An ex vivo comparison of partial thromboplastin time and activated clotting time for heparin anticoagulation in an ovine model.

ABSTRACT

BACKGROUND: Sheep are a primary model of mechanical circulatory support (MCS) with heparin anticoagulation therapy frequently being monitored by activated clotting time (ACT) due to ease and cost. In patients undergoing long-term heparin therapy, other anticoagulation monitoring strategies, such as activated partial thromboplastin time (aPTT), have proven to be more reliable indicators for the adequacy of anticoagulation, frequently determined by heparin concentration. As there is a paucity of similar studies in sheep, we sought to investigate the correlation between heparin concentration and ACT and aPTT using whole sheep blood in an ex vivo model. METHODS: Fresh whole blood was serially drawn from an adult female Dorset-hybrid sheep and aliquots were placed into tubes containing heparin saline solutions with concentrations ranging from 0 to 7.81 U heparin per mL of whole blood. ACT and aPTT values were measured on each of the samples. The experiment was performed four times with the same animal. A simple linear regression was performed to determine correlation, and subgroup analysis was performed on low versus high heparin concentrations typically seen in human patients on long-term MCS, such as extracorporeal membrane oxygenation (ECMO), versus cardiopulmonary bypass, respectively. RESULTS: aPTT measurements versus the heparin concentration had an R2  = 0.7295. ACT measurements versus the heparin concentration had a R2  = 0.4628. aPTT measurements versus the ACT measurements had a R2  = 0.2974. The strength of the correlation between aPTT and heparin concentration increased at low heparin concentrations (R2  = 0.8392). CONCLUSION: aPTT had a more reliable correlation to heparin concentration and thus anticoagulation level than ACT. This was particularly true at lower heparin concentrations, similar to ranges seen for patients on ECMO. The correlation between aPTT and ACT values was poor. Further in vivo studies should be performed to confirm our results.