Table 1 Characteristics of the included trials.
| Study, year, country of origin | Participants, inclusion and exclusion criteria | Interventions | Outcomes and observed period | Allocation concealment, blinding, % loss to follow‐up, intention to treat analysis, and Jadad's scale (0–5) |
|---|---|---|---|---|
| Kiziltepe et al., 2003, Turkey17 | 20 adult patients after cardiac surgery in ICU with any supraventricular tachyarrhythmias (60% with AF & the duration before treatment <1 week). Exclusion: arrhythmias before surgery, hypokalaemia, hypotension, receiving antiarrhythmics or beta‐blockers | Amiodarone group (n = 10): 150 mg +60 mg/h for 6 h then 30 mg/h for 18 h. | Rhythm conversion, bradycardia or atrioventricular block, hypotension, requiring rescue antiarrhythmics. Observed for 2 h | Allocation concealment unclear, unblinded, all patients completed the study, analysis by intention to treat. 2 |
| Magnesium group (n = 10): 1.5 g over 5–10 min and repeat if no response in 15 min | ||||
| Joshi et al., 1995, India20 | 86 patients in ICU with AF or atrial flutter >160/min (the proportion of AF and flutter & the exact duration of AF before treatment were not reported). | Verapamil group (n = 45): 5 mg intravenous verapamil and repeat if no response after 15 min. | Rhythm conversion, ventricular response rate <100/min, bradycardia or hypotension. Observed for 20 min | Allocation concealment unclear, unblinded, all patients completed the trial, analysis by intention to treat. 2 |
| Exclusion: renal dysfunction, hypotension | Magnesium group (n = 41): 2.0 g and repeat if no response after 15 min | |||
| Walker et al., 1996, Australia21 | 42 patients with either AF (90.2%) or atrial flutter (9.8%) >100/min duration <48 h, at ED. | Controlled group (n = 20): Plain 20 ml normal saline, intravenous digoxin at the discretion of the treating physician after first 30 min (86% at the end). | Rhythm conversion, ventricular response rate <100/min, bradycardia or atrioventricular block, flushing, tingling, dizziness, requiring rescue antiarrhythmics. Observed for 30 min initially followed by a second period of 6 h | Allocation concealment unclear, double‐blinded, 2.3% not able to complete the trial, not by intention to treat. 4 |
| Exclusion: hypotension, renal dysfunction, drug overdose, receiving antiarrhythmic drug other than digoxin | Magnesium group (n = 22): 5 g magnesium in 20 ml normal saline, intravenous digoxin at the discretion of the treating physician after first 30 min (70% at the end) | |||
| Caron et al., 2003, USA22 | 20 patients in cardiology department with AF (60%) or flutter (40%) before intravenous ibutilide treatment (minimal rate and duration not defined). | Controlled group (n = 9): 50 ml normal saline10 min before intravenous ibutilide (1 mg). | Rhythm conversion. Observed for 60 min | Allocation concealment unclear, double‐ blinded, all patients completed the trial, analysis by intention to treat. 4 |
| Exclusion: renal dysfunction, magnesium treatment within 8 h, pregnant, receiving antiarrhythmics | Magnesium group (n = 11): 2 g magnesium over 10 min then 2 g over 1 h, intravenous ibutilide after first 2 g magnesium | |||
| Hays et al., 1994, USA23 | 15 patients with AF >99/min, for a mean duration of 4 to 6 days at ED. | Controlled group (n = 8): placebo (not defined), intravenous 0.5 mg digoxin after first 30 min. | Rhythm conversion, hypotension, flushing, tingling, dizziness. Observed for 4 h | Allocation concealment unclear, double‐blinded, all patients completed the trial, analysis by intention to treat. 3 |
| Exclusion: receiving antiarrhythmics, digoxin, beta‐blockers, calcium antagonists, hypokalaemia, atrioventricular block, hypotension | Magnesium group (n = 7): 2 g over 1 min, then 1 g/h for 4 h, intravenous 0.5 mg digoxin after first 30 min | |||
| Davey et al., 2005, Australia24 | 199 patients with AF >120/min, 62% <24 h in duration, at ED. | Controlled group (n = 97): 100 ml 5% dextrose, intravenous digoxin at the discretion of treating physician (86.2% at the end). | Rhythm conversion, ventricular response rate <100/min, bradycardia or atrioventricular block, flushing, tingling, dizziness, requiring rescue antiarrhythmics. Observed for 150 min | Adequate allocation concealment, triple‐blinded, 8.5% could not complete the trial, analysis by intention to treat. 5 |
| Exclusion: hypotension, renal dysfunction, atrioventricular block, myocardial infarction | Magnesium group (n = 102): 2.5 g over 20 min then 2.5 g over 2 h, intravenous digoxin at the discretion of treating physician (77.6% at the end) | |||
| Brodsky et al., 1994, USA25 | 18 cardiology outpatients with AF >100/min & <7 days in duration. | Controlled group (n = 8): 100 ml 5% dextrose + simultaneous intravenous digoxin 0.375 to 0.625 mg (adjusted with body weight and then up to 3 more doses of digoxin from 0.125 to 0.375 mg). | Rhythm conversion, ventricular response rate <90/min, bradycardia or atrioventricular block. Observed for 24 h | Allocation concealment unclear, double‐blinded, all patients completed the trial, analysis by intention to treat. 4 |
| Exclusion: hypotension, renal dysfunction, receiving antiarrhythmics, digoxin blood concentration >0.8 nmol/l | Magnesium group (n = 10): 2 g in 100 ml 5% dextrose over 15 min then 8 g over 6 h. | |||
| Chiladakis et al., 2001, Greece26 | 46 patients with AF >100/min & <12 h in duration in cardiology department. | Diltiazem group (n = 23): intravenous diltiazem 25 mg over 15 min, then 12.5 mg/h infusion for 6 h. | Rhythm conversion, bradycardia or atrioventricular block, hypotension, flushing, tingling, dizziness, requiring rescue antiarrhythmics. Observed for 6 h | Allocation concealment unclear, single‐blinded, all patients completed the trial, analysis by intention to treat. 2 |
| Exclusion: myocardial infarction, hypotension, atrioventricular block, sick sinus syndrome | Magnesium group (n = 23): 2.5 g over 15 min, then 7.5 g over 6 h | |||
| Moran et al., 1995, Australia27 | 34 ICU patients with AF >120/min, duration 1 h to <6 months. | Amiodarone group (n = 18): intravenous amiodarone 5 mg/kg over 15–20 min, then 10 mg/kg over 24 h. | Rhythm conversion, requiring rescue antiarrhythmics. Observed for 24 h | Adequate allocation concealment, unblinded, 4.8% could not complete the trial, analysis by intention to treat. 3 |
| Exclusion: hypotension, renal dysfunction if not on dialysis, serum potassium concentration <4.0 mmol/l | Magnesium group (n = 16): 0.15 mmol (0.6 g)/kg/h over 5 min, then 0.1 mmol (0.4 g)/kg/hr for 24 h | |||
| Gullestad et al., 1993, Norway28 | 35 hospitalised patients (exact location not defined) with AF >100/min & <1 week in duration. | Verapamil group (n = 20): intravenous verapamil 5 mg over 5 min, repeat 5 mg if no response after 10 min, followed by 0.1 mg/min infusion. | Rhythm conversion, bradycardia ventricular response rate <90/min, or atrioventricular block, hypotension, flushing, tingling, dizziness, requiring rescue antiarrhythmics. Observed for 4 h | Adequate allocation concealment, single‐blinded, 5.7% could not complete the trial, analysis by intention to treat. 3 |
| Exclusion: severe congestive heart failure, hypotension, atrioventricular block, sick sinus syndrome, renal dysfunction, myocardial infarction. | Magnesium group (n = 15): 1.2 g over 5 min, repeat if no response after 10 min, followed by infusion 0.04 mmol (0.01 g)/min |
AF, atrial fibrillation; ICU, intensive care unit.