Clinical effects of anticoagulants in suspected acute myocardial infarction

BMJ 1997; 314 doi: https://doi.org/10.1136/bmj.314.7075.222a (Published 18 January 1997) Cite this as: BMJ 1997;314:222

Adding heparin seems justified

  1. Michael A James, Consultant cardiologista
  1. a Taunton and Somerset Hospital, Taunton TA1 5DA
  2. b Victoria Infirmary, Glasgow G42 9TY
  3. c Dumfries and Galloway Royal Infirmary, Dumfries DG1 4AP
  4. d BHF/MRC/ICRF Clinical Trial Service Unit, University of Oxford, Oxford OX2 6HE
  5. e Leicester Royal Infirmary, Leicester LE1 5WW
  6. f Department of Medicine and Therapeutics, Leicester University, Leicester Royal Infirmary, PO Box 65, Leicester LE2 7LX

    Editor-Rory Collins and colleagues cannot justify the conclusions of their systematic overview of the clinical effects of anticoagulant treatment in suspected acute myocardial infarction; at best the conclusions can be put forward only as a personal view.1 Heparin, even when added to aspirin, showed a small but (given the large numbers) significant benefit, and without significant harm. A reduction in mortality of five cases per 1000 is of the order of benefit gained from accelerated tissue plasminogen activator in the global utilisation of strategies to open occluded coronary arteries (GUSTO) trial,2 and, given the huge numbers of patients treated, the total number of lives saved will be clinically important. That this benefit is achieved without significant harm is a most important observation. What treatment that saves five lives per 1000 without significant harm at so little cost could be rejected?

    The study has severe limitations. The authors start with the premise that because patients now receive aspirin the results of studies of heparin without aspirin are meaningless. In the studies that they used to analyse this, however, patients were receiving aspirin for the first time. Currently, most patients admitted to hospital with myocardial infarctions are already receiving aspirin and therefore cannot benefit from a new prescription; thus this review is equally meaningless. A patient with infarction who is taking aspirin represents a failure of aspirin; clearly this is a different population. By Bayesian theory we can predict that patients in whom aspirin fails are more …

    View Full Text

    Sign in

    Log in through your institution