Rapid Responses to:

EDUCATION AND DEBATE: John G F Cleland For Debate: Preventing atherosclerotic events with aspirin BMJ 2002; 324: 103-105 [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Antiplatelet efficacy in high risk patients

Colin Baigent, Rory Collins, Richard Peto   (11 January 2002)

ASPIRIN OR SALMON STEAK FOR HEART DISEASE ?

Eddie Vos   (15 January 2002)

Chronic aspirin therapy is effective - if you massage the data sufficiently

John GF Cleland   (15 January 2002)

More inaccuracies in the commentaries on the antiplatelet therapy meta-analysis

Cathie LM Sudlow, Peter Sandercock, Charles Warlow   (30 January 2002)

Re: Chronic aspirin therapy is effective - if you massage the data sufficiently

Colin Baigent, Rory Collins and Richard Peto   (31 January 2002)

The Editor of the BMJ displays errors of judgement

Nick Manassiev   (13 February 2002)

Aspirin for Coronary Disease - More spin than substance.

John GF Cleland, HU6 7RX   (26 April 2002)

Antiplatelet efficacy in high risk patients 11 January 2002
Colin Baigent Radcliffe Infirmary, Oxford, Rory Collins, Richard Peto Send response to journal: Re: Antiplatelet efficacy in high risk patients	The worldwide meta-analysis of antiplatelet trials1 shows that low- dose aspirin (or some other effective antiplatelet regimen) reduces nonfatal myocardial infarction, nonfatal stroke and vascular death in a wide range of patients who are at high risk of occlusive vascular disease. A paper disputing this2 was published concurrently in the "For Debate" section of the journal, but the arguments in it (some of which the author also published on the same date in a Lancet editorial3) depend strongly on quite simple mistakes about the randomised evidence, and could cause unnecessary deaths. Consider, for example, the ISIS-2 trial of short-term antiplatelet therapy, in which 17,187 patients with suspected acute myocardial infarction were randomised, half to active aspirin and half to placebo4, demonstrating a clear reduction in 5-week all-cause mortality (811/8587 [9.4%] aspirin versus 1030/8600 [12.0%] placebo deaths, 2P<0.00001)5. Bizarrely, in a section entitled "Trials do not show that aspirin saves lives", the For Debate paper attempts to dismiss the ISIS-2 findings by suggesting that "all patients lost to follow-up in the active group should be considered to have died and none of those in the control arm. Such an analysis would neutralise the benefit observed in one of the few seemingly convincingly positive studies of aspirin, the ISIS-2 trial" 2. This is not even arithmetically correct, and such a statement should not be part of any serious debate in the BMJ. The 5-week follow-up was 97% complete when this trial was first reported4, and 99% complete when further follow-up was reported in the BMJ5. This slightly greater completeness yielded, in fact, only 13 extra deaths (6 aspirin, 7 placebo), and the even slighter incompleteness that remains cannot, of course, be of any material relevance (especially since most of the few still untraced at 5 weeks are known to have been discharged alive from hospital: for only 0.2% of the aspirin and 0.2% of the placebo patients is there no follow-up at all). Likewise, among about 20,000 patients in the 12 trials of long-term (mean 2 years) antiplatelet therapy among patients with a history of previous myocardial infarction, the odds of having a nonfatal reinfarction recorded was reduced by 30% (SE6; 2P<0.00001), with no significant heterogeneity between the results in different studies1. The For Debate paper purports to account for this 30% reduction by suggesting (without good evidence of any such effect) that the proportion of nonfatal infarcts that would be reported might be 70% with aspirin and 75% without. Again, however, this argument is arithmetically wrong, for 70 versus 75 would represent a reduction of only 7%, not 30%. Furthermore, having suggested earlier that it is only analyses of all -cause mortality that can be trusted, the paper then goes on to elaborate a curious theory that involves trusting the somewhat arbitrary distinction between mortality attributed to sudden death and to other cardiac causes. From this it eventually concludes that aspirin could be producing "an increased risk of sudden death among concealed, and therefore untreated, events" 2. But, there is no good evidence that this is true. More importantly, in the worldwide meta-analysis, "vascular mortality" — which is highly significantly reduced1 — already included both sudden death and death from unknown causes (as well as death from any type of stroke). In the 12 trials of long-term antiplatelet therapy during the years after myocardial infarction, the reduction in vascular mortality was 15% (SE5; 2P=0.002) again with no significant heterogeneity between the effects in different antiplatelet trials (or 17% [2P<0.001], with even less heterogeneity, if the imbalance in prognostic features in the AMIS trial is appropriately allowed for6). Moreover, all-cause mortality was also reduced, as there was no significant excess of non-vascular deaths in this category of patient, or in any of the other four main categories of high- risk patient. Indeed, taking the 135,000 patients in all five categories together, non-vascular mortality was 1.1% with antiplatelet therapy and 1.2% without, which looks pretty safe. Thus, there is no good evidence from these trials that non-vascular mortality offsets the highly significant reduction in vascular death, or in non-fatal myocardial infarction or stroke, among high risk patients. A recent study of the costs of the secondary prevention of such vascular events by aspirin is cited in the For Debate paper as concluding that the cost per event prevented would be over £3000. If true, this could be money well spent, but it is included in a section misleadingly entitled "Neither safe nor cheap"2! (No other cost estimates in that section are relevant to secondary prevention.) The author also suggests that "the greatest potential detriment of aspirin on health care, however, is that it diverts attention away from treatments that are of unequivocal benefit". No good evidence for this assertion is provided and, moreover, there is no good reason why other effective treatments (such as angiotensin converting enzyme inhibitors, beta-blockers and statins) should not be used in addition to aspirin, conferring additional benefit7. There are several other errors of judgement, partly from failure to understand the proper role of meta-analysis in the interpretation of randomised evidence. Given this, none of the substantive points in the For Debate article is of material relevance (except, perhaps, as a warning about the power of prejudice), and the chief ones have been dealt with adequately in the current1 or previous antiplatelet reports. In retrospect, it would perhaps have been better for the BMJ to have sought review of the paper from, inter alia, those whose work it criticises, as this would have given the journal the opportunity to avoid publication of arguments and conclusions that are wrong for trivial reasons, and potentially damaging to patients. References 1. Antithrombotic Trialists' Collaboration. Collaborative meta- analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction and stroke in high risk patients. BMJ 2002; 324: 71-86. 2. Cleland JGF. For debate: preventing atherosclerotic events with aspirin. BMJ 2002; 324: 103-5. 3. Cleland JGF. No reduction in cardiovascular risk with NSAIDS — including aspirin? Lancet 2002; 359: 92-3. 4. ISIS-2 Collaborative Group. Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction. Lancet 1988; ii: 349-60. 5. Baigent C, Collins R, Appleby P, Parish S, Sleight P, Peto R on behalf of the ISIS-2 Collaborative Group. ISIS-2: 10-year survival among patients with suspected acute myocardial infarction in randomised comparison of intravenous streptokinase, oral aspirin, both, or neither. BMJ 1998; 316: 1337-43. 6. Antiplatelet Trialists' Collaboration. Secondary prevention of vascular disease by prolonged antiplatelet treatment. BMJ 1988; 296: 320- 31. 7. Flather MD, Yusuf S, Køber L, Pfeffer M, Hall A et al. Long-term ACE-inhibitor therapy in patients with heart failure or left ventricular dysfunction: a systematic overview of data from individual patients. Lancet 2000; 355: 1575-81. Colin Baigent, Rory Collins, Richard Peto Radcliffe Infirmary, Oxford Correspondence to: Dr Colin Baigent (colin.baigent@ctsu.ox.ac.uk)
ASPIRIN OR SALMON STEAK FOR HEART DISEASE ? 15 January 2002
Eddie Vos, maintains site: www.health-heart.org Sutton (Qc) JOE 2K0 Canada Send response to journal: Re: ASPIRIN OR SALMON STEAK FOR HEART DISEASE ?	This is in support of Dr. Cleland's questioning of the over-all survival benefits of "antiplatelet" (aspirin) prevention strategies in heart disease (1). Since about 90% of heart and vascular attacks are of the ischaemic (blocking) nature, statistics and logic support a survival benefit from immediate and short term (weeks) aspirin use, which is where there seems to be some current consensus. Dr. Cleland points out that several studies do not support a long- term survival benefit from prophylactic aspirin use. He further suggests that the aspirin approach may represent a "distraction" from other avenues of prevention such as with statins and other drugs. While aspirin itself is cheap, bleeding and congestive heart problems that may result from its use are not as statin drugs taken for prevention threaten the health of government drug-plans and their use may also be a distraction surpassing the aspirin story in scope. While highly touted as beneficial by reducing LDL-cholesterol, they often raise Lp(a) cholesterol, a player in blood clotting, and they always slash production of mevalonate derivatives like ubiquinone (Coenzyme Q10), so vital in heart cell energy production. The January 2002 North American consumer television ads for the highly prescribed statin Lipitor feature in fact a 6 second warning reading: "Lipitor has not been shown to prevent heart disease or heart attacks." Primary and secondary prevention are long-term strategies where the data about both aspirin and statins are complex and far from clear, and both are without question distractions from non-drug prevention approaches. For example, aspirin's main action is via the inhibition of the (omega-6) arachidonic acid cascade which generates circulation restricting factors. Increasing dietary omega-3 fatty acids (flax/linseed oil, canola oil, English walnuts and fatty fish) and reducing omega-6 linoleic acid intake (corn, sunflower, safflower and cottonseed oils, e.g. most margarines) promote blood fluidity "naturally" by several pathways, including via omega-3 competition for the cyclooxygenase enzymes that, without such competition (or aspirin), produce inflammatory and circulation reducing arachidonate derivatives. The worth of such food-based, circulation promoting strategies is being lost in the discussion about drug-approaches. This author proposes that more emphasis should be directed at the potential of nutritional changes affecting eicosanoid (aspirin and omega-3 controlled) pathways. For instance, the switch from sunflower and corn oils to high omega-3 flax, canola and fish oils can be made with little if any extra expense, but such changes will require education. Quoting Dr. A. Leaf (2): "..relatively simple dietary changes achieved greater reductions [70%] in risk of all-cause and coronary heart disease mortality ... than any of the cholesterol-lowering studies to date. This is emphasized by the finding that the unprecedented reduction ... was not associated with differences in total cholesterol levels." This comment concerned the Lyon Diet Heart [secondary prevention] Study where the main intervention was a canola based margarine, daily supplying the omega-3 alpha-linolenic acid found in 2 tablespoons of canola or in 1 teaspoon of flax/linseed oil. This suggests that long-term blood circulation enhancement with omega -3 oils, acting through changes in cell-wall function and anti-coagulation effects, may well generate blood circulation and anti-inflammation benefits in heart disease that eclipse similar effects of aspirin (and statins) but without the well known and inherent negative drug side- effects. If one has to argue about the statistics of mega-prescribed drugs like aspirin or statins, the real debate is likely to be elswhere. Eddie Vos; Sutton (Qc) Canada; e-mail: vos@health-heart.org (1) Cleland JGF. Preventing Atherosclerotic events with aspirin. BMJ 2002: 324: 103-5. (2) Leaf A. Dietary prevention of coronary heart disease: the Lyon Diet Heart Study. Circ. 1999 Feb 16;99(6):733-5
Chronic aspirin therapy is effective - if you massage the data sufficiently 15 January 2002
John GF Cleland, Professor of Cardiology University of Hull HU6 7RX Send response to journal: Re: Chronic aspirin therapy is effective - if you massage the data sufficiently	Baigent, Collins and Peto argue with passion for the effectiveness of aspirin but at the same time warn against the power of prejudice. They fail to address any of the important concerns that I raised in my article. A little less evangelism and a more objective view of the information they present might be helpful. Perhaps there is more than one reasonable perspective on the aspirin meta-analysis, which the authors should have the courage to admit. On the subject of arithmetic:- ISIS-2 lost 250 patients to follow-up on aspirin and a similar number in the control group, whilst there were 216 less deaths on aspirin (1). 250 (or 501 if you include the control group) is larger than 216 in most peoples calculations. Subsequently, long after the blind was broken, the outcome was reassessed (2). The difference in 35-day mortality had risen to 249 and the number lost to follow-up had shrunk to about 206. It is reasonable for purposes of generating a hypothesis to revise results retrospectively but it is the original report that must stand. I am sure that with enough post-hoc revision that many failed medicines could be shown to be effective. However, my major concern is about the long-term efficacy and risks of aspirin, rather than its short-term efficacy for which there is at least one conventionally positive clinical trial (ISIS-2). Should aspirin be seen as a discreet, short-term intervention after a vascular event or a long-term agent for cardiovascular prophylaxis? Baigent et al appear to have trouble in accepting the simple fact that the original data from the long-term post-infarction studies failed to show a reduction in mortality, individually or overall (3). They ignore the fact that the reason the meta-analysis shows any reduction in long- term mortality with anti-platelet agents after myocardial infarction is due to retrospective, unblinded, re-analysis of studies that had completed many years previously (3). This retrospective re-analysis identified patients who were supposed to be dead that were found to be alive and patients that were supposed to be alive that were in fact dead. We are given no reassurance about how bias during this exercise was prevented nor on why this newly acquired data was so much more in favour of anti- platelet agents than the original reports had been. Considering that we are discussing a small treatment effect (the meta-analysis suggests 6 lives saved for every thousand patients treated for a year) such biases could easily explain the purported effects of aspirin on long-term mortality. On top of this there is evidence of publication bias amongst trials of aspirin (4). Baigent et al twist the argument about the value of all-cause mortality in a curious way. All-cause mortality is the only robust way of reporting death and is a very robust endpoint for trials that assess the effects of treatment designed to improve prognosis. Assessing the mode of death can help in the interpretation of why a treatment was or was not, in the case of aspirin, effective. The increased risk of sudden death with aspirin, that was reported in the original trials but may have been 'lost' in the meta-analyses, suggests either that aspirin increases the risk of fatal arrhythmias or that it increases the proportion of vascular events that rapidly result in death. Baigent et al suggest that a subtle reduction by aspirin in the number of patients with overt non-fatal infarction would be insufficient to account for the reduction in non-fatal myocardial infarction or the increase in sudden death with aspirin. Perhaps the ability of aspirin to conceal non-fatal events is not so subtle? Perhaps aspirin reduces the chance of a non-fatal myocardial infarction being reported from 70% to 50%? This would certainly help explain the results of the US Physician's trial, in which a 39% reduction in non-fatal myocardial infarction was observed, but no reduction in mortality or non-fatal stroke and a trend to an excess of sudden death (5). Meta-analysis was arbitrarily chosen by some authorities as the best evidence in support of efficacy when creating guidelines. Recently, other professional bodies, such as the European Society of Cardiology, have questioned the validity of recommendations based on meta-analysis alone, due to the difficulty of ensuring that meta-analysis is free from major bias and the frequency with which meta-analysis of individually inconclusive trials has come up with seemingly wrong answers. At least one guideline, has decided that meta-analysis should be viewed as a method for deciding whether the mass of data on a particular topic are consistent with the outcome in the best available trials rather than perceiving meta- analysis by itself as sufficient evidence (6). Since the original reports of the best available trials on aspirin failed to show a reduction in mortality, guideline recommendations may now have to be revised, at least in terms of the quality of the supporting evidence. The authors conclude that it were better that my article had not been published at all. I am thankful that we live in a society where censorship of the medical press by a few powerful people with huge resources and a message that they are passionate to sell cannot silence criticism. We need to be sure that long-term aspirin is effective. The consequences of a wrong evaluation for patients and future therapeutic developments would be disastrous. 1. ISIS-2 Collaborative group. Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction. Lancet 1988; ii:349-360. 2. Baigent C, Collins R, Appleby P, Parish S, Sleight P, Peto R on behalf of the ISIS-2 Collaborative Group. ISIS-2: 10-year survival among patients with suspected acute myocardial infarction in randomised comparison of intravenous streptokinase, oral aspirin, both, or neither. BMJ 1998; 316: 1337-43. 3. Cleland JGF, Bulpitt CJ, Falk RH, Findlay IN, Oakley CM, Murray G, et al. Is aspirin safe for patients with heart failure? British Heart Journal 1995; 74:215-219. 4. Cleland JGF. No reduction in cardiovascular risk with NSAIDS - including aspirin? Lancet 2002; 359: 92-3. 5. Steering Committee of Physicians Health Study Group. Final report on the aspirin component of the ongoing physician's health study. New Engl J Med 1989; 321:129-135. 6. Task Force for the Diagnosis and Treatment of Chronic Heart Failure ES. Guidelines for the diagnosis and treatment of chronic heart failure. Eur Heart J 2001; 22:1527-1560.
More inaccuracies in the commentaries on the antiplatelet therapy meta-analysis 30 January 2002
Cathie LM Sudlow, Wellcome Clinician Scientist Dept of Clinical Neurosciences, Western General Hospital, University of Edinburgh, EH4 2XU, Peter Sandercock, Charles Warlow Send response to journal: Re: More inaccuracies in the commentaries on the antiplatelet therapy meta-analysis	Sir, We fully endorse the response of Baigent and others to Cleland's inaccurate and potentially dangerous For Debate commentary on the Antithrombotic Trialists' (ATT) Collaboration meta-analysis of antiplatelet therapy for the prevention of death, myocardial infarction and stroke in high risk patients.[1-3] We would like to add some further comments in response both to Cleland's article and to the editorial by Reilly and FitzGerald that accompanied the ATT meta-analysis in the same issue.[4] Both suggest that the ATT data were revised retrospectively. In fact, the methods of this and former cycles of the collaboration's overviews (e.g. trial inclusion criteria, outcomes to be assessed) were planned prospectively. Differences between the data presented in some trial publications and that used in the ATT analyses occurred where trialists provided additional, unpublished information on the numbers of patients originally randomised, on outcomes not included in the trial publications, or on subsequently available vital status or event classification for small numbers of patients. Minor differences in the data between the current and previous antiplatelet overviews do not materially affect any of the results or conclusions and generally relate to the new availability of individual patient data from a handful of trials. Reilly and FitzGerald claim that the absolute reduction in serious vascular events is smaller in patients with acute stroke than in other high risk patients. In fact, because the benefits of treatment with antiplatelet therapy (generally aspirin) in acute stroke patients accrue over such a short time, the opposite is true. About 10 events will be prevented per 1000 acute ischaemic stroke patients in the first month after onset; thereafter, a little more than one event per 1000 patients per month will be prevented with long term treatment.[3] Reilly and FitzGerald also suggest that meta-analysis is no longer needed because large enough trials are now being done. This view fails to acknowledge that: (i) meta-analysis of large trials provides opportunities to assess not just whether a treatment works but also for whom and by how much;[5] (ii) trials comparing different antiplatelet regimens have rarely been of sufficient size (i.e. tens of thousands of patients)to detect the small differences in treatment effect that might reasonably be expected. Cleland states that 'It is remarkable…how seldom trials of antiplatelet agents have shown benefit on their selected primary outcome'. It is indeed true that, as for many other interventions, many of the earlier trials of antiplatelet therapy were often too small reliably to detect moderate benefits on clinically important outcomes. It was for precisely this reason that the first meta-analyses were needed. For example, by 1980, enough patients with a previous MI (over 10,000) had been randomised in several trials of aspirin versus placebo for all the randomised evidence in aggregate to show moderate, but statistically and clinically significant, reductions in both vascular mortality and morbidity.[6] However, none of the individual trials had sufficient power convincingly to demonstrate this in isolation. Cleland suggests that the results of 'inconvenient trials' are ignored in the discussion section of the ATT paper. However, all large trials that have added substantial new data since the ATT database was closed, such as the clopidogrel in unstable angina to prevent recurrent events (CURE) trial (including around 12,500 patients), were discussed. In support of his statement, Cleland cites a reference referring to a meeting report of the results of four trials.[7] Only one of these was an antithrombotic trial, and, as far as we are aware, its results have not yet been published in full. It included less than 200 patients in the comparison between aspirin and control, and only about 50 vascular events were recorded in these two groups. Unsurprisingly, for a small study, no significant difference in the effects on vascular events between aspirin and control was found. Including this trial in the ATT meta-analysis would make no difference to the results. In discussing costs, Cleland refers to an economic appraisal of aspirin, which he co-authored.[8] The full report is not yet available from the Scottish Chief Scientist Office, but we were able to obtain the executive summary. When one considers the views that Cleland puts forward in his commentary, it is interesting to note that the first sentence of this summary reads 'Aspirin is a cheap drug that is effective for the prophylaxis of cardiovascular events.' Finally, it is worth reminding readers that, unlike the ATT meta- analysis, neither Reilly and FitzGerald's editorial nor Cleland's commentary have been endorsed by hundreds of collaborating trialists worldwide. Furthermore, Cleland's commentary was published despite a reviewer pointing out that his views are maverick, and despite the fact that the conclusions of his article are based on basic errors of fact.[1] Cathie Sudlow (corresponding author) Wellcome Clinician Scientist Peter Sandercock Professor of Medical Neurology Charles Warlow Professor of Medical Neurology Department of Clinical Neurosciences, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU Email: csudlow@skull.dcn.ed.ac.uk Cathie Sudlow, Peter Sandercock, and Charles Warlow are members of the steering committee of the Antithrombotic Trialists' collaboration. Cathie Sudlow was involved in co-ordinating the current cycle of the collaborative overview and was a member of the writing committee for the antiplatelet meta-analysis. 1. Baigent C, Collins R, Peto R. Article makes simple errors and could cause unnecessary deaths. BMJ 2002; 324: 167. 2. Cleland JGF. Preventing atherosclerotic events with aspirin. BMJ 2002; 324: 103-105. 3. Reilly M, FitzGerald GA. Gathering intelligence on antiplatelet drugs: the view from 30000 feet. BMJ 2002; 324: 59-60. 4. Antithrombotic Trialists' Collaboration. Collaborative meta- analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ 2002; 324: 71-86. 5. Fibrinolytic Therapy Trialists' Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet 1994; 324: 311- 322. 6. Anonymous. Aspirin after myocardial infarction. Lancet 1980; 1: 1172-1173. 7. Jones CG, Cleland JGF. Meeting report - LIDO, HOPE, MOXCON and WASH Studies. Eur J Heart Failure 1999; 1: 425-431. 8. McMahon AD, MacDonald TM, Davey PG, Cleland JGF. The impact of low -dose aspirin prescribing on upper gastrointestinal toxicity, renal toxicity and healthcare utilisation. Edinburgh: Chief Scientist Office, 2001: 1.
Re: Chronic aspirin therapy is effective - if you massage the data sufficiently 31 January 2002
Colin Baigent, MRC Scientist CTSU, Radcliffe Infirmary, Oxford, OX2 6HE, Rory Collins and Richard Peto Send response to journal: Re: Re: Chronic aspirin therapy is effective - if you massage the data sufficiently	Actual antiplatelet evidence EDITOR-The 1988 report of the ISIS-2 trial [1] had 97% follow-up, and the 1998 report of that trial [2] had 99% follow-up. The unfounded suggestion [3] that incompleteness of follow-up could have accounted for the substantial difference in 5-week mortality between aspirin and placebo can be dismissed [4] by comparison of these two reports. With 97% follow- up this mortality difference was 218 (805 aspirin versus 1023 placebo deaths), and with 99% follow-up it was 219 (811/8587 [9.4%] versus 1030/8600 [12.0%]). (Confusingly, Professor Cleland's most recent reply [5] slightly miscalculates both these mortality differences.) Thus, the mortality difference was virtually identical in the 1988 and 1998 reports: indeed, getting most of the missing follow-up added only 6 aspirin and 7 placebo deaths. By 1998 there remained only 87 aspirin (and 97 placebo) patients without 5-week follow-up of mortality: by no plausible assumptions could the findings among them be of material relevance. Our purpose in comparing the two ISIS-2 reports [4] was not "to revise results retrospectively" [5], nor was it "post-hoc revision" to rescue a "failed medicine" [5]; it was merely to illustrate that slight incompleteness of follow-up in such a placebo-controlled comparison cannot account for the ISIS-2 findings on aspirin. The claim that it could do so is numerically wrong, and the suggestion that authors be asked by journals to avoid publishing such gross errors is hardly "censorship of the medical press" [5]. Moreover, the ATT meta-analysis shows that the long-term antiplatelet trials did in aggregate show a significant reduction both in vascular mortality and in overall mortality [4], so why say otherwise [5]? Colin Baigent MRC Scientist colin.baigent@ctsu.ox.ac.uk Rory Collins Professor of Medicine & Epidemiology Richard Peto Professor of Medical Statistics & Epidemiology 1 ISIS-2 Collaborative Group. Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction. Lancet 1988; ii: 349-60. 2 Baigent C, Collins R, Appleby P, Parish S, Sleight P, Peto R on behalf of the ISIS-2 Collaborative Group. ISIS-2: 10-year survival among patients with suspected acute myocardial infarction in randomised comparison of intravenous streptokinase, oral aspirin, both, or neither. BMJ 1998; 316: 1337-43. 3 Cleland JGF. Preventing atherosclerotic events with aspirin. BMJ 2002; 324: 103-5. 4 Baigent C, Collins R, Peto R. Article makes simple errors and could cause unnecessary deaths. BMJ 2002; 321: 167. 5 Cleland JGF. Chronic aspirin therapy is effective - if you massage the data sufficiently. BMJ Rapid Response to [4] www.bmj.com 15 January 2002.
The Editor of the BMJ displays errors of judgement 13 February 2002
Nick Manassiev, GP registrar Little London Surgery, Little London, Walsall, West Midlands, WS1 3EP Send response to journal: Re: The Editor of the BMJ displays errors of judgement	Sir In his paper1, Prof Cleland states, “Antiplatelet agents seem to be substantially more effective in reducing the incidence of non-fatal event than in reducing death. Indeed, among large long-term trials after myocardial infarction there is no evidence that aspirin saves lives.” He continues to say, “All cause mortality and, arguably, disabling stroke are the only robust markers of benefit with an antiplatelet agent. It is not clear that antiplatelet agents reduce the risk of either.” If he is right, these statements imply that in Prof Cleland’s view, it is acceptable to have a heart attack and stroke as long as you don’t die or are not disabled by them. As he is more likely to be wrong, he may find himself under GMC investigation for denying his patients effective treatment for potentially disabling and/or fatal conditions. In either case, I feel fortunate I am not one of his patients, but I would like to know where the referees live so that I can avoid settling there too. I should like to ask Prof Cleland how he knows that low doses of aspirin (75-325 mg) is an analgesic strong enough to convert the severe visceral pain of acute myocardial infarction into a silent event. Or is this a subhypothesis to support the main hypothesis? As for diversion, in the healthcare settings I work, aspirin does not create diversion but is one of the treatments we offer our patients, together with ACE inhibitors, beta blockers and statins. Does Prof Cleland have any data to support his statement that such diversion exists or is this yet another subhypothesis in support of his main hypothesis? I suspect that the reader is not interested in speculations on what may or may not happen(1) but rather in facts of what actually happens(2). Lastly, on the usefulness of his final sentence: I am just trying to imagine the average reader following his advice ‘to read the original papers on which the conclusions are based in order to form an own opinion’(1) – all 278 of them starting from 1974. And if the Editor of the BMJ could help me here, what is the issue that needs to be debated? All in all the Editor of the BMJ should resign. The way the editorial was commissioned for the BMJ I find rather bizarre – fancy bumping into your colleague and inviting him to write about his pet hate in an objective way! Using visual cliches on the cover and verbal ones inside (education in bite sizes (3) ) betray dearth of imagination and shows journalistic impotency. The aspiration of the Editor to turn the BMJ in Cosmopolitan is not the one I share. Many of my colleagues are clever, well read and intelectually gifted and if bite size education may be good for people with learning difficulties I doubt it will of use for doctors. The BMJ has wide circulation, tens of thousands in the UK alone. And yet people either don’t read it or don’t like what is in it because they don’t quote it. The impact factor of the BMJ is 4,994 i.e. about half that of the Journal of Clinical Investigation - 9.667, and below journals like Circulation, Blood, Diabetes, Radiology, to mention only a few. ‘The forbidding journal’(3) Brain has also higher impact factor meaning that it is read and quoted more than the less forbidding BMJ. This indicates that the readers of Brain don’t seem to find it forbidding though it is so for the Editor of the BMJ. The New England Journal of Medicine, the Lancet and the Journal of the American Medical Association are well out of reach. Turning the BMJ into Cosmopolitan or giving more space to the hypothesis, and maverick one too, of prof Cleland (1) then to a commentary on the mountain of evidence (2) will not help the decline. An Editor who is striving for the utopia that all readers should ‘read every word of every page’(3) of a such broad based medical journal as the BMJ has no place in the reality of running the journal. Chillingly, what if I am wrong. What if the 15 minute attention span is what my colleagues relish? What if they prefer ‘learning in bite sizes’ the way many people like coke and crisps ‘on the go’? What if they like the BMJ to be like Cosmopolitan, half full of advertisements, as deep as a puddle and as thought provoking as a slice of toast? What if they prefer fringe opinions to the well researched ones (Wakefield and MMR, changing the UK practice of administering vit K to the newborn from i.m. to oral on the basis of a single small study from Sweden, to my knowledge never replicated elsewhere in the world, spring to mind). Then I shall have to find those who think like me, organise ourselfs, stop our subscription to BMA, part of which goes to pay the salary of the Editor of BMJ, and start up our own medical organisation and our own journal to represent it. 1. Cleland JFG. Preventing atherosclerotic events with aspirin. BMJ 2002;324:103-5 2. Antithrombotic Trialists’ Collaboration. Colaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ 2002;324:71- 86 3. Smith R. The BMJ: moving on. BMJ 2002;324:5-6 Yours faithfully: Dr Nick Manassiev No competing interests.
Aspirin for Coronary Disease - More spin than substance. 26 April 2002
John GF Cleland, Professor of Cardiology University of Hull, HU6 7RX Send response to journal: Re: Aspirin for Coronary Disease - More spin than substance.	I read with interest the comments of Sudlow and colleagues (1). Their lack of insight into the flaws of meta-analysis is suprising! To say that the antiplatelet meta-analysis was prospective when it used an end-point which few individual trials used and when it allowed the investigators to revise data after the blind on the studies had been broken is an interesting 'spin'. The retrospective re-evaluation of deaths in the post-infarction data-set is considerable and biased in favour of aspirin. Between the 1988 and 1994 meta-analyses, the net increase in deaths on anti-platelet agents was 14 but on placebo was 33, which is substantial considering the small difference under consideration (2). A reporting convention that magnified the benefit with anti-platelet agents when randomisation was not 1:1 (2) and evidence of publication bias (3) must also be taken into account when interpreting the meta-analyses that the authors admit showed only small differences. Sudlow correctly points out that the long-term trials were too small to show benefit. The meta-analysis provides the evidence of the size of trial that would be required to prove efficacy, but the meta-analysis itself is so prone to bias and error that it should not be accepted as evidence of a long-term benefit from anti-platelet agents. Sudlow failed to mention the PEP trial (4), a trial of 16,000 patients without a recent acute vascular event, which showed a 33% (p<0.05) excess of fatal and non-fatal ischaemic heart disease events with aspirin compared to placebo. I accept that most people believe that anti-platelet agents are effective. However, they may have been convinced more by 'spin' than by evidence. Mass deception by misleading propaganda causes much misery in this world. Pignone and Mulrow (1) suggest that aspirin should be used more often for primary prevention. If vascular disease is so deadly and anti-platelet agents are so effective why have none of the long-term trials, especially of primary prevention, shown a reduction in mortality, especially cardiovascular mortality. More than 234,000 patient-years exposure in primary prevention trials has resulted in no change in mortality (3.33% to 3.10%); hardly a resounding effect (2). Mortality was a more common event than non-fatal myocardial infarction in these trials, so this cannot account for the anomaly of a reduction in non-fatal myocardial infarction (of perhaps 30%) but no effect on non-fatal stroke or mortality. Aspirin's ability to conceal minor, non-fatal events is striking. Its ability to do anything useful is dubious. Professor John GF Cleland; Castle Hill Hospital, University of Hull. 1) Sudlow C, Sandercock P, Warlow C, Pignone M, Mulrow C. Anti- platelet therapy and atherosclerosis events. BMJ 2002; 324: 917-918 2) Cleland JGF. Is Aspirin 'The Weakest Link' in Cardiovascular Prophylaxis. The Surprising Lack of Evidence Supporting the Use of Aspirin for Cardiovascular Disease. Prog Cardiovasc Dis 2002; 44: 275-292 3) Cleland JGF. No reduction in cardiovascular risk with NSAIDs - including aspirin? Lancet 2002; 359: 92-93 4) Pulmonary Embolism Prevention (PEP) Trial Collaborative Group. Prevention of pulmonary embolism and deep vein thrombosis with low dose aspirin: Pulmonary Embolism Prevention (PEP) trial. Lancet 2000; 355: 1295 -1302.