The possibility of using implantable cardioverter-defibrillators (ICDs) for primary prevention of sudden death in selected high-risk patients has prompted a series of prospective controlled studies. Recently, the MADIT II study highlighted the possibility of effective primary prevention of sudden death in patients with coronary artery disease selected by straightforward clinical data and without expensive screening (electrophysiological study). For patients with previous myocardial infarction and low left ventricular ejection fraction (</=30%), ICD implantation may reduce mortality risk by approximately 31% in the following 2 years. Implementation of this therapeutic strategy threatens to impact on public health-care spending. Possible cost-limiting mechanisms include price cuts because of increasing usage (market forces); identification of subgroups at higher risk of sudden death and use of cheaper devices with limited diagnostic and therapeutic options. Further long-term evaluation of the cost-effectiveness and cost-utility of ICDs should identify subgroups of patients for whom implantation is affordable despite current economic constraints. For heart failure patients, randomized controlled trials are currently evaluating the effects on overall survival of both conventional ICDs and devices with biventricular pacing capabilities. In this perspective, data from the COMPANION trial are expected to stimulate the use of devices with defibrillation back-up in candidates for biventricular pacing.