Intra- and inter-individual variation in glucose response to white bread and oral glucose in healthy women

Abstract

Objectives: To determine within- and between-individual variation in plasma glucose and serum insulin after the intake of the two standards, glucose and white bread, used to calculate the glycaemic index (GI) of foods, and to determine the number of subjects necessary for GI testing. Design and methods: Nine healthy female volunteers randomly took four repeats each of 50 g glucose or 50 g available carbohydrate from white bread. Venous blood samples were drawn before, and every 15 minutes for 1 hour after the test meals, and thereafter every 30 minutes for 1 hour. The incremental areas under the glucose and insulin response curves were calculated using two different methods (AUCmin and AUC0). Coefficients of variation (CV) were measured for each time interval and for the AUCs. The GI of bread, taken on four occasions by the same subjects, was calculated using the mean of four glucose responses as standard. Power calculations were done to determine the number of subjects necessary for determining the GI of foods. Results: No significant differences (p > 0.05) for between- or within-subject variation in plasma glucose were found at any time interval. After glucose, mean plasma glucose was significantly greater at the 15-minute time interval, and the incremental AUCs were significantly greater for glucose. The mean GI of bread varied from 46% to 77%. Power calculations, based on this variation, indicated that if a 10% range for the GI of a food is sought with 80% confidence, between 24 and 90 subjects should participate in GI testing. Conclusion: Patients and clients should be made aware that responses in the same person may differ from week to week, and that ranges for low, medium and high GI foods, rather than specific values, best describe the glycaemic response to foods.