Prof Kenneth Dickstein
Professor of Medicine, University of Bergen, Stavanger University Hospital, Norway
Taskforce co-chairman, 2010 Focused Update of ESC Guidelines on device therapy in heart failure
Although most countries have their own guidelines for important, clinically relevant topics, those issued by the European Society of Cardiology (ESC) have an influence that spreads well beyond Europe, to some 52 countries, or 1.3 billion people. This does not include official affiliations with countries such as Brazil and Argentina and, more recently, with China. In large measure, this is because the ESC’s Committee for Practice Guidelines (CPG) has a reputation for carrying out extremely high-quality, rigorously conducted reviews.
Certainly, this is true of the new European guidelines updating the use of device therapy in heart failure.1 But more than this, they also represent a new approach aimed at ensuring that the guidance relates to the patients doctors actually need to treat, rather than being based on outcomes in patients defined by the strict entry criteria of relevant clinical trials. There are often substantial differences between the two.
Trying to ensure that new trial evidence can be realistically applied in daily clinical practice was very much our starting point in taking a fresh look at the 2008 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure, which I chaired, and the 2007 ESC guidelines for cardiac and resynchronisation therapy (CRT), chaired by Panos Vardas, professor of cardiology at the University of Heraklion in Crete.
The evidence from two recent trials, REVERSE2 and MADIT-CRT3 proved particularly helpful in this regard. Both were randomised, prospective, multicentre trials that demonstrated reduced morbidity following device insertion in mild heart failure (HF), a condition in which the heart cannot adequately support the requirements of the circulation.
Revised guidelines are due in 2012, but we have produced this focused update – in itself a new initiative for an ESC guideline – not only because of the recent research evidence, but also because of new developments in device technology and performance and a more extensive understanding of treatment options and responses. We felt it would not be right to keep people waiting a further two years to get the benefit.
The guidelines set another precedent, too, in that they represent the first-ever joint effort between the ESC’s Heart Failure Association (HFA) and the European Heart Rhythm Association (HRA), a collaboration which recognises that devices may be implanted by both cardiologists and cardiac arrhythmia specialists. In acknowledging that some interventions cross two disciplines, we hope that this will stimulate other organisations to do the same.
ESC guidelines are developed by a taskforce chosen for its expertise by the relevant ESC working groups and associations. In the case of the device guidelines, the taskforce was made up of five HFA and five HRA members. Professor Vardas and I were the co-chairs.
The guideline process takes some 18 months to two years to complete. The first draft from the taskforce goes to a review committee for extensive criticism and comment and, after several revisions, once an agreed revised version has been put together by the review co-ordinator, this goes back to the taskforce to see if it is happy with any proposed amendments. The agreed version is then sent to the CPG for final review.
While I personally would like to see doctors reaching consensus opinion, the point about guidelines is that they are just that – expert guidance, not hard and fast rules. In Europe, while guidelines are taken very seriously by administrators as well as clinicians, there is flexibility in the interpretation and the clinical decision-making process.
At the CPG we stick with evidence-based medicine. We’re not the European Medicines Agency. Guidelines are, by nature, conservative and we don’t permit any citations that aren’t published in peer review journals. Doctors have to evaluate for themselves whether the approaches discussed are appropriate for their units. All the same, if clinicians deviate, they should first know what the guidelines are.
Cardiac resynchronisation therapy
The updated guidelines put the emphasis on the use of cardiac resynchronisation therapy (CRT) in heart failure and include the indications in conditions such as atrial fibrillation and situations in which patients require pacemakers, two of five areas not covered in the earlier guidelines.
For example, CRT is helpful in reducing electrical dyssynchrony, caused by abnormal intraventricular conduction within the heart’s system of electrical impulses. Instead of the ventricles of the heart contracting simultaneously, one side does so, followed by the other after a delay; this is known as a widened QRS interval and is measured in milliseconds (ms). Either a CRT pacemaker (CRT-P) or a CRT-implantable cardioverter defibrillator (CRT-D) device may be used to shorten the delay.
To determine the functional level of patients with heart failure, we referred to the New York Heart Association (NYHA) classification system for heart failure, which relates the symptoms patients are experiencing to the everyday activities that cause their symptoms. Classes I and II are considered mild, class III moderate and class IV severe.
Reassuringly, all the relevant clinical trials we examined confirmed that CRT brings about a sustained, significant alleviation of symptoms and an increase in exercise capacity. On average, the patient’s NYHA function class decreased by 0.5%–0.8 points, the six-minute walk test increased by 20%, and peak oxygen consumption increased by 10%–15% after treatment.
We give a class IA recommendation – our highest – to two situations, because the evidence to back them is so strong. We recommend extending the use of CRT from the current NYHA class III to IV patients, where the intention is to reduce morbidity and mortality, to include those with mild NYHA class II heart failure, with the aim of reducing morbidity and, importantly, curtailing disease progression.
This decision derives in particular from findings related to the NYHA class I and class II patients in the MADIT-CRT trial. Class I patients, who generally are symptom-free on medication, accounted for just 15% of the trial population; subgroup analysis showed no significant benefit from CRT.
In addition, although the trial’s entry criteria included a QRS interval of 130ms or more, the CRT benefit was concentrated in patients with a QRS of 150ms or more. Importantly, the 150ms cut-off was pre-specified subgroup in the trial protocol, therefore the decision to restrict the recommendation to patients with a QRS width of at least 150ms is in line with statistical principles. In the interests of defining the patient groups most likely to benefit from CRT, we recommend including those in NYHA class II who are in sinus rhythm and on optimal medication, with a left ventricular ejection fraction (LVEF) of 35% or less, and a QRS interval of 150ms or more.
Reducing morbidity and mortality
Our second recommendation of class IA standing concerns the use of devices to reduce morbidity and mortality in NYHA classes III and IV. Patients must be in sinus rhythm and on optimal medication, with an LVEF of 35% or less, and a QRS interval of 120ms or more. Class IV patients must be ambulatory. Patients being considered for CRT-D therapy must be expected to survive with good functional status for at least a year.
We drew the support for this decision from the findings of the COMPANION4 and CARE-HF5 studies, which looked at the combined primary endpoints of morbidity and mortality. In the COMPANION trial, the use of either CRT-D or CRT-P among the 217 NHYA class IV patients included showed two-year mortality rates of 55% and 45% respectively, compared with 62% in the control group, alongside significant functional improvement.
Patients who have permanent atrial fibrillation (AF), account for about a fifth of patients receiving CRTs in Europe, and the prevalence of AF in heart failure rises with the severity of the disease, from about 5% in NYHA class I to some 25%–50% in classes III and IV.6 Patients in these classes are typically older with a higher prevalence of comorbidity and a worse prognosis than patients in sinus rhythm, who have mainly been the subjects in the randomised clinical trials of CRT carried out to date.
For patients with permanent AF who are either pacemaker-dependent with bradycardia, an abnormally slow heartbeat, or following atrial-ventricular ablation, carried out to alleviate rapid and irregular heart pump activity, and who are in NYHA classes III and IV with an LVEF of 35% or less and a QRS of 130ms or higher, and are on optimal medication, we make a class IIa B recommendation to consider CRT-P or CRT-D insertion, in order to reduce morbidity. Again, there should be a reasonable expectation of survival with good functional status for at least a year in patients given CRT-D.
Ineligible for transplant
We also make a class IIb B recommendation for patients in NYHA classes IIIB and IV who have severe heart failure but are ineligible for transplant, and who have an LVEF of 25% or less. The IIb B classification means that providing the funding is available, it is worth considering the implantation of a left ventricular assist device (LVAD), which works on external batteries that need to be recharged twice a day.
This can be used as a bridge while awaiting a transplant, or as destination therapy. In fact, around 10% of patients in clinical practice receive an LVAD as destination therapy, according to data from the INTERMACS registry, which is supported by the US National Institutes of Health. Encouragingly, in the light of the shortage of organs available for transplantation, we are increasingly finding that patients are not proceeding to transplant, as planned, but are learning to live surprisingly well with the device; this situation is called destination therapy.
The evidence to date suggests that continuous-flow LVAD devices as destination therapy, such as Thoratec’s HeartMate II, are superior to pulsatile versions in patients with end-stage heart failure who are ineligible for transplant. In one recent randomised comparison study,7 the endpoints measured were freedom at two years from disabling stroke or further surgery to repair or replace the device (46% versus 11%), and survival at two years (58% versus 24%).
It is important to note that this is an area where patient selection is crucial. Most patients are on continuous inotropic support and candidates should not have significant renal, pulmonary or hepatic dysfunction. Nor should they have active infection, nor be in cardiogenic shock.
The guidelines’ implications
What are the implications of the guidelines? We believe that the finding that CRT sustainably improves the quality of life by alleviating symptoms and increasing exercise capacity is most encouraging news for patients and their families.
For doctors, the guidelines give a clinically relevant indication of the mortality, morbidity, symptoms and functional capacity to be expected within the different patient classes described, enabling them to target the population likely to respond to CRT.
Although the financial implications are not really discussed in the guidelines, they are substantial. CRT-Ds in particular are very expensive, and most hospitals cap their purchase annually, according to available resources. But although the devices are costly, they do last years and play an important role in avoiding expensive hospitalisation – my own unit costs £800 a day. In fact, overall, hospitalisation accounts for 55% of all money spent on the heart and, given today’s economic climate, it may well be appropriate to include financial considerations in future guidelines.
As for ourselves, in the effort to be clinically relevant, we try to include a ‘gaps chapter’ in a guideline. This helps to direct attention towards potentially fruitful lines of research. Or, as the late, great pioneer of research into heart failure – Prof Philip Poole-Wilson – put it: “I like the gaps chapter. It is an exercise in humility.”
- Dickstein K, Vardas P et al. Eur Heart J 2010;31:2677-2687. Accessed via
- Linde C et al. J Am Coll Cardiol 2008;52: 1834-1843.
- Moss AJ et al. N Engl J Med 2009;361: 1329-1338.
- Bristow MR et al. N Engl J Med 2004;350: 2140-2150.
- Cleland JG et al. N Engl J Med 2005;352: 1539-1549.
- Dickstein K et al. Eur Heart J 2009;30(20):2450-2460. First published online August 31, 2009.
- Slaughter MS et al. N Engl J Med 2009;361:2241-2251.