Abstract
OBJECTIVE To determine the associations of stress ulcer prophylaxis with gastrointestinal (GI) bleeding, nosocomial pneumonia (NP), mortality, and length of stay in the pediatric intensive care unit (PICU). STUDY DESIGN We conducted a systematic review and meta-analysis of randomized controlled trials (RCTs) and observational studies in the English language assessing the effects of proton pump inhibitors and histamine-2 receptor antagonists on patients in the PICU published before October 2018 from the PubMed, Embase, CINAHL, and Cochrane Central Register of Controlled Trials databases. A random-effects Mantel-Haenszel risk difference (MHRD) model was used to pool all the selected studies for meta-analysis. Primary outcomes were the incidences of GI bleeding and NP. Secondary outcomes included mortality and length of PICU stay. RESULTS Seventeen studies (4 RCTs and 13 observational studies) with a total of 340 763 patients were included. The overall incidence of GI bleeding was 15.2%. There was no difference in the risk of GI bleeding based on stress ulcer prophylaxis status (MHRD, 5.0%; 95% CI, -1.0% to 11.0%; I2 = 62%). There was an increased risk of NP in patients who received stress ulcer prophylaxis compared with those who did not (MHRD, 5.3%; 95% CI, 3.5%-7.0%; I2 = 0%). An increased risk of mortality was seen in patients receiving stress ulcer prophylaxis (MHRD, 2.1%; 95% CI, 2.0%-2.2%; I2 = 0%), although this association was no longer found when 1 large study was removed in a sensitivity analysis. There was no statistically significant difference in length of PICU stay between the groups (standardized mean difference, 0.42 days; 95% CI, -0.16 to 1.01 days; I2 = 89.8%). CONCLUSIONS Stress ulcer prophylaxis does not show a clear benefit in reducing GI bleeding or length of PICU stay. Observational studies suggest an increased risk of NP and mortality with stress ulcer prophylaxis, which remains to be validated in clinical trials.
