The effect of macronutrient composition on energy expenditure and fat balance – is it true that a calorie is a calorie?

EAB logos 3Purpose

Current research and public health policy on obesity is largely based on the hypothesis that the fundamental cause of the condition is an energy imbalance between calories consumed and expended. By this hypothesis, the interaction between diet and body fat is determined by the caloric content of the foods consumed, while the macronutrient content of the diet (the proportion and type of carbohydrates, fats, and protein) has no meaningful effect. This is often summed up by the assertion that a “calorie-is-a-calorie,” shorthand for the hypothesis that a calorie’s worth of protein has an equivalent effect on the accumulation and storage of fat in the human body (on “adiposity”) as does a calorie of carbohydrate or a calorie of fat. An alternative hypothesis is that the macronutrient composition of the diet influences adiposity through its effect on the hormones that regulate uptake of fat (technically “fatty acids”) by fat cells and their subsequent mobilization and use for fuel (that is, oxidation).

This study is one of two studies currently funded by Nutrition Science Initiative (NuSI) that are designed to test this calorie-is-a-calorie hypothesis – in this case, in the context of fixed energy intake under maximum control on human subjects. While studies dating back to the 1930s have attempted to investigate the relationship between dietary macronutrient composition and body fat under controlled conditions, these studies were typically based on untested assumptions and lacked either the rigor or power to provide meaningful results.

This study is designed to utilize the most rigorous environmental and dietary controls to isolate the effect of dietary carbohydrate and fat on metabolism. This study confines participants in a metabolic ward and carefully determines their caloric requirements while they maintain a constant body composition on a standard American diet (50 % carbohydrates, 35% fat and 15% protein). After four weeks of this diet, the participants are shifted to a diet of identical caloric content, but radically different macronutrient composition. In this diet, known as a “ketogenic” diet, the carbohydrate content is reduced to only 5 % of calories; fat content is increased to 80% (protein content remains unchanged). By keeping calories constant while making this radical shift in carbohydrate and fat content, the study constitutes a robust test of the hypothesis that the interaction between diet and body fat is determined by the caloric content of the diet independent of macronutrient composition.

This is a pilot study that will provide a trial run of the methodologies that in turn will be used in a larger, randomized controlled trial of longer duration that will constitute an even more rigorous test of the hypothesis. The results will also provide the necessary data to “power” the follow-up study – that is, to determine how many participants will be necessary for a reliable test of the hypothesis. Because this pilot study may be too short in duration to detect a significant change in fat mass under these conditions, the primary outcome measure of the study is the change in total energy expenditure of the participants during the four weeks on the ketogenic diet. If the restriction of carbohydrates in the ketogenic diet reduces fat mass, the crossover to the ketogenic diet will be accompanied by an increase in energy expenditure.

This study will provide insight into whether the macronutrient composition of the diet – specifically fat and carbohydrate – influences adiposity and energy expenditure, independent of total calories consumed.

Condition: Overweight and obesity
Intervention: Ketogenic diet


Primary outcome

  • Total energy expenditure [Measured weekly]
    Indirect calorimetry

Secondary outcomes

  • Body composition [Measured biweekly]
    Dual-energy X-ray absorptiometry
  • Insulin resistance [Measured weekly]
    Oral glucose tolerance test
  • Changes in circulating hormones [Measured weekly]
    Insulin, glucagon, leptin, thyroid hormones
  • Lipid profile [Measured weekly]
    Liquid chromatography-mass spectrometry
  • Changes in gut microbiome
    Stool analysis
  • Appetitive behavior
    Meal test

Enrollment: 16 subjects
Study start date: 1/5/2014

Assigned interventions

Standard American diet (4-week run-in)
50% carbohydrate, 35% fat, 15% protein (14% of energy as sugar to match the estimated sugar consumption of an average American); caloric intake matched to participants’ energy expenditure

Ketogenic diet (4-week intervention)
5% carbohydrate, 80% fat, 15% protein (0% of energy as sugar); caloric intake equivalent to intake and expenditure at the end of the run-in period, once the subjects are deemed to be in perfect energy balance

Detailed description

A total of 16 overweight/obese subjects, aged 18-50, will be enrolled across four sites: Pennington Biomedical Research Center (Baton Rouge, Louisiana), National Institutes of Health (Bethesda, Maryland), Translational Research Institute (Orlando, Florida), or Columbia University (New York, New York).

Each subject will be monitored as an outpatient for 1 week on a standard American diet at a caloric level estimated to maintain a stable body composition.

As inpatients, participants continue on a standard American diet designed to maintain a stable body composition and energy expenditure – both fat mass and “free-fat mass” remain constant. Patients spend 2 days in a “metabolic chamber” (indirect calorimeter) every week to assess their total energy expenditure. Caloric intake is adjusted after each session, as necessary, to match expenditure. Before moving into the intervention phase, participants will have to be in “energy balance” – expenditure matching intake and so a stable body composition – for at least 2 weeks.

During the 4 weeks of the intervention phase, subjects are fed a ketogenic diet with caloric intake matched to their intake and expenditure at the end of the run-in period. Each participant’s total energy expenditure will then continue to be assessed weekly and compared to that of the completion of the run-in.

If the interaction of diet and adiposity is dependent only on caloric content, subjects should show no change in energy expenditure when they cross over from the standard American diet to the ketogenic diet. If the macronutrient content of the diet influences fat mass – specifically the carbohydrate content – fat balance should be negative during the 4 weeks of the ketogenic diet phase (more fat should be oxidized than is consumed) and energy expenditure, the primary endpoint, should increase.

These results will then be used to determine the number of participants and the study design of Phase II, which will randomize allocation of participants to multiple diets and for longer periods of time. Phase II will be designed with the expectation that any differences in body mass and composition will be detectable with the longer duration of the intervention. This study will provide critical insight into the relationship between the macronutrient content of the diet, body composition, and energy expenditure.

Subject criteria

Body mass index 25-35
Aged 18-50


Pennington Biomedical Research Center
Baton Rouge, Louisiana, 70808

National Institutes of Health, NIDDK
Bethesda, Maryland, 20800

Translational Research Institute
Orlando, Florida, 32804

Columbia University
New York, New York, 10027


Pennington Biomedical Research Center
Eric Ravussin, Ph.D. (Co-principal investigator)

National Institutes of Health, NIDDK
Kevin Hall, Ph.D. (Co-principal investigator)
Marc Reitman, M.D., Ph.D.
Kong Y. Chen, Ph.D.

Translational Research Institute for Metabolism and Diabetes
Steven R. Smith, M.D.

Columbia University
Rudolph Leibel, M.D.
Michael Rosenbaum, M.D.
Laurel Mayer, M.D.
B. Timothy Walsh, M.D.

University of California, San Francisco
Peter J. Turnbaugh, Ph.D.

  • Donate or get involved Connect and talk to us