[Fish consumption and coronary mortality in the general population: a meta-analysis of cohort studies]

In order to estimate the effect of fish intake on ischemic heart disease mortality in the general population, we performed a meta-analysis of the epidemiologic studies involving participants free of disease at baseline published on the topic. All of the 7 studies published to date were cohort studies; however, only 5 of them reported the results with enough detail to be used in a formal meta-analysis. The total number of participants in these studies was 27,656, with an average follow-up in each study between 7.5 and 25 years and a total of 1,731 coronary deaths. The combined estimate of the relative risk for an intake of 30 g/day of fish compared to no intake was 0.96 (95% CI: 0.93-1.00; P = 0.058). Due to the presence of statistically significant heterogeneity among the studies, unexplained by a priori factors, we combined the studies assuming a random effects model, obtaining a relative risk estimate of 0.92 (95% CI: 0.84-1.01; P = 0.090). These results, together with the results of the only clinical trial of fish intake performed in post-myocardial infarction patients, in which an intake of 200-400 g/week of fatty fish reduced total mortality by 29% (relative risk of intake compared to no intake 0.71; 95% CI: 0.54-0.93), suggest a moderate beneficial effect of fish intake on coronary mortality.