Introduction: Vitamin K antagonist (VKA) therapy is monitored using the international normalized ratio (INR) calculated from the prothrombin time (PT). It has an individual dosage and a narrow therapeutic index. As a consequence, patients spend more than one third of their time outside the therapeutic range. Bleeding is one of the most important complications of VKA therapy. Although the risk of bleeding increases sharply when the INR exceeds 5, in extremely over-anticoagulated patients when the INR cannot be determined (INR>10), the exact bleeding risk in the individual patient is difficult to define. Thrombin generation is a global haemostasis assay that measures the overall capacity of the plasma to form thrombin after initiating coagulation with tissue factor. It is decreased in patients on VKA therapy and shows an inverse, but not linear correlation with INR. Our aim was to measure TG in over-anticoagulated patients and correlate it with their bleeding symptoms.
Materials and methods: Twelve patients receiving vitamin K antagonist treatment with INR>10 were included. Based on complete clinical evaluation of the bleeding symptoms, severity was classified as life threatening, major, minor, non-relevant and no bleeding. Blood was collected on the day of overdose and one day after. Thrombin generation was measured by the Calibrated Automated Thrombogram method. In order to achieve thrombin formation in the highly anticoagulated plasma, we applied tissue factor in high concentration.
Results: Out of the twelve patients, two had major and three had minor bleeding symptoms. In five cases, the bleeding was non-relevant and no bleeding symptoms were detected in two cases. In the two patients with major bleeding, TG was considerably lower than in the other patients. We found no difference in TG parameters between the minor, non-relevant and no bleeding groups.
Discussion: In patients with severe vitamin K antagonist overdose, the actual risk of bleeding is unknown. Thrombin generation can be measured in the highly anticoagulated plasma (INR>10) using a modified protocol. This way, it may provide an estimation of the individual bleeding tendency.