Clinical question
Is intermittent proton pump inhibitor (PPI) therapy comparable with continuous-infusion PPI for the treatment of patients with high-risk bleeding ulcers who have undergone endoscopic therapy?
Bottom line
For patients with high-risk bleeding ulcers who have been treated endoscopically, treatment with intermittent proton pump inhibitor (PPI) therapy is as effective as continuous infusion of PPIs for the prevention of rebleeding. This systematic review, however, was not able to determine the most optimal intermittent PPI regimen for this purpose because the included studies included used various dosing schedules and administration routes. (LOE = 1a)
Reference
Study design: Meta-analysis (randomized controlled trials)
Funding source: Government
Allocation: Uncertain
Setting: Inpatient (ward only)
Synopsis
Current guidelines recommend that patients with bleeding ulcers and endoscopic evidence of active bleeding, nonbleeding visible vessels, and adherent clots should receive an intravenous bolus dose of PPI followed by a continuous PPI infusion for 72 hours to prevent rebleeding. In this study, investigators searched multiple databases including MEDLINE, EMBASE, and the Cochrane Register to find randomized clinical trials that evaluated this continuous PPI regimen versus the use of intermittent PPIs for the treatment of these high-risk bleeding ulcers. The intermittent PPI regimens differed in both dosage and administration, from pantoprazole 40 mg given orally every 12 hours to pantoprazole 80 mg given intravenously once, followed by 40 mg intravenously every 6 hours.
Two authors independently performed the search, selected studies for inclusion, extracted data, and assessed the risk of bias for included studies. Ten of the 13 selected studies reported on the primary outcome of recurrent bleeding within 7 days and found that intermittent PPI use was noninferior to continuous-infusion PPI therapy, with the noninferiority margin predefined as an absolute risk difference of 3%. Noninferiority criteria were also met for the secondary outcomes, including rebleeding at 3 days or 30 days, mortality, need for surgical intervention, need for transfusion, and hospital length of stay. No publication or reporting biases were detected.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.