Approximately 20 percent of patients have recurrent bleeding of peptic ulcers following endoscopic treatment. These recurrences often require emergency surgery and are associated with a high rate of mortality. In vitro data suggest that platelet aggregation is enhanced at higher levels of gastric pH. However, studies using intravenous histamine H₂-receptor antagonists to prevent recurrent bleeding have produced conflicting results. In addition, tolerance to this class of drugs develops within 72 hours when they are given intravenously, negating their ability to suppress acid production. Lau and colleagues performed a randomized study to determine whether high-dose therapy with a proton-pump inhibitor would decrease the risk of subsequent upper gastrointestinal bleeding after endoscopy.

Patients enrolled in the study were adults admitted to the hospital with a diagnosis of upper gastrointestinal bleeding. All study subjects underwent upper endoscopy within 24 hours of admission, at which time any active bleeding ulcer or nonbleeding ulcer with visible vessels was injected with epinephrine followed by direct thermocoagulation with a heater probe. Testing for Helicobacter pylori infection was also performed at the time of endoscopy.

After endoscopy, the patients were randomized to receive an 80-mg intravenous dose of omeprazole or placebo. This was followed by an intravenous infusion of omeprazole or placebo at 8 mg per hour for 72 hours. Intensive monitoring for any sign of recurrent bleeding was performed until patients were determined to be stable enough for discharge from the hospital. If rebleeding was evident on clinical or laboratory evaluation, endoscopy was repeated along with epinephrine and thermocoagulation if clinically indicated. All patients from the treatment and placebo arms of the study were treated with 20 mg of oral omeprazole per day for eight weeks after discharge. Patients who had a positive test for H. pylori were also given amoxicillin and clarithromycin for one week. The primary end point of the trial was recurrent bleeding within 30 days of endoscopy. All patients underwent repeat endoscopy at eight weeks.

A total of 739 patients with upper gastrointestinal bleeding were admitted during the study period, of which 267 underwent endoscopic treatment. From this group, 120 patients were randomized to the omeprazole arm and 120 to the placebo group. Eighty men were in each group (66.7 percent), and the average age was about 65 years. Recurrent bleeding occurred in eight patients in the omeprazole group and in 27 patients in the placebo group. Most cases of recurrent bleeding (five of eight in the omeprazole group and 24 of 27 in the placebo group) occurred during the three-day infusion period. Three patients in the omeprazole group required surgery, compared with nine in the placebo group.

Of patients who presented with actively bleeding ulcers, recurrent bleeding was found in three of the omeprazole patients compared with 10 in the placebo group. Another factor noted was the number of blood transfusions needed during the first 30 days after endoscopy, which was significantly less in the treatment group. The average duration of hospitalization was also significantly shorter in the omeprazole group compared with the placebo group. Five deaths occurred in the treatment group compared with 12 in the placebo group. However, none of the deaths in the treatment group were directly associated with gastrointestinal bleeding.

The authors conclude that high-dose intravenous omeprazole significantly reduces the risk of recurrent bleeding following endoscopic treatment of peptic ulcers. This therapeutic intervention also appears to decrease the need for blood transfusions and surgery, and to reduce the length of the hospital stay.

editor's note: Although the results of this study are impressive, intravenous omeprazole is not currently available in the United States. A new proton-pump inhibitor, pantoprazole, was recently labeled by the U.S. Food and Drug Administration for this purpose and is available for intravenous administration. In an accompanying editorial, Libby notes that only about one third of the patients admitted in this study had endoscopic evidence of bleeding. This therapy cannot be extrapolated to include any patient admitted with a diagnosis of upper gastrointestinal bleeding. Intravenous omeprazole can rapidly raise intragastric pH to more than 6, a level shown to enhance platelet aggregation and formation of fibrin clot. Oral proton-pump inhibitors take several days to produce this level of acid suppression and thus may not be as effective as the intravenous formulation in preventing early recurrence of bleeding.—j.t.k.