MYLES et al. performed a meta-analysis of all randomized trials of adult cardiac surgery in which patients underwent coronary artery bypass graft surgery with cardiopulmonary bypass and/or vascular surgery with cardiopulmonary bypass to examine the question of whether anesthetic technique affected the outcome of cardiac surgery. Ten randomized trials met the inclusion and exclusion criteria for adequacy of study design with a total of 1,800 patients. It is a bit surprising that only 10 randomized trials of anesthetic technique could be found in the world literature for a surgical procedure that is performed more than 500,000 times a year. The even more surprising finding is that 1,012 of the 1,800 patients came from a single study. More than 56% of the total patients studied in randomized trials in the world literature for anesthetic technique in cardiac surgery came from a single study by Slogoff and Keats published in 1989. 1Does this affect the results of Myles et al. ?
Slogoff and Keats 1performed a prospective, randomized, blinded-analysis, clinical trial of four different anesthetics—sufentanil (15–30 μg/kg) or fentanyl (10 μg/kg) in combination with isoflurane, halothane, or enflurane—and found no difference in primary outcome myocardial infarction or death. For a study that was underpowered to demonstrate a difference in hard outcomes, it is not surprising that none was found. The authors calculated that 7,844 patients would need to be studied to have an 80% power to show a difference in myocardial infarctions and 3,287 for death. It is not surprising, given the calculated sample sizes, that no further large-scale trials of anesthetic agents were designed with myocardial infarction or death as an endpoint.
Slogoff and Keats also used surrogate endpoints. They found no difference in their surrogate outcome variables of myocardial ischemia or creatine phosphokinase and isoenzyme release of creatine kinase containing M and B subunits between the anesthetic techniques. They found that tachycardia led to ischemia, ischemia led to ischemia, and β-blockers reduced tachycardia. They also found that sufentanil 15–30 μg/kg caused longer periods of tracheal intubation (22.8 ± 12.3 vs . 15.3 ± 6.3, P = 0.001) than fentanyl (10 μg/kg) in combinations with an inhaled agent. They explained that the “… duration[s] of postoperative intubation were as expected from the known pharmacologic effects of the primary anesthetic agents.” The definitive work on the use of fast-track versus slow-track cardiac anesthesia was published in 1989, with the patients studied between 1985 and 1987.
This Editorial View accompanies the following article: Myles PS, Daly DJ, Djaiani G, Lee A, Cheng DCH: A systematic review of the safety and effectiveness of fast-track cardiac anesthesia. Anesthesiology 2003; 99:000–00.
What additional information can we glean from the subsequent 15 yr of experience with fast-track anesthesia? Myles et al. conclude that fast-track anesthesia is safe. No significant outcome differences were found in 30-day all-cause mortality, myocardial infarction, sepsis, wound infection, stroke, acute renal failure, prolonged intensive care unit stay, or surgical reexploration for bleeding. The duration of tracheal intubation and intensive care unit length of stay were shorter for fast-track cardiac anesthesia. As expected, the duration of postoperative intubation was shorter, given the known pharmacologic effects of the primary anesthetic agents.
Unless someone decides to study between 3,000 and 8,000 patients, the world literature will be left with the conclusions of Slogoff and Keats that there is no definitive difference between high-dose opioid anesthesia and low-dose opioid anesthesia with an inhaled agent. That is not to say there is no difference—there is just little chance of showing a difference without studying very large numbers of patients. It is important to remember that the failure to demonstrate a difference is not proof of similarity. The two techniques could still be different, but the difference must be small. On the other hand, demonstrating a statistically significant difference indicates a high probability of a true difference. When deciding one of the fundamental questions of anesthetic technique, we are left with the conclusions of Slogoff and Keats. Controlling the heart rate prevents ischemia, preventing ischemia prevents myocardial infarctions and deaths, and the best way to prevent myocardial ischemia is with a β-blocker, not with the choice or dose of opioids. There is no benefit from high-dose opioid anesthesia for cardiac surgery as compared to low-dose opioid anesthesia with an inhaled agent. The low-dose opioid anesthesia with an inhaled agent allows for fast-track cardiac surgery, which may reduce costs. The surrogate endpoints of a 15-yr-old study are as important today as they were in the past. Myles et al. have provided an important systematic review of the world literature on the safety and effectiveness of fast-track versus slow-track cardiac anesthesia. It is time to get on the fast track.