Prolonged Glycemic Adaptation Following Transition From a Low- to High-Carbohydrate Diet: A Randomized Controlled Feeding Trial

OBJECTIVE Consuming >= 150 g/day carbohydrate is recommended for 3 days before an oral glucose tolerance test (OGTT) for diabetes diagnosis. For evaluation of this recommendation, time courses of glycemic changes following transition from a very-low-carbohydrate (VLC) to high-carbohydrate diet were assessed with continuous glucose monitoring (CGM). RESEARCH DESIGN AND METHODS After achieving a weight loss target of 15% (3%) on the run-in VLC diet, participants (18-50 years old, BMI >= 27 kg/m(2)) were randomly assigned for 10 weeks to one of three isoenergetic diets: VLC (5% carbohydrate and 77% fat); high carbohydrate, high starch (HC-Starch) (57% carbohydrate and 25% fat, including 20% refined grains); and high carbohydrate, high sugar (HC-Sugar) (57% carbohydrate and 25% fat, including 20% sugar). CGM was done throughout the trial (n = 64) and OGTT at start and end (n = 41). All food was prepared in a metabolic kitchen and consumed under observation. RESULTS Glucose metrics continued to decline after week 1 in the HC-Starch and HC-Sugar groups (P < 0.05) but not VLC. During weeks 2-5, fasting and 2-h glucose (millimoles per liter per week) decreased in HC-Starch (fasting -0.10, P = 0.001; 2 h -0.10, P = 0.04). During weeks 6-9, 2-h glucose decreased in HC-Starch (-0.07, P = 0.01) and fasting and 2-h glucose decreased in HC-Sugar (fasting -0.09, P = 0.001; 2 h -0.09, P = 0.003). The number of participants with abnormal glucose tolerance by OGTT remained 10 (of 16) in VLC at start and end but decreased from 17 to 9 (of 25) in both high-carbohydrate groups. CONCLUSIONS Physiological adaptation from a low- to high-carbohydrate diet may require many weeks, with implications for the accuracy of diabetes tests, interpretation of macronutrient trials, and risks of periodic planned deviations from a VLC diet.