Proton Pump Inhibitors Versus Histamine-2-Receptor Antagonists for Stress Ulcer Prophylaxis in Critically Ill Patients: A Meta-analysis and Trial Sequential Analysis

Link to article at PubMed

J Clin Gastroenterol. 2021 May 28. doi: 10.1097/MCG.0000000000001562. Online ahead of print.

ABSTRACT

GOALS AND BACKGROUND: Stress ulcer prophylaxis has been shown to lower gastrointestinal bleeding (GIB) rates. Various agents have been studied, and the optimal strategy continues to be contested. This study evaluates the efficacy between proton pump inhibitors (PPIs) and histamine-2-receptor antagonists. Small sample sizes and methodology flaws limited prior studies.

STUDY: A systematic search of MEDLINE, EMBASE, CENTRAL, Web of Science, and ClinicalTrials.gov for randomized controlled trials reporting the use of PPI and histamine-2-receptor antagonist reporting rates of GIB and standardized intensive care outcomes. Risk ratios (RR) and standardized mean difference (SMD) with 95% confidence intervals (CIs). A trial sequential analysis was performed to guard against errors.

RESULTS: A total of 14 randomized controlled trials of 28,526 patients with a mean age of 57.83±17.35 years and 30.82% females. In our pooled analysis, PPI outperformed its comparator (RR: 0.68; 95% CI: 0.57-0.82) in clinically significant GIB. PPI re-demonstrated significant reduction in overt GIB (RR: 0.61; 95% CI: 0.39-0.97). No differences between groups was noted toward all-cause mortality (RR: 1.05; 95% CI: 1.00-1.10) or incidence of pneumonia (RR: 1.11; 95% CI: 0.82-1.51). Duration of stay (SMD: 0.07; 95% CI: -0.04-0.17) and ventilator days (SMD: 0.01; 95% CI: -0.01-0.04) were indifferent between the groups.

CONCLUSIONS: Among critically ill patients, PPI was associated with reduced clinically significant or overt GIB. No differences in pneumonia were seen with the use of either agent. Trial sequential analysis for clinically significant GIB ruled out the risk for false-positive results, and thereby it is unlikely that future trials will affect our conclusions.

PMID:34049377 | DOI:10.1097/MCG.0000000000001562

Leave a Reply

Your email address will not be published. Required fields are marked *