Development of a two-hit lethal liver injury model in swine.

Abstract

Noncompressible truncal hemorrhage remains a leading cause of preventable death in the prehospital setting. Standardized and reproducible large animal models are essential to test new therapeutic strategies. However, existing injury models vary significantly in consistency and clinical accuracy. This study aims to develop a lethal porcine model to test hemostatic agents targeting noncompressible abdominal hemorrhages. We developed a two-hit injury model in Yorkshire swine, consisting of a grade IV liver injury combined with hemodilution. The hemodilution was induced by controlled exsanguination of 30% of the total blood volume and a 3:1 resuscitation with crystalloids. Subsequently, a grade IV liver injury was performed by sharp transection of both median lobes of the liver, resulting in major bleeding and severe hypotension. The abdominal incision was closed within 60s from the injury. The endpoints included mortality, survival time, serum lab values, and blood loss within the abdomen. This model was lethal in all animals (5/5), with a mean survival time of 24.4 ± 3.8min. The standardized liver resection was uniform at 14.4 ± 2.1% of the total liver weight. Following the injury, the MAP dropped by 27 ± 8mmHg within the first 10min. The use of a mixed injury model (i.e., open injury, closed hemorrhage) was instrumental in creating a standardized injury while allowing for a clinically significant hemorrhage. This novel highly lethal, consistent, and clinically relevant translational model can be used to test and develop life-saving interventions for massive noncompressible abdominal hemorrhage.