BART

Blood Conservation using Antifibrinolytics: A Randomized Trial in High-Risk Cardiac Surgery Patients

P.C. Hébert, D.A. Fergusson, R. Arellano, M.A. Blajchman, D. Côté, P. Duke, S. Fremes, J. Karski, R. Martineau, D. Mazer, J. Murkin, M. Rodger, J. Robblee, K. Teoh, G. Wells

Background and significance:  Antifibrinolytic drug therapy has proven to be effective in reducing the number of cardiac surgical patients exposed to blood products and one meta-analysis has suggested that aprotinin saves lives and decreases re-operations for bleeding.  There is no consensus on the best drug, largely because there are limited data comparing agents against one another.  Furthermore, there is also a 10-fold cost differential between aprotinin and epsilon-aminocaproic acid. 

Objectives:  To definitively determine if aprotinin is superior to epsilon-aminocaproic acid and tranexamic acid in terms of:  decreasing massive postoperative bleeding; minimizing exposure to any blood product; decreasing both fatal/life-threatening or serious post-operative complications.

Study Design: Multi-centre (25 Canadian cardiac surgical sites), triple-blind randomized clinical trial comparing three antifibrinolytic agents. 

Study Population: 2,970 high-risk cardiac surgical patients undergoing either re-operation for coronary heart bypass graft (CABG) or aortic valve replacement, or combined valves or valve/CABG procedures.

Interventions:  Patients will be randomized to receive one of the following: 1) Aprotinin (2 million unit bolus + 2 million units in pump prime + 2 million units via infusion over 4 hours) 2) Tranexamic acid (30mg/kg bolus+ 2mg/kg/hr in pump prime+ 16mg/kg/hr infusion +) or 3) Epsilon-aminocaproic acid (10g bolus + 250 cc NaCl in pump prime + 2g/hr infusion).  Transfusion protocols will be suggested. 

Outcomes: The primary outcome will be massive postoperative bleeding, massive transfusion, death due to hemorrhage, or re-operation for hemorrhage and tamponnade.  Secondary outcomes will include the proportion of patients exposed to any blood products over 30 days, fatal and/or life-threatening clinical consequences and serious postoperative complications. Tertiary outcomes will include 6-month survival, health-related quality of life, re-hospitalizations and costs of care. 

Sample Size: Assuming a 3% absolute risk difference (from 6% to 3%) in massive postoperative bleeding between aprotinin and the other two therapies, an alpha error of <0.025, a beta error of 0.20, and a 1% non-compliance rate we require 990 patients per arm or 2,970 patients in total.

Interpretation of potential results: If aprotinin is found to decrease allogeneic exposure and decrease clinical complications, then its adoption will be strongly endorsed.  If aprotinin only decreases exposure to allogeneic blood products, then an economic evaluation will ascertain which option is most cost-effective.  If, on the other hand, the three therapies are shown definitively to be equally effective, then routine use of epsilon-aminocaproic acid would be justified based on cost considerations alone. 

 Timeline: Four years