The context of this study is that lower energy expenditure resulting from weight loss is considered to contribute to ulterior weight gain.
This is something that most people engaged in dieting or who want to lose weight through one way or the other fail to remember. As you lose weight, your energy expenditure tends to go down. There are cases when this does not apply, but in most of the cases it does.
One may think that eating X amount of kcals/day and doing some exercise is a rigid strategy that will apply continuously. But as weight goes down, that X amount of kcals/day may need to be lowered as well if one wants to see on-going weight loss.
This is a very simplified context that only considers food and exercise as the tools, but the equation of weight is by far more complex. And very few people seem to get it, at least partially correct.
Low-Fat vs. Low-Glycemic vs. Very-Low Carb.
I do not want to start by criticizing the article. The authors acknowledge its short-comings and its strong points very well (see the ending of the study). I want to focus on the important lessons we can draw from it.
They wanted to examine 3 different diets in 21 overweight and obese subjects. They made the subjects lose 10-15% of their weight through a run-in diet.
The selection of the subjects was rigorous. It started with 681 participants who were screened by telephone, out of which 547 were excluded because they didn’t meet the required BMI, body weight, or because of other reasons.
The remaining 134 were screened in person, out of which 102 were excluded. The remaining 32 began the run-in diet and 8 were excluded.
The remaining 24 were randomly assignment one of the 3 diets (low-fat, low-glycemic, very-low carb), but only 21 were able to comply with all the requirements and be included in the study.
So I would say this study was well designed in terms of screening and selecting the participants.
The 3 diets:
Low-Fat: 60% carbs, 20% fat, 20% protein. High-glycemic diet.
Low-Glycemic: 40% carbs, 40% fat, 20% protein.
Very-Low-Carb: 10% carbs, 60% fat, 30% protein.
Duration: 4 weeks.
Participants: men and women, 18-40 years old, BMI >= 27.
Here’s a more accurate view of the diets (please excuse my interventions).
What’s nice about this study is that the subjects were compensated with $$$:
$500 for those who ended the run-in diet and had at least 10% weight loss, and
$2,000 for those who completed the study.
I’d personally enroll in a study that rewards me with $2,500. 🙂
Here are the results:
Interpretations (+my personal interpretation)
The researchers acknowledge that different diets can elicit different metabolic outcomes:
However, because metabolic pathways vary in energetic efficiency, dietary composition could affect energy expenditure directly by virtue of macronutrient differences or indirectly through hormonal responses to diet that regulate metabolic pathways. (Ebbeling et al., 2012)
They specifically point to the implications of low blood glucose.
As I said above, many people fail to encompass that energy expenditure seems to decrease as one loses weight. However, as you can see in the results this is not valid for the very-low carb diet.
The decrease in REE (resting energy expenditure) from pre–weight-loss levels, measured by in-direct calorimetry in the fasting state, was:
greatest for the low-fat diet (mean of –205 kcal/d),
intermediate with the low–glycemic index diet (mean of –166 kcal/d), and
least for the very low-carbohydrate diet ( mean of −138 kcal/d);
The decrease in TEE (total energy expenditure), assessed using the doubly-labeled water method, also differed significantly by diet:
−423 kcal/d for low fat;
−297 kcal/d for low glycemic index; and
−97 kcal/d for very low carbohydrate;
In my interpretation, this means that the very-low-carbers spent more energy at rest and had a greater TEE compared to the other two dietary protocols.
Leptin levels were higher in the low-fat group and lower in the very-low-carb group. Though, we are talking about overweight and obese subjects here so we do not know about their leptin sensitivity.
On the other hand, insulin sensitivity was much better in the very-low-carb group compared to the low-fat group. I’m not mentioning the low-glycemic group because most of their results seem to be in-between the results of the other two groups.
One of the non-surprising facts is that even though the very-low-carb group ate ~4 times more saturated fat compared to the low-fat group, the lipid profile of the very-low-carb group was better than that of the low-fat group (see HDL and TAG in the results table).
The researchers warn about the higher urinary cortisol and lower thyroid function (measured by TSH and T3) in the very-low-carb group. I didn’t find consistent results in other studies when it comes to urinary cortisol excretion (though I may have not researched it enough). Plus, we should not forget that we are dealing with overweight and obese people and their hormonal panels may not be as good-looking as their leaner counterparts.
When it comes to the lower thyroid function, I would challenge the general thinking on what “lower thyroid function” is. I wrote about my thoughts on this subject here (it’s a long post – skip to the conclusion section if you have a very-narrow-attention-span).
The researchers also warn about the higher CRP levels in the very-low-carb group. But I would say the difference between the low-fat group and very-low-carb group is very small (~0.78 vs. 0.87 mg/L), both mean values are lower than 1 mg/L and they both point-out to lower levels of inflammation in the body. When compared to the run-in diet (~1.75 mg/L), both values are excellent.
Personally, my HS-CRP was extremely low after 1 year of strict ketosis.
I would urge not to make generalizations from this study. Even though it supports some of the low-carb theories that have been well-voiced recently, stronger conclusions can be drawn if they are based on many more studies done with many more methods, approaches, and subjects.
Let’s try to take the good out of it (experiment with it individually), learn from the bad, and leave out the rest.
The good, the bad, and the rest are specific to the individual. What are your personal good and bad take-away messages from this study? Please share your thoughts in the comment sections below and let’s try to learn from each other.
Ebbeling, C. B., Swain, J. F., Feldman, H. A., Wong, W. W., Hachey, D. L., Garcia-Lago, E., & Ludwig, D. S. (2012). Effects of dietary composition on energy expenditure during weight-loss maintenance. Jama, 307(24), 2627-2634.
Photos: here and here