Résumé
Objective To assess the relationship between fruit intake and gastric cancer.Methods Articles published during January 2001 and October 2012 that assessed the relationship between fruit intake and gastric cancer were searched on PubMed,Ovid and Willey database.Adjusted relative risk (RR) and 95％ confidence intervals (CI) were calculated by using fixed-effect or random-effect model.Variants of subgroup analysis included living regions,followup duration,and adjusted factors.Results A total of 3679 gastric cancer patients and 1 173 859 subjects from 7 prospective cohorts were included in this metaanalysis,and the pooled RR was 0.89 (95％ CI:0.78-1.01).In the subgroup analysis,the pooled RR in the 10-year followup group was 0.94 (95％ CI:0.85-0.99) before age,cigarette smoking and alcohol drinking were adjusted (RR =0.81,95％ CI:0.54-0.99).Conclusion Fruit intake may prevent the development of gastric cancer.