Abstract P273: Dietary Saturated Fatty Acids and Coronary Heart Disease Risk in a Dutch Elderly Population: Food Source and Carbon Chain Length Matter

Introduction: The association between dietary saturated fatty acids (SFA) and the risk of coronary heart disease (CHD) is heavily debated and appears to be more complex than initially thought. The food source and carbon chain length of SFA, as well as the macronutrient replacing SFA in the diet may affect the association between dietary SFA and CHD. Hypothesis: We assessed the hypothesis that the association between dietary SFA and incident CHD is dependent on the food source of SFA, the carbon chain length of SFA, and the substituting macronutrient. Methods: From the Rotterdam Study, we included 4,722 healthy men and women, aged 55 years or older. Baseline (1998-1993) SFA intake was assessed using a validated food frequency questionnaire. Incident CHD comprised fatal and non-fatal myocardial infarctions and coronary mortality up to January 2011, and was assessed through digital linkage with municipality records and general practitioners. We used multivariable Cox’ proportional hazard models to calculate CHD risks for higher intakes of total SFA, SFA from specific food sources, SFA differing in carbon chain length, and for higher intakes of other macronutrients at the expense of total SFA. Results: During a median follow up of 16.3 years 569 CHD events occurred. Median total SFA intake was 15.7 en% per day, and the top three food sources were cheese (20.0%), meat (17.8%), and milk and milk products (13.1%). Total SFA intake was not significantly associated with CHD risk (Hazard Ratio (HR) per 5 en%: 1.13, 95%CI: 0.94, 1.22). Neither was SFA from specific food sources, although the data suggested an increased CHD risk for a higher intake of SFA from meat (HR per 1 en%: 1.06, 95%CI: 0.99, 1.14). Regarding SFA differing in carbon chain lengths, a higher CHD risk was observed for C16:0 intake (HR 1en% : 1.18, 95%CI: 1.03, 1.35), but not for the remaining SFA. No significant associations with CHD risk were observed for each 5% lower intake of energy from SFA and a concomitant higher intake of 5% of energy from carbohydrates (HR 5en% : 0.90: 0.80, 1.02), polyunsaturated fatty acids (HR 5en% : 0.90, 95%CI: 0.71, 1.15), cis -monounsaturated fatty acids (HR 5en% :1.09, 95%CI: 0.98, 1.22), or vegetable protein (HR 5en% : 0.88, 95%CI: 0.50, 1.53).The exception was substitution of SFA with animal protein, which was associated with higher CHD risk (HR 5en% : 1.24, 95%CI: 1.01, 1.51). Conclusions: In this Dutch elderly population, with a high median intake of total SFA, the association between SFA intake and CHD risk depends on the SFA food source and carbon chain length. Future intervention studies are needed to determine the importance of the SFA food sources, and to confirm that C16:0 is more strongly related with CHD risk than other SFA.