Summary Background & aims Systematic reviews of case–control studies evaluating the relationship between dietary salt intake and gastric cancer showed a positive association, however a quantitative analysis of longitudinal cohort studies is lacking. Therefore, we carried out a meta-analysis to assess the association between habitual salt intake and risk of gastric cancer in prospective studies. Methods We performed a systematic search of published articles (1966–2010). Criteria for inclusion were: original articles, prospective adult population studies, assessment of salt intake as baseline exposure and of gastric cancer as outcome, follow-up of at least 4 years, indication of number of participants exposed and events across different salt intake categories. Results Seven studies (10 cohorts) met the inclusion criteria (268 718 participants, 1474 events, follow-up 6–15 years). In the pooled analysis, “high” and “moderately high” vs “low” salt intake were both associated with increased risk of gastric cancer (RR = 1.68 95% C.I. 1.17–2.41, p  = 0.005 and respectively 1.41 1.03–1.93, p  = 0.032), with no evidence of publication bias. The association was stronger in the Japanese population and higher consumption of selected salt-rich foods was also associated with greater risk. Meta-regression analyses did not detect specific sources of heterogeneity. Conclusions Dietary salt intake was directly associated with risk of gastric cancer in prospective population studies, with progressively increasing risk across consumption levels.