Mechanism

treatment

comparator

Amiodarone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	1.60 [0.53 4.86]	p=1.00	0	673	4	Channer, GEFACA, Kochiadakis (amiodarone vs placebo), SAFE-T (amiodarone vs placebo),
All cause death	1.60 [0.53 4.86]	p=1.00	0	673	4	Channer, GEFACA, Kochiadakis (amiodarone vs placebo), SAFE-T (amiodarone vs placebo),
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	5.59 [1.24 25.27]	p=0.04	0	274	3	Channer, GEFACA, Kochiadakis (amiodarone vs placebo),
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	2.06 [0.62 6.82]	p=1.00	0	673	4	Channer, GEFACA, Kochiadakis (amiodarone vs placebo), SAFE-T (amiodarone vs placebo),
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	0.53 [0.35 0.81]	p=0.04	0	673	4	Channer, GEFACA, Kochiadakis (amiodarone vs placebo), SAFE-T (amiodarone vs placebo),
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
Channer, 2004	Amiodarone 200 mg/d	placebo	patients with Persistent AF
GEFACA, 2001	Amiodarone 200 mg/d	placebo	Persistent AF lasting > 2 months
Kochiadakis (amiodarone vs placebo), 2000	Amiodarone 200 mg/d	placebo	Any documented symptomatic previous or persistent AF
SAFE-T (amiodarone vs placebo), 2005	Amiodarone 300 mg/d	placebo	Persistent AF lasting 3 days to 1 year

Amiodarone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	0.59 [0.34 1.03]	p=1.00	0	405	2	AFFIRM Substudy (amiodarone vs class I drugs), AFFIRM Substudy (sotalol vs class I drugs),
All cause death	0.59 [0.34 1.03]	p=1.00	0	405	2	AFFIRM Substudy (amiodarone vs class I drugs), AFFIRM Substudy (sotalol vs class I drugs),
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	0.33 [0.18 0.61]	p=0.04	0	275	1	AFFIRM Substudy (amiodarone vs class I drugs),
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	0.30 [0.16 0.57]	p=0.04	0	531	2	AFFIRM Substudy (sotalol vs class I drugs), AFFIRM Substudy (amiodarone vs class I drugs),
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	0.63 [0.18 2.20]	p=1.00	0	222	1	AFFIRM Substudy (amiodarone vs class I drugs),
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	0.75 [0.46 1.22]	p=1.00	0	405	2	AFFIRM Substudy (amiodarone vs class I drugs), AFFIRM Substudy (sotalol vs class I drugs),
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
AFFIRM Substudy (amiodarone vs class I drugs), 2003	Amiodarone 200 mg/d ,	class I drugs	patients with AF likely to be recurrent and to cause ilness or deathpj
AFFIRM Substudy (sotalol vs class I drugs), 2003	Amiodarone 200 mg/d , , sotalol	class I drugs	patients with AF likely to be recurrent and to cause ilness or deathpj

Amiodarone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Amiodarone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Amiodarone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Amiodarone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	0.70 [0.42 1.17]	p=1.00	0	910	3	AFFIRM Substudy (amiodarone vs sotalol), Kochiadakis (amiodarone vs sotalol), SAFE-T (amiodarone vs sotalol),
All cause death	0.70 [0.42 1.17]	p=1.00	0	910	3	AFFIRM Substudy (amiodarone vs sotalol), Kochiadakis (amiodarone vs sotalol), SAFE-T (amiodarone vs sotalol),
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	1.11 [0.61 2.00]	p=1.00	0	415	2	AFFIRM Substudy (amiodarone vs sotalol), Kochiadakis (amiodarone vs sotalol),
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	0.61 [0.30 1.24]	p=1.00	0	943	3	AFFIRM Substudy (amiodarone vs sotalol), Kochiadakis (amiodarone vs sotalol), SAFE-T (amiodarone vs sotalol),
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	0.95 [0.27 3.38]	p=1.00	0	256	1	AFFIRM Substudy (amiodarone vs sotalol),
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	0.69 [0.53 0.91]	p=0.04	0	910	3	AFFIRM Substudy (amiodarone vs sotalol), Kochiadakis (amiodarone vs sotalol), SAFE-T (amiodarone vs sotalol),
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
AFFIRM Substudy (amiodarone vs sotalol), 2003	Amiodarone 200 mg/d	Sotalol 240 mg/d	patients with AF likely to be recurrent and to cause ilness or deathpj
Kochiadakis (amiodarone vs sotalol), 2000	Amiodarone 200 mg/d , Amiodarone 200 mg/d		Any documented symptomatic previous or persistent AF
SAFE-T (amiodarone vs sotalol), 2005	Amiodarone 300 mg/d , Amiodarone 300 mg/d		Persistent AF lasting 3 days to 1 year

Amiodarone

cardiac arrest, in out hospital patients

Amiodarone

cardiac arrest, in out hospital patients

Amiodarone

heart failure, in all type of heart failure

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	0.69 [0.53 0.89]	p=0.04	0	-36	2	EPAMSA, GESICA,
All cause death	0.69 [0.53 0.89]	p=0.04	0	-36	2	EPAMSA, GESICA,
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	0.66 [0.42 1.03]	p=1.00	0	-36	2	EPAMSA, GESICA,
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	no data
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	no data
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	no data
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
EPAMSA, 1985	amiodarone 400 mg/day	no treatment	patients with reduced left ventricular ejection fraction ( < 35%) and asymptomatic ventricular arrhythmias (Lown classes 2 and 4) LVEF <=35% and Lown class 2–5
GESICA, 1994	amiodarone 300 mg/day	no treatment	patients with severe heart failure Any two of CTR >0·55, LVEF<=35%, echo LVED >3·2 cm/m2

Amiodarone

heart failure, in all type of heart failure

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	0.96 [0.74 1.25]	p=1.00	0	-54	3	Hamer, Nicklas, STATCHF,
All cause death	0.96 [0.74 1.25]	p=1.00	0	-54	3	Hamer, Nicklas, STATCHF,
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	0.87 [0.60 1.24]	p=1.00	0	-54	3	Hamer, Nicklas, STATCHF,
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	no data
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	no data
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	no data
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
Hamer, 1989	amiodarone 200 mg/day	placebo	patients with severe congestive heart failure but no sustained ventricular arrhythmia
Nicklas, 1991	amiodarone 200 mg/day	placebo	patients with ejection fractions less than 30%, New York Heart Association class III or IV symptoms, and frequent but asymptomatic spontaneous ventricular ectopy (Lown class II to V) LVEF <=30% and Lown class 2–5
STATCHF, 1995	amiodarone 200 mg/day	placebo	patients with symptoms of congestive heart failure, cardiac enlargement, 10 or more premature ventricular contractions per hour, and a left ventricular ejection fraction of 40 percent or less LVEF <=40% and >=10 VPD/h and LVED >=55 mm or CTR >0·55

Amiodarone

heart failure, in all type of heart failure

Amiodarone

post myocardial infarction, in all type of patients

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	0.44 [0.22 0.86]	p=0.04	0	350	2	BASIS, Navarro-Lopez,
All cause death	0.44 [0.22 0.86]	p=0.04	0	350	2	BASIS, Navarro-Lopez,
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	no data
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	no data
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	0.56 [0.24 1.30]	p=1.00	0	350	2	BASIS, Navarro-Lopez,
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	no data
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
BASIS, 1990	amiodarone 1 g for 5 d; then 200 mg/d started within 4 weeks of AMI	no amiodarone (usual care)	patients with persisting asymptomatic complex arrhythmias after myocardial infarction (Lown class 3 or 4b in > 2 of 24 h)
Navarro-Lopez, 1993	amiodarone 600 mg/d for 1 week, 400 mg/d for 1 week then 200 mg/d started 10-30 d after AMI	no amiodarone	patients who have had MI with a left ventricular ejection fraction of 20 to 45% and > or = 3 ventricular premature complexes per hour (pairs or runs) - 3 VPOs/h, pairs or runs of VT

Amiodarone

post myocardial infarction, in all type of patients

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	0.84 [0.64 1.09]	p=1.00	0	724	3	CAMIAT , Ceremuzynski, Hockings,
All cause death	0.84 [0.64 1.09]	p=1.00	0	724	3	CAMIAT , Ceremuzynski, Hockings,
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	no data
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	no data
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	0.68 [0.45 1.01]	p=1.00	0	724	3	CAMIAT , Ceremuzynski, Hockings,
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	no data
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
CAMIAT , 1991	amiodarone 10 mg/kg per d for 3 weeks then 300-400 mg/d started 6-45 d after AMI	placebo	patients with acute myocardial infarction within the previous 6-30 days and > 10 VPDs/h for 18 h or a run of VT
Ceremuzynski, 1992	amiodarone 800 mg/d for 1 week then 200-400 mg/d started 5-7 d after AMI	placebo	No need for antiarrhythmic therapy
Hockings, 1987	amiodarone 200 mg 3 times daily for 1 wk; then 200 mg/d started < 8-10 d after AMI	placebo	patients with AMI - Absence of VF or VT > 3 beats

Moricizine

acute myocardial infarction, in all type of patients

Moricizine

post myocardial infarction, in all type of patients

Disopyramide

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	2.59 [0.23 29.18]	p=1.00	0	146	2	Karlson, Lloyd (Disopyramide vs placebo),
All cause death	2.59 [0.23 29.18]	p=1.00	0	146	2	Karlson, Lloyd (Disopyramide vs placebo),
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	3.50 [0.69 17.86]	p=1.00	0	92	1	Karlson,
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	0.93 [0.06 15.16]	p=1.00	0	146	2	Karlson, Lloyd (Disopyramide vs placebo),
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	0.77 [0.39 1.52]	p=1.00	0	146	2	Karlson, Lloyd (Disopyramide vs placebo),
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
Karlson, 1998	Disopyramide 500 mg/d	palcebo	Persistent AF between 6 weeks and 1 year
Lloyd (Disopyramide vs placebo), 1984	Disopyramide 450 mg/d ,	placebo	Persistent AF lasting 1 month to 3 years

Disopyramide

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Disopyramide

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Quinidine

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Quinidine

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	1.75 [0.54 5.65]	p=1.00	0	1361	4	Lloyd (quinidine vs placebo), PAFAC (quinidine vs placebo), SOPAT (quinidine vs placebo), Byrne Quinn,
All cause death	1.75 [0.54 5.65]	p=1.00	0	1361	4	Lloyd (quinidine vs placebo), PAFAC (quinidine vs placebo), SOPAT (quinidine vs placebo), Byrne Quinn,
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	0.89 [0.66 1.19]	p=1.00	0	1362	4	Lloyd (quinidine vs placebo), PAFAC (quinidine vs placebo), SOPAT (quinidine vs placebo), Byrne Quinn,
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	1.76 [0.71 4.37]	p=1.00	0	1361	4	Lloyd (quinidine vs placebo), PAFAC (quinidine vs placebo), SOPAT (quinidine vs placebo), Byrne Quinn,
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	0.85 [0.63 1.15]	p=1.00	0	1361	4	Lloyd (quinidine vs placebo), PAFAC (quinidine vs placebo), SOPAT (quinidine vs placebo), Byrne Quinn,
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
Lloyd (quinidine vs placebo), 1984	quinidine 1.4 g/d	placebo	Persistent AF lasting 1 month to 3 years
PAFAC (quinidine vs placebo), 2004	Quinidine 0,480 g/d (+ verapamil	placebo	Persistent AF lasting > 7 daysil
SOPAT (quinidine vs placebo), 2004	Quinidine 0,320 or 0,480 g/d (+ verapamil)	placebo	Paroxysmal AF documented in the last 1 month@4
Byrne Quinn, 1979	Quinidine 1.2 g/d	placebo	Persistent AF

Quinidine

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Quinidine

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	1.00 [0.02 53.89]	p=1.00	0	30	1	Steinbeck (quinidine vs flecainide),
All cause death	1.00 [0.02 53.89]	p=1.00	0	30	1	Steinbeck (quinidine vs flecainide),
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	5.00 [0.22 114.28]	p=1.00	0	30	1	Steinbeck (quinidine vs flecainide),
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	2.00 [0.16 24.77]	p=1.00	0	30	1	Steinbeck (quinidine vs flecainide),
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	1.67 [0.38 7.39]	p=1.00	0	30	1	Steinbeck (quinidine vs flecainide),
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
Steinbeck (quinidine vs flecainide), 1988	Quinidine 1 g/d (+ digoxine) , Quinidine 1 g/d (+ digoxine)		Paroxysmal symptomatic AF of any duration

Quinidine

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	0.73 [0.36 1.47]	p=1.00	0	1978	6	Hohnloser, Juul-Moller, Kalusche, SOCESP, PAFAC (quinidine vs sotalol), SOPAT (quinidine vs sotalol),
All cause death	0.73 [0.36 1.47]	p=1.00	0	1978	6	Hohnloser, Juul-Moller, Kalusche, SOCESP, PAFAC (quinidine vs sotalol), SOPAT (quinidine vs sotalol),
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	1.06 [0.85 1.33]	p=1.00	0	1978	6	Hohnloser, Juul-Moller, Kalusche, SOCESP, PAFAC (quinidine vs sotalol), SOPAT (quinidine vs sotalol),
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	0.98 [0.57 1.67]	p=1.00	0	1978	6	Hohnloser, Juul-Moller, Kalusche, SOCESP, PAFAC (quinidine vs sotalol), SOPAT (quinidine vs sotalol),
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	0.97 [0.80 1.18]	p=1.00	0	1978	6	Hohnloser, Juul-Moller, Kalusche, SOCESP, PAFAC (quinidine vs sotalol), SOPAT (quinidine vs sotalol),
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
Hohnloser, 1995	Quinidine 1 g/d	sotalol 240-320 mg/dt	Persistent AF between 2 days and 6 months
Juul-Moller, 1990	Quinidine 1,2 g/d	Sotalol 160-320 mg/dt	Persistent AF between 2 months and 1 year
Kalusche, 1994	Quinidine 1 g/d	Sotalol 240-400 mg/dt	AF lasting from 2 weeks to 2 years
SOCESP, 1999	Quinidine 700 mg/d	sotalol 240 mg/d	AF lasting < 6 months
PAFAC (quinidine vs sotalol), 2004	Quinidine 0,480 g/d (+ verapamil , Quinidine 0,480 g/d (+ verapamil		Persistent AF lasting > 7 daysil
SOPAT (quinidine vs sotalol), 2004	Quinidine 0,320 or 0,480 g/d (+ verapamil) , Quinidine 0,320 or 0,480 g/d (+ verapamil)		Paroxysmal AF documented in the last 1 month@4

Quinidine

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Flecainide

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Flecainide

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Flecainide

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Flecainide

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Flecainide

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Propafenone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	0.34 [0.07 1.72]	p=1.00	0	1098	5	Bellandi (propafenone vs placebo), Dogan, Kochiadakis b (propafenone vs placebo), RAFT, Stroobandt,
All cause death	0.34 [0.07 1.72]	p=1.00	0	1098	5	Bellandi (propafenone vs placebo), Dogan, Kochiadakis b (propafenone vs placebo), RAFT, Stroobandt,
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	1.56 [0.96 2.53]	p=1.00	0	1098	5	Bellandi (propafenone vs placebo), Dogan, Kochiadakis b (propafenone vs placebo), RAFT, Stroobandt,
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	0.96 [0.27 3.39]	p=1.00	0	1098	5	Bellandi (propafenone vs placebo), Dogan, Kochiadakis b (propafenone vs placebo), RAFT, Stroobandt,
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	0.67 [0.50 0.89]	p=0.04	0	1098	5	Bellandi (propafenone vs placebo), Dogan, Kochiadakis b (propafenone vs placebo), RAFT, Stroobandt,
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
Bellandi (propafenone vs placebo), 2001	Propafenone 900 mg/d/d after pharmacological or electrical cardioversion	placebo	patients with paroxysmal recurrent or persistent AF
Dogan, 2004	Propafenone 450 mg/d	placebo	AF of duration 3 hours to 3 months ???
Kochiadakis b (propafenone vs placebo), 2004	Propafenone 450 mg/d	placebo	Any documented symptomatic previous or persistent AF
RAFT, 2003	Propafenone at various doses (450, 650, 850 mg/d)	placebo.	Previous symptomatic AF documented in the last year
Stroobandt, 1997	Propafenone 450 mg/d	placebo	Recent-onset AF or persistent AF lasting > 2 weeks

Propafenone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Propafenone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Propafenone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Metoprolol

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Azimilide

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Azimilide

post myocardial infarction, in all type of patients

Dofetilide

acute myocardial infarction, in all type of patients

Dofetilide

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Dofetilide

post myocardial infarction, in all type of patients

Sotalol

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	1.73 [0.76 3.90]	p=1.00	0	2078	8	Benditt, Singh, Bellandi (sotalol vs placebo), Carunchio (sotalol vs placebo), Kochiadakis b (sotalol vs placebo), PAFAC (sotalol vs placebo), SAFE-T (sotalol vs placebo), SOPAT (sotalol vs placebo),
All cause death	1.73 [0.76 3.90]	p=1.00	0	2078	8	Benditt, Singh, Bellandi (sotalol vs placebo), Carunchio (sotalol vs placebo), Kochiadakis b (sotalol vs placebo), PAFAC (sotalol vs placebo), SAFE-T (sotalol vs placebo), SOPAT (sotalol vs placebo),
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	1.10 [0.80 1.51]	p=1.00	0	1685	7	Benditt, Singh, Bellandi (sotalol vs placebo), Carunchio (sotalol vs placebo), Kochiadakis b (sotalol vs placebo), PAFAC (sotalol vs placebo), SOPAT (sotalol vs placebo),
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	2.06 [1.05 4.04]	p=0.04	0	2078	8	Benditt, Singh, Bellandi (sotalol vs placebo), Carunchio (sotalol vs placebo), Kochiadakis b (sotalol vs placebo), PAFAC (sotalol vs placebo), SAFE-T (sotalol vs placebo), SOPAT (sotalol vs placebo),
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	0.79 [0.64 0.99]	p=0.04	0	2078	8	Benditt, Singh, Bellandi (sotalol vs placebo), Carunchio (sotalol vs placebo), Kochiadakis b (sotalol vs placebo), PAFAC (sotalol vs placebo), SAFE-T (sotalol vs placebo), SOPAT (sotalol vs placebo),
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
Benditt, 1999	Sotalol various doses (80, 120, 160 mg/d) after cardioversion	placebo	patients with AF or AFl documented in the last 3 months
Singh, 1991	Sotalol 80 - 320 mg/d	placebo	Persistent AF or AFl lasting 2 weeks to 1 year
Bellandi (sotalol vs placebo), 2001	sotalol 240 mg/d	placebo	patients with paroxysmal recurrent or persistent AF
Carunchio (sotalol vs placebo), 1995	sotalol 240 mg/d ,	placebo	patients with recurrent AF with > 3 episodes in previous 1 year
Kochiadakis (sotalol vs placebo), 2000	sotalol 320 mg/d	placebo	Any documented symptomatic previous or persistent AF
Kochiadakis b (sotalol vs placebo), 2004	sotalol 300 mg/d	placebo	Any documented symptomatic previous or persistent AF
PAFAC (sotalol vs placebo), 2004	sotalol 320 mg/d ,	placebo	Persistent AF lasting > 7 daysil
SAFE-T (sotalol vs placebo), 2005	sotalol 320 mg/d	placebo	Persistent AF lasting 3 days to 1 year
SOPAT (sotalol vs placebo), 2004	sotalol 320 mg/d ,	placebo	Paroxysmal AF documented in the last 1 month@4

Sotalol

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Sotalol

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	1.01 [0.02 51.58]	p=1.00	0	171	1	Kochiadakis b (sotalol vs propafenome),
All cause death	1.01 [0.02 51.58]	p=1.00	0	171	1	Kochiadakis b (sotalol vs propafenome),
Coronary death	no data
Cardiovascular death	no data
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	no data
Adverse events leading to treatment discontinuation	1.01 [0.28 3.63]	p=1.00	0	171	1	Kochiadakis b (sotalol vs propafenome),
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	1.52 [0.25 9.32]	p=1.00	0	171	1	Kochiadakis b (sotalol vs propafenome),
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	1.24 [0.68 2.28]	p=1.00	0	171	1	Kochiadakis b (sotalol vs propafenome),
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
Kochiadakis b (sotalol vs propafenome), 2004	sotalol 300 mg/d , , sotalol 300 mg/d ,		Any documented symptomatic previous or persistent AF

Sotalol

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Dronedarone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	0.92 [0.72 1.17]	p=1.00	0	9213	4	DAFNE, EURIDIS ADONIS (pooled analysis), ATHENA, PALLAS,
All cause death	0.92 [0.72 1.17]	p=1.00	0	9213	4	DAFNE, EURIDIS ADONIS (pooled analysis), ATHENA, PALLAS,
Coronary death	no data
Cardiovascular death	0.71 [0.51 0.98]	p=0.04	0	4628	1	ATHENA,
Heart failure	no data
Major bleeding	no data
stroke (fatal and non fatal)	2.35 [1.04 5.34]	p=0.04	0	4386	2	EURIDIS ADONIS (pooled analysis), PALLAS,
myocardial infarction (fatal and non fatal)	1.00 [0.20 4.95]	p=1.00	0	3149	1	PALLAS,
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	0.99 [0.78 1.26]	p=1.00	0	7777	2	ATHENA, PALLAS,
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	0.84 [0.75 0.95]	p=0.04	0	7777	2	ATHENA, PALLAS,
Hypotension	no data
Adverse events leading to treatment discontinuation	1.55 [1.30 1.85]	p=0.04	0	6064	3	DAFNE, EURIDIS ADONIS (pooled analysis), ATHENA,
non cardiovascular death	1.09 [0.74 1.62]	p=1.00	0	4628	1	ATHENA,
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	0.32 [0.01 16.24]	p=1.00	0	199	1	DAFNE,
Ventricular fibrillation	no data
II-III atrio ventricular block	no data
Cardiogenic shock	no data
severe arrhythmia	no data
Bradycardia	no data
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	0.55 [0.34 0.89]	p=0.04	0	4628	1	ATHENA,
Vascular noncardiac death	0.84 [0.46 1.53]	p=1.00	0	4628	1	ATHENA,
Cardiac nonarrhythmic death	0.96 [0.49 1.86]	p=1.00	0	4628	1	ATHENA,
hospitalization for AF	0.66 [0.57 0.77]	p=0.04	0	4628	1	ATHENA,
Atrial fibrillation recurrence	0.71 [0.64 0.79]	p=0.04	0	6064	4	EURIDIS, ADONIS, DAFNE, ATHENA,
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	0.85 [0.77 0.94]	p=0.04	0	9014	4	EURIDIS, ADONIS, ATHENA, PALLAS,
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
EURIDIS, 2007	dronedarone 400 mg twice daily	placebo	patients with at least one episode of atrial fibrillation in the preceding 3 months, and in sinus rhythm for at least 1 hour before randomization
ADONIS, 2007	dronedarone 400 mg twice daily	placebo	patients with at least one episode of atrial fibrillation in the preceding 3 months, and in sinus rhythm for at least 1 hour before randomization
DAFNE, 2003	Dronedarone various doses (800, 1200, 1600 mg/d)	placebo	patients with Persistent AF
EURIDIS ADONIS (pooled analysis), 2009	Dronedarone 800 mg/d	placebo	AF documented in the previous 3 months
ATHENA, 2009	dronedarone 400 mg twice a day	placebo	patients (>70y) with paroxysmal or persistent atrial fibrillation and additional risk factors for death
PALLAS, 2011	Dronedarone	placebo	patients with permanent atrial fibrillation and additional risk factors

Dronedarone

atrial fibrillation, in maintaining sinus rhythm after cardioversion

Dronedarone

atrial fibrillation, in rate control

Dronedarone

atrial fibrillation, in prevention of cardiovascular events

Dronedarone

heart failure, in all type of heart failure

Magnesium

acute myocardial infarction, in all type of patients

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	1.01 [0.93 1.09]	p=1.00	0	38129	3	ISIS-4, Wu, Zhu,
All cause death	1.01 [0.93 1.09]	p=1.00	0	38129	3	ISIS-4, Wu, Zhu,
Coronary death	no data
Cardiovascular death	no data
Heart failure	1.07 [1.03 1.12]	p=0.04	0	58277	2	ISIS-4, Wu,
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	1.11 [1.06 1.16]	p=0.04	0	58050	1	ISIS-4,
Adverse events leading to treatment discontinuation	no data
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	no data
Ventricular fibrillation	0.89 [0.82 0.97]	p=0.04	0	61456	3	ISIS-4, Wu, Zhu,
II-III atrio ventricular block	1.07 [1.03 1.12]	p=0.04	0	58277	2	ISIS-4, Wu,
Cardiogenic shock	1.10 [1.01 1.19]	p=0.04	0	61229	2	ISIS-4, Zhu,
severe arrhythmia	0.39 [0.20 0.74]	p=0.04	0	227	1	Wu,
Bradycardia	7.42 [4.06 13.57]	p=0.04	0	58050	1	ISIS-4,
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	no data
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
ISIS-4, 1995	24 h of intravenous magnesium sulphate (8 mmol initial bonus injection over about 15 minutes followed by 72 mmol in about 50 mLinfused over 24 h)4	no magnesium infusion	patients entering 1086 hospitals up to 24 h (median 8 h) after the onset of suspected acute myocardial infarction with no clear contraindications4
Wu, 1992	2.5 g MgSO4 once or twice a day for 7-14 dayssce	usual care	suspected AMI
Zhu, 2002	100 mL (4 g) potassium-magnesium aspartate IV. for the first day, 50 ml for rest 4 datio	routine AMI treatmentƒkB	AMI

Magnesium

acute myocardial infarction, in all type of patients

Endpoint	TE [95% CI]	p val	I2	n	k	trials
All cause death	0.88 [0.78 0.99]	p=0.04	0	11020	19	Abraham, MAGIC, Bhargava, Ceremuzynski, Feldstedt, Gyamlani, Morton, Nakashima, Raghu, Rasmussen, Santoro, Shechter, Shechter, Shechter, Singh, Smith, Thogersen, Urek, Woods,
All cause death	0.88 [0.78 0.99]	p=0.04	0	11020	19	Abraham, MAGIC, Bhargava, Ceremuzynski, Feldstedt, Gyamlani, Morton, Nakashima, Raghu, Rasmussen, Santoro, Shechter, Shechter, Shechter, Singh, Smith, Thogersen, Urek, Woods,
Coronary death	no data
Cardiovascular death	no data
Heart failure	0.97 [0.88 1.09]	p=1.00	0	9714	10	MAGIC, Feldstedt, Gyamlani, Morton, Nakashima, Parikka, Rasmussen, Santoro, Shechter, Woods,
Major bleeding	no data
stroke (fatal and non fatal)	no data
myocardial infarction (fatal and non fatal)	no data
Death from any cause or hospitalization for any reason	no data
hospitalisation for heart failure	no data
Deep vein thrombosis	no data
Sudden death	no data
cardiovascular events	no data
Hypotension	1.07 [0.76 1.51]	p=1.00	0	2618	4	Abraham, Morton, Singh, Woods,
Adverse events leading to treatment discontinuation	no data
non cardiovascular death	no data
stroke or TIA	no data
Peripheral embolism	no data
Atrial fibrillation follow-up	no data
Hospitalization for any reason	no data
Pro-arrhythmia	no data
Ventricular fibrillation	0.81 [0.39 1.72]	p=1.00	0	936	8	Abraham, Bhargava, Ceremuzynski, Chen, Feldstedt, Gyamlani, Morton, Nakashima,
II-III atrio ventricular block	0.97 [0.88 1.08]	p=1.00	0	9924	13	MAGIC, Feldstedt, Gyamlani, Morton, Nakashima, Parikka, Rasmussen, Santoro, Shechter, Shechter, Shechter, Urek, Woods,
Cardiogenic shock	0.53 [0.30 0.93]	p=0.04	0	1105	10	Abraham, Bhargava, Ceremuzynski, Gyamlani, Nakashima, Rasmussen, Shechter, Shechter, Shechter, Singh,
severe arrhythmia	0.85 [0.71 1.03]	p=1.00	0	7126	9	MAGIC, Chen, Rasmussen, Shechter, Shechter, Shechter, Singh, Smith, Urek,
Bradycardia	1.47 [1.12 1.93]	p=0.04	0	3218	7	Abraham, Bhargava, Gyamlani, Raghu, Rasmussen, Santoro, Woods,
Flushing	no data
non fatal VF	no data
fatal and non fatal ventricular fibrillation	no data
fatal ventricular fibrillation	no data
Arrhythmic/sudden death	no data
Cardiac arrhythmic death	no data
Vascular noncardiac death	no data
Cardiac nonarrhythmic death	no data
hospitalization for AF	no data
Atrial fibrillation recurrence	no data
survival to hospital admission	no data
Failure of return of spontaneous circulation	no data
Death before hospital admission	no data
Death before hospital discharge	no data
death or hospitalization for cardiac causes	no data
death by cardiac arrhythmia	no data
combined endpoint	no data

TE: treatment effect; CI: confidence interval; k: nulmber of trials; n: total number of patients

Trial	Studied treatment	Control	Patients
Abraham, 1987	2.4g of magnesium sulfate in 50 ml of 5% glucose solution intravenously over a 20 minutes period for 3 days	50 ml of 5% glucose solution alone, imag	patients with AMI
MAGIC, 2000	2 g intravenous bolus of MgSO4 over 15 minutes, followed by a 17 g infusion of MgSO4 over 24 h	matched intravenous bolus and 24 h infusion of sterile waterndºã	AMi patients within 6 h of onset of symptomsm
Bhargava, 1995	8 mmol magnesium sulphate over 5 min followed by 65 mmol over 24-h infusion	isotonic saline infusion	proven AMI patients with chest pain of 1-6h
Ceremuzynski, 1989	8 g MgSO4 in 500 mL 15% glucose for 24 h intravenouslypj	conventional treatmenton	patients with AMI within 12 h from onset of symptomslypj
Chen, 1991	MgSO4 2g/day for 3 daysn	5% glucose	patients with AMI
Feldstedt, 1991	continuous infusion of 80 mmol magnesium chloride in 1000 mL dextrose˜K	matching placebo	patients, aged 75 y or less, with suspected AMI less than 8 h+—²
Gyamlani, 2000	magnesium 12g (50 mmol) in the first 24h, 3g (12 mmol) in the second 24h used within 2h after admission and within 30 minutes of thrombolytic therapy™	equal volume of isotonic glucose	patients with proven AMI
Ising, 1990	81 mval/daymagnesium sulphate infusion 13+/-9h after the onset of severe pain for 3 days	80 mval/day NaCl infusion for 3 daysimag	patients with AMI
Morton, 1984	36 h intravenous infusion of magnesium sulphate (0.75 mEq/kg/body weight/12 h).	saline solution infusion	patients with AMI within 8 h of onsetmag
Nakashima, 2004	bolus injection of 8 mmol of magnesium followed by an infusion of 24 mmol over 24 h	equivalent amount of normal saline, imag	patients with successful PCI weree, imag
Parikka, 1990	8mmol MgSO4 in 10 min, 62 mmol in 24hmag	NaCl ¯B	patients with < 12 h from onset of chest pain AMImage/pj
Raghu, 1999	18 g (75.6 mmol) of Mg sulphate over 24 h started immediately after completion of thrombolytic therapy	equivalent amount of salinexbitm	confirmed AMI < 6 h from the onset of symptomsce
Rasmussen, 1986	50 mmol MgCl2 during the first 24 h, 12 mmol during the second 24 h	isotonic glucose	patients with suspected AMIxbitm
Santoro, 2000	MgSO4 7 g (28 mmol) with 5 hon	matching saline solution
Shechter, 1990	magnesium 22 g (91.6 mmol) within 48 h (67 mmol within first 24 h).	isotonic glucose.	patients with admission diagnosis of AMI
Shechter, 1991	22 g (91.6 mmol) within 48 h (67 mmol within first 24 h).	isotonic glucose.	patients with documented AMIbitm
Shechter, 1995	magnesium 22 g (91.6mmol) within 48 h (67mmol within first 24 h)pj	isotonic glucose	suspected with AMI and considered unsuitable candidates for thrombolysis
Singh, 1990	5 g (8.12 mmol) of MgSO4 daily for 4 daysptomsce	2% dextrose solution for 3 daysm	patients suspected with AMI within 8-12h of the onset of MI
Smith, 1986	65 mmol MgSO4 given over 24 h	Saline	patients with suspected AMI h.tm
Thogersen, 1995	magensium 50 mmol within 24 h	isotonic NaCl.	patients with suspected AMI
Urek, 1996	17 g MgSO4 with first 24 h.xbitm	saline.	patients with documented AMIbitm
Woods, 1992	magnesium 8 mmol over 5 min, 65 mmol over 24h imag	physiological saline hon	patients with suspected AMI in the preceding 24h

antiarrythmic drugs