Dicarbonyl compounds such as methylglyoxal (MGO), glyoxal (GO), and 3-deoxyglucosone (3-DG) are present in numerous foods. They are pro-inflammatory and pro-oxidative, but their potential role in cardiovascular disease (CVD) development has been scarcely studied. We explored associations between dietary dicarbonyls with fatal and non-fatal CVD.

We conducted a case-cohort analysis based on 32 873 subjects drawn from 346 055 participants of the multi-national prospective EPIC cohort. Cases (15 863 subjects) were CVD-free at baseline and later developed CVD [coronary heart disease (CHD) and/or stroke] with non-fatal (n CVD = 17 837; n CHD = 12 003; n stroke = 6791; not mutually exclusive) and/or fatal (n CVD = 2894; n CHD = 2284; n stroke = 908) outcomes. Dietary intake of dicarbonyl compounds was estimated using country-specific questionnaires linked to a food composition database of dicarbonyl compounds. Multivariable prentice weighted Cox proportional hazards regression models were used to estimate hazard ratios (HRs) and 95% CIs for incident non-fatal and fatal CVD. The main food sources of dicarbonyl compounds include cereals, sugar and confectionaries, coffee, fruits, and vegetables. Higher dietary dicarbonyl intakes were inversely associated with non-fatal CVD (per 1 SD increase, GO: HR = 0.95, 95% CI 0.92–0.98; 3-DG: HR = 0.95, 95% CI 0.92–0.98), fatal CVD (MGO: HR = 0.92, 95% CI 0.87–0.97; GO: HR = 0.91, 0.86–0.96; 3-DG: HR = 0.93, 0.86–0.99), non-fatal CHD (3-DG: HR = 0.95, 0.92–0.99), non-fatal stroke (MGO: HR = 0.93, 95% CI 0.90–0.96; GO: HR = 0.90, 95% CI 0.86–0.95; 3-DG: HR = 0.92, 95% CI 0.89–0.96), and fatal CHD (MGO: HR = 0.94, 95% CI 0.88–0.99; GO: HR = 0.92, 0.86–0.98; 3-DG: HR = 0.89, 0.82–0.96).

Higher intakes of dietary MGO, GO, and 3-DG intake are associated with lower risk of non-fatal or fatal CVD. Further research is required to confirm these findings, assess circulating levels of dicarbonyls, and explore potential underlying mechanisms for their observed CVD risk associations.