Endoscopic Hemostatic Treatment Research Articles

High-dose versus low-dose intravenous proton pump inhibitor treatment for bleeding peptic ulcers

Evaluation of: Liu N, Liu L, Zhang H et al. Effect of intravenous proton pump inhibitor regimens and timing of endoscopy on clinical outcomes of peptic ulcer bleeding. J. Gastroenterol. Hepatol. 27(9), 1473–1479 (2012).Peptic ulcer bleeding is a common medical emergency associated with significant mortality and healthcare costs. All recent guidelines agree on the beneficial role of proton pump inhibitor treatment, but there is still controversy regarding the optimal dose and route of administration of proton pump inhibitors. The evaluated article reports on a large, single-center randomized controlled trial that compared the clinical efficacy of a low-dose twice-daily intravenous bolus regimen with a high-dose continuous intravenous infusion regimen in 875 patients with acute bleeding from peptic ulcers. The high-dose regimen was associated with significant reductions in rebleeding, blood transfusion requirements and length of hospital stay. There was no demonstrable difference in mortality or the need for endoscopic hemostatic treatment or surgery. We discuss the strengths and limitations of the evaluated article, as well as the implications for clinical practice.

Prospective randomized controlled trial to compare the effects of omeprazole and famotidine in preventing delayed bleeding and promoting ulcer healing after endoscopic submucosal dissection

Proton pump inhibitors (PPIs) are generally used to prevent delayed bleeding after endoscopic submucosal dissection (ESD) and to heal the artificial ulcers. However, it remains controversial whether PPIs or histamine-2 receptor antagonists (H(2) RAs) are more effective in preventing delayed bleeding after ESD. We prospectively compared the effects of omeprazole and famotidine in preventing delayed bleeding and promoting artificial ulcer healing after ESD. A total of 158 patients (155 early gastric cancers and three adenomas) were randomly assigned to the PPI group (omeprazole 20 mg/day) or H(2) RA group (famotidine 40 mg/day) in a prospective randomized controlled trial. The primary end point was the incidence of hematemesis, melena, and/or a decrease in hemoglobin level of 2 g/dL or more requiring endoscopic hemostatic treatment. ESD-induced ulcer healing and changes in ulcer size were also compared at 6 weeks after ESD as a secondary end point. Of the 158 patients, two were excluded from analysis because they had been treated with a PPI before the present study. Accordingly, data from 77 PPI and 79 H(2) RA subjects were included for analysis. Delayed bleeding after ESD occurred in 6.5% of subjects (PPI group) and in 6.3% (H(2) RA group); there was no significant difference between the two groups. Likewise, the two groups were not significantly different with respect to ulcer stage or ulcer size reduction rate. Proton pump inhibitors are not superior to H(2) RAs for the prevention of delayed bleeding or the healing of artificially induced ulcers after ESD.

Predictive Factors for Colonic Diverticular Rebleeding: A Retrospective Analysis of the Clinical and Colonoscopic Features of 111 Patients

Background/AimsColonic diverticular bleeding can stop spontaneously or be stopped by endoscopic hemostasis. We analyzed the clinical and colonoscopic features of patients with colonic diverticular bleeding to establish the predictive factors for rebleeding.MethodsA total of 111 patients (median age, 72 years) with colonic diverticular bleeding in Aso Iizuka Hospital between April 2007 and July 2010 were enrolled. Age, sex, body mass index (BMI), comorbidity, medication, location of bleeding, colonoscopic findings and hemostatic methods were analyzed retrospectively from the hospital records.ResultsThe most common sites of bleeding were the ascending (39.6%) and sigmoid (29.7%) colon. Overt rebleeding occurred in 30 patients (27.0%). Spontaneous hemostasis was seen in 81 patients (73.0%), and endoscopic hemostatic treatment was performed in 30 patients. The BMI in the patients with colonic diverticular rebleeding was significantly higher than in patients without rebleeding. Colonoscopic findings of actively bleeding or nonbleeding visible vessels in the responsible diverticula were more frequent in the group with rebleeding.ConclusionsA higher BMI and colonoscopic findings of actively bleeding or nonbleeding visible vessels can be used as predictive factors for colonic diverticular rebleeding. Patients with such findings should be carefully followed up after hemostasis of the initial colonic diverticular bleeding.

Osler-Weber-Rendu Syndrome: A Rare Cause of Iron Deficiency

A 52-year-old man with frequent epistaxis presented with fatigue and exertional dyspnea. On physical examination, multiple telangiectasias were noted on the patient’s face, lips, tongue (Fig. 1), and finger tips (Fig. 2). Laboratory analysis demonstrated iron deficiency anemia (hemoglobin, 8.2 g/dl; ferritin: 15 ng/ml). Colonoscopy revealed multiple colonic telengiectasias, and endoscopic hemostatic treatment was performed on bleeding lesions. Figure 1 Multiple telangiectasias on the patient’s face, lips, and tongue. Figure 2 Multiple telangiectasias on the patient’s finger tips. Osler-Weber-Rendu syndrome (OWR), also known as hereditary hemorrhagic telangiectasia, is an autosomal dominant disease that affects all ethnic and racial groups with an overall frequency of 1 per 5,000 to 10,000 persons. Although the number and location of lesions vary widely, most telangiectasias are found in the oral, nasal, and gastrointestinal mucosa, as well as the fingertips, whereas arteriovenous malformation occurs most commonly in the lungs, liver, and central nervous system. The prevalence of intestinal telangiectasia varies from 10% to 33% in patients with OWR. Approximately 25% of individuals older than 60 years present with melena or anemia. Endoscopic ablation using electrocoagulation or laser techniques is useful for control of gastrointestinal bleeding in the short term.1

WITHDRAWN: Proton pump inhibitor treatment for acute peptic ulcer bleeding.

Randomised controlled trials (RCTs) evaluating the clinical effect of proton pump inhibitors (PPIs) in peptic ulcer (PU) bleeding yield conflicting results. To evaluate the efficacy of PPIs in acute bleeding from PU using evidence from RCTs. We searched CENTRAL, The Cochrane Library (Issue 4, 2004), MEDLINE (1966 to November 2004), EMBASE (1980 to November 2004), proceedings of major meetings to November 2004, and reference lists of articles. We contacted pharmaceutical companies and experts in the field. RCTs of PPI treatment (oral or intravenous) compared with placebo or H(2)-receptor antagonist (H(2)RA) in acute bleeding from PU. Two reviewers extracted data independently, assessed study validity, summarised studies and undertook meta-analysis. The influence of study characteristics on the outcomes was examined by subgroup analyses and meta-regression. Twenty-four RCTs comprising 4373 participants in total were included. Statistical heterogeneity was found among trials for rebleeding (P = 0.04), but not for all-cause mortality (P = 0.24) or surgery (P = 0.45). There was no significant difference in all-cause mortality rates between PPI and control treatment; pooled rates were 3.9% on PPI versus 3.8% on control (odds ratio (OR) 1.01; 95% CI 0.74 to 1.40). PPIs significantly reduced rebleeding compared to control; pooled rates were 10.6% with PPI versus 17.3% with control treatment (OR 0.49; 95% CI 0.37 to 0.65). PPI treatment significantly reduced surgery compared with control; pooled rates were 6.1% on PPI versus 9.3% on control (OR 0.61; 95% CI 0.48 to 0.78). There was no evidence to suggest that results on mortality and rebleeding were dependent on study quality, route of PPI administration, type of control treatment or application of initial endoscopic haemostatic treatment. PPIs significantly reduced surgery compared with placebo but not when compared with H(2)RA. There was no evidence to suggest that study quality, route of PPI administration or application of initial endoscopic haemostatic treatment influenced results on surgery. PPI treatment appeared more efficacious in studies conducted in Asia compared to studies conducted elsewhere. All-cause mortality was reduced only in Asian studies; reductions in rebleeding and surgery were quantitatively greater in Asian studies. Among patients with active bleeding or non-bleeding visible vessel, PPI treatment reduced mortality (OR 0.53; 95% CI 0.31 to 0.91), rebleeding and surgery. PPI treatment in PU bleeding reduces rebleeding and surgery compared with placebo or H(2)RA, but there is no evidence of an overall effect on all-cause mortality.

Rectorragie abondante tardive après biopsies de prostate : à propos d’un cas

The biopsy of the prostate is a common medical act, which is little invasive and can easily be practiced in external care. Some cases of early rectal or urinary bleedings, which mainly stopped spontaneously, have been described in the literature. The case reported here is that of a patient whose hemorrhagic syndrome arose more than two weeks after the biopsy and required an endoscopic haemostatic treatment.

Immediate‐release oral proton pump inhibitors: Panacea or a flash in the pan?

Use of proton pump inhibitors (PPI) after endoscopic hemostatic treatment of bleeding peptic ulcers is a standard therapy for preventing early re-bleeding.1 However, controversy continues regarding the optimal dose and route of administration of PPI.2 While both i.v. and oral PPI are effective in preventing early re-bleeding, meta-analysis of randomized trials found that i.v. infusion of high-dose PPI is superior to low-dose (i.e. oral or repeated i.v. injections) PPI in terms of the need for surgery.1 Oral PPI have two major limitations: a slow onset of action and failure to maintain a stably high intragastric pH. One attempt to overcome the limitations of conventional oral PPI is the development of immediate-release (IR) formulations. In this issue, Banerjee et al.3 shows that healthy volunteers who received buffered IR esomeprazole 40 mg p.o. had superior intragastric pH profile compared to those who received i.v. pantoprazole 40 mg every 12 h for 24 h. After the first dose of buffered IR esomeprazole, the intragastric pH was rapidly raised to 6 in a median time of 2 min whereas it took 80 min for i.v. pantoprazole to achieve this level of intragastric pH. Furthermore, the mean percentage time to an intragastric pH of more than 6.0 was 91% in the buffered IR esomeprazole group compared to 20% in the i.v. pantoprazole group in a 24-h period. While many readers would agree that the intragastric pH profile achieved by buffered IR esomeprazole was close to perfection, was the result too good to be true? The rapid rise in intragastric pH with buffered IR esomeprazole was not unexpected because the bicarbonate content of the formulation would neutralize gastric acid. Similar findings were also reported in an earlier study with buffered IR omeprazole.4 In the previous study, however, IR omeprazole rarely raised the intragastric pH above 6 (reported median 24-h intragastric pH was 3.7 only).4 By contrast, the reported median 24-h intragastric pH was 6.18 with IR esomeprazole in the current study.3 In fact, the pH profile achieved by IR esomeprazole was comparable to that with continuous infusion of high-dose PPI.5 What factors could possibly account for such impressive results with just two p.o. doses of IR esomeprazole in 24 h? One could argue that esomeprazole is superior to omeprazole because the former has a better pharmacodynamic profile. The modest advantage of esomeprazole over omeprazole, however, is unlikely to explain the striking difference between the current and earlier studies.3,4 When interpreting the effects of PPI on intragastric pH, one needs to be aware of factors that may influence the outcome. Helicobacter pylori infection is well known to enhance the acid-suppressing effect of PPI.6 Investigators often recognize the importance of excluding H. pylori infection in pH studies. However, a negative rapid urease test does not exclude hypochlorhydria in patients with past H. pylori infection. In Asian countries where the prevalence of H. pylori infection is high, a considerable proportion of the population with hypochlorhydria associated with past H. pylori infection is not unexpected. It would be useful to know the proportion of subjects with hypochlorhydria or histological evidence of gastric atrophy when interpreting pH studies. Another factor is gastric parietal cell mass. PPI are thought to be more effective in Asians because of the smaller parietal cell mass. Small parietal cell mass may partly explain the observation that PPI reduce mortality associated with peptic ulcer bleeding in Asian studies but not in Western studies.1 Presumably, the study by Banerjee et al.3 was conducted in Indian subjects whereas the IR omeprazole study was done on white subjects.4 The efficacy of PPI is also influenced by genetic polymorphism of certain human drug-metabolizing cytochrome P450 (CYP) enzymes. In particular, polymorphism of CYP2C19 has been reported to affect the efficacy of some PPI,7 and substantial racial difference exists in terms of the relative proportions of extensive and poor metabolizers.8 Unlike other PPI such as omeprazole, the CYP2C19 genetic polymorphism is thought to have little influence on the disposition of esomeprazole.9 Surprisingly, a recent study found that the intragastric pH and plasma level of esomeprazole was affected by the CYP2C19 genotype status, and that a multiple dosing regimen of oral esomeprazole improved acid control compared to a single daily regimen.10 In summary, there is little doubt that buffered IR PPI have certain advantages. Whether they are an alternative to i.v. infusion of high-dose PPI in patients with bleeding peptic ulcers can only be addressed by large-scale, head-to-head trials using predefined clinical outcomes. Until further data are available, i.v. infusion of high-dose PPI after endoscopic treatment of bleeding peptic ulcers remains the most studied and best proven strategy.

Clinical and endoscopic characteristics of acute haemorrhagic rectal ulcer, and endoscopic haemostatic treatment: a retrospective study of 95 patients

Acute haemorrhagic rectal ulcer (AHRU) is characterized by sudden onset of painless and massive rectal bleeding in elderly bedridden patients who have serious illness. Endoscopic diagnosis and management of AHRU is, however, still controversial. We retrospectively investigated 95 AHRU patients to elucidate the clinical characteristics, endoscopic findings and haemostatic strategies. Between January 1999 and March 2007, 95 patients were diagnosed with AHRU in our hospital. Medical records and colonoscopy files were reviewed. Clinical features, colonoscopic findings, haemostatic treatment and outcome of the patients were evaluated. Eighty per cent of the patients were bedridden at the onset. The most frequent underlying disorder was cerebrovascular disease (36.8%). Hypoalbuminaemia (< 3.5 g/dl) was seen in 92.6% of the patients. Endoscopic findings of AHRU were classified as circumferential ulcer (41.1%), linear or nearly round small ulcer(s) (44.2%), circumferential and small ulcer(s) (7.4%) and Dieulafoy-like ulcer (7.4%). Primary endoscopic haemostatic treatment was performed in 45.3% of cases. Recurrent bleeding occurred in 24.2% of patients. Permanent haemostasis was achieved by secondary endoscopic treatment in 82.6% of re-bleeding patients. Understanding the typical clinical and endoscopic findings and careful endoscopic examination are important for the accurate diagnosis of AHRU, and endoscopic haemostatic therapy may be effective for bleeding patients.

Pharmacologic treatment of peptic ulcer bleeding

Over the last 3 decades, there has been extensive clinical research on the pharmacologic treatment of peptic ulcer bleeding. A critical review of randomized controlled trials and meta-analyses reveals insufficient evidence to recommend histamine-2 receptor antagonists (H(2)RAs), somatostatin, octreotide, or tranexamic acid in the routine management of patients with peptic ulcer bleeding. In contrast, there is good-quality evidence for recommending proton-pump inhibitor (PPI) treatment for patients with peptic ulcer bleeding. PPI treatment, compared with an H(2)RA or placebo, reduces rebleeding and the need for surgical intervention and, in patients with high-risk endoscopic stigmata, also reduces all-cause mortality. Patients with ulcers that demonstrate only low-risk endoscopic stigmata (clean base or flat pigmented spot) can be treated with an oral PPI at double the standard clinical dose. Patients with ulcers that demonstrate high-risk endoscopic stigmata (spurting, oozing, or nonbleeding visible vessel) should receive high-dose intravenous PPI treatment following appropriate endoscopic hemostatic treatment. The currently recommended dose is an initial intravenous bolus equivalent to 80 mg of omeprazole followed by an intravenous infusion equivalent to 8.0 mg/h of omeprazole for up to 72 hours. A switch to high-dose oral PPI treatment may be appropriate before completion of a 72-hour treatment period in some patients whose clinical status stabilizes early. Once the initial bleeding episode has been dealt with, patients will require standard pharmacologic treatment to heal the ulcer and prevent recurrence.

To Establish the Efficacy of PPI Therapy for Ulcer Bleeding in the United States, Do We Need More Patients or More PPIs?

Proton pump inhibitor (PPI) therapy is widely used in the management of patients with ulcer bleeding. However, most randomized controlled trials (RCTs) have been conducted outside of the United States. Jensen and colleagues have conducted the first United States-based RCT to compare a PPI with an H(2)-receptor antagonist following appropriate endoscopic hemostatic treatment for patients with ulcer bleeding. Unfortunately, the trial had to be discontinued prematurely because of problems with enrolment. There was a statistically nonsignificant difference in rebleeding rates between the two treatment arms. A type II statistical error probably explains this apparent lack of efficacy. However, U.S. patients may also require higher doses of PPI for adequate control of intragastric acidity.

Proton pump inhibitor treatment for acute peptic ulcer bleeding.

Peptic ulcer (PU) bleeding is associated with substantial morbidity, mortality and healthcare cost. Randomised controlled trials (RCTs) evaluating the clinical effect of proton pump inhibitors (PPIs) in peptic ulcer bleeding have yielded conflicting results. To evaluate the efficacy of PPIs in the management of acute bleeding from PU using evidence from RCTs. We performed a search of CENTRAL, The Cochrane Library (Issue 3, 2003), MEDLINE (1966 to February 2003) and EMBASE (1980 to February 2003) and proceedings of recent major meetings through to February 2003. We searched the reference lists of articles and contacted pharmaceutical companies and experts in the field for additional published or unpublished data. RCTs of PPI treatment (oral or intravenous) compared with either placebo or H(2)-receptor antagonist (H(2)RA) in patients with acute bleeding from PU were included if they met pre-defined criteria. Two reviewers extracted data independently on a purpose-designed data extraction form. Validity of included studies was assessed by adequacy of randomisation method and other pre-defined criteria. Studies were summarised and meta-analysis was undertaken. The influence of factors on the outcomes was assessed. Twenty-one RCTs with a total of 2915 participants were included. Statistical heterogeneity was found among trials for rebleeding (P = 0.05), but not for mortality (P = 0.26) or surgery (P = 0.42). There was no significant difference in mortality rates between PPI and control treatment; pooled rates were 5.2% on PPI versus 4.6% on control (odds ratio (OR) 1.11; 95% CI 0.79 to 1.57). PPI treatment significantly reduced rates of surgical intervention compared with control; pooled rates were 8.4% on PPI versus 13.0% on control (OR 0.59; 95% CI 0.46 to 0.76). PPIs significantly reduced rebleeding compared to control; pooled rates were 10.6% with PPI (range: 0% to 24.4%) versus 18.7% with control treatment (range: 2.3% to 39.1%), the OR was 0.46 (95% CI 0.33 to 0.64). Results on mortality and rebleeding rates were independent of route of PPI administration, type of control treatment or application of initial endoscopic haemostatic treatment. Surgical intervention rates varied with type of control (PPI significantly reduced surgical intervention rates compared with placebo and not when compared with H(2)RA) but not with route of PPI administration or application of initial endoscopic haemostatic treatment. PPI treatment in PU bleeding reduces rebleeding and surgical intervention rates in studies comparing treatment with placebo or H(2)RA, but there is no evidence of an effect on mortality.

Which types of non-bleeding visible vessels in gastric peptic ulcers should be treated by endoscopic hemostasis?

Because non-bleeding visible vessels (NBVV) of gastric peptic ulcers have the potential to re-bleed, endoscopic hemostatic treatment may be necessary during the first emergency endoscopy. However, not all NBVV re-bleed, and endoscopic hemostasis sometimes causes fatal side-effects. Therefore, we have evaluated the risk of re-bleeding from various NBVV in gastric peptic ulcers to determine which types should be treated by endoscopy to prevent re-bleeding. A total of 227 NBVV in 202 patients with gastric peptic ulcers that were endoscopically followed without endoscopic hemostatic procedures were classified by the following factors: vessel color, form, location of the NBVV in the ulcer crater, and location of the ulcer in the stomach. The re-bleeding rate was then analyzed for each type of NBVV. Significantly high rates of re-bleeding were observed in cases with white, protruded and peripheral NBVV. In particular, white NBVV located in the peripheral zone of the ulcer crater were frequent re-bleeding sources. The location of the ulcer in the stomach was not a statistically significant factor in determining re-bleeding rates. We found that white, protruded and peripherally located NBVV in gastric ulcers have a higher chance of re-bleeding if preventive endoscopic hemostatic procedures are not performed.

A randomized trial comparing heater probe plus thrombin with heater probe plus placebo for bleeding peptic ulcer

Background & Aims: This multicenter, double-blind, controlled trial compared the efficacy of combined endoscopic hemostatic treatment using the heater probe plus thrombin injection with that of the heater probe plus placebo injection as treatment for peptic ulcers with active bleeding or nonbleeding visible vessels. Efficacy was defined in terms of primary hemostasis, prevention of rebleeding, and need for urgent surgery. Methods: Two hundred forty-seven patients presenting with major peptic ulcer bleeding were randomized to heater probe plus thrombin or to heater probe plus placebo. The groups were well matched for all risk categories including age, endoscopic stigmata, shock, and severity of comorbid diseases. Endoscopic therapy was applied using the heater probe followed by injection of thrombin or placebo. Results: Successful primary hemostasis was achieved in 97% of patients. Rebleeding developed in 19 (15%) of thrombin plus heater probe patients and 17 (15%) of placebo plus heater probe patients. Emergency surgery was necessary in 16 and 13 patients, respectively. Eight patients in the thrombin group had adverse events compared with 4 in the placebo group. Eight (6%) of thrombin plus heater probe patients and 14 (12%) of placebo plus heater probe patients died ( P = 0.21). Conclusions: The combination of thrombin and the heater probe does not confer an additional benefit over heater probe and placebo as endoscopic treatment for bleeding peptic ulcer. Our trial does not support the use of this combination of hemostatic therapy.

Endoscopic treatment of upper GI hemorrhage with a novel irrigating hood attached to the endoscope

Despite advances in endoscopic treatment methods for upper GI hemorrhage, hemostasis is often difficult to achieve, particularly when the endoscopic view at the site of hemorrhage is poor because of the presence of mucus and blood. The investigators developed an end hood that facilitates endoscopic hemostatic procedures while simultaneously allowing irrigation of the bleeding site. The usefulness of this end hood for treatment of upper GI hemorrhage, excluding hemorrhage from varices, was evaluated. The end-hood piece was fabricated by drilling a side hole in the cap portion of a conventional transparent hood. An irrigation tube then was glued to the exterior surface of the hole. The fabricated transparent hood was placed on the tip of an endoscope. With the hood piece in place, hemoclip placement and other endoscopic hemostatic procedures were performed in 15 patients with nonvariceal upper GI hemorrhage. Hemostasis was successfully achieved in all cases. In all cases of active hemorrhage, hemostatic treatment was enhanced by simultaneous irrigation beneath the hood. The median time required for the hemostatic procedure with the attached hood was 4.8 minutes. The end-hood irrigation device and technique facilitate endoscopic hemostatic treatment of nonvariceal upper GI hemorrhage.

7141 Endoscopic treatment of hemorrhagic gastric ulcer in patients aged 80 years or more.

Background: Hemorrhagic gastric ulcer is a more serious disorder in elderly patients than in young patients. Emergency surgery in elderly patients carries high risk. Thus, less invasive treatment by endoscopy may be advantageous. This study aimed to evaluate the efficacy of endoscopic treatment for hemorrhagic gastric ulcer in patients aged 80 or more. Methodology: A total of 269 patients underwent endoscopic hemostatic treatment for hemorrhagic gastric ulcer over a four year period. As hemostatic measures, hemoclipping and pure ethanol injection, either alone or in combination, were employed. These patients were divided into two groups: elderly (≥80 years old, n=26) and younger (

Effect of Omeprazole on the Outcome of Endoscopically Treated Bleeding Peptic Ulcers Randomized Double-Blind Placebo-Controlled Multicentre Study

Background: Haemostasis is highly pH-dependent and severely impaired at low pH. However, there is no clear evidence that acid-suppressing drugs have beneficial effects in peptic ulcer haemorrhage. Endoscopic haemostatic treatment provides important reduction in morbidity and may be more efficient when a neutral intragastric pH is maintained. Methods: We conducted a double-blind, placebo-controlled multicentre study of intravenous infusion of omeprazole (80 mg as bolus, followed by 8 mg/h) or placebo for 72 h. All patients received 20 mg omeprazole orally from day 3 until follow-up on day 21. Only patients with ulcer haemorrhage, endoscoped within 12 h after admission, and with a history or signs of circulatory failure and spurting bleeding, oozing bleeding, visible vessel, or clot, were included. Endoscopic intervention was aimed at when spurting bleeding, oozing bleeding, or a visible vessel was observed. The primary efficacy measure was the worst ranking on an overall outcome scale (5 = death, 4 = surgery, 3 = additional endoscopic treatment, 2 = more than 3 units of blood, and 1 = no more than 3 units of blood transfused). Base-line prognostic factors of treatment success by day 3 and of other binary outcomes were considered in a logistic regression model. Results: Two hundred and seventy-four patients were randomly assigned to omeprazole (134 patients) or placebo (140 patients). The number of patients included in the ‘intention-to-treat’ analysis was 130 in the omeprazole group and 135 in the placebo group. The primary variable, the overall outcome at 72 h, showed a difference (P = 0.004) between the two treatments in favour of omeprazole. Treatment success by 72 h defined as no death, no operation, or no additional endoscopic treatment was 91.0% in the omeprazole group and 79.7% in the placebo group (therapeutic gain, 11.3 percentage units; 95% confidence interval, 2.3 to 20.4 percentage units). Significant differences in favour of omeprazole were also found for secondary variables such as number of blood transfusions, duration and degree of bleeding, and the need for surgery and additional endoscopic treatments on day 3 and day 21. However, the numbers of deaths by day 3, 21, or 35 were very similar. Conclusions: We found a beneficial effect of intravenous omeprazole in severe ulcer haemorrhage, with a reduction in the number of operations, in endoscopic treatments, and in the duration and severity of bleeding.