Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

Pre and post endoscopic use of IV proton pump inhibtors in acute non-variceal upper gastrointestinal bleeds, bolus vs infusion

Three Part Question

In [acute non variceal upper GI bleed] is [ continuous infusion or bolus infusions of iv proton pump inhibitor better] at [reducing mortality] pre-endoscopy

Clinical Scenario

A 45 years old man with known history of peptic ulcer disease is admitted with haematemesis and malaena. The is no history or stigmata of chronic liver disease to suspect varices. You want to know whether giving proton pump inhibitors (PPI) as intermittent intravenous (IV) bolus dose is as effective as as a continuous IV infusion pre endoscopy.

Search Strategy

Pubmed search
Date: all to present

((((((("proton pump inhibitors"[MeSH Terms] OR ("proton"[All Fields] AND "pump"[All Fields] AND "inhibitors"[All Fields]) OR "proton pump inhibitors"[All Fields]) OR ("lansoprazole"[Substance Name] OR "lansoprazole"[All Fields])) OR ("pantoprazole"[Substance Name] OR "pantoprazole"[All Fields])) OR ("omeprazole"[MeSH Terms] OR "omeprazole"[All Fields] OR "esomeprazole"[All Fields])) OR ("omeprazole"[MeSH Terms] OR "omeprazole"[All Fields])) AND ("ulcer"[MeSH Terms] OR "ulcer"[All Fields])) AND infusion[All Fields]) AND bolus[All Fields] AND ("humans"[MeSH Terms] AND English[lang] AND "adult"[MeSH Terms])

Search Outcome

34 papers found. 5 papers were relevant although there were no papers directly comparing PPI continuous infusion agasint PPI intermittent bolus doses pre endoscopy.

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Yüksel I, Ataseven H, Köklü S, Ertuğrul I, Başar O,
Ankara, Turkey
97 patients admitted with non-variceal upper GI bleed over 3 years. All had some form of endoscopic intervention and were subsequently randomised to recieve either a bolus of 80mg Pantoprazole followed by a continuous infusion of 8mg/day for 72 hours or IV bolus infusion of 40mg 12 hourly for 72 hoursProspective randomized studyRebleeding rate No significant difference between the 2 groupsDid not distinguish between H.pylori status. Although there was no significance between the 2 groups. Relatively short follow up time of 3 days.
Duration of stay in hospitalNo significant difference between the 2 groups
Need for total blood transfusionNo significant difference between the 2 groups
Need for surgeryNo significant difference between the 2 groups
Garrido A et al
41 patients randomized into 2 groups. Group 1 recieved pantoprazole 80mg bolus followed by 8mg/hr continuous infusion for 72 hours and group 2 recieved daily bolus of pantoprazole 40mg for 72 hours.Prospective randomised studyRecurrance of bleeding no significant difference between the 2 groupsRelatively small numbers. Not much detail on endoscopic therapies performed. Some may argue that it was not originally in English.
Blood transfusion requirementsno significant difference between the 2 groups
Need for surgeryno significant difference between the 2 groups
Mortalityno significant difference between the 2 groups
Wai Ka Hung et al
Hong Kong
153 patients post successful endoscopic therapy for bleeding peptic ulcers were randomised to receive either pantoprazole 80mg IV bolus followed by 8mg/hr infusion, pantoprazole 80mg IV bolus followed by 40mg IV bolus every 12 hours or no treatment. Treatments were conducted for 72 hours.Prospective randomised trial. Rebleeding within 30 days, Fewer in bolus group but no difference between infusion or intermittent bolusNo placebo. Not blinded. Sub groups small numbers.
Transfusion requirementsFewer in bolus group but no difference between infusion or intermittent bolus
MortalityNo difference between groups
Duration of Hospital StayNo difference between groups
Need for operation interventionNo difference between groups
Gastric pHpH was higher in infusion compared to intermittent bolus but no statistical significance
Choi KD et al
August 2009
South Korea
61 patients with bleeding ulcers were randomized into one of three groups after endoscopic hemostasis: pantoprazole 80 mg bolus followed by 8mg/h; 40mg, 4mg/h infusion; and bolus injection of 40mg every 24hProspective, randomised study. No mention of blinding. Over period of 4 years.Rebleeding rates within 30 daysNo significant difference between groupsSolely Asian population. Unblinded.
Mean percentage time of patients with intragastric pH >6A continuous infusion, regardless of high or low dose, was more effective for acid suppression than a 40 mg bolus PPI injection in Korea. H. pylori infection was an important factor for the maintenance of an intragastric pH > 6.
Angelo Andriulli, M.D.; Silvano Loperfido, M.D et al
January 2009
A total of 482 patients with bleeding ulcers which received endoscopic haemostasis were randomised into 2 grouos. Group 1 received a intensive PPI regime; 80mg IV Bolus followed by 8mg/hr for 72 hours. Groupd 2 received a standard regime; 40mg bolus daily followed by saline infusion for 72hrs. Both pantoprazole and omeprazole were used in both groups.Prospective, randomized, double-blind study.Rebleeding ratesNo significant differences between the groupsNo significant weakness.
Mean units of blood transfusedNo significant differences between the groups
Duration of hospital stayNo significant differences between the groups
Need for surgical interventionNo significant differences between the groups


PPI treatment initiated for non-variceal gastrointestinal bleed is shown to reduce proportion of patients with stigmata of recent haemmorhage at endscopy and reduces the requirement for endoscopy therapy. However, there is no evidence that PPI treatment affects mortality, need for blood transfusion, rebleeding rates or need for surgery. These results were also mirrored in studies whereby PPI was started post endoscopic haemostasis. IV PPI provides more rapid increase in pH, reaching mean pH of 6 approximately 1 hour sooner than oral PPI.

Clinical Bottom Line

There were no papers directly comparing benefits of continuous infusion PPI against intermittent IV bolus PPI pre-endoscopy. In the papers that compared IV PPI regimes post-endoscopic haemostasis, there was no significant difference in outcomes between the continuous IV infusion and intermittent IV bolus group.


  1. Yüksel I, Ataseven H, Köklü S, Ertuğrul I, Başar O, Odemiş B, Ibiş M, Saşmaz N, Sahin B Intermittent versus continuous pantoprazole infusion in peptic ulcer bleeding: a prospective randomized study Digestion 2008;78(1):39-43. Epub 2008 Sep 29
  2. Garrido A, Giráldez A, Trigo C, Leo E, Guil A, Márquez JL. Intravenous proton-pump inhibitor for acute peptic ulcer bleeding--is profound acid suppression beneficial to reduce the risk of rebleeding? Rev Esp Enferm Dig. 2008 Aug;100(8):466-9.
  3. Hung WK, Li VK, Chung CK, Ying MW, Loo CK, Liu CK, Lam BY, Chan MC. Randomized trial comparing pantoprazole infusion, bolus and no treatment on gastric pH and recurrent bleeding in peptic ulcers ANZ Journal of Surgery ANZ J Surg. 2007 Aug;77(8):677-81
  4. Choi KD, Kim N, Jang IJ, Park YS, Cho JY, Kim JR, Shin JM, Jung HC, Song IS. Optimal dose of intravenous pantoprazole in patients with peptic ulcer bleeding requiring endoscopic hemostasis in Korea. J Gastroenterol Hepatol. 2009 Oct;24(10):1617-24. Epub 2009 Aug 3
  5. Angelo Andriulli, M.D.; Silvano Loperfido, M.D.; Rosaria Focareta, M.D.; Pietro Leo, M.D.; Fabio Fornari, M.D.; Antonietta Garripoli, M.D.; Paolo Tonti, M.D.; Sergio Peyre, M.D.; Antonio Spadaccini, M High- Versus Low-Dose Proton Pump Inhibitors After Endoscopic Hemostasis in Patients With Peptic Ulcer Bleeding: A Multicentre, Randomized Study The American Journal of Gastroenterology The American Journal of Gastroenterology. 2008;103(12):3011-3018