Comparison of isocaloric very low carbohydrate/high saturated fat and high carbohydrate/low saturated fat diets on body composition and cardiovascular risk

Abstract

Background: It is speculated that high saturated fat very low carbohydrate diets (VLCARB) have adverse effects on cardiovascular risk but evidence for this in controlled studies is lacking. The objective of this study was to compare, under isocaloric conditions, the effects of a VLCARB to 2 low saturated fat high carbohydrate diets on body composition and cardiovascular risk.

Methods: Eighty three subjects, 48 +/- 8 y, total cholesterol 5.9 +/- 1.0 mmol/L, BMI 33 +/- 3 kg/m2 were randomly allocated to one of 3 isocaloric weight loss diets (6 MJ) for 8 weeks and on the same diets in energy balance for 4 weeks: Very Low Fat (VLF) (CHO:Fat:Protein; %SF = 70:10:20; 3%), High Unsaturated Fat (HUF) = (50:30:20; 6%), VLCARB (4:61:35; 20%)

Results: Percent fat mass loss was not different between diets VLCARB -4.5 +/- 0.5, VLF-4.0 +/- 0.5, HUF -4.4 +/- 0.6 kg). Lean mass loss was 32-31% on VLCARB and VLF compared to HUF (21%) (P < 0.05). LDL-C increased significantly only on VLCARB by 7% (p < 0.001 compared with the other diets) but apoB was unchanged on this diet and HDL-C increased relative to the other 2 diets. Triacylglycerol was lowered by 0.73 +/- 0.12 mmol/L on VLCARB compared to -0.15 +/- 0.07 mmol/L on HUF and -0.06 +/- 0.13 mmol/L on VLF (P < 0.001). Plasma homocysteine increased 6.6% only on VLCARB (P = 0.026). VLCARB lowered fasting insulin 33% compared to a 19% fall on HUF and no change on VLF (P < 0.001). The VLCARB meal also provoked significantly lower post prandial glucose and insulin responses than the VLF and HUF meals. All diets decreased fasting glucose, blood pressure and CRP (P < 0.05).

Conclusion: Isocaloric VLCARB results in similar fat loss than diets low in saturated fat, but are more effective in improving triacylglycerols, HDL-C, fasting and post prandial glucose and insulin concentrations. VLCARB may be useful in the short-term management of subjects with insulin resistance and hypertriacylglycerolemia.

Figure 1

Schematic representation of randomization.

Figure 2

Schematic representation of study design. U = 24 hour urine. MTT = meal tolerance test. GTT = glucose tolerance test. DXA = Dual Xray Absorptiometry. ↑ = blood sample. VLCARB = very low carbohydrate diet (n = 24) VLF = very low fat diet (n = 22) HUF = high unsaturated fat (n = 21).

Figure 3

Plasma ketones during weight loss and weight maintenance, according to dietary treatment¹. ¹mean ± SEM. * denotes a significant difference of VLCARB from VLF and HUF (p < 0.05) by one way ANOVA at each time point for main effect of diet (p < 0.05) and post hoc Tukeys test to detect differences (p < 0.05). VLCARB = very low carbohydrate diet (n = 24) VLF = very low fat diet (n = 22) HUF = high unsaturated fat (n = 21)

Figure 4

Weight loss according to dietary treatment¹. ¹mean ± SEM. VLCARB = very low carbohydrate diet (n = 24) VLF = very low fat diet (n = 22) HUF = high unsaturated fat (n = 21). There were no significant differences in absolute weight loss according to dietary treatment.

Figure 5

Plasma glucose and insulin response for 3 h meal tolerance test (MTT) by dietary treatment¹. ¹Mean (± SEM) plasma glucose and insulin concentrations at baseline and 30, 60, 120 and 180 minutes and total AUC after the ingestion of the test meals (Table 2B) at weeks 0 and 12. The main effect of the test meals at week 0 and 12 were compared by repeated-measures ANOVA with week and blood sampling time as within subject factors, and diet as between subject factors. The main effect of time (weight loss) for each diet was compared using repeated-measures ANOVA with AUC at week 1 and 12 as within subject factors. VLCARB = very low carbohydrate diet (n = 24) VLF = very low fat diet (n = 22) HUF = high unsaturated fat (n = 21). * VLCARB test meal significantly different from VLF and HUF test meals at week 0 and week 12, P < 0.01. a denotes no significant effect of weight loss within diet group b = significant effect of weight loss within diet group (P < 0.05)

Figure 6

Plasma glucose and insulin response (mean ± SEM) for 3 h 75 g oral glucose tolerance test (GTT) by dietary treatment¹. ¹Mean (± SEM) plasma glucose and insulin concentrations at baseline and 30, 60, 120 and 180 minutes and total AUC after the ingestion of the 75 g oral glucose drink at weeks 0 and 12. The main effect of diet was compared using repeated-measures ANOVA with week and blood sampling time as within subject factors, and diet as between subject factors. The main effect of time (weight loss) for each diet was compared using repeated-measures ANOVA with AUC at week 1 and 12 as within subject factors. * VLCARB significantly different overall from VLF and HUF at week 0 and week 12, P < 0.01. a = no significant effect of weight loss within diet group. b = significant effect of weight loss within diet group (P < 0.05). VLCARB = very low carbohydrate diet (n = 24) VLF = very low fat diet (n = 22) HUF = high unsaturated fat (n = 21).