OUP user menu

The economical challenge in the treatment of chronic heart failure: is primary prophylactic ICD therapy cost-effective in Europe?

Mathias Meine, Tim Smith, Richard N.W. Hauer
DOI: http://dx.doi.org/10.1093/europace/eup140 689-691 First published online: 26 May 2009

Despite great scepticism and opposition in the early years of implantable cadioverter/defibrillator (ICD) therapy, 25 years of clinical practice and multiple randomized multicentre studies have proven that ICD therapy is highly effective to prevent sudden cardiac death.1,2 In patients with or without symptomatic heart failure and impaired left ventricular ejection fraction (LVEF) ≤35%, primary prophylactic ICD therapy reduces mortality. This was clearly shown in patients with ischaemic cardiomyopathy;3 however, the evidence in non-ischaemic cardiomyopathy patients is less convincing.47 In heart failure patients who are candidates for primary prophylactic ICD therapy, an optimal pharmacological treatment is mandatory to improve LV function and to reduce disease progression. Conventional ICD therapy comes on top of an optimal heart failure treatment and does not slow disease progression or reduce symptoms of heart failure. This implicates that primary prophylactic ICD therapy cannot be cost-saving in heart failure patients and raises the question about cost-effectiveness. Cost-effectiveness of this therapy has been thoroughly evaluated in the American population,810 but information about the costs and benefits of prophylactic ICD implantation in Europe is lacking. The milestone trials of primary prophylactic ICD therapy are mainly based on American patients in the American health-care system. Therefore, a cost-effectiveness analysis for European conditions remains a challenge.

Cowie et al.12 present a modelling study in which the cost-effectiveness of primary prophylactic ICD implantation according to the current European guidelines11 is being analysed in the European population. Cowie et al. present a cost-effectiveness analysis based on the following six trials of primary prevention of sudden cardiac death: AMIOVIRT,7 CAT,4 DEFINITE,6 MADIT,13 MADIT II,3 and SCD-HeFT.5 To obtain probability estimates for a Markov model, a meta-analysis of patients with an LVEF ≤ 35% and whether chronic heart failure in NYHA class II and III (non-ischaemic aetiology) or prior myocardial infarction with or without heart failure was performed. The analysis that was based on Belgian reimbursement costs of ICD therapy was performed from the health-care system perspective over the lifetime of patients. Under these conditions, primary prophylactic ICD therapy led to an additional cost of €46 413 and the cost-effectiveness of ICD compared with conventional treatment was €31 717 per quality-adjusted life-year (QALY) gained. For the estimation of a QALY, a utility score of 0.85 was assigned for both ICD and conventional treatment groups (1 = perfect health and 0 = death).

Two other published cost-effectiveness analysis of primary prophylactic ICD therapy revealed a worse cost-effectiveness of $91 300 in MADIT II patients10 and $58 510 in SCD-HeFT patients8 to gain 1 QALY over a period of 12 years, respectively. Because of higher estimated costs of ICD therapy in the calculation of the MADIT II population, the cost-effectiveness was lower than compared with the SCD-HeFT population, despite of a higher ICD efficacy in the MADIT II than in the SCD-HeFT trial. Because of unpredictable disease progression in heart failure patients, it may be very difficult to calculate cost-effectiveness over a long period of time, especially when evidence is only available over the study duration of 3.5 years in MADIT II and 5 years in SCD-HeFT.

The lacking data of ICD effectiveness in Europe and the different health-care system within the European countries make it difficult to calculate reliable ICD cost-effectiveness. Another problem of extrapolation of the data of the milestone trials of primary prophylactic ICD therapy is the fact that these trials enrolled patients from 1990 to 2001, a period in which the treatment of acute myocardial infarction—one of the key players in developing a substrate for sudden death—drastically changed (thrombolysis vs. PCI).

Sensitivity analysis showed that the incremental cost-effectiveness of ICD therapy compared with medical treatment alone was most sensitive to variation in five factors: the efficacy of the ICD in reducing mortality, the cost of ICD implantation, the frequency of generator replacement, the quality of life, and the time horizon used in the analysis.9 In the present study of Cowie et al., the cost-effectiveness of primary prophylactic ICD therapy may be overestimated because of the following reasons.

  1. The cost of ICD implantation in Belgian (single-chamber ICD: €16 650 lead: €1772 + implantation fee: €7789.59) is not representative for Europe. There are huge differences of the health-care systems between the European countries. For example, the reimbursement costs for ICD implantation in the Netherlands are three times higher (€46 56214) than in Germany (€13 726–€16 612, depending on the calculation in the diagnosis-related groups15).

  2. The longevity of the ICDs that were implanted the last years were lower (<5 years)16 than the assumption of 6.5 years in the present analysis. However, cost-effectiveness studies should be based on actual ICD longevity.

  3. The present study was based on a constant ICD efficacy during lifetime. Substudies of MADIT II and SCD-HeFT showed an increased mortality after appropriate ICD shock of, respectively, 50% mortality over 2 years in MADIT II17 and a median survival of 1 year in SCD-HeFT.18 When assuming an incidence of appropriate ICD shock of 7.5% per year,5 Cowie et al. assumed mortality in ICD patients of 8.3% per year may increase dramatically during lifetime. Furthermore, when recurrent appropriate and inappropriate ICD shocks additionally decreased quality of life, the cost-effectiveness of ICD therapy would be less favourable.

  4. In daily practice, the patients who were screened for primary prophylactic ICD therapy are older than the typical SCD-HeFT patient with a mean age of 60 years (range 51–69 years). In patients receiving ICDs, increased age is significantly and progressively associated with shorter survival times after ICD implantation19 that lower the ICD cost-effectiveness. In heart failure patients over age 65, the short-term mortality was in the first year after primary prophylactic ICD implantation 13% and increased in the second year up to 17%,20 much higher than in the cost-effectiveness analysis by Cowie assumed (annual mortality of 8.3%).

  5. The annual mortality of pharmacologically treated heart failure patients without ICD was in the DEFINITE study 6.2% and in the CAT study 5.6%, much lower than the predicted annual mortality in the analysis of Cowie et al. (11.8 %).

  6. The complication rate that was in the present study estimated by the advisory board may be higher in daily practice, especially when the ICD is implanted by a non-electrophysiologist.21

To increase the cost-effectiveness of primary prophylactic ICD therapy, several options have to be discussed.

  1. More effective risk stratification for sudden cardiac death may improve cost-effectiveness of primary prophylactic ICD therapy. As shown in the subanalysis of the SCD-HeFT trial, patients with heart failure in NYHA class III did not benefit from ICD therapy (HR 1.16, P = 0.30).5 There would be no reason to implant an ICD in this subgroup, whereas NYHA class II had a significant relative hazard reduction of 46% and ICD therapy in NYHA class II may be economically attractive.8 In SCD-HeFT, only 17% of patients with an EF >30% were enrolled; in this subgroup, there was not any benefit of the ICD (HR 1.08, 95% CI 0.57–2.07).5 In the MADIT II trial, the upper limit of LVEF was 30%. More extensive risk stratification in all MADIT II-eligible patients with microvolt T-wave alternans improved the cost-effectiveness.22 This highlights the need for an appropriate selection of patients for primary prophylactic ICD implantation that is not only based on the inclusion criteria of the studies but also based on the interpretation of the (sub-)study results.

  2. An optimal programming of the ICD reduced inappropriate and appropriate ICD shocks23 that were independent predictors of mortality.18

  3. Lowering the cost and increasing the longevity of the ICD improve significantly the cost-effectiveness.9

  4. A further improvement of ICD cost-effectiveness may be obtained by focusing the ICD implantation in specialized electrophysiological centres with a high number of ICD implantation per electrophysiologist and the possibility for ablation therapy.

It remains a difficult challenge to analyse the cost-effectiveness of primary prophylactic ICD therapy for Europe. The analysis of ICD cost-effectiveness that is based on meta-analysis of relevant randomized trials in which the majority of study patients were enrolled in centres ‘out of Europe’ and that is calculated with direct medical costs of a small West-European country may not be representative for all the different European health-care settings. Because of the higher reimbursement costs of ICD therapy in several countries and other factors which mainly influence the disease progression of heart failure, primary prophylactic ICD therapy may be less cost-effective over the patient's lifetime than the studies with a maximum follow-up of 5 years have shown. If we were implementing our knowledge about risk stratifying of sudden cardiac death in our decision-making for primary prophylactic ICD implantation, reducing the high complication rate for ICD implantation, increasing the device longevity, improving the lead performance, and lowering the initial device cost—to name but a few—ICD therapy would become economical attractive in Europe.

Conflict of interest: M.M. has received speaker's honoraria from Boston Scientific and St. Jude Medical. T.S. and R.N.W.H.: none declared.


  • The opinions expressed in this article are not necessarily those of the Editors of Europace or of the European Society of Cardiology.


View Abstract