Varying dietary protein with overfeeding in a metabolic ward

A new paper by George Bray and colleagues in JAMA examines the effect of altering dietary protein content on weight gain, energy expenditure, and body composition in response to overeating in metabolic units. The rationale being to test the hypothesis of metabolic inefficiency on a low or high protein diet that existing research has suggested. Scroll down for a video summary from Bray.

25 healthy people with BMIs between 19 and 30 completed the study who spent 10-12 weeks in a metabolic unit (what fun) where food was provided. They were randomized to 3 groups: diets containing 6% energy from protein (52% fat, 42% carbohydrate) (low pro), 15% from protein (44% fat, 41% carbohydrate) (normal pro), or 26% energy from protein (33% fat, 41% carbohydrate) (high pro). These diets were given during the last 8 weeks of their stays, which provided on average 39.4% (954 kcal/day) greater intake than their weight stabilized intakes (determined in the first couple weeks).

Resting energy expenditure was measured each week with a ventilated hood system, total daily energy expenditure during baseline and weeks 7-8 via doubly labeled water, and body composition at baseline and biweekly with DXA. Body weight, lean body mass, fat mass, total daily energy expenditure, and resting metabolic rate were not significantly different at baseline between the 3 groups.

The weight gain after the 8 weeks of overfeeding resulted in a 3.16 kg weight gain in the low pro group, significantly less than the 6.05 kg in the normal pro and 6.51 kg in the high pro groups. This is explained by the reduction in lean body mass by 0.70 kg in the low pro group compared to gains of 2.87 kg and 3.18 kg in the others. The average fat gain was not significantly different between the 3 groups (P=.89). Resting energy expenditure increased in the normal pro and high pro groups but not in the low pro. Likewise, total energy expenditure increased in both the normal and high pro groups compared with the low pro. These likely reflect the increased cost of protein turnover/storage. Here are the graphs for resting and total EE:

Metabolic efficiency (excess energy/weight gain) was higher in the low pro group than the high pro group, but adjustment for age, sex, and baseline weight removed the significance (P=.15).

They show that the calories consumed predicts the increase in lean body mass and body fat but protein intake fails to predict the body fat changes.

So, calories matter more than protein in an overfeeding context on body fat, but low protein reduces lean body mass.

Edit: this study used a fixed amount of calories for overfeeding, but we know from research that protein is critical to satiety and weight maintenance in lesser controlled contexts. Just want to make sure nobody misinterprets.

Also see JAMA’s summary of this work here.  NPR also has a well written article here.

Here is a video summary from Bray:


George A. Bray, Steven R. Smith, Lilian de Jonge, Hui Xie, Jennifer Rood, Corby K. Martin, Marlene Most, Courtney Brock, Susan Mancuso, & Leanne M. Redman (2012). Effect of Dietary Protein Content on Weight Gain, Energy Expenditure, and Body Composition During Overeating JAMA : 10.1001/jama.2011.1918

  • Julie

    I would imagine protein levels might affect lean mass even more if calories were in deficit?  That might shed some light on skinny-fat people, of which I know a few.  And one very strange guy who likes some anti-cancer quack, thinks broccoli is high-protein food, and has so little muscle tone that he looks about 20 years older.

  • Arielevehealth

    I. I wonder what or if there would have been a difference in the result of the low protein diet if the fat percentage had also been low (5-10 percent), similar to the 80/10/10 approach?

  • Endodoc

    Dr. Bray, are you tracking these volunteers to see how quickly they lose weight and what happens to their % fat and % lean body mass as they return to usual diets?