New fuel for the calorie debate

The big news in obesity this week is the publication of this paper Effects of Dietary Composition on Energy Expenditure During Weight-Loss Maintenance, by Ebbeling et al. in JAMA. The study examines the effects of three types of diets – low fat, low carbs, and low glycemic index – on energy expenditure and weight loss maintenance. It was a cross-over study on 21 obese young adults, where all three diets were given to each subject consecutively. The basic result of the study was that people on the low carb (i.e. Atkins) diet had the highest total energy expenditure (TEE), followed by the low glycemic index, and coming in last place was the low fat diet. The study certainly bolsters the claims of  those on the “calorie is not a calorie” and  “carbs are bad” side. The implication being that you can eat more on a low carb diet than on a low fat diet. While the study was carefully done, there are some discrepancies that may call into question some of the results. Here is my colleague Kevin Hall’s take:

The resting RQ values in Table 3 seem too high. Generally, resting RQ should be lower than 24hr RQ which should match the daily food quotient (FQ) when the subjects are in macronutrient balance. My rough calculation of the FQ values for the test diets give about 0.9, 0.84, and 0.76 for the LF, LGI, and VLC diets, respectively. The reported resting RQ values are 0.9, 0.86, and 0.83. This would usually suggest a degree of overfeeding during the weight loss maintenance phases since RQ generally exceeds FQ during positive energy balance. However, the reported energy intake during the test diet phase was 2626 +/- 686 kcal/d which is lower than the TEE in all 3 test diets reported in Table 3. Something is odd here.

For those not up on metabolism lingo, the RQ is the respiratory quotient, which is the ratio between carbon dioxide expired and oxygen inhaled and gives a measure of what types of fuel the body is burning. The RQ works because carbs, fat and protein are all oxidized slightly differently.  Carbs have an RQ of 1 meaning every mole of oxygen consumed produces a mole of carbon dioxide.  Protein has an RQ that is between 0.8 and 0.9 and fat has an RQ around 0.7. Different types of proteins and fats will have slightly different RQs. The FQ is the expected RQ given the diet. The resting RQ values were measured by wearing a mask that measures the air you breath for twenty minutes after an overnight fast. Generally, when you fast, your body switches from burning carbs to burning fat. Hence, fasting RQs are usually lower than FQs. However, in this study the opposite is true. Fasting RQs can be higher than FQs if the person is not in energy balance and increasing in weight. However, the study also reports that the energy intake was lower than the TEE. Their estimates for TEEs are based on results from doubly labelled water measurements which uses the reported RQs as one of the input paramters. Hence, the differences in TEE that they report could be explained by experimental error.