Randomized Controlled Trial

A - Click here for explanation of classification scheme.

NEUTRAL: See Quality Criteria Checklist below.

The objective of this research was to compare the effects of altering the glycemic index or the amount of carbohydrates on glycated hemoglobin (HbA1c), plasma glucose, lipids and C-reactive protein (CRP) in patients with type 2 diabetes mellitus.

• Subjects were 35 years to 75 years old and had HbA1c up to 130% of the upper limit of normal and a body mass index (BMI in kg/m²) of 24 to 40
• Provided written informed consent.

• Use of insulin or any hypoglycemic or anti-hyperglycemic medication
• Stroke
• Myocardial infarction
• Major surgery within six months of randomization
• Serum triacylglycerol concentrations over 10mmol per L
• Any major debilitating disorder
• Any condition or drug likely to alter nutrient absorption
• Use of oral steroids, substance or alcohol abuse
• Allergy or intolerance to more than one of the study's key foods
• Expectation of being on vacation and unable to take study foods for more than eight weeks in a row or a total of more than 12 weeks.

• Recruitment: Men or non-pregnant women with type 2 diabetes mellitus [fasting plasma glucose of at least 7.0mmol per L or plasma glucose of at least 11.1mmol per L two hours after a 75-g oral glucose tolerance test (OGTT) on at least one occasion within two months of randomization] that was managed by diet alone were recruited.
• Design: Randomized controlled trial
• Blinding used: Implied with measurements.

Intervention

• Subjects were randomly assigned to receive one of three diets for one year:
  • High carbohydrate and high glycemic index (high GI diet)
  • High carbohydrate and low glycemic index diet (low GI diet)
  • Low carbohydrate and high monounsaturated fat diet (low-CHO diet).
• The high-GI, low-GI, and low-CHO diets contained, respectively, 47%, 52% and 39% of energy as carbohydrates and 31%, 27% and 40% of energy as fat. They had GIs of 63, 55 and 59, respectively.
• Subjects were instructed how to record all foods and drinks consumed during two typical weekdays and one weekend day (three-day food record). After the food record was completed, subjects underwent a baseline 75-g OGTT and the food record was reviewed by a registered dietitian (RD) who provided dietary advice with the aim of a diet containing 55% of energy as carbohydrates, 15% of energy as protein, and 30% of energy as fat and with no more than 10% saturated fatty acids (SFAs), no more than 10% polyunsaturated fatty acids and the remainder as monounsaturated fatty acids (MUFAs).
• Composition of the three diets:
  • Higher carbohydrate and higher glycemic index (higher GI diet, 47% carbohydrate, 31% fat, GI of 63, 10.2±0.4% saturated fat, 12.3±0.3% monounsaturated fat, 5.5±0.2% polyunsaturated fat)
  • Higher carbohydrate and lower glycemic index diet (lower GI diet, 52% carbohydrate, 27% fat, GI of 55, 8.2±0.4% saturated fat, 10.7±0.4% monounsaturated fat, 5.1±0.2% polyunsaturated fat)
  • Lower carbohydrate and higher monounsaturated fat diet (lower-CHO diet, 39% carbohydrate, 40% fat, GI of 59, 10.8±0.3% saturated fat, 18.3±0.3% monounsaturated fat, 8.2±0.2% polyunsaturated fat). For the lower carbohydrate and higher monounsaturated fat diet, subjects consumed specific foods, such as olive or canola oils or spreads, nuts and nut butters (almonds, peanuts, cashews, pecans, hazelnuts and macadamia), avocados and olives.
• Subjects were seen two and four weeks after randomization and then every four weeks for weighing, review of key-food diaries and pick-up of supplies of key foods
• During each 30-minute visit, dietitians provided individualized dietary advice and discussed any challenges that subjects encountered in following the study protocol and their solutions
• Three-day food records were recorded twice during the run-in period and at one, three, six, nine and 12 months after randomization.

Statistical Analysis

• The nutrient composition of the test meals and diets was assessed by using an in-house program with a nutrient database based on the Canadian nutrient file and with values for GI added
• Glycemic load (GL) was calculated as the sum of GI x g for each food in the diet, where g represents grams of carbohydrate
• Longitudinal analyses of primary and secondary outcomes were carried out by using a general linear mixed model in SAS PROC MIXED software
• Analysis of CRP, fasting insulin and free fatty acids was carried out on the natural logarithm of the values to improve the symmetry and homoscedasticity of the distributions
• For the breakfast profiles, the significance of differences in plasma glucose increments and AUC from baseline values were assessed by using a paired T-test
• The significance of differences in glucose increments and AUC between baseline and one year were compared across the different diets by using one factor analysis of variance (ANOVA) with Tukey's test used to control for multiple comparisons
• Differences were considered significant if two-tailed P-values were less than 0.05.

Timing of Measurements

Measurements made at baseline and then one, three, six, nine and 12 months after randomization.

Dependent Variables

• Glycated hemoglobin (HbA1c) and blood glucose, lipids, and C-reactive protein measured through standard laboratory methods
• Subjects underwent 75-g OGTTs with blood samples for plasma glucose and insulin taken fasting and at 30, 60 and 120 minutes after the subjects started to consume the glucose
• Blood samples were analyzed centrally and HbA1c was measured with HPLC by using a hexokinase method. Insulin was measured by using an electrochemiluminescence immunoassay.
• Free fatty acids were measured enzymatically and so was serum cholesterol, triacylglycerol, apolipoproteins A-I and apo B.
• CRP was analyzed by using nephelometry.

Independent Variables 

• Subjects were randomly assigned to receive one of three diets for one year:
  • High carbohydrate and high glycemic index (high GI diet)
  • High carbohydrate and low glycemic index diet (low GI diet)
  • Low carbohydrate and high monounsaturated fat diet (low-CHO diet).
• The high-GI, low-GI and low-CHO diets contained, respectively, 47%, 52% and 39% of energy as carbohydrates and 31%, 27% and 40% of energy as fat. They had GIs of 63, 55 and 59, respectively.
• Subjects were instructed how to record all food and drinks consumed during two typical weekdays and one weekend day (three-day food record). After the food record was completed, subjects underwent a baseline 75-g OGTT and the food record was reviewed by an RD who provided dietary advice with the aim of a diet containing 55% of energy as carbohydrates, 15% of energy as protein and 30% of energy as fat and with no more than 10% saturated fatty acids (SFAs), no more than 10% polyunsaturated fatty acids and the remainder as monounsaturated fatty acids (MUFAs).
• Subjects were seen two and four weeks after randomization and then every four weeks for weighing, review of key-food diaries and pick-up of supplies of key foods
• During each 30-minute visit, dietitians provided individualized dietary advice and discussed any challenges that subjects encountered in following the study protocol and their solutions
• Three-day food records were recorded twice during the run-in period and at one, three, six, nine and 12 months after randomization.

• Initial N: 162 subjects randomized. A total of 52 (50% female) were in the high-GI diet, 56 (66% female) in the low-GI diet and 54 (47% female) in the low-CHO diet
• Attrition (final N): A total of 156 were included in the final analysis (48 in the high-GI diet, 55 in the low-GI diet and 53 in the low-CHO diet)
• Age: Mean age, 60.4 years in the high-GI diet group; 60.6 years in the low-GI diet group; 58.6 years in the low-CHO diet group
• Ethnicity: Not reported
• Other relevant demographics: Mean HbA1c of 6.2% in the high-GI diet group, 6.2% in the low-GI diet group and 6.1% in the low-CHO diet group
• Anthropometrics: Mean BMI, 30.1kg/m² in the high-GI diet group, 31.6kg/m² in the low-GI diet group and 31.1kg/m² in the low-CHO diet group. The only significant differences among diet groups at baseline were lower LDL-cholesterol in subjects following the high-GI diet than in those following the low-GI diet and lower CRP in those following the low-CHO diet than in those following the high-GI diet.
• Location: Toronto, Canada.

Key Findings

• Body weight, A1C and fasting insulin did not significantly differ between diets
• There was a main effect of time squared (P=
  0.0005) for body weight, which fell over the first eight weeks and then rose steadily over the remainder of the trial
• Fasting glucose was higher (P=0.041), but two-hour post-load glucose was lower (P=0.010) after 12 months of the low-GI diet
• With the low-GI diet, overall mean triacylglycerol was 12% higher and HDL-cholesterol 4% lower than with the low-CHO diet (P<0.05), but the difference in the ratio of total to HDL-cholesterol disappeared by six months (time x diet interaction, P=0.044)
• Overall mean CRP with the low-GI diet (1.95mg per L) was 30% less than that with the high-GI diet, (2.75mg per L), P=0.0078; while the concentration with the low-CHO diet (2.35mg per L) was intermediate
• There were no significant differences between groups in terms of total cholesterol, LDL-cholesterol or blood pressure.

• In patients with type 2 diabetes mellitus treated with diet alone who have optimal glycemic control, long-term HbA1c was not affected by altering the source or the amount of dietary carbohydrates
• The deleterious effects of the high-carbohydrate diets on the total-to-HDL-cholesterol ratio had disappeared by six months
• The low-GI diet elicited sustained reductions in post-prandial glucose and CRP and for these reasons it may be preferred for the dietary management of T2DM.

Government:

Canadian Institute of Health Research (CIHR-MCT-44205)

The only significant differences among diet groups at baseline were lower LDL-cholesterol in subjects following the high-GI diet than in those following the low-GI diet and lower CRP in those following the low-CHO diet than in those following the high-GI diet.