Use of Aprotinin versus Tranexamic Acid in Cardiac Surgery Patients with High-Risk for Excessive Bleeding (APACHE) trial: a multicentre retrospective comparative non-randomized historical study

Abstract

Abstract OBJECTIVES Following the reintroduction of aprotinin into the European market, the French Society of Cardiovascular and Thoracic Anaesthesiologists recommended its prophylactic use at half-dose for high-risk cardiac surgery patients. We examined whether the use of aprotinin instead of tranexamic acid could significantly reduce severe perioperative bleeding. METHODS This multicentre, retrospective, historical study included cardiac surgery patients treated with aprotinin or tranexamic acid between December 2017 and September 2020. The primary efficacy end point was the severe or massive perioperative bleeding (class 3–4 of the universal definition of perioperative bleeding). The safety secondary end points included the occurrence of thromboembolic events and all-cause mortality within 30 days after surgery. RESULTS Among the 693 patients included in the study, 347 received aprotinin and 346 took tranexamic acid. The percentage of patients with severe or massive bleeding was similar in the 2 groups (42.1% vs 43.6%, Adjusted odds ratio [ORadj] = 0.87, 95% confidence interval: 0.62–1.23, P = 0.44), as was the perioperative need for blood products (81.0% vs 83.2%, ORadj = 0.75, 95% confidence interval: 0.48–1.17, P = 0.20). However, the median (Interquartile range) 12 h postoperative blood loss was significantly lower in the aprotinin group (383 ml [241–625] vs 450 ml [290–730], P < 0.01). Compared to tranexamic acid, the intraoperative use of aprotinin was associated with increased risk for thromboembolic events (adjusted Hazard ratio 2.30 [95% Cl: 1.06–5.30]; P = 0.04). CONCLUSIONS Given the modest reduction in blood loss at the expense of a significant increase in thromboembolic adverse events, aprotinin use in high-risk cardiac surgery patients should be based on a carefully considered benefit–risk assessment.