American Journal of Clinical Nutrition, Vol 64, 78-86, Copyright © 1996 by The American Society for Clinical Nutrition, Inc

ORIGINAL RESEARCH COMMUNICATIONS

Thermic and metabolic responses to oral glucose in obese subjects with non-insulin-dependent diabetes mellitus treated with insulin or a very- low-energy diet

R Gougeon
McGill Nutrition and Food Science Centre, Royal Victoria Hospital, Montreal, Canada.

Increased resting energy expenditure (REE) and a blunted thermic effect of glucose (TEF) have been reported in obese subjects with non-insulin- dependent diabetes mellitus (NIDDM). I questioned whether the abnormal REE and TEF would be corrected by normalizing glycemia with insulin or a very-low-energy diet (VLED). Three male and four female obese subjects with NIDDM [weighing 108 +/- 6 kg and with body mass index (in kg/m2) of 39 +/- 2] received a weight-maintaing formula diet containing 95 g protein/d for 15 d then a 1.7-MJ, 93-g-protein VLED for 27 d. Insulin was given from days 1 to 8 in doses sufficient to normalize glycemia. REE was measured weekly and TEF was measured on days 8 and 15 of isoenergetic feeding and 28 d after the VLED by using a ventilated- hood indirect calorimeter. Weight decreased 9.8 +/- 1 kg during the VLED. REE was 3% lower with insulin treatment than during hyperglycemia (7878 +/- 364 compared with 8125 +/- 381 kJ/d, P = 0.002). REE decreased by 20% to 6494 +/- 280 kJ/d by week 4 of the VLED. After 112 g oral glucose, increments in energy expenditure were significantly greater during isoenergetic feeding with insulin than without (7.5 +/- 1.3% compared with 4.3 +/- 0.9% above REE) and after the VLED (10.5 +/- 1.0% above REE, P < 0.05). Plasma glucose excursions were greatest without exogenous insulin (peak 21.5 +/- 1.8 mmol/L at 120 min, 16.3 +/- 1.9 mmol/L at 225 min). Plasma fatty acid excursions were the lowest with insulin treatment. The integrated plasma glucose and fatty acid responses above baseline did not differ among studies; the integrated insulin and C-peptide responses were greater after the VLED. Cumulative nonoxidative glucose disposal (stored glucose) was higher with insulin therapy than without (52 +/- 6 compared with 35 +/- 7 g/210 min, P < 0.05) and increased significantly to 66 +/- 6 g after the VLED (compared with the isoenergetic diet without insulin). TEF correlated significantly with integrated C-peptide and insulin responses. The percentage increase in TEF with euglycemia (with insulin and VLED) correlated with the percentage increase in stored glucose (P < 0.05). The greater TEF was associated with a greater insulin response, which was probably responsible for the greater stored glucose.