Click to enlarge
On the first day of my second year of residency, I showed up to the cardiac surgery intensive care unit (CSICU) terrified for what awaited me. It wasn’t uncommon for patients recovering from cardiac surgery to be on three different vasopressors while intubated with four chest tubes and pacing wires. Even after a year of general surgery residency, the CSICU was a bit of a foreign land to me. One thing I learned quickly, though, was that bleeding after cardiac surgery was relatively common.
Many patients required blood transfusions and occasionally had to return to the operating room due to ongoing bleeding. In an effort to reduce postoperative bleeding, transfusion, and reoperation, most cardiac surgery units administer antifibrinolytic agents such as tranexamic acid, a lysine analogue. Although studies have shown that tranexamic acid can reduce the risk of blood loss and transfusion after cardiac surgery, it also has a prothrombotic effect, and the risk of myocardial infarction, stroke, or other complications is unclear. Further, some studies have shown an increased risk of seizures with tranexamic acid, potentially due to cerebral infarction.
A recent multicenter, double-blind, randomized-controlled study by Myles et al. published in NEJM investigates the risks and benefits of intra-operative tranexamic acid infusion in cardiac surgery. According to the trial’s 2-by-2 factorial design, patients at increased surgical risk undergoing coronary-artery surgery (either on- or off-pump and with or without concomitant valve procedures) were randomized to receive aspirin or placebo (results reported in a separate study) and tranexamic acid or placebo. The primary outcome was a composite of death and thrombotic complications (nonfatal myocardial infarction, stroke, pulmonary embolism, renal failure, or bowel infarction) within 30 days after surgery. Secondary outcomes included all components of the composite primary outcome as well as reoperation due to major hemorrhage or cardiac tamponade, transfusion, and seizures (which was added later in the study and was based on clinical diagnosis).
Among the 2311 patients randomized to receive tranexamic acid and the 2320 patients who received placebo, baseline characteristics and comorbidities were evenly distributed between groups. The composite primary outcome was reported in 17% of patients in the tranexamic acid group and 18% of patients in the placebo group (relative risk, 0.92; 95% confidence interval [CI], 0.81 to 1.05; P = 0.22). This result was not affected by whether or not patients received aspirin. Rates of the individual components of the composite outcome were similar between groups.
Tranexamic acid was associated with lower rates of bleeding, blood loss, number of patients transfused, number of blood products transfused (4331 vs. 7994, P<0.001), and major hemorrhage or cardiac tamponade requiring reoperation (1.4% vs. 2.8%, P=0.001). However, postoperative seizures were more common in the tranexamic acid group (15 vs. 2 patients; relative risk, 7.62; 95% CI, 1.77 to 68.71; P = 0.002 by Fisher’s exact test). In a post-hoc analysis, seizure was associated with poorer outcomes such as stroke (relative risk, 21.88; 95% CI, 10.06 to 47.58; P<0.001) and death (relative risk, 9.52; 95% CI, 2.53 to 35.90; P = 0.02). Several years into the study, the dose of tranexamic acid administered was halved after some studies showed a possible dose-related risk of seizure. The dose reduction did not affect the risk of seizure (although the study was underpowered for this analysis).
Overall, tranexamic acid seems to reduce bleeding complications after cardiac surgery, most impressively by reducing the number of transfusions required by 47%. In addition, tranexamic acid did not increase the risk of death, myocardial infarction, or stroke. However, this study corroborates data linking tranexamic acid to higher risk of seizure.
How will this study change current practice in cardiac surgery? Will cardiac surgery units that do not currently use tranexamic acid change their practice? NEJM Deputy Editor John Jarcho comments, “Surgeons really don’t like bleeding, for good reasons. I think they will see these findings as a good reason to use antifibrinolytic therapy, if they weren’t using it previously.”
Browse more From Pages to Practice »
Andrea Merrill, MD
Andrea was a 2015-2016 NEJM Group Editorial Fellow. She is currently in the middle of her General Surgery residency at Massachusetts General Hospital and is also conducting research focusing on improvements in breast cancer surgery. She plans to pursue a fellowship in Surgical Oncology at the completion of her residency.