Effect of Temporal Distribution of Calories on Diurnal Patterns of Glucose Levels and Insulin Secretion in NIDDM

The effect of different temporal patterns of calorie intake on plasma glucose, serum insulin, and insulin secretion rates was examined in six patients with moderately well controlled non-insulin-dependent diabetes mellitus (NIDDM). Patients were studied on three separate occasions over 26 h. Total calories and food composition (50% carbohydrate, 15% protein, and 35% fat) were kept constant, but the pattern of calorie intake was varied. In study A (similar meal size), calories were distributed as 30, 40, and 30% at breakfast, lunch, and dinner, respectively. In study B (3 snacks, 3 meals), each subject ate three meals of 20, 20, and 30% of calories for breakfast, lunch, and dinner, respectively, and three snacks, each comprising 10% of calories, presented 2.5 h after the meal. In study C (large dinner), 10% of calories were consumed at breakfast, 20% at lunch, and 70% at dinner. Glucose, insulin, and Cpeptide concentrations were measured at 15- to 30-min intervals. Insulin secretion rates were calculated from C-peptide levels with individually derived C-peptide clearance parameters. The different eating patterns were associated with only modest differences in overall levels of glucose and insulin secretion. Daytime insulin secretion was lowest when most of the daily calorie intake occurred in the form of a large dinner. Overnight levels of glucose and insulin secretion rates did not differ for the three eating patterns, and the morning glucose levels were also unaffected by the pattern of calorie intake on the previous day. A morning rise of glucose of >0.28 mM occurred consistently only when dinner was of moderate size (30% of total calories). We conclude that if total calorie content and food composition are kept constant, overall blood glucose control is only modestly affected by the temporal pattern of food intake. The partition of food intake into meals and snacks does not improve glucose control, and a large dinner does not result in significantly higher overnight or next-morning glucose levels.