Objectives <p>Proton pump inhibitors (PPIs) are recommended for the prevention of aspirin-associated gastrointestinal mucosal injury, but emerging evidence indicates that PPIs may increase the risk of lower gastrointestinal bleeding (LGIB). This meta-analysis aimed to comprehensively evaluate the effects of PPIs on gastrointestinal bleeding (GIB).</p> Methods <p>Major electronic databases were searched to identify the studies evaluating the differences between PPIs versus placebo/blank, no-PPI, histamine-2 receptor antagonists (H₂RAs), mucosal protective agents (MPAs), and potassium-competitive acid blockers (P-CABs) in preventing aspirin-associated GIB. Risk ratios (RRs) with their 95% confidence intervals (CIs) were calculated.</p> Results <p>Sixteen randomized controlled trials (RCTs) and 14 non-RCTs were included. PPIs group had significantly lower risk of upper gastrointestinal bleeding (UGIB) than placebo/blank group (7 RCTs, RR = 0.30, 95%CI = 0.19–0.46; 2 non-RCTs, RR = 0.05, 95%CI = 0.01–0.42), no-PPI group (7 non-RCTs, RR = 0.50, 95%CI = 0.38–0.66), H₂RAs group (5 RCTs, RR = 0.21, 95%CI = 0.09–0.53; 4 non-RCTs, RR = 0.12, 95%CI = 0.03–0.52), and MPAs group (4 RCTs, RR = 0.24, 95%CI = 0.10–0.58), but statistically similar (2 non-RCTs, RR = 1.18, 95%CI = 0.77–1.79) or significantly higher (1 RCT, log-rank <i>P</i> = 0.018 and 0.019) risk of UGIB than P-CABs group. PPIs group had significantly higher risk of lower gastrointestinal bleeding (LGIB) than no-PPI group (3 non-RCTs, RR = 2.85, 95%CI = 1.85–4.38), but statistically similar risk of LGIB to H₂RAs group (2 RCTs, RR = 1.88, 95%CI = 0.30-11.82).</p> Conclusions <p>The evidence from RCTs and non-RCTs suggests the advantages of PPIs over placebo/blank, no-PPI, H₂RAs, and MPAs for the prevention of UGIB in aspirin users. The evidence regarding PPIs and LGIB in aspirin users is from non-RCTs alone, and suggests their potential harms.</p> Graphical abstract <p></p>