This study aimed (i) to design an in vivo model to study fat digestibility, and (ii) to apply this design to test the in vivo digestibility of a highly thermoxidized olive oil. True digestibility of unheated olive oil was tested 2, 4, 6, and 7 h after administering 1 g of olive oil/100 g body weight to young adult Wistar rats by means of esophageal probes. Remaining gastrointestinal lumen fat showed an inversely linear relationship (r= -0.9932; P < 0.001) with the length of the experiment. A 4-h test was considered adequate because after this period, half of the oil administer still remains in the lumen, making it possible to accurately measure the different nondigested, nonabsorbed thermoxidized compounds. In a second experiment, fresh olive oil (3.6 mg polar content/100 mg oil) was heated at 180 degrees C for 50 h in the presence of air; the polar content in this oil rose to 46.0 mg/100 mg oil. After 4 h, the true digestibility coefficient of 50-h heated olive oil did not significantly change, although it tended to decrease (24%) with respect to the unheated oil. Silica gel column chromatography and high-performance size exclusion chromatography were used to quantify nonthermoxidized and thermoxidized products present in the oils and in the gastrointestinal lumen after these test periods. True digestibility of the different thermoxidized compounds from the heated oil was 30-40%, whereas that of thermoxidized compounds from the fresh oil was much higher (approximately 80%). Nonoxidized triacylglycerol hydrolysis was negatively affected by the presence of large amounts of thermoxidized compounds. The present proposed model seems to be a useful tool for the study of thermoxidized oils. Data also show that thermoxidized compounds from abused olive oil are poorly but actively hydrolyzed and absorbed in vivo.