BMJ 2007;335:895-896 (3 November), doi:10.1136/bmj.39353.546933.BE (published 11 October 2007)
Whether combined cardiac resynchronisation and implantable cardioverter defibrillator therapy offers benefit above either device alone remains uncertain
Over the past decade, two devices have emerged as adjuncts to optimal medical treatment for people with heart failure—cardiac resynchronisation therapy, which is delivered by atrial synchronised biventricular pacemakers, and implantable cardioverter defibrillators. The evidence base underpinning these treatments has evolved rapidly. In this week's bmj, Lam and Owen have summarised mortality data from 12 randomised trials using a Bayesian network meta-analysis.1
Their analysis confirms earlier meta-analyses showing that both devices independently improve survival in patients with heart failure (these earlier analyses also showed that resynchronisation therapy, but not implantable cardioverter defibrillators, improved functional status and reduced hospital admissions). More importantly, Lam and Owen found insufficient evidence to prove that combined devices were better than either device alone even using network meta-analysis incorporating all major trials. This analysis has four important implications for clinicians.
Firstly, while these devices work in selected patients (table), their role in many patients with heart failure remains uncertain. Only 1-3% of people with heart failure meet the eligibility criteria for the trials that proved the efficacy of these devices.2 The criteria were New York Heart Association (NYHA) class III or IV symptoms despite optimal medical management, left ventricular ejection fraction 35%, electrical dyssynchrony, and sinus rhythm for trials of cardiac resynchronisation, whereas criteria were NYHA class II or III symptoms and left ventricular ejection fraction 35% for trials of implantable cardioverter defibrillators.
In addition, not all people eligible for trials will benefit. For example, 41% of trial participants had no improvement in functional status after successful implantation of their cardiac resynchronisation device,3 and two thirds of patients with implantable cardioverter defibrillators never received therapeutic defibrillation discharges.4 Only 10% of patients who receive an implantable cardioverter defibrillator for primary prevention have been estimated to receive an appropriate shock and survive at least one year.5 Thus, we urgently need to develop and validate cheap and easily applied tools that can accurately identify those people most likely to benefit from either (or both) of these devices.
Secondly, although the benefits of both devices seemed to be relatively stable over time in the randomised trials (survival curves of patients with and without the devices did not converge as follow-up lengthened), these trials were of relatively short duration. Thus, long term follow-up data are essential to confirm that the benefits do not attenuate over time, particularly in light of the frequency with which Food and Drug Administration advisories and device recalls arise years after efficacy studies are completed.6 Everyone who has one of these devices implanted should therefore be entered into a registry and followed for long term risks and benefits. This would have the added benefit of tracking changes in the success rates and complications of implantation procedures over time as the sophistication of devices and technical expertise of those implanting them evolves.
Thirdly, data are lacking on the safety and effectiveness of either device when applied outside the confines of a clinical trial. These trials could possibly have overestimated the potential benefit-safety ratio because half of the people randomised were drawn from trials in which they were randomised after successful implantation of the device (thus, the 7% implantation failure seen for cardiac resynchronisation devices and 0.5-1% peri-procedural mortality seen with both devices are not factored into the effect estimates presented by Lam and Owen). Moreover, while the trials proving the efficacy of both devices enrolled relatively young patients, these devices are now being implanted into older people with more comorbidities.7 They are also being implanted by less experienced providers working in hospitals with lower implant volumes.8 Moreover, people with heart failure with less severe symptoms (NYHA class I or II) or common comorbidities (such as bradyarrhythmias, atrial fibrillation, or chronic kidney disease) were largely excluded from the published trials. Again, this emphasises the potential value of a prospective registry in defining the benefits and safety of these devices in a wider spectrum of patients.
Finally, the finding by Lam and Owen of uncertain incremental benefits for the combined device over single devices is important. Is it plausible that a combined device would not be better than either device alone? The answer is yes. Increases in the use of angiotensin converting enzyme inhibitors and blockers over the past decade mean that people with heart failure are now more likely to die of progressive heart failure than sudden death.9 Also, cardiac resynchronisation alone seems to reduce the frequency of ventricular arrhythmias10 and—at least in the trial with the longest follow-up—the incidence of sudden death.11
While several trials are ongoing to define the incremental benefits of combined devices over implantable cardioverter defibrillators alone in patients with heart failure, I agree with Lam and Owen that further randomised trials are needed to define the incremental benefit of combined devices over cardiac resynchronisation alone. As both devices have been shown to prevent morbidity and mortality in similar patient populations (table), some may argue that such trials are not needed and would advocate the use of combined devices in all patients except those with severe comorbidities. However, consideration of the costs of the two options (£5074 (7249; $10 253) for cardiac resynchronisation therapy versus £17 266 for a combined device)12 brings the matter clearly into focus—in publicly funded healthcare systems with finite budgets, should we reallocate resources from other areas to pay for expensive technologies that offer uncertain incremental benefits over cheaper options?12 Until a trial definitively proves that a combined device is better than cardiac resynchronisation alone, such devices should be reserved for people with heart failure who are eligible for cardiac resynchronisation and have the highest risk of sudden cardiac death.
Finlay A McAlister, associate professor
Division of General Internal Medicine, Department of Medicine, University of Alberta, Edmonton, AB, Canada T6G 2R7
Competing interests: None declared.
Provenance and peer review: Commissioned; not externally peer reviewed.