In vitro dialyzability using meal approach as an index for zinc and iron absorption in humans.

With a standardized protocol under simulated gastrointestinal conditions, 65Zn and 59Fe dialyzability was measured for 38 diets from 7 different published studies on human absorption. The compositions of these diets were available in the form of the amounts of food ingredients used for breakfast, lunch, and dinner. Considering each of these types as a separate meal, percent dialyzability was measured. The weighted average of breakfast, lunch, dinner, and snacks, if any, was taken to represent the entire day's bioavailability. The correlation between in vitro percent dialyzability and reported human absorption was 0.92 for zinc and 0.96 for iron and both were statistically significant (p = 0.0001). The prediction equations for zinc and iron were obtained as y = -0.7718 + 1.1038x and y = 0.3197 + 0.9084x, respectively. This indicated that in vitro dialyzability using the meal approach can be used as an index with good discriminating power for different levels of human absorption.