Statistics from Altmetric.com
- ACE, angiotensin converting enzyme
- ARB, angiotensin II receptor blocker
- ATLAS, assessment of treatment with lisinopril and survival
- BNP, B-type natriuretic peptide
- CPAP, continuous positive airway pressure
- CONSENSUS, cooperative North Scandinavian enalapril survival study
- EPHESUS, eplerenone neurohormonal efficacy and survival study
- GMS, General Medical Services
- LVDD, left ventricular diastolic dysfunction
- LVSD, left ventricular systolic dysfunction
- MI, myocardial infarction
- MINAP, Myocardial Infarction National Audit Project
- NICE, National Institute for Clinical Excellence
- NYHA, New York Heart Association
- RALES, randomized aldactone evaluation study
- TRACE, trandolapril cardiac evaluation
Question: Regarding the timing of assessment of left ventricular function after myocardial infarction, many of the speakers have talked about the need to pick this up early and some have talked about the use of diuretics being a strong indicator for poor prognosis. What would be the best approach to identifying this high risk group of patients after myocardial infarction and when should we be doing this?
Professor Karl Swedberg: As we do not have a good surrogate for measuring left ventricular function in this context, we have to look at left ventricular function in all patients after MI using imaging such as echocardiography. We know that a high BNP level is associated with increased risk, however at present we are uncertain how to handle that in the context of treatment decisions. Hopefully, the routine measurement of BNP might be a tool for guiding care.
Professor Martin Cowie: But there is a problem in that BNP goes up in big infarcts and then you might not know quite what you are dealing with.
Dr Theresa McDonagh: I think obviously measuring left ventricular function as early as possible is good but the problem can be that some patients do improve—they have heart failure temporarily but myocardial stunning can be the initial problem, leading to decreased left ventricular function which may then recover. I would have thought the minimum standard is that every patient should have left ventricular function assessed at six weeks because that gets over that particular problem. Ideally, they should have an echocardiogram in hospital, but if this is not possible you have to screen at six weeks.
Professor John Cleland: The implication of that is that people need to be assessed twice: they need to be assessed acutely within three days, and then several weeks later to pick up the people with chronic ventricular dysfunction.
Dr Ahmet Fuat: I would like to make a comment about BNP. In our current study in Darlington, we had 600 patients, with 47 who had previously had an MI (many of them many years ago). All 47 had raised BNP ranging from 180 pg/ml to 13 000 pg/ml, and only 19 of those had LVSD on echocardiography. So I am not sure about the use of BNP, particularly in the acute setting.
Question: There is quite a push towards incorporating patients with heart failure into cardiac rehabilitation programmes, but there is very little research or literature available for either prescribing or, more importantly, delivering exercise to these patients.
Professor Swedberg: The European Society of Cardiology advises that patients receive structured care. It is extremely important that the care is organised, and that has been shown in several randomised trials, so there is an evidence based rationale for doing it. However, when providing this rehabilitation package I encourage you to look deeper into what is important and what is not. The content of the care, and what advice to give, is not “evidence based” in the strict sense of the words. To address the point of the question, we don’t know the exact components of cardiac rehabilitation that may benefit patients with heart failure. Those who don’t participate in such programmes tend to do worse than those who do, but it is difficult to take all the possible reasons for this into account, such as the extent of co-morbidities which limit the ability to exercise.
Professor Cleland: Along the same theme, the GMS contract includes a target for influenza immunisation. Here is a wonderful example of poor evidence based medicine—there is very little evidence that vaccinating patients makes any difference in terms of outcome for hospitalisation, mortality or anything else, and yet the government spends huge amounts of money on a potentially quite wasteful exercise and puts it in all sorts of guidelines as well.
Dr McDonagh: The questioner raises a good point. After an MI, UK patients are usually enrolled in a cardiac rehabilitation programme, which includes exercise. That is very different from the patient with chronic heart failure, for whom some studies suggest that exercise may be beneficial and others that it is not. The position is unclear. But you have raised an interface issue. In some areas, where the resources to deliver specialist heart failure nurse care for patients with LVSD and heart failure post-MI may not be available, then surely this is something that the rehabilitation team should take on rather than have no one do it at all. I would suggest looking at the evidence base: if there is a specialist heart failure nurse multidisciplinary management programme then perhaps patients with heart failure or LVSD would be best in that programme once their rehabilitation programme is complete. But I understand the question and it is a dilemma. We don’t want too many chronic disease management programmes otherwise many patients would not only have to cope with polypharmacy but also six or seven different disease management programmes as well!
Professor Cowie: It is true that cardiac rehabilitation uptake is lowest among those at highest risk. It is usually the young, middle class, white males who come to rehabilitation but the patients who you actually want to reach, the ones who are at highest risk, are the ones least likely to come. It is a complicated issue.
Dr Fuat: Later this year, we hope to start a randomised study comparing rehabilitation with “usual” care for patients with heart failure in Darlington. This will provide data highly relevant to today’s discussion.
Professor Cowie: In the new NICE guideline on post-MI care, heart failure has been specifically excluded from the suggested scope because it was said that the topic had been covered by the chronic heart failure guideline published last year. Perhaps I could ask Professor Dargie, as chairman of the British Society of Heart Failure, whether this is a reasonable position or whether the professional bodies should be pushing for post-MI heart failure to be in the post-MI guideline?
Professor Henry Dargie: It really depends on the point at which the patient is diagnosed as having chronic heart failure. Dr McDonagh and others have made the point that patients can have heart failure transiently in the coronary care unit but that their left ventricular function actually may show a surprising amount of recovery by the time they get to a post-MI clinic, which may be related to the fact that it is stunning or other factors that have improved. But those patients who have important left ventricular dysfunction who remain symptomatic at the post-MI clinic, de facto, have chronic heart failure and clearly then should follow the appropriate guideline. I think that part of the problem is that we have two “streams”. Dr Dancy recommended that we should follow the post-MI chest pain guideline in the first instance because that is the one that is understood. But the interface point becomes the clinic at six weeks and it would seem to me that at that point there should be referral to the heart failure service. Acute MI is specifically excluded from our own heart failure liaison nurse service. That was because of logistic reasons in the first instance but I think that with the success of that programme undoubtedly the acute MI patients who then end up having chronic heart failure should go there for sure.
Professor Cowie: Isn’t there a danger that if a lot of benefit is in the first few weeks after infarction, when the ventricle is remodelling, then if you wait for them to get into the territory of chronic heart failure at a six week clinic then you are robbing those patients of the potential benefit of treatments such as eplerenone? Some of the patients won’t come back or will have died in that period when perhaps you could have saved them.
Professor Dargie: It is important to remember that the coronary care unit is a place where the patient is very much under “probation” and can benefit from all that goes on there. Initiation of effective medication should start there undoubtedly. Of course, with an effective post-MI programme involving specialist nurses, who are actually called rehabilitation nurses in that context, that process can continue, but the medication really ought to be started and titrated as much as is possible in the coronary care unit. But, as has been pointed out, the time in hospital is becoming ever shorter. Starting ACE inhibitors, β blockers, maybe aldosterone antagonists and so on, all at the same time does pose quite a challenge for most patients in terms of their blood pressure. So there does need to be something in between, as you have pointed out, to make sure the titration does continue and the patient does not lose out.
DIASTOLIC HEART FAILURE
Question: We have largely discussed LVSD today and I am conscious it is difficult to manage heart failure when there is a normal ejection fraction or left ventricular diastolic dysfunction. What is the place of LVDD?
Professor Dargie: That is a very interesting question and we had a good debate about that at the European Society of Cardiology congress about exactly how you diagnose it. I think it is very difficult. I don’t think that at the moment there are accepted criteria for diastolic dysfunction, and one can only rely on clinical factors plus the fact that the ejection fraction may be normal. At that point there is a dilemma: what does this patient have? In a non-MI situation, patients of course could have heart failure with preserved systolic function due to something else, like atrial fibrillation for example. But in other situations I think it is really quite difficult. Natriuretic peptides probably do have a role in this clinical setting. If I am faced with a patient and an absolutely normal looking echocardiogram I have to think hard before diagnosing heart failure, even if they have symptoms. Of course if they have all the signs of heart failure then fine, but if they don’t I think it is quite difficult and I do really need some further evidence to start them on drug treatment for the rest of their life.
I think that the extent of diastolic or “preserved systolic function” heart failure is probably smaller than is being suggested elsewhere but I would be interested to hear the comments of my colleagues on that matter.
Professor Swedberg: First of all, this syndrome of preserved left ventricular function is more common among women and that has been documented now in several studies, including the EuroHeart failure survey. The symptoms are very similar in both impaired and preserved LV function heart failure. We should remember that according to the Braunwald criteria, all heart failure is diastolic heart failure because it involves increased end diastolic pressures! Clinically, it is important that both systolic and diastolic function are evaluated, and that the European Society of Cardiology diagnostic criteria are applied. We do know that the cardiovascular morbidity and mortality in patients with “preserved systolic function” heart failure is about half that of those with LVSD heart failure, although the non-cardiovascular morbidity and mortality is similar. It has been difficult to demonstrate the benefits of therapy with lower event rates in such patients. There is only one outcome trial so far among these patients, the CHARM Preserved study among 3000 patients, in which we could show that hospitalisations were significantly reduced by the addition of candesartan to their drug therapy. More studies are needed and some trials are ongoing, but this has been a neglected patient population.
Question: I would like to ask Dr Fuat and Ms Grange what mechanisms they use for prescription of drugs for their heart failure patients: is it the nurse and pharmacist referring to doctors for prescribing, or do the nurse and pharmacist prescribe and the doctor countersigns? Is it a patient group direction or supplementary prescribing?
Ms Julie Grange: There are many different systems throughout different hospitals in the country: some people go down the supplementary prescribing route, some centres prefer to use patient group directions, while in some centres nurses simply ask for a doctor’s signature in the corridor. It is just whatever works in the local centre, because what works in one centre won’t necessarily work in another.
Dr Fuat: What currently happens in Darlington is that the nurse asks us to sign the prescription, but she will shortly be prescribing as a supplementary prescriber as she has gone through a detailed training course. So from now on our heart failure nurse will actually be prescribing β blockers and other agents for heart failure. But I agree that there are different models in different parts of the country and a lot of it depends on what the senior pharmaceutical adviser allows; ours wouldn’t let us initially allow nurses to prescribe β blockers.
Professor Cowie: It would be interesting to get a view from a pharmacist—where is that professional group going in terms of its steer for pharmacy role in chronic disease management?
Reply: I am a pharmacist working in North Wales and I work full time with a heart failure team. I have one dispensary slot a week and apart from that I work as part of the heart failure team and am involved in running my own clinics, up-titrating drugs. I am also going through a supplementary prescribing course with one of the nurses as we decided that a multidisciplinary approach was the best way forward. Our Drug and Therapeutics Committee and the various other people in the hospital are happy with us changing drugs; they decided that was the safest way to do things.
Question: Patients often become hypotensive as more agents are given. Is it better for a patient to be just on an ACE inhibitor, with the dose being increased until the blood pressure drops, or if the blood pressure drops should you reduce the ACE inhibitor so you can introduce and up-titrate the β blocker until the blood pressure drops? Should we be aiming to initiate multiple agents at lower doses or just going for one at an optimal dose?
Professor Dargie: Having been involved in a clinical trial in which we were doing just that, experience was that it proved impossible to reach “target” dose for all agents together. If we wanted to get the β blocker on board it was rarely possible to get the patient titrated up to the full dose of ACE inhibitor before we did that. So I think we compromised; the patient had to be on at least the starting dose of an ACE inhibitor, preferably higher, before the β blocker was introduced. But it comes back to the point we were discussing earlier, that in a trial situation of course the patients then became up-titrated as we went through the trial; to what extent can clinical practice mirror clinical trials? How do we achieve that? We should remember that most of the evidence for the use of β blockers in heart failure is on top of ACE inhibitors. I am not sure, however, if it really matters what you should aim to get titrated up to the maximum first—this can be left to local judgment.
Professor Cowie: If I may just clarify your recommendation—you would not start several agents together, but start with an ACE inhibitor and then add the other drugs?
Professor Dargie: Indeed. I would normally start the ACE inhibitor first. I am very much aware of the remodelling effects of ACE inhibitors when given quite early. That is one of the reasons why it is important to diagnose left ventricular dysfunction and I then have no hesitation in prescribing ACE inhibitors in the acute MI situation because I know that it is going to be valuable. I am a little more hesitant in giving ACE inhibitors in this acute situation when there is no left ventricular dysfunction because I am aware that the fall in blood pressure can be hazardous.
Professor Cleland: If you look at the trial evidence, we don’t have a great deal of evidence that one dose is better than another. The dose ranging issues for β blockade have only been looked at in very small trials, but one could argue that if you want to keep people feeling well and out of hospital, 6.25 mg twice daily of carvedilol is superior to 25 mg twice daily, but if you want to save lives it may be the other way round, so maybe you have to make your choice! With ACE inhibitors the water is even murkier, because there was a large outcome study, ATLAS, which compared low doses and high doses and it really didn’t show huge differences between them. But we have to look at the data a little more carefully. ATLAS did show that if you had mild heart failure with reasonably well preserved systolic blood pressure, then you really did benefit from high dose ACE inhibition, so the mildest patients are the ones who benefit from high doses. For sick patients who are having difficulty tolerating therapy it really doesn’t seem to matter which dose you use, perhaps because they only tolerate one dose, but I would prefer to use small doses of several agents in a patient who was intolerant of high doses of any one agent. Patients with milder heart failure very often will tolerate full doses of all of these agents.
Professor Cowie: So you initiate ACE inhibitor and β blocker together?
Professor Cleland: I would prefer a rapid initiation sequence so that you start ACE inhibitor on day one, β blocker on day 2 and maybe eplerenone on day 3. Eplerenone gives a significant mortality reduction within 30 days so starting your eplerenone at the six week outpatient clinic is too late and doesn’t seem to be optimal use according to the clinical trial data.
Professor Dargie: I would like to ask Professor Cleland if he is basing his opinion on only one trial, the ATLAS trial?
Professor Cleland: Well, we don’t have a wealth of evidence from dose ranging trials. The largest dose ranging trial in the UK, which was NETWORK, was neutral. I think that ATLAS clearly shows that if you can titrate to “full” dose of ACE inhibitors, the patients tend to benefit and it is indeed the mildest patients, the patients you don’t think need it, who are the ones who get the most benefit.
Dr Mark Dancy: Although we conventionally think of the appropriate dose ranges for ACE inhibitors being the same for post-MI patients as for heart failure patients, of course the dose ranges for β blockers in post-MI patients without heart failure are considerably greater than those we use for people with chronic heart failure. Therefore if you have got patients who, say, go into pulmonary oedema when given relatively large doses of β blockers early after acute MI, then one of the compromises is to fall back to heart failure type doses, really small doses of β blocker, in which case you may be able to introduce the ACE inhibitor as well.
Dr Fuat: It is important to remember that these patients are on lots of other drugs, some of which also drop blood pressure. Do they still need their calcium channel blocker for their hypertension? Do they need high doses of antidepressants or benzodiazepines? There are lots of other agents. We need to address that issue and look at whether we can drop diuretic doses back down, rather than keeping them on higher doses.
Question: What about treatment of acute pulmonary oedema on the coronary care unit? What about nitrates, diuretics and CPAP?
Professor Swedberg: In general, for acute care, nitrates have an important role to play in symptom relief and haemodynamic stabilisation. The problem is that we are uncertain about the importance of this for the long term outcome. The nitrate is the best treatment for reducing increased LV filling pressures and pulmonary and venous pressures, and usually in combination with a diuretic.
Professor Cowie: Would you use an ACE inhibitor or eplerenone in such patients?
Professor Swedberg: I have listened to the discussion about what, when and how to initiate and of course I have a number of thoughts. We tried in the CONSENSUS II trial to use ACE inhibitors very early. Although patients in that trial were not in acute failure, this was the only ACE inhibitor trial to “fail”. We wished to use ACE inhibition as early as possible, and so gave it intravenously on the first day after infarction. That was probably too early. My current practice in patients with heart failure early after myocardial infarction is to stabilise such a patient using vasodilators such as nitrate, and after stabilisation but as early as possible initiate ACE inhibitor. Definitely I would not use a β blocker at that early stage. I would use eplerenone also after the initial stabilisation.
Dr McDonagh: Eplerenone was started at day 3 of the trial so there is no evidence for its use before then.
EPLERENONE IN CLINICAL PRACTICE
Question: While we are waiting for eplerenone I have been using spironolactone in patients with left ventricular dysfunction and heart failure after myocardial infarction. Do I have reason to change my practice unless there are side effects? Unfortunately this trial was again disadvantaged by the mean age being 64, so a lot of elderly people who would be coming into coronary care unit would be suffering from side effects.
Professor Cleland: One can ask whether you prevent side effects or treat side effects. The way to prevent side effects is to use eplerenone; the way to treat them is to switch from spironolactone to eplerenone. So what would you want as a male patient: being told that you needed to go onto spironolactone? Would you be willing to pay your extra 50p per day to avoid what might be considered a relatively minor side effect but is distressing when you get it?
Professor Cowie: I think one has to be careful. The mean age of patients admitted to coronary care units in most of Europe is actually only in the high 60s so the EPHESUS trial was recruiting from a representative coronary care unit population, and there are a few patients that are quite elderly in that study—I don’t think there was an upper age exclusion. But the key issue in answer to your question is class effect and the interpretation of “evidence based” medicine. Professor Swedberg’s talk based on the draft of the new European Society of Cardiology guideline uses the generic phrase “aldosterone antagonist”, when strict evidence based medicine enthusiasts would say it is eplerenone for post-MI and that we should not extrapolate to other patient groups. It would be useful to get some opinions from the panel on whether they will start using eplerenone in post-MI heart failure once it becomes available, or whether they will use spironolactone (or neither!).
Dr McDonagh: For a patient with chronic heart failure, I would switch to eplerenone if the patient had oestrogenic side effects. I would also make sure they weren’t on digoxin: the incidence of gynaecomastia is far higher in patients on digoxin. In terms of using it in the chronic heart failure situation I would await the result of the clinical trial, which I think Pfizer is undertaking in patients with NYHA class II and III heart failure with systolic dysfunction, and those are going to be patients over the age of 60. In terms of post-MI, I think I would add eplerenone in a patient who could tolerate it at day 3.
Professor Cleland: I guess we have to look at the bigger picture: 93% of health care is nothing to do with drug costs, and in fact just one or two fewer visits to the doctor to manage the side effects of the wrong treatment might well pay for the more expensive therapy. I think that while spironolactone will remain the mainstream treatment for the management of chronic heart failure, my threshold for using eplerenone will be very low—25% of our patients are in atrial fibrillation and therefore on digoxin, so there is already a target group in which I would probably go for eplerenone as the agent of choice in the management of heart failure rather than spironolactone. There are a number of other groups. Also, if you are giving informed consent to your patients when you are giving them drugs, some of them will be rather alarmed at the side effects. Knowing that there is an alternative, which although slightly more expensive is free of these side effects, becomes an issue of patient choice. It will be interesting to watch what happens. I think that the uptake in the heart failure population will be quite high.
Professor Dargie: Very rarely have I prescribed spironolactone in the coronary care unit. I have regarded that as a treatment for chronic heart failure, and so I think my practice in chronic heart failure will probably be the same, and if there are side effects then I may well move on to eplerenone in the chronic heart failure situation, even though we don’t yet have the evidence. But we do now have a large clinical trial with the evidence in the coronary care unit, and I am inclined to follow the trials. I think these trials do indicate a dose and a scheme which has been effective and safe, and I don’t know whether using spironolactone in the coronary care unit in the same way would be as effective. I guess that if the health providers will not fund eplerenone for any reason, then I might consider using spironolactone, but in the first instance I would be inclined to follow the clinical trial evidence.
Professor Cleland: I would just point out that the EPHESUS study went on for 16 months and the proportion of time spent in the coronary care unit was less than 1% of follow up. We are looking at the evolution of a disease. It is not like thrombolysis where you give it one day and you are looking for an effect over 16 months. This is something you are continuing to give during the evolution of the disease.
Professor Dargie: But this is a different situation, we are looking at an evolving situation of left ventricular dysfunction.
Professor Cowie: Now that there is a drug that is superior to spironolactone but which is slightly more expensive in terms of the head to head comparison of unit cost, the cost of care is clearly going to be a big issue for drug and therapeutic committees. Do you think specialists and people interested in heart failure should be giving a lead to their local health care community here?
Dr Dancy: I am always conscious of the fact that if you spend an unnecessary amount on new drugs you are effectively robbing some other part of the health care system. My own personal experience of spironolactone has not led me to believe that the incidence of side effects, particularly gynaecomastia, is very common. I would have thought it perfectly reasonable to use spironolactone and change to eplerenone if side effects arise; otherwise you are going to have to go lock, stock and barrel over to the new drug in a common disease, with fairly major health care costs associated with it.
Julie Grange: As a non-prescribing nurse, I would be guided by whichever drug my physician would want to use. But I would monitor whichever drug was used and report back any side effects and provide alternative suggestions should they be required.
Professor Cowie: That is an important message, as both drugs influence potassium balance and people do need to be very conscious of this and monitor the blood biochemistry, whichever aldosterone antagonist they decide to use.
Professor Swedberg: We have two very well documented classes of drugs used after MI: ACE inhibitors (and if not tolerated valsartan), and a β blocker. To initiate on top of that, rapidly, an aldosterone blocker may lead to difficulties. I just want to raise one concern that hasn’t really been tackled in clinical practice, and that is the concern about renal function. Unfortunately, there is a need for good renal function to be able to tolerate eplerenone. In addition to age and systolic function, renal function is maybe the third most important prognostic factor in the coronary care unit, and a number of these patients will not be eligible for eplerenone if we use the entry criteria used in the EPHESUS study (serum creatinine less than 220 μmol/l).
Professor Cleland: I would just point out that in the EPHESUS study there was a significant interaction with the other treatments for ventricular dysfunction in heart failure, such that patients who got an ACE inhibitor, a β blocker, and eplerenone did best. So in fact we have evidence of stronger benefit when these drugs are used in combination then when they are used one by one.
Dr McDonagh: What we didn’t point out was that there was a similar cut off point of creatinine to be eligible for EPHESUS as there was for RALES, which was 220 μmol/l, which is quite high. Not many patients in the coronary care unit have a creatinine of that level.
Professor Dargie: I think there is no doubt that if one is going to introduce eplerenone, then it has to have appropriate monitoring, but then you are introducing it into the highest risk group of patients and why should there not be an effective disease management programme for such patients in any case?
IDENTIFYING MARKERS OF APPROPRIATE CARE
Question: Professor Pearson highlighted the need to audit heart failure and has challenged us as a group is to come up with something to measure in these patients that is relevant and can help the health service move forward, just as MINAP helped with thrombolysis. Can the panel suggest one or two points in heart failure care that they think would be good measurements for us to recommend to our national societies?
Professor Cleland: I think that there are four elements: the symptoms and signs of the clinical syndrome, which do have prognostic value (although the problem is you are trusting very much to the doctor at bedside to record these accurately and faithfully, and to have equal skills up and down the country); secondly, whether an echocardiogram is performed or not (already being audited in MINAP); thirdly, plasma natriuretic peptide concentration (although there are a lot of problems about their diagnostic use there doesn’t seem to be much problem about their prognostic use, and it may be that if we had had natriuretic peptides before we had the echocardiogram we would be choosing to use natriuretic peptides as the objective criteria for the diagnosis of heart failure, and treating according to natriuretic peptide levels); and finally, diuretic usage (a powerful prognostic marker). Heart failure is a syndrome and you should look at a cluster of things rather than just one thing. My four suggestions are relatively simple to measure:
Did your doctor think you had heart failure or ventricular dysfunction?
Does your echocardiogram show that your ejection fraction is impaired?
Is your natriuretic peptide up?
And have you required diuretic treatment?
Professor Cowie: I would like to ask the GP on the panel, Dr Fuat, about the new GMS contract, which provides quality points for the use of echocardiography to confirm the diagnosis of left ventricular systolic dysfunction. Of course, the only reason one would know that the patient has left ventricular systolic dysfunction is because an echocardiogram has been performed. Do you think echocardiography should be one of the key quality measures of heart failure care in the UK?
Dr Fuat: I think it has to be. I think the other measurable things are the drugs. We know that the drugs work, and therefore auditing the uptake of those is rightly in the GMS contract. Sadly it stopped short at having β blockers and went only for ACE inhibitors, but I don’t doubt that when it is revamped in a year’s time and we have all met our targets, they will add harder ones. I think quality of life is also important but measuring this is difficult: we tried to use the Minnesota Living with Heart Failure questionnaire when we first started the clinic in Darlington, but found it quite hard to administer and analyse. We are about to start a joint project with the University of York that will look at different measures of quality of life.
Dr Dancy: On the diagnostic components that John McMurray has alluded to, the only thing that rather surprises me, although I will take the evidence as stated, is that even transient heart failure as evidenced by the need for a single dose of diuretic is a prognostically significant event. These sorts of patients, if they only have one dose of diuretic, are probably on the absolute borderline situation between overt heart failure and not heart failure. The type of decision the junior doctor is making in that circumstance is usually on the basis of crackles on lung auscultation or something of that sort, and we all know that if you stick anyone in a bed for 24 hours they get crackles. I think that the diagnosis of heart failure is very difficult, so it surprises me that a single shot of diuretic is so significant in terms of prognosis.
Professor McMurray: Even transient heart failure after acute MI, diagnosed on clinical grounds, is associated with increased risk of poor subsequent outcome. This was clearly shown by the TRACE investigators where, in a multivariable analysis, transient heart failure was an independent predictor of all cause mortality. The risk ratio was 1.5, though this was lower than for persistent failure (risk ratio 2.8). Importantly, this increased risk was independent of left ventricular function.
Professor Cowie: But, Dr Dancy, what would you measure as the chair of the CHD Collaborative looking at how heart failure management varies across England?
Dr Dancy: Well, quality and quantity of life are the only other two major outcome measures, both of which are difficult—quality of life for the reasons we have heard and quantity of life because each area will only have small numbers of patients and any difference may not be statistically significant.
Professor Cleland: I think that we are being confused by the quality of life aficionados. There are two issues: one is what your quality of life is, and the other is why is it that your quality of life is not the way that you want it to be? The problem with most quality of life questionnaires is that they are neither fish nor fowl. They do not ask the simple question—“how do you rate your quality of life”—and yet often cannot determine why your quality of life is impaired.
Dr Dancy: The other measure is hospital admissions, which is commonly used and seems a good marker.
Professor Cowie: It wouldn’t be unreasonable to say that use of echocardiography, the use of life saving drug therapy of which we have a few to choose from, and the hospitalisation rate for heart failure in the district would be three measures that we could perhaps put forward as being justifiable measures of quality.
Question: Professor Cleland mentioned this morning his concern about aspirin in heart failure. I am concerned because most patients with heart failure are on aspirin, so are we killing them? What is your advice?
Professor Cleland: The evidence that we are killing them is not in. The evidence that we may not be doing them any good is beginning to come in. Results from two randomised control trials, WASH and WATCH, have been presented. They have both shown about a 35% increase in heart failure related hospitalisations among patients randomised to aspirin compared to warfarin. The smaller WASH study suggested there was no difference in hospitalisation rates between warfarin and no antithrombotic therapy. So the suggestion is that aspirin increases your risk of heart failure hospitalisation substantially, warfarin probably has little or no effect, and that maybe if you are concerned about polypharmacy stopping antithrombotic therapy in this population might be to their advantage.
Professor Cowie: The NICE chronic heart failure guideline committee had to consider this type of emerging evidence, and it took the view that it would be irresponsible at this stage to change what is generally accepted: that if a patient has atherosclerosis, particularly with unstable plaque, there is good evidence to use antiplatelet therapy. As Professor Cleland has stated, the evidence is accumulating about the role of aspirin in heart failure. The consensus message from today is certainly not that you should stop aspirin in all your patients with heart failure.
Professor Cleland: I would like to point out that there is pretty good evidence that for the 6–12 weeks after an acute vascular event, aspirin is of benefit. The issue really is about whether you should continue the antithrombotic therapy beyond 6–12 weeks after an acute event, and there the data are accumulating that you should not.
Question: Is the chest radiograph worthwhile in the diagnosis of heart failure?
Professor Swedberg: In the CONSENSUS I trial, the very first survival trial with ACE inhibitors, in very sick patients the only entry criteria for documenting heart failure was a chest x ray, and several of these patients in a substudy turned out to have preserved systolic function: this is often forgotten. For the diagnosis of heart failure, the guidelines say that if there is an enlarged heart on the chest radiograph there is every reason to go on and image the heart using echocardiography, which will provide much more information about the heart structure and function, including that of the valves. But for sure, after an MI a large heart on the chest radiograph would be adequate for making a diagnosis of heart failure if there are symptoms. Echocardiography will then, most likely, provide you with additional information.
Dr McDonagh: Looking at the chronic heart failure situation, there is really a two edged side to this coin. If you use the chest x ray as your sole investigative tool to diagnose heart failure, it is not very good: it is like tossing a coin. However, you are dealing with a breathless patient and they should have a chest x ray because there are many other pulmonary causes of breathlessness that you may miss. So a chest x ray is mandated in all the guidelines for diagnosing heart failure because you really don’t want to miss a lung tumour.
Professor Cowie: I wish to sum up some of the key points that I have heard throughout the day. Many patients have heart failure. Many patients have failure after myocardial infarction. Many of these patients are not in the obvious places: they may not be in the coronary care unit, they may not have had an ST elevation infarct, and they may be diabetics without obvious signs of problems. But these are patients that we need to deal with better: they must not be allowed to slip through the net.
The evidence base for treatment of patients with heart failure after a myocardial infarct is really not contested: we have more evidence to go on than in many areas of medicine. Proper diagnosis is the basis for proper treatment, which now includes several drugs such as ACE inhibitors, β blockers, and eplerenone. Initiation and up-titration of these drugs is an organisational challenge, with implications for both primary and secondary care.
Perhaps it is your responsibility, as a local champion for this area, to get things moving. The solutions will be different in different locations, although we probably all agree about what we have to achieve. We have heard about different models that have worked in chronic heart failure and some components will undoubtedly transfer across into the post-MI setting. I think Dr Fuat’s three bullet points sum up the situation very neatly: we need to move forward by addressing the care pathway from three different viewpoints: that of the patient, the physician or health care professional, and the “system”. We need to look hard at the barriers to implementation at those three levels.
The challenge to you is to return to your own practice, to think about the issues, and to move practice forward by implementing the evidence you have heard. Review of progress, with audit of key parts of the pathway, will be essential to demonstrate both improvements and areas of continuing difficulty.