One-year effects of increasingly fat-restricted, carbohydrate-enriched diets on lipoprotein levels in free-living subjects

Restriction of all dietary fat is a popular strategy for restricting saturated fat intake to lower LDL cholesterol. Some authorities advise the restriction of fat intake to the extreme of less than 10% of daily energy on the assumption that more fat restriction is better. The two studies described herein address questions relating to whether increasing fat restriction produces proportionally increasing benefit on cardiovascular risk factors in hyperlipidemic subjects. The first study is the Dietary Alternatives Study (DAS). The DAS was conducted in 531 male Boeing employees over a 2-year period. Subjects were defined as hypercholesterolemic (HC) or combined hyperlipidemic (CHL) based on age-specific 75th percentiles for plasma LDL-C and triglyceride levels. Hypothesis test analyses were performed at 1 year. HC subjects were randomized to diets taught to attain fat intakes of 30, 26, 22, and 18% (Diets levels 1-4, respectively). CHL subjects (slightly fewer in number) were randomized to Diets IS. After 1 year, subjects' total fat intakes were 27, 26, 25, and 22% of energy (en%), resulting in saturated fat intakes of 8, 7, 7, and 6%, respectively. In HC subjects the greatest LDL-C decrease was with Diet 2 (mean of 13.4%) and in CHL subjects with Diet 1 (7.0%). Surprisingly, plasma triglyceride concentrations rose in HC subjects 20% and 40% above baseline on Diets 3 and 4, respectively, with reciprocal reductions in HDL cholesterol of 2.5% and 3%, respectively. Furthermore, apo B reductions were attenuated below Diet 2 in HC subjects and Diet 1 in CHL subjects, end no further reductions were seen in plasma glucose and insulin concentrations, blood pressure, or body weight. Measurements of plasma total fatty acid composition showed a slight increase in plasma palmitate, whereas stearate decreased slightly, supporting the idea that de novo synthesis of palmitic acid was increased in the chronic high-carbohydrate feeding condition. The second study asked if the most effective diet in HC subjects, Diet 2, has an equivalent effect in women and men. To answer this question, men and women Boeing employees were taught the closely similar National Cholesterol Education Program (NCEP) Step II diet. After 6 and 12 months, equivalent reductions in LDL cholesterol were observed in women compared with men. HDL cholesterol levels in men were unchanged from baseline at 6 and 12 months, but were reduced 8% in HC women, with accompanying decreases of 18% in HDL2-cholesterol end 5% in apoprotein A-I (all P < 0.01). These data indicate that intakes of fat below about 25 en% and carbohydrate intake above 60 en% yield no further LDL-C lowering in HC and CHL male subjects and can be counterproductive to triglyceride, HDL-C, and apo B levels. This lack of benefit appears to be explained by an enhanced endogenous synthesis of palmitic acid, which negates the benefit of further saturated fat restriction. The HDL-C decrease in HC women may have a similar cause and points to an underlying male-female difference. Alternative dietary approaches to limit saturated fat intake deserve intensive study.