The aim of this study was to develop a standardized effective thrombogenic arterial anastomosis model, as usually encountered in clinical practice, and to offer a detailed evaluation of the antithrombotic effect of thrombin's direct inhibitors, antithrombin III and hirudin, as locally applied. Wistar rats were divided into four groups of 12 animals each. The carotid artery sustained a standardized crush-avulsion-type injury (groups B-D). A segment of the afflicted area was removed and replaced by a microvenous graft. Group A had no crush-avulsion injury inflicted; a microvenous graft replaced a simple resection from the center of the carotid artery. During microvascular anastomoses, normal saline (groups A and B), recombinant hirudin (group C), or antithrombin III (group D) were locally applied. Bleeding times were recorded, and patency tests were performed 20 min, 48 h, and 1 week after blood flow reestablishment. All grafts were harvested and examined histologically. Patency tests, 1 week postrevascularization, demonstrated that this experimental crush-avulsion injury model ensured low patency in group B (25%), whereas group A, which had no injury inflicted, achieved a 100% patency rate. The local application of hirudin and antithrombin III significantly increased bleeding times as well as the patency rate (92% and 75%, respectively) compared to group B. These findings indicate the efficiency of the experimental model and the potential use of thrombin's direct inhibitors in microvascular surgery.