BMI modifies the effect of dietary fat on atherogenic lipids: a randomized clinical trial

Background: SFA intake increases LDL cholesterol whereas PUFA intake lowers it. Whether the lipid response to dietary fat differs between normal-weight and obese persons is of relevance to dietary recommendations for obese populations. Objectives: We compared the effect of substituting unsaturated fat for saturated fat on LDL cholesterol and apoB concentrations in normal-weight (BMI <= 25 kg/m(2)) and obese (BMI: 30-45) subjects with elevated LDL cholesterol. Methods: We randomly assigned 83 men and women (aged 21-70 y) stratified by BMI (normal: n = 44; obese: n = 39) and elevated LDL cholesterol (mean +/- SD, normal weight 4.6 +/- 0.9 mmol/L; obese 4.4 +/- 0.8 mmol/L) to either a PUFA diet enriched with oil-based margarine (n = 42) or an SFA diet enriched with butter (n = 41) for 6 wk. Results: Seven-day dietary records showed differences of similar to 9 energy percent (E%) in SFA and similar to 4 E% in PUFA between the SFA and PUFA groups. In the total study population, the PUFA diet compared with the SFA diet lowered LDL cholesterol (-0.31 mmol/L; 95% CI: -0.47, -0.15 mmol/L, compared with 0.32 mmol/L; 95% CI: 0.18, 0.47 mmol/L; P < 0.001) and apoB (-0.08 g/L; 95% CI: -0.11, -0.05 g/L, compared with 0.07 g/L; 95% CI: 0.03, 0.10 g/L; P < 0.001). Tests of the BMI x diet interaction were significant for total cholesterol, LDL cholesterol, and apoB (P values <= 0.009). In normal-weight compared with obese participants post-hoc comparisons found that the respective changes in LDL cholesterol were 9.7% (95% CI: 5.3%, 14.2%) compared with 5.3% (95% CI: -0.7%, 11.2%), P = 0.206, in the SFA group, and -10.4% (95% CI: -15.2%, -5.7%) compared with -2.3% (95% CI: -7.4%, 2.8%), P = 0.020, in the PUFA group. ApoB changes were 7.5% (95% CI: 3.5%, 11.4%) compared with 3.0% (95% CI: -1.7%, 7.7%), P = 0.140, in the SFA group, and -8.9% (95% CI: -12.6%, -5.2%) compared with - 3.8% (95% CI: -6.3%, -1.2%), P = 0.021, in the PUFA group. Responses to dietary fat were not associated with changes in polyprotein convertase subtisilin/kexin type 9 concentrations. Conclusions: BMI modifies the effect of PUFAs compared with SFAs, with smaller improvements in atherogenic lipid concentrations in obese than in normal-weight individuals, possibly supporting adjustment of dietary recommendations according to BMI.