprazole, men + or enteric-coated granules 15 minutes before breakfast
- SEM, in eleven healthy given 60 mg single, oral doses of A enteric-coated granules 2 minutes before breakfast
omeprazole as:

enteric-coated granules; 65.5 070 when given 15 Six healthy volunteers were given, in random
minutes before breakfast and 66.6 070 when given 2 order, enteric-coated granules with and without
minutes before breakfast. The variability in the concomitant administration of an aluminium-
AUC between doses within subjects was small, as magnesium hydroxidekarbonate suspension. The
can be seen in Figure 8. The conclusion of this study dose was given on an empty stomach and venous
is that the omeprazole dose can be given before the blood samples collected during a period of seven
morning meal, and that there is no need to specify hours. In one experiment the subjects were
any time between the administration and the start pretreated with antacid the day before the ome-
of the morning meal. prazole administration. 10 ml doses of a liquid ant-
acid with an acid-binding capacity of 85 mmol
Bioavailability of enteric- per dose (Novaluzid@, AB Hassle, Sweden) were
given one and three hours after each meal and at
coated granules - interaction bed time; a total of seven doses. On the morning of
with antacids the next day, another 10ml dose was given just prior
In the clinical treatment of ulcer, antacids are often to the dose of omeprazole enteric-coated granules.
prescribed together with inhibitors of gastric acid In the other experiment omeprazole enteric-coated
secretion. Antacids may interfere with the function granules were administered without antacid treat-
of an enteric-coated dosage form and cause ment.
dissolution of the coating in the stomach. For The results of the plasma analyses are summarised
omeprazole this could mean an increased risk of in Figure 9, which shows the individual and mean
degradation in the stomach. We therefore tested the AUCs. The mean AUC was practically identical in
influence of a liquid antacid on the bioavailability the two experiments. As can be seen in Figure 9, the
of omeprazole enteric-coated granules. variability in the AUC within each subject is small,
 120

lpniol I h I prnol x hi1

Scand J Gastroenterol Downloaded from informahealthcare.com by York University Libraries on 08/04/13

' buffered enteric coated granules

suwersion 15 rnn 2 min
betore belore
breakfasr breakfast

Figure 8. AUC (umolx h/l) in eleven healthy volunteers

given 60 mg single, oral doses of omeprazole as a buffered
suspension and enteric-coated granules 15 or 2 minutes
For personal use only.

before breakfast.
_ _ individual values
..__ mean
. values

whereas the variability between subjects is WI t hout

antacids antacids
substantial.
Figure 9. AUC (umolxh/l) in six healthy volunteers
Conclusion: Co-administration of antacids has no given 60 mg single, oral doses of omeprazole enteric-
coated granules with and without antacids.
influence on the bioavailability of omeprazole -individual values
given as enteric-coated granules. _ _ - - mean values

6. Bechgaard H. Critical factors influencing

References gastrointestinal absorption - What is the role of
I . Wallmark B, Berglindh T , Mirdh S C't al. The pellets? Acta Pharm Technolog 1982;28:149-57.
mechanism of action of omeprazole. Scand J 7. Bogentoft C, Jonsson UE, Eriksson R, Alpsten M .
Gastroenterol 1985;2O(suppl 108j:37-51. Gastric emptying of pellets and tablets in healthy
2. Larsson H, Mattsson H, Sundell G, Cdrlsson E. subjects under fasting and nonfasting conditions.
Animal pharmacodynamics of omeprazole. A 43rd International Congress of Pharmaceutical
survey of the pharmacological properties of Sciences FIP, Montreux 1983, p 102.
omeprazole in animals. Scand J Gastroenterol 8. Levy G , Hayes BA. Physicochemical basis of the
1985;?O(~~ppl 108):23-35. buffered acetylsalicylic acid controversy. N Engl J
3 . Erickson GEM. Unpublished results 1980. Med 1960;262: 1053-8.
-1. Blythe RH, Grass GM, MacDonell DR. The for- 9. Persson B-A, Lagerstrom P - 0 , Grundevik 1. Deter-
mulation and evaluation of enteric-coated aspirin mination of omeprazole and metabolites in plasma
tablets. Am J Pharm 1959;131:206-16. and urine. Scand J Gastroenterol 1985;2O(suppl
.% . Bogentoft C, Carlsson 1, Ekenved G, Magnusson A. 108):71-7.
Influence of food o n the absorption of 10. Cederberg C, Ekenved G, Lind T, Olbe L. Acid in-
acetylsalicylic acid from enteric-coated dosage hibitory characteristics of omeprazole in man.
forms. Eur J Clin Pharmacol 1978;14:351-5. Scand J Gastroenterol 1985;2O(suppl 108):105-12.
Dig Dis Sci
DOI 10.1007/s10620-013-3017-y

ORIGINAL ARTICLE

Once-Daily Omeprazole/Sodium Bicarbonate Heals Severe

Refractory Reflux Esophagitis with Morning or Nighttime Dosing
Diana M. Orbelo • Felicity T. Enders • Yvonne Romero • Dawn L. Francis •
Sami R. Achem • Tushar S. Dabade • Michael D. Crowell • Debra M. Geno •
Ramona S. DeJesus • Vikneswaran Namasivayam • Steven C. Adamson •
Amindra S. Arora • Andrew J. Majka • Jeffrey A. Alexander • Joseph A. Murray •

Matthew Lohse • Nancy N. Diehl • Mary Fredericksen • Kee Wook Jung •

Margaret S. Houston • Angela E. O’Neil • David A. Katzka

Received: 5 July 2013 / Accepted: 27 December 2013

 Springer Science+Business Media New York 2014

Abstract were completed at baseline and follow-up. Intention-to-treat

Background Morning dose or twice-daily proton pump and per-protocol analyses were performed.
inhibitor (PPI) use is often prescribed to heal severe reflux Results Ninety-two of 128 (72 %) eligible subjects par-
esophagitis. ticipated [64 (70 %) male, mean age 58 (range 19–86),
Aim Compare the effect of single dose morning (control median BMI 29 (range 21–51), 58 C:34 D]. Overall, 81
arm) versus nighttime (experimental arm) omeprazole/ (88 %) subjects healed [n = 70 (76 %)] or improved
sodium bicarbonate (Zegerid) (IR-OME) on esophagitis [n = 11 (12 %)] erosions. There was no significant differ-
and gastroesophageal reflux symptoms. ence (morning vs. night) in mucosal healing [81 vs. 71 %,
Methods Adult outpatients with Los Angeles grade C or D (p = 0.44)] or symptom resolution [heartburn (77 vs. 65 %,
esophagitis were allocated to open-label 40 mg IR-OME p = 0.12), acid regurgitation (82 vs. 73 %, p = 0.28)].
once a day for 8 weeks in a prospective, randomized, parallel Prevalence of newly identified Barrett’s esophagus was
design, single center study. Esophagogastroduodenoscopy 14 % with half diagnosed only after treatment.
(EGD) and validated self-report symptom questionnaires Conclusions Once-daily IR-OME (taken morning or
night) effectively heals severe reflux esophagitis and
improves GERD symptoms. Results support the clinical
ClinicalTrials.gov, Number: NCT00693225. practice recommendation to repeat EGD after 8 weeks PPI

D. M. Orbelo (&) J. A. Murray

Department of Otolaryngology, Mayo Clinic, 200 First St., SW, e-mail: [email protected]
Rochester, MN 55905, USA
M. Fredericksen
e-mail: [email protected]
e-mail: [email protected]
F. T. Enders D. A. Katzka
Division of Biomedical Statistics and Informatics, Mayo Clinic, e-mail: [email protected]
200 First St., SW, Rochester, MN 55905, USA
e-mail: [email protected] D. L. Francis
S. R. Achem
Division of Gastroenterology and Hepatology, Mayo Clinic,
Y. Romero
D. M. Geno
A. S. Arora
4500 San Pablo Road, Jacksonville, FL 32224, USA
J. A. Alexander
J. A. Murray
M. Fredericksen
D. A. Katzka e-mail: [email protected]
Division of Gastroenterology and Hepatology, Mayo Clinic, 200
S. R. Achem
First St., SW, Rochester, MN 55905, USA
e-mail: [email protected]
e-mail: [email protected]
D. M. Geno T. S. Dabade
e-mail: [email protected] Department of Dermatology, Tufts Medical Center, 1615
Tremont St., Boston, MA 02120, USA
A. S. Arora
e-mail: [email protected]
J. A. Alexander
e-mail: [email protected]

123
 Dig Dis Sci

therapy in severe esophagitis patients to assure healing and with severe Los Angeles (LA) grade C or D erosive
exclude Barrett’s esophagus. esophagitis do not heal after 8 weeks of PPI therapy [2–5].
Among reasons cited for sub-optimal healing is poor con-
Keywords Gastroesophageal reflux disease
trol of nocturnal acid exposure [6–9] due to reduced
Esophagitis
Barrett’s esophagus
Prospective esophageal clearance from hypotensive peristalsis, reduc-
randomized controlled trial tion of swallowing during sleep, loss of normal upright
gravitational factors that enhance esophageal clearance
Abbreviations [10], obesity [11], and insufficient effect of PPI’s on noc-
BE Barrett’s esophagus turnal acid secretion [12]. Pharmacological attempts to
BMI Body mass index control nocturnal esophageal acid reflux have been diffi-
EGD Esophagogastroduodenoscopy cult. Reasons for this include the standard recommendation
GERD Gastroesophageal reflux disease to take their PPI upon arising in the morning. Inherent to
HH Hiatal hernia this standard is an adherence burden, specifically the need
HRA Histamine receptor antagonist to swallow their PPI immediately upon awakening, waiting
IMC Intestinal metaplasia of the cardia 20–60 min prior to masticating a solid. Arguably, this is
IR-OME Immediate release-omeprazole less burdensome than trying to remember to take a PPI
ITT Intention-to-treat analysis dose on an empty stomach prior to the last meal of the day;
LA Los Angeles classification system for erosive a recommendation that is often added when twice-daily
reflux esophagitis PPIs are empirically prescribed for those with severe, LA
LSBE Long segment Barrett’s esophagus Grade C or D, erosive esophagitis [13].
MDQ-30 Mayo Dysphagia Questionnaire–30 day A newer formulation of omeprazole (Zegerid), which
PPI Proton pump inhibitor combines omeprazole with sodium bicarbonate in lieu of
SSBE Short segment Barrett’s esophagus an enteric coating, is the only immediate-release PPI; with
demonstrated ability to control gastric pH when adminis-
tered at bedtime, in the absence of a meal [14, 15]. It has
Introduction been theorized that the rapid elevation of gastric pH with
sodium bicarbonate taken before bedtime may be a suffi-
Severe gastroesophageal reflux disease (GERD) can lead to cient stimulus to release gastrin and activate proton pumps,
gross esophageal epithelial injury [1]. While the advent of facilitating pump binding without the need for food stim-
proton pump inhibitors (PPIs) has revolutionized the ability ulus [16]. Thus, our a priori hypotheses are that the timing
to heal erosive esophagitis, as many as 10–40 % of patients of administration of immediate release omeprazole sodium
bicarbonate (IR-OME) will impact its efficacy in healing

M. D. Crowell M. Lohse
Division of Gastroenterology and Hepatology, Mayo Clinic, Department of Surgery, Virginia Mason Medical Center, 901 8th
13400 E. Shea Blvd., Scottsdale, AZ 85259, USA Ave., Seattle, WA 98104, USA
e-mail: [email protected] e-mail: [email protected]

R. S. DeJesus N. N. Diehl
Division of Primary Care Internal Medicine, Mayo Clinic, 200 Division of Biostatistics, Mayo Clinic, 4500 San Pablo Road,
First St., SW, Rochester, MN 55905, USA Jacksonville, FL 32224, USA
e-mail: [email protected] e-mail: [email protected]

V. Namasivayam K. W. Jung
Department of Gastroenterology and Hepatology, Singapore Department of Internal Medicine, Asan Medical Center, Seoul,
General Hospital, Outram Road, Singapore 169608, Singapore Korea
e-mail: [email protected] e-mail: [email protected]

S. C. Adamson M. S. Houston
A. E. O’Neil

Department of Family Medicine, Lake City Medical Center, Department of Family Medicine, Mayo Clinic, 200 First St., SW,
Mayo Clinic Health System, 500 W. Grant St., Lake City, Rochester, MN 55905, USA
MN 55041, USA e-mail: [email protected]
e-mail: [email protected]
A. E. O’Neil
e-mail: [email protected]
A. J. Majka
Division of General Internal Medicine, Mayo Clinic, 200 First
St., SW, Rochester, MN 55905, USA
e-mail: [email protected]

123
Dig Dis Sci

esophagitis and that convienient once-daily administration bleeding diathesis, neoplasm of the esophagus or stomach,
at nighttime, immediately prior to sleep, (experimental previous upper GI surgery [esophagectomy, Heller myot-
arm) will be superior in healing esophagitis compared to omy, hiatal hernia (HH) repair], diabetic gastroparesis,
morning administration 20–60 min prior to a meal (control esophageal motility disorder, Zollinger-Ellison syndrome,
arm). For a full list of a priori and post hoc aims, see infection with human immunodeficiency virus, history of
‘‘Appendix’’. gastric or small bowel obstruction, reported an inability to
read due to blindness, cognitive dysfunction or English
language illiteracy, diagnosis of a disorder which predis-
Methods poses to unreliable responses such as Schizophrenia or
dementia. Pregnant and lactating females, children younger
The study and manuscript adhere to the 2010 CONSORT than 18, and vulnerable populations were excluded. Those
Statement for parallel group randomized trials [17]. The with primary residence outside the USA or in prison, and
Mayo Clinic Institutional Review Board approved the anyone opting out of research activities at the Mayo Clinic,
study on March 18, 2008. Trial registration: Clinical Tri- were excluded.
als.gov Identifier: NCT00693225.
Interventions
Trial Design
All subjects received a one-on-one educational session,
A single center, randomized, open-label, parallel-group developed for a 6th grade reading level, describing the
pilot study was conducted. Subjects were randomly allo- normal physiology of the upper gastrointestinal tract, the
cated to receive daily study medication, either in the pathophysiology of HH and reflux esophagitis, food stuffs
morning or at night, for 8 weeks. No changes in trial design that contribute to reflux, and lifestyle modifications they
were needed [17]. can take to limit reflux prior to their invitation to partici-
pate in the trial as per our standard clinical care. IR-OME
Participants powder for oral suspension was supplied in individual
packets that are emptied into a small cup containing
Target population: Adults with severe (LA grade C or D) 15–30 ml (1–2 tablespoons) of water, one per day, for
erosive reflux esophagitis seen at an academic center in 8 weeks per standardized instructions. Instructions to sub-
Rochester, Minnesota, that provides both primary and jects allocated to the experimental nighttime arm specified
specialty care. Study accural occurred from June 20, that they keep the medication by their bedside; take it in a
2008, to May 24, 2010. All patients underwent a clini- standing or seated upright position immediately before
cally indicated esophagogastroduodenoscopy (EGD) as turning off the lights with the intention to sleep and not to
advised by their primary health care provider. Those use any food until the next morning. See ‘‘Appendix’’ for
meeting inclusion and exclusion criteria were then invi- complete instruction details.
ted to participate. GelusilTM use was recorded and distributed as an ‘‘on
demand’’ rescue antacid (dried aluminum hydroxide gel
Eligibility Criteria 200 mg, magnesium hydroxide 200 mg, simethicone
25 mg); no more than 12 tablets in a 24-h period, as
Inclusion Criteria directed on the box. All other antacids, HRAs, PPIs and
sucralfate were prohibited. No other medication was
Subjects were age 18 years or older. Patients with reflux altered.
esophagitis despite use of a non-omeprazole PPI(s) or Study coordinators conducted a structured phone inter-
histamine receptor antagonist (HRAs) were invited to dis- view at 3 weeks to: (1) clarify the start date of the study
continue their PPI and/or HRAs and participate without a medicine, (2) assure medication tolerance, (3) review the
wash-out period. proper medication administration, including clarifying dose
time in relationship to meals, (4) review concomitant
Exclusion Criteria medications, and (5) inquire about adverse events.
After 8 weeks, follow-up EGD was performed by a
Patients were excluded if they were already using some clinical (not research) endoscopist who was blind to the
formulation of omeprazole when diagnosed with esopha- study and subject allocation. The Mayo Dysphagia Ques-
gitis or reported intolerance or previously failed omepra- tionnaire–30 Day (MDQ-30) [18] was administered at
zole. Patients were excluded if they were using clopidogrel baseline and after 8 weeks of study medication. Chart
(Plavix), or had one or more of the following diagnoses: review was performed by a reviewer blind to endoscopic

123
 Dig Dis Sci

Assessed for eligibility (n=387)

Excluded (n=295)
♦ Did not meet inclusion criteria (n=259)
Already on omeprazole = 64
Previously failed omeprazole = 21
Intolerant of PPI therapy = 1
Age < 18 = 4
Bleeding diathesis = 7
Clopidogrel use = 2
Diabetic gastroparesis = 1
Esophageal varices = 3
Gastric/small bowel obstruction = 13
Morbid obesity = 1
Motility disorder = 28
Multiple medical issues = 33
Neoplasm of esophagus or stomach = 13
Pacemaker/defibrillator = 2
Previous GI surgery = 42
Provider would not permit contact = 19
Residence outside the US/prison = 2
Significant memory loss = 3

Randomized ♦ Declined to participate (n=18)

(n=92) ♦ Other reasons (n=18)

- Could not reach = 1
- Not planning to return to MN = 17

Allocated to morning dose (n=43) Allocated to bedtime dose (n=49)

♦ Received allocated intervention (n=43) ♦ Received allocated intervention (n=48)
♦ Did not receive allocated intervention (n=0) ♦ Did not receive allocated intervention (n=1)
Changed mind soon after randomization

Lost to follow-up (n=1) Did not return to Lost to follow-up (n=2) Did not return to
Rochester Rochester

Discontinued intervention (n=1) Perceived side Discontinued intervention (n=3) Perceived

effect side effect

N = 80 N = 80

Fig. 1 Flow diagram of the progress through the phases of a parallel-randomized trial of two groups (morning vs. nighttime dosing)

results and outcomes to categorize baseline symptoms in offered to each participant only after consent had been
four subjects who did not complete the MDQ-30. Subjects gained to avoid the appearance of coercion.
were invited to undergo a pH study while on IR-OME
during the final week of the study while using study Outcomes Measured
medication. Those with persistent erosions at follow-up
EGD were offered twice-daily dosing IR-OME for an Endoscopic outcome measures included the presence and
additional 8 weeks as a courtesy. Remuneration was degree of esophageal erosions. Results were categorized

123
Dig Dis Sci

based on initial severity, LA grade C or D, and level of Table 1 Subject demographics at baseline, intention-to-treat analysis
improvement at follow-up: healed, improved, or same or Factor Morning Nighttime p value
worse. See ‘‘Appendix’’ for endoscopic outcomes details. dosing dosing
GERD symptoms were measured as frequent, infrequent n = 43 n = 49
or absent. Symptom outcomes were categorised into three
Male sex [n (%)] 30 (70) 34 (69) 1.00
groups: complete resolution, partial resolution, and no
Age [years median (range)] 59 (19–86) 58 (26–86) 0.92
improvement. See ‘‘Appendix’’ for MDQ-30 scoring
BMI [kg/m2 median 29.4 30.1 0.98
details. (range)] (24.5–51.1) (21.2–47.8)
Additional outcomes measures included Barrett’s Hiatal hernia (HH)
esophagus (BE) defined by 1 cm or greater length salmon- Overall [n (%)] 36 (84) 40 (83) 1.00
colored mucosa in the tubular esophagus at EGD with Of those with HH [cm 3.9 (1–13) 3.4 (1–6) 0.25
histologic confirmation of intestinal metaplasia with goblet average (range)]
cells, most severe degree of dysplasia, and presence and Esophagitis severity [n (%)]
length of HH. Long segment BE (LSBE) is C3 cm length, LA grade C 26 (60) 32 (65) 0.67
while short segment BE (SSBE) ranges from 1 to 3 cm in LA grade D 17 (40) 17 (35)
length. Intestinal metaplasia of the cardia (IMC) is defined Using a PPI at time of baseline EGD [n (%)]
as salmon-colored mucosa in the esophagus shorter than Overall 12 (28) 15 (31) 0.82
1 cm in length demonstrating intestinal metaplasia with Among asymptomatic 3 of 9 (33) 3 of 7 (43) 1.00
goblet cells at histology. The 24-h pH test outcomes subjects
included total, upright, and supine time with pH \ 4 %. Among symptomatic 9 of 34 (26) 12 of 42 (29) 1.00
Body Mass Index (BMI) (kg/m2), sex (m/f), and date of GERD subjects
birth were collected. Primary and secondary outcomes Using a H2 blocker and or OTC antacid at time of baseline EGD
were pre-specified and unaltered per CONSORT guidelines [n (%)]
[17]. Overall 31 (74) 32 (70) 1.00
p values from Fisher’s exact test for categorical variables and the
Sample Size Wilcoxon rank sum test for continuous variables

Our clinical a priori estimate was that 70 % of subjects

would heal their esophagitis with daily morning IR-OME, underwent review of their medical records to assure eligi-
while 90 % would heal taking IR-OME at night. Resources bility before undergoing randomization.
were available to test 100 subjects and, assuming 50 sub-
jects were randomized to each arm (for a total of 100 Blinding
subjects), we would have 74 % power to detect a 20 %
difference in the timing of the doses. Although large Subjects, study coordinators, and Research Pharmacy per-
numbers of subjects met criteria for LA C or D esophagitis, sonnel were aware of treatment assignment (morning vs.
the majority were inpatients with at least one exclusion nighttime dosing). Subjects were encouraged to report their
criteria and, hence, not eligible to participate. See Fig. 1. use of ‘‘once-daily omeprazole/sodium bicarbonate’’ only
Due to slow accrual, this study was stopped early at if an inquiry was made by a health care provider (and to not
n = 92, resulting in 71 % power to detect a 20 % differ- volunteer further detail such as the time of day of admin-
ence in esophageal healing between morning and nighttime istration), to maintain blinding. Managing clinicians, en-
dosing. doscopists, investigators, and the biostatistical team
remained blinded to treatment time until the analysis was
Randomization and Masking complete. Over 14,000 EGDs are conducted annually by 58
individual endoscopists facilitating blinding.
Study medication was allocated in a 1:1 ratio (morning vs.
nighttime) using stratification (in blocks of 6) to balance: Statistical Methods
(1) sex, (2) current use of a PPI (yes/no), and (3) severity of
esophagitis (C/D). The Mayo Clinic Research Pharmacy An intention-to-treat (ITT) analysis was performed in
generated the random allocation sequence and assigned which subjects with missing follow-up symptom data
participants to intervention. After obtaining permission were presumed to have frequent symptoms, and missing
from their primary care provider, study coordinators con- EGD data were presumed to have severe LA D esopha-
tacted the patient. Patients who gave permission then gitis. Additionally, a sensitivity analysis assuming the

123
 Dig Dis Sci

LA Grades C & D Healing of Severe Erosive Reflux Esophagitis

Morning v. Nighttime Dosing of Omeprazole/Sodium Bicarbonate
Intention-To-Treat
Overall 81 of 92 (88 %) subjects healed [n = 70 (76 %)]
Overall LA C LA D or improved [n = 11 (12 %)] their severe reflux esopha-
p = 0.44 p = 0.19 p = 1.0 gitis following treatment with IR-OME once per day for
8 weeks. Figure 2 shows the ITT frequency of healing in
all subjects, and in subgroups with either LA C or D
Percent %

esophagitis, comparing results between morning versus

nighttime dosing. There was no statistically significant
difference in healing for the morning group compared to
the nighttime group (81 vs. 71 %, p = 0.44) or for subjects
with LA C versus D esophagitis (78 vs. 74 %, p = 0.80).
Lack of healing was not associated with elevated BMI
(p = 0.48). When comparing healing rates between the
small subset of subjects with severe esophagitis while on a
Morning Nighttime Morning Nighttime Morning Nighttime non-omeprazole PPI at baseline versus the majority of
Total N = 43 N = 49 N=26 N = 22 N = 17 N = 17 subject who were PPI naive at baseline, there was no sta-
Healed Improved Same or No EGD at tistically significant difference in healing (74 vs. 77 %,
Worse follow-up
p = 0.79).
GelusilTM use was reported in 23 of 43 (53 %) subjects
Fig. 2 Endoscopic outcomes following 8 weeks of omeprazole/
sodium bicarbonate once a day in subjects with LA grade C or D allocated to the morning group (mean 10 capsules over an
erosive esophagitis: morning versus nighttime dosing. Intention-to- 8-week period, range 1–75 capsules); and in 30 of 49
treat (61 %) subjects allocated to the nighttime group (mean 16
capsules over 8 weeks, range 2–100 capsules) (p = 0.22).
best-case scenario was completed. A per-protocol ana-
lysis was also performed. Categorical variables are GERD Symptoms at Baseline and Follow-up
summarized as proportions. Fisher’s exact test was used
to compare categorical variables between groups. Ordinal Table 2 shows the per-protocol distribution of heartburn,
logistic regression was used to adjust the association acid regurgitation, and overall GERD symptoms for sub-
between esophageal healing and time of dosing for jects allocated to morning versus nighttime dosing. ITT
baseline symptom frequency. For the exploratory aim results are shown in the Appendix (Table 6). At baseline,
assessing the association between nocturnal acid expo- when scored independently, heartburn and acid regurgita-
sure in subjects who healed versus those with persistent tion were reported equally. However, when the overall
esophagitis, statistical analysis was not performed due to GERD symptom score was calculated, subjects allocated to
limited sample size. the morning dose met criteria for frequent GERD more
often than subjects allocated to nighttime dosing (68 vs.
50 %, p = 0.02). Overall, there was no statistically sig-
Results nificant difference in symptom response comparing morn-
ing versus nighttime dosing after adjusting for baseline
Ninety-two of 128 (72 %) eligible patients provided writ- symptom frequency. Both dose strategies demonstrated
ten informed consent to participate (Fig. 1) and were used efficacy in resolving symptoms [heartburn (77 vs. 65 %,
in the ITT analysis (Table 1). Of these, 43 were random- p = 0.12), acid regurgitation (82 vs. 73 %, p = 0.28), and
ized to morning and 49 to nighttime dosing (due to ran- GERD (73 vs. 60 %, p = 0.14)].
domization based on blocks of 6). There were no At baseline, 16 of 92 (17 %) subjects with severe
significant group differences in sex, age, BMI, esophagitis esophagitis were asymptomatic, denying heartburn and
grade, presence or length of HH, or prestudy PPI use acid regurgitation, while 6 of 16 (38 %) were using a PPI at
(Table 1). Eight subjects dropped out by: changing their the time of diagnosis. Another 23 (25 %) had infrequent
mind prior to receiving study medication (n = 1), not symptoms. Only 53 (58 %) subjects with severe esopha-
returning for follow-up (n = 3), or stopping study medi- gitis had frequent (at least weekly) symptoms at baseline.
cation due to perceived side effect (n = 4). Eighty subjects Figure 3a, b shows per protocol heartburn and acid
underwent baseline and follow-up EGD and entirely regurgitation symptoms at baseline and after 8 weeks of
completed all questionnaires, and thus were used in the IR-OME for each individual participant. The ITT results
per-protocol analysis (Table 5 in the Appendix). (in which missing data are analyzed assuming a worst case

123
Dig Dis Sci

Table 2 Frequency of heartburn, acid regurgitation, and overall GERD symptoms in subjects allocated to morning (n = 40) versus nighttime
(n = 40) dosing of omeprazole/sodium bicarbonate at entry and following 8 weeks therapy, per-protocol analysis
Symptom Time of study drug administration Symptom frequency pc
At baseline After 8 weeks therapy
a
Absent Infrequent Freq. p Absent Infrequent Freq. pb
n (%) n (%) n (%) n (%) n (%) n (%)

Heartburn Morning dose 11 (28) 5 (12) 24 (60) 0.24 31 (77) 7 (18) 2 (5) 0.31 0.12
Nighttime dose 8 (20) 11 (27) 21 (53) 26 (65) 8 (20) 6 (15)
Acid regurgitation Morning dose 19 (47) 11 (28) 10 (25) 0.31 33 (82) 4 (10) 3 (8) 0.63 0.28
Nighttime dose 17 (42) 17 (43) 6 (15) 29 (73) 7 (18) 4 (10)
GERD Morning dose 9 (22) 4 (10) 27 (68) 0.02 29 (73) 7 (17) 4 (10) 0.51 0.14
Nighttime dose 5 (12) 15 (38) 20 (50) 24 (60) 9 (23) 7 (17)
a
p values from Fisher’s exact test comparing symptom frequency (absent, infrequent or frequent) for patients with morning and nighttime doses
b
p values from Fisher’s exact test comparing symptom frequency (absent, infrequent or frequent) for patients with morning and nighttime doses
c
Using ordinal logistic regression, this p value assesses whether time of dose is associated with symptom frequency at 8 weeks, after adjusting
for baseline symptom frequency

scenario) are depicted in Fig. 4a, b in the Appendix, while segment did not change after the esophagitis had healed as
sensitivity analyses (in which estimates include best case the erosions were located at the proximally displaced
scenarios) are presented in the Appendix (Table 7; Fig. 5a, aspect of the Barrett’s segment at the squamocolumnar
b). Notably, the outcome did not meaningfully change with junction.
per protocol versus ITT analysis. Overall, 76 subjects with At follow-up EGD, once the esophagitis had healed, 7
severe esophagitis reported GERD symptoms at baseline. additional subjects were diagnosed with BE for the first
Although the majority healed their esophagitis at follow-up time, the majority [5 of 7 (71 %)] of whom had LA D
[57 of 76 (75 %)] and reported complete symptom reso- esophagitis at baseline. Overall, 13 of 92 (14 %) subjects
lution [35 of 57 (61 %)], 22 of 57 (39 %) reported per- [7 (54 %) males, median age 58 years (range 30–77),
sistent reflux symptoms despite demonstrating complete median BMI 29.1 (range 24.6–51)] with severe esophagitis
mucosal healing. Five of 19 (26 %) subjects who reported had concomitant BE. In over half (54 %) of the cases, the
GERD symptoms at baseline but who did not heal at fol- diagnosis of BE would have been missed had a post-PPI
low-up became asymptomatic despite demonstrating per- research follow-up EGD not been done, including three
sistent mucosal injury. subjects with C5 cm segment lengths. Detailed demo-
In contrast, 16 (17 %) subjects denied heartburn and graphics including HH length at baseline EGD are shown
acid regurgitation at baseline, despite having severe in Table 8 in the Appendix. Overall, 7 of 58 (12 %) LA C
esophagitis. Endoscopy was done to investigate dysphagia and 6 of 34 (18 %) LA D subjects had BE.
and chest pain. Thirteen of 16 (81 %) subjects healed of After healing the esophagitis, 2 additional subjects met
whom 12 (92 %) remained asymptomatic. criteria for IMC (1 male, both age 55, BMIs of 30.8 and
33.4). One subject with SSBE had indefinite dysplasia. No
Prevalence of BE, Intestinal Metaplasia of the Cardia, subject had low- or high-grade dysplasia, or cancer. All BE
and HH in Patients with Severe Reflux Esophagitis subjects [n = 13 (100 %)] had a HH [mean length 5 cm
at Baseline (95 % CI 3–7); see Appendix Table 8 for specifics] while
HH was diagnosed in 64 of 79 (81 %) non-BE subjects
None of the subjects in this study were known to have BE [mean length 3 cm (95 % CI 3–4)] (p = 0.15).
prior to undergoing a clinically indicated EGD; at which
time severe reflux esophagitis was diagnosed thus, Effect of IR-OME on Esophageal Acid Exposure
prompting their recruitment into this study. At baseline, 6
subjects were diagnosed with prevalent BE and esophagitis Nineteen subjects underwent 24-h pH testing. Seven sub-
(LA C = 5, LA D = 1). Table 3 shows the initial extent of jects inadvertently discontinued the study medication prior
BE as measured using Prague criteria [19] and the degree to placement of the pH probe and were excluded from this
and extent of the erosions, if described. For subjects a priori exploratory subgroup analysis. Table 4 shows the
simultaneously diagnosed with both BE and esophagitis results of the 12 subjects who underwent pH testing while
during the baseline EGD, the length of the Barrett’s on study medication (n = 6 morning, n = 6 night). There

123
 Dig Dis Sci

Fig. 3 a Frequency of (a)

heartburn before and after
Heartburn Per-Protocol
Morning vs Nighttime: Before and After Treatment with Omeprazole/Sodium Bicarbonate
8 weeks of omeprazole/sodium
bicarbonate therapy comparing Morning Dose Nighttime Dose
morning versus nighttime
dosing. Per protocol.
b Frequency of acid
regurgitation before and after Frequent
8 weeks of omeprazole/sodium
bicarbonate therapy comparing
morning versus nighttime
dosing. Per protocol
Infrequent

Absent

Before After Before After

(b) Acid Regurgitation Per-Protocol

Morning vs Nighttime: Before and After Treatment with Omeprazole/Sodium Bicarbonate

Morning Dose Nighttime Dose

Frequent

Infrequent

Absent

Before After Before After

was 100 % healing with the morning dose (6/6) and 83 % Common PPI side effects (headache, nausea, abdominal
healing with the nighttime dose (5/6). None of the 6 sub- pain, and diarrhea) were not reported.
jects taking morning IR-OME had abnormal total esopha-
geal acid exposure (based on % duration of pH \ 4 of 4 %
or greater), and only 1/6 had abnormal nocturnal acid Discussion
exposure at 5.7 % time pH \ 4. Three of 6 subjects using
IR-OME at night demonstrated both abnormal total and Given the difficulty of healing severe erosive reflux
nocturnal acid exposure; with one subject demonstrating esophagitis, this study adds several important points to the
16.1 % time pH \ 4 when supine. literature: (1) once-daily IR-OME is highly effective in
healing LA grades C and D esophagitis; (2) our data do not
Safety and Tolerability support the hypothesis that nighttime administration of IR-
OME is superior in healing esophagitis compared to
Discontinuation resulting from adverse events occurred in morning IR-OME administration but do provide the basis
1 subject allocated to morning dosing (joint and muscle on which to calculate the sample size for a future equiva-
pain) and three subjects allocated to nighttime dosing lency study; (3) among subjects with esophagitis who
(peripheral edema and/or increased blood pressure). perceive heartburn and/or acid regurgitation at baseline,

123
Dig Dis Sci

Table 3 Endoscopic features of subjects ultimately diagnosed with Barrett’s esophagus

EGD at which BE Subject ID Baseline Extent of Baseline EGD Follow-up EGD
was diagnosed LA grade erosions (cm)
Prague criteria Prague criteria LSBE vs. SSBE
C M C M

Baseline 104 C 15 15 15 15 15 LSBE

a a
183 C 14 14 14 LSBE
163 C 11 9 11 7 10 LSBE
187 C 3 3 3 2 3 LSBE
190 C 2 0 2 0 2 SSBE
182 D 8 0 2 0 2 SSBE

Follow-up 185 D 4 6 6 LSBE

164 C ND 5 6 LSBE
131 C ND 4 5 LSBE
109 D 3 2 2 SSBE
103 D 1 2 2 SSBE
110 D 5 0 2 SSBE
149 D ND 0 2 SSBE
Rows 1–4 and 7–9 indicate morning (control group) and rows 5, 6 and 10–13 indicate nighttime (experimental group)
C circumferential extent of Barrett’s esophagus mucosa per the Prague Classification System, LSBE long segment Barrett’s esophagus,
M maximal extent of Barrett’s esophagus mucosa per the Prague Classification System, ND not described, SSBE short segment Barrett’s
esophagus
a
Lost to follow-up

Table 4 pH test results among subjects who permitted testing while on omeprazole/sodium bicarbonate once daily, either in the morning or at
nighttime
Time of dose Subject no. Esophagitis severity Abnormal post-treatment % time pH \ 4
pH test?
Pre- Post- Total Upright Supine

Morning omeprazole/sodium 1 C Healed No 3.3 0 5.7

bicarbonate users 3 D Healed No 1.9 3.3 0.6
5 C Healed No 0.8 1.1 0.5
9a C Healed No 0.3 b b

10 D Healed No 2 2.7 0
11 D Healed No 1 1.7 0
Mean % time pH \ 4 for 1.5 1.5 2.7
morning users
Nighttime omeprazole/sodium 2 D Healed No 1.9 3.1 0
bicarbonate users 4 C Healed No 0.8 1.6 0
6 C Healed Yes 6.6 0 16.1
7 C Healed No 0 0 0
8 D Healed Yes 10.8 15.7 2.9
12 C B Yes 11.5 11.3 12
Mean % time pH \ 4 for nighttime users 5.7 5.3 3.6
a
Wireless pH catheter used
b
Information not available

morning and nighttime administration of IR-OME are presence of severe esophagitis nor predict its resolution;
equally likely to resolve their symptoms; (4) the absence or and (5) there was no association between elevated BMI and
presence of esophageal symptoms may not predict the failure to heal esophagitis [11].

123
 Dig Dis Sci

The complete healing rate (76 %) found in this study and focus of GERD questionnaires or differences in
compares to 60–90 % of subjects with LA grade C or D severity of reflux esophagitis. The limited reliability of
erosive esophagitis depending on PPI type and dose in the self-reported GERD symptoms to predict the presence or
literature [2–5]. It has been hypothesized that lack of healing of esophagitis further reinforces the need for
efficacy of PPIs might be explained by poor control of endoscopic evaluation of chronic GERD patients before
nocturnal acid secretion. Pharmacologic therapies for and after medical therapy.
severe esophagitis have included twice-daily-administered Retrospective studies have estimated that 9–27 % of
PPIs, sometimes in combination with nighttime HRAs. In patients with severe esophagitis have concomitant BE [21–
this study, once-daily IR-OME achieved complete healing 24]. In this prospective study, 12 % LA C and 18 % LA D
in 76 % of subjects with severe erosive esophagitis, with subjects met stringent criteria for long- and short-segment
improvement in another 12 %. As a result, nearly all sub- BE (demonstrating intestinal metaplasia with goblet cells at
jects in this study had improvement in endoscopic histology). An additional 2 % (2/92) subjects had IMC. To
appearance. Notably, these favorable results were as robust limit the likelihood of over-diagnosing a condition with
in subjects with grade D esophagitis as those with grade C. elevated neoplastic risk, namely BE, it is our group’s
It is important to highlight that our team could not standardized clinical practice to not biopsy salmon-colored
control for the various types, dose, duration, or adherence mucosa unless it is at least 1 cm in length. Thus, we may
to pre-study PPI prior to their baseline EGD. Therefore, the have under-diagnosed the frequency of IMC and ultrashort
high effectiveness of once-daily dosing must be taken in BE in this cohort (that was endoscoped in a clinical
the context of the study. Specifically, upon enrollment, endoscopy suite by 1 of 58 endoscopists). This clinical
study participants were provided with structured education practice, however, strengthens the reliability of the BE
and counseling regarding how their PPI should be taken prevalence calculations. Unlike retrospective analyses, the
and had close follow-up while taking the study medication. prevalent BE values in this prospective study were not
This may help explain the high healing rates as well as the overinflated by including subjects with IMC who were
finding that those previously prescribed a non-omeprazole misclassified as having SSBE.
PPI healed at the same rate as those who were PPI naı̈ve. Most subjects simultaneously diagnosed with BE and
It is also noteworthy that this level of improvement was esophagitis during the baseline EGD had obvious long-seg-
achieved with both morning and nighttime administration ment BE and LA C erosions at the proximal aspect of the BE
of IR-OME. This was surprising, as it was hypothesized segment. Once the esophagitis healed, the measured BE
that better control of nocturnal gastric acid secretion with segment length generally remained the same (not longer). It
nighttime administration would be more effective in heal- did not appear that the erosions were hiding additional sal-
ing esophagitis. On the other hand, our 24-h pH data, mon-colored mucosa beneath them. Conversely, most of the
though limited, demonstrated at least equal, if not better, subjects in whom BE was only appreciated once the esoph-
control of both day and nighttime esophageal acid exposure agitis had healed initially had LA D esophagitis, in which the
with morning administration of IR-OME. These results erosions involved more than 75 % of the circumference of
could provide a valuable practical and theoretical economic the esophagus. The diagnosis of BE would have been missed
advantage to the patient by allowing the single use of IR- in 54 % of BE patients had a post-PPI research follow-up
OME in either the morning or nighttime to heal their severe EGD not been done, including three subjects with C5 cm
erosive esophagitis, without the need for twice-daily PPI segment lengths.
dosing or supplementation with an HRA. The efficacy of Some have informally postulated that the extent of a
once-daily administration may also lead to improved Barrett’s segment is due to the proximal extent of each
patient adherence, especially in patients with minimal patient’s individual esophagitis or refluxate. Proximal
symptoms. extent of reflux was not measured in this study, but, given
At the time of their diagnosis with severe esophagitis, the length of erosions at baseline, shown in Table 3, it
83 % of subjects reported perception of GERD symptoms, appears that this prospective study has demonstrated that
and when healed, 61 % reported symptom resolution. For the extent of a subject’s erosions does not strongly corre-
39 %, however, despite healing, their symptoms persisted. late with and often underestimates the BE length (measured
This observation is consistent with the concept that acid when the esophagitis has healed).
suppression medications alter the pH of the refluxate but This study supports the role for HH in advanced
they do not alter the mechanics of reflux. Our findings that esophageal inflammatory disease. HH is an independent
17 % reported no reflux symptoms despite severe esopha- risk factor for nocturnal reflux and erosive esophagitis [25,
gitis differs from those of others. Specifically, Lee et al. 26]. Therefore, it is not surprising that 81 % of non-BE and
[20] found that 43 % with erosive esophagitis did not 100 % of BE subjects with concomitant severe esophagitis
report symptoms. This difference may be due to quality in our study have a HH.

123
Dig Dis Sci

Several weaknesses to this study should be considered. On the other hand, this study has several strengths. Most
Firstly, our methods relied upon pill counting as a proxy for importantly, this was a prospective randomized trial
adherence to medical therapy. Thus, unrecognized medi- exclusively for LA grades C and D erosive esophagitis in a
cation adherence issues may explain the failure to achieve relatively large cohort. Endoscopists judging the primary
mucosal healing in some subjects. Conversely, adherence outcome of esophageal mucosal status (normal vs. esoph-
may have been better than real-world situations due to agitis) were not aware that the subject was enrolled in a
careful education of subjects and close follow-up. Other clinical trial; hence, they were blind, as were the statisti-
factors may also be involved in non-healing such as hy- cian and remainder of the investigative team. This large
permetabolizer CYP2C19 genotype status [27] or greater academic endoscopic group has previously demonstrated
degrees of reflux in some subjects. In this open-label study, near-perfect incorporation and use of the Los Angeles
subjects were not blind as to when they took their medi- grading system in the past [28]. This study is generalizable
cation. Limited resources to conduct the study made to patients found to have severe reflux esophagitis despite
employment of a double dummy design unfeasible. Also, being prescribed (or self-prescribed) non-omeprazole PPI
one of the benefits of participating in this study, from the therapy.
subject’s perspective, was the opportunity to avoid twice- In conclusion, our study demonstrates the high efficacy
daily dosing to heal their severe erosive esophagitis. Sec- of single dose IR-OME administered either in the morning
ondly, despite implementation of stratification for treat- or at night for healing or improving severe erosive
ment allocation, symptom frequencies were not equally esophagitis. It lays a foundation for consideration of
distributed between the two groups at baseline and, by equivalency studies comparing morning versus evening
chance, subjects allocated to morning dosing met criteria PPI treatment as well as comparison of once-daily versus
for overall GERD symptoms statistically more often than twice-daily PPI administration to heal severe erosive
did their counterparts allocated to nighttime dosing. Also, esophagitis. The aim of this study was to compare con-
there was an imbalance that occurred among the subjects ventional morning PPI dosing that requires a higher burden
who completed their follow-up questionnaires; this resulted of adherence to convenient nighttime dosing. Had the
in our reliance on the per-protocol analysis to assess the experimental arm of taking the PPI at nighttime proven
impact of dose timing on symptom response (Fig. 3a, b). superior, it could have potentially increased adherence to
Nine subjects allocated to nighttime IR-OME dosing PPI treatment and greatly improved our ability to heal
compared to three subjects allocated to morning dosing, did erosive reflux esophagitis. Unfortunately, that did not
not fully complete their follow-up questionnaires. The ITT work. At this time, adherence is still required for optimal
analysis assumption is that subjects missing follow-up benefit. Also, although symptom correlation to healing can
questionnaire information have severe symptoms, hence be demonstrated in a majority of subjects, in a sizable
the statistically significant findings shown in the ITT ana- number of subjects, symptom profiles do not reliably pre-
lysis (Fig. 4a, b) in the Appendix are deemed artifactual dict esophageal healing with therapy. Therefore, clinicians
and not clinically relevant. Figures 5a, b in the Appendix should continue evaluating these patients by objective
show the sensitivity analysis, in which missing data are means, specifically endoscopy, to ensure endoscopic heal-
assumed to reflect complete lack of GERD symptoms and ing and to allow for accurate assessment for the presence of
healing of esophageal injury (i.e. the best case scenario). BE after pharmacologic therapy; especially in patients with
Because more patients in the nighttime dosing group were LA D esophagitis.
missing follow-up data, the sensitivity analyses produced
statistically significant results only for the worst case sce- Acknowledgments We thank Lori R. Anderson for help typing and
submitting the manuscript; Kaiser Lim, M.D., Prasad Iyer, M.D., Adil
nario ITT results. Thirdly, now that PPIs are available A. Abdalla, M.B.B.S. and Judith McElhiney, M.D., for help recruiting
over-the-counter, the frequency that outpatients present de subjects; Rayna M. Grothe, M.D., Joanna M. Peloquin, M.D., Sha-
novo with LA C and D esophagitis is probably low. In this bana F. Pasha, M.D., and Darlene E. Graner, CCC-SLP, for help
study, although 387 consecutive patients were found to conceptualizing the study design; and Michael D. Van Norstrand,
M.D., Ph.D., for editorial assistance. The project was supported in
have LA C or D esophagitis over a 23-month period of part by the Miles and Shirley Fiterman Center for Digestive Diseases
time, the majority were inpatients with at least one exclu- at Mayo Clinic, Rochester, Minnesota. Yvonne Romero, M.D., was
sion criteria. The subjects in this study were medically supported in part by a grant from the National Institutes of Health
stable outpatients. In order to attain 80 % power to detect (NIDDK 02956) and the Robert Wood Johnson Foundation Harold
Amos Medical Faculty Development Program.
the difference we observed for esophageal healing between
morning and nighttime dosing, a total of 610 subjects Conflict of interest This study was funded in part by an Investi-
would be needed. Finally, the results of this project are not gator-Initiated Research Award from Santarus, Inc., IRB #
generalizable to patients using an omeprazole product at 07-008503. Santarus Inc. provided study medications [omeprazole/
sodium bicarbonate (Zegerid) and GelusilTM] at no charge. No
the time of their diagnosis with severe esophagitis.

123
 Dig Dis Sci

assistance for manuscript preparation was received from Santarus, • Assess the frequency with which a diagnosis of
Inc. Unpaid academicians voluntarily wrote this paper. All data Barrett’s esophagus was made once the over-
analysis was undertaken by Felicity T. Enders, Diana M. Orbelo,
Nancy N. Diehl, Debra M. Geno, Yvonne Romero, David A. Katzka, lying severe esophagitis had healed.
Dawn L. Francis, Michael D. Crowell, Sami R. Achem, Ramona
• A priori exploratory and hypothesis-generating
DeJesus, Vikneswaran Namasivayam, Steven C. Adamson, Amindra
S. Arora, Andrew J. Majka, Jeffrey A. Alexander, Joseph A. Murray, objective: assess objective control of esophageal
Kee Wook Jung, Margaret S. Houston and Angela O’Neil, all of acid exposure as determined by ambulatory pH
whom are or were Mayo Clinic employees at the time. Santarus Inc. monitoring.
provided salary support to FTE, NND, and DMG for data analysis,
and DMG and MF for data collection. The Santarus study sponsors, 2. IR-OME powder instruction details
William F. Bleker, M.D., Philip Yeung, Pharm.D., Gregg Checani,
M.D., and E. David Ballard, M.D., participated in study design but Omeprazole/sodium bicarbonate powder for oral sus-
did not participate in data collection, analysis or interpretation, and pension 40 mg was supplied in individual packets that are
did not participate in drafting the manuscript or in the decision to
emptied into a small cup containing 15–30 ml (1–2
publish the results of this trial. Study sponsors were permitted to
review the manuscript prior to its submission for publication. Yvonne tablespoons) of water, one per day, for 8 weeks. They
Romero has received research funding from Aptalis, AstraZeneca, were asked to stir well and drink immediately then refill
Meritage Pharma and Takeda, has served as an ad hoc consultant for the cup with water and drink. Subjects assigned to
Kala, and receives royalties from the licensed use of the Mayo
morning dosing were instructed to take the medication on
Dysphagia Questionnaire—30 day. Dawn L. Francis has received
research funding from AstraZeneca, and receives royalties from the an empty stomach, immediately upon rising, 20–60 min
licensed use of the Mayo Dysphagia Questionnaire—30 day. Jeffrey prior to chewing a solid. Subjects assigned to nighttime
A. Alexander has received research funding from Aptalis and Merck, dosing were instructed to keep the medication by their
serves as a consultant and has stock holdings of Meritage Pharma and
bedside; taking the medication in a standing or seated
receives royalties from the licensed use of the Mayo Dysphagia
Questionnaire—30 day. Joseph A. Murray has an ownership interest upright position immediately before turning off the lights
with Torax Medical Inc. and Vysera Biomedical, and receives roy- with the intention to sleep. The subject was instructed to
alties from the licensed use of the Mayo Dysphagia Questionnaire— not use other liquids or foods for 20 min after taking their
30 day. Felicity T. Enders, Michael D. Crowell, Debra M. Geno,
study medication for those allocated to morning dosing,
Matthew Lohse, Nancy N. Diehl, and Mary Fredericksen receive
royalties from the licensed use of the Mayo Dysphagia Question- and until the next morning for those allocated to night-
naire—30 day. time dosing.
3. Endoscopic outcomes
LA classification system [1]
Appendix
• LA grade A: one or more erosions on non-confluent
1. Hypotheses and aims folds in which the longest erosion is \5 mm length
• LA grade B: one or more erosions on non-confluent
• A priori hypotheses: folds in which the longest erosion is C5 mm length
• The timing of administration of omeprazole/ • LA grade C: confluent erosions of more than one fold
sodium bicarbonate will impact its efficacy in involving \75 % of the circumference.
healing esophagitis • LA grade D: confluent erosions involving C75 % of the
• Nighttime administration (experimental arm) circumference of the distal esophagus.
will be superior in healing esophagitis com- The endoscopic evaluation results were grouped for
pared to morning administration (control arm) outcomes analysis as follows, for subjects with:
prior to a meal.
• LA C esophagitis at baseline:
• A priori primary aim: quantify the efficacy of
omeprazole/sodium bicarbonate in healing severe • Healed = no erosions in the esophagus
erosive esophagitis when used either as a morning • Improved = LA grades A or B
or before bedtime dose. • Same or worse = C or D at follow-up
• A priori secondary aims:
• LA D esophagitis at baseline:
• Determine the percent of patients who improve,
• Healed = no erosions in the esophagus
but did not resolve, their mucosal damage.
• Improved = LA grades A, B or C
• Determine the efficacy of omeprazole/sodium
• Same = LA grade D at follow-up
bicarbonate on symptom resolution in patients
with erosive esophagitis 4. MDQ-30 scoring details

123
Dig Dis Sci

• Symptom analysis: GERD symptoms (absent, the breast bone in the chest), at least once per
infrequent, frequent) at baseline and follow-up week, with at least one of the following:
were measured by the MDQ-30. heartburn that improves with antacids, awakens
them at night, or radiates to the neck. Respon-
• Frequent: The respondent reports experiencing
dents would also be categorized as having
heartburn, (a burning pain or discomfort behind
frequent GERD symptoms if they report expe-
riencing acid regurgitation, (a bitter or sour-
tasting fluid coming up from the stomach into
Table 5 Subject demographics at baseline, per-protocol analysis
the mouth or throat), at least once per week.
Factor Morning Nighttime p value • Absent: The respondent is completely asymp-
dosing dosing
tomatic; denies experiencing heartburn or acid
n = 40 n = 40
regurgitation.
Male sex [n (%)] 28 (70) 28 (70) 1.00 • Infrequent: Any heartburn or acid regurgitation
Age [years median (range)] 59 (19–86) 61 (30–86) 0.61 in those who do not meet criteria for frequent
BMI [kg/m2 median 29.9 29.9 0.58 symptoms.
(range)] (24.5–51.1) (21.2–45.6)
Hiatal hernia (HH) • Symptom outcomes at follow-up include:
Overall [n (%)] 34 (85) 33 (83) 1.00 • Complete resolution (subjects with either fre-
Of those with HH [cm 4.0 (1–13) 3.3 (1–6) 0.27 quent or infrequent GERD symptoms at base-
average (range)]
line report the absence of GERD symptoms at
Esophagitis severity [n (%)] follow-up)
LA grade C 24 (60) 26 (65) 0.82 • Partial resolution (subjects with frequent
LA grade D 16 (40) 14 (35) GERD symptoms at baseline meet criteria for
Using a PPI at time of baseline EGD [n (%)] infrequent GERD symptoms at follow-up)
Overall 12 (30) 10 (25) 0.80 • No improvement (subjects with either frequent
Among asymptomatic 3 of 9 (33) 2 of 5 (40) 1.00 or infrequent GERD symptoms at baseline
subjects
continue to report those symptoms at follow-
Among symptomatic 9 of 31 (26) 8 of 35 (29) 0.59
up)
GERD subjects
Using a H2 blocker and or OTC antacid at time of baseline EGD 5. Per-protocol demographic results (Table 5)
[n (%)] 6. Table 6 is an ITT (worst case) version of Table 2. Table
Overall 29 (73) 29 (73) 1.00 6 was not used in the body of the manuscript because
p values from Fisher’s exact test for categorical variables and the there was imbalance in the subjects who com-
Wilcoxon rank sum test for continuous variables pleted their questionnaires. Nine subjects allocated to

Table 6 Frequency of heartburn, acid regurgitation and overall GERD symptoms in subjects allocated to morning (n = 43) versus nighttime
(n = 49) dosing of omeprazole/sodium bicarbonate at entry and following 8 weeks therapy, intention-to-treat analysis (worst case)
Symptom Time of study drug administration Symptom frequency pc
At baseline After 8 weeks therapy
Absent Infrequent Freq. pa Absent Infrequent Freq. pb
n (%) n (%) n (%) n (%) n (%) n (%)

Heartburn Morning dose 11 (26) 6 (14) 26 (60) 0.26 31 (72) 7 (16) 5 (12) 0.09 0.03
Nighttime dose 10 (20) 14 (29) 25 (51) 26 (53) 8 (16) 15 (31)
Acid regurgitation Morning dose 21 (49) 11 (26) 11 (26) 0.19 33 (77) 4 (9) 6 (14) 0.21 0.06
Nighttime dose 20 (41) 21 (43) 8 (16) 29 (59) 7 (14) 13 (27)
GERD Morning dose 9 (21) 5 (12) 29 (67) 0.02 29 (67) 7 (16) 7 (16) 0.15 0.04
Nighttime dose 7 (14) 18 (37) 24 (49) 24 (49) 9 (18) 16 (33)
a
p values from Fisher’s exact test comparing symptom frequency (absent, infrequent or frequent) for patients with morning and nighttime doses
b
p values from Fisher’s exact test comparing symptom frequency (absent, infrequent or frequent) for patients with morning and nighttime doses.
For this ITT analysis, missing values were presumed to have frequent symptoms
c
Using ordinal logistic regression, this p value assesses whether time of dose is associated with symptom frequency at eight weeks, after
adjusting for baseline symptom frequency. For this ITT analysis, missing values at follow-up were presumed to have frequent symptoms

123
 Dig Dis Sci

Table 7 Frequency of heartburn, acid regurgitation and overall GERD symptoms in subjects allocated to morning (n = 43) versus nighttime
(n = 49) dosing of omeprazole/sodium bicarbonate at entry and following 8 weeks therapy, Intention-To-Treat analysis (best case)
Symptom Time of study drug administration Symptom frequency pc
At baseline After 8 weeks therapy
a
Absent Infrequent Freq. p Absent Infrequent Freq. pb
n (%) n (%) n (%) n (%) n (%) n (%)

Heartburn Morning dose 11 (26) 6 (14) 26 (60) 0.26 34 (79) 7 (16) 2 (5) 0.46 0.25
Nighttime dose 10 (20) 14 (29) 25 (51) 35 (79) 8 (16) 6 (12)
Acid regurgitation Morning dose 21 (49) 11 (26) 11 (26) 0.19 36 (84) 4 (9) 3 (7) 0.79 0.47
Nighttime dose 20 (41) 21 (43) 8 (16) 38 (78) 7 (14) 4 (8)
GERD Morning dose 9 (21) 5 (12) 29 (67) 0.02 32 (74) 7 (16) 4 (9) 0.76 023
Nighttime dose 7 (14) 18 (37) 24 (49) 33 (67) 9 (18) 7 (14)
a
p values from Fisher’s exact test comparing symptom frequency (absent, infrequent or frequent) for patients with morning and nighttime doses
b
p values from Fisher’s exact test comparing symptom frequency (absent, infrequent or frequent) for patients with morning and nighttime doses.
For this ITT analysis, missing values were presumed to have absent symptoms
c
Using ordinal logistic regression, this p value assesses whether time of dose is associated with symptom frequency at 8 weeks, after adjusting
for baseline symptom frequency. For this ITT analysis, missing values at follow-up were presumed to have absent symptoms

Table 8 Endoscopic features of subjects ultimately diagnosed with Barrett’s esophagus showing raw symptom, sex, age and Body Mass Index
(BMI) data
EGD at which BE was Study ID Baseline Extent of Baseline Follow-up Baseline
diagnosed LA grade erosions (cm) EGD EGD
Prague Prague LSBE vs. GERD Sex Age BMI HH
criteria criteria SSBE symptoms (cm)
C M C M

Baseline 104 C 15 15 15 15 15 LSBE Frequent F 57 29.0 4.5

a a
183 C 14 14 14 LSBE Frequent M 39 24.6 6.0
163 C 11 9 11 7 10 LSBE Infrequent M 53 43.6 3.0
187 C 3 3 3 2 3 LSBE Frequent M 73 29.8 4.0
190 C 2 0 2 0 2 SSBE Frequent M 75 28.9 2.0
182 D 8 0 2 0 2 SSBE Frequent M 65 29.4 2.0

Follow-up 185 D 4 6 6 LSBE Absent M 30 51.0 13.0

164 C ND 5 6 LSBE Frequent M 71 26.3 4.0
131 C ND 4 5 LSBE Frequent F 70 28.4 6.0
109 D 3 2 2 SSBE Frequent F 58 32.3 6.0
103 D 1 2 2 SSBE Frequent F 51 29.0 5.0
110 D 5 0 2 SSBE Absent F 77 28.8 6.0
149 D ND 0 2 SSBE Infrequent F 43 45.6 3.0
Rows 1–4 and 7–9 indicate morning (control group) and rows 5, 6 and 10–13 indicate nighttime (experimental group)
HH hiatal hernia length in cm at baseline endoscopy, C circumferential extent of Barrett’s esophagus mucosa per the Prague Classification
System, GERD sx gastroesophageal Reflux Disease symptoms, LSBE long segment Barrett’s esophagus, M maximal extent of Barrett’s
esophagus mucosa per the Prague Classification System, ND not described, SSBE short segment Barrett’s esophagus
a
Lost to follow-up

nighttime IR-OME dosing, compared to 3 subjects assume subjects without follow-up questionnaire infor-
allocated to morning dosing did not fully complete their mation have severe symptoms at follow-up. Thus, the
follow-up questionnaires. In the ITT analysis, we statistically significant findings shown in Table 6 are

123
Dig Dis Sci

(a) Heartburn Intention-To-Treat (worst case)

Morning vs Nighttime: Before and After Treatment with Omeprazole/Sodium Bicarbonate

Morning Dose Nighttime Dose

Frequent

Infrequent

Absent

Before After Before After

Subjects had complete questionnaires

Key Subjects missing follow-up questionnaires

(b)
Acid Regurgitation Intention-To-Treat (worst case)
Morning vs Nighttime: Before and After Treatment with Omeprazole/Sodium Bicarbonate

Morning Dose Nighttime Dose

Frequent

Infrequent

Absent

Before After Before After

Subjects had complete questionnaires

Key Subjects missing follow-up questionnaires

Fig. 4 a Frequency of heartburn before and after 8 weeks of omeprazole/sodium bicarbonate therapy comparing morning versus nighttime
dosing. Intention-to-treat (worst case). b Frequency of acid regurgitation before and after 8 weeks of omeprazole/sodium bicarbonate therapy
comparing morning versus nighttime dosing. Intention-to-treat (worst case)

artifactual. Thus, we opted to show the Per-protocol questionnaire information are symptom free at 8
results in the manuscript (see Table 2). weeks.
7. Table 7 is the ITT (best case) version of Table 2. 8. Detailed demographic information of 13 subjects
It shows the results for ITT best case scenario ultimately diagnosed with Barrett’s esophagus
in which we assume subjects without follow-up (Table 8).

123
 Dig Dis Sci

(a) Heartburn Intention-To-Treat (best case)

Morning vs Nighttime: Before and After Treatment with Omeprazole/Sodium Bicarbonate
Morning Dose Nighttime Dose

Frequent

Infrequent

Absent

Before After Before After

Subjects had complete questionnaires

Key Subjects missing follow-up questionnaires

(b) Acid Regurgitation Intention-To-Treat (best case)

Morning vs Nighttime: Before and After Treatment with Omeprazole/Sodium Bicarbonate

Morning Dose Nighttime Dose

Frequent

Infrequent

Absent

Before After Before After

Subjects had complete questionnaires

Key Subjects missing follow-up questionnaires

Fig. 5 a Frequency of heartburn before and after 8 weeks of omeprazole/sodium bicarbonate therapy comparing morning versus nighttime
dosing. Intention-to-treat (best case). b Frequency of acid regurgitation before and after 8 weeks of omeprazole/sodium bicarbonate therapy
comparing morning versus nighttime dosing. Intention-to-treat (best case)

References 3. 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 Gastro-
1. Lundell LR, Dent J, Bennett JR, et al. Endoscopic assessment of
enterol. 2001;96:656–665.
oesophagitis: clinical and functional correlates and further vali-
4. Labenz J, Armstrong D, Lauritsen K, et al. A randomized com-
dation of the Los Angeles classification. Gut. 1999;45:172–180.
parative study of esomeprazole 40 mg versus pantoprazole 40 mg
2. Chiba N, De Gara CJ, Wilkinson JM, Hunt RH. Speed of healing
for healing erosive oesophagitis: the expo study. Aliment Phar-
and symptom relief in grade II to IV gastroesophageal reflux
macol Ther. 2005;21:739–746.
disease: a meta-analysis. Gastroenterology. 1997;112:1798–1810.

123
Dig Dis Sci

5. Kinoshita Y, Hongo M. Efficacy of twice-daily rabeprazole for 16. Castell D. Review of immediate-release omeprazole for the
reflux esophagitis patients refractory to standard once-daily treatment of gastric acid-related disorders. Expert Opin Phar-
administration of PPI: the Japan-based TWICE Study. Am J macother. 2005;6:2501–2510.
Gastroenterol. 2012;107:522–530. 17. Moher D, Hopewell S, Schulz KF, et al. Consort 2010 explana-
6. Adachi K, Fujishiro H, Katsube T, et al. Predominant nocturnal tion and elaboration: updated guidelines for reporting parallel
acid reflux in patients with los angeles grade c and d reflux group randomised trials. J Clin Epidemiol. 2010;63:e1–37.
esophagitis. J Gastroenterol Hepatol. 2001;16:1191–1196. 18. McElhiney J, Lohse MR, Arora AS, et al. The mayo dysphagia
7. Campos GR, Peters JH, DeMeester TR, Öberg S, Crookes PF, questionnaire-30: documentation of reliability and validity of a
Mason RJ. The pattern of esophageal acid exposure in gastro- tool for interventional trials in adults with esophageal disease.
esophageal reflux disease influences the severity of the disease. Dysphagia. 2010;25:221–230.
Arch Surg. 1999;134:882–888. 19. Sharma P, Dent J, Armstrong D, et al. The development and vali-
8. Orr WC, Allen ML, Robinson M. The pattern of nocturnal and dation of an endoscopic grading system for Barrett’s esophagus: the
diurnal esophageal acid exposure in the pathogenesis of erosive Prague C & M criteria. Gastroenterology. 2006;131:1392–1399.
mucosal damage. Am J Gastroenterol. 1994;89:509–512. 20. Lee D, Lee KJ, Kim KM, Lim SK. Prevalence of asymptomatic
9. Frazzoni M, De Micheli E, Savarino V. Different patterns of erosive esophagitis and factors associated with symptom presentation
oesophageal acid exposure distinguish complicated reflux disease of erosive esophagitis. Scand J Gastroenterol. 2013;48:906–912.
from either erosive reflux oesophagitis or non-erosive reflux 21. Hanna S, Rastogi A, Weston AP, et al. Detection of Barrett’s
disease. Aliment Pharmacol Ther. 2003;18:1091–1098. esophagus after endoscopic healing of erosive esophagitis. Am J
10. Bredenoord AJ. Mechanisms of reflux perception in gastro- Gastroenterol. 2006;101:1416–1420.
esophageal reflux disease: a review. Am J Gastroenterol. 2012; 22. Modiano N, Gerson LB. Risk factors for the detection of Barrett’s
107:8–15. esophagus in patients with erosive esophagitis. Gastrointest En-
11. Peura DA, Pilmer B, Hunt B, Mody R, Perez MC. The effects of dosc. 2009;69:1014–1020.
increasing body mass index on heartburn severity, frequency and 23. Rodriguez S, Mattek N, Lieberman D, Fennerty B, Eisen G.
response to treatment with dexlansoprazole or lansoprazole. Ali- Barrett’s esophagus on repeat endoscopy: should we look more
ment Pharmacol Ther. 2013;37:810–818. than once? Am J Gastroenterol. 2008;103:1892–1897.
12. Peghini PL, Katz PO, Castell DO. Ranitidine controls nocturnal 24. Gilani N, Gerkin RD, Ramirez FC, Hakim S, Randolph AC. Prev-
gastric acid breakthrough on omeprazole: a controlled study in alence of Barrett’s esophagus in patients with moderate to severe
normal subjects. Gastroenterology. 1998;115:1335–1339. erosive esophagitis. World J Gastroenterol. 2008;14:3518–3522.
13. Cicala M, Emerenziani S, Guarino MP, Ribolsi M. Proton pump 25. Gaddam S, Maddur H, Wani S, et al. Risk factors for nocturnal reflux
inhibitor resistance, the real challenge in gastro-esophageal reflux in a large GERD cohort. J Clin Gastroenterol. 2011;45:764–768.
disease. World J Gastroenterol. 2013;19:6529–6535. 26. Chiba H, Gunji T, Sato H, et al. A cross-sectional study on the
14. Castell D, Bagin R, Goldlust B, Major J, Hepburn B. Com- risk factors for erosive esophagitis in young adults. Intern Med.
parison of the effects of immediate-release omeprazole powder 2012;51:1293–1299.
for oral suspension and pantoprazole delayed-release tablets on 27. Egan LJ, Myhre GM, Mays DC, Dierkhising RA, Kammer PP,
nocturnal acid breakthrough in patients with symptomatic Murray JA. Cyp2c19 pharmacogenetics in the clinical use of
gastro-oesophageal reflux disease. Aliment Pharmacol Ther. proton-pump inhibitors for gastro-oesophageal reflux disease:
2005;21:1467–1474. variant alleles predict gastric acid suppression, but not oesopha-
15. Katz PO, Koch FK, Ballard ED, et al. Comparison of the effects geal acid exposure or reflux symptoms. Aliment Pharmacol Ther.
of immediate-release omeprazole oral suspension, delayed- 2003;17:1521–1528.
release lansoprazole capsules and delayed-release esomeprazole 28. Abdalla AA, Petersen BT, Ott BJ, et al. Impact of feedback and
capsules on nocturnal gastric acidity after bedtime dosing in didactic sessions on the reporting behavior of upper endoscopic
patients with night-time gerd symptoms. Aliment Pharmacol findings by physicians and nurses. Clin Gastroenterol Hepatol.
Ther. 2007;25:197–205. 2007;5:326–330.

123
Aliment Pharmacol Ther 1993: 7: 385-392.

Relapse rate of patients after healing of oesophagitis-a prospective

study of alginate as self-care treatment for 6 months
T. POYNARD and a French Co-operative Study Group
Service &Hipato-Gastroentirologie,H6pital Antoine Bicli.re, Clamart F92 141 Cedex, France
Accepted for publication 16 April 1 9 9 3

were seen every 2 months: they filled out diary-cards

SUMMARY
describing symptoms and drug consumption. Data
In order to estimate the rate of symptomatic relapse were available for 883 patients during follow up. The
and to identify factors associated with this relapse, cumulative percentage of patients without relapse at 6
1030 patients were included in a multicentre study. months estimated by Kaplan Meier method was
All patients included had clinically and endoscopically 76 f2 % (mean & S.E.M.). Among these patients
healed oesophagitis (grade I (erosion, 57%).grade 11 without relapse, 9 5 % took less than two sachets per
(confluent erosion, 3 3 %) or grade I11 (circular erosion, day. Only two factors were identified as associated with
9 %)) after treatment with an H,-blocker or omeprazole. relapse-the grade of oesophagitis ( P < 0.007), and
Patients were given conservative advice and were told the occurrence of stress during follow-up ( P < 0.05).
to take sodium alginate only in response to pain. The percentage of patients free from relapse at six
Relapse was defined as pain justifying more than eight months was 85 f 2 % in patients with grade I
sodium alginate sachets over 48 hours or treatment oesophagitis, 69 k 3 % in patients with grade 11, and
with another drug, or the need for a new endoscopy. 56 6% in patients with grade 111 oesophagitis ( P <
Forty risk factors were recorded initially and patients 0.00011.

The aim of this study was to assess prospectively the re-

INTRODUCTION
lapse rate of patients after healing of oesophagitis, whilst
Very few prospective studies have described the fate of only receiving symptomatic treatment with sodium
patients after healing oe~ophagitis.l-~ The common forms alginate. This strategy was considered a failure when the
of oesophagitisare not severe and are mostly grade I (non- dose of alginate necessary to relieve symptoms was
confluent erosion). It is not known if these patients after greater than 8 sachets in 48 h, or when it was necessary
healing need costly long-term treatment with an H,- to give a drug other than alginate, or if a new endoscopy
blocker or proton pump inhibitor. The simplest approach was required. These 3 situations defined a relapse. The
is for the patient to take an inexpensive treatment only in severity of the disease during follow-up was also assessed
response to symptoms (self-caretreatment). by the number of alginate sachets consumed.
Alginate, which floats on the gastric content and
prevents contact of irritant refluxed material with the METHODS
oesophageal mucosa, has been shown to be an effective
treatment for symptomatic gastro-oesophageal refl~x.~-’
Patients
It was chosen for this study because it is widely used in
France, and is 7 times less expensive than the H,-blockers Patients were aged 18-75 years. They had experienced
and 14 times less expensive than omeprazole. symptoms of reflux oesophagitis and had healed grade I
Correspondence to: Professor 2’. Poynard. Service d’Hipato-Gastro-
(erosion), grade I1 (confluent erosion) or grade 111
entirologie. Hipita1 Antoine BicPre, Clamart F92141 Cedex, France. (circular erosion) oesophagitis, verified by endoscopy

385
27 BAP 7
386 T. P O Y N A R D & FRENCH STUDY G R O U P

before and after treatment using an H,-blocker or with a potential effect on gastro-oesophageal reflux. A list
omeprazole. Patients must have been asymptomatic after of 116 forbidden drugs were given to each gastro-
the initial treatment and have no erosions at endoscopy enterologist. All patients were given advice about con-
to be included in this study. servative measures : postural treatment by raising the
They were recruited from 2 3 5 French gastro- head of the bed, reduction in volume of meals, limiting fat
enterologists who were asked to include patients con- consumption, reduction of excess weight, elimination or
secutively. A large number of gastroenterologists had major reduction in the consumption of alcohol and
been chosen in order to be representative of the popu- smoking. Patients were seen by the gastroenterologist
lation of prescribers. every 2 months, returning diary cards describing
All patients were informed about the aims of the study, symptoms and drugs consumption.
as well as the benefits and risks associated with the
treatment. Only those who gave their written informed
Clinical assessment
consent to participate were enrolled. This protocol was
approved by the Ethical Committee of the Faculte de During the 6 months of follow-up, relapse was defined as
Medecine-Paris Sud. symptoms not relieved by 8 alginate sachets in 2
The following exclusion criteria applied : pregnancy and consecutive days, or the necessity to prescribe another
lactation: grade 0 or grade IV oesophagitis: oesophageal drug for gastro-oesophageal symptoms, or the necessity
or gastric malignancy, gastric or duodenal ulcer: pre- to perform another endoscopy.
vious surgery of the upper digestive tract: chemical or Before entering the study, each patient received a
biochemical evidenceof significant renal, hepatic, cardiac physical examination. Before inclusion (pre-entry) and at
or other severe diseases impairing follow-up or one-year each subsequent visit ( 2 , 4 and 6 months), patients were
survival: sclerodermia, oesophageal varices. disease re- questioned in a standardized way about the type,
quiring treatment by aspirin, non-steroidal anti- frequency and severity of symptoms which they had
inflammatory drugs (NSAIDs), or potassium tablets : any experienced. For heartburn (pyrosis),the number of days
condition associated with poor compliance, concomitant without pyrosis was recorded, together with the number
treatment with salt-free diet: oesophagitis with initial of days with pyrosis not interfering with normal life, the
treatment shorter than 2 weeks or greater than 3 number of days with pyrosis interfering with normal life,
months. the number of nights with pyrosis episodes without
insomnia, and the number of nights with pyrosis episodes
leading to insomnia. The presence of the following
symptoms were recorded : postural factors, epigastric
Study design
pain, other abdominal pain, dysphagia, regurgitations,
After inclusion patients were instructed not to take belching, nausea, vomiting, dyspepsia, abdominal dis-
alginate, or any drug for reflux, if they were free of reflux tension, diarrhoea, constipations and bad taste.
symptoms or epigastric pain. If they had reflux or To assess all the digestive symptoms an analogue scale
epigastric pain they were allowed to take a minimal of the ‘digestive state’ was scored by the patient from 0
number of alginate sachets, from 1to 4 sachets per day, (terrible) to 100 (perfect).
in order to relieve symptoms. If the symptoms persisted The quantity of alginate sachets consumed during each
despite 8 sachets during 2 consecutive days, they must period of 2 months was considered as a pragmatic
contact the gastroenterologist for another treatment or estimate of the disease severity in patients during follow-
an endoscopy, and were considered as having relapsed. up. This estimation has been used to describe the pattern
After recruitment to the study, and after each visit at 2 of the disease.
and 4 months, patients received a box with 240 sachets The following potential risk factors were also recorded :
of alginate. Each sachet included 500mg sodium age, sex, weight, ethnic origin, size of residence, pro-
alginate, 267 mg sodium bicarbonate and 160 mg cal- fession, unemployment, tobacco, alcohol, aspirin, NSAID
cium carbonate (liquid Gaviscon. SmithKline Beecham, consumption, stress, vacation, fat-rich diet, spicy diet,
Paris, France). A six-month period of follow-up with this duration of pyrosis, history of gastro-oesophageal reflux,
symptomatic treatment was scheduled. During this duration of the last episode before inclusion, type of
period patients were not allowed to take any other drug treatment (H,-blocker or omeprazole) used for

Aliment Pharmacol Ther 1993, vol. 7.

Table 1. Clinical characteristics of patients recruited

Quantitative characteristics

Age (years)
History of reflux (years)
Number of
patients with
data available Mean S.D.
9 74
964
+
51 f 1 4
5k6 50%c
r
t
t
-
ALGINATE AS SELF-CARE TREATMENT FOR OESOPHAGITIS 387

Duration of last episode 955 45 f63

(days)
Duration of last treatment 958 45 22
(days)
Previous number of endoscopies
Previous number of X-rays
980
936
1.3 1.2
0.6+0.9
0
t
I
2
I I

4 6
I

% Time (months)
Males 9 74 59 Number of subjects at risk 883 778 643 445
Smokers 990 29 Number of subjects with relapse 0 51 111 180
Drinkers 985 43 (cumulative)
Spicy diet 980 21 Truncated data 0 54 129 258
Fat diet 9 78 32 Total 883 883 883 883
Retired 975 27
Figure 1. Percentage of subjects free of relapse during
City > 100,000 inhabitants 957 44
symptomatic treatment with sodium alginate.
North African origin 934 4
Type of last treatment :
H,-blocker 1030 88
Omeprazole 1030 12 the Kaplan-Meier method and curves compared by the
Prokinetic 1030 66 log-rank test. Multidimensional analysis used regression
Antacid 1030 59 analysis and the Cox model.
Grade of oesophagitis*
1 non-confluent erosion 9 79 57.4
2 confluent erosion 979 33.2 RESULTS
3 circulary erosion 979 9.4
A total of 1030 subjects were recruited initially. Clinical
* For 51 patients, it was stated by the endoscopist ‘grade 1-111’
characteristics are given in Table 1. For 147 patients no
without details. Therefore, these patients were excluded from the
analysis according to the grade of oesophagitis. This concerns more data were available after their inclusion and they
the endoscopic evaluation before the initial treatment. All were excluded from the analysis. For 883 patients, data
patients were treated before starting the study. were available during the follow-up2 58 subjects were
not followed for 6 months: 21 subjects refused to follow
the protocol, 59 stopped the treatment for reasons not
related to efficacy, and 178 were lost to follow-up. These
oesophagitis healing before inclusion, number of pre- 258 subjects were included and analysed as truncated
vious upper endoscopies, number of previous data.
oesophageal X-rays, grade of initial oesophagitis healed A total of 180 subjects were defined as having relapsed.
by treatment before inclusion. All of these 180 subjects were treated by another drug
than alginate, 129 needed a new endoscopy which
demonstrated erosions in 54 subjects. The cumulative
percentage of patients free from relapse at 6 months was
Statistical analysis
76 & 2 % (Figure 1).
For comparison between groups the Student’s t-test or For 482 patients, the distribution of alginate sachet
non-parametric Mann-Whitney U-test were used when consumption according to the period of follow-up was
appropriate for quantitative variables, and either x2 or available and is given in Table 2. For 104 patients (22 %),
Fisher exact test for qualitative variables. Relapse was the consumption of alginate sachet was very low-that
considered as a time-dependent end-point. Occurrence of is, less than 1 sachet per week: 95 % of subjects took less
relapse during the 6-month follow-up was assessed by than two sachets per day and only 5 % took between 2 to

Ahtent Pharmacol Ther 1993. vol. 7.

27-2
388 T. POYNARD & FRENCH STUDY GROUP

Table 2. Pattern of alginate sachet consumption during follow-up

Mean alginate sachet consumption during 2 months

( L- S.D.)

n % 1-2nd month 3 4 t h month 5-6th month

Category of alginate consumer

Nothing (0 sachet) during 6 months 14 2.9 0 0 0
Very few ( < 1 sachet/week) during 90 18.7 3f 2 2f2 3L-2
6 months
Very few (< 1 sachet/week) during first 4 0.8 3L-3 3f1 29+33
4 months
Very few (< 1 sachet/week) during last 56 11.6 36f38 2+2 2f2
4 months
Very few ( < 1 sachet/week) during 6rst 16 3.3 3f 2 62 L- 82 64f 76
2 months
Very few ( < 1 sachet/week) during 3rd 16 3.3 33+29 4f3 25+15
and 4th month
Very few ( < 1 sachet/week) during last 43 8.9 61+55 31+54 2+2
2 months
Few ( < 1 sachet/day) during 6 months 211 43.8 81 62 + +
61 50 51 f 52
High (2-4 sachets/day) during 6 months 24 5.0 180 f90 151L-86 142 f 82
Mean sachet consumption 482 100.0 48 + 57 35f50 26+ 10

Table 3. Intensity of pyrosis and digestive score during follow-up

Period

1st-2nd months 3rd-4th months 5th-6th months

n mean*S.D. n meanfS.D. n mean 5 S.D.
~ ~~

Days with pyrosis not interfering with activity 682 13L-14 626 8f10 590 7fll
Days with pyrosis interfering with activity 685 5+11 625 2L-6 592 lf6
Nights with pyrosis but without insomnia 686 4f9 621 2L-6 591 7f5
Nights with pyrosis and insomnia 688 1+5 619 0.5+3 588 0.6L-4
Digestive Score* (0-100) 703 59f16 620 65+15 597 68f15

* The digestive score (mean+S.D.) is from 0 = terrible to 100 = perfect.

Unidimensional Multidimensional Table 4. Factors associated with relapse

analysis analysis (Cox model)

Factors identified: Log rank P-value Coefflcient P-value

Grade of oesophagitis 46 0.0001 0.44 0.007
Initial treatment by 4.6 0.03 0.40 0.12
omeprazole
Stress during follow-up 3.2 0.07 0.46 0.05

* All the other factors were not significantly associated with relapse.

4 sachets per day. The evaluation of symptoms during conservative measures was 81% 82% and 79% at 2 , 4
the follow-up is shown in Table 3. and 6 months, respectively. The percentages of patients
The percentage of patients stating that they followed stating that they experienced familial stress were 9 %, 6 %
'

Aliment Pharmacol Ther 1993. vol. 7.

 ALGINATE AS SELF-CARE TREATMENT FOR OESOPHAGITIS 389

100% Table 5. Factors associated with the alginate consumption

85 i 2% Regression analysis

I I
Standardized
e d e 2 regression
Factors identified coefficient P-value

Grade 3 Grade of oesophagiti 0.19 0.001

Initial treatment by omeprazole 0.15 0.02
50%
+
56 6% Length of initial attack treatment 0.16 0.001

All the other factors were not Significantly associated with

alginate consumption.

The percentages of subjects free from relapse at 6

months were 85 f2% in subjects with initial grade I
oesophagitis, 69 k 3 % in subjects with initial grade I1
I I 1
oesophagitis, and 56 f6 % in subjects with initial grade
2 4 6 I11 oesophagitis :these differences were highly significant
Time (months) (log-rank test = 46, P < 0.0001) (Figure 2).
Patients at risk The following variables were independently and
Grade 1 489 440 369 262 significantly associated with alginate consumption
Grade 2 305 264 211 142 (Table 5): the grade of initial oesophagitis (P = 0.001),
Grade 3 85 74 60 36
initial treatment by omeprazole ( P = 0.02), and the
Relapse (cumulative)
Grade 1 0 11 36 61 length of initial acute healing treatment (P = 0.001).
Grade 2 0 28 55 83
Grade 3 0 10 19 34
Truncated data 0 52 129 261
DISCUSSION
Total 879 879 879 879 This is one of the few large prospective studies of patients
Figure 2. Percentage of subjects free of relapse according to the after healing of non-severe oesophagitis. As expected in
initial grade of oesophagitis. The total of patients included is 879
such a large study, the mean number of patients included
and not 883 as in Figure 1, because for 4 patients, it was only
stated ‘ oesophagitis grade 1-111 ’, without details between grades. per physician was low, less than 5, and around 15% of
patients were lost to follow-up. It is possible that, in the
worst case, this 15% of patients included a higher rate of
relapse than in those who remained in the study. Patients
and 7%, and those stressed during work were 6% 5% were given a series of rules (‘conservative measures’)
and 4 %, respectively. Side-effects were minor (nausea, which they were supposed to obey. During the follow-up,
diarrhoea, constipation) and observed in l.O%, 0.3% every 2 months it was asked if these conservative
and 0.3%at 2 , 4 and 6 months, respectively. measures were observed, but no attempt was made to
Factors significantly associated with relapse during check the answers by interview of spouses or relatives.
follow-up in multidimensional analysis are shown in This is a weakness of a large multicentre study in out-
Table 4. All the other variables were not associated with patients, but this type of follow-up is offered by most
relapse. Using the Cox model, two variables were in- general practitioners. Similarly, despite the drug list
dependently associated with relapse: the initial grade of given to each gastroenterologist, it is not possible to be
oesophagitis (P < 0.01) and the occurrence of stress sure that other effective drugs were not taken by some
during follow-up (P < 0.05). Age, sex, ethnic origin, patients.
profession, alcohol or tobacco consumption, fat or spicy We have observed the fate of a large group of patients
diet, length of history, duration of acute treatment, after healing an oesophagitis for whom an ‘intention to
number of previous endoscopies, conservative measures treat ’ by conservative measures and alginate as self-care
were not associated with relapse. treatment, was decided by the gastroenterologist.Within

Aliment Pharmacol Ther 1993, vol. 7.

390 T. POYNARD & FRENCH STUDY GROUP

these limitations, this study is unique for the number of ACKNOWLEDGEMENT

patients included prospectively after two endoscopies.
This study was supported by a grant from SmithKline
Because of the number of drugs now available on the
Beecham Laboratoires Pharmaceutiques, Paris, France.
market, this type of study will be more and more difficult
to complete in the future.
Treatment with alginate was only taken to relieve
APPENDIX 1
symptoms, and therefore alginate consumption is an
estimate of disease activity. Less than 2 % of patients took
List of participants
no alginate during the 6 months, which confirms that
gastro-oesophageal reflux is mostly a recurrent disease. Abadia Rene (Yerres), Abbas Ali (Sarreguemines),
However, this disease is not usually severe, as the mean Abensur Robert (Blenod Les Pont a Mousson),Alle Jean-
consumption of alginate sachets was less than 1per day Pierre (Palaiseau),Allemoz Didier (Marseille),Amar Irene
(but the most severely symptomatic patients would have (Paris), Amerien Gerard (Sarreguemines), Amsellem
tended to withdraw from this trial). With this self-care Albert (Paris), Arramon-Tucoo Dominique (Orthez),
treatment, the mean number of days with pyrosis was A m 1 Claude (Nantes), Aubert Herve (Grenoble).
less than 6 per month, including less than 2 days per Aucomte Alain (Aubagne), Audegon Gilbert (Reims),
month with pain interfering with daily activity, and less Auffret Yves (Dinan), Azzi Antoine (Saint Avold), Badetti
than 2 nights per month with symptoms. Alginate was Jean (Marseille),Bagnouls Gerard (Metz),Bandu Martine
not sufficient to prevent a new endoscopy or the (Paris), Basin Norbert (Forbach), Berck Chantal
prescription of another medical treatment in 24%, of (Sarreguemines), Becqwort Jean-Pierre (Dunkerque),
subjects. 30% of these patients had an oesophageal Begue Jean-Yves (Nancy), Bengualid Jacques (Paris),
erosion during the new endoscopy-that is, 70% of Benhaim Serge (Marseille),Ekraud Guy (Le Puy). Billard
symptomatic relapses not controlled by alginate were J. L. (Saint Avold), Blime Jean-Paul (Epinal), Boehm
associated with gastro-oesophageal reflux but no Christian (Saverne), Bonin Noelle (La Garenne
endoscopic oesophagitis. The benefit of maintenance Colombes), Bottos Jean-Pierre (La Rochelle), Boucher
treatment with an antisecretory drug, in comparison Anne-Marie (Le Cannet), Boulenger Francis (Asnieres),
with alginate self-care treatment, after healing of non- Bouquet Jean-Paul (Decazeville),Boyer Jean-Dominique
severe oesophagitis is unknown and needs to be proved. (Vienne), Brajer Sylvain (Paris), Breban Pierre (Lens),
The initial grade of oesophagitis was the factor most Bredoux Paul (Pontivy), Brodschii Francois (Metz).
significantly associated with both alginate consumption Burguiere Pierre (Le Havre), Butel Joel (Abbeville),Calvet
and relapse (Figure 2). In patients with initial grade I or Bernard (Toulon), Cambray Sylvaine (Paris), Canard
grade I1 oesophagitis, which represents most patients Jean Marc (Paris), Causse Jean-Louis (Rodez), Cazals
with oesophagitis, the disease is mild and controlled by Jean-Brice (Agen), Cerf Gerard (Charleville-Mezieres),
simple self-care treatment. However patients who Chanteloup Jean Michel (Saintes), Chapalain Jean-
presented with grade HI oesophagitis had a 44%chance Claude (Montbrison). Charbit Michel (Levallois),
of needing a treatment other than alginate in the 6 Chauveinc Lionel (Dreux), Chiesa Alain (Sarcelles).
months of follow-up. In this sub-group of patients Colonna Patrick (Marseille), Colson Michel (Istres),
another therapeutic strategy needs to be tested by clinical Comuce Claude (Marseille), Congrad Patrick (Caen),
trial. Consigny Pierre (Carcassonne), Coomans Didier
Initial treatment using omeprazole was significantly (Bordeaux), Cotinat Martine (Creil), Cotten Jean-Claude
associated with alginate consumption, independently of (Guingamp), Coudert Philippe (Chalons sur Marne),
the initial severity of oesophagitis.It is possible that these Coulanjon Gilles (Issoire), Coulom Pierre (Tarbes).
patients had more severe symptoms initially, and hence Courtial Philippe (Nimes). Cuturello Laurent (Millau),
were treated with omeprazole and not an H,-blocker. It is Dantin Bruno (Le Bouscat), Decroix Bruno (Noyon),
also possible that patients treated with omeprazole are at Delette Olivier (Lille),Deltombe Bernard (Riom),Diacono
a higher risk of severe symptomatic relapse. Michel (Figeac), Diaz Jacques (Toulouse), Du Reau
In conclusion, after healing of grade I or grade 11 Antoine (Angers), Dudicourt Jean-Christophe (Rennes),
oesophagitis, self-care treatment using alginate is an Duperie Jean (Bordeaux), Dupin-Nizard Beatrice
effective strategy for the majority of patients. (Marseille),Dupont Thierry (Paris), Dury Philippe (Saint-

Aliment Pharmacol Ther 1993, vol. 7.

 ALGINATE AS SELF-CARE TREATMENT FOR OESOPHAGITIS 391

Avold), Edouard Regis (Tarbes), El Haik Elie (Paris), Ciotat), Parmentier Thierry (Antony), Parois Lionel
Elouard-Blanc Lisa (Paris),Estable Patrick (Guinganmp), (Epinay sur Orge), Pautrat Jean-Marie (Bergerac),
Figarol-Boutin Marie-Francoise (Tarbes), Fortas Loic Pauwels Bernard (Rivery Les Amiens), Pecriaux Olivier
(Toulouse),Fortier Bruno (Nice),Foucard Michel (Paris), (Vincennes), Pellat Bernard (Marseille), Peroua Jean
Fourguette Edouard ((Pau), Femond Luc (Le Havre), Jacques (Bordeaux), Perriguey Marthe (Montbeliard),
Gabriele Robert (Forbach), Gabriellini Francoise Pflumio Francis (Saverne), Picard Jean (Toulon). Plante
(Marseille), Galtie Guy (Montauban), Gaussen J. M. Michel (Oloron sur Marie), Plihon Genevieve (Rennes),
(Bordeaux), Gautier Jean-Michel (Bourges), Gobert Pommelet Pierre (Lille), Portefaix Jean-Pierre (Bagnols
Florence (Chalons sur Marne), Goldfain Denis (Dreux), sur Ceze),QuivigerJean-Francois (Ploumillau),Raimbert
Gomez Henri (Toulouse). Grange Martine (St Chamond), Philippe (Rueil Malmaison), Rapport Gilles (Paris),
Grunberg Bernard (Miramas), Halperin Sophie (Paris), Raynal Jean-Noel (Toulouse),Resensberg Michel (Paris),
Herbet Thierry (Marseille), Hilpert Gerard (Asnieres), Reig Jean-Jacques (Langon), Renault Guy (Neuilly sur
Houdee Gerard (Perpignan), Jacques Jean-Paul Seine), Revol Christian (Paris), Richard Mireille (Valff),
(Toulouse), Jonas Claude (Lille), Julien Pierre-Emile Richard Philippe (Rueil Malmaison), Riera Jean-Claude
(Boulogne), Julien Henri (Marseille), Jute1 Philippe (Longjumeau), Rion Jean-Louis (Rodez), Roblin Xavier
(Nantes), Kaffy Louis (Toulouse),Kalt Philippe (Epinal), (Saint Jean D’Angely), Rogier Patrice (Bourges),
Karsenty Claude (Nice), Kherroubi Hubert (Marseille), Rotenberg Andre (Dreux), Roussy Pierre (Bordeaux),
Kiriakos Alfred (Rouen), Kolopp Jean-Louis (Metz), Roux Michel (Marseille), Rumeau Jean-Michel
Lamiraud Bertrand (Mouilleron le Captif), Laplane (Bordeaux), Salvatori Gerard (Saint Etienne), Sanchez
Bertrand (Brive), Laredo Jean-Claude (Saint-Raphael), Jean-Yves (Nice), Saurfelt Marc-Michel (Selestat),
Larrue Philippe (Bordeaux), Laugros A. (Freyming- Schlosserg Patrick (Lens), Schmitz Daniel (Forbach),
Merlebach), Laurent Philippe (Palaiseau), Le Bayon Scotto Jean-Michel (Joue-Les-Tours), Seng G. (Freming-
Andre (Away), Le Bihan Pierrick (Guingamp), Le Merlebach), Servent Laurence (Le Mesnil Saint Denis),
Couteulx Christian (Yvetot),Le Texier Alain (Cannes),Le Simon Jean-Francois (Meudon),Soulie Jean-Marc (Nice),
Tortu Remy (Saint-Malo), Leclerc Philippe (Perigueux), Spreux Thierry (Cannes), Taoubi Nazem (Villefrance du
Leduc Jean-Louis (Fougeres),Lefevre Raymond (Auch), Rouergue), Teste Yves (Caen), Traissac Jean-Paul
Lelong Patrick (Lille),Lemarchand Nicolas (Paris),Lemee (Bergerac), Tubiana Bernard (Paris), Tubiana Georges-
Patrice (Dman), Antoine Leplat (Wattrelos), Leurent Eric (Cannes), Vandro Pierre (Hyeres). Veber Pierre
Bruno (Saint Omer), Lhermie Michel (Douai), Loison (Metz). Vedel Jean-Pierre (Nice), Venot Jacques (Saint-
Patrick (Creil), Lombard0 Gerard (Bordeaux), Lombart Junien), Vermeulen Laurie (Monaco), Verneau Alain
Jean (Nice), Loth Francois (Saverne), Loubiere Michel (Poitiers), Vernisse Bernard (Saint Germain en Laye),
(Draguignan),Loyer Robert (Marseille),Luciani Francois Veyres Bruno (Nice),Viaud Stephane (Perigueux), Vielh
(Paris), Lunaud Bernard (Clermont-Ferrand),Macherey Jacques (Metz), Villand Jacques (Dijon), Villard Jack
Denis (Marseille),Magen Michel (Libourne),Magois Jean- (Agen), Vitaux Jean (Paris), Wallez Jean-Yves (Lille),
Yves (Mouilleron Le Captif),Mahe Michel (Saint-Brieuc), Wantier Michel (Rivery Les Amiens),Weber Jean-Etienne
Maire Philippe (Epinal), Majeau Jean-Marc (Prades), (Perigueux), Zarka Alfred (Ecully).
Maniere Alex (Cannes), Marchetti Bernard (Marseille),
Margulies Alain (Joue Les Tours), Marois Francis
(Sannois), Martin George (Bethune), Masbou Pierre Statistical analysis and monitoring
(Cahors), Massabie Bernard (Toulouse), Mattei Alain Helene Benand Agostini, Quanta Medical (Rueil-
(Istres), Merlin Patrick (Verdun), Messerschmitt Claude Malmaison).
(Compiegne),Mir Wies Sidiqian (Grasse),Monges Bruno
(Aubagne), Mourrut Christian (Carcassonne), Navarro
Philippe (Marseille), Nayrac Aline (Antony), Neumeier REFERENCES
Martin (Sarreguemeines), Nguyen Phuong Thao
1 Festen H P M. Driassen W M M, Lamers C B M. Von Tongeren
(Courbevoie).Nizard Gilles (Garges Les Gonesses),Noizet
J H M. Cietidine in the treatment of severe ulcerative reilux
Jean-Marie (Marseille), Noun Kheirallah (Bron), Nury oesophagitis:results of an 8-week double blind study and of a
Bernard (Montigny Les Cormeilles), Ollivier Bernard subsequent long term treatment. Neth J Med 1980; 23:
(Ploumillau), Pariente Edouard (Pau), Paris Yves (La 2 3 740.

Aliment Pharmacol Ther 1993. vol. 7.

392 T. POYNARD & FRENCH STUDY GROUP

2 Bright-Asare P, Behar J, Brand DL. Effects of long term 6 Beeley M, Warner J 0. Medical treatment of symptomatic
maintenance cimetidine (CIM) therapy on gastrooesophageal hiatus hernia with lowdensity compounds. Curr Med Res Opin
reflux disease (GEM). Gastroenterology 1982: 82: 1025 1972; 1: 63-9.
(Abstract). 7 Stanch C. Bennett J R. Alginate-antacid in the reduction of
3 Kaul B, Petersen H, Ericksen H. et al. Gastrooesophageal reflux gastro-oesophageal reflux. Lancet 1974 : i : 109-1 1.
disease. Acute and maintenance treatments with cimetidine. 8 Barnardo D E. Lancaster-Smith M, Strickland I D, Wright J T. A
Scan J Gastroenteroll986; 21: 1 3 9 4 5 . double-blind controlled trial of ‘Gaviscon’ in patients with
4 Koelz H R, Birchler R. Brethloz A, et al. Healing and relapse of symptomatic gastro-oesophageal reflux. Curr Med Res Opin
reflux oesophagitis during treatment with ranitidine. Gastro- 1975; 3 : 388-91.
enterology 1986: 91: 1198-205. 9 Filoche B, Carteret E, Couzigou B, et al. Essai randomise d u n e
5 Liebermann DA. Medical therapy for chronic reflux oeso- supension buvable d’alginate dans le traitement du syndrome
phagitis. Long term follow-up. Arch Intern Med 1987; 147: douloureux du reflux gastro-oesophagien. GastroentCrol Clin
1717-20. Bioll991: 15: 984-5.

Aliment Pharmacol Ther 1993. vol. 7.

 RESEARCH

Stability of Omeprazole Mixtures: Implications for

Paediatric Prescribing
Linda V Graudins, Naghmeh J Kivi, Bruce N Tattam

ABSTRACT An audit at the Sydney Children’s Hospital found

Background: Commercial omeprazole paediatric mixtures are that around 9% of all inpatients were prescribed
not available. The stability of omeprazole mixtures prepared omeprazole and 85% needed a liquid formulation.6 The
from Probitor capsules or omeprazole powder has not been Sydney Children’s Hospital pharmacy department
established. prepares 12 L of omeprazole liquid formulation per month
Aim: To determine the physicochemical stability of two and the Royal Children’s Hospital, Melbourne prepares
omeprazole mixtures, prepared from Probitor capsules or 2 L of omeprazole suspension each working day. A survey
omeprazole powder BP, over a 45-day period under refrigeration of hospital pharmacies in New Zealand found that
or at room temperature, shaken or unshaken. omeprazole mixture was the most common
Method: Omeprazole 2 mg/mL mixtures were prepared using extemporaneously compounded medicine.7
standard procedures by either emptying the contents of Probitor Commercial omeprazole mixtures are not available and
capsules or by adding omeprazole powder to preserved sodium must be prepared extemporaneously (off-label). 8 As
bicarbonate 8.4% solution. Each formulation was stored in 100 pharmacists in public hospitals are exempt from the
mL amber glass bottles under refrigeration (2–8 oC) or at room requirements for Good Manufacturing Practice licensing,
temperature (21 ± 2 oC). One bottle of each formulation was compounded products do not undergo regular testing
shaken and the other not shaken, pH was measured and colour and most pharmacy departments do not have the
change determined using a visual scale. Omeprazole concentrations resources to do so.9 In the Australian Pharmaceutical
were measured using liquid chromatograph tandem mass Formulary, the omeprazole 2 mg/mL children’s mixture
spectrometry on Days 1, 7, 14, 28 and 46. involves dispersing the contents of omeprazole capsules
Results: Omeprazole 2 mg/mL mixture prepared from Probitor into sodium bicarbonate 8.4% and is given a 28-day expiry
capsules was stable over 45 days when refrigerated and regularly date at room temperature. 10 Quercia et al. found
shaken. When unshaken, the omeprazole concentration omeprazole dispersed in sodium bicarbonate 8.4%
decreased rapidly over 7 days. Mixtures prepared from injection retained 90% potency for up to 14 days when
omeprazole powder had lower concentrations compared to the stored at 24 oC and up to 30 days when stored at 5 oC and
mixtures prepared from Probitor capsules, but tended to be -20 oC in glass vials. 11 Ogden et al. reported on an
more stable when stored at room temperature. Observed colour omeprazole mixture prepared using the contents of Losec
changes did not correlate with measured omeprazole capsules with a one-month expiry under refrigeration.12
concentrations and pH remained constant. The Sydney Children’s Hospital’s omeprazole liquid
Conclusion: Omeprazole mixture prepared from Probitor formulation uses Probitor capsules and is based on the
capsules was stable over 45 days, when refrigerated and regularly Australian Pharmaceutical Formulary’s formula
shaken. Omeprazole concentration markedly decreased in the (Appendix).10 The Probitor product information states
mixture prepared from omeprazole powder when refrigerated. that ‘there is no experience of Probitor in children’,
J Pharm Pract Res 2008; 38: 276-9. therefore this method is not endorsed by the
manufacturer.2 The process of preparing the suspension
INTRODUCTION is costly both in preparation time and drug costs.
Omeprazole is a ‘top ten’ Pharmaceutical Benefits Scheme Omeprazole powder BP is a less expensive alternative
drug by cost and volume and is widely prescribed for ($31 vs $160 for capsules per 4 g, based on the 2007
gastrooesophageal reflux disease in children. 1 As hospital acquisition price) and preparation takes less time,
omeprazole is broken down in gastric acid, it is available but stability data are not available.
either as enteric-coated spheres in capsules or enteric- Although it is well established that a liquid
coated pellets in dispersible tablets.2,3 Omeprazole liquid formulation that contains dispersed undissolved
preparations are indicated in children who are intubated, components needs to be regularly shaken to ensure dose
infants whose dose is a fraction of a tablet and infants consistency, this study aimed to simulate normal use
with swallowing or gut disorders requiring nasogastric, conditions in the home or ward, so that one sample from
percutaneous endoscopic gastrostomy or other feeding each formulation was not shaken to determine the impact
tubes. Dispersible tablets are an option but cannot be of this practice on the omeprazole concentration in the
used via fine bore tubes.4,5 formulation. If the omeprazole concentration decreases
because of an unstable formulation or incorrect storage,
Linda V Graudins, BPharm, DHP, GradDipClinEpid, FSHP, Medicines Safety the patient is at risk of therapeutic failure and the return
and QUM Pharmacist, Paediatric Therapeutics Program, Sydney Children’s
Hospital, and Conjoint Lecturer, School of Women’s and Children’s Health,
of reflux symptoms. Omeprazole stability is pH-dependent
University of NSW, Naghmeh J Kivi, MPharmSci, PharmD, Research Assistant, (stable at pH above 8), with omeprazole degradation
Bruce Tattam, Facility Manager, Mass Spectrometry Analytical & NMR Facility, occurring within minutes at a pH of 1 to 2.13
Faculty of Pharmacy, The University of Sydney, Sydney, New South Wales The aim of this study was to determine the
Address for correspondence: Linda Graudins, Paediatric Therapeutics Program,
Sydney Children’s Hospital, Randwick NSW 2031, Australia. physicochemical stability of two omeprazole mixtures,
E-mail: [email protected] prepared from Probitor capsules or omeprazole powder
BP, over a 45-day period under refrigeration or at room
temperature, shaken or unshaken.

276 Journal of Pharmacy Practice and Research Volume 38, No. 4, 2008.
METHOD concentration range of 125 to 2000 mg/mL. The intra-day
The omeprazole mixtures were prepared in the pharmacy variation of the assay method was determined by
department as follows (Appendix): replicate analysis (n = 5) run on the same day. The inter-
• Capsule formulation: contents of sufficient Probitor day precision was evaluated by measuring replicates of
capsules (Sandoz) were emptied into sodium the same samples at three different concentrations over
bicarbonate 8.4% with methyl hydroxybenzoate 1% a period of two weeks. The precision was expressed as a
solution to achieve a final omeprazole concentration percentage by calculating the intra-day and inter-day
of 2 mg/mL and allowed to disperse over four hours relative standard deviations. The accuracy (error) was
in the refrigerator. estimated by the per cent difference of the mean
• Powder formulation: sufficient omeprazole powder concentration determined from the known concentration
BP (Rhenochem AG, from Trapeze Associates, [(measured concentration - nominal concentration) ÷
Melbourne) was weighed on a Mettler balance nominal concentration × 100%]. The analysis was
(Model P1210) and dispersed in sodium bicarbonate repeated a number of times on the same day to test
8.4% solution with methyl hydroxybenzoate 1% repeatability and stability and on different days to test
solution to achieve a final omeprazole concentration instrument reproducibility. This was used to establish
of 2 mg/mL. the robustness of the sample analysis method. As the
The capsule and powder mixtures were each samples were non-biological, measurement of ionisation
transferred into 6 x 100 mL amber glass bottles and matrix effect in the electrospray ionisation/ liquid
labelled with the storage and handling conditions – chromatography-mass spectrometry/mass spectrometry
refrigerated and shaken; room temperature and shaken; was deemed unnecessary. The assay intra-day and inter-
refrigerated and undisturbed. day reproducibility data are summarised in Table 1. The
The omeprazole concentration was analysed on a peak area ratio of omeprazole to the internal standard
ThermoFinnigan TSQ 7000 LCMS/MS system. This was used for estimation of omeprazole concentration.
analytical method and equipment was chosen as it had
Table 1. Omeprazole mixture assay precision and accuracy
been validated for this assay and could be adapted if the
study were expanded to include assays of biological Intra-day Inter-day
Nominal omeprazole
fluids. A Hewlett Packard 1090 liquid chromatograph concentration (mg/mL) RSD Error RSD Error
controlled by the TSQ 7000 software was coupled with
125 4.5% -5 % 6.5% -3%
the system. An electrospray ionisation interface was used
in positive ionisation mode. The temperature of heated 500 3% -1% 3.5% 1%
capillary was set to 275 oC. Selected reaction monitoring 2000 1.5% 2% 2% 4%
was used for the analytical assay, monitoring 346 to 198
for omeprazole and 180 to 110 for phenacetin (internal
standard) the collision energies were set to 13 ev and 22 The colour of the samples were measured visually
ev, respectively. Argon was used as the collision gas at on a linear scale from 1 (black) to 10 (white). The pH of
2.0 mtorr. Separation of omeprazole and phenacetin the samples was determined using pH indicator strips
(internal standard) was achieved using a 5 µm Alltima (Merck). Omeprazole recovery was measured by spiking
C18 (150 × 2.1 mm ID) narrow-bore high pressure liquid standard concentrations of omeprazole into sodium
chromatography column equipped with a 5 µm C18 (20 × bicarbonate solution at pH 9 and using the extraction
2.1 mm ID) guard column cartridge. The mobile phase procedure described above and comparing the ratio to
was a mixture of acetonitrile:water (50:50) with a flow rate that of the standard (Table 2).
of 300 mL/min, vacuum filtered through a 0.45 mm high
volume low pressure filter (Millipore) before use. Table 2. Omeprazole recovery from samples with liquid-liquid
extraction
On each study day, duplicate samples were prepared
from each omeprazole stock solution. All the samples Omeprazole concentration (mg/mL) Recovery (% of total)
(except those undisturbed) were inverted three times and 125 93.5 + 3.5
3 mL of the mixture was transferred into clean glass tubes.
500 95.0 + 5.0
Immediately after, 100 µL of the sample was put into a
clean tube and 900 µL of distilled water added. Omeprazole 2000 96.5 + 3.0
was extracted by the addition of ethyl acetate 2 mL, mixed
on a vortex mixer for one minute and then centrifuged RESULTS
(3000 rpm) for three minutes. The organic (upper) layer The physical stability of the omeprazole mixtures
was transferred to a clean labelled tube and the aqueous assessed by colour change is summarised in Table 3.
layer was extracted again (as above) and the organic The colour of all of the samples were white (‘10’) after
layers where combined. The internal standard (50 µL, preparation but by Day 46, all of the samples were rated
phenacetin methanol 500 ng/mL) was added to the at or below ‘8’, with mixtures stored at room temperature
organic layer, then dried under a stream of nitrogen before the darkest colour. The pH remained stable at pH 9 over
reconstitution in 100 µL of mobile phase and injected (20 the seven weeks of the study.
µL) onto liquid chromatography-mass spectrometry/ The omeprazole concentration data are presented in
mass spectrometry. The omeprazole (5 mg/mL) and Table 4 and Figure 1. Omeprazole concentrations are the
phenacetin (5 mg/mL) stock solutions were prepared in average taken from three readings from two of each
methanol and stored at -20 ºC. On each day that samples samples (n = 6). Omeprazole concentration in the mixtures
were prepared, the multipoint standard curves were decreased over 46 days for all formulations except those
constructed from the omeprazole stock solution (2000, prepared from capsule content, refrigerated and shaken,
1500, 1000, 500, 250 and 125 mg/mL). Standard curves for where concentrations above 1800 mg/L (90% of initial
omeprazole were linear (r 2 0.982–0.996) over the concentration) were observed throughout the study

Journal of Pharmacy Practice and Research Volume 38, No. 4, 2008. 277
Table 3. Colour rating of the omeprazole samples (1 = black, omeprazole powder BP retained only 3% of the original
10 = white) omeprazole concentration after one day. The omeprazole
Day of Prepared from omeprazole Prepared from concentration was around 50% of the initial concentration
analysis capsule content omeprazole powder BP at Day 46 for the powder formulation that was regularly
Ref RmTemp Ref Ref RmTemp Ref shaken and refrigerated. This formulation was most stable
at room temperature when shaken, with concentrations
Shk Sh k Notshk Shk Shk Notshk varying from 65 to 90% over the 46 days.
1 10 10 10 10 10 10
DISCUSSION
7 10 8 10 10 8 10
The omeprazole mixture prepared from Probitor capsules
14 8.5 8 9 9. 5 6 9.5 when regularly shaken retained 93% of its initial
28 8 7 8 8.5 5 8.5 concentration under refrigeration for at least 46 days and
101% of its concentration at room temperature for seven
46 7.5 6.5 7.5 7.5 4 8
days. Although this formulation does not include a
Ref = refrigerated; RmTemp = room temperature; Shk = shaken; suspending agent, the capsule excipients (hypromellose,
Notshk = not shaken sucrose, mannitol, macrogol, polysorbate) may have
improved the quality of the mixture, when compared with
Table 4. Percentage of omeprazole concentration remaining in the formulation prepared with omeprazole powder BP.2
samples under various storage conditions
Surprisingly, the powder formulation retained 73% of the
Day of Prepared from omeprazole Prepared from initial omeprazole concentration over 46 days at room
analysis capsule contents omeprazole powder BP temperature when regularly shaken, even though this
Ref RmTemp Ref Ref RmTemp Ref formulation had the most significant colour change.
All of the omeprazole mixtures changed colour,
Shk Shk Notshk Shk Shk Notshk
however, colour change did not necessarily relate to the
100% 100% 100% 100% 100% 100% omeprazole concentration in the formulation. The reason
1 93% 88% 86 % 39% 75% 2.7% for the formulation colour change was not elucidated.
Quercia et al. also found the greatest colour change for
7 103% 101% 31% 54% 90% 0
samples kept at room temperature but did not speculate
14 94% 87% 10 % 48% 74% 0 on the possible cause of this phenomenon.11 The pH of
28 97% 81% 0 53% 87% 0 each omeprazole mixture was stable, indicating that
sodium bicarbonate was able to maintain the optimum
46 93% 77% 0 54% 73% 0
pH for omeprazole. This study did not investigate
Ref = refrigerated; RmTemp = room temperature; Shk = shaken; microbial contamination, but each formulation contained
Notshk = not shaken a standard preservative. DiGiancinto et al. used an
unpreserved formulation of omeprazole in sodium
period. The omeprazole mixture prepared using capsule bicarbonate and found no microbiological growth over
content was stable at room temperature, retaining 87 to 45 days. 14 This observation supported the microbial
101% of the original concentration (2000 mg/L) over 14 stability of the mixtures used in this investigation.
days. The fastest decrease in omeprazole concentration The importance of shaking the bottle before each
occurred in both the unshaken formulations. The dose administration should be emphasised to nursing
unshaken refrigerated formulation prepared using staff, parents and carers. Parents and prescribers should

2500

2000
Omeprazole concentration (mg/L)

CAP REF SHK

1500 CAP NOT REF SHK
CAP REF NOT SHK
POW REF SHK
1000 POW NOT REF SHK
POW REF NOT SHK

500

0
0 10 20 30 40 50
Time (days)
Figure 1. Omeprazole concentrations (mg/L) over the study period (days)

278 Journal of Pharmacy Practice and Research Volume 38, No. 4, 2008.
be aware of symptoms of overdosing or under-dosing 8. Kelly M, Gazarian M, McPhee J. Off-label prescribing. Aust Prescr 2005; 28.
9. National Coordinating Committee on Therapeutic Goods. Discussion paper
when formulations are changed. Even though on regulation of extemporaneously prepared medicines in non-hospital pharmacies.
refrigeration is ideal, the omeprazole mixture prepared Canberra: National Coordinating Committee on Therapeutic Goods. Available
from capsule content was found to be adequately stable from <www.tga.gov.au/meds/extempcomp2.pdf>. Accessed 10 April 2008.
10. Pharmaceutical Society of Australia. Australian pharmaceutical formulary
for seven days at room temperature, which is useful when and handbook. 20th ed. Canberra: Pharmaceutical Society of Australia; 2006.
transporting medications or travelling. The Australian 11. Quercia R, Fan C, Liu X, Chow MS. Stability of omeprazole in an
Pharmaceutical Formulary’s expiry date could be extemporaneously prepared oral liquid. Am J Health Syst Pharm 1997; 54:
1833-6.
extended from 28 to 46 days and the formulation should 12. Ogden S, Asghar J. Preparing omeprazole liquid from Losec capsules. N Z
be refrigerated and regularly shaken.10 Pharm 2005; 25: 20.
The product information for lansoprazole endorses 13. Kromer W, Kruger U, Huber R, Hartmann M, Steinijans VW. Differences in
its use in children over one year old and via a nasogastric pH activation rates of substituted benzimidazoles and biological in vitro
correlates. Pharmacology 1998; 56: 57-70.
tube, if contents of the sachet are mixed with apple juice.15 14. DiGiacinto, Olsen KM, Bergman KL, Hole EB. Stability of suspension
However, when Dunn et al. compared the delivery of formulations of lansoprazole and omeprazole stored in amber-coloured plastic
omeprazole and lansoprazole granules via a nasogastric oral syringes. Ann Pharmacother 2000; 34: 600-5.
15. Zoton (lansoprazole) Australian approved product information. Sydney:
tube in vitro, they found per cent delivery to be variable Wyeth Australia. Approved 14 September 2004, safety related notification 13
and poor. 16 Lansoprazole was also found to have a April 2007.
shorter stability than omeprazole liquid preparation when 16. Dunn A, White CM, Reddy P, Quercia RA, Chow MS. Delivery of omeprazole
and lansoprazole granules through a nasogastric tube in vitro. Am J Health Syst
refrigerated (14 days vs 45 days for omeprazole). 14 Pharm 1999; 56: 2327-30.
Flavoured sachets of omeprazole powder with sodium 17. Burnett JE, Balkin ER. Stability and viscosity of a flavoured omeprazole
bicarbonate are available in the US and tested for stability oral suspension for pediatric use. Am J Health Syst Pharm 2006; 63: 2240-7.
18. Johnson CE, Cober MP, Ludwig JL, Soler EM, Robert S. Stability of partial
and usability via small bore tubes (5, 6 and 8 French doses of omeprazole-sodium bicarbonate oral suspension. Ann Pharmcother
[French unit = 0.33 mm]).17 Adding the contents of the 2007; 41: 1954-61
omeprazole-sodium bicarbonate sachets to water to make 19. Tan E, Cranswick NE, Rayner CR, Chapman CB. Dosing information for
paediatric patients: are they really “therapeutic orphans”? Med J Aust 2003;
a 2 mg/mL suspension, Johnson et al. found a
179: 195-8.
concentration of 98% (initial concentration) 25 days when 20. Glass BD, Haywood A. Stability considerations in liquid dosage forms
stored under refrigeration.18 extemporaneously prepared from commercially available products. J Pharm
Dispensing off-label formulations for children’s Pharmaceut Sci 2006; 9: 398-426.
21. Nahata M. Pediatric drug formulations: challenges and potential solutions.
medicines is a common occurrence, as there are Ann Pharmacother 1999; 33: 247-9.
inadequate dose forms marketed specifically for
children.19 There are many implications for patient safety Received: 22 June 2008
Revisions requested after external review: 18 August 2008
and quality use of medicines in providing off-label Revised version received: 14 October 2008
formulations, such as the variability in clinical response Accepted: 18 October 2008
with different formulations.20,21 Consequently, different
omeprazole formulations may lead to unforeseen clinical
outcomes and may require dose adjustment when Appe ndix. Sydne y Childre n's Hos pital ome prazole 2 mg/mL
liquid formulation
changing formulations.
Commercial paediatric formulations would be the Ingre die nts Quantity
ideal solution. In the meantime, funding research on the Omeprazole 20 mg (Probitor) capsules 10
effect of stability and bioavailability of extemporaneous
formulations on patient outcomes should be encouraged Sodium bicarbonate 8.4 g
for all medicines commonly prescribed for children. Methyl hydroxybenzoate 1% solution APF 1 mL
In conclusion, omeprazole mixtures prepared from
Water for irrigation to 100 mL
Probitor capsules were stable over 45 days, when
refrigerated and regularly shaken. There needs to be more M e thod
discussion and action to provide suitable dose Weigh the sodium bicarbonate and transfer to a measuring cylinder.
formulations for paediatric patients.
Dissolve in a portion of water for irrigation.
Acknowledgments Add methyl hydroxybenzoate 1% solution APF.
The authors would like to thank SHPA for sponsorship of this study via the
DBL Research and Development Grants program. The authors also wish to Make up to volume with water and stir until dissolved.
thank the Faculty of Pharmacy, University of Sydney for the use of laboratory
space and the use of the LCMS/MS. Thanks to Julie Arena (Senior Pharmacist, Empty required quantity of Probitor capsules into the sodium
SCH) for help with the Grant application and to Ann Stubley (Pharmacy bicarbonate solution.
Department, SCH) for preparation of the omeprazole formulations.
Stir and leave refrigerated until dispersed (may take 4 to 5 hours).
Competing interests: None declared
Mix with hand- held blender or stir well before transferring into amber
References glass bottles.
1. Anon. Top 10 drugs. Aust Prescr 2007; 30: 142.
2. Probitor (omeprazole) Australian approved product information. Pyrmont:
Sandoz Pty Ltd. Approved 19 November 2001, safety related notification
December 2003.
3. Acimax (omeprazole) Australian approved product information. Glebe: Alp-
hapharm. Approved 18 August 1999, safety related notification 3 July 2008.
4. Elliott R. Nasogastric administration of omeprazole. Aust J Hosp Pharm 1998;
28: 174-6.
5. White R, Bradnam V. Handbook of drug administration via enteral feeding
tubes. Pharmaceutical Press. UK, 2007
6. Graudins LV. Variation in omeprazole paediatric dosing. J Pharm Pract Res
2008; 4: 277-9.
7. Kairuz T, Chhim S, Hasan F, Kumar K, Lal A, Patel R, et al. Extemporaneous
compounding in a sample of New Zealand hospitals: a retrospective study. N Z
Med J 2007; 120: 48-56.

Journal of Pharmacy Practice and Research Volume 38, No. 4, 2008. 279
 RESEARCH REPORTS

Pediatrics

Stability of Partial Doses of Omeprazole–Sodium Bicarbonate

Oral Suspension

Cary E Johnson, Mary Petrea Cober, and Jennifer L Ludwig

n healthy infants and children, the phys-

I iological process of gastric contents re-
fluxing into the esophagus is defined as BACKGROUND: Omeprazole–sodium bicarbonate powder for oral suspension has
recently been marketed. The manufacturer provides no information regarding the
gastroesophageal reflux (GER).1 Typical-
acceptability of using partial doses and recommends that the reconstituted
ly, this is common in infants, with vomit- suspension be administered immediately after preparation.
ing as the most common symptom.2 OBJECTIVES: To determine the stability of the powder for oral suspension over 45
Chronic symptoms or mucosal damage days, evaluate the stability of a partial dose (<20 mg) following exposure to
produced by the abnormal reflux of gastric Simulated Gastric Fluid USP (SGF) over 2 hours, and determine the feasibility of
contents into the esophagus is defined as administering the suspension through neonatal and pediatric nasogastric feeding
gastroesophageal reflux disease (GERD). tubes compared with lansoprazole.
The pathogenesis of GERD is multi- METHODS: Three identical samples of omeprazole–sodium bicarbonate suspen-
sion 2 mg/mL were stored in the refrigerator (3–5 ˚C) and assayed by high-
faceted, involving gastric acidity, gastric
performance liquid chromatography immediately after preparation and at 7, 15,
emptying, visceral hypersensitivity, and 30, and 45 days. Stability of a 1 mg/kg dose with an estimated volume of SGF for
frequency of reflux.3 Children with GERD a simulated 12.7 kg pediatric patient was determined in triplicate over 2 hours at
commonly experience manifestations such 37 ˚C. The ability to administer a typical dose of omeprazole suspension and lan-
as vomiting, esophagitis, poor weight soprazole suspension (microgranules and water compounded from lansoprazole
gain, respiratory disorders, and dysphagia. oral disintegrating tablets) was assessed in triplicate using 3 different sizes of
neonatal/pediatric nasogastric feeding tubes.
Management of GERD is imperative in
RESULTS: At least 98% of the initial concentration of omeprazole remained
controlling the symptoms and disorders
throughout the 45 day study period in all suspensions. The suspension maintained
associated with the disease. at least 93% of the initial concentration following exposure to SGF for 2 hours. The
Typically, the cornerstones of GERD lansoprazole bead mixture partially clogged the 6 French feeding tube and com-
treatment have included histamine2-re- pletely clogged the 5 French feeding tube. The omeprazole–sodium bicarbonate
ceptor antagonists (H2RAs) and more re- suspension was easily administered through all 3 sizes of neonatal/ pediatric
cently, proton pump inhibitors (PPIs). feeding tubes.
Both treatments are used to reduce the CONCLUSIONS: Omeprazole–sodium bicarbonate suspension 2 mg/mL prepared
from 20 mg packets was stable for at least 45 days when stored at 3–5 ˚C. A
amount of acid and therefore encourage
partial dose of 12.7 mg was stable following exposure to SGF for 2 hours at
healing and avoidance of symptoms. 37 ˚C. This suspension can be easily administered through 5, 6, and 8 French
H2RAs work by inhibiting the histamine2 neonatal/pediatric feeding tubes and, when taking time and ease of preparation
receptor on the gastric parietal cell, there- into account, it is cost competitive with simple omeprazole suspension.
fore decreasing acid secretion.3 PPIs work KEY WORDS: omeprazole suspension, pediatrics, stability.

Ann Pharmacother 2007;41:xxxx.

Published Online, 23 Oct 2007, www.theannals.com, DOI 10.1345/aph.1K246
Author information provided at the end of the
text.

www.theannals.com The Annals of Pharmacotherapy n 2007 December, Volume 41 n

Copyright © 2007 by Harvey Whitney Books Company. All rights reserved.

CE Johnson et al.

by covalently bonding and deactivating the hydrogen–potas- by a 20 mg packet of omeprazole–sodium bicarbonate for
sium–adenosine triphosphatase pump and are considered to oral suspension (20 mEq). Stability of the reconstituted oral
be the most effective acid suppressant medications.4 Numer- suspension may also be a concern; however, a recent report
ous randomized controlled trials in adults have shown the su- has shown the product to be stable in concentrations ranging
periority of PPIs compared with H2RAs in relieving symp- from 0.6 to 4 mg/mL for up to 28 days when stored at 4 ˚C.13
toms associated with GERD.5 In children, omeprazole and To increase the flexibility in preparation and storage
other PPIs have been studied for various indications (eg, time and to allow patient- specific, weight-based dosing of
esophagitis, peptic ulcer disease, pulmonary disease).6 omeprazole–sodium bicarbonate suspension, it would be
PPIs are available as extended-release capsules contain- highly desirable to have a longer expiration time following
ing enteric-coated granules, enteric-coated tablets, and as reconstitution and the ability to administer doses less than
an intravenous solution. The pH-sensitive enteric coating 20 mg.
is necessary to protect these acid-labile molecules and en- The purpose of this study was to determine the stability of
sure bioavailability.7 These dosage forms pose a variety of omeprazole–sodium bicarbonate oral suspension 2 mg/mL
problems in the pediatric and neonatal population, primari- over 45 days, evaluate the stability of a partial dose (<20 mg)
ly the inability of this age group to swallow them. To over- following exposure to Simulated Gastric Fluid, USP (SGF)
come this barrier in patients with swallowing disorders, over 2 hours, and determine the feasibility of administering
many healthcare providers administer the medication by the suspension through neonatal/pediatric nasogastric feeding
flushing the enteric-coated granules through nasogastric tubes compared with a water and bead mixture of lansopra-
tubes. In many cases, especially with small diameter tubes zole prepared from the oral-disintegrating tablets.
used in neonates, either the beads from the capsules or the
microgranules from the rapidly disintegrating tablets have Methods
irreversibly clogged the tubes.6 To overcome this problem,
liquid products of both lansoprazole and omeprazole have STABILITY DETERMINATION
been formulated by dissolving the granules in 8.4% sodi-
um bicarbonate solution. These dosage forms have been A liquid preparation of omeprazole 2 mg/mL was pre-
ideal for children with nasogastric, orogastric, or gastrosto- pared by slowly adding water to the contents of six 20-mg
my tubes and have allowed for individualized dosing using packets of omeprazole–sodium bicarbonate for oral sus-
milligrams per kilogram. However, for normal oral admin- pensiona in a glass mortar. The suspension was thoroughly
istration, these extemporaneous products have a very bitter mixed to produce a final volume of 60 mL. Details of the
taste, and physicians and patients have reported to our hos- preparation procedure are provided in Appendix I.
pital pharmacy on numerous occasions that some commu- Three identical samples were prepared, placed in 60 mL
nity pharmacies will not compound these products.6 plastic bottles with child-resistant caps,b and stored in the
An immediate-release, pleasantly flavored omepra- refrigerator (3–5 ˚C). A 1.25 mL sample was withdrawn
zole–sodium bicarbonate powder for oral suspension is from each of the 3 bottles with a 5 mL micropipette imme-
available as a single-use packet containing either 20 or 40 diately after preparation, and at 7, 15, 30, and 45 days. The
mg of omeprazole, each with 20 mEq (1680 mg) of sodi- opening in the pipette tip for the 5 mL micropipette was
um bicarbonate. The manufacturer provides no informa- determined to be large enough to accurately and repro-
tion regarding the acceptability of using partial doses and ducibly measure the viscous samples. After further dilution
recommends that the reconstituted suspension be adminis- to an expected concentration of 50 µg/mL with mobile
tered immediately after preparation.8 phase, the samples were assayed in duplicate by high-per-
The primary concern related to partial doses of omepra- formance liquid chromatography (HPLC).14,15
zole–sodium bicarbonate suspension is based on the possi- The stability-indicating HPLC method developed by
bility that the reduced amount of sodium bicarbonate in a Quercia et al.16 was used for the omeprazole assay. The in-
partial dose could be insufficient to neutralize and maintain strumentation included a constant-flow solvent-delivery sys-
the gastric fluid at a pH high enough to protect omeprazole temc and a 150 × 3.9 mm inside diameter, 4-µm particle
from degradation prior to absorption and therefore reduce columnd maintained at 35 ˚C with a column heater.e A vari-
bioavailability.9,10 However, several investigators have evalu- able volume injector,f an ultraviolet light detectorg set at 302
ated the efficacy of simplified omeprazole suspension 2
mg/mL prepared in sodium bicarbonate injection 84 mg/mL
in a dose of 10 mL (20 mg omperazole and 10 mEq sodium
a
bicarbonate) for adults and reported intragastric pH control Omeprazole powder for oral suspension, 20 mg packets, lot
C6A38851. Santarus, Inc., San Diego, CA.
representative of adequate omeprazole bioavailability.11,12 The b
2-oz amber prescription bottles with child-resistant caps, Owens
amount of sodium bicarbonate in simplified omeprazole sus- Illinois Prescription Products, Owens Illinois, Toledo, OH.
pension (10 mEq or 840 mg) is one-half the amount provided c
HPLC pump, model 501. Waters Associates, Inc., Milford, MA.

n The Annals of Pharmacotherapy n 2007 December, Volume 41 www.theannals.com

 Stability of Partial Doses of Omeprazole–Sodium Bicarbonate Oral Suspension

nm, and a recording integratorh were also used. The mobile ated for odor, pH tested,j and taste tested on each day of
phase consisted of 25:75 by volume acetonitrile and aqueous analysis. Microbiologic testing was not performed; howev-
0.05 M monobasic sodium phosphate buffer (pH 8.5 with di- er, other investigators have found no bacterial or fungal
lute phosphoric acid) delivered at a flow rate of 1 mL/min.16 growth in similar samples following storage.13
The stability-indicating capability of the omeprazole as- The stability of omeprazole was determined by evaluat-
say was reevaluated in our laboratory. Decomposition of ing the percentage of the initial concentration remaining at
omeprazole was forced by allowing 2 separate 2 mg/mL each time interval. Stability of the omeprazole oral liquid
samples to stand at room temperature for 15 minutes after formulation was defined as the retention of at least 90% of
adjusting the pH to 12 with 1 N sodium hydroxide or to a the initial concentration.
pH of 5 with dilute hydrochloric acid solution. The pH was
corrected to 7, and the solutions were diluted with mobile STABILITY OF A PARTIAL DOSE IN SIMULATED GASTRIC
phase to an expected concentration of 50 µg/mL and as- FLUID
sayed. Approximately 70% degradation was achieved with
the acid solution; however, no short-term degradation was An evaluation of the stability of omeprazole sodium bi-
achieved with the basic solution. No interfering peaks were carbonate suspension following oral administration can be
found following assay at 3 different wavelengths. The peak simulated in the laboratory based on data from published
for omeprazole appeared at 7.75 minutes, with unknown articles reporting fasting gastric fluid volumes, rates of
degradation products at 2.9 and 4.6 minutes. The basic solu- production, and hydrochloric acid concentrations in pedi-
tion was evaluated again at 45 days following heating to atric patients by age and weight (Table 1). The fasting gas-
60 ˚C for 2 hours. Approximately 20% degradation was tric fluid volume data should be representative of typical
achieved with an unknown degradation product at 3.8 min- dosing conditions since omeprazole is recommended to be
utes. administered on an empty stomach at least 1 hour before a
A 1 mg/mL stock solution of analytical-grade omepra- meal.8 For example, a 3-month-old infant weighs about
zolei was prepared in mobile phase on each day of sample 5.6 kg (50th percentile) and, based on 2 studies of children
analysis. Standard samples of omeprazole were prepared 0 –24 months of age, has a reported mean fasting gastric
by diluting the stock solution with mobile phase to concen- fluid volume ranging from 0.19 mL/kg to 0.25 mL/kg17,18
trations of 40, 45, 50, 55, and 60 µg/mL. A 50 µg/mL con- with a hydrochloric acid concentration of 34.8 mEq/L.18,19
centration of omeprazole was assayed in duplicate approxi- HCl production is 0.1 mEq/kg/h19,20 for a total 2 hour pro-
mately every tenth sample as an external control. A standard duction of 1.12 mEq. Using the higher reported value
curve was produced on each day of sample analysis using (0.25 mL/kg), the fasting gastric fluid volume calculates to
linear regression of the peak heights of omeprazole against be 1.4 mL with 0.049 mEq HCl. The total HCl (fasting + 2
omeprazole concentration. The standard curve was linear (r2 h production) is 1.17 mEq. The normal omeprazole dose
> 0.995) over the working range of concentrations. The be- for this child would be 1 mg/kg (5.6 mg) once or twice
tween-day and within-day coefficients of variation for the
omeprazole assay were 1.67% and 0.84%, respectively.
Each of the omeprazole samples was shaken thoroughly Table 1. Gastric Hydrochloric Acid Concentration and
by hand immediately before assay. Following dilution, all Production Rate in Pediatric Patients17-21
samples were centrifuged at 1000 rpm for 2 minutes to Fasting
separate possible insoluble components. Five microliters Gastric
Age Weight Volume HCl HCl Output HCl NaHCO3
of each sample were injected into the HPLC system and (mo) (kg) (mL/kg) (mEq/L) (mEq/kg/h) (mEq)a (mEq)b
each sample was assayed in duplicate. In addition, each
1c 2.0 0.25 26.4 0.08 0.33 2.0
sample was visually inspected for any color change, evalu- 1 4.2 0.25 26.4 0.08 0.70 4.2
3 5.6 0.25 34.8 0.10 1.17 5.6
6 7.54 0.25 49.2 0.24 3.71 7.54
12 10.4 0.25 114.2 0.24 5.29 10.4
d
Nova-Pak C-18 (4-µm particle size, 3.9 mm diameter × 150 mm), 12 12.7d 1.10 114.2 0.24 7.69 12.7
Waters Associates, Inc. 60 18 0.43 114.2 0.24 9.52 18.0
e
Eppendorf CH-30 column heater, PJ Cobert Associates, St. a
Total of fasting and 2 h production.
Louis, MO. b
f
Total based on an omeprazole dose of 1 mg/kg.
Variable-volume injector, model U6K. Waters Associates, Inc. c
g
Premature.
Tunable absorbance detector, model 486. Waters Associates, d
97th percentile for weight.
Inc.
h
Integrator-recorder, model 3394. Hewlett Packard Company,
Avondale, PA.
i
Omeprazole analytic grade powder, lot 086K1405. Sigma-
Aldrich, Inc., St. Louis, MO. j
Altex pH 41 meter. Beckman Instruments, Irvine, CA.

www.theannals.com The Annals of Pharmacotherapy n 2007 December, Volume 41 n

CE Johnson et al.

daily. Using the omeprazole–sodium bicarbonate suspen- ed. The beaker contents were maintained at body tempera-
sion 2 mg/mL with sodium bicarbonate 2 mEq/mL, the child ture (37 ˚C) and covered around the probe of the pH meter
would receive 2.8 mL suspension containing 5.6 mEq of and the tip of the burette with laboratory film to simulate a
sodium bicarbonate for each dose. The amount of sodium bi- relatively closed environment. A magnetic stir-bar continu-
carbonate provided by a normal dose (1 mg/kg) of omepra- ously mixed the contents to simulate stomach agitation. The
zole–sodium bicarbonate suspension for each of the case ex- initial concentration of omeprazole in the suspension prior to
amples is about double or more the total estimated baseline placement in the beaker was determined by HPLC as previ-
plus 2 hour production of HCl (Table 1). If one assumes a 1:1 ously described. A 0.5 mL sample was removed immediate-
ratio for neutralization, the excess sodium bicarbonate should ly after addition to the beaker and at 30, 60, 90, and 120 min-
maintain the pH above 7 and therefore protect omeprazole utes, diluted to an expected concentration within the range of
from degradation during even a prolonged absorption phase the standard curve with mobile phase and assayed in dupli-
of 2 hours, as noted in some pediatric patients with reflux. cate. The entire experiment was repeated 3 times.
We propose at least 2 potential problems with this as- Based on the volume of SGF added and the cumulative
sumption. First, the minimum pH for omeprazole stability volumes and milligrams of omeprazole removed with each
has been reported to be 7.5, with degradation escalating as 0.5 mL sample at each time point, the expected omepra-
the pH decreases.7 Second, sodium bicarbonate is a rela- zole concentration was calculated and compared with the
tively poor buffering agent. The neutralization reaction be- assayed concentration to determine the percent of omepra-
tween HCl (gastric fluid) and sodium bicarbonate (NaHCO3) zole remaining.
generates carbon dioxide (CO2), which temporarily reacts
with water to form carbonic acid (H2CO3 as in carbonated ADMINISTRATION THROUGH FEEDING TUBES
beverages). The buffering capacity of the sodium bicar-
bonate is temporarily blunted until the CO2 dissipates as a Another problem with the use of PPIs in some infants
gas from the gastric fluid. The duration and pH effect of and small children is associated with the need to adminis-
this reaction in gastric fluid is unknown. ter these agents through feeding tubes. The enteric-coated
An in vitro alternative to a bioavailability or efficacy products, even when supposedly dissolved in an 8.4%
study would be to evaluate the stability of omeprazole– sodium bicarbonate solution, have the potential to irre-
sodium bicarbonate suspension in an appropriate volume versibly clog the 5 French and 6 French feeding tubes and
of SGF, as determined for a selected pediatric patient, over have done so at our institution.
an absorption phase of 2 hours. We chose a potentially To determine the feasibility of administering the ome-
worst-case scenario of a 1-year-old child weighing 12.7 kg prazole–sodium bicarbonate powder for oral suspension
(97th percentile), as noted in Table 1. We used the reported through a feeding tube, a size range of oral feeding tubes
maximum fasting gastric fluid volume for children aged were injected with 2 different PPI suspensions: omepra-
1–14 years of 1.1 mL/kg.21 The HCl concentration for this zole–sodium bicarbonate suspension compounded according
age was not reported, so we chose the higher value for the to the method presented in Appendix I and a lansoprazole
4 –9 year age group of 114.2 mEq/L and the highest HCl suspension compounded from oral disintegrating tablets.k
production rate of 0.24 mEq/kg/h.19,20 The baseline fasting The dose of each PPI suspension was based on approximate-
gastric fluid volume was calculated to be 13.97 mL with ly 1 mg/kg/dose for some of the smallest patients typically
1.595 mEq of HCl, and the 2 hour HCl production was using each of the various pediatric feeding tubes. The ability
calculated to be 6.096 mEq. We converted these HCl totals to partially dose the lansoprazole suspension made from the
into equivalent volumes of SGF, which has a HCl concen- oral disintegrating tablets is more challenging than the partial
tration of 84 mEq/L with sodium chloride 2 g/L and pepsin doses made from the omeprazole–sodium bicarbonate pack-
3.2 g/L as follows: 1.595 mEq ÷ 0.084 mEq/mL (SGF) = ets. To administer a lansoprazole 15 mg oral disintegrating
18.99 mL SGF. The fasting volume of SGF (18.99 mL) tablet for use through a nasogastric tube, the manufacturer
was placed in a beaker. The 2 hour HCl production was recommends suspending the tablet in 10 mL of water.22 To
6.096 mEq ÷ 0.084 mEq/mL (SGF) = 72.57 mL (SGF). To more closely approximate a 1 mg/kg/dose of lansoprazole for
simulate gastric production, this volume of SGF was patients weighing less than 10 kg, only 5 mL of the suspen-
placed in a calibrated burette and was dripped into the sion (7.5 mg of lansoprazole) is needed. The problem with
beaker over 120 minutes at an approximate rate of 0.6 this method is the difficulty of accurately administering one-
mL/min. The beaker contents were continuously moni- half of the lansoprazole microgranules through an oral sy-
tored with a pH meter.j At the start of the simulation,
omeprazole–sodium bicarbonate suspension 2 mg/mL was
prepared from a 20 mg packeta; the 1 mg/kg calculated
dose (12.7 mg/6.35 mL) for this simulated patient was k
Lansoprazole, 15 mg oral-distintegrating tablet, lot 426952E21.
added to the beaker and the burette drip of SGF was initiat- TAP Pharmaceuticals, Inc., Lake Forest, IL.

n The Annals of Pharmacotherapy n 2007 December, Volume 41 www.theannals.com

 Stability of Partial Doses of Omeprazole–Sodium Bicarbonate Oral Suspension

ringe. The microgranules often adhere to the plunger of the in the refrigerator for at least 45 days. This supply can then
oral syringe, and it is essentially impossible to determine ex- be used to prepare variable unit doses of up to 20 mg. Up
actly how much of the dose has been administered.6 to a 45 day supply of the suspension can be easily and ac-
The ability to administer a standard age/weight appro- curately prepared in the community pharmacy and dis-
priate dose was determined with 3 different sizes of neona- pensed for home use when doses less than 20 mg are need-
tal/pediatric nasogastric feeding tubes (Table 2). Each sus- ed for infants and young children. Compared with simple
pension was drawn into an oral dosing syringel and inject- omeprazole suspension prepared from enteric-coated gran-
ed into the feeding tube, followed by an NaCl 0.9% flush ules in 8.4% sodium bicarbonate solution, there is no need
to ensure delivery of the entire dose. The procedure was for concern about undissolved particles clogging even the
repeated 3 times with each suspension. smallest feeding tube when using omeprazole–sodium bi-
carbonate powder for oral suspension, and the product has
Results a very pleasant taste if oral therapy is needed.
Based on the average wholesale price in the 2007 Red
STABILITY OF THE SUSPENSION Book,23 the drug cost of 100 mL of omeprazole–sodium bi-
At least 98% of the initial concentration of omeprazole carbonate powder for oral suspension, 2 mg/mL, prepared
remained throughout the 45 day study period in all suspen- with Zegerid 20 mg packets, is $51.00 compared with
sions (Table 3). There was no detectable change in odor or $42.44 for the least expensive generic omeprazole cap-
taste and no visible microbial growth in any sample. At 45 sules and the two 50-mL unit-dose injectable sodium bi-
days, there was a very slight yellowish tint in all samples, carbonate 8.4% solution containers commonly used to pre-
which were creamy white when first prepared. There was pare simple omeprazole suspension. However, considering
no appreciable change in the initial mean ± SD pH (8.25 ± flexibility of use as well as the time and ease of prepara-
0.05) in any of the samples. The samples were sweet, with tion, the overall difference is appreciably smaller and may
a pleasant peach–mint flavor with little or no aftertaste on even favor the former product.
all study days.
Limitations
STABILITY OF SUSPENSION IN SIMULATED GASTRIC FLUID
Although we have evaluated only one patient simula-
The omeprazole–sodium bicarbonate suspension main- tion, we chose a high, if not the highest, ratio of acid–base
tained at least 93% of the initial concentration following exposure for a given dose of omeprazole. The limitations
exposure to a baseline quantity of SGF with an additional of this simulation relate primarily to the reported values for
volume added slowly over 2 hours. The pH initially HCl concentration and production as being representative
dropped, rose at 30 minutes with evolution of carbon diox- of the general pediatric population; however, we tested the
ide, and then decreased but remained above 7.0 for the full upper limits. Our simulated stomach (beaker) had no outlet
2 hours (Table 4). to allow acid, drug, or base to flow out as might occur with
normal gastric emptying; however, only the selective loss of
sodium bicarbonate would have altered the pH of the envi-
ADMINISTRATION THROUGH FEEDING TUBES
ronment. We feel that the potential loss of bicarbonate, if any,
The lansoprazole bead suspension, irrespective of dose, would have been offset by the simultaneous loss of acid and
partially clogged the 6 French feeding tube and completely by omeprazole more rapidly reaching the higher pH environ-
clogged the 5 French feeding tube. The omeprazole–sodi- ment of the duodenum. Based on these results, we feel confi-
um bicarbonate suspension made from the 20 mg packets dent that following administration of doses less than 20 mg,
was easily and completely administered through all 3 sizes the omeprazole in the oral suspension prepared from the 20
of neonatal/pediatric feeding tubes. The results were the mg packets will not degrade below acceptable standards dur-
same for all 3 trials (Table 2). ing a reasonable absorption period of 2 hours. In vivo
bioavailability and/or efficacy studies would be desirable to
Discussion confirm our results.

Based on the results of this study, the omeprazole–sodi- Conclusions

um bicarbonate powder for oral suspension can be pre-
pared in the hospital pharmacy in bulk supply and stored Omeprazole–sodium bicarbonate suspension 2 mg/mL
prepared from the 20 mg packets was stable for at least 45
days when stored at 3–5 ˚C. A partial dose of 12.7 mg was
stable following exposure to SGF for 2 hours at 37 ˚C. This
l
Oral syringe, 10 mL, lot 197629. Baxa Corp., Englewood, CO. suspension can be easily administered through 5, 6, and 8

www.theannals.com The Annals of Pharmacotherapy n 2007 December, Volume 41 n

CE Johnson et al.

French neonatal/pediatric feeding tubes and, when taking Reprints: Dr. Johnson, College of Pharmacy, University of Michigan,
428 Church St., Ann Arbor, MI 48109, fax 734/763-4480, cejohn@
time and ease of preparation and other factors into account, it umich.edu
is cost competitive with simple omeprazole suspension.

Cary E Johnson PharmD FASHP, Professor of Pharmacy, College of

References
Pharmacy, University of Michigan; Clinical Pharmacist—Pediatrics, 1. Nelson SP, Chen EH, Syniar GM, et al. Prevalence of symptoms of gas-
University of Michigan Hospitals and Health Centers, Ann Arbor, MI
troesophageal reflux during childhood: a pediatric practice–based survey.
Mary Petrea Cober PharmD, Clinical Assistant Professor of Phar- Pediatric Practice Research Group. Arch Pediatr Adolesc Med 2000;154:
macy, College of Pharmacy, University of Michigan; Clinical Phar-
150- 4.
macist—Pediatric Surgery, University of Michigan Hospitals and
Health Centers 2. Nelson SP, Chen EH, Syniar GM, et al. Prevalence of symptoms of gas-
troesophageal reflux during infancy. A pediatric practice–based survey.
Jennifer L Ludwig PharmD, at time of study, Adjunct Clinical In-
structor of Pharmacy, College of Pharmacy, University of Michigan; Pediatric Practice Research Group. Arch Pediatr Adolesc Med 1997;151:
Pharmacy Practice Resident, University of Michigan Hospitals and 569-72.
Health Centers; now, Specialty Resident in Hematology/Oncology, 3. Rudolph CD, Mazur LJ, Liptak GS, et al. Guidelines for evaluation and
The Nebraska Medical Center, Omaha, NE treatment of gastroesophageal reflux in infants and children: recommen-

Table 2. Administration of Two Different PPI Suspensions Through Three Sizes of Pediatric Nasogastric Feeding Tubes
Size of
Pediatric
Feeding PPI Suspension Administered/Observations
Tube
(French) Omeprazole–Sodium Bicarbonatea Lansoprazole Oral-Disintegrating Tabletb

8c 10 mg administered easily through pediatric feeding tube 7.5 mg administered easily through pediatric feeding tube
6d 5 mg administered easily through pediatric feeding tube 7.5 mg administered through partially clogged feeding tube; microgranules
stuck in feeding tube hub-barrel junction, but NaCl 0.9% flush cleared the tube
5e 2.5 mg administered easily through pediatric feeding tube 7.5 mg unable to be administered through completely clogged feeding tube;
microgranules stuck in feeding tube hub-barrel junction; unable to administer
NaCl 0.9% flush

PPI = proton pump inhibitor.

a
Omeprazole–sodium bicarbonate suspension dosing based on approximately 1 mg/kg/dose for smallest patient for each feeding tube; administered
10 mL NaCl 0.9% flush with 8 and 6 French feeding tubes and 5 mL NaCl 0.9% flush with 5 French feeding tube.
b
Lansoprazole dosing based on dissolving 15 mg oral-disintegrating tablet in 10 mL of water and administering 5 mL of resulting suspension; ad-
ministered 10 mL NaCl 0.9% flush with 8 and 6 French feeding tubes and 5 mL NaCl 0.9% flush with 5 French feeding tube.
c
Corflo-Ultra Lite, 8 French Pediatric Feeding Tube, lot 25168, Viasys MedSystems, Wheeling, IL.
d
Corflo-Ultra Lite, 6 French Pediatric Feeding Tube, lot 25160, Viasys MedSystems, Wheeling, IL.
e
Corflo-Ultra Lite, 5 French Neonatal Feeding Tube, lot 24986, Viasys MedSystems, Wheeling, IL.

Table 3. Stability of Omeprazole–Sodium Bicarbonate for Oral Suspension 2 mg/mL

% Initial Concentration Remaininga

Actual Initial Drug
Concentrationa (mg/mL) Day 7 Day 15 Day 30 Day 45

2.00 ± 0.04 98.96 ± 1.39 100.12 ± 1.88 100.18 ± 0.55 98.05 ± 1.66

a
Mean ± SD of duplicate determinations for 3 samples.

Table 4. Stability of Omeprazole for Oral Suspension 2 mg/mL with Simulated Gastric Fluid USPa
% Initial Concentration Remainingb
Initial 30 min 60 min 90 min 120 min

Concentrationb,c (mg/mL) 0.52 ± 0.03 99.97 ± 1.03 96.80 ± 2.36 95.68 ± 1.24 93.11 ± 1.89
pH 7.33 ± 0.07 7.72 ± 0.01 7.29 ± 0.07 7.03 ± 0.01 7.01 ± 0.02
a
Simulated patient 12.7 kg: 18.96 mL Simulated Gastric Fluid USP with 12.7 mg/1066.8 mg (6.35 mL) omeprazole–sodium bicarbonate suspension
(see text for calculations).
b
Mean ± SD of duplicate determinations for 3 samples.
c
Initial concentration following mixing; expected concentration was 0.50 mg/mL; pH of omeprazole prior to mixing was 8.27 ± 0.03; pH of Simulated
Gastric Fluid USP was 1.16 ± 0.04.

n The Annals of Pharmacotherapy n 2007 December, Volume 41 www.theannals.com

 Stability of Partial Doses of Omeprazole–Sodium Bicarbonate Oral Suspension

dations of the North American Society for Pediatric Gastroenterology 22. Package insert. Prevacid SoluTab (lansoprazole). Lake Forest, IL: TAP
and Nutrition. J Pediatr Gastroenterol Nutr 2001;32(suppl):S1-S31. Pharmaceuticals, Inc., September 2006.
4. Wolfe MM, Sachs G. Acid suppression: optimizing therapy for gastro- 23. Red Book 2007. Montvale, NJ: Thomson Healthcare Inc., 2007:591,
duodenal ulcer healing, gastroesophageal reflux disease, and stress-relat- 681,753.
ed erosive syndrome. Gastroenterology 2000;118(suppl 1):S9-31.
5. Chiba N, De Gara CJ, Wilkinson JM, et al. Speed of healing and symp-
tom relief in grade II to IV gastroesophageal reflux disease: a meta-anal-
ysis. Gastroenterology 1997;112:798-810.
Estabilidad de Dosis Parciales de Suspensión Oral de
6. Israel DM, Hassall E. Omeprazole and other proton pump inhibitors:
pharmacology, efficacy, and safety, with special reference to use in chil-
Omeprazol/Bicarbonato Sódico
dren. J Pediatr Gastroenterol Nutr 1998;27:568-79. CE Johnson, MP Cober, y JL Ludwig
7. Tutunji MF, Qaisi AM, El-Eswed B, et al. An in vitro investigation on
Ann Pharmacother 2007;41:xxxx.
acid catalyzed reactions of proton pump inhibitors in the absence of elec-
trophile. Int J Pharm 2006;323:110-6.
8. Package insert. Zegerid (omeprazole–sodium bicarbonate). San Diego, EXTRACTO
CA: Santarus, February 2006. TRASFONDO: Recientemente se ha comenzado la comercialización de un
9. Sharma VK. Comparison of 24-hour intragastic pH using four liquid for- compuesto en polvo de omeprazol/bicarbonato sódico para suspensión
mulations of lansoprazole and omeprazole. Am J Health Syst Pharm oral. El fabricante no proporciona información sobre la aceptabilidad del
1999;56(suppl):S18-21. empleo de dosis parciales y recomienda que se administre la suspensión
10. Sharma VK, Peyton B, Spears T, et al. Oral pharmacokinetics of ome- reconstituida de forma inmediata tras su preparación.
prazole and lansoprazole after single and repeated doses as intact cap- OBJETIVOS: Determinar la estabilidad del polvo para suspensión oral
sules or as suspensions in sodium bicarbonate. Aliment Pharmacol Ther durante 45 días, evaluar la estabilidad de una dosis parcial (menos de 20
2000;14:887-92. mg) tras su exposición a Fluido Gástrico simulado USP (SGF) durante 2
11. Phillips JO, Metzler M, Palmieri TL, et al. A prospective study of simpli- horas, y determinar la viabilidad de la administración de la suspensión a
fied omeprazole suspension for the prophylaxis of stress-related mucosal través de una sonda nasogástrica de alimentación neonatal y pediátrica
damage. Crit Care Med 1996;24:1793-800. en comparación con lansoprazol.
12. Lasky MR, Metzler MH, Phillips JO. A prospective study of omeprazole MÉTODOS: Se almacenaron tres muestras idénticas de suspensión de
suspension to prevent clinically significant gastrointestinal bleeding from omeprazol/bicarbonato sódico 2 mg/mL en la nevera (3–5 ºC) y se
stress ulcers in mechanically ventilated trauma patients. J Trauma Inj In- evaluó por cromatografía líquida de alta resolución (HPLC, por sus
fect Crit Care 1998;44:527-33. siglas en inglés) inmediatamente después de su preparación y a los 7, 15,
13. Burnett JE, Balkin ER. Stability and viscosity of a flavored omeprazole 30, y 45 días. Se determinó por triplicado la estabilidad de una dosis de
oral suspension for pediatric use. Am J Health Syst Pharm 2006;63: 1 mg/kg con un volumen estimado de SGF para un paciente pediátrico
2240-7.
hipotético de 12.7 kg durante 2 horas a 37 ºC. Se evaluó por triplicado la
capacidad de administrar una dosis típica de la suspensión de omeprazol y
14. Trissel LA. Avoiding common flaws in stability and compatibility stud-
de la suspensión de lansoprazol (microgránulos y agua procedente de la
ies of drugs. Am J Health Syst Pharm 1983;40:1159-60. desintegración de comprimidos orales de lansoprazol) a través de 3
15. Trissel LA, Flora KP. Stability studies: five years later. Am J Health Syst tamaños diferentes de sondas nasogástricas de alimentación pediátrica.
Pharm 1988;45:1569-71.
RESULTADOS: Al menos el 98% de la concentración inicial de omeprazol
16. Quercia RA, Chengde F, Liu X, et al. Stability of omeprazole in an ex- permaneció estable durante el periodo del estudio de 45 días en todas las
temporaneously prepared oral liquid. Am J Health Syst Pharm 1997;54: suspensiones. La suspensión mantuvo al menos un 93% de la
1833-6. concentración inicial tras la exposición a SGF durante dos horas. La
17. Cook-Sather SD, Harris KA, Rosetta Chiavacci R, et al. A liberalized suspensión de lansoprazol colapsó parcialmente la sonda de alimentación
fasting guideline for formula-fed infants does not increase average gas- 6 F y colapsó completamente la sonda de alimentación 5 F. La
tric fluid volume before elective surgery. Pediatr Anesth 2003;96:965-9. suspensión de omeprazol/bicarbonato sódico se administró sin
18. Splinter WM, Schaefer JD, Bonn GE. Unlimited clear fluid ingestion by problemas con los 3 tipos de sondas de alimentación
infants up to 2 hours before surgery is safe (abstract). Can J Anesth neonatales/pediátricas.
1990;37(suppl):S95. CONCLUSIONES: La suspensión de omeprazol/bicarbonato sódico 2
19. Agunod M, Yamaguchi N, Lopez R, et al. Correlative study of hy- mg/mL preparado a partir de un envase de 20 mg permaneció estable
drochloric acid, pepsin, and intrinsic factor secretion in newborns and in- durante al menos 45 días cuando se conserva a 3–5 ºC. Una dosis
fants. Am J Dig Dis 1969;14:400-14. parcial de 12.7 mg permaneció estable tras su exposición a SGF durante
20. Grand RJ, Williams JB, Torti FM. Development of the human gastroin- 2 horas a 37 ºC. Esta suspensión puede administrarse sin problemas
testinal tract. A review. Gastroenterology 1976;70:790-810.
mediante sondas de alimentación neonatal/pediátrica de tamaño, 5, 6, y
8 F y si tenemos en cuanta el tiempo y la facilidad de preparación, es
21. Sandhar BK, Goresky GV, Maltby JR, et al. Effect of oral liquids and coste competitivo con la suspensión de omeprazol simple.
ranitidine on gastric fluid volume and pH in children undergoing outpa-
tient surgery. J Anesth 1989;71:327-30. Traducido por Enrique Muñoz Soler

Appendix I. Procedure for Compounding Omeprazole Oral Suspension 2 mg/mL

1. Select six 20-mg omeprazole–sodium bicarbonate packets.

2. Empty contents of packets into a glass mortar.
3. Gradually add water to the powder in 3 steps:
a. Add approximately 30 mL to the powder, triturate well, and transfer contents to a 60 mL child-resistant amber plastic prescription bottle.
b. Rinse mortar with about 10 mL of water and transfer contents to the amber prescription bottle.
c. Repeat step b as necessary with enough water to bring the final volume to 60 mL.
4. Label the bottle: “Shake Well Before Use,” the calculated expiration date (45 days), and “Keep Refrigerated.”

www.theannals.com The Annals of Pharmacotherapy n 2007 December, Volume 41 n

CE Johnson et al.

Stabilité d’une Suspension d’Oméprazole et de Bicarbonate de bicarbonate de sodium par rapport à celle de la suspension magistrale
Sodium extemporanée de lansoprazole a finalement été évaluée à trois reprises
en utilisant des tubes nasogastriques pédiatriques de tailles différentes
CE Johnson, MP Cober, et JL Ludwig (grosseur 5, 6, 8).
Ann Pharmacother 2007;41:xxxx. RÉSULTATS: Plus de 98% des concentrations initiales d’oméprazole
demeurent présentes dans les échantillons de suspension qui sont
conservés à une température de 3 à 5 ºC durant une période de 45 jours.
RÉSUMÉ Près de 93% des concentrations initiales d’oméprazole se retrouvent
OBJECTIF: Une poudre pour suspension orale d’oméprazole et de après une exposition de 2 heures à un liquide gastrique USP. La
bicarbonate de sodium vient d’être récemment commercialisée. Le suspension d’oméprazole et de bicarbonate de sodium s’administre
fabricant ne fournit toutefois aucune information quant à la stabilité de facilement par les 3 tubes nasogastriques pédiatriques testés (grosseur 5,
cette suspension reconstituée et quant à son administration possible par 6, 8). Par contre, les tubes nasogastriques pédiatriques de taille 6 se sont
la voie nasogastrique avec une alimentation entérale. L’objectif de cette partiellement obstrués suite à l’administration de la suspension
étude était donc de déterminer (1) la stabilité de la suspension extemporanée de lansoprazole alors qu’une obstruction totale des tubes
reconstituée sur une période de 45 jours; (2) la stabilité d’une dose de taille 5 a été notée dans de telles circonstances.
partielle de cette suspension après avoir été mélangée à un liquide CONCLUSIONS: La reconstitution d’un sachet de poudre de 20 mg
gastrique USP pendant une période de 2 heures; (3) la facilité d’oméprazole et de bicarbonate de sodium permet d’obtenir une
d’administration de cette suspension par la voie nasogastrique en suspension d’une concentration finale de 2 mg/mL qui demeure stable
présence d’une alimentation entérale, comparativement à celle d’une pour un minimum de 45 jours à une température variant entre 3 et 5 ºC.
suspension magistrale extemporanée de lansoprazole. Une dose partielle de 12.7 mg est stable pour une durée de 2 heures à la
MÉTHODOLOGIE: Trois échantillons d’une suspension d’oméprazole et de température de 37 ºC lorsqu’elle est mélangée à un liquide gastrique
bicarbonate de sodium ont été analysés par une technique de USP. La suspension d’oméprazole et de bicarbonate de sodium
chromatographie à haute pression en phase liquide (HPLC) s’administre facilement par les tubes nasogastriques pédiatriques de
immédiatement après la reconstitution de la suspension et après 7, 15, taille 5, 6, et 8. Le coût associé à la préparation et à la facilité
30, et 45 jours d’entreposage à une température variant entre 3 et 5 ºC. d’administration de la suspension d’oméprazole et de bicarbonate de
La stabilité d’une dose de 12.7 mg (à raison de 1 mg/kg pour un enfant sodium demeure compétitif lorsqu’il est comparé à celui des
de 12.7 kg) de cette suspension après avoir été mélangée à un suspensions magistrales extemporanées de lansoprazole ou
échantillon de liquide gastrique USP à une température de 37 ºC a aussi d’oméprazole.
été déterminée par 3 analyses s’échelonnant sur une période de 2 heures.
La facilité d’administration de la suspension d’oméprazole et de Traduit par Sylvie Robert

n The Annals of Pharmacotherapy n 2007 December, Volume 41 www.theannals.com