Antiplatelet therapy with or without PPIs for the secondary prevention of cardiovascular diseases in patients at high risk of upper gastrointestinal bleeding: A systematic review and meta-analysis

Figures

• 

  Figure 1. PRISMA flow chart for the selection process of eligible studies.

• 

  Figure 2. Forest plots showing the relative strength of treatment effects of aspirin plus PPIs and clopidogrel groups. The forest plots demonstrate the pooled odds ratios between the aspirin plus PPIs and clopidogrel groups for (A) recurrent UGI events in the overall meta-analysis, (B) recurrent UGI events and UGI bleeding using data obtained only from randomized studies, (C) recurrent UGI events using data obtained only from the observational studies of Hsiao et al (10) and Tsai et al (8), (D) recurrent CV events in the overall meta-analysis, (E) recurrent CV events using data obtained only from randomized studies and (F) overall mortality and vascular death. PPIs, proton pump inhibitors; UGI, upper gastrointestinal; CV, cardiovascular.

• 

  Figure 3. Publication bias assessment. The funnel plot for overall meta-analysis reports on recurrent UGI events between the aspirin plus PPIs and clopidogrel groups. The funnel plot represents minimal publication bias. OR: odds ratio; SE: standard error. UGI, upper gastrointestinal; PPIs, proton pump inhibitors.

• 

  Figure 4. Publication bias assessment. The funnel plot for overall meta-analysis reports on recurrent CV events between the aspirin plus PPIs and clopidogrel groups. Visual assessment of the funnel plot shows no asymmetry and no obvious publication bias. CV, cardiovascular; PPIs, proton pump inhibitors.

• 

  Figure 5. Forest plots showing the relative strength of treatment effects of the aspirin plus PPIs and clopidogrel plus PPIs groups. Two observational studies by Hsiao et al (10) and Tsai et al (8) compare recurrent UGI events between the aspirin plus PPIs group and the clopidogrel plus PPIs group. PPIs, proton pump inhibitors; UGI, upper gastrointestinal.

• 

  Figure 6. Publication bias assessment. The funnel plot for reports on recurrent UGI events between the aspirin plus PPIs and clopidogrel plus PPIs groups. Visual assessment of the funnel plot shows no asymmetry and no obvious publication bias. UGI, upper gastrointestinal; PPIs, proton pump inhibitors.

• 

  Figure 7. Forest plots showing the relative strength of treatment effects of clopidogrel plus PPIs and clopidogrel groups. The forest plots demonstrate pooled odds ratios between the clopidogrel plus PPIs and clopidogrel groups for (A) recurrent UGI events in the overall meta-analysis, (B) recurrent UGI events using data obtained only from the observational studies of Hsiao et al (10) and Tsai et al (8) and (C) recurrent CV events. PPIs, proton pump inhibitors; UGI, upper gastrointestinal; CV, cardiovascular.

• 

  Figure 8. Publication bias assessment. The funnel plot for overall meta-analysis reports on recurrent UGI events between the clopidogrel plus PPIs and clopidogrel groups. Visual assessment of the funnel plot shows no asymmetry and no obvious publication bias. UGI, upper gastrointestinal; PPIs, proton pump inhibitors.