Sedation for cardioversion in the emergency department: analysis of effectiveness in four protocols

We compare effectiveness, adverse effects, and recovery times of propofol, etomidate, and midazolam (with and without flumazenil) for cardioversion in the emergency department (ED).

Thirty-two hemodynamically stable adult patients undergoing cardioversion in the ED were randomly assigned to receive etomidate (n=9), propofol (n=9), midazolam (n=8), or midazolam followed by flumazenil (n=6). For all patients, we measured induction time, awakening time, total recuperation time, global time, and adverse effects. Arterial pressure, cardiac and respiratory rate, and peripheral oxygen saturation were monitored throughout the procedure. Descriptive and nonparametric tests were used.

Demographic data were similar in all groups. Deep sedation and successful cardioversion were achieved in all cases. Hemodynamic assessment at baseline, after induction, after cardioversion, and at recovery demonstrated no significant difference between the 4 groups. Induction time was short in all groups. Awakening time was longer in the midazolam group (median 21 minutes, range 1 to 42 minutes) compared with that of the other groups (etomidate group: median 9.5 minutes, range 5 to 11 minutes; propofol group: median 8 minutes, range 3 to 15 minutes; midazolam/flumazenil group: median 3 minutes, range 2 to 5 minutes), and the same occurred with total recuperation time (etomidate group: median 14 minutes, range 5 to 20 minutes; propofol group: median 10 minutes, range 5 to 15 minutes; midazolam group: median 45 minutes, range 20 to 60 minutes; midazolam/flumazenil group: median 5 minutes, range 2 to 90 minutes). All patients in the midazolam/flumazenil group but 1 became resedated after flumazenil was discontinued. Four patients who had received etomidate exhibited myoclonus, which was pronounced and seizure-like in 1 case.

Four sedative regimens (propofol, etomidate, midazolam, and midazolam/flumazenil) were uniformly effective in facilitating ED cardioversion in hemodynamically stable adults. Propofol was well tolerated and lacked the myoclonus, prolonged sedation, and resedation noted with the latter 3 respective groups. Larger studies are needed to generalize these conclusions.