You are on page 1of 8

ORI GI NAL ARTI CLEjgh_5900 1236..

1243
Comparison of p.o. or i.v. proton pump inhibitors on 72-h
intragastric pH in bleeding peptic ulcer
Gul Javid, Showkat Ali Zargar, Riyaz-u-saif., Bashir Ahmad Khan, Ghulam Nabi Yatoo,
Altaf Hussain Shah, Ghulam Mohammad Gulzar, Jaswinder Singh Sodhi and Mushtaq Ahmad Khan
Department of Gastroenterology, Sher-i-Kashmir Institute of Medical Sciences, Soura, Srinagar, Kashmir, India
Abstract
Background and Aims: After successful endoscopic hemostasis in bleeding peptic ulcer,
addition of proton pump inhibitors reduce the rate of recurrent bleeding by maintaining
intragastric pH at neutral level. The aim of the present study was to evaluate the effect of
various proton pump inhibitors given through different routes on intragastric pH over 72 h
after endoscopic hemostasis in bleeding peptic ulcer.
Methods: Ninety consecutive patients who had successful endoscopic therapy of bleeding
peptic ulcer underwent 72-h continuous ambulatory intragastric pH study, were randomly
assigned to receive p.o. omeprazole 80 mg bolus followed by 40 mg every 12 h for 72 h or
i.v. 80 mg omeprazole followed by infusion 8 mg/h for 72 h. Oral pantoprazole 80 mg
bolus followed by 80 mg every 12 h for 72 h or i.v. 80 mg pantoprazole followed by
infusion of 8 mg/h for 72 h. Oral rabeprazole 80 mg bolus followed by 40 mg every 12 h
for 72 h or i.v. 80 mg rabeprazole followed by infusion 8 mg/h for 72 h. Five patients
received no treatment after successful endoscopic therapy and underwent 72-h pH study.
Results: Mean 72-h intragastric pH for p.o. omeprazole was 6.56 versus 6.93 for ome-
prazole infusion (P = 0.48). Mean 72-h intragastric pH for p.o. pantoprazole was 6.34
versus 6.32 for pantoprazole infusion (P = 0.62). Mean 72-h intragastric pH for rabeprazole
p.o. was 6.11 versus 6.18 rabeprazole i.v. (P = 0.55). Mean 72-h pH for the no proton pump
inhibitor group was 2.04.
Conclusion: There was no signicant difference among various proton pump inhibitors
given through different routes on raising intragastric pH above 6 for 72 h after successful
endoscopic hemostasis in bleeding peptic ulcer.
Key words
72-h intragastric pH, bleeding peptic ulcer,
endoscopic therapy, proton pump inhibitors.
Accepted for publication 27 February 2009.
Correspondence
Dr Gul Javid, Department of
Gastroenterology, Gulistan Manzil, Amira
Kadal Srinagar Kashmir India 190001.
Email:gul_javid@rediffmail.com
Introduction
Acute gastro duodenal ulcer bleeding remains a therapeutic chal-
lenge with signicant morbidity and mortality. Endoscopic
therapy using various modalities signicantly reduces re-bleeding,
need for surgery and mortality in patients with peptic ulcer bleed-
ing.
1
Endoscopic therapy achieves successful hemostasis in more
than 90% of patients, and re-bleeding occurs in 1030% of
patients.
24
Re-bleeding has an important impact on prognosis.
5,6
Addition of proton pump inhibitors (PPI) after successful endo-
scopic therapy reduces the risk of further re-bleeding.
79
Gastric
acidity may play an important role in pathogenesis of recurrent
bleeding. Ahigh gastric pH may help stabilize the clot by avoiding
pepsinogen activation.
10,11
The rst major defense against uncontrolled hemorrhage aside
from transient vasoconstriction which follows vascular injury is
the formation of a platelet plug. A platelet plug alone can secure
adequate hemostasis for several hours (except in lesions of major
vessels), but thereafter the plug disintegrates unless it has been
reinforced by a brin clot.
12
Signicant disorder at almost any
level of coagulation cascade can thus lead to hemorrhagic tenden-
cies. The physiology of hemostasis demands a pH value near 7. A
pH of less than 2 is responsible for pepsin activity. Pepsin activity
is inversely inactivated and denatured at pH 7. A pH of more than
4 is necessary for clot lysis prevention and a pH of more than 6 is
necessary for hemostasis.
13
Pharmacologically, PPI can quickly
achieve an optimal intragastric pH condition in support of the
physiological cascade of hemostasis.
14
Previously, effects of PPI on intragastric pH have been investi-
gated in single comparator studies
15,16
and cross-over study.
17
In a
meta-analysis on PPI and outcome of endoscopic hemostasis in
bleeding peptic ulcers, the authors concluded that in bleeding
peptic ulcer patients who have undergone successful endoscopic
therapy, the benet of PPI as adjuvant to endotherapy is indepen-
dent of route and dose of administration of PPI.
18
We performed a
randomized trial in order to compare the effect of high p.o. doses
and continuous infusion of PPI on intragastric pH over 72 h after
endoscopic therapy.
doi:10.1111/j.1440-1746.2009.05900.x
1236 Journal of Gastroenterology and Hepatology 24 (2009) 12361243 2009 The Authors
Journal compilation 2009 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd
Methods
Study population
This study was conducted in the Department of Gastroenterology,
Sher-i-Kashmir Institute of Medical Sciences, from May 2004 to
January 2007 and was approved by the Postgraduate Clinical
Research and Ethical Committee of Medicine. All consecutive
patients above 18 years of age with proven peptic ulcer (gastric or
duodenal) bleeding were included in this study.
After the documentation of gastrointestinal bleeding, patients
were resuscitated. All patients admitted to this hospital with a
history of peptic ulcer bleeding (i.e. hematemesis and/or melena)
or who bled while in hospital, underwent emergency endoscopy as
soon as possible, always within 12 h of bleeding or immediately
after resuscitation in patients with massive bleeding or shock. The
possibility of endoscopic therapy was discussed with the patient
and/or their relatives and written informed consent was obtained
before the endoscopy. Endoscopic therapy was given if endoscopy
showed a peptic ulcer in the stomach or duodenum with active
bleeding (spurting hemorrhage, oozing hemorrhage) or stigmata of
recent hemorrhage (a non-bleeding visible vessel). Assessment of
presence of those stigmata was made after adherent clots and
debris of the ulcer base had been vigorously washed away.
19
Initial
hemostasis was dened as no visible hemorrhage lasting for 5 min
after endoscopy therapy. Patients who achieved hemostasis with
endoscopic therapy (witnessed by endoscopy) were eligible for
entry into the study.
Method of endoscopic treatment
Endoscopic hemostasis was achieved using a combination of heat
probe preceded by epinephrine injection. Epinephrine (1 : 10 000
diluted in normal saline) was injected in aliquots of 0.51 mL into
and around the bleeding area. After injection therapy, heater probe
thermocoagulation was given to the ulcer using an Olympus heater
probe unit with 2.8-mm probe (Olympus USE-2; Olympus, Tokyo
Japan). The energy output of the heater probe was set at 25 J and
coaptive pulses (minimum of three) were applied until cavitation
and adequate coagulation was obtained. The bleeding site was
observed for 5 min and challenged with maximum water irrigation
for 10 s. If any further bleeding occurred, the above procedure was
repeated.
20
The endoscope used in the study was a bro-optic
endoscope, FG-29V (Pentax, Tokyo, Japan).
Exclusion criteria
Patients were excluded if they: were under 18 years of age; unable
or unwilling to give written informed consent; pregnant or lactat-
ing; taking anticoagulants; had more than one possible source of
bleeding; had severe coagulopathy (prothrombin time 30% less
than normal) or platelet count less than 50 000/mm
3
;
2
had previous
acid reducing surgeries (vagotomy, gastric resection); were mori-
bund because of terminal cancer or severe comorbid illness; or had
bleeding gastric cancer. Patients who did not obtain initial hemo-
stasis with endoscopic therapy or re-bled within 3 days were also
excluded.
pH monitoring
Eligible patients after successful endoscopic hemostasis under
went 72-h intragastric pH metery using a pH electrode connected
to a data recorder (Proxima Light 2; Mentova, Italy). The patients
were enrolled after written informed consent was obtained and
received treatment protocol as described. The electrode was posi-
tioned in the gastric body under uoroscopic guidance and con-
tinuously recorded pH at 6-s intervals.
Randomization and pharmacological treatment
Immediately after endoscopic control of bleeding and successful
placing probe of the pH monitor in the corpus of the stomach
(uoroscopically documented), eligible patients were randomly
assigned to receive different PPI (pantoprazole, omeprazole and
rabeprazole) through different routes (p.o. and infusion). No other
treatment in the form of antacid, H
2
-receptor antagonist or others
was given.
In the omeprazole group (Dr Reddys Laboratories, Hyderabad,
India), patients were randomly assigned to receive omeprazole
either as an i.v. bolus of 80 mg followed by continuous infusion of
8 mg/h for 72 h, or 80 mg p.o. bolus followed by 40 mg after every
12 h for 72 h and i.v. saline.
In the pantoprazole group (Sun Pharmaceutical Industries,
Mumbai, India), patients were randomly assigned to receive pan-
toprazole either as an i.v. bolus of 80 mg followed by 8 mg/h
infusion for 72 h, or 80 mg p.o. bolus followed by 80 mg after
every 12 h for 72 h and i.v. saline.
In the rabeprazole group (Cadila Pharmaceutical, Ahmedabad,
India), patients were randomly assigned to receive rabeprazole
either as an i.v. bolus of 80 mg followed by 8 mg/kg continuous
infusion for 72 h, or 80 mg p.o. bolus followed by 40 mg after
every 12 h for 72 h and i.v. saline. Patients in all groups were
allowed to take clear liquids only.
Randomization was carried out in the endoscopy laboratory by
opening an opaque sealed numbered envelope by the senior endo-
scopy technologist. Treatment assignments were made based on
random numbers derived from a table of random numbers in
blocks of four. Patients and the attending physicians taking care of
the patient were blinded to the nature of treatment. The treatment
code was known only to the senior endoscopy technologist and
pharmacist.
We also studied 72-h intragastric pH in ve patients who had
successful endoscopic hemostasis and did not receive any PPI
except i.v. saline thus acted as control group. After 72 h, patients
who were Helicobacter pylori-infected received triple therapy and
p.o. PPI daily for 8 weeks and patients who were not H. pylori-
infected received daily PPI for 8 weeks irrespective of treatment
protocol. Three out of ve patients in no PPI group were H.
pylori-infected and received triple therapy and p.o. PPI for
8 weeks; those who were not H. pylori-infected received p.o. PPI
for 8 weeks.
Clinical monitoring
Patients were observed for re-bleeding in a high-care facility of the
gastroenterology ward. Every patient was serially monitored for
vital signs, hemoglobin concentration, need for blood transfusion,
G Javid et al. i.v. vs p.o. PPI in bleeding peptic ulcer
1237 Journal of Gastroenterology and Hepatology 24 (2009) 12361243 2009 The Authors
Journal compilation 2009 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd
need for surgery and length of hospital stay. Demographic fea-
tures, comorbid illness, ulcer size, initial hemoglobin level, use of
non-steroidal anti inammatory drugs (NSAIDs) and H. pylori
status as determined by enzyme-linked immunosorbent assay
(ELISA) were recorded. Re-bleeding was dened by fresh hemate-
mesis, melena or both, with either shock (systolic blood pressure
of 100 mmHg or a pulse rate of 100 b.p.m., accompanied by
cold sweats, pallor and oliguria); or a fall in hemoglobin of 2 g/dL
or more over a 24-h period after initial stabilization of vital signs.
Patients meeting these criteria were excluded from our study and
underwent emergency endoscopy within 4 h to conrm the diag-
nosis of re-bleeding. Re-bleeding was managed in both groups
with repeat endoscopic therapy as before or by surgery as
necessary.
The pH monitor was also continuously checked to see the posi-
tion of the electrode and was subsequently manipulated under
uoroscope if needed. The primary end-point was to see the time
for which intragastric pH remained above 6 over 72 h with various
PPI when given through different routes.
Statistical analysis
The primary comparison was between the mean 72-h intragastric
pH with p.o. PPI and that with PPI infusion. If the difference
between these two measurements was less than 10% of the mean
pH with PPI infusion, then the two group forms were considered
therapeutically equivalent, provided the minimum pH achieved
with p.o. PPI was more than 6. For that reason, it was estimated
that 43 evaluable patients would be required to provide 95% over
all power to detect a difference of 10% between the p.o. PPI group
and PPI infusion group, assuming the variance to be 0.16 and the
signicance level (type 1 error) to be 0.05. The trial was designed
to randomize 45 patients in each group so that 15 patients each
would receive omeprazole, pantoprazole or rabeprazole p.o., or by
infusion.
Statistical analysis was performed on an intention-to-treat basis.
All data were expressed as the mean standard deviation. Quan-
titative data between different treatment groups were compared
using the Students t-test and MannWhitney U-test. Categorical
data were compared with Pearsons test and Fishers exact test.
21
The 95% condence intervals (CI) associated with proportions
were calculated. A two-tailed P-value < 0.05 was considered sig-
nicant. Statistical analysis was carried out with SPSS for
Windows (ver. 11.5; SPSS, Chicago, IL, USA).
Results
During the study period, 326 patients with peptic ulcer bleeding
were seen. The prevalence of various types of stigmata of recent
hemorrhage were as follows: 45 (13.80%) active bleeders (spurt-
ing and oozing); 78 (23.90%) non-bleeding visible vessel; 54
(16.56%) adherent clot or at red spot; and 149 (45.70%) clean
base. Of these, 123 patients who revealed active bleeding (45
patients) or non-bleeding visible vessel (78 patients) underwent
emergency endoscopic therapy. Initial endoscopic hemostasis was
obtained in 119 (96.74%) patients. Endoscopic treatment was
unsuccessful in four patients due to torrential bleeding that
obscured the bleeding area and prevented adequate endoscopic
treatment; these patients were treated by emergency surgery. Thus,
119 patients were taken for our study; of them, seven patients
refused to give consent for pH monitoring, and in six patients there
was difculty in placing the pH probe because of deformed nasal
septa or previous surgery in the nose. Re-bleeding occurred in
eight patients during 72-h pH monitoring who required second
endoscopy or surgery as necessary and eight patient refused 72-h
pH monitoring after giving initial consent. The remaining 90
patients underwent randomization to receive omeprazole, panto-
prazole and rabeprazole either in p.o. or infusion form in higher
doses. Five patients were taken for 72-h intragastric pH study, who
received endoscopic therapy but no PPI and i.v. saline thus acted as
the control group.
Omeprazole was given either as 80 mg p.o. bolus followed by
40 mg p.o. every 12 h for 72 h and i.v. saline or 80 mg i.v. bolus
followed by 8 mg/h infusion for 72 h. A total of 15 patients were
included in each group. Pantoprazole was given either as 80 mg
p.o. bolus followed by 80 mg p.o. every 12 h for 72 h and i.v.
saline or 80 mg i.v. bolus followed by 8 mg/h infusion for 72 h. A
total of 15 patients were included in each group. Rabeprazole was
given either 80 mg i.v. bolus followed by 8 mg/h infusion for 72 h
or 80 mg p.o. bolus followed by 40 mg p.o. every 12 h for 72 h and
i.v. saline. A total of 15 patients were included in each group. The
groups were well matched with respect to the patient and endo-
scopic features (Table 1).
The mean 72-h intragastric pH in the omeprazole group when
given through the p.o. route was 6.56 0.99 (standard error of
mean [SEM], 0.11) versus 6.93 0.96 (SEM, 0.10) when given as
infusion with a statistically insignicant difference between the
two groups of P = 0.48. The median 72-h intragastric pH with p.o.
omeprazole was 6.76 versus 7.14 for omeprazole infusion. The
minimum pH for omeprazole p.o. was 0.96 versus 1.12 for ome-
prazole infusion group. The maximum pH for omeprazole p.o. and
for omeprazole infusion was 7.26 and 7.54, respectively. The 25th
and 75th percentile of 72-h intragastric pH with p.o. omeprazole
was 6.60 and 6.94 versus 6.90 and 7.28, respectively with ome-
prazole infusion. The 95% CI for mean 72-h intragastric pH with
p.o. omeprazole was 6.336.79 versus 6.717.15 with omeprazole
infusion. There was no statistically signicant difference between
the two groups.
The mean 72-h intragastric pH in the pantoprazole group when
given through the p.o. route was 6.34 0.90 (SEM, 0.10) versus
6.32 0.81 (SEM, 0.09) when given as infusion with no statisti-
cally signicant difference between the two groups (P = 0.62).
The median 72-h intragastric pH with pantoprazole p.o. was 6.52
versus 6.48 with pantoprazole infusion. The minimum pH for
pantoprazole p.o. was 1.22 versus 1.54 for the pantoprazole infu-
sion group. The maximum pH for pantoprazole p.o. was 6.84
versus 6.88 for the pantoprazole infusion group. The 25th and 75th
percentile of 72-h intragastric pH with pantoprazole p.o. was 6.38
and 6.68 versus 6.32 and 6.64, respectively, with pantoprazole
infusion. The 95% CI for the mean with pantoprazole p.o. was
6.136.55 versus 6.126.50 with pantoprazole infusion. There was
no statistically signicant difference between the two groups.
The mean 72-h intragastric pH in the rabeprazole group, when
given through the p.o. route was 6.11 0.86 (SEM, 0.99) versus
6.18 0.83 (SEM, 0.09) with rabeprazole infusion with no statis-
tically signicant difference between the two groups P = 0.55. The
minimum pH for p.o. rabeprazole was 1.44 versus 1.56 with
rabeprazole infusion. The maximum pH for p.o. rabeprazole was
i.v. vs p.o. PPI in bleeding peptic ulcer G Javid et al.
1238 Journal of Gastroenterology and Hepatology 24 (2009) 12361243 2009 The Authors
Journal compilation 2009 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd
6.78 versus 6.78 with rabeprazole infusion. The 25th and 75th
percentile pH with rabeprazole p.o. was 6.16 and 6.44 versus 6.26
and 6.58, respectively, with the rabeprazole infusion group. The
95% CI for the mean with rabeprazole p.o. was 5.916.30 versus
5.996.37 with rabeprazole infusion. There was no statistically
signicant difference between the two groups (Table 2).
The mean pH on day 1 with omeprazole p.o. was 6.63 versus
6.61 with omeprazole infusion. The mean pH on day 2 with ome-
prazole p.o. was 6.82 versus 6.92 with omeprazole infusion. The
mean pH on day 3 with omeprazole p.o. was 6.72 versus 6.92 with
omeprazole infusion. The median time to reach a gastric pH of 6 or
above was 60 min. (range, 50140 min) after an initial p.o. bolus
Table 1 Demographic characteristics of patients of p.o. and infusion group of PPI*
Oral group n = 45 Infusion group n = 45
Omeprazole Pantoprazole Rabeprazole Omeprazole Pantoprazole Rabeprazole
n = 15 n = 15 n = 15 n = 15 n = 15 n = 15
No. of patients 15 15 15 15 15 15
Mean age (years) 35.6 7 35 15.8 35.6 11.7 35.8 10 35 15.8 33.4 10.8
Age range (years) 2845 1860 2250 2250 1860 2250
Presentation
Hemetemesis 7 6 8 6 5 7
Melena 12 13 14 13 13 14
Both 7 6 7 6 6 6
Shock at presentation 1 1 0 0 0 1
Mean Hb at presentation
(g/dL) range
9.2 2.1 (513) 9.5 2.1 (612) 9.1 2.1 (613) 9.1 2.1 (613) 9.3 2.2 (513) 9.5 2.1 (612)
Comorbid illness 0 0 0 0 0 0
H. pylori infected 10 9 10 9 10 8
NSAID intake 2 1 1 1 2 1
Stigmata of bleeding
Spurting 4 3 2 3 3 2
Oozing 3 4 3 4 2 4
Non-bleeding vessel 8 8 10 8 10 9
Mean ulcer size (cm) 1.4 07 1.4 07 1.4 07 1.2 0.8 1.2 0.8 1.2 0.8
Mean dose of epinephrine
(mL)
11.6 11.6 11.5 11.5 11.5 11.6
No. of heater probe pulses
used (mean)
4 4 4 4 4 4
*No statistical signicant difference between the two groups. Helicobacter pylori. NSAID, non-steroidal anti-inammatory drug; PPI, proton pump
inhibitors.
Table 2 Effects of various PPI when given in p.o. or infusion form on 72-h intragastric pH*
Oral group n = 45 Infusion group n = 45
Omeprazole Pantoprazole Rabeprazole Omeprazole Pantoprazole Rabeprazole
n = 15 n = 15 n = 15 n = 15 n = 15 n = 15
Mean 6.56 6.34 6.11 6.93 6.32 6.18
Median 6.76 6.52 6.32 7.14 6.48 6.42
Standard deviation 0.99 0.90 0.86 0.96 0.81 0.83
SEM 0.11 0.10 0.99 0.10 0.09 0.09
Minimum 0.96 1.22 1.44 1.12 1.54 1.56
Maximum 7.26 6.84 6.78 7.54 6.88 6.78
Percentiles
25 6.60 6.38 6.16 6.90 6.32 6.26
50 6.76 6.52 6.32 7.14 6.48 6.42
75 6.94 6.68 6.44 7.28 6.64 6.58
95% CI for Mean 6.33- 6.79 6.136.55 5.916.30 6.717.15 6.126.50 5.996.37
*P-value for: omeprazole p.o. vs omeprazole infusion (P = 0.48); omeprazole infusion vs pantoprazole infusion (P = 0.15); pantoprazole p.o. vs
pantoprazole infusion (P = 0.62); omeprazole infusion vs rabeprazole infusion (P = 0.09); rabeprazole p.o. vs rabeprazole infusion (P = 0.55); panto-
prazole infusion vs rabeprazole infusion (P = 0.73). This shows there is statistically no difference between PPI when given p.o.ly or as infusion. CI,
condence interval; PPI, proton pump inhibitors; SEM, standard error of the mean.
G Javid et al. i.v. vs p.o. PPI in bleeding peptic ulcer
1239 Journal of Gastroenterology and Hepatology 24 (2009) 12361243 2009 The Authors
Journal compilation 2009 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd
of omeprazole and persisted above 6 for approximately 98% of the
time in the rst 24 h and 100% of the time in the next 2 days of
72-h pH monitoring. The median time to reach intragastric pH of
6 or above was 45 min (range, 35125 min) after the initial i.v.
bolus of omeprazole infusion and persisted above 6 for approxi-
mately 98% of the time in the rst 24 h and 100% of the time in the
next 2 days of 72-h pH monitoring (Fig. 1).
The mean pH on day 1 with pantoprazole p.o. was 6.36 versus
6.61 with pantoprazole infusion and on day 2 mean pH with
pantoprazole p.o. was 6.58 versus 6.71 with pantoprazole infusion.
The mean pH for day 3 with p.o. pantoprazole was 6.60 versus
6.77 with pantoprazole infusion. The median time to reach a
gastric pH of 6 or above was 52 min (range, 44135 min) after the
initial p.o. bolus of pantoprazole and persists above 6 for approxi-
mately 97% of the time in the rst 24 h and 100% of the time in the
next 2 days of 72 h. The median time to reach a gastric pH above
6 was 45 min (range, 30118 min) after the initial i.v. bolus of
pantoprazole and persisted above 6 for 98% of the time in the rst
24 h and 100% of the time in the next 2 days of 72-h pH monitor-
ing (Fig. 2).
The mean pH on day 1 with rabeprazole p.o. was 6.16 versus
5.73 with rabeprazole infusion. The mean pH on day 2 with
rabeprazole p.o. was 6.29 versus 6.82 with rabeprazole infusions
and for day 3 were 6.35 versus 6.79, respectively. The median time
to reach a gastric pH of 6 or above was 58 min (range,
48130 min) after the initial p.o. bolus of rabeprazole and per-
sisted above 6 for 96% of the time in the rst 24 h and 100% of the
time in the next 2 days of 72-h pH monitoring. The median time to
reach a gastric pH of 6 or above was 42 min (range, 35115 min)
after the initial i.v. bolus of rabeprazole and persisted above 6 for
approximately 97% of the time in the rst 24 h and remained
above 6 for 100% of the time in the next 2 days of 72-h pH
monitoring (Fig. 3).
Subgroup analysis
The mean 72-h intragastric pH with omeprazole infusion was 6.93
versus 6.32 pantoprazole infusion, there was no signicant statis-
tical difference between the two groups (P = 0.15). The mean 72-h
intragastric pH for pantoprazole infusion was 6.32 versus 6.18
with rabeprazole infusion which was statistically insignicant
(P = 0.73). The mean 72-h intragastric pH for omeprazole infusion
was 6.93 versus 6.18 for rabeprazole infusion and the difference
was statistically insignicant (P = 0.09). The comparison of intra-
gastric pH for 72 h of different PPI when given through the i.v.
route and with no PPI is shown in Figure 4, and the difference
between the PPI group and no PPI group was statistically signi-
cant (P < 0.05). The mean 72-h intragastric pH with omeprazole
p.o. was 6.56 versus 6.34 with pantoprazole p.o. and the difference
was statistically insignicant (P = 0.61). The mean 72-h intragas-
tric pH with pantoprazole p.o. was 6.34 versus 6.11 with rabepra-
zole p.o. and the difference was statistically insignicant
(P = 0.57). The mean 72-h intragastric pH with omeprazole p.o.
was 6.56 versus 6.11 with rabeprazole p.o. the difference was
statistically insignicant (P = 0.29). The comparison of intragas-
tric pH for 72 h of different PPI when given through the p.o. route
and with no PPI is shown in Figure 5 and the difference between
the PPI group and no PPI group was statistically signicant
(P < 0.05).
No PPI group
Five patients after successful endoscopic hemostasis of bleeding
peptic ulcer underwent 72-h intragastric pH monitoring, and these
patients received only normal saline for 72 h. The mean pH was
0
1
2
3
4
5
6
7
8
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70
Time in h
pH
Figure 1 Graph showing effect of administration of omeprazole
through different routes on 72-h intragastric pH. Mean pH (omepra-
zole p.o.); Mean pH (omeprazole infusion).
0
1
2
3
4
5
6
7
8
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70
Time in h
pH
Figure 2 Graph showing effect of administration of pantoprazole
through different routes on 72-h intragastric pH. Mean pH (pantopra-
zole p.o.); Mean pH (pantoprazole infusion).

0
1
2
3
4
5
6
7
8
1 4 7 101316192225283134374043464952555861646770
Time in Hours
pH
Figure 3 Graph showing effect of administration of rabeprazole
through different routes on 72-h intragastric pH. Mean pH (Rabepra-
zole Oral); Mean pH (Rabeprazole Infusion).
i.v. vs p.o. PPI in bleeding peptic ulcer G Javid et al.
1240 Journal of Gastroenterology and Hepatology 24 (2009) 12361243 2009 The Authors
Journal compilation 2009 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd
2.05, and median pH was 2.20 (SEM, 0.06). The 25th and 75th
percentile of 72-h intragastric pH without any PPI therapy was
1.65 and 2.35 with a mean 95% CI was 1.922.17 (Table 3). The
pH never raised above 4 in these patients and when this pH was
compared with the other PPI group the difference was statistically
signicant P < 0.05.
Outcome
There was no statistically signicant difference between hospital
stay, re-bleeding, surgery and need of transfusion in patients
among different groups (Table 4). Five patients who did not
receive any PPI for 72 h after endoscopic therapy did not have
re-bleeding or surgery.
Discussion
The result of this study demonstrates that i.v. and high p.o. doses of
various PPI are equal in their ability to suppress gastric acid
secretion in patients of bleeding peptic ulcer treated with
endoscopic therapy. There was no difference in intragastric pH for
72 h with various PPI (omeprazole, pantoprazole and rabeprazole)
when given through different routes (p.o./infusion). But when the
PPI group was compared with the non-PPI group there was a
statistically signicant difference in 72-h intragastric pH. The pH
remained more than 6 for most of the time with PPI and it remained
less than 4 most of the time with the non-PPI group. Regarding the
outcome of patients there was no statistical signicance.
The pharmacotherapy of bleeding peptic ulcer after successful
endoscopic therapy is directed at attempting to keep the intragas-
tric pH above the protolytic range of pepsin to stabilize the clotting
process. Fresten et al.
22
conducted a study to compare 24-h intra-
gastric pH effects of lansoprazole 30 mg administered nasogastri-
cally with pantoprazole 40 mg administered i.v. Healthy adults
were enrolled in an open-label two-way cross-over single-centre
study. Lansoprazole produced a signicantly higher mean 24-h
intragastric pH value in relation to pantoprazole on day 1 (3.05 vs
2.76, P < 0.001) and on day 5 (3.65 vs 3.45, P = 0.024).
Figure 4 Graph showing effect of administration of various proton
pump inhibitors (PPI) in infusion form on 72-h intragastric pH and in
non-PPI group. - - - MEAN pH (Omeprazole Infusion); MEAN pH
(Rabeprazole infusion); MEAN pH (Pantoprazole infusion).
Figure 5 Graph showing effect of administration of various proton
pump inhibitors (PPI) in p.o. form on 72-h intragastric pH and in non-PPI
group. Mean pH (Rabeprazole Oral); MEAN pH (Pantoprazole Oral);
- - - MEAN pH (Omeprazole Rabeprazole Oral).
Table 3 Study of 72-h intragastric pH study of ve patients in non-PPI
group
Mean 2.05
Median 2.20
Standard deviation 0.55
SEM 0.06
Minimum 1.15
Maximum 3.45
Percentiles
25 1.65
50 2.20
75 2.35
95% CI (mean) 1.922.17
CI, condence interval; PPI, proton pump inhibitors; SEM, standard error
of the mean.
Table 4 Outcome of patients with bleeding peptic ulcer according to
treatment received after successful endoscopic hemostasis
Omeprazole p.o. Omeprazole infusion
Blood transfusion (units) 4 3
Hospital stay (days) 3.5 3.5
Surgery 1 0
Re-bleeding 2 1
Death 0 0
Pantoprazole p.o. Pantoprazole infusion
Blood transfusion (units) 3 3
Hospital stay (in days) 3.7 3.5
Surgery 1 l
Re-bleeding 1 1
Death 0 0
Rabeprazole p.o. Rabeprazole infusion
Blood transfusion (units) 3 2
Hospital stay (days) 3.5 3.7
Surgery 0 l
Re-bleeding 1 2
Death 0 0
These patients were excluded from 72-h pH study.
G Javid et al. i.v. vs p.o. PPI in bleeding peptic ulcer
1241 Journal of Gastroenterology and Hepatology 24 (2009) 12361243 2009 The Authors
Journal compilation 2009 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd
Philips et al.
17
conducted a ve-way cross-over study to eva-
luate and compare 24-h intragastric pH following standard doses
of esomeprazole, lansoprazole, omeprazole, pantoprazole and
rabeprazole, administered once daily in 34 H. pylori-negative
patients aged 1860 years with signicant gastro-esophageal
reux disease. This randomized open-label comparative ve-way
cross-over study evaluated 24-h intragastric pH prole. On day 5,
intragastric pH was maintained for a mean of 14.0 h with esome-
prazole, 12.1 h with rabeprazole, 11.8 h with omeprazole, 11.5 h
with lansoprazole and 10.1 h with pantoprazole (P 0.001).
Esomeprazole at the standard dose of 40 mg once daily provided
more effective control of gastric acidity at steady state than stan-
dard doses of lansoprazole, omeprazole, pantoprazole and rabepra-
zole (P < 0.05). We conducted a 72-h intragastric pH study on i.v.
pantoprazole and found that the median time to reach a gastric pH
of 6 was 45 min (range, 29118) after the initial bolus of panto-
prazole infusion and persisted approximately 5.67.1 for the
remaining 72 h.
9
Oral and i.v. dosage of pantoprazole have been
found to be equivalent in their ability to suppress gastric acid
secretion in patients with gastro-esophageal reux disease.
23
Pantoickova et al.
24
conducted a cross-over study on 18 H.
pylori-negative subjects. Twenty-four hour intragastric pH moni-
toring was performed on day of treatment (once daily for rabepra-
zole 20 mg, lansoprazole 30 mg, pantoprazole 40 mg, omeprazole
20 mg, omeprazole multiple unit pellet system [MUPS] tablets
20 mg or placebo). The intragastric pH (3.4) and time at pH of
more than 4 during the 24-h post-dose (8 h) were signicantly
greater with rabeprazole than with lansoprazole, pantoprazole,
omeprazole capsule, omeprazole MUPS or placebo (P 0.04 for
rabeprazole vs others).
Observations of therapeutic and physiological variability among
the ve different PPI continue to arise as our experience with these
agents increases. Theories behind these observations include dif-
ferences in parietal cell mass among different populations and
cytochrome P450 polymorphism resulting in different PPI phar-
macokinetic proles among individual patients. PPI therapy for
ulcer bleeding has been more efcacious in Asia than else-
where.
25,26
Most pH studies have been on volunteers or cross-over
studies or in small groups; this is a rst randomized trial to study
72-h pH in various PPI after successful endoscopic therapy in
patients of peptic ulcer bleed. Katz and colleagues
27
reported the
results of their evaluation of gastric acid suppression achieved with
ve available PPI at a standard recommended dose for erosive
esophagitis, and found that healing was better with esomeprazole
than other PPI. However, all of the PPI resulted in a pH value of
more than 4 for at least 810 h. The acid suppressive efcacy of
rabeprazole 20 mg has been shown to be similar to that of ome-
prazole,
28,29
as is the case with omeprazole and esomeprazole.
30
We
used high p.o. and i.v. forms of three PPI (omeprazole, pantopra-
zole, rabeprazole); as in these three PPI, both p.o. and i.v. forms
are available, and only recently has esomeprazole i.v. become
available; this why we could compare only these three PPI.
We used high doses of PPI as it has been shown that a rapid
increase of intragastric pH above 6 can be reliably achieved only
by continuous infusion with a large initial bolus dose of PPI.
31
In
bleeding peptic ulcer patients, the risk of ulcer re-bleeding is
highest during rst 3 days and most re-bleed in the rst 24 h;
32
this
was the rationale of choosing a high-dose treatment period of
3 days with PPI. The treatment regime was aimed at elevating and
maintaining the intragastric pH at a level 6 or higher and thereby
promoting hemostasis. In our study, intragastric pH for 72 h with
omeprazole when given in different routes (infusion vs p.o.)
showed that intragastric pH raised to above 4 within 1 h in both
groups and remained above 6 for more than 98% of the time for the
remaining 72 h. Almost the same results were obtained with the
pantoprazole and rabeprazole groups of drugs on intragastric pH
when given through different routes. There was no signicant
statistical difference on intragastric pH with group of drugs (ome-
prazole, rabeprazole and pantoprazole) when given in different
routes (p.o. vs infusion; P > 0.05). The median pH remained above
6 in all forms of drugs for more than 98% of the time in the rst
day and 100% of the time in the second and third days.
We excluded patients with re-bleeding or continuous bleed as
presence of blood would interfere with the study of pH. There
were eight patients who re-bled and were evenly distributed in
each group. We studied other outcome parameters like number of
blood transfusions, hospital stay, surgery, re-bleed and deaths.
There was no statistical signicance in outcome.
In conclusion, our results demonstrated that high doses of dif-
ferent groups of PPI (omeprazole, rabeprazole, pantoprazole)
when given in different forms (p.o. or infusion) after successful
endoscopic therapy achieved an intragastric pH of 6 and above
within 1 h of administration and maintained a pH of more than 6
for more than 98% of the time. There was an insignicant differ-
ence among various PPI on 72-h intragastric pH, however, there
was a signicant difference between the PPI group and non-PPI
group. Thus, we conclude that a high dose of PPI maintains intra-
gastric pH above 6 regardless of route of administration.
Acknowledgments
The authors thank Mr Nazir Ahmad Patoo, chief pharmacist, and
his staff; Mr Ghulam Nabi Dar, technical ofcer of the endoscopic
unit; Mr Ghulam Mustafa, technical ofcer of the motility labora-
tory, and his staff for assistance in this study. We also thank
Mr Gulzar Ahmad and Mr Nazir Ahmad for their secretarial
assistance.
References
1 Consensus Development Panel. Consensus statement on therapeutic
endoscopy and bleeding ulcer. JAMA 1989; 262: 136972.
2 Lin J, Wang K, Perng CL, Lee CH, Lee SD. Heat probe thermal
coagulation and multiple electrocoagulation for arrest of peptic ulcer
bleeding a prospective randomized comparative trial. J. Clin.
Gastroenterol. 1995; 21: 99102.
3 Kubba AK, Murphy W, Palmer KR. Endoscopic injection for
bleeding peptic ulcer, a comparison of adrenaline alone with
adrenaline plus human thrombin. Gastroenterology 1996; 111:
6238.
4 Laine L, Peterson WL. Bleeding peptic ulcer. N. Engl. J. Med. 1994;
331: 71727.
5 Rockall TA, Logan RFA, Devlin HD, Northeld TC. Risk
assessment after acute upper gastro intestinal hemorrhage. Gut 1996;
38: 31621.
6 Clason AE, Macleod DAD, Elton RA. Clinical factors in the
prediction of further hemorrhage mortality in acute upper
gastro-intestinal hemorrhage. Br. J. Surg. 1986; 73: 9857.
i.v. vs p.o. PPI in bleeding peptic ulcer G Javid et al.
1242 Journal of Gastroenterology and Hepatology 24 (2009) 12361243 2009 The Authors
Journal compilation 2009 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd
7 Lau JYW, Sug JJY, Lee KC et al. Effect of intravenous Omeprazole
on recurrent bleeding after endoscopic treatment of bleeding peptic
ulcers. N. Engl. J. Med. 2000; 343: 31016.
8 Javid G, Masoodi I, Zargar SA. Omeprazole as adjuvant therapy to
endoscopic combination injection sclerotherapy for treating bleeding
peptic ulcer. Am. J. Med. 2001; 111: 28084.
9 Zargar SA, Javid G, Khan BA. Pantoprazole infusion as adjuvant
therapy to endoscopic treatment in patients in peptic ulcer bleeding.
Prospective randomized controlled trial. J. Gastroenterol. Hepatol.
2006; (4): 71621
10 Barkham P, Tocantins TM. Action of human gastric juice on human
blood clot. J. Appl. Physiol. 1953; 6: 17.
11 Peterson WI, Cook DJ. Anti secretary therapy for bleeding peptic
ulcer. JAMA 1998; 280: 8778.
12 Green FW, Kaplan MM, Curtis LE, Levine PH. Effect of acid
and pepsin on blood coagulation and platelet aggregation.
Gastroenterology 1978; 74: 3843.
13 Patchett SE, Enright H, Atdhal N, OConnell W, ODonoghue DP.
Clot lysis by gastric juice, an in vitro study. Gut 1989; 30: 17047.
14 Yacyshyn BR, Thomson ABR. Clinical importance of proton pumps
inhibitor pharmacokinetics. Digestion 2002; 66: 678.
15 Rohan K, Hassel G, Hedenstron H. Effects of Esomeprazole 40 mg
versus Omeprazole 40 mg on 24 h intragastric pH in patients with
symptoms of gastrroesophageal reux disease. Dig. Dis. Sci. 2002;
47: 9548.
16 Rohss K, Wilder Smith C, Claor Nilsson C et al. Esomeprazole
40 mg provided effective control than standard doses of all other
proton pump inhibitors [A 418 Abstract]. Gastroenterology 2001;
120: 2140.
17 Miner P Jr, Kartz PO, Chen Y, Sostek M. Gastric acid control with
Esomeprazole, Lansoprazole, Omeprazole, Pantoprazole and
Rabeprazole. A ve way cross over study. Am. J. Gastroenterol.
2003; 98: 261620.
18 Andrivlli A, Aneese V, Caruso N et al. Proton pump inhibitors and
outcome of endoscopic hemostasis in bleeding peptic ulcer. A series
of meta analysis. Am. J. Gastroenterol. 2005; 100: 20719.
19 Johnston JH. Endoscopic risk factors of bleeding peptic ulcer.
Gastrointest. Endosc. 1990; 36 (Suppl. 5): 1620.
20 Lin HJ, Lo WC, Cheng YC, Perng CL. Role of intravenous
Omeprazole in patients with high risk, peptic ulcer bleeding after
successful endoscopic epinephrine injection. A prospective
randomized comparative trial. Am. J. Gastroentrol. 2006; 101:
5005.
21 Altman DG. Practical Statistics for Medical Research. London:
Chapman & Hall, 1991.
22 Fresten J, Chiu Y-L, Pan WJ, Lukasik N, Tubel J et al. Effect on 24
intragastric pH: A comparison of Lansoprazole administered
nasogastrically in apple juice and Pantoprazole administered
intravenously. Am. J. Gastroenterol. 2001; 96: 205865.
23 Metz DC, Pratha V, Martin P et al. Oral and intravenous dosage
forms of pantoprazole are equivalent in their ability to suppress
gastric acid secretion in patients within gastroenterooesophageal
reux disease. Am. J. Gastroenterol. 2000; 95: 62633.
24 Pantackova D, Dorta G, Ravic M, Jornod P, Blum AL. Acid
inhibition on rst day of dosing Comparison of four Proton pump
inhibitors. Aliment. Pharmacol. Ther. 2003; 17: 150714.
25 Leontidias GI, Sharma VK, Horiden CW. Systemic review and
meta analysis, enhanced efcacy of proton pump inhibitors
therapy for peptic ulcer bleeding in Asia a posthoc analysis from
Cochrane collaboration. Aliment. Pharmacol. Ther. 2005; 21:
105561.
26 Leher JK, Gideon RM, Braitman L et al. Intragastric pH control with
Esomeprazole once daily all ethnic groups are not equal [Abstract
#100]. Am. J. Gastroenterol. 2004; 99: S33.
27 Katz P, Miner P, Chen Y et al. Effect of 5 marketed proton pump
inhibitors on acid suppression relative to a range of pH threshold
[Abstract #102]. Am. J. Gastroenterol. 2004; 99: S34.
28 Galamiche JP, Zerbib F, Ducrotte P et al. Decreasing esophageal
acid exposure in patients with GERD. A comparison with
Rabeprazole and Omeprazole. Aliment. Pharamacol. Ther. 2001; 15:
134350.
29 Stedman CAM, Barclay ML. Review article comparison of
pharmacokinetics acid suppression and efcacy of proton pump
inhibitors. Aliment. Pharmacol. Ther. 2000; 14: 96378.
30 Andersson T, Rhss K, Bredberg E, Hassan-Alin M.
Pharmacokinetics and pharmacodynamics of Esomeprazole the
S-isomer of Omeprazole. Aliment. Pharmacol. Ther. 2001; 15:
15639.
31 Brunner G, Luna P, Hartman M, Wurst W. Optimising the
intra-gastric pH as a supportive therapy in upper GI bleeding.
Yale J Biol Med 1996; 69: 22531.
32 Lau JYW, Chan SCS, Lenuing JW. Evaluation of stigmata of
hemorrhage in bleeding peptic ulcer a sequential endoscopic study.
Endoscopy 1989; 30: 5138.
G Javid et al. i.v. vs p.o. PPI in bleeding peptic ulcer
1243 Journal of Gastroenterology and Hepatology 24 (2009) 12361243 2009 The Authors
Journal compilation 2009 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd