Coronary artery bypass graft (CABG) is among the most frequently performed major surgical procedures worldwide. It is still considered a high-risk procedure, and even in high-income countries, remains associated with a relatively high incidence of complications, with a one-year mortality of about 3%. Modern anesthesia can be performed by administration of intravenous agents, inhalational halogenated agents (volatile anesthetics), or a combination of both intravenous and inhalational agents. Evidence from animal studies and initial studies in human suggest that volatile agents may have some pharmacological properties that help to protect the heart during the critical phase of the procedure when patient heart is stopped and cardiac and respiratory function are taken over by the heart-lung machine. These studies suggested that administration of volatile anesthetics during coronary artery bypass graft may reduce mortality from 3% to about 2%. However, a clear and definitive evidence on this beneficial effect is lacking. If this was confirmed, it would translate in saving several thousands of lives every year. Therefore, we planned and conducted a clinical trial comparing an anesthetic regimen including volatile agents with total intravenous anesthesia in patients undergoing CABG.

The study was performed in 36 hospitals in 13 countries. Patients undergoing elective CABG were randomly assigned to receive either volatile agents or total intravenous anesthesia (TIVA) during surgery. Patients undergoing emergency surgery, combined cardiac surgical procedures (e.g. CABG + valve surgery) and receiving medications that could interfere with cardioprotection induced by volatile agents were excluded. The study was single-blinded, meaning that patients were unaware of trial group assignment. Outcome assessors and statistician performing analysis for the primary outcome were also blinded to study group assignment. The primary outcome of the study was mortality one-year after the procedure. Additional outcomes were 30-days mortality, 1-year and 30-days cardiac mortality, occurrence rate of postoperative myocardial infarction, kidney injury and failure, stroke, bleeding and length of stay in the intensive care unit (ICU) and in the hospital. Safety outcomes included anesthesia-related adverse events. With the exception of study treatment, all of the perioperative management strategies were left at discretion of the attending clinicians. Anesthesiologists were instructed to performed strategies to enhance the cardioprotective effects of volatile agents: these include maintaining a specific dose for at least 30 minutes, and perform wash-in/wash-out periods. The study originally planned to enroll 10 600 patients, but was interrupted early as interim analyses (performed after enrolment of 50% of planned sample size) demonstrated that the probability to observe a significant difference in outcome between the two groups was close to 0, even after completing enrolment. From 2014 to 2017, a total of 5400 patients were enrolled in the study, with 2709 patients assigned to the volatile group and 2691 assigned to the TIVA group. More than 99% of enrolled patients were followed up to one-year after randomization. Only 2.3% of enrolled patients crossed-over from one treatment arm to the other. In the volatile group, more than 97% of patients received at least one of the recommended strategies to enhance the cardioprotective effects. After one year from the operation, we observed no significant difference in the occurrence of death from any cause between the two groups: mortality was 2.8% in the volatile anesthetics group and 3.0% in the TIVA group; relative risk = 0.94; 95% confidence interval [CI] = 0.69 to 1.29; P = 0.71 (with data available for 5353 patients [99.1%]). Similarly, mortality rate were similar also at 30 days (1.4% and 1.3%, respectively; relative risk = 1.11; 95% CI = 0.70 to 1.76, with data available for 5398 patients [99.9%]). One-year and 30-days mortality due to cardiac causes were comparable between groups. There were no significant differences between the two groups in terms of incidence of perioperative myocardial infarction, kidney injury, stroke, bleeding or length of ICU and hospital stay. The overall incidence of anesthesia-related adverse events was negligible, with only 9 cases (0.2%) of allergic reactions reported. We did not identify any subgroup of patients in which one of the treatments was superior to the other. Our study is the largest study ever performed on this topic and showed that use of volatile anesthetics during CABG did not reduce one-year mortality or other major clinical outcomes. In addition, it demonstrated that modern anesthesia techniques are both equally safe in this patient population. The multicenter design and large sample size ensure that our results have a high external validity. Even if the study was interrupted early, the analyses we performed suggested that it is highly unlikely that enrolling the planned sample size would have yielded different results. Our study has some limitations. We did not specify a specific anesthetic management protocol, because we aimed at replicating real-life clinical practice. We allowed for co-administration of propofol (the most commonly used iv anesthetic) in the volatile group. Some studies suggest that propofol administration may jeopardize the cardioprotective effect of volatile agents; however, previous studies also showed that the beneficial effect of volatile anesthetics is maintained also when propofol is co-administered. In addition, we enrolled relatively low-risk patients undergoing elective, isolated CABG. Although these comprises the vast majority of CABG patients, we cannot exclude a beneficial effect of volatile agents in patients with a higher risk of perioperative complications or undergoing different type of cardiac surgery. Nevertheless, previous studies showed no evidence of benefit from volatile agents in patients undergoing high-risk combined cardiac surgery. In conclusion, our study showed that among patients undergoing elective, isolated CABG, an intraoperative anesthetic regimen including volatile anesthetics did not result in significantly lower 30 days or 1 year mortality than a regimen of TIVA. Furthermore, both techniques can be considered safe.