The effects of a daily intake of 30 g fructose on blood glucose regulation, erythrocyte insulin receptors, and lipid metabolism have been studied in type II (non-insulin-dependent) diabetic subjects. Eight well-controlled patients received, in a randomly assigned crossover design over two 2-mo study periods, 30 g of fructose in exchange for an isocaloric amount of starch. Fructose could be taken at any time during the day as part of the 1400-1600 kcal allowed diet (50% carbohydrate, 30% fat, 20% protein). No significant difference was observed concerning body weight, HbA1c, fasting plasma glucose, fasting plasma insulin, uric acid, total cholesterol, high-density lipoprotein cholesterol, and triglycerides, nor was there any change in insulin binding to erythrocytes between the fructose and the control starch period. However, the mean plasma triglyceride levels after the fructose period, although still in the normal range, were significantly higher than baseline values (P less than .05). We conclude that moderate amounts of fructose incorporated into the diet of well-controlled type II diabetic subjects have no significant deleterious effect on glycemic control, insulin receptors of erythrocytes, or lipid metabolism.

The effects of mixed meals containing varying amounts of carbohydrate (CHO) on blood glucose levels and insulin delivery by an artificial pancreas were studied in seven insulin-dependent diabetes mellitus subjects. Each patient received, at random over 3 consecutive days, three mixed meals containing 60, 80, and 140 g complex CHOs. There was a high and linear correlation between total amount of insulin delivered to restore blood glucose values and amount of CHO consumed: 12.1 +/- 1.3 to 31.2 +/- 5.2 U insulin were needed for 116 +/- 16 to 198 +/- 24 min. However, neither the time lapse between the beginning of meal intake and blood glucose increase nor the peaking time for blood glucose variation were significantly different between meals. We suggest that some of the data obtained in this study might be useful in programming an open-loop insulin-infusion system.