Background—Diet is a major modifiable risk factor for cardiovascular disease, but it varies markedly in different regions of the world. The objectives of the present study were to assess the association between dietary patterns and acute myocardial infarction (AMI) globally.

Methods and Results—INTERHEART is a standardized case-control study involving participants from 52 countries. The present analysis included 5761 cases and 10 646 control subjects. We identified 3 major dietary patterns using factor analysis: Oriental (high intake of tofu and soy and other sauces), Western (high in fried foods, salty snacks, eggs, and meat), and prudent (high in fruit and vegetables). We observed an inverse association between the prudent pattern and AMI, with higher levels being protective. Compared with the first quartile, the adjusted ORs were 0.78 (95% CI 0.69 to 0.88) for the second quartile, 0.66 (95% CI 0.59 to 0.75) for the third, and 0.70 (95% CI 0.61 to 0.80) for the fourth (P for trend <0.001). The Western pattern showed a U-shaped association with AMI (compared with the first quartile, the adjusted OR for the second quartile was 0.87 [95% CI 0.78 to 0.98], whereas it was 1.12 [95% CI 1.00 to 1.25] for the third quartile and 1.35 [95% CI 1.21 to 1.51] for the fourth quartile; P for trend <0.001), but the Oriental pattern demonstrated no relationship with AMI. Compared with the first quartile, the OR of a dietary risk score derived from meat, salty snacks, fried foods, fruits, green leafy vegetables, cooked vegetables, and other raw vegetables (higher score indicating a poorer diet) increased with each quartile: second quartile 1.29 (95% CI 1.17 to 1.42), third quartile 1.67 (95% CI 1.51 to 1.83), and fourth quartile 1.92 (95% CI 1.74 to 2.11; P for trend <0.001). The adjusted population-attributable risk of AMI for the top 3 quartiles compared with the bottom quartile of the dietary risk score was 30%.

Conclusions—An unhealthy dietary intake, assessed by a simple dietary risk score, increases the risk of AMI globally and accounts for 30% of the population-attributable risk.