Effect of changes in the thrombin-antithrombin (TAT) half-life used in the model on the predicted thrombin generation rate in subject 1. (A) The measured level of TAT (•) versus the predicted level of TAT in the circulatory model using 3 different TAT half-lives: 4, 8, and 16 minutes. The lowest error between measured and predicted TAT levels occurred with a half-life of 8 minutes. (B) The effect of different TAT half-lives in the circulatory model on the predicted thrombin generation rates. The gray curve corresponds to the best fit between measured and predicted TAT levels and represents the best estimate of the in vivo thrombin generation rate. The solid black box at the top of the figure indicates the period of cardiopulmonary bypass (CPB).