Computational simulation and comparison of prothrombin complex concentrate dosing schemes for warfarin reversal in cardiac surgery

Abstract

Prothrombin complex concentrate (PCC) is increasingly used for acute warfarin reversal. We hypothesized that computational modeling of thrombin generation (TG) could be used to optimize the timing and dose of PCC during hemodilution induced by cardiopulmonary bypass (CPB). Thrombin generation patterns were modeled in anticoagulated patients (n=59) using a published computational model. Four dosing schemes were evaluated including single full dose (median, 41.2IU/kg) of PCC before or after CPB, ½-dose before and after CPB, or 1/3-dose before CPB plus 2/3-dose after CPB. Hemodilution was modeled as 40 or 60% dilution of factors from baseline. The lag time (s) of TG, and peak thrombin level (nM) were evaluated. Prolonged lag time, and reduced peak TG were due to low vitamin K-dependent (VKD) factors, and pre-CPB PCC dose-dependently restored TG to near-normal or normal range. After 40% dilution, TG parameters were similar among 4 regimens at the end of therapy. The recovery of VKD factors was less when PCC was given before CPBafter 60 % dilution, but TG parameters were considered hemostatically effective (>200nM) with any regimen. Withholding the full dose of PCC until post-CPB resulted in severely depressed TG peak (median, 47nM) after 60 % dilution, and some supra-normal TG peaks after treatment. Pre-CPB administration of full or divided doses of PCC prevents extremely low TG peak during surgery, and maintains hemostatic TG peaks in both 40 and 60% hemodilution models. Although PCC's hemostatic activity appears to be highest using the full dose after CPB, hypercoagulability may develop in some cases.