Outcome of Major Hemorrhage at a Major Cardiothoracic Center in Patients with Activated Major Hemorrhage Protocol versus Nonactivated Protocol

Abstract

This study aimed to determine the impact of major hemorrhage (MH) protocol (MHP) activation on blood administration and patient outcome at a UK major cardiothoracic center. MH was defined in patients (> 16 years) as those who received > 5 units of red blood cells (RBCs) in < 4 hours, or > 10 units in 24 hours. Data were collected retrospectively from patient electronic records and hospital transfusion databases recording issue of blood products from January 2016 to December 2018. Of 134 patients with MH, 24 had activated MHP and 110 did not have activated MHP. Groups were similar for age, sex, baseline hemoglobin, platelet count, coagulation screen, and renal function with no difference in the baseline clinical characteristics. The total number of red cell units (median and [IQR]) transfused was no different in the patients with activated (7.5 [5-11.75]) versus nonactivated (9 [6-12]) MHP (p = 0.35). Patients in the nonactivated MHP group received significantly higher number of platelet units (median: 3 vs. 2, p = 0.014), plasma (median: 4.5 vs. 1.5, p = 0.0007), and cryoprecipitate (median: 2 vs. 1, p = 0.008). However, activation of MHP was associated with higher mortality at 24 hours compared with patients with nonactivation of MHP (33.3 vs. 10.9%, p = 0.005) and 30 days (58.3 vs. 30.9%, p = 0.01). The total RBC and platelet (but not fresh frozen plasma [FFP]) units received were higher in deceased patients than in survivors. Increased mortality was associated with a higher RBC:FFP ratio. Only 26% of patients received tranexamic acid and these patients had higher mortality at 30 days but not at 24 hours. Deceased patients at 30 days had higher levels of fibrinogen than those who survived (median: 2.4 vs. 1.8, p = 0.01). Patients with activated MHP had significantly higher mortality at both 24 hours and 30 days despite lack of difference in the baseline characteristics of the patients with activated MHP versus nonactivated MHP groups. The increased mortality associated with a higher RBC:FFP ratio suggests dilutional coagulopathy may contribute to mortality, but higher fibrinogen at baseline was not protective.