Comparison of low- and high-carbohydrate diets for type 2 diabetes management: A randomized trial

Abstract

Background: Few well-controlled studies have comprehensively examined the effects of very-low-carbohydrate diets on type 2 diabetes (T2D). Objective: We compared the effects of a very-low-carbohydrate, high-unsaturated fat, low'saturated fat (LC) diet with a highcarbohydrate, low-fat (HC) diet on glycemic control and cardiovascular disease risk factors in T2D after 52 wk. Design: In this randomized controlled trial that was conducted in an outpatient research clinic, 115 obese adults with T2D [mean ± SD age: 58 ± 7 y; body mass index (in kg/m2): 34.6 ± 4.3; glycated hemoglobin (HbA1c): 7.3 ± 1.1%; duration of diabetes: 8 ± 6 y] were randomly assigned to consume either a hypocaloric LC diet [14% of energy as carbohydrate (carbohydrate ,50 g/d), 28% of energy as protein, and 58% of energy as fat (,10% saturated fat)] or an energymatched HC diet [53% of energy as carbohydrate, 17% of energy as protein, and 30% of energy as fat (,10% saturated fat)] combined with supervised aerobic and resistance exercise (60 min; 3 d/wk). Outcomes were glycemic control assessed with use of measurements of HbA1c, fasting blood glucose, glycemic variability assessed with use of 48-h continuous glucose monitoring, diabetes medication, weight, blood pressure, and lipids assessed at baseline, 24, and 52 wk. Results: Both groups achieved similar completion rates (LC diet: 71%; HC diet: 65%) and mean (95% CI) reductions in weight [LC diet: -9.8 kg (-11.7, -7.9 kg); HC diet: -10.1 kg (-12.0, -8.2 kg)], blood pressure [LC diet: -7.1 (-10.6, -3.7)/-6.2 (-8.2, -4.1) mm Hg; HC diet:-5.8 (-9.4,-2.2)/-6.4 (-8.4,-4.3) mm Hg], HbA1c [LC diet: -1.0% (-1.2%, -0.7%); HC diet: -1.0% (-1.3%, -0.8%)], fasting glucose [LC diet: -0.7 mmol/L (-1.3, -0.1 mmol/L); HC diet: -1.5 mmol/L (-2.1, -0.8 mmol/L)], and LDL cholesterol [LC diet: -0.1 mmol/L (-0.3, 0.1 mmol/L); HC diet: -0.2 mmol/L (-0.4, 0.03 mmol/L)] (P-diet effect $ 0.10). Compared with the HC-diet group, the LC-diet group achieved greater mean (95% CI) reductions in the diabetes medication score [LC diet: -0.5 arbitrary units (-0.7, -0.4 arbitrary units); HC diet: -0.2 arbitrary units (-0.4, -0.06 arbitrary units); P = 0.02], glycemic variability assessed by measuring the continuous overall net glycemic action-1 [LC diet: -0.5 mmol/L (-0.6, -0.3 mmol/L); HC diet: -0.05 mmol/L (-0.2, -0.1 mmol/L); P = 0.003], and triglycerides [LC diet: -0.4 mmol/L (-0.5, -0.2 mmol/L); HC diet: -0.01 mmol/L (-0.2, 0.2 mmol/L); P = 0.001] and greater mean (95% CI) increases in HDL cholesterol [LC diet: 0.1 mmol/L (0.1, 0.2 mmol/L); HC diet: 0.06 mmol/L (-0.01, 0.1 mmol/L); P = 0.002]. Conclusions: Both diets achieved substantial weight loss and reduced HbA1c and fasting glucose. The LC diet, which was high in unsaturated fat and low in saturated fat, achieved greater improvements in the lipid profile, blood glucose stability, and reductions in diabetes medication requirements, suggesting an effective strategy for the optimization of T2D management. This trial was registered at www.anzctr.org.au as ACTRN12612000369820.