OUP user menu

Implantable cardioverter defibrillators: from evidence of trials to clinical practice

Giuseppe Boriani, Renato Ricci, Tiziano Toselli, Roberto Ferrari, Angelo Branzi, Massimo Santini
DOI: http://dx.doi.org/10.1093/eurheartj/sum060 I66-I73 First published online: 26 November 2007


A series of landmark randomized trials has validated the role of implantable cardioverter defibrillators (ICDs) not only in the setting of secondary prevention of sudden cardiac death (SCD) but also in the challenging subset of primary prevention of SCD, i.e. for high-risk patients without previous malignant ventricular tachyarrhythmias. Despite definite indications provided by consensus guidelines, the use of ICDs in clinical practice still encounters a series of barriers mainly related to the characteristics of such treatment (a ‘rescue’ treatment) and its up-front cost, resulting in substantial under-referral and rationing. Cost is likely to remain a major determinant of full acceptance, and implementation of ICD therapy and the problem of how broadened evidence-based indications for implantation can be translated into routine clinical practice require an analysis of available economic resources and identification of priorities for health care. Economic analysis (cost-effectiveness, cost–utility and cost–benefit estimates) provides the most appropriate tool for weighing ICD costs against likely eventual outcome benefits. A series of data indicate that the use of ICDs in appropriately selected high-risk SCD patients is associated with cost-effectiveness ratios similar to, or better than, other accepted treatments, such as renal dialysis. Improvement in risk stratification for SCD and assessment of the cost-effectiveness profile of ICD treatment in specific subgroups of patients appears to be a crucial step in any attempt to maximize health outcomes in a context of limited economic resources.

  • Cardioverter defibrillator
  • Cost
  • Cost-effectiveness
  • Health care system
  • Heart failure
  • Sudden cardiac death

Prevention of sudden cardiac death: size of the problem

Sudden cardiac death (SCD) is defined as an unexpected death from cardiac causes following sudden cardiac arrest occurring within 1 h of onset of acute symptoms.1 Prevention of SCD is an important objective of health care systems. Indeed, SCD is the most common cause of death in developed western countries,1,2 accounting for more victims each year than AIDS, lung cancer, breast cancer, or stroke. Worldwide, over 3 million people are estimated to die from SCD each year and despite great efforts in this field <3–10% of patients survive an episode of cardiac arrest.1,2 Although a reduction in total cardiac mortality was observed between 1989 and 1999 in the USA, the proportion of deaths with characteristics of SCD was shown to have increased.3

Preventing SCD is a complex issue in view of a wide range of subjects who may be exposed to the risk of an unexpected malignant ventricular tachyarrhythmia and the variable chain of pathophysiogical events involved in the sequence that ultimately leads to ventricular fibrillation.4 Thus, termination of ventricular tachyarrhythmias through an electrical shock delivered by an ICD represents a ‘rescue’ treatment to be combined with a wide range of preventive strategies, such as correction of life style, correction of risk factors, pharmacological treatment of underlying heart disease and modulating factors, coronary revascularization, etc. As a matter of fact the limitations of preventive strategies and the evidence collected on the efficacy of ICD in avoiding SCD by prompt termination of malignant ventricular tachyarrhythmias have validated the role of ICD within an integrated approach for preventing SCD in patients judged to be at high risk of SCD.

The role of ICDs in secondary and primary prevention of sudden cardiac death

Current knowledge on the role of ICD in preventing SCD derives from a series of randomized clinical trials that have demonstrated that ICD therapy improves survival both in the setting of secondary prevention of SCD (i.e. in patients with a previous aborted cardiac arrest or a previous life-threatening ventricular tachycardia) or in the setting of primary prevention of SCD (i.e. in patients who had not previously experienced a life-threatening ventricular tachyarrhythmia or an equivalent symptom).

The ICD was conceived >30 years ago as a treatment for highly selected patients who had survived at least two episodes of cardiac arrest. In the first years of clinical use, implants were limited to the field of secondary prevention of SCD and the overall implant number remained low, both in the USA and in Europe.57 In subsequent years, as a consequence of growing evidence of efficacy from randomized clinical trials,7 the use of the ICD in the field of secondary prevention gradually expanded. However, since only 3–10% of patients experiencing cardiac arrest are effectively resuscitated and survive the event, the potential benefit of secondary prevention of SCD is necessarily limited to a small number of patients and has a limited impact on the overall burden of SCD.4

In the last decade, a series of large randomized trials have provided strong evidence that in appropriately selected patients with left ventricular dysfunction, the use of ICDs improves overall survival at 2–5 years.68 Demonstrated efficacy of ICDs in primary prevention was initially established in patients with previous myocardial infarction and left ventricular dysfunction (MADIT I, MUSTT, MADIT II trials),58 and was then extended to patients with left ventricular dysfunction and heart failure (NYHA class II and III) of either ischaemic or non-ischaemic aetiology (SCD-HeFT trial).9 The benefits of ICDs in primary prevention of SCD are additional to optimized pharmacological treatment, and the evidence supporting them is even stronger than for secondary prevention of SCD7 (Figure 1). A meta-analysis incorporating data from 10 primary prevention trials involving over 9000 patients (with around 5500 interventions and 3700 controls) indicated that ICD treatment led to a reduction in arrhythmic mortality and this translates also in a reduction of all-cause mortality, although with some degree of heterogeneity among the trials.7 Considering the evidence from the individual trials (Figure 1), as well as the results of various meta-analyses,58 it seems clear that in appropriately selected patients with left ventricular dysfunction at high risk of SCD, ICDs are effective in improving overall survival at 2–5 years. In this patient setting, current guidelines provide definite indications for device therapy, with an ICD or a biventricular ICD for cardiac resynchronization therapy (the so-called CRT-D devices) (Table 1).

Figure 1

Results of studies evaluating the efficacy of ICDs in various clinical conditions and underlying heart diseases: data from either randomized clinical trials (AVID, CIDS, MADIT, MADIT-II, and SCD-HeFT) or observational studies are reported in terms of annual rates of death, appropriate and inappropriate shocks, respectively. ARVD, arrhythmogenic right ventricular dysplasia; CAD, coronary artery disease; HCM, hypertrophic cardiomyopathy; LQTS, long QT syndrome; Pr Prev, primary prevention; Sec Prev, secondary prevention.

View this table:
Table 1

Current indications for device therapy in patients with left ventricular ejection fraction ≤35% (ischaemic or non-ischaemic aetiology)

QRS <120 msQRS ≥120 ms
NYHA class IICD if LVEF ≤30% and previous MI
ICD if previous MI ≥40 days + LVEF 31–35% + nSVT + VT/VF inducibility at EPS
NYHA class IVNo device (CRT if dyssynchrony detected on echo?)CRT/CRT-D
  • CRT, cardiac resynchronization therapy; CRT-D, cardiac resynchronization therapy + defibrillation capabilities; EPS, electrophysiological study; ICD, implantable cardioverter-defibrillator; LVEF, left ventricular ejection fraction; MI, myocardial infarction; nSVT, non-sustained ventricular tachycardia; NYHA, New York Heart Association; VT/VF, ventricular tachycardia/ventricular fibrillation.

Implantable cardioverter defibrillators are currently used also in clinical conditions at high risk of premature SCD, in the absence of heart failure or left ventricular dysfunction. These conditions include hypertrophic cardiomyopathy and a series or ‘purely electrical’ diseases (arrhythmogenic right ventricular cardiomyopathy, Brugada syndrome, long QT syndrome, short QT syndrome, idiopathic ventricular fibrillation, or other genetically determined arrhythmic diseases) where SCD may represent the major determinant of prognosis. Effective prevention by an ICD may, in these cases, confer a ‘normal’ or ‘near-normal’ life expectancy.6,8,1014 In these diseases, which usually affect young active subjects, there is a growing effort to develop appropriate strategies for identifying subjects at high risk of SCD, to be considered for prophylactic ICD implant. This trend is obviously motivated by the potentially devastating consequences of a malignant ventricular tachyarrhythmia in a young active subject, but it also represents a further important evolution in the role of ICD therapy, with relevant consequences on the methods for documenting ICD efficacy. Indeed, although SCD is the major determinant of prognosis in such conditions, the risk of ventricular tachyarrhythmias and SCD, in terms of events per year, is lower than that in the conventional population of candidates for prophylactic ICD, affected by left ventricular dysfunction (at 2 years 5–10% vs. 10–30%) and therefore the perspective for ICD benefit has a very distant horizon, not merely in the range of 2–5 years. In this setting, it has not been possible to plan randomized controlled trials for assessing the efficacy of ICDs, which has instead been inferred from observational studies or registries. The results of such observations (in part reported in Figure 1) have provided some evidence on the role of ICD in these substrates, with incorporation of specific indications for prophylactic ICD implant in the current guidelines.6,8

Barriers to full implementation of mplantable cardioverter defibrillators in primary prevention of sudden cardiac death

On the basis of the existing evidence from randomized controlled trials, the ICD has emerged in the last few years as a key tool for any strategy targeted at reducing the impact of SCD in high-risk patients. Verification of these benefits in real-world medical practice, outside the strictly controlled setting of randomized trials (i.e. analysis of effectiveness as opposed to efficacy), appears to be of crucial importance for health care providers and health policy-makers,15,16 and it has been suggested that multicentre registries should be set up to monitor and validate ICD use in routine clinical practice.15,17 Indeed, some subpopulations of recipients (in terms of age, gender, clinical profile and co-morbidities) have limited representation in clinical trials, and ICD use in these settings requires further evaluation.

No clinical trial, moreover, has investigated the ‘(un)natural history’ of patients with left ventricular dysfunction beyond 3–5 years after ICD implant and in this perspective re-evaluation, in long-term effectiveness studies, of the ratios between costs/risks and benefits/outcomes appears mandatory. A first effectiveness study performed on a cohort of 770 patients with ischaemic heart disease and left ventricular dysfunction showed that the mortality reduction associated with ICD use in primary prevention trials can be translated into routine practice without a substantial dilution of benefit.16 In this perspective, it is noteworthy that the institution of prospective registries on ICD use in primary prevention according to the results of MADIT II and SCD-HeFT15 has been promoted in the USA as a key component of Medicare financial coverage.

Substantial barriers presently exist with regard to full implementation of ICD therapy in clinical practice as a standard treatment for preventing SCD. At a first glance it appears that implementation of ICD use according to consensus guidelines is more problematic in comparison with other treatments. Indeed, a complex combination of cultural factors, related to wrong perception of the characteristics of ICD therapy,5 financial issues, and current priorities in allocating budgets for health care create the basis for both under-referral of potential candidates for ICD therapy in daily clinical practice as well as explicit or, more commonly, subtle, hidden rationing of such a therapeutic option.

Evaluation of implantable cardioverter defibrillator therapy according to available economic estimates

The traditional view of the ICD is that this treatment is expensive, which has been driven by the high up-front costs of the device itself and of the implant procedure, followed over time by maintenance costs for device replacement. Despite marked price reductions in the last decade, the cost issue continues to limit full acceptance and application of ICD therapy, especially in the perspective of primary prevention of sudden death.5 To overcome the limitations of an approach to the issue of ICD treatment implementation based simply on an analysis of costs, approaches aimed at measuring benefits against costs have been proposed as an aid to decision makers, policy-makers, and health care providers.

Cost-effectiveness and cost–benefit analyses have been performed in various fields of medicine to determine which particular treatment option is most likely to provide maximum health benefits for a given level of financial resources, or to assess which is the least expensive way of delivering a given level of health benefits.18 While cost-effectiveness estimates measure clinical outcomes in terms of ‘years of added life’ or ‘quality-adjusted life years gained’, cost–benefit analysis directly assigns a monetary value to therapeutic benefits.5,19,20 An approach based on cost-effectiveness analysis seems to be particularly suitable for approaching the problem of ICD treatment, which is an expensive form of treatment, with both high up-front costs due to the device itself and deferred evidence of clinical benefit. In cost-effectiveness analysis, the ‘cost’ of a therapy includes both the direct costs (initial cost of therapy, those of maintaining the therapy and those from any adverse effects) and the indirect costs paid by patients, their families, and/or the community. Effectiveness is measured as the mean extra number of years survived as a result of a treatment. Incremental cost-effectiveness analysis involves comparison of alternative therapeutic strategies. The cost-effectiveness ratio is commonly expressed as dollars per year of life saved ($/YLS). A treatment is usually considered very attractive if the cost-effectiveness ratio ranges between 0 and 20 000 $/YLS; attractive between 20 000 and 40 000 $/YLS; borderline between 40 000 and 60 000 $/YLS; unfavourable between 60 000 and 100 000 $/YLS; and absolutely unfavourable over 100 000 $/YLS.5,1821

A brief summary of cost-effectiveness estimates related to a series of commonly used medical interventions, as well as for ICD treatment, is shown in Table 2. Currently available cost-effectiveness estimates of ICD treatment, generated by observational data, projections based on decision models (retrospective analysis) and by randomized trials5,1921 have shown that the use of ICDs in selected patients, or in trial subgroups, at high-risk of sudden death has often produced cost-effectiveness estimates that are comparable with, or lower than, other accepted treatments (antihypertensive drugs, coronary angioplasty, coronary artery bypass graft, etc.) and even more favourable than a standard benchmark for economic evaluation such as renal haemodialysis. (Table 2).5,1921 However, in a comprehensive review of the available literature, it emerges that a broad range of cost-effectiveness ratios have been reported for the ICD, ranging from economically attractive to very unfavourable values,5,1921 which stresses the need for further evaluations in this field.

View this table:
Table 2

Cost-effectiveness of a series of medical or interventional treatments

TreatmentCost per year of life saved ($/YLS)
Valve replacement for aortic stenosis2200
CABG for left main stenosis9200
CABG for three-vessel disease18 500
Beta-blockers in low-risk post-infarction CAD20 200
ICD in secondary prevention with EPS25 700
Primary stent in PTCA26 800
ICD for MADIT I profile27 000
Clopidogrel for secondary prevention in patients with CAD ineligible for aspirin31 000a
ICD for SCD-HeFT profile38 389
ICD for MADIT II profile39 000
Anti-hypertensive tx (DAP 95–104 mmHg)41 900
ICD for MADIT II profile50 500
ICD for SCD-HeFT profile50 700
Hospital haemodialysis59 500
CABG in two-vessel disease72 900
Lovastatin in primary prevention, chol ≥300 mg/dL, no RF, male, age 55–64 years78 300
ICD for MADIT II profile78 600
PTCA in one-vessel disease109 000
Clopidogrel for secondary prevention in all patients with CAD130 000a
Lovastatin in primary prevention, chol ≥ 300 mg/dL, no RF, female, age 45–54 years2 024 800
  • Data on cost-effectiveness were derived from the literature.5,19–21 As shown, different estimates have been reported in the literature for ICD implantation according to MADIT II or SCD-HeFT profile. Estimates are reported in $ per year of life saved (in $ per quality-adjusted year of life gained, when reported).

  • CABG, coronary artery by-pass graft; CAD, coronary artery disease; chol, cholesterol; DAP, diastolic arterial pressure; EPS, electrophysiological study; PTCA, percutaneous transluminal coronary angioplasty; RF, risk factor; Tx, therapy; CRT, cardiac resynchronization therapy.

  • aDollars per quality-adjusted year of life gained.

An important source of variability is the duration of therapy within which cost-effectiveness is estimated. When Hlatky and Bigger22 projected the results of all the trials published up to 2001 to gauge the full gain in life expectancy, they obtained a cost-effectiveness ratio of 31 500 $/YLS, in line with what is currently considered fully acceptable in developed countries. Indeed, the lack of data on long-term benefits constitutes an important limitation of currently available ICD cost-effectiveness estimates. Since prospective trials that validated the ICD treatment were systematically stopped as soon as demonstration of improved survival was reached, the follow-up of the patients enrolled has tended to be far shorter than the life expectancy of many patients who receive an ICD treatment in clinical practice. This bias is highly relevant since the high initial cost of the device can markedly affect cost-effectiveness estimates, particularly when the follow-up is not long enough to assess the full benefits of ICD treatment.5,22 Prospective evaluations specifically designed for long-term assessment of cost-effectiveness should be planned in order to provide valuable data to health care providers.

The cost-effectiveness analysis of SCD-HeFT trial data generated a value of 38 389 $/YLS,23 indicating that implementation of the study’s patient selection strategy for prophylactic ICD implant may be economically justifiable. SCD-HeFT explored the possibility of implanting a single-chamber ICD on an outpatient basis, and this option surely contributed to the favourable cost-effectiveness value. Further evaluations are required to identify the specific subset of patients in whom this approach is safe and appropriate in current everyday clinical practice.

In terms of investment for health care systems, ICDs must be considered a long-term investment, involving a high initial cost. It is quite clear that our health care systems, often dealing with fixed budgets, are particularly resistant to approve such investments, whereas they may be much more tolerant of approving lower initial investments, which have accumulating long-term costs, as in the case of chronic pharmacological treatments. It is noteworthy that the cost-effectiveness profile of some pharmacological treatments may be unattractive, as in the case of long-term administration of lipid lowering agents in patients with a relatively low risk profile, or the use of antihypertensive agents in mild hypertension, or unselected use of antithrombotic treatment with clopidogrel in coronary artery disease.5,19

Cost-effectiveness allows expenditure to be assessed in terms of eventual clinical benefit. A further approach to the issue of ICD cost is to estimate the cost per day of the treatment, over a predefined period. According to this approach, we recently calculated that >5–7 years, the average daily costs associated with ICD treatment were estimated to be around $7.90–11.40 (around $6.70–9.60 for single-chamber devices).17 The calculation of cost per day of ICD treatment may allow comparisons with other treatments currently adopted in clinical practice and therefore may be an additional interesting approach to be applied in order to evaluate the affordability of ICD treatment beyond its high up-front cost.

Analyses of ICDs in terms of cost–benefit have been largely underused. However, in a study recently reported by Caro et al.24 and based on SCD-HeFT data, cost–benefit was estimated comparing prophylactic ICD treatment with the most prominent antiarrhythmic drug, amiodarone. The conclusion is that in countries where society values a life at more than €2 million, an ICD is a worthwhile investment compared with amiodarone for primary prevention.24 It is noteworthy that this cost–benefit evaluation completely changes the perspective of ICD use in high-risk patients, supporting the view that an initially expensive ICD rather than a long-term cost can be a worthwhile investment for society, as well as for the patient.25

In recent years, health care systems have had to face a growing financial burden related to the costs of evolving technology and it is surprising that economic analysis based on cost-effectiveness estimates, as well as health technology assessment analysis, has yet to become the fundamental criteria for deciding whether a new treatment should have financial coverage by public services. As a matter of fact it appears that the decisions of both policy-makers and health care providers are not mainly based on in-depth economic analysis, but are still largely influenced by mere financial projections, with a consequent tendency towards rationing or even rejection of costly innovative treatments, despite evidence of clinical efficacy.18

Implantable cardioverter defibrillator in primary prevention of sudden cardiac death: patient selection and implications for cost-effectiveness

Selection of candidates for prophylactic ICD is nowadays quite straightforward, and application in clinical practice of the criteria used in the MADIT II and SCD-HeFT trials is expected to induce an impressive rise in the potential number of implants among patients with left ventricular dysfunction or heart failure.5 Table 1 reports a summary of current indications to ICD implant for primary prevention of SCD in patients with heart failure and/or left ventricular dysfunction. As shown, the current selection criteria in accordance with the above-mentioned trials are mainly based on the assessment of left ventricular ejection fraction and this easily applicable approach may result in an important number of patients potentially eligible for a prophylactic ICD. As a consequence, there is now growing fear that increased use of ICDs, according to current indications and current guidelines, may cause a dramatic burden on health care systems.5 Of note, the most crucial issues regarding clinical use of ICDs are not related to doubts about the overall efficacy of this treatment, but to considerations about risks, benefits, and costs among different subgroups of patients with heart disease who are potential candidates for such treatment. The answer to such crucial questions may be obtained by post hoc analysis of available trials, pooling data in order to obtain larger populations and, above all, by analysis of registries, which directly represent ‘real world’ daily practice. The latter option is really interesting since it is well known how patients enrolled in clinical trials may differ from those in real world practice with regard to age, co-morbidity, concurrent treatments, and overall accuracy of care. In this heterogeneous scenario, the challenge is now to asses to what extent the efficacy demonstrated in clinical trials translates into effectiveness and affordability, in specific subgroups of patients for whom limited or no data are available (the elderly, patients with co-morbidity, patients with recent revascularization, patients with valvular heart disease, etc).

As for any treatment with a substantial cost, one potential approach to improve the cost-effectiveness profile of ICDs is based on improving patient targeting, i.e. selecting those patients who have the highest probability of gaining substantial benefit. At present, there are no well-defined tools for identifying in advance the patients most likely to die from SCD rather than from non-sudden cardiac death. An improvement in patient risk stratification may be important for identifying subgroups of patients for whom the option of ICD implant appears more favourable or attractive and might be helpful in attempts to maximize health outcomes in a context of limited economic resources.26 Such an approach should consider a detailed analysis of the patient’s profile and of his/her co-morbidity, as well as the use of further investigations or testing to be combined with the assessment of left ventricular function for defining ICD eligibility.

An improvement in patient selection for ICD therapy could imply pairing the assessment of left ventricular ejection fraction with another marker of high risk for sudden death. However, it should be considered that attempts to improve patient selection may imply the risk of excluding some patients at potential risk of SCD who could benefit, or even be saved, if implanted with an ICD. Indeed, any attempt to improve risk stratification will probably be imperfect, since an increase in specificity will probably lead to some decrease in sensitivity. In view of the strong evidence supporting ICD therapy, prospective studies are required to validate the acceptability of excluding potential candidates for ICD therapy on the basis of specific risk stratification algorithms.8

In the perspective of improving risk stratification for SCD and of identifying those patients who may obtain the greatest benefit from ICD implant, a series of tests has been proposed as potential risk stratifiers to be used in combination with left ventricular ejection fraction: heart rate variability, signal-averaged ECG, baroreflex sensitivity, evaluation of ventricular tachyarrhythmia inducibility at electrophysiological testing, or T-wave alternans testing. In MADIT II, a pre-planned substudy assessing the role of electrophysiological study as a risk stratification tool failed to demonstrate a reliable significance of inducibility of ventricular tachyarrhythmias in predicting appropriate ICD discharge in follow-up.27

Promising results have recently been obtained with microvolt T-wave alternans testing, with the potential of improving the cost-effectiveness profile of ICD in primary prevention of SCD. T-wave alternans testing is a non-invasive method of measuring beat-to-beat microvolt variations in the shape, amplitude, or timing of the T-wave that are linked with the development of clinical ventricular arrhythmias. The test has a high negative predictive value, i.e. a negative test result identifies a patient at very low risk of SCD. Chan et al.28 used a mathematical model to compare the lifetime costs and benefits of a therapeutic strategy to prevent death by implanting ICDs in all currently eligible MADIT II-like patients, or implanting defibrillators in only those patients considered to be at higher risk according to T-wave alternans testing (67% of the whole population, estimated to gain 83% of the total potential ICD benefit). According to this estimate, a strategy of making ICDs available to all eligible patients according to MADIT II criteria would be less cost-effective compared with a more discriminate strategy of implanting ICDs only in patients selected on the basis of T-wave alternans testing. The cost-effectiveness estimate of implanting an ICD in the group judged to be at lower risk according to microvolt T-wave alternans testing was close to US$ 90 000/QALY, while implanting an ICD in MADIT II-like patients selected on the basis of non-negative T-wave alternans testing would result in a much more favourable cost-effectiveness profile (US$ <50 000/QALY).28 However, the precise value of T-wave alternans testing in selecting/excluding potential candidates to implant an ICD in ‘real world’ clinical practice still need to be assessed.

Apart from risk stratification on the basis of specific testing, there is plenty of room for enhancing clinical risk stratification and evaluation for better targeting of candidacy for ICD therapy. Assessment of co-morbidities (systemic illness, major organ compromise, etc.), which may preclude obtaining a real benefit from ICD therapy or may enhance the risk of complications (infections, etc.), could aid more precise targeting of ICD therapy. In particular, identification of renal insufficiency proved to be a strong determinant of the risk of non-sudden death, thus limiting the potential for a benefit of ICD therapy.29 Another important issue of clinical judgement of potential candidates for ICD therapy is evaluation of concurrent medical therapy. Optimization of medical therapy, and particularly of beta-blockers and ACE-inhibitors, is a crucial step in targeting the candidates for ICD. It is known that beta-blockers and ACE-inhibitors may improve NYHA class and left ventricular ejection fraction and that both titration of the drug and full development of drug effects may require some months. Waiting for some months for the full effect of concurrent pharmacological treatment is a reasonable choice since all the trials supporting ICD treatment were performed on patients on ‘optimized medical treatment’.

Moreover, it has to be considered that the benefit of ICD in primary prevention of SCD, according to MADIT II and SCD-HeFT, emerges only in a mid- to long-term follow-up (at 2–5 years), thus patients with an expected survival of at least >1 year, and preferentially much longer than 2–3 years, will have to be selected as candidates for ICD in primary prevention of SCD.

Implantable cardioverter defibrillators in clinical practice: limitations and open issues

Complications may occur following ICD implantation, including infection, lead dislodgment, lead fracture, or frequent shocks. The occurrence of frequent shocks may be related to appropriate interventions for storms due to ventricular tachyarrhythmias (so called ‘arrhythmic storms’) or, much more commonly, may be related to inappropriate interventions for supraventricular tachyarrhythmias and, especially, atrial fibrillation with fast ventricular response.30 Since experiencing shocks delivered by the device may be particularly distressing and influence the patient’s psychological reaction to the device, there is now a strong commitment to adopt specific device programming and/or discrimination algorithms to minimize the occurrence of inappropriate shocks,31,32 as well as to increase the use of antitachycardia pacing as a painless treatment for interrupting ventricular tachyarrhythmias.33 Some inappropriate shocks may be related to lead failure and this reflects the need to improve the reliability of the leads, which is a critical determinant of long-term safety of ICD systems.

An important limitation of current ICDs is device longevity. Indeed, extending the indications to clinical conditions with a long expected survival once SCD is prevented, as in genetic arrhythmic diseases, implies that several device replacements will occur during the life of a patient.1014,34 Extension of device longevity will, therefore, represent advancement with important benefits for ICD recipients.34

In summary, despite the technological improvements of current devices, the large series of randomized controlled trials performed in the last few years, and the availability of consensus guidelines helping clinicians in decision-making about the indication for ICD therapy, a series of issues concerning ICD use remain open and will must be addressed over the next few years (Table 3).

View this table:
Table 3

Open issues regarding ICD therapy

Barriers to implementation of ICD in daily practice
Effectiveness vs. efficacy
Time course of ICD-associated benefit
Long-term benefits of ICD therapy
Residual risk of SCD in patients implanted with an ICD
Cost-effectiveness of ICDs
Risk stratification for improving patient targeting
Implant setting (in-patient vs. out-patient)
Implant technique in children
Need for VF induction at ICD implant
ICD programming
Inappropriate interventions
Arrhythmic storms
Device longevity
Methods for ICD follow-up and role of telemedicine
Infections and skin erosions
Long-term reliability of devices and leads
  • ICD, implantable cardioverter-defibrillator; SCD, sudden cardiac death; VF, ventricular fibrillation.


The ICD has represented a revolutionary step in the fight against SCD. Its clinical use is related to appropriate risk stratification and has to be integrated in a comprehensive approach including a wide range of preventive measures (pharmacological treatment of underlying disease, revascularization). The technological evolution of ICDs is still in progress, with the aim of broadening the spectrum of therapies these devices may deliver, as well as of implementing new diagnostic functions and improving patient tolerability. A series of landmark randomized trials has validated the role of ICDs not only in the setting of secondary prevention of SCD but also in the challenging subset of primary prevention, i.e. for patients judged to be at high risk of SCD but without previous malignant ventricular tachyarrhythmias. Despite definite indications provided by consensus guidelines, the use of ICDs in clinical practice still encounters a series of barriers related to the characteristics of such treatment (a ‘rescue’ treatment) and its up-front cost, resulting in substantial under-referral and rationing. Cost is likely to remain a major determinant of full acceptance and implementation of ICD therapy and the problem of how broadened evidence-based indications for implantation can be translated into routine clinical practice require analysis of available economic resources and identification of priorities for health care. Economic analysis (cost-effectiveness, cost-utility, and cost-benefit estimates) provides the most appropriate tool for weighing ICD costs against likely eventual outcome benefits. A series of data indicate that the use of ICDs in appropriately selected patients at high risk of SCD is associated with cost-effectiveness ratios similar to, or better than, other accepted treatments, such as renal dialysis. Improvement in risk stratification for SCD and assessment of ICD cost-effectiveness in specific subgroups of patients appear mandatory for any attempt at maximizing health outcomes in a context of limited economic resources.

Within this complex scenario, offering appropriate care to individual patients even in an era of economic constraints may be a difficult task and the individual cardiologist responsible for decisions regarding the well-being of individual patients may often be in trouble in daily clinical practice, influenced both by limited economic funding and the need to offer the best for each individual patient.

Conflict of interest: none declared.


The work was partly supported by a grant from ‘Fondazione Luisa Fanti Melloni’, Bologna, Italy.


View Abstract