Human glycemic response curves after intake of carbohydrate foods are accurately predicted by combining in vitro gastrointestinal digestion with in silico kinetic modeling

Background Frequent high blood glucose concentrations are associated with increased risks of metabolic diseases. Knowledge about the glycemic response after food intake is essential in relation to human health. The American Association of Cereal Chemists recommends the development of reliable in vitro methods for standardized assessment of the human glycemic response after intake of carbohydrates. Aim To realize a cost-efficient in vitro–in silico technology to predict reliably the human glycemic concentration curve after intake of different carbohydrate products or meals. Methods We developed and validated a combined technology based on in vitro mastication of foods, digestion of the carbohydrates, availability for absorption of glycemic saccharides, and (based on these in vitro data as input) in silico prediction of glycemic response curves in humans. Results The predicted curves were compared with human clinical data for 22 different food products. The results showed a correlation coefficient for glucose iAUC 0 – 120 and glucose C max of 0.89 and 0.94, respectively. Also the shape of the curves and t max were very similar for 18 out of 22 products, while 4 products showed an ‘early' in vitro t max compared to the human data. Conclusion Based on the demonstrated accuracy and predictive quality, this in vitro–in silico technology can be used for the testing of food products on their glycemic response under standardized conditions and may stimulate the production of (s)low carbs for the prevention of metabolic diseases.