Effect of blood sampling schedule and method of calculating the area under the curve on validity and precision of glycaemic index values

To evaluate the suitability for glycaemic index (GI) calculations of using blood sampling schedules and methods of calculating area under the curve (AUC) different from those recommended, the GI values of five foods were determined by recommended methods (capillary blood glucose measured seven times over 2.0 h) in forty-seven normal subjects and different calculations performed on the same data set. The AUC was calculated in four ways: incremental AUC (iAUC; recommended method), iAUC above the minimum blood glucose value (AUCmin), net AUC (netAUC) and iAUC including area only before the glycaemic response curve cuts the baseline (AUCcut). In addition, iAUC was calculated using four different sets of less than seven blood samples. GI values were derived using each AUC calculation. The mean GI values of the foods varied significantly according to the method of calculating GI. The standard deviation of GI values calculating using iAUC (20.4), was lower than six of the seven other methods, and significantly less (P<0.05) than that using netAUC (24.0). To be a valid index of food glycaemic response independent of subject characteristics, GI values in subjects should not be related to their AUC after oral glucose. However, calculating GI using AUCmin or less than seven blood samples resulted in significant (P<0.05) relationships between GI and mean AUC. It is concluded that, in subjects without diabetes, the recommended blood sampling schedule and method of AUC calculation yields more valid and/or more precise GI values than the seven other methods tested here. The only method whose results agreed reasonably well with the recommended method (ie. within +/-5 %) was AUCcut.