Background: In cardiac surgery, protamine is used to reverse the effects of heparin after separation from cardiopulmonary bypass (CPB). Excess of protamine has been demonstrated to cause platelet dysfunction and coagulopathy. A protamin-to-heparin ratio of less than 1:1 is endorsed by the European guidelines. Pharmacokinetic models of heparin decay have been proposed to allow for individualised dosing rather than fixed ratios. The objective of this study is to compare three such models in a large cohort of simulated patients. Methods: The doses of protamine were calculated using the models proposed by Meesters et al., Miles et al., and in the PRODOSE trial. We employed data from the VitalDB database to calculate the doses of heparin and randomly generated time intervals in-between. We tested two scenarios: with an initial dose and heparin added to the priming solution, and where additional heparin was administered over the course of the CPB. Results: We simulated 1166 cases with a mean interval between heparin and protamine administration of 90 ± 22 minutes in the first and 140 ± 28 minutes in the second scenario. The PRODOSE formula produced the lowest protamine-to-heparin ratios, followed by Meesters' formula in the first scenario (0.68:1 vs 0.72:1, P < .001) and the Miles' formula in the second scenario (0.55:1 vs 0.62:1, P < .001). Conclusion: The doses calculated with pharmacokinetic models exhibited discrepancies of up to 13.6%. While confirmation of complete reversal with alternative methods is necessary, the models facilitate a more individualised dose selection than the fixed ratios proposed in the literature.