Background and aims: This study aims to investigate the relationship of serum and dietary advanced glycation endproducts (AGEs) with cardiac function and structure after eight years of follow-up. Methods and results: We included 370 Hoorn Study participants (aged 66.4 ± 6.1, 47% women). Serum protein-bound AGEs [Nε-(carboxymethyl)lysine, Nε-(carboxyethyl)lysine, and pentosidine], as well as echocardiography to assess left atrium volume index (LAVI), left ventricle ejection fraction (LVEF), and left ventricle mass index (LVMI), were measured at baseline and after 8 years of follow-up. Dietary AGEs [Nε-(carboxymethyl)lysine and Nε-(carboxyethyl)lysine] were estimated at baseline with a validated food-frequency questionnaire and an AGEs database. Increased pentosidine [-1.4% (−2.6;-0.2)] and overall serum AGEs Z-scores over time [-2.1% (−3.8;-0.5)] were associated with decreased LVEF at follow-up, adjusted for confounders. Glucose metabolism status was an effect modifier (P-for-interaction = 0.04). In participants with impaired glucose metabolism, but not type 2 diabetes, increased pentosidine was associated with decreased LVEF [-4.2 (−8.0;-0.3)%]. Higher dietary Nε-(carboxyethyl)lysine [1.9 (0.1; 3.7)%] and overall dietary AGEs Z-scores [2.1 (0.1; 4.2)%] were associated with higher LVEF at follow-up. However, prior cardiovascular disease (CVD) was an effect modifier (P = 0.02). We found a stronger, non-significant, association of higher dietary (carboxyethyl)lysine with higher LVEF at follow-up in participants without CVD [2.3 (−0.1; 4.7)%] compared to participants with CVD [0.6 (−2.1; 3.4)%]. Conclusion: Overall serum AGEs were longitudinally associated with impaired systolic function. Future research should focus on including changes in dietary AGEs intake over time and the relation of dietary AGEs with cardiac measures needs to be established in intervention studies using low AGEs diets.