Q J Med 2002; 95: 419-421
© 2002 Association of Physicians
Editorial |
What do Cochrane reviews tell us about anti-thrombotic therapy in heart failure with sinus rhythm?
Viscount Royston Professor of Clinical Cardiology, National Heart and Lung Institute, Imperial College School of Medicine at Royal, Brompton Hospital, London e-mail: ajscoats{at}aol.com
Although major advances have been made in the treatment of many cardiovascular disorders,1 many patients with advanced cardiovascular disease remain symptomatic and are far from cured. For most of the history of medicine, treatments have been designed to relieve symptoms and suffering, and to render the patient more fully functional. In the modern era, the concepts of disease and cure have taken on new meaning, and the idea of curative therapy has become widespread. Even more recent is the concept of preventive therapy, whose sole aim is to prolong life or prevent some future catastrophe in people who are presently well. These are not people complaining of something: they are people we seek out to recommend therapies that themselves may cause disability or unpleasant symptoms, in the hope of reducing later illness. Even modern patients sometimes fail to grasp the significance of this change in the contract between doctor and patient.2
Successes abound in cardiovascular medicine. Anti-platelet, hyoplidaemic and hypotensive strategies are effective in a broad range of patients at risk of major cardiovascular disease, and success has also been established for beta-blockers, ACE inhibitors, and the specific aldosterone antagonist spironolactone, in severe heart failure.3 In the former category, people are happy to take reasonably well tolerated medicines for purely preventive purposes. In the latter, very symptomatic patients with a clearly lethal disease are prepared to take therapies that fortunately improve both survival and patient well-being. A more difficult scenario is in recommending anticoagulation with warfarin. This therapy is not easy. It requires considerable effort by the patient, has obvious side-effects, and does nothing to improve the patient's quality of life. Its only value is that it reduces the chance of something really nasty (death or a major stroke).
Patients with artificial heart valves or with atrial fibrillation and mitral stenosis have been convinced that the risk of stroke is so high that anticoagulation is essential.4 Only a minority default. For more borderline indications, the use of full anticoagulation is much less universal. Full anticoagulation is usually recommended in the presence of atrial fibrillation and at least one other risk factor for embolic stroke, including softer risk factors such as age >65 years, but only a minority of such patients actually receive it. Why is this? Is the evidence unconvincing? One concern is whether the much more acceptable anti-platelet agents might be nearly as good. This question of whether to pursue anti-platelet therapy, anticoagulation, both or neither, needs to be investigated, and the results publicized, for each and every patient group where the result might be different. We will be highly unlikely to get definitive trials in every patient disease cohort, and most trials are too small to be definitive. There are also too many trials for the average practising doctor to keep abreast of the evidence. The Cochrane collaboration is designed to be a permanent (and regularly updated) repository of the synthesized trial evidence for a variety of disease-treatment scenarios. A practising doctor faced with a particular management issue, could turn to see if a Cochrane review addresses the question, and patients might eventually do the same.5
Imagine a common clinical setting. A physician is looking after a 73-year-old man with NYHA class II chronic heart failure. The patient has had a previous inferior myocardial infarction and is already taking an ACE inhibitor and a diuretic. The doctor would obviously want to offer him evidence-based therapy, and to start a statin to reduce cholesterol and a beta-blocker proven to reduce mortality because of his heart failure. Should the doctor also tell the patient he needs aspirin or warfarin, or both? Would we expect the doctor to know which trials are specifically relevant to this patient and what the outcomes were? Would he or she be likely to remember which, if both strategies worked, would be preferable? Would the patient tolerate being asked to take so many new therapies unless the evidence presented were sound and convincing? In this setting, the Cochrane database could be very helpful.
Two relevant reviews are presented in this issue of QJM, both by Lip and Gibbs, on the questions of anticoagulation or anti-platelet therapy for heart failure with sinus rhythm,6,7 They have been formatted to adhere to Cochrane Collaboration format, and could form the basis of regularly updated reviews. They are therefore very timely, given our uncertainties in clinical practice. It has long been known that patients with heart failure are at increased risk of thrombotic events, even when in sinus rhythm,8 but of course this does not mean that treating this tendency will necessarily be beneficial.
Lip and Gibbs conclude in the first review that little if any evidence exists to recommend aspirin for patients with sinus rhythm and heart failure, even though this group of patients overlaps with the group with known ischaemic heart disease where chronic anti-platelet therapy is recommended. Many patients with extensive vascular disease, where stroke prevention has been shown for both older and newer anti-platelet agents,9 will go on to develop heart failure. Despite recent controversy in the press fuelled by speculative reports, and post-hoc and selective analyses of trials designed to test alternative questions,10 there is a dearth of evidence to suggest we should stop anti-platelet therapy in a patient with vascular disease simply because heart failure has developed. Post-hoc analyses can be seriously distorted because of the effects of prior therapy, because anti-platelet therapy in particular changes the outcome of myocardial infarction, and because patients with aspirin therapy are inherently different to those not taking aspirin, in any non-randomized comparison.11 At the same time there is also a lack of evidence that we should start anti-platelet therapy merely because the patient has developed heart failure. In other words, in heart failure with sinus rhythm, the jury is out and the evidence is balanced; we should use other factors to decide whether the patient needs anti-platelet therapy. Assuming the authors have done a good job (and the Cochrane accreditation, if given here, would be a very useful marker of reliability) then we have a synthesis of opinion that most of us could not better without extensive literature searching.
For the second question, that of anticoagulation for heart failure in sinus rhythm, the authors found supportive randomized controlled trial evidence only in three small trials on average 
50 years old. These did not demonstrate the precision of methods and standards we expect in modern trials, and certainly were not against the background of modern heart failure therapy. What did the authors conclude? Again, that the evidence is contradictory and does not support mandatory use of anticoagulation in this particular setting. Here the doctor cannot know what is best to do, and need not feel compelled to recommend a treatment that has no symptomatic benefit, and whose prognostic benefit is unproven. Again the review helps, because it conforms to the standards of a Cochrane collaborative review, the best form of evidence we are likely to get until further trial evidence becomes available. It is also reassuring that a virtually simultaneous review came to almost identical conclusions.12 In the situation (which is likely) that the patient has been on aspirin himself because of press campaigns and the evidence of benefit after his older inferior myocardial infarction, we should recommend he stays on it. The outcome of stopping therapy is sometimes not the same as having never taken it.13
Further trials of both anti-platelet agents and anticoagulation are sorely needed, and we await the results of the WATCH trial of aspirin and warfarin in heart failure with sinus rhythm with interest. In the meantime doctors will continue to use anti-platelet agents for vascular disease and warfarin in the settings where it has been truly proven. Our hypothetical 73-year-old is saved from yet another treatment, and one requiring regular attendance at the anticoagulant clinic, until definitive trial evidence arrives.
References
1. Coats AJ. Cardiovascular research in the dock: a case for the defence. Int J Cardiol2000; 75:121–2.[Web of Science][Medline]
2. Coats AJ. Life, quality of life and choice in an ageing society. Int J Cardiol2001; 78:1–3.[Medline]
3. Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, Palensky J, Wittes J. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized Aldactone Evaluation Study Investigators. N Engl J Med1999; 341:709–17.
4. Abdelhafiz AH. A review of anticoagulation with warfarin in patients with nonvalvular atrial fibrillation. Clin Ther2001; 23:1628–36.[Medline]
5. White PJ. Evidence-based medicine for consumers: a role for the Cochrane Collaboration. JAMA2002; 90:218–22.
6. Lip GYH, Gibbs CR. Anticoagulation for heart failure in sinus rhythm: A Cochrane Systematic Review. Q J Med2002; 95:451–9.
7. Lip GYH, Gibbs CR. Antiplatelet agents versus control or anticoagulation for heart failure in sinus rhythm: A Cochrane Systematic Review. Q J Med2002; 95:461–8.
8. Davis CJ, Gurbel PA, Gattis WA, Fuzaylov SY, Nair GV, O'Connor CM, Serebruany VL. Hemostatic abnormalities in patients with congestive heart failure: diagnostic significance and clinical challenge. Int J Cardiol2000; 75:15–21.[Web of Science][Medline]
9. CAPRIE Steering Committee. A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). Lancet1996; 348:1329–39.[Web of Science][Medline]
10. Cleland JGF. Commentary. No reduction in cardiovascular risk with NSAID's: including aspirin? Lancet2002; 359:92–3.[Medline]
11. Lancaster GI, Lancaster CJ, Radley D, Cohen M. Prior aspirin use in unstable angina predisposes to higher risk: the aspirin paradox. Int J Cardiol2001; 80:201–7.[Medline]
12. Sirajuddin RA, Miller AB, Geraci SA. Anticoagulation in patients with dilated cardiomyopathy and sinus rhythm: a critical literature review. J Card Fail2002; 8:48–53.[Web of Science][Medline]
13. Collet JP, Himbet F, Steg PG. Myocardial infarction after aspirin cessation in stable coronary artery disease patients. Int J Cardiol2000; 76:257–8.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||