Induction drug and outcome of patients admitted to the intensive care unit after emergency laparotomy.

BACKGROUND AND

OBJECTIVE

Etomidate is often used to induce anaesthesia in sick patients owing to its relative cardiovascular stability. However, etomidate affects adrenal cortical activity, and there is concern that this could impair outcome in patients undergoing emergency surgery.

METHODS

We retrospectively analysed data from 176 patients admitted to an ICU after emergency laparotomy. We retrieved ASA status, surgical diagnosis, induction drug use, blood pressure before and after induction and any vasopressor administration, steroid and vasopressor therapy in ICU and patient outcome. Choice of induction drug was at the discretion of the attending anaesthetist.

RESULTS

The drugs (numbers of patients) used to induce anaesthesia were etomidate (52), thiopental (90), propofol (16), midazolam (12) and ketamine (4). Fifty-two patients (30%) died in hospital. ASA status was the only independent predictor of hospital outcome (P < 0.001). Choice of induction drug was related to ASA status. As ASA status worsened, the likelihood of using etomidate or midazolam/ketamine increased (P = 0.001). We found no association between etomidate and dying in hospital, though our study might not have had sufficient power to show a difference between induction drugs. The relative risks [95% confidence interval (CI)] of dying in hospital were etomidate 1.16 (0.72-1.87), thiopental 0.82 (0.52-1.30), propofol 0.40 (0.11-1.49) and midazolam/ketamine 1.84 (1.09-3.12). Vasopressor and steroid therapy in the ICU was not related to induction drug. The risk of developing hypotension at induction or of receiving vasopressor to treat hypotension was least with etomidate.

CONCLUSION

We found no evidence that etomidate is associated with worse outcome than thiopental or propofol in patients undergoing emergency laparotomy, but we cannot be certain that etomidate is well tolerated in this group of patients. More data are required to address this issue definitively.