Correlation between in vitro and in vivo data on food digestion. What can we predict with static in vitro digestion models?

During the last decade, there has been a growing interest in understanding food's digestive fate in order to strengthen the possible effects of food on human health. Ideally, food digestion should be studied in vivo on humans but this is not always ethically and financially possible. Therefore, simple in vitro digestion models mimicking the gastrointestinal tract have been proposed as alternatives to in vivo experiments. Thus, it is no surprise that these models are increasingly used by the scientific community, although their various limitations to fully mirror the complexity of the digestive tract. Therefore, the objective of this article was to call upon the collective experiences of scientists involved in Infogest (an international network on food digestion) to review and reflect on the applications of in vitro digestion models, the parameters assessed in such studies and the physiological relevance of the data generated when compared to in vivo data. The authors provide a comprehensive review in vitro and in vivo digestion studies investigating the digestion of macronutrients (i.e., proteins, lipids, and carbohydrates) as well as studies of the bioaccessibility and bioavailability of micronutrients and phytochemicals. The main conclusion is that evidences show that despite the simplicity of in vitro models they are often very useful in predicting outcomes of the digestion in vivo. However, this has relies on the complexity of in vitro models and their tuning toward answering specific questions related to human digestion physiology, which leaves a vast room for future studies and improvements.