Standardisation of in vitro carbohydrate digestion methods for predicting the relative glycemic response to foods : a thesis presented in partial fulfilment of the requirements for the degree of doctor of philosophy in the nutritional sciences, Massey University, Palmerston North campus

Abstract

Global incidence of type II diabetes is driving the need for communication, via foodlabelling, of the likely glycemic effects of foods. In vivo methods for measuring the glycemic response are costly, time-consuming and hence unsuitable for routine food analysis. In vitro carbohydrate digestion methods offer an alternative to in vivo testing. Foods are incubated sequentially with pepsin and pancreatin under simulated in vivo conditions and the pattern of sugar release used as a predictor of the food’s likely glycemic effect. In vitro methods are well-suited to routine food analysis since they are inexpensive, high-throughput and yield highly precise results. Application of in vitro technology is hindered by the lack of standardised methodology. Countless in vitro methods are described in the literature. All differ in their approach to replicating in vivo conditions. It is not known what effect such differences in methodology might exert on relative estimates of glycemic response. A systematic investigation was undertaken to characterise the relative effect of method variables on subsequent in vitro digestion results, using five standard test foods. Variables investigated include mode of comminution, pepsin inclusion versus omission, amylolytic enzyme concentration, incubatum pH and stirring method. A rudimentary framework for a standardised in vitro method is proposed. Comminution and stirring were the method factors most influential to in vitro starch digestion kinetics. Thus, the standardised method features differing approaches to comminution and incubatum stirring depending on the structural properties of the food to be analysed. In vitro methods, in their current format, do not account for the effect of gastric emptying rate on the glycemic response. The glycemic response and gastric emptying rate of 13C-labelled flatbreads containing either 5, 15 or 30 % fat, known to slow gastric emptying, was measured in ten healthy subjects via a GI test and breath testing. The objective was to obtain in vivo data for gastric emptying that might be applied as a correction to parallel in vitro digests of the flatbreads improving their predictive power. Gastric emptying rate reduced significantly with increased flatbread fat content. There was no difference in the glycemic response to each flatbread. Due to the lack of glycemic effect in vivo, no adjustments to in vitro curves could be made.