Low-Carb Diet Trumps Caloric Restriction in Type 2 Diabetes

WASHINGTON — Patients with type 2 diabetes may be able to improve glycemic control, weight loss, and cholesterol levels significantly more with low-carbohydrate diets than with diets emphasizing a caloric deficit, according to the results of two randomized trials presented at the annual scientific sessions of the American Diabetes Association.

Uncontrolled pilot studies of low-carbohydrate diets (ranging from 7% to 20% of total calories) in patients with type 2 diabetes have shown declines in HbA_1c values ranging from 1.2 to 4.1 percentage points and decreases in weight of 7%–12% in up to a year of follow-up.

With such promising results in mind, Dr. Eric C. Westman and colleagues at Duke University, Durham, N.C., conducted a comparison of a low-glycemic index diet (LGI) with a low-carbohydrate, ketogenic diet (LCKD) in 97 type 2 diabetes patients, average age 50–51 years, who wanted to lose weight.

To be included in the trial, patients had to have no history of diabetic ketoacidosis, have an HbA_1c of greater than 6% at screening, and be able to self-monitor blood glucose levels if on insulin or oral hypoglycemic therapy. Patients were excluded if they had unstable cardiovascular disease (allowable if hypertension and hyperlipidemia were under control), kidney disease, or liver disease, or were pregnant or breast-feeding.

Participants in each group received exercise recommendations and nutritional supplementation (a multivitamin, chromium 1,000 mcg/day, and vanadyl sulfate 60 mg/day) and attended group meetings. The LGI diet consisted of an individualized 500-kcal reduction per day in energy intake, deriving 55% of energy from low-glycemic carbohydrates, whereas people on the LCKD restricted their carbohydrate intake to less than 20 g per day.

Among the 50 participants who completed the 24-week study, HbA_1c levels improved significantly more in the LCKD group (8.8% to 7.3%) than in the LGI group (8.3% to 7.8%). When dropouts were included in analyses using the last observation, the LCKD group still had a significantly greater, albeit smaller, HbA_1c improvement than the LGI group did (8.4% to 7.6% vs. 8.3% to 8%).

In a multivariate regression model, treatment group was independently predictive of change in HbA_1c after weight loss was controlled for “suggesting that [the LCKD] diet, even beyond weight loss, had an effect on HbA_1c,” said Dr. Westman, professor in the department of medicine at Duke University.

By the end of the trial, 18 of the 29 people who completed the LGI diet had eliminated or reduced the number of medications they were on, compared with 20 of 21 individuals in the LCKD group. Body weight changed significantly more among LCKD participants (107 kg to 97 kg) than in LGI participants (104 kg to 98 kg).

Both groups had similar improvement in total and low-density lipoprotein cholesterol as well as triglycerides. LCKD participants had a significant 5.5% increase in high-density lipoprotein cholesterol (HDL-C), compared with no change in the LGI individuals.

The study was funded by the Robert C. Atkins Foundation. Dr. Atkins is credited with popularizing a low-carbohydrate diet.

In a separate 2-year trial, an interim analysis of 6 months of data showed that a diet restricted to about 70 g of carbohydrates (half from fruit and milk) significantly improved weight and waist circumference in patients with type 2 diabetes, compared with a diet with a 500-kcal per day calorie restriction, reported Dr. Michael E. Daly, a diabetologist at the Royal Devon and Exeter NHS Trust, Exeter, England.

Six-month data were available for 206 of 259 (80%) patients randomized in the trial. This 20% dropout rate was lower than that of other low-carbohydrate studies, Dr. Daly said (N. Engl. J. Med. 2003;348:2074–81; 2082–90). Participants had an average age of 59, a mean weight of 99 kg, and a mean BMI of 35 kg/m

Significantly more weight loss occurred in carbohydrate-restricted patients (3.7 kg) than in those on the energy-deficit diet (1.2 kg). Both diets improved body fat percentage, HbA_1c, total cholesterol:HDL cholesterol ratio, triglycerides, and blood pressure, but no differences emerged between the two groups on any of these measures. In both groups, calcium excretion declined to a similar extent and microalbumin excretion did not change substantially (Diabet. Med. 2006;23:15–20).