Statistics from Altmetric.com
Atrial fibrillation (AF) remains a common arrhythmia. Electrical cardioversion is commonly employed in its management. Cardioversion needs to be carried out in a timely manner as the longer the duration of AF, the lower the success rate. Traditionally, an anaesthetist is present and administers a short acting general anaesthetic. It is often difficult to schedule cardioversions at a mutually acceptable time for both the anaesthetist and cardiologist. Recently, cardiologists have become more accustomed to the administration of conscious sedation during electrophysiology studies, and pacemaker and cardioverter-defibrillator implantations.1
We describe the use of intravenous midazolam in the setting of external electrical cardioversion for atrial flutter/fibrillation without the direct supervision of an anaesthetist.
One hundred and forty nine consecutive unselected patients (112 men and 37 women), mean (SD) age 67 (11.8) years, with haemodynamically stable persistent AF were included in this study (December 1998 to June 2000). These included patients from cardiology and general medical/geriatric outpatient departments. The 149 patients underwent a total of 169 cardioversions with 20 patients requiring more than one cardioversion on separate occasions because of recurrence of AF.
The protocol involved obtaining informed consent, ensuring adequate anticoagulation (international normalised ratio (INR) of 2.0–3.0) for at least four weeks before cardioversion. Patients were asked to fast from midnight before the procedure. Cardioversions were performed in an endoscopy suite equipped with a full resuscitation trolley. The procedure was carried out under the direct supervision of the cardiologist (consultant or specialist registrar) with the assistance of a specialist cardiology nurse. Continuous pulse oximetry monitoring was used to measure oxygen saturation and cardiac rhythm was continuously monitored on a cardiac monitor. Patients routinely received low flow (2 l/min) oxygen by nasal cannula before and after the procedure. Midazolam was administered intravenously by the physician, 2.5 mg over 30 seconds, and repeated if necessary in 1 mg increments (maximum 12 mg) until the patient developed slurred speech and was not easily arousable by verbal and physical stimuli (Ramsay sedation score 5).2 Pethidine (25–50 mg) was given intravenously, at the discretion of the physician, to potentiate midazolam sedation.
When adequate sedation was achieved, cardioversion was performed with 200-360-360 J of synchronised energy (100 J for atrial flutter). The defibrillator paddles were positioned over the ventricular apex and in the right infraclavicular area. At each cardioversion attempt, serial shocks using higher energy levels were used if necessary. The procedure was discontinued if a patient failed to revert to sinus rhythm after at least three synchronised shocks, the latter two shocks being 360 J.
Following the cardioversion, the patient was turned on his or her left side and sedation was immediately reversed in all patients with flumazenil, a competitive benzodiazepine receptor antagonist. The dosage schedule for flumazenil was 200 μg over 15 seconds, then 100 μg at 60 second intervals if required, to a maximum total dose of 1 mg. An anaesthetist was always available on site for emergencies.
Once the procedure was completed the patient recovered for two hours with vital signs (blood pressure/respiratory rate) assessed every 15 minutes for the first hour and every 30 minutes for the second hour. Patients were asked to walk for 30 minutes before discharge. All patients were routinely assessed by a specialist cardiac nurse before discharge by use of a questionnaire which asked: (1) Did you find the procedure: intolerable; very unpleasant; mildly unpleasant; not unpleasant? (2) Do you remember anything about the test being done? (3) Would you be prepared to have another cardioversion done: yes; no.
Multiple regression analysis and analysis of variance (ANOVA, three way) was used to compare dose of midazolam, age, and number of synchronised shocks. A probability value of p < 0.05 was considered significant.
The mean (SD) dose of intravenous midazolam was 8.6 (2.1) mg. The requirement of midazolam varied inversely with age (p < 0.001) (table 1).
The mean (SD) level of synchronised energy necessary for cardioversion was 263 (88) J.
The requirement of midazolam varied inversely with the number of synchronised shocks required for cardioversion (table 1). Four of 35 patients (11.4%) who required two shocks needed additional midazolam for the second shock. Eight patients of 28 (28%) who required three shocks needed additional midazolam following the first shock.
Pethidine was administered to 54 (31.9%) patients in addition to midazolam to augment sedation. The requirement of pethidine varied inversely with age (χ2 linear trend p = 0.001).
No procedure was abandoned because of failure to sedate the patient adequately. The mean (SD) dose of flumazenil was 223 (72.1) μg.
Cardioversion with reversion to sinus rhythm before discharge was achieved in 134 procedures (79%).
No patient found the procedure intolerable and only five found it very unpleasant. All patients had total amnesia in regard to the procedure. All patients were prepared to have another cardioversion. One patient developed symptomatic hypotension post-procedure which responded immediately to intravenous fluids. This did not delay discharge. No patient required intubation, or hospital admission.
Our findings show that conscious sedation with midazolam can be safely administered to patients undergoing elective electrical cardioversion by physicians without the direct supervision of an anaesthetist. The patients in our study were unselected and consecutive and a large number of elderly patients (31% > 75 years) were included. This is the largest study in the UK on conscious sedation in patients undergoing electrical cardioversion. Three previous smaller studies (n = 12–33) have shown safety and efficacy of conscious sedation in a similar setting.3-5
Concern regarding the safety of conscious sedation in absence of an anaesthetist is valid. In our study, all the attending physicians and the specialist cardiac nurse were trained in airway management and resuscitation. We consider close monitoring and a short sedation period with its immediate reversal following cardioversion contributed to the low complication rate. Flumazenil allowed patients to recover quickly and was associated with no adverse effects consistent with other studies.5 6 There was no evidence of a wearing off effect of flumazenil which could result in a recurrence of drowsiness following an initial recovery from sedation.
In summary, conscious sedation is a safe and effective method and an alternative to general anaesthesia in patients undergoing electrical cardioversion.