Non-mechanical Hemorrhage in Severe Liver Injury

Abstract
Coagulopathy, or non-mechanical hemorrhage, complicated the operative course of 17 of 33 (51.5%) patients suffering severe liver trauma. The highest incidence of non-mechanical hemorrhage (66.7%) occurred in patients undergoing anatomic lobectomy. Serial hemostatic parameters were assessed and thrombocytopenia was the most striking abnormality in patients with non-mechanical hemorrhage. The degree of thrombocytopenia was directly correlated with the number of blood transfusions administered. The mean operative blood transfusion requirement was significantly greater in patients with non-mechanical hemorrhage, 25.1 .+-. 2.87 (standard error of the mean) units, than in those without, 12.2 .+-. 1.83 units (P < 0.001). The bulk of transfusion was given before the onset of clinically overt coagulopathy. Massive transfusion of stored blood was felt to be the most important factor in causing non-mechanical hemorrhage. Convincing evidence for disseminated intravascular coagulation was lacking, and abnormal fibrinolysis was infrequent and mild when observed. Although uneventful in most, in 6 patients non-mechanical hemorrhage resulted in excessive blood transfusion, unnecessary operation or death. Infusions of platelet concentrate, fresh frozen plasma and fresh blood were used to successfull treat most cases of non-mechanical hemorrhage. In all cases, components were not started until non-mechanical hemorrhage was clinically apparent. The value of prophylactic use of blood components is stressed. Because of troublesome side effects associated with the use of prothrombin complex concentrates, these agents are contraindicated in patients with severe liver injury. After receiving concentrates, 1 patient developed severe hypothesion leading to ventricular fibrillation, 2 developed transient thrombocytopenia and 2 others demonstrated multiple pulmonary microthrombi at autopsy, a finding unobserved in autopsied patients not receiving the concentrates.