Summary of randomised controlled trials of implantable cardioverter-defibrillators versus drug treatment to reduce sudden cardiac death: main results
| Author, year | Relative reduction in risk | Absolute results | NNT (95% CI)* |
|---|---|---|---|
| *Calculated by the authors using Arcus software. | |||
| †Mean (SD). | |||
| ARR, absolute risk reduction; CI, confidence interval; ICD, implantable cardioverter-defibrillator; NNT, number needed to treat; RR, relative risk; RRR, relative risk reduction. | |||
| Primary prevention of VT/VF | |||
| Moss (1996)7: MADIT (multicenter automatic defibrillator implantation trial) | RR, ICD arm: 0.46 (95% CI, 0.26 to 0.82; p=0.009); RRR, 54% | Absolute mortality: ICD, 15.8%; conventional treatment, 38.6%; ARR, 22.8% | NNT=5 (3 to 10) |
| Buxton (1999)9,31: MUSTT (multicenter unsustained tachycardia trial) | Absolute all cause mortality in randomised comparison: conservative v EP guided; RRR, 13% | Absolute all cause mortality in randomised comparison: conservative, 48%; EP guided, 42%; ARR 6% | NNT=17 |
| In EPG arm (non-randomised comparison) ICD compared with drug treatment; RRR 56% | In EPG arm (non-randomised comparison) ICD compared with drug treatment: total mortality ICD 24%; drug treatment 55%; ARR 31% | NNT=3 | |
| Bigger (1997)8: CABG patch (coronary artery bypass patch trial) | RR in ICD arm, 1.07 (95% CI, 0.81 to 1.42); p=0.64. Adjusted RR, 1.03 (95% CI, 0.75 to 1.41) | Absolute mortality: ICD, 22.6% at 32 months; control, 20.9% at 32 months. ARR in usual treatment group, 1.7% | NNH=58 (14 to infinity) |
| Secondary prevention (recurrent VT/VF) | |||
| Zipes (1997)10: AVID (antiarrhythmic versus implantable defibrillator) | Relative reduction in total mortality (adjusted) in ICD arm†: 37 (22)% at 1 y; 24 (22)% at 2 y; 29 (23)% at 3 y; p<0.02 | Absolute mortality: ICD, 10.7% at 1y; 18.4% at 2 y; 24.6% at 3 y; amiodarone/sotalol, 17.7% at 1 y; 25.3% at 2 y; 35.9% at 3 y; ARR, 7% at 1 y; 6.9% at 2 y; 11.3% at 3 y | NNT=9 (95% CI, 6 to 18) |
| Kuck (2000)11; Siebels (1993)12: CASH (cardiac arrest study Hamburg) | At 2 years: RR, 0.766 (upper 97.5% CI 1.112); RRR, 23.4%; p=0.081 | Total mortality: ICD, 13.6%; propefenone, 29.3%. Trial stopped Absolute total mortality at 2 years: ICD, 36.4% (95% CI 26.9% to 46.6%); amiodarone/metoprolol, 44.4% (95% CI 37.2% to 51.8%); ARR, 8.0% | NNT=13 (6 to infinity) |
| % Reduction in mortality, year 1 to 9: 41.9, 39.3, 28.4, 27.7, 22.8, 11.4, 9.1, 10.6, 24.7 | |||
| Connolly (2000)14,32: CIDS (Canadian implantable defibrillator study) | RRR at 5 years: 19.7% with ICD (p=0.142) | Absolute mortality at 5 years: ICD, 23%; amiodarone, 27%; ARR, 3.7% | NNT=24 (10 to infinity) |
| Wever (1995)13 | RR of death in ICD arm: 0.27 (0.09 to 0.85); p=0.02 | Absolute mortality: early ICD, 14% at 2 years; conventional group, 35% at 2 years; ARR, 21% at 2 years | NNT=5 (3 to infinity) |