Science

Cara B. Ebbeling, PhD Janis F. Swain, MS, RD Henry A. Feldman, PhD William W. Wong, PhD David L. Hachey, PhD Erica Garcia-Lago, BA David S. Ludwig, MD, PhD

Context: Reduced energy expenditure following weight loss is thought to contribute to weight gain. However, the effect of dietary composition on energy expenditure during weight-loss maintenance has not been studied.
Objective: To examine the effects of 3 diets differing widely in macronutrient composition and glycemic load on energy expenditure following weight loss.
Design, Setting, and Participants: A controlled 3-way crossover design involving 21 overweight and obese young adults conducted at Children’s Hospital Boston and Brigham and Women’s Hospital, Boston, Massachusetts, between June 16, 2006, and June 21, 2010, with recruitment by newspaper advertisements and postings.
Intervention: After achieving 10% to 15% weight loss while consuming a run-in diet, participants consumed an isocaloric low-fat diet (60% of energy from carbohydrate, 20% from fat, 20% from protein; high glycemic load), low–glycemic index diet (40% from carbohydrate, 40% from fat, and 20% from protein; moderate glycemic load), and very low-carbohydrate diet (10% from carbohydrate, 60% from fat, and 30% from protein; low glycemic load) in random order, each for 4 weeks.
Main Outcome Measures: Primary outcome was resting energy expenditure (REE), with secondary outcomes of total energy expenditure (TEE), hormone levels, and metabolic syndrome components.
Results: Compared with the pre–weight-loss baseline, the decrease in REE was great- est with the low-fat diet (mean [95% CI], –205 [–265 to –144] kcal/d), intermediate with the low–glycemic index diet (–166 [–227 to –106] kcal/d), and least with the very low-carbohydrate diet (−138 [–198 to –77] kcal/d; overall P = .03; P for trend by glycemic load=.009). The decrease in TEE showed a similar pattern (mean [95% CI], −423 [–606 to –239] kcal/d; −297 [–479 to –115] kcal/d; and −97 [–281 to 86] kcal/d, respectively; overall P = .003; P for trend by glycemic load < .001). Hormone levels and metabolic syndrome components also varied during weight maintenance by diet (leptin, P