LAUSR

Learning Objectives: Recent studies have brought into question the efficacy of stress ulcer prophylaxis (SUP) in the ICU. Prior systematic reviews have not extensively examined bleeding outcomes (clinically significant bleeding, overt bleeding) in trials utilizing SUP versus control or enteral nutrition. The hypothesis of this systemic review is that SUP maintains efficacy for specefic bleeding outcomes in the ICU patient Methods: The purpose of this study is to evaluate the efficacy of SUP with proton pump inhibitors (PPIs) and histamine-2 receptor antagonists (H2RAs) against placebo/control or enteral nutrition. A systematic review was performed utilizing a random-effects meta-analysis with trial sequential analysis (TSA) according to a predefined protocol. Randomized controlled trials comparing PPIs or H2RAs with either placebo/control or enteral nutrition alone were identified with a comprehensive search. Two blinded reviewers independently assessed studies for inclusion, risk of bias, and extracted data utilizing Cochrane Collaborative methodology. The predefined primary outcome was gastrointestinal bleeding (clinically important, overt, and any bleeding). Secondary outcomes included pneumonia, Clostridium difficile associated diarrhea (CDAD), and mortality. Subgroup analyses were conducted for the primary outcome by PPI or H2RA, ICU status, studies after early goal-directed therapy (EGDT) publication, presence of risk factors for stress ulceration, enteral nutrition use Results: Thirty-four trials (N=3220) were included; 33 were judged as high risk of bias and 1 was judged at low risk. Use of SUP significantly reduced clinically important bleeding (RR 0.53, 95% CI 0.34–0.74; P<0.00; I(2)=0%), overt bleeding (RR 0.55, 95% CI 0.39–0.76; P=0.0003; I(2)=53%), and any bleeding (RR 0.54, 95% CI 0.41–0.71; P<0.00001; I(2)=58%). TSA confirmed these findings. There were no significant differences in pneumonia, CDAD, or mortality. Subgroup analyses revealed significant reductions in clinically significant bleeding with SUP in neurosurgical patients (RR 0.37, P<0.05), but not surgery/trauma or medical ICU patients with risk factors. SUP provided no benefit in studies after EGDT. SUP significantly reduced clinically significant bleeding with or without use of enteral nutrition (P<0.05). Conclusions: This meta-analysis demonstrated SUP use was associated with significant reductions in bleeding but not mortality. SUP should not be abandoned until large randomized trials demonstrate the futility of this intervention.