Best Practice & Research Clinical Gastroenterology xxx (xxxx) xxx

Contents lists available at ScienceDirect

Best Practice & Research Clinical Gastroenterology

journal homepage: https://ees.elsevier.com/ybega/default.asp

Proton pump inhibitors for upper gastrointestinal bleeding

Omar Kherad a, *, Sophie Restellini b, c, Myriam Martel c, Alan Barkun c
a
Internal Medicine Department, La Tour Hospital and University of Geneva, Switzerland
b
Department of Specialties, Division of Gastro-enterology and Hepatology, Geneva University Hospital, Switzerland
c
Division of Gastroenterology, McGill University, Montreal, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Acute upper gastrointestinal bleeding (UGIB) remains a public health burden with a persistent high
Received 31 December 2018 mortality despite advances in modern day management. Proton pump inhibitors (PPI) as medical therapy
Accepted 15 April 2019 is an attractive adjuvant to endoscopic treatment in UGIB but the method and dose of PPI therapy re-
mains controversial. This chapter aims to describe the current evidence addressing acute PPI use in the
Keywords: management of UGIB. It will explore the evidence behind the timing, the dosage and the mode of
Proton pump inhibitor
administration of PPI during initial UGIB management, prior to and immediately following endoscopy, as
Upper gastrointestinal bleeding
well as in the short-term following discharge.
Hemorrhage
Endoscopy
© 2019 Elsevier Ltd. All rights reserved.

Introduction PPI during the initial management of UGIB prior and immediately
following, as well as in the short-term following discharge.
Acute upper gastrointestinal bleeding (UGIB) remains a com-
mon cause of hospitalization with an annual incidence of 78/ Proton pump inhibitor and the role of acid suppression
1000 000 population and a reported mortality that decreased in the
United States over the last 2 decades from 4.7% to 2.1% [1,2]. UGIB In vitro data have explored the important role of acid in
can be categorized into variceal and non-variceal UGIB (NVUGIB) impairing hemostasis and causing clot digestion [7], highlighting
causes, as there are important differences in management strate- the important therapeutic role of acid suppressive drugs in the
gies. Peptic ulcer (PU) disease remains the most common cause of acute setting of UGIB. Maintenance of a high intragastric pH (above
NVUGIB and hospital admission diagnosis in 2012 among all 6.0) during the management of UGIB is indeed warranted, as the
gastro-intestinal related disorders in the United States [3]. Despite ability for platelets to form the primary hemostatic platelet plug is
advances in modern day management of UGIB, including optimized deeply impaired by an acid environment, being reduced by 75% at a
use of endoscopic therapy, the morbidity and mortality associated pH of 6.8 relative to a pH of 7.4. When the pH falls to 5 or 4, platelets
with UGIB remains significant, as does its health economic burden start to disaggregate (Fig. 1). Low pH levels also alter the platelet
[1,2,4,5]. Medical therapy is an attractive adjuvant to endoscopic aggregation response to ADP, collagen and adrenaline by both
treatment in UGIB and acid suppression with the use of high-dose inhibiting initial platelet aggregation and causing disaggregation
proton pump inhibitors (PPI) remains a cornerstone in the medical [8]. Acid suppression may also be beneficial in preventing fibrino-
management of acute UGIB; the optimal route of administration lysis in patients with upper GI bleeding and ensuring the integrity
and dosing however remain controversial [6]. of the mucus/bicarbonate barrier.
This review summarizes the protective pathophysiological For a long time, available agents did not permit such a sustained
mechanisms and the current evidence pertaining to the efficacy targeted elevation in gastric pH. It has been postulated that this is
and cost-effectiveness of PPI therapy in the management of UGIB. It why studies using H2-receptor antagonists did not demonstrate
will discuss the timing, dosage and the route of administration of significant improvements in important patient outcomes, such as
rebleeding, surgery or mortality [8].
Since the approval of omeprazole by the US Food and Drug
Administration (FDA) in 1991, PPI have been extensively used to
* Corresponding author. Internal Medicine Department, Ho ^ pital de la Tour, 3
avenue JD Maillard, 1217, Geneva, Switzerland.
treat a variety of conditions in the upper gastrointestinal tract,
E-mail addresses: [email protected] (O. Kherad), sophie.restellini@hcuge. including peptic ulcer bleeding as they are capable of producing
ch (S. Restellini), [email protected] (A. Barkun). profound acid suppression in the stomach. Due to their generic

https://doi.org/10.1016/j.bpg.2019.04.002
1521-6918/© 2019 Elsevier Ltd. All rights reserved.

Please cite this article as: Kherad O et al., Proton pump inhibitors for upper gastrointestinal bleeding, Best Practice & Research Clinical
Gastroenterology, https://doi.org/10.1016/j.bpg.2019.04.002
2 O. Kherad et al. / Best Practice & Research Clinical Gastroenterology xxx (xxxx) xxx

followed by continuous infusion of omeprazole before patients

underwent endoscopy. The mean (±SD) duration of infusion before
endoscopy was 14.7 ± 6.3 h in the omeprazole group and
15.2 ± 6.2 h in the placebo group. Endoscopic treatment was
required in 19.1% of patients who received omeprazole compared to
28.4% of patients who received placebo (P ¼ .007). Similarly, among
patients with peptic ulcer disease, active bleeding was significantly
less common in patients who received omeprazole (6.4% vs 14.7%;
P ¼ .01), and clean-based ulcers were found more often (64.2% vs
47.4%; P ¼ .001) [14]. Notably, rebleeding rate and mortality within
30 days were not reduced in patients receiving omeprazole (11% vs
8%, p ¼ .49 and 8% vs 7% p ¼ .78, respectively). A subsequent
Fig. 1. pH and Platelet aggregation (adapted from Ref. [9]). Cochrane meta-analysis of 6 randomized trials (n ¼ 2223 patients)
evaluating the use of a PPI before endoscopic evaluation also found
that PPI therapy prior to endoscopy (using different doses and
mechanism of action, their use was extended to include all non- varying routes of administration) did not significantly reduce
variceal causes of UGIB [10,11]. All currently approved PPI are mortality (odds ratio [OR] ¼ 1.12; 95% CI, 0.72e1.73), rebleeding
benzimidazole derivatives, heterocyclic organic molecules that (OR ¼ 0.81; 95% CI, 0.61e1.09), or the requirement for surgery
include both a pyridine and benzimidazole moiety linked by a (OR ¼ 0.96; 95% CI, 0.68e1.35) [15]. There was only a significantly
methylsulfinyl group(12). lower proportion of peptic ulcers with high-risk stigmata at
They are specifically designed to interact with the proton pump endoscopy (OR ¼ 0.67; 95% CI, 0.54e0.84) and significantly lower
of gastric parietal cells to induce profound and sustained acid in- rates of endoscopic treatment performed (OR ¼ 0.68; 95% CI,
hibition. Dexlansoprazole, esomeprazole, omeprazole, lansopra- 0.50e0.93).
zole, rabeprazole and pantoprazole are the most commonly Given these data and along with the favorable risk profile of
available worldwide as oral preparations; omeprazole, esomepra- early PPI use, the 2012 American college of Gastroenterology (ACG)
zole and pantoprazole are also available in i.v. form(12) (Table 1). guidelines and the 2015 European Society of Gastrointestinal
PPI interact by covalent binding with the proton pump of the Endoscopy (ESGE) guidelines recommend the use of PPI prior
parietal cell, thereby causing an irreversible inhibition of acid endoscopy in order to decrease the proportion of patients with
secretion. Consequently, only a de novo synthesis of proton pumps ulcers with high-risk stigmata and the requirement for endoscopic
will enable the parietal cell to further produce acid [8]. Theoreti- treatment(16, 17). In contradistinction, as PPI use prior endoscopy
cally, maximal acid inhibition should be obtained with a proton does not affect rates of rebleeding, surgery, or mortality NICE rec-
pump inhibitor if a significant amount of drug is available at the ommendations in UK did not recommend routine PPI administra-
precise moment the parietal cells become activated. Rationally, this tion prior to endoscopy(18). Likewise, the 2018 Asia-Pacific
is obtainable by continuous, not intermittent, i.v. administration of working group consensus on NVUGIB PPI rejects indiscriminate use
a PPI. However, the mode of administration has been recently of iv PPI in stable patients with suspected NVUIGB prior to
challenged by the publication of a meta-analysis supporting the endoscopy as there is no proven value(19).
non-inferiority of intermittent (oral or iv) vs continuous iv PPI
administration with regards to rebleeding rate and mortality [13].
Eventually, the rationale behind acid-suppressive therapy in UGIB Route of administration
is to rapidly increase the intragastric pH above 6.0 and so maintain Several trials of repeated bolus administration have demon-
it for 3e4 days. Although every PPI has different substitutions on strated that an initial loading dose of 80 mg omeprazole or pan-
the pyridine and/or benzimidazole rings, in general they are all toprazole is preferable to 40 mg in achieving fast and sustained
remarkably similar in their pharmacological properties. Besides, increases in intragastric pH, without additional influence if using a
head-to-head data are few and taken overall do not suggest sig- larger loading dose [8].
nificant differences(10). Therefore, any benefit of the PPI is assumed Intravenous administration prior endoscopy seemed legitimate
to be a class effect in the setting of UGIB. as the aim of acid-suppressive therapy in UGIB is to rapidly increase
the intragastric pH above 6.0 and people are often fasting awaiting
PPI therapy prior endoscopy upper endoscopy. Despite their high bioavailability, iv PPI route
should be then preferred, where indication before endoscopy is
Current evidence appropriate. The 2012 ACG guidelines and the 2015 ESG guidelines
The guidance on whether to administer PPI therapy prior to for NVUGIB recommend the iv route of PPI administration with
endoscopy is conflicting. In 2007, a randomized trial (n ¼ 638 pa- 80 mg bolus following by a continuous perfusion (8 mg/h) while
tients) by Lau et al. reported the benefit of a high-dose bolus awaiting an early endoscopy, based on the best evidence [16,17].

Table 1
Pharmacological difference between different proton pump inhibitors (PPI) [12]

PPI Equivalent oral dose Routes of administration Time to peak plasma level (tmax, hr) Bioavailability (%)

Omeprazole 20 mg oral/IV/DR 0.5e3.5 30e40

Esomeprazole 20 mg oral/IV/DR 1.5 64e90

Pantoprazole 40 mg oral/IV/DR 2e3 77
Lansoprazole 30 mg oral/DR 1.7 80e85
Dexlansoprazole MR 30 mg oral/DR 1-2; 4-5 na
Rabeprazole 20 mg oral/DR 12.1 52

IV: intravenously; DR: delayed release; MR ¼ produces two distinct releases of drug, na ¼ not available.

Please cite this article as: Kherad O et al., Proton pump inhibitors for upper gastrointestinal bleeding, Best Practice & Research Clinical
Gastroenterology, https://doi.org/10.1016/j.bpg.2019.04.002
 O. Kherad et al. / Best Practice & Research Clinical Gastroenterology xxx (xxxx) xxx 3

Cost-effectiveness consideration of PPI use prior to endoscopy  If endoscopic evaluation has to be delayed or cannot be per-
Cost-effectiveness analyses of PPI therapy prior endoscopy have formed, PPI therapy should be started and continued to reduce
shown mixed results. the risk of further bleeding.
Cost-effectiveness considerations are important because most  When PPI are considered, the high dose iv (80 mg) followed by
of randomized trials assessing PPI prior endoscopy found no sta- 8 mg/h should be the preferred regimen
tistically significant between-group differences in rebleeding, sur-  Additional data are required to better determine the efficacy and
gery, or mortality rates. Tsoi et al. provided an important economic cost-effectiveness of PPI prior to endoscopy and the subgroup of
analysis on the use of PPI while awaiting an early endoscopy [20]. patients more likely to benefit from such an approach
Overall, 631 patients were recruited, and 377 were eventually
found to be bleeding from ulcers. Among these, 60 (19.1%) patients
in the PPI and 90 (28.4%) patients in the placebo group required PPI therapy after endoscopy
endoscopic hemostasis at the time of gastroscopy (performed on
average 14e15 h after the onset of PPI administration). Authors Current evidence in non-variceal bleeding
tabulated direct costs and overall average per patient costs were It is important to contrast the pre-endoscopy from post-
U.S. $2813 for the intravenous PPI (80 mg bolus followed by 8 mg/h endoscopic hemostasis uses of PPI in patients with high risk stig-
infusion till the endoscopy), and U.S. $2948 in the placebo group. mata bleeding ulcers. The use of intravenous high-dose PPI has
PPI administration also reduced endoscopic therapy by 7.4% and become standard practice in the management of ulcer UGIB as it is
was thus both less costly and more effective than placebo, making it effective and less costly in most settings [10,16,17,19,24]. As
a dominant strategy in economic terms. mentioned above, increased gastric pH has been linked with
However, the conclusions of an economic analysis can vary ac- improved clot stability. A landmark placebo-controlled randomized
cording to the choice of unit of effectiveness or utility that in turn trial from Hong Kong showed that the administration of continuous
usually reflects a clinically meaningful outcome. Another economic omeprazole infusion (80 mg intravenous bolus followed by 8 mg/h
analysis assessing the cost-effectiveness of using PPI prior to for 72 h) after endoscopic therapy for bleeding peptic ulcers was
endoscopy concluded this strategy was slightly more effective and superior to placebo in reducing recurrent bleeding, transfusion
costlier than no administration(21). In a Canadian economic envi- requirements and hospital stay [25]. Subsequently, a meta-analysis
ronment, this approach becomes dominant as the duration of of randomized trials of intravenous PPI therapy (80 mg bolus fol-
hospitalization for high-risk ulcer patients increases or that of low- lowed by 8 mg/h continuous infusion) vs. placebo/no treatment for
risk ulcer patients decreases(21). This economic analysis also sug- 72 h after endoscopic therapy of high-risk stigmata revealed a
gested that PPI use pre-endoscopy is most cost-effective if endos- significant reduction in further bleeding (RR ¼ 0.40; 95% CI
copy is to be delayed for more than 16 h or in patients most likely to 0.28e0.59), surgery (RR ¼ 0.43; 95% CI 0.24e0.76), and mortality
be bleeding from a non-variceal UGIB cause, or a source likely to be (RR ¼ 0.41; 95% CI 0.20e0.84) [26]. The improvement in mortality
exhibiting a high-risk lesion [21]. was specifically in patients having first undergone successful
Pragmatically, if endoscopic evaluation has to be delayed or endoscopic hemostasis. Those authoritative data have been repli-
cannot be performed, PPI therapy should be continued to reduce cated in other meta-analysis [24,27] and unanimously incorporated
the risk of further bleeding. A Cochrane meta-analysis of random- into guidelines addressing the post-endoscopic management of
ized trials (n ¼ 4373 patients) of patients with UGIB who did not NVUIGB [10,16,17].
consistently receive endoscopic hemostatic therapy reported that Notably, in a large randomized trial of bolus followed by
PPI therapy was associated with reduced rebleeding (OR ¼ 0.38; continuous infusion PPI vs. placebo after successful endoscopic
95% CI, 0.18e0.81) and surgery (OR ¼ 0.62; 95%, CI 0.44e0.88), but hemostasis, subgroup analysis of patients with oozing bleeding
not mortality [22]. This suggests that if endoscopy will be delayed showed a very low rebleeding rate with placebo (8/163 (4.9%)). The
or cannot be performed, PPI therapy may improve clinical out- results of this subgroup analysis suggest that intensive PPI therapy
comes. Therefore, the above-mentioned observed lesion down- may not be needed for a subgroup of lesions exhibiting oozing
staging attributable to PPI therapy prior endoscopy may be even without other stigmata of recent hemorrhage [28], and that the
more beneficial in situations in which early endoscopy may be Forrest classification of Ia lesions may need revisiting.
delayed or when available endoscopic expertise may be
suboptimal. Route of administration
Additional data that are required to better determine the cost- It is recommended that after endoscopy, the route of adminis-
effectiveness of indiscriminate PPI administration before endos- tration and PPI dosing should be tailored to the identified source of
copy including the relationship between duration of PPI adminis- bleeding(10).
tration before endoscopy and subsequent endoscopic severity of
the lesion (Forrest class), the effects of differing drug doses, regi- Low risk stigmata
mens, and the determination of any possible effect of PPI on pa- Hemodynamically stable patients without serious comorbid
tients bleeding from non-ulcer lesions [21,23]. Efforts should be conditions who exhibit low-risk endoscopic lesions and therefore
made to better identify subgroups of patients or situations in which do not require endoscopic hemostasis (e.g., clean-based, flat, pig-
this use will be more likely to be cost-effective. mented spots) can be discharged on a once-daily oral PPI (40 mg)
[10,16]. Indeed, once-daily PPI therapy has demonstrated effective
ulcer healing for patients with peptic ulcer disease in short term
use [29]. Recommendations on duration of treatment remain
Practice points disparate, varying between 4 and 8 weeks, depending on the cause
of bleeding: 4 weeks for duodenal ulcer and 8 weeks for gastric
PPI prior endoscopy ulcer or until endoscopic reassessment [10,16]. These recommen-
dations are based on expert opinion, as only one trial has studied
 PPI started prior to endoscopy only decrease the proportion of different regimens and duration of treatment, finding that twice-
patients with high-risk stigmata ulcers and the requirement for daily oral dosing for the first 11 days may be preferable in pa-
endoscopic treatment but do not affect rebleeding or mortality tients at high risk of rebleeding. In this trial, patients with a Rockall

Please cite this article as: Kherad O et al., Proton pump inhibitors for upper gastrointestinal bleeding, Best Practice & Research Clinical
Gastroenterology, https://doi.org/10.1016/j.bpg.2019.04.002
4 O. Kherad et al. / Best Practice & Research Clinical Gastroenterology xxx (xxxx) xxx

score 6 who were given a twiceedaily oral PPI had a lower endoscopic hemostasis, classical administration of high-dose
rebleeding rate than those who received a once-daily oral PPI (11 intravenous PPI therapy for 3 days has proven to be both more
versus 29% at 28 days) [30]. effective and less costly than not doing so, as demonstrated by
numerous economic analyses [36e38]. PPI treatment initiated after
High risk stigmata endoscopy is cost-effective as it significantly reduces the incidence
Conversely, patients with evidence of high-risk stigmata after of re-bleeding and the need for surgery compared with placebo or
successful endoscopic hemostasis benefit from a higher-dosing of other antacid drugs [27]. High-dose intravenous PPI therapy is a
PPI therapy for 72 h but the optimal regimen remains controversial dominant strategy mainly because the cost of the medications is
[10,16,17]. Based upon previously published metanalyses, evidence- relatively lower than the incremental expenses of one additional
based guidelines on NVUGIB have recommended that PPI therapy rebleeding episode. Cost analyses should be repeated if additional
be given as an 80 mg intravenous bolus followed by 8 mg/h high-quality data become available on oral PPI use after UGIB.
continuous infusion for 72 h to reduce rebleeding, surgery, and Indeed, the administration of intravenous PPI requires nursing
mortality in patients with high risk ulcers that had undergone supervision leading to high costs, while oral administration is
successful endoscopic hemostasis [10,16,17]. In comparison to H(2)- attractive due to widespread availability, ease of implementation
receptor antagonist or placebo, the benefits of PPI seem to persist and economical [31].
regardless of the route of administration (iv or oral) and the dose Based on the results of an economic model, the strategy of
(high dose, defined as 80 mg bolus followed by 8 mg/h for 72 h, or administering oral PPI both before and after endoscopy with
lower dose) [24]. endoscopic therapy in NVUIGB would likely to be the most cost-
More recently, high-dose infusion of PPI (80 mg iv bolus effective but warrants further investigations [27]. A significant
following by a continuous perfusion (8 mg/h) for 72 h has been consideration in such economic suppositions needs however to be
challenged by meta-analyses [13,31e33]. Indeed, these have sug- the non-evidence based and questionable safety of premature
gested non-inferiority in rebleeding risk and other clinical out- discharge from hospital before the usual 72-h post-endoscopy
comes including the need for blood transfusion, surgery, length of period that represents the conventional highest risk period for ul-
hospital stay and mortality, when comparing continuous vs inter- cer rebleeding following endoscopic hemostasis.
mittent IV doses [13,31e33]. These results however are hampered
by poor methodological quality in the intermittent bolus studies Current evidence in variceal bleeding
and a confusion in concluding equivalence when only non- The best available evidence supports the use of short-course oral
inferiority can be surmised [34]. once-daily dose PPI for 10 days post-endoscopic variceal ligation to
Furthermore, additional data have suggested that a high dose reduce ulcer size if ulcer healing is a concern as a complication of
oral PPI regimen (40 mg every 12 h for 3 days) are non-inferior in sclerotherapy or variceal ligation [39].Practices such as high-dose
preventing recurrent bleeding from peptic ulcers when comparted infusion and prolonged use should be discouraged unless evi-
to iv regimens [13,31e33,35]. Generalizability of those results dence of such additional attributable benefits becomes available
coming mainly from Asian studies are however limited due to [39,40].
differences in underlying bleeding etiologies, genetic cytochrome
polymorphisms in the metabolism of PP, and the higher prevalence Practice points
of H.pylori in Asian populations. Additionally, most of the included
trials informing these meta-analyses were relatively small and  Authoritative guidelines support use of high dose PPI for 3 days
rates of rebleeding were overall low (3e14%). after successful endoscopic hemostasis for NVUGIB, in particular
There is no formal consensus regarding the different PPI regi- high risk peptic ulcer bleeding
mens according to international recommendations that favour  Additional data are required to prove non-inferiority, let alone
high-dose PPI infusion post-endoscopic therapy. For patients who equivalence of intermittent (oral/iv) dosing in comparison to
receive endoscopic hemostasis and for patients with an adherent continuous infusion of PPI
clot not receiving endoscopic hemostasis, the 2015 ESGE guidelines  PPI after endoscopy in NVUGIB is a dominant strategy in eco-
suggest considering PPI therapy as intermittent intravenous bolus nomic terms, i.e.: more efficacious and less costly
dosing (at least twice-daily) for the 72 h following endoscopy. If the  In variceal UGIB, only a short 10-day course of a once-daily oral
patient's condition permits, high dose oral PPI may also be an op- PPI following endoscopic banding may be reasonable
tion in those able to tolerate oral medications(16). The NICE
guidelines recommend routine administration of PPI to patients
with NVUGIB and stigmata of recent hemorrhage shown at Concerns associated with PPI use
endoscopy, but do not specify the route, dosage or duration of
administration [18]. The recent 2018 Asia-Pacific working group Safety concern
consensus on NVUGIB have included intermittent high-dose oral In a recent systematic review, Vaezi et al. summarize the evi-
PPI for 3 days following endoscopic therapy as an approach to dence for the various proposed complications of PPI therapy [41].
prevent rebleeding [19]. The authors found moderate strength of evidence to suggest that
Even if the intermittent regimen warrants further investigation PPI use may be associated with bacterial enteric infections,
before becoming treatment of choice, future guidelines should be including Clostridium difficile [42]. However, the remaining associ-
clearer as to the possible recommended use of high dose PPI infu- ations, including myocardial infarction, hepatic encephalopathy,
sion versus bolus IV dosing, and the possible more widespread hospital-acquired pneumonia and spontaneous bacterial perito-
adoption of high dose oral PPIs, although the highest-quality data nitis were weak and were most likely explained by residual con-
remain those favoring the PPI high-dose IV infusion (8 mg/h for founding due to study design limitations. These data are supported,
72 h). with regards to short-term PPI use, by a recent randomized trial
addressing the value of PPI prophylaxis for gastrointestinal
Cost-effectiveness considerations addressing PPI use after bleeding in the ICU [43]. In this study comparing pantoprazole vs
endoscopic hemostasis placebo, no difference was found in the new-onset rate of adverse
In patients with UGIB who have undergone successful events (pneumonia or Clostridium difficile) but the follow-up was

Please cite this article as: Kherad O et al., Proton pump inhibitors for upper gastrointestinal bleeding, Best Practice & Research Clinical
Gastroenterology, https://doi.org/10.1016/j.bpg.2019.04.002
 O. Kherad et al. / Best Practice & Research Clinical Gastroenterology xxx (xxxx) xxx 5

Fig. 2. Algorithm: Therapeutic management of UGIB with PPI.

only 90 days [43]. Eventually, these drugs seem generally safe with patients or situations in which this use will be more likely to be
a favorable risk-benefit profile, assuming they are being given for an cost-effective so that authoritative recommendations can be
appropriate indication. confidently issued, based on best evidence.

Cost concern Research agenda

In addition to safety concerns, inappropriate overuse of PPI also
has economic repercussions. In USA, the annual cost was $ 14  Cost effectiveness analysis for PPI use prior endoscopy and
billion and $ 24 billion worldwide [44]. In the state of New Jersey, better identify who can benefit
the additional cost due to inappropriate PPI prescribing at one of  Additional trials to assess whether intermittent regimens are
their institutions in 2014 was estimated at approximately $ 1 equivalent to continuous iv PPI use
million [45]. Inappropriate use of PPI is still common even as iv  To determine the optimal dosing regimen once the iv PPI
administration [46]. The main reasons for this overuse of PPI are the regimen is completed in patients with acute NVUGIB undergo-
prevention of gastro-duodenal ulcers in low-risk patients or, ing successful hemostasis
arguably perhaps in stress ulcer prophylaxis [47]. Anemia with no
evidence of digestive bleeding was also identified as an important
inappropriate indication [48]. Reducing inappropriate PPI use has Conflicts of interest
become a priority and is one of the privileged themes of the
Choosing Wisely campaign across the world that seeks to help None.
physicians and patients engaging in conversations about unnec-
essary tests, treatments and procedures [49]. A PPI deprescribing References
tool uploaded into an electronic medical record of a Canadian
[1] Lewis JD, Bilker WB, Brensinger C, Farrar JT, Strom BL. Hospitalization and
Hospital has been studied as a reminder to guide reassessment and
mortality rates from peptic ulcer disease and GI bleeding in the 1990s: rela-
deprescribing. This tool permitted to reassess 93% of PPI prescrip- tionship to sales of nonsteroidal anti-inflammatory drugs and acid suppres-
tion, resulting in 26% of patients having their PPI safely and suc- sion medications. Am J Gastroenterol 2002;97(10):2540e9.
[2] Abougergi MS, Travis AC, Saltzman JR. The in-hospital mortality rate for upper
cessfully deprescribed [50].
GI hemorrhage has decreased over 2 decades in the United States: a nation-
wide analysis. Gastrointest Endosc 2015;81(4):882e888 e1.
[3] Peery AF, Crockett SD, Barritt AS, Dellon ES, Eluri S, Gangarosa LM, et al.
Conclusion Burden of gastrointestinal, liver, and pancreatic diseases in the United States.
Gastroenterology 2015;149(7):1731e17341 e3.
Adherence to evidence-based practice and ensuring robust [4] Longstreth GF. Epidemiology and outcome of patients hospitalized with acute
lower gastrointestinal hemorrhage: a population-based study. Am J Gastro-
quality assurance of UGIB management are crucial for optimizing enterol 1997;92(3):419e24.
patient outcomes (Fig. 2). Clear guidelines for adequate PPI use in [5] Laine L. Rolling review: upper gastrointestinal bleeding. Aliment Pharmacol
order to standardize best practices are urgenty required. Although Ther 1993;7(2):207e32.
[6] Neumann I, Letelier LM, Rada G, Claro JC, Martin J, Howden CW, et al. Com-
high-dose PPI after endoscopy is unanimously incorporated into
parison of different regimens of proton pump inhibitors for acute peptic ulcer
guidelines for management of NVUGIB, many gaps persist in our bleeding. Cochrane Database Syst Rev 2013;(6). CD007999.
knowledge of best practice for PPI use prior endoscopy and to [7] Patchett SE, Enright H, Afdhal N, O'Connell W, O'Donoghue DP. Clot lysis by
determine the optimal dosing regimen once the iv regimen is gastric juice: an in vitro study. Gut 1989;30(12):1704e7.
[8] Barkun AN, Cockeram AW, Plourde V, Fedorak RN. Review article: acid sup-
completed. pression in non-variceal acute upper gastrointestinal bleeding. Aliment
We should focus our efforts atbetter identifying subgroups of Pharmacol Ther 1999;13(12):1565e84.

Please cite this article as: Kherad O et al., Proton pump inhibitors for upper gastrointestinal bleeding, Best Practice & Research Clinical
Gastroenterology, https://doi.org/10.1016/j.bpg.2019.04.002
6 O. Kherad et al. / Best Practice & Research Clinical Gastroenterology xxx (xxxx) xxx

[9] Green Jr FW, Kaplan MM, Curtis LE, Levine PH. Effect of acid and pepsin on [30] Cheng HC, Wu CT, Chang WL, Cheng WC, Chen WY, Sheu BS. Double oral
blood coagulation and platelet aggregation. A possible contributor prolonged esomeprazole after a 3-day intravenous esomeprazole infusion reduces
gastroduodenal mucosal hemorrhage. Gastroenterology 1978;74(1):38e43. recurrent peptic ulcer bleeding in high-risk patients: a randomised controlled
[10] Barkun AN, Bardou M, Kuipers EJ, Sung J, Hunt RH, Martel M, et al. Interna- study. Gut 2014;63(12):1864e72.
tional consensus recommendations on the management of patients with [31] Jian Z, Li H, Race NS, Ma T, Jin H, Yin Z. Is the era of intravenous proton pump
nonvariceal upper gastrointestinal bleeding. Ann Intern Med 2010;152(2): inhibitors coming to an end in patients with bleeding peptic ulcers? Meta-
101e13. analysis of the published literature. Br J Clin Pharmacol 2016;82(3):880e9.
[11] Laine L. CLINICAL PRACTICE. Upper gastrointestinal bleeding due to a peptic [32] Tringali A, Manta R, Sica M, Bassotti G, Marmo R, Mutignani M. Comparing
ulcer. N Engl J Med 2016;374(24):2367e76. intravenous and oral proton pump inhibitor therapy for bleeding peptic ulcers
[12] Strand DS, Kim D, Peura DA. 25 Years of proton pump inhibitors: a compre- following endoscopic management: a systematic review and meta-analysis.
hensive review. Gut Liver 2017;11(1):27e37. Br J Clin Pharmacol 2017;83(8):1619e35.
[13] Sachar H, Vaidya K, Laine L. Intermittent vs continuous proton pump inhibitor [33] Tsoi KK, Hirai HW, Sung JJ. Meta-analysis: comparison of oral vs. intravenous
therapy for high-risk bleeding ulcers: a systematic review and meta-analysis. proton pump inhibitors in patients with peptic ulcer bleeding. Aliment
JAMA Intern Med 2014;174(11):1755e62. Pharmacol Ther 2013;38(7):721e8.
[14] Lau JY, Leung WK, Wu JC, Chan FK, Wong VW, Chiu PW, et al. Omeprazole [34] Leontiadis GI, Barkun AN. Commentary: what is the optimal PPI dosing
before endoscopy in patients with gastrointestinal bleeding. N Engl J Med following endoscopic haemostasis in acute ulcer bleeding? Aliment Pharma-
2007;356(16):1631e40. col Ther 2012;35(11):1351e2. author reply 2.
[15] Sreedharan A, Martin J, Leontiadis GI, Dorward S, Howden CW, Forman D, [35] Javid G, Zargar SA, R.U.S, Khan BA, Yatoo GN, Shah AH, et al. Comparison of p.o.
et al. Proton pump inhibitor treatment initiated prior to endoscopic diagnosis or i.v. proton pump inhibitors on 72-h intragastric pH in bleeding peptic ulcer.
in upper gastrointestinal bleeding. Cochrane Database Syst Rev 2010;(7). J Gastroenterol Hepatol 2009;24(7):1236e43.
CD005415. [36] Barkun AN, Herba K, Adam V, Kennedy W, Fallone CA, Bardou M. High-dose
[16] Gralnek IM, Dumonceau JM, Kuipers EJ, Lanas A, Sanders DS, Kurien M, et al. intravenous proton pump inhibition following endoscopic therapy in the
Diagnosis and management of nonvariceal upper gastrointestinal hemor- acute management of patients with bleeding peptic ulcers in the USA and
rhage: european society of gastrointestinal endoscopy (ESGE) guideline. Canada: a cost-effectiveness analysis. Aliment Pharmacol Ther 2004;19(5):
Endoscopy 2015;47(10):a1e46. 591e600.
[17] Laine L, Jensen DM. Management of patients with ulcer bleeding. Am J Gas- [37] Erstad BL. Cost-effectiveness of proton pump inhibitor therapy for acute
troenterol 2012;107(3):345e60. quiz 61. peptic ulcer-related bleeding. Crit Care Med 2004;32(6):1277e83.
[18] nice.org.uk/guidance/cg141. Acute upper gastrointestinal bleeding in over [38] Lee KK, You JH, Wong IC, Kwong SK, Lau JY, Chan TY, et al. Cost-effectiveness
16s: management [Available from: nice.org.uk/guidance/cg141. analysis of high-dose omeprazole infusion as adjuvant therapy to endoscopic
[19] Sung JJ, Chiu PC, Chan FKL, Lau JY, Goh KL, Ho LH, et al. Asia-Pacific working treatment of bleeding peptic ulcer. Gastrointest Endosc 2003;57(2):160e4.
group consensus on non-variceal upper gastrointestinal bleeding: an update [39] Lo EA, Wilby KJ, Ensom MH. Use of proton pump inhibitors in the manage-
2018. Gut 2018;67(10):1757e68. ment of gastroesophageal varices: a systematic review. Ann Pharmacother
[20] Tsoi KK, Lau JY, Sung JJ. Cost-effectiveness analysis of high-dose omeprazole 2015;49(2):207e19.
infusion before endoscopy for patients with upper-GI bleeding. Gastrointest [40] Kang SH, Yim HJ, Kim SY, Suh SJ, Hyun JJ, Jung SW, et al. Proton pump inhibitor
Endosc 2008;67(7):1056e63. therapy is associated with reduction of early bleeding risk after prophylactic
[21] Al-Sabah S, Barkun AN, Herba K, Adam V, Fallone C, Mayrand S, et al. Cost- endoscopic variceal band ligation: a retrospective cohort study. Medicine
effectiveness of proton-pump inhibition before endoscopy in upper gastro- (Baltim) 2016;95(8):e2903.
intestinal bleeding. Clin Gastroenterol Hepatol 2008;6(4):418e25. [41] Vaezi MF, Yang YX, Howden CW. Complications of proton pump inhibitor
[22] Leontiadis GI, Sharma VK, Howden CW. Proton pump inhibitor therapy for therapy. Gastroenterology 2017;153(1):35e48.
peptic ulcer bleeding: Cochrane collaboration meta-analysis of randomized [42] Leonard J, Marshall JK, Moayyedi P. Systematic review of the risk of enteric
controlled trials. Mayo Clin Proc 2007;82(3):286e96. infection in patients taking acid suppression. Am J Gastroenterol 2007;102(9):
[23] Barkun AN. Should every patient with suspected upper GI bleeding receive a 2047e56. quiz 57.
proton pump inhibitor while awaiting endoscopy? Gastrointest Endosc [43] Krag M, Marker S, Perner A, Wetterslev J, Wise MP, Schefold JC, et al. Pan-
2008;67(7):1064e6. toprazole in patients at risk for gastrointestinal bleeding in the ICU. N Engl J
[24] Leontiadis GI, Sharma VK, Howden CW. Proton pump inhibitor treatment for Med 2018;379:2199e208.
acute peptic ulcer bleeding. Cochrane Database Syst Rev 2006;(1):CD002094. [44] Sheen E, Triadafilopoulos G. Adverse effects of long-term proton pump in-
[25] Lau JY, Sung JJ, Lee KK, Yung MY, Wong SK, Wu JC, et al. Effect of intravenous hibitor therapy. Dig Dis Sci 2011;56(4):931e50.
omeprazole on recurrent bleeding after endoscopic treatment of bleeding [45] Ladd AM, Panagopoulos G, Cohen J, Mar N, Graham R. Potential costs of
peptic ulcers. N Engl J Med 2000;343(5):310e6. inappropriate use of proton pump inhibitors. Am J Med Sci 2014;347(6):
[26] Laine L, McQuaid KR. Endoscopic therapy for bleeding ulcers: an evidence- 446e51.
based approach based on meta-analyses of randomized controlled trials. [46] Craig DG, Thimappa R, Anand V, Sebastian S. Inappropriate utilization of
Clin Gastroenterol Hepatol 2009;7(1):33e47. quiz 1-2. intravenous proton pump inhibitors in hospital practice–a prospective study
[27] Leontiadis GI, Sreedharan A, Dorward S, Barton P, Delaney B, Howden CW, of the extent of the problem and predictive factors. QJM 2010;103(5):327e35.
et al. Systematic reviews of the clinical effectiveness and cost-effectiveness of [47] Savarino V, Dulbecco P, de Bortoli N, Ottonello A, Savarino E. The appropriate
proton pump inhibitors in acute upper gastrointestinal bleeding. Health use of proton pump inhibitors (PPIs): need for a reappraisal. Eur J Intern Med
Technol Assess 2007;11(51):1e164. iii-iv. 2017;37:19e24.
[28] Jensen DM, Eklund S, Persson T, Ahlbom H, Stuart R, Barkun AN, et al. Reas- [48] Chia CT, Lim WP, Vu CK. Inappropriate use of proton pump inhibitors in a local
sessment of rebleeding risk of forrest IB (oozing) peptic ulcer bleeding in a setting. Singap Med J 2014;55(7):363e6.
large international randomized trial. Am J Gastroenterol 2017;112(3):441e6. [49] choosingwisely.org.
[29] Klok RM, Postma MJ, van Hout BA, Brouwers JR. Meta-analysis: comparing the [50] Walsh K, Kwan D, Marr P, Papoushek C, Lyon WK. Deprescribing in a family
efficacy of proton pump inhibitors in short-term use. Aliment Pharmacol Ther health team: a study of chronic proton pump inhibitor use. J Prim Health Care
2003;17(10):1237e45. 2016;8(2):164e71.

Please cite this article as: Kherad O et al., Proton pump inhibitors for upper gastrointestinal bleeding, Best Practice & Research Clinical
Gastroenterology, https://doi.org/10.1016/j.bpg.2019.04.002