LAUSR

Background and aim A number of recent studies have been published evaluating the chemopreventive effect of aspirin against gastric cancer, and an updated meta-analysis is required to evaluate this relationship further. This study presents a meta-analysis of studies examining the effect of aspirin on gastric cancer incidence and death. Methods The PUBMED and Cochrane Central Registration of Controlled Trials databases were searched for eligible studies published up to December 2018. Pooled risk ratios for gastric cancer incidence and death in aspirin users versus nonusers were determined using fixed- and random-effects models. The influence of the frequency of aspirin use, duration of aspirin use, and geographic location on gastric cancer incidence was evaluated. Results The meta-analysis comprised 33 studies with a total of 1 927 971 patients. The pooled risk ratios for gastric cancer incidence in the fixed- and random-effects models were 0.890 (95% confidence interval, 0.871-0.909) and 0.826 (0.740-0.922), respectively. In Asia and North America, the maximum preventive benefit of aspirin use was observed with weekly or daily use. Aspirin use was most effective for noncardiac gastric cancer. The pooled risk ratios for gastric cancer death in the fixed- and random-effects models were 0.798 (0.749-0.850) and 0.894 (0.780-1.024), respectively. Significant heterogeneity was observed among studies of gastric cancer incidence but not gastric cancer death. Conclusion Aspirin use may reduce the risk of gastric cancer incidence and death; however, the relationship may be limited to a specific frequency and duration of aspirin use and geographic location.