Abstract Background: Our aim was to characterize the association of lipoprotein-associated phospholipase A2 (Lp-PLA2) with coronary and aortic atherosclerosis in a large population-based study. Methods: Lp-PLA2 mass and activity were measured in 2171 subjects 30–65 years old participating in the Dallas Heart Study. We examined the association of Lp-PLA2 levels with 3 atherosclerosis phenotypes: coronary artery calcium (CAC) measured by electron-beam computed tomography and abdominal aortic plaque (AAP) and aortic wall thickness (AWT) measured by magnetic resonance imaging. Results: CAC and AAP were detected in 21% and 40% of subjects, respectively, and mean AWT (SD) was 1.70 (0.32) mm. In univariable analyses, Lp-PLA2 mass (but not activity) was higher in both men (P = 0.04) and women (P = 0.02) with detectable CAC. Lp-PLA2 mass and activity were higher (P = 0.004 and P = 0.01, respectively) and AWT was greater (P < 0.001 and P = 0.02, respectively) in women with aortic atheroma, but not in men. After adjustment for traditional atherosclerosis risk factors and C-reactive protein concentrations, Lp-PLA2 mass and activity were not associated with AAP or AWT in either sex, but Lp-PLA2 mass remained modestly associated with detectable CAC only in men (odds ratio 1.20 per 1 standard deviation increase, 95% CI 1.01–1.42, P = 0.04). Conclusions: Although Lp-PLA2 mass was independently associated with CAC in men, it was not associated with AAP or AWT in men or with any of the atherosclerosis phenotypes in women. These findings suggest that if Lp-PLA2 independently influences clinical events, it does so by promoting atherosclerotic plaque instability rather than by stimulating atherogenesis.