Introduction: External electrical cardioversion under hypnotics, even when combined with opioids, has been consistently described as distressing or painful. The main objective of the present study was to determine if adding an opioid to a hypnotic, in comparison to the same hypnotic alone, would decrease the incidence of unpleasant or painful recall during anaesthesia for external electrical cardioversion.
Methods: This was a single-centre, prospective, randomised, double-blinded clinical trial that took place from September 2011 to March 2012. Fifty-two adult patients with persistent atrial fibrillation, scheduled for external direct current cardioversion, were enrolled. Exclusion criteria were age >80 years, previous cardiac surgery, implanted pacemaker or defibrillator, and haemodynamic instability. Patients received intravenously either (group EF) fentanyl 50 g and after 60 s etomidate 0.1 mg/kg, or (group E) only etomidate 0.1 mg/kg. If the patients did not lose their eyelid reflex, repeated doses of etomidate 4 mg were given. Cardioversion was attempted with an extracardiac biphasic electrical shock from 200 to 300 J, at most three times. The primary endpoint was recall of something unpleasant or painful. Secondary outcome measures were predictors of the requirement for repeat doses of etomidate, and the number of shocks needed.
Results: Fifty-one patients (35 male, 16 female), aged 62.1 ± 10.2 years, completed the study. There were no differences between group EF and group E regarding recall (unpleasant recall 0 vs. 2 patients, p=0.235; painful recall 1 vs. 0 patients, p=0.510). The administration of etomidate alone was a significant predictor for subsequent repeated doses of etomidate (p=0.049, odds ratio 4.312, 95% confidence interval 1.007-18.460). The number of shocks needed to restore sinus rhythm did not differ between the groups (p=0.846).
Conclusions: In the present study, the addition of fentanyl to etomidate did not diminish distressing or painful experience during anaesthesia for external cardioversion.