We aimed to study the effects of chronic ingestion of short-chain fructooligosaccharides (FOS), an indigestible carbohydrate, on hepatic glucose production, insulin-mediated glucose metabolism, erythrocyte insulin binding, and blood lipids in healthy subjects. Twelve healthy volunteers received either 20 g FOS/d or sucrose for 4 wk in a double-blind crossover design. FOS did not modify fasting plasma glucose and insulin concentrations. Mean (+/- SEM) basal hepatic glucose production was lower after FOS than after sucrose consumption (2.18 +/- 0.10 compared with 2.32 +/- 0.09 mg.kg-1, min-1, respectively; P < 0.02, paired Student's t test). However, neither insulin suppression of hepatic glucose production nor insulin stimulation of glucose uptake measured by hyperinsulinemic clamp was significantly different between the two dietary periods. Erythrocyte insulin binding was also comparable. Serum triacylglycerols, total and high-density- lipoprotein cholesterol, apolipoproteins A-I and B, and lipoprotein(a) were not modified by FOS. To try to understand why FOS did not increase serum lipids, the in vitro production of short-chain fatty acids from FOS was evaluated by using human fecal inoculum and compared with that from lactulose, which was found to increase serum lipids. FOS produced an acetate-propionate ratio two times lower than that of lactulose. We conclude that 4 wk of 20 g FOS/d decreased basal hepatic glucose production but had no detectable effect on insulin-stimulated glucose metabolism in healthy subjects. The colonic fermentation pattern of undigestible carbohydrates may be relevant to predicting their metabolic effects.

Objective: To evaluate the in¯uence of three dietary protein types (casein, gelatin, soy protein) on satiety and food intake, at two levels of loading (total energy of test meals: 3.6 or 1.8 MJ). Design: The study employed a repeated measures design. Test meals were controlled for energy, macronutrients, ®ber and palatability, and contained about 23% energy as protein (of which about 65% was experimentally manipulated). Postprandial subjective satiety and hunger, plasma glucose, insulin and glucagon were assessed for 8 h, and energy and macronutrient intakes were monitored for 24 h. Subjects: Nine healthy normal-weight men. Results: No effect of the type of protein on 24 h energy and macronutrient intakes was observed despite a signi®cant effect of protein source on the kinetics of peripheral metabolic responses (but only after 3.6 MJ lunches), and inconsistent effects on subjective hunger and satiety responses A casein-enriched lunch delayed glucose and insulin responses for 1.5 h, compared with soy protein, probably due to a lag in gastric emptying. Conclusion: Varying the protein source in a mixed meal modi®es glucose, insulin and glucagon kinetics in healthy men, but these variations in satiety-implicated factors have inconsistent effects on subjective satiety and fail to affect food intake.

MYARA, ISAAC, CATHERINE ALAMOWITCH, ODILE MICHEL, DIDIER HEUDES, JEAN BARIETY, BERNARD GUY-GRAND, AND JACQUES CHEVALIER. Lipoprotein oxidation and plasma vitamin E in nondiabetic normotensive obese patients. Obes Res. 2003;11:112-120. Objective: To correlate the susceptibility of low-(LDL) and very-low-density lipoprotein to oxidation in vitro and the concentrations of serum antibodies against malondialdehyde-modified LDL and plasma vitamin E with the anthropometric and laboratory characteristics of obesity. Research Methods and Procedures:A total of 75 nondiabetic, normotensive obese patients were assigned to one of four groups according to their body mass index (BMI): moderately obese (30 Յ BMI Յ 34.9 kg/m 2 , n ϭ 11), severely obese (35 Յ BMI Յ 39.9 kg/m 2 , n ϭ 20), morbidly obese (40 Յ BMI Յ 50 kg/m 2 , n ϭ 29), and very severely obese (BMI Ͼ 50 kg/m 2 , n ϭ 15). Results: The oxidation lag time for LDL from patients with a BMI Ն35 kg/m 2 was shorter than that for LDL from non-obese controls (n ϭ 13), whereas very-low-density lipoprotein oxidation lag times were not significantly different. The serum antibodies against modified LDL were similar in all groups, whereas the plasma vitamin E concentrations of obese patients were decreased (p Յ 0.01). There was a negative correlation between LDL oxidation lag time and BMI (r ϭ Ϫ0.35, p ϭ 0.0008), and between plasma vitamin E and BMI (r ϭ Ϫ0.53, p Ͻ 0.0001) and waist-tohip ratio (r ϭ Ϫ0.40, p ϭ 0.0003). Discussion: The LDL of nondiabetic, normotensive obese patients is more readily oxidized, and plasma vitamin E concentrations are low. These are both risk factors for coronary heart disease.

The results validate the OC-Clamp method to monitor the posthepatic appearance of glucose after carbohydrate ingestion and to discriminate between different carbohydrate sources. The results suggest that the technique be used in humans.