Continuous Etomidate Infusion for the Management of Severe Cushing Syndrome

Validation of a Standard Protocol

Ty B. Carroll; William J. Peppard; David J. Herrmann; Bradley R. Javorsky; Tracy S. Wang; Hina Patel; Katarzyna Zarnecki; James W. Findling


J Endo Soc. 2019;3(1):1-12. 

In This Article


Clinical Features

Seven patients representing nine distinct episodes of standardized intravenous etomidate infusion for the medical treatment of severe hypercortisolemia are described. Patient clinical features are reported in Table 1. The diagnosis of neoplastic hypercortisolism was confirmed by elevations of serum, urine, late-night salivary cortisol, and/or ACTH measurements. Etomidate was administered to reduce preoperative cortisol before surgery in five episodes. In three of these episodes, the patient ultimately underwent surgery, and in the other two episodes, both in the same patient, therapy was discontinued in pursuit of medical management as an alternative to surgery. Etomidate was eventually discontinued in this patient as hospice care was ultimately pursued. In three episodes, patients had previously received long-term oral therapy but acutely became unable to tolerate oral therapy; therefore, etomidate was used as temporary bridge therapy. Another patient received etomidate while awaiting a special order of metyrapone to arrive to pursue long-term medical management. Baseline cortisol was elevated in all patients (Table 1).


All episodes were dosed per protocol, including initial dose and subsequent dose adjustments. Treatment began with an optional 5-mg intravenous bolus of etomidate. The bolus was at the discretion of the treating endocrinologist and based on desired rapidity of cortisol reduction. Starting dose was consistently 0.02 mg/kg/h, with the exception of one patient who had previously tolerated a higher infusion rate, in whom etomidate was reinitiated at 0.04 mg/kg/h. Serum cortisol levels were then measured every 6 hours. The infusion rate was maintained if the cortisol level was trending toward goal. If the cortisol level was not trending to goal, then the infusion rate was adjusted in increments of 0.01 to 0.02 mg/kg/h to achieve desired cortisol levels. The infusion rate was not up titrated more frequently than every 6 hours. A maximum allowed infusion rate was set at 0.3 mg/kg/h. This maximum infusion rate was not reached in any of the nine episodes. The maximum achieved etomidate infusion rate was identical to the infusion rate when goal cortisol levels were achieved in all episodes with a median of 0.081 mg/kg/h (range, 0.033 to 0.150 mg/kg/h). The infusion was continued for a median duration of 63.6 hours (range, 36.3 to 168.8 hours).


In eight of the nine episodes, the target cortisol concentration was <20 μg/dL (one episode targeted <40 μg/dL) and was achieved in all episodes except for one where the patient entered hospice care before achieving goal. Detailed results of cortisol dynamics on therapy are shown in Table 1. In the episodes where etomidate therapy reached completion, median baseline cortisol was 105 μg/dL (range, 32 to 245 μg/dL). It took a median of 38 hours (range, 26 to 134 hours) to achieve a median nadir serum cortisol of 15.8 μg/dL (range, 6.9 to 27 μg/dL). The median cortisol reduction rate was 1.09 μg/dL/h (range, 0.78 to 4.97 μg/dL/h), resulting in a median reduction in cortisol of 86.8 μg/dL (range, 19.8 to 179.6 μg/dL), or a median 80% reduction (range, 67% to 95%) (Figure 2). The one patient who did not complete etomidate therapy was on a trajectory to do so before dying, representing one of the fastest rate reductions observed in this case series.

Figure 2.

Plot of serum cortisol vs rate of etomidate infusion in seven of the nine episodes with available data.


All patients tolerated their etomidate infusion well. In all but one episode, mentation was measured via the RASS. At baseline, seven patients had a RASS score of 0, which remained largely unchanged throughout the infusion. One episode who was admitted with hypercortisolemia-induced psychosis had a RASS score of −1 during the infusion and improved to 0 upon termination of etomidate therapy. The one episode without documented RASS score revealed no mentation changes per physician and nursing progress notes. At no time throughout the etomidate infusion did any RASS score go < −1 or > 0, nor did a change in RASS score necessitate discontinuation of etomidate.

Patients did not experience any electrolyte abnormalities or substantial changes in renal function during the etomidate infusion, nor did they experience metabolic acidosis. Glucose control improved slightly overall from the beginning of therapy compared with discontinuation of therapy.

All episodes except one realized a gradual reduction in blood pressure, whereas the heart rate remained almost unchanged for all patients and did not correlate with changes in blood pressure. No patient experienced hypotension while receiving the infusion protocol.

Two episodes experienced nausea and vomiting toward the end of etomidate infusion (35 hours into a 38-hour infusion and 33 hours into a 36-hour infusion, respectively) and may be attributed to rapid reduction of cortisol levels, given a relative reduction of cortisol of 68% and 72%, respectively (cortisol of 10.3 μg/dL down from baseline level of 32.5 μg/dL and 53.6 μg/dL down from baseline level of 192.2 μg/dL). No patients required a dose adjustment as a result of adverse effects.

Cumulative median propylene glycol exposure was 90.3 g (range, 32 to 187 g) or 818 mg/kg (range, 284 to 1762 mg/kg) or 232 mg/kg/d (range, 112 to 328 mg/kg/d). No propylene glycol–related adverse effects, such as acidosis, intravascular hemolysis, altered mentation, seizure, hypoglycemia, or renal failure, were noted, although they were prospectively monitored for during the entire etomidate infusion.