Interventions for Preventing Upper Gastrointestinal Bleeding in People Admitted to Intensive Care Units

Abstract

What are the effectiveness and harms of interventions to prevent upper gastrointestinal bleeding (UGIB) in patients admitted to the intensive care unit (ICU)?Despite advancements in medical treatments and interventions, UGIB results in more than 300 000 hospital admissions annually in the United States and has a mortality rate of 3.5% to 14%.1,2 Upper gastrointestinal bleeding is frequently the result of stress ulcers—superficial damage of the mucosal lining of the stomach or intestines that can occur as the result of shock, sepsis, or trauma. With improvements in treatment options and standards of care, the incidence of UGIB in ICUs has been decreasing,3 yet stress ulcer prophylaxis is still commonplace. Because there are potential risks associated with stress ulcer prevention, this practice needs to be examined.The disadvantage of stress ulcer prophylaxis is that many of the interventions used to suppress gastric acid can alter the gastric flora and promote the colonization of bacteria, leading to infection. For example, stress ulcer prophylaxis is associated with ventilator-associated pneumonia (VAP). By inhibiting gastric acid secretion, the colonization of gastric bacteria can also lead to colonization of the pharynx, leading to aspiration and VAP.4 Hence, it is necessary to evaluate strategies that safely decrease the incidence of UGIB.The focus of previous systematic reviews related to stress ulcer prophylaxis and UGIB has been on either older therapies or the effects of a single drug class versus another. As such, a systematic review of the current pharmacologic and nonpharmacologic prophylaxis of UGIB, comparing them with no treatment or other treatments, was warranted to better understand the effectiveness and harms.This summary is based on a Cochrane systematic review by Toews et al5 that included data from 107 studies with a total of 15 057 patients. The review included 27 comparisons involving 14 different treatment modalities (most notably proton pump inhibitors and H2-receptor antagonists). The primary outcome of the review was clinically important gastrointestinal bleeding (as defined by the authors of the individual studies); the secondary outcomes were nosocomial pneumonia and VAP; death in the ICU; duration of ICU stay; duration of mechanical ventilatory support; number of patients requiring blood transfusions; number of units of blood transfused; and serious adverse events (eg, thrombocytopenia) of interventions.5 The authors independently assessed the risk of bias for each study, including selection bias, performance bias, detection bias, attrition bias, reporting bias, and publication bias; they resolved any disagreements by reviewing the data together and through discussion.5Risk ratio (RR) and mean difference with confidence intervals (CIs) were used as the measures of treatment effect between different comparisons and outcomes.5 The internationally approved Grades of Recommendation, Assessment, Development and Evaluation approach was used to provide a level of the quality of evidence for each outcome.6 The 4 levels are summarized as follows: (1) high-quality level indicates further research is very unlikely to change the confidence in the conclusions; (2) moderate-quality level indicates further research is likely to have an important impact on the confidence in the conclusion and may change the conclusion; (3) low-quality level indicates further research is very likely to have an important impact on the confidence in the conclusion and is likely to change the conclusion; and (4) very low-quality level means there is uncertainty about the conclusion.6The following points summarize the main results of the Toews et al5 review:In regard to the comparison of H2-receptor antagonists and proton pump inhibitors, results were the following:In their systematic review, Toews et al5 found evidence of moderate certainty that, compared with placebo or no prophylaxis, the stress ulcer prevention interventions examined might be effective in preventing UGIB in patients in the ICU. When any treatment was compared with no prophylaxis, the risk of VAP developing was about the same. There was evidence of low certainty to suggest that proton pump inhibitors might be more effective than H2-receptor antagonists.5 As such, decisions on what interventions to use should be based on clinician assessments of the patient’s needs and underlying conditions.The low quality of the evidence for several aspects of this review indicates larger and high-quality randomized controlled trials are warranted to confirm the results. Such studies include high-quality trials assessing the risk of VAP in patients receiving stress ulcer prevention medication.Nurses caring for critically ill patients should advocate for the best evidence-based treatment. We must always consider the best available evidence and understand the feasibility, appropriateness, meaningfulness, and effectiveness of any intervention to determine if it is the best approach to implement in our individual context.