Low-Carb Ketogenic Diet Beats Low-Glycemic Index Diet in Overweight Type 2 Diabetes


Avoid the needle with a low-carb ketogenic diet

Duke University (U.S.) researchers demonstrated better improvement and reversal of type 2 diabetes with an Atkins-style diet, compared to a low-glycemic index reduced-calorie diet.


Ninety-seven overweight and obese adults, 78% women and 40% black, were randomly assigned to either:

  • a very low-carb ketogenic diet (Atkins induction phase, as in Atkins Diabetes Revolution) or
  • a low glycemic-index index calorie-restricted diet (The GI Diet by Rick Gallop). 

Thirty-eight were in the Atkins group; 46 in the low-glycemic index (low-GI) group.  Seventeen dropped out of each group before the end of the 24-week study.  Average weight was 234.3 pounds (106.5 kg); average body mass index was 37.  The Atkins group averaged 13% of total calories from carbohydrate; the low-GI cohort averaged 44%. 


Both groups lost weight and had improvements in hemoglobin A1c, fasting insulin, and fasting glucose. 

The Atkins group lowered their hemoglobin A1c by 1.5% (absolute drop, not relative) versus 0.5% in the other group. 

The Atkins group lost 11.1 kg versus 6.9 kg in the other group. 

The Atkins group increased HDL cholesterol by5.6 mg/dl versus no change in the other group. 

All the aforementioned comparisons were statistically significant. 

Diabetes medications were stopped or reduced in 95% of the Atkins group versus 62% of the low-GI group.

Total and LDL cholesterol levels were unchanged in both groups. 

Triglycerides fell significantly only in the Atkins group.

My Comments

You may be interested to know that this study was funded by the Robert C. Atkins Foundation.

One strength of this study is that it lasted for 24 months.  Many similar studies last only eight to 12 weeks.  A drawback is that, with all the drop-outs,  the number of participants is low. 

The GI Diet performed pretty well, too, all things considered.  Sixty-two percent reduction or elimination of diabetes drugs—not bad.  For a six-year-old book, it’s still selling fairly well at Amazon.com.  That may be why they chose it as the comparison diet.

The diet with fewer carbohydrates—Atkins induction—was most effective for  improving control of blood sugars.  So effective, in fact, that the researchers sound a note of warning:

For example, participants taking from 40 to 90 units of insulin before the study were able to eliminate their insulin use, while also improving glycemic control.  Because this effect occurs immediately upon implementing the dietary changes, individuals with type 2 diabetes who are unable to adjust their own medication or self-monitor their blood glucose should not make these dietary changes unless under close medical supervision.  

[Not all insulin users were able to stop it.]

Overall, lipids were improved or unchanged in the Atkins group, despite the lack of limits on saturated fat intake.  A common criticism of the Atkins diet is that it has too much saturated fat, leading to higher total and LDL cholesterol levels, which might raise long-term cardiovascular risks.  Not so, here. 

When you reduce carbohydrate intake, the percentages of fat and protein in the diet also change.  In this Atkins diet, protein provided 28% of daily calories, and fat 59%.  In the low-GI diet, protein provided 20% of daily calories, fat 36%.  The beneficial effects of the Atkins diet probably reflect the low carbohydrate consumption rather than high protein and fat. 

The Atkins induction-phase diet was clearly superior to the low-glycemic index diet in this overweight diabetic sample, without restricting calories.

Steve Parker, M.D.

Reference:  Westman, Eric, et al.  The effect of a low-carbohydrate, ketogenic diet versus a low-glycemic index diet on glycemic control in type 2 diabetes mellitusNutrition & Metabolism 2008, 5:36   doi:10.1186/1743-7075-5-36

Additional Reading

Samaha, F., et al.  A low-carbohydrate as compared with a low-fat diet in severe obesity.  New England Journal of Medicine, 348 (2003): 2,074-2,081.

Boden, G., et al.  Effect of a low-carbohydrate diet on appetite, blood glucose levels, and insulin resistance in obese patients with type 2 diabetes.  Annals of Internal Medicine, 142 (2005): 403-411.

Vernon, M., et al.  Clinical experience of a  carbohydrate-restricted diet: Effect on diabetes mellitus.  Metabolic Syndrome and Related Disorders, 1 (2003): 233-238.

Yancy, W., et al.  A pilot trial of a low-carbohydrate ketogenic diet in patients with type 2 diabetes.  Metabolic Syndrome and Related Disorders, 1 (2003): 239-244.


Filed under Carbohydrate, Glycemic Index and Load, ketogenic diet, Overweight and Obesity

3 responses to “Low-Carb Ketogenic Diet Beats Low-Glycemic Index Diet in Overweight Type 2 Diabetes

  1. Steve

    I hate these results lol 🙂

    Honestly – I’m confused. Here I am, trying to eat AMD style (after a couple years of learning about med. benefits and AMD giving me a “plan” to do it) – thinking it’s the “best way”. After all, healthy cultures actually eat this way!

    Now all I’m hearing is:
    (a) Low carb is going to be healthier for me
    (b) Low carb will help me lose weight
    (c) Saturated fat is nothing to worry about (sorry Ancel Keys)

    What the?

    It’s simply frustrating. I know that science, as you have said, often gives contradictory results. But still!

    And here I was, thinking that AMD style was (as Simopoulos recommends in the Omega Diet) “moderate carbs”
    After all (on 1900 calorie plan) – 4 serves of grain carbs is not a huge amount! Unless you count fruit as well – then you could call it high carb? But who said fruit is bad for you??


  2. I know what you mean, Steve!

    Overall, a Mediterranean-style diet has the best evidence to support it as the healthiest way for most people to eat. Very low-carb diets haven’t been put to the test in terms of long-term observational studies looking at longevity and chronic disease rates.

    I’m working on identifying the healthiest amounts and types of fruits, vegetables, legumes, and whole grains, while avoiding adverse effects on blood sugar and insulin levels. These are issues primarily for people with type 2 diabetes.

    It’s going to be a balancing, high-wire act for some them. Too much carb: glucose levels out of control, with diabetic complications. Too little carb: miss the health benefits of fruits,veggies, legumes, and whole grains.

    No easy answers.


  3. Steve

    Thanks Steve.

    Maybe this “nutrogenomics” thing I’m reading about is the answer? i.e. We are all different and react to things differently? It’s all a gene thing?

    Still, I take a good amount of comfort in simply recognising 2 things:

    (a) Med diet has so much cultural and scientific evidence, as you say.
    (b) I feel better when I eat that way!!! Fish, legumes, nuts, fruits, veg, whole grains etc all make me feel good!

    Maybe it’s the European in me (from my mother’s side) that leans me toward that way of eating anyway, wich could support the gene thing!

    Thanks for the encouraging ‘reminder’ – the Med diet has the best evidence thus far – I needed that 🙂