Abstract Background Antithrombin is one of the main natural coagulation system inhibitors. It is potentiated by heparin, and may be a key component of heparin response, particularly in infants aged <1 year. We sought to determine the impact of baseline antithrombin activity on response to heparin and thrombin generation during cardiopulmonary bypass (CPB). Methods Secondary analysis was performed using linear regression analyses, which combined patients from a trial of individualized versus weight-based heparin management for 90 infants aged <1 year undergoing cardiac surgery. Results Mean baseline antithrombin activity was 0.69 ± 0.16 U/mL, and it was lower in neonates than in older infants (0.57 ± 0.15 vs 0.77 ± 0.12 U/mL; P  < .001). Lower baseline antithrombin activity was associated with lower postheparin anti-Xa activity (EST SE: +0.47 (0.19) U/mL per 100 U/kg heparin; P  = .01) and higher heparin doses during surgery (EST SE: +51 (17) U/kg per hour; P  = .003). The administration of fresh frozen plasma attenuated the effect of low baseline antithrombin activity (interaction P value = .009). Patients with lower anti-Xa activity recorded during CPB had higher levels of thrombin-antithrombin complex (EST SE: +12.8 (4.7) ng/mL per −1 U/mL anti-Xa; P  = .006); prothrombin activation fragment 1.2 (EST SE: +0.13 (0.07) log pg/mL per −1 U/mL anti-Xa; P  = .06); and D-dimer (EST SE: −0.25 (0.09) log ng/mL per −1 U/mL anti-Xa; P  = .009) in the postoperative period after adjustment for baseline antithrombin activity, duration of CPB, amount of fresh frozen plasma and heparin used throughout surgery in multivariable models. Conclusions Low circulating antithrombin activity is associated with lower heparin efficacy, which ultimately leads to a lower ability to suppress thrombin generation during CPB. Determination of risk factors for heparin resistance, and potentially, antithrombin replacement therapy, may individualize and improve anticoagulation treatment.