부산대학교 도서관

The implantable cardioverter defibrillator (ICD) is a therapy for patients at risk of sudden cardiac death due to ventricular tachycardia (VT) or ventricular fibrillation (VF). But the apparent high cost of ICD therapy relative to antiarrhythmic drugs such as amiodarone has raised questions about the cost effectiveness of ICD therapy versus drug therapy. To inform this debate we reviewed the literature on ICD cost effectiveness. An electronic and manual search was conducted for articles published since 1980 reporting original data on the cost effectiveness of ICD versus drug therapy for patients at risk of VT/VF. Data on costs and life-years gained were abstracted and studies were grouped into those that used decision-analysis models and those that were trial-based analyses. Cost-effectiveness ratios were inflated to 2001 US dollars. Nine studies were included in the review; five studies were modelling studies and four were part of randomised trials of ICD therapy. Studies varied in time horizon, but all except one indicated that ICD therapy was more costly than drug therapy. Early decision models assumed larger survival benefits than those observed in subsequent trials and therefore had attractive incremental cost-effectiveness ratios in the range of dollars US 27000 to dollars US 60000 per life-year gained. Trial-based studies, with the exception of one small trial, indicated cost per life-year gained in the range dollars US 44000 to dollars US 144000. Stratified analysis shows clearly that patients with a greater risk of mortality due to structural heart disease (e.g. left ventricular ejection fraction < or =35%) benefit more from ICD therapy and therefore have a more attractive cost effectiveness ratio than patients at lower risk. ICD therapy is still evolving over time with implant costs declining and device technology improving. Current evidence is that, in selected patients who are at high risk of VT/VF, ICD therapy can be a cost-effective option. Future research should focus on (i) patient selection to optimise benefits for available resources; and (ii) more comprehensive outcome measures to include health-related quality of life. [ABSTRACT FROM AUTHOR]