Patients with poorly controlled type 2 diabetes who for 1 year consumed a diet with a low glycemic index had reductions in their hemoglobin A1c level similar to those seen with the diet recommended by the American Diabetes Association, according to research published in Nutrition.
Dr. Yusheng Ma of the University of Massachusetts, Worcester, and colleagues randomly assigned 40 individuals (age range 33–77 years; mean age 54 years) with type 2 diabetes and an HbA1c of 7% or higher either to a diet based on the ADA recommendations (21) or to a diet that instead encouraged consumption of foods with a low glycemic index (19).
Participants received counseling monthly during the first 6 months and again at months 8 and 10, and completed a 7-day dietary recall at each visit.
In the ADA group, 52% of participants were men compared with 42% in the low-glycemic index (GI) group, though the difference was not statistically significant (Nutrition 2008;24:45–6).
Both mean HbA1c level and total cholesterol level decreased over 12 months, with no significant difference between the two dietary groups.
Over the same period, HDL cholesterol increased in both groups. However, the mean LDL cholesterol level dropped from 89 mg/dL at baseline to 71 mg/dL at 12 months in the ADA group, while in the low-GI group, it rose, from 93 mg/dL at baseline to 95 mg/dL at 12 months.
Mean weight dropped in both groups, but mean waist circumference rose in both. Caloric intake in the ADA group was higher at 6 and 12 months while in the low-GI group it was lower at both time points.
Those participants in the low-GI group also were less likely to add or switch diabetes medications than were those in the ADA group (odds ratio 0.26), but this finding was also significantly associated with higher body mass index, high HbA1c level, and male gender.
These results led the investigators to conclude that a low- glycemic index diet is equally able to reduce and control HbA1c and blood lipids as the ADA-recommended diet, which entails carbohydrate counting without regard to glycemic index.
They acknowledged several limitations of the study, including lower attendance at counseling sessions among the low-GI group, and they cautioned that the findings warrant replication in a larger randomized controlled trial.