Abstract

Some dietary patterns are associated with inflammation, while others lower inflammation and improve health. However, many people cannot follow a complete, healthy diet. Therefore, this study's aim was to identify specific foods associated chronic inflammation and mortality. The study used Multi-Ethnic Study of Atherosclerosis (MESA) research materials from the NHLBI Biologic Specimen and Data Repository Information Coordinating Center. Three plant-based and three animal-based MESA food categories were chosen based on perceived availability in the western diet. The assessed food categories were avocado, ham, sausage, eggs, greens, and broccoli. Inflammatory markers assessed were interleukin-6 (IL-6), fibrinogen antigen, C-reactive protein, D-Dimer, interleukin-2, matrix metalloproteinase 3, necrosis factor-a soluble receptors, oxidized LDL (oxLDL), and total homocysteine. The primary outcome was the multivariable association of foods and inflammatory markers with all-cause mortality. All inflammatory makers, except oxLDL, were associated with mortality in univariate analysis. The effect was largest with IL-6 and D-dimer. The category of broccoli had the most consistent association in univariate analyses with lower inflammation and lower mortality odds. Low and high broccoli consumption versus no consumption were associated with lower mortality odds in the multivariable models with IL-6 and D-dimer. Consumption of the MESA<b>-</b>defined food category "broccoli" (<i>i.e.</i>, broccoli, cabbage, cauliflower, brussels sprouts, sauerkraut, and kimchee) was associated with lower inflammation and lower mortality odds. These findings should be validated in randomized controlled trials testing a "food is medicine" approach to identify which, if any, of these foods may have potential as an herbal therapeutic for chronic inflammation.