Simulation of Human Small Intestinal Digestion of Starch Using an In Vitro System Based on a Dialysis Membrane Process

Abstract

This work deepens our understanding of starch digestion and the consequent absorption of hydrolytic products generated in the human small intestine. Gelatinized starch dispersions were digested with α-amylase in an in vitro intestinal digestion system (i-IDS) based on a dialysis membrane process. This study innovates with respect to the existing literature, because it considers the impact of simultaneous digestion and absorption processes occurring during the intestinal digestion of starchy foods and adopts phenomenological models that deal in a more realistic manner with the behavior found in the small intestine. Operating the i-IDS at different flow/dialysate flow ratios resulted in distinct generation and transfer curves of reducing sugars mass. This indicates that the operating conditions affected the mass transfer by diffusion and convection. However, the transfer process was also affected by membrane fouling, a dynamic phenomenon that occurred in the i-IDS. The experimental results were extrapolated to the human small intestine, where the times reached to transfer the hydrolytic products ranged between 30 and 64 min, according to the flow ratio used. We consider that the i-IDS is a versatile system that can be used for assessing and/or comparing digestion and absorption behaviors of different starch-based food matrices as found in the human small intestine, but the formation and interpretation of membrane fouling requires further studies for a better understanding at physiological level. In addition, further studies with the i-IDS are required if food matrices based on fat, proteins or more complex carbohydrates are of interest for testing. Moreover, a next improvement step of the i-IDS must include the simulation of some physiological events (e.g., electrolytes addition, enzyme activities, bile, dilution and pH) occurring in the human small intestine, in order to improve the comparison with in vivo data.