The correlation of in vitro and in vivo assays for determining bioavailable Cd amounts in vegetables is limited. Herein, the correlations between Cd relative bioavailability (Cd-RBA) in rat models and Cd bioaccessibility in four in vitro assays were examined in vegetables. Results showed that the combined liver plus kidney data provided the appropriate endpoint and was used as a biomarker to estimate Cd-RBA. The Cd-RBA was negatively correlated with the mole ratio of Ca/Cd and Fe/Cd in vegetables. Strong in vivo-in vitro correlations were found from physiologically based extraction test (PBET) and in vitro gastrointestinal (IVG) (R2 = 0.66-0.69). We concluded that PBET and IVG were optimal models for Cd-RBA determination in vegetables. The nutritional elements in the vegetables could affect Cd absorption. Furthermore, the Cd bioavailability in vegetables should be considered because risk estimates solely based on the total Cd concentration in vegetables would overestimate Cd intake.
Keywords: bioaccessibility; bioavailability; cadmium; in vivo−in vitro correlation; vegetable.