Washington—Noncardiac patients given etomidate may be up to three times more likely than those given propofol to die within 30 days, a new study suggests.

“Although etomidate is sometimes used for general anesthesia induction in critically ill patients, the drug is known to cause prolonged adrenal impairment by blocking cortisol release,” said Ryu Komatsu, MD, a resident at the Cleveland Clinic in Ohio. “However, the potential link between etomidate and worsened postoperative outcomes has not been systematically studied in general surgical patients.”

Dr. Komatsu and his colleagues evaluated the electronic records of 31,148 adult patients who underwent noncardiac surgery under general anesthesia requiring at least one night of hospitalization at the institution between 2005 and 2009. Anesthesia was induced with etomidate in 2,616 patients; 28,532 received propofol.

The researchers performed propensity score matching for 2,143 patients who received etomidate and 5,231 given propofol using a host of potential demographic and intraoperative confounders. The matched groups were compared with respect to duration of postoperative hospitalization and 30-day mortality.

Dr. Komatsu’s group found that 139 patients given etomidate died within 30 days of surgery, versus 134 in the propofol group (odds ratio, 2.3; 97.5% confidence interval, 1.73-3.06; P<0.001). The mean length of stay was seven days for patients given etomidate (range, three to 13) and six days for those who received propofol (range, two to 11). Patients who received etomidate were 18% less likely to be discharged from the hospital at any time after surgery than were those treated with propofol.

“We also looked at the systolic blood pressure profile during surgery,” added Dr. Komatsu, who reported the findings at the 2012 annual meeting of the American Society of Anesthesiologists (abstract 015). “We found that from intubation to incision, propofol patients had lower systolic blood pressure, with mean difference of 8 to 9 mm Hg, which I think could be clinically significant.”

Dr. Komatsu acknowledged the limitations of the retrospective study. “We could only show an association between worse outcomes and etomidate use,” he said. “Randomized controlled trials are necessary to determine if there is another relationship between etomidate use and outcomes, as well as to define precisely the effect of etomidate. In the meantime, however, [clinicians] should use etomidate judiciously, considering that etomidate’s improved hemodynamic stability at induction might be accompanied by substantially worse longer-term outcomes.”

Jesse Ehrenfeld, MD, an assistant professor of anesthesiology at Vanderbilt University School of Medicine, in Nashville, Tenn., said the authors did the best analysis they could with the available data, However, “while their analysis adjusted for a number of important variables [that] impact mortality and length of stay, it is still unclear what the potential impact of unmeasured confounding might be,” he said. “In my experience, most clinicians choose to use etomidate in sicker patients under particular clinical scenarios. And without a prospective randomized trial, those unaccounted-for circumstances cannot be controlled for. It, therefore, would not be appropriate to attribute the excess mortality seen in the study to use of etomidate alone. I congratulate the authors for investigating this important subject, and hope that their efforts and this study will lead to future, more conclusive, work.”