Confidential: for Review Only Efficacy and Safety of Gastrointestinal Bleeding Prophylaxis in Critically Ill Patients: Systematic Review and Network Meta-Analysis
Total Page:16
File Type:pdf, Size:1020Kb
BMJ Confidential: For Review Only Efficacy and safety of gastrointestinal bleeding prophylaxis in critically ill patients: systematic review and network meta-analysis Journal: BMJ Manuscript ID BMJ-2019-052088 Article Type: Research BMJ Journal: BMJ Date Submitted by the 08-Aug-2019 Author: Complete List of Authors: Wang, Ying; Beijing Chao-Yang Hospital, Capital Medical University, Department of Pharmacy YE, ZHIKANG; McMaster University, Department of Health Research Methods, Evidence and Impact Ge, Long; School of Public Health, Lanzhou University, Evidence Based Social Science Research Center; School of Basic Medical Sciences, Lanzhou University, Evidence Based Medicine Center Siemieniuk, Reed; McMaster University, Health Research Methods, Evidence, and Impact Wang, Xin; Beijing Chaoyang Hospital, Capital Medical University, Beijing, China Wang, Yingkai; Beijing Chaoyang Hospital, Capital Medical University, Beijing, China Hou, Liangying; Evidence Based Social Science Research Center, School of Public Health, Lanzhou University, Lanzhou, China Ma, Zhuo; Beijing Chaoyang Hospital, Capital Medical University, Beijing, China Agoritsas, Thomas; McMaster University, Department of Clinical Epidemiology and Biostatistics; University Hospitals of Geneva, Division of General Internal Medicine & Division of Epidemiology Vandvik, Per; Innlandet Hospital Trust-divisjon Gjøvik, Department of Medicine Perner, Anders; Copenhagen University Hospital Rigshospitalet, Dept. of Intensive Care 4131 Møller, Morten; Copenhagen University Hospital Rigshospitalet, Dept. of Intensive Care 4131 Guyatt, Gordon; McMaster University, Liu, Li-Hong; Department of Pharmacy, Beijing Chao-Yang Hospital, Capital Medical University, Department of Pharmacy gastrointestinal bleeding prophylaxis, systematic review and network Keywords: meta-analysis, proton pump inhibitors, histamine-2 receptor antagonists, sucralfate https://mc.manuscriptcentral.com/bmj Page 1 of 109 BMJ 1 2 3 4 5 6 7 8 9 10 11 Confidential: For Review Only 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 2 of 109 1 2 3 4 Efficacy and safety of gastrointestinal bleeding prophylaxis in critically ill 5 6 patients: systematic review and network meta-analysis 7 8 9 1 1,2 3 2,4 1 10 Ying Wang, Zhikang Ye, Long Ge, Reed A Siemieniuk, Xin Wang, Yingkai 11 Wang,1 Liangying Hou,3 Zhuo Ma,1 Thomas Agoritsas,2,5 Per Olav Vandvik,6 Anders 12 Confidential: For Review Only 13 Perner,7 Morten H Møller,7 Gordon H Guyatt,2 Lihong Liu1 14 15 16 17 1Department of Pharmacy, Beijing Chaoyang Hospital, Capital Medical University, 18 19 Beijing, China 20 21 2Department of Health Research Methods, Evidence and Impact, McMaster 22 23 University, Hamilton, Canada 24 25 3Evidence Based Social Science Research Center, School of Public Health, Lanzhou 26 27 University, Lanzhou, China 28 29 4Department of Medicine, University of Toronto, Toronto, Canada 30 31 5Division General Internal Medicine, and Division of Clinical Epidemiology, 32 33 University Hospitals of Geneva, Geneva, Switzerland 34 35 6Institute of Health and Society, University of Oslo, Norway 36 37 7Department of Intensive Care, Copenhagen University Hospital Rigshospitalet, 38 39 Copenhagen, Denmark 40 41 Correspondence to: L Liu [email protected] 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 1 https://mc.manuscriptcentral.com/bmj Page 3 of 109 BMJ 1 2 3 4 ABSTRACT 5 6 OBJECTIVE 7 8 To address the relative impact of gastrointestinal bleeding prophylaxis (often referred 9 10 to as stress ulcer prophylaxis or SUP) with proton pump inhibitors (PPIs), histamine-2 11 receptor antagonists (H2RAs), sucralfate or no prophylaxis in critically ill patients. 12 Confidential: For Review Only 13 DESIGN 14 15 Systematic review and network meta-analysis. 16 17 DATA SOURCES 18 19 Medline, PubMed, Embase, Cochrane Central Register of Controlled Trials, trial 20 21 registers, and grey literature up to March 2019. 22 23 STUDY SELECTION 24 25 Randomised controlled trials comparing gastrointestinal bleeding prophylaxis with 26 27 PPIs, H2RAs, or sucralfate versus one another or placebo or no prophylaxis in adult 28 29 critically ill patients. 30 31 REVIEW METHODS 32 33 Two reviewers independently screened studies for eligibility, extracted data and 34 35 assessed risk of bias. A parallel guideline committee (BMJ Rapid Recommendation) 36 37 provided critical oversight of the systematic review, including identifying outcomes 38 39 important to patients. We performed random-effects pairwise and network 40 41 meta-analyses (NMA) and used GRADE to assess certainty of evidence for each 42 43 outcome. When results differed between low and high risk of bias studies we used the 44 45 former as best estimates. 46 47 RESULTS 48 49 Seventy-two trials including 12 660 patients proved eligible. For patients at highest 50 (above 8%) or high (4% to 8%) risk of bleeding, both PPIs (odds ratio 0.61, 95% 51 52 confidence interval 0.42 to 0.89, 3.3% fewer for highest and 2.3% fewer for high risk 53 54 patients, moderate certainty) and H2RAs (OR 0.46, 0.27 to 0.79, 4.6% fewer for 55 56 highest and 3.1% fewer for high risk patients, moderate certainty) probably reduce 57 58 clinically important gastrointestinal bleeding compared with placebo or no 59 60 2 https://mc.manuscriptcentral.com/bmj BMJ Page 4 of 109 1 2 3 4 prophylaxis. PPIs (OR 1.38, 0.97 to 2.08, 4.8% more, low certainty) and H2RAs (OR 5 6 1.26, 0.90 to 1.86, 3.4% more, low certainty) may increase pneumonia compared with 7 8 no prophylaxis. It is likely that neither PPIs nor H2RAs impact on mortality; 9 10 otherwise, results provided no support for any impacting on mortality, Clostroides 11 difficile infection, length of ICU stay, length of hospital stay, or duration of 12 Confidential: For Review Only 13 mechanical ventilation (varying certainty of evidence). 14 15 CONCLUSIONS 16 17 Evidence failed to support differences in a number of outcomes, including mortality. 18 19 For higher risk critically ill patients, PPIs and H2RAs compared with no prophylaxis 20 21 likely result in important reductions in gastrointestinal bleeding; for patients at low 22 23 risk, the reduction in bleeding may be unimportant. Both PPIs and H2RAs may 24 25 increase pneumonia. 26 27 SYSTEMATIC REVIEW REGISTRATION 28 29 PROSPERO CRD42019126656. 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 Introduction 59 60 3 https://mc.manuscriptcentral.com/bmj Page 5 of 109 BMJ 1 2 3 4 Critically ill patients in the intensive care unit (ICU) are at risk for gastrointestinal 5 1 6 bleeding from a number of sources, one of which is stress ulceration. Authorities 7 8 have suggested gastrointestinal bleeding prophylaxis (often referred to as stress ulcer 9 10 prophylaxis or SUP) as essential to optimize the care of critically ill patients, and most 11 patients considered high-risk receive acid suppression during an ICU stay.2 3 Proton 12 Confidential: For Review Only 13 pump inhibitors (PPIs) are the most common prophylactic agent, followed by 14 15 histamine-2 receptor antagonists (H2RAs); clinicians seldom use sucralfate and 16 17 antacids.2 4 18 19 Recently, however, the practice has generated controversy.5 Doubts include 20 21 questions of whether bleeding occurs frequently enough, or whether its consequences 22 23 are serious enough, to warrant prophylaxis.5 Moreover, observational studies have 24 25 reported increased rates of nosocomial pneumonia and Clostroides difficile infection 26 27 with the use of acid-suppressive drugs,6 7 raising concern that harms may outweigh the 28 29 benefits. 30 31 Many published systematic reviews and meta-analyses have summarized 32 33 randomised controlled trial (RCT) evidence addressing the efficacy and safety of 34 35 interventions for gastrointestinal bleeding prophylaxis,8-13 including a network 36 37 meta-analysis (NMA) conducted by members of our team.8 Since the publication of 38 39 the last NMA, several trials have been published,14-17 including a large, international, 40 17 41 multicenter RCT (the SUP-ICU trial). This trial compared pantoprazole with 42 43 placebo and concluded that pantoprazole did not reduce mortality or a composite 44 45 secondary outcome of ‘clinically important events’, and questioned the routine use of 46 47 PPIs in critically ill adults. 48 49 We therefore conducted an updated systematic review and NMA addressing the 50 potential benefits and harms of gastrointestinal bleeding prophylaxis with PPIs, 51 52 H2RAs, and sucralfate in critically ill patients. This review is part of the BMJ Rapid 53 54 Recommendations project, a collaborative effort from the MAGIC Evidence 55 56 Ecosystem Foundation (www.magicproject.org) and BMJ. The aim of the project is to 57 58 respond to new potentially practice changing evidence and provide a trustworthy 59 60 4 https://mc.manuscriptcentral.com/bmj BMJ Page 6 of 109 1 2 3 18 4 practice guideline in a timely manner. In this case, the stimulus was the SUP-ICU 5 17 6 trial. This systematic review informed the parallel guideline published in a 7 8 multilayered electronic format on bmj.com and MAGICapp 9 10 (https://app.magicapp.org/app#/guideline/3096). 11 Methods 12 Confidential: For Review Only 13 Protocol registration 14 15 We registered the protocol for this systematic review with PROSPERO 16 17 (CRD42019126656).19 18 19 Guideline panel and patient involvement 20 21 In accordance with the BMJ Rapid Recommendations process, a guideline panel 22 23 provided critical oversight during the review process, which included identifying 24 25 populations, subgroups, and outcomes of interest.