For healthcare professionals only


Prophylaxis of venous thromboembolism

BMJ 1997; 314 doi: (Published 26 April 1997) Cite this as: BMJ 1997;314:1281

Article did not give enough information

  1. Thomas H S Dent, Consultant in public health medicinea,
  2. Monica A Dent, Senior registrar in public health medicinea
  1. a West Surrey Health Authority, Camberley GU16 5QE
  2. b Bristol Royal Infirmary, Bristol BS2 8W
  3. c Salisbury District Hospital, Salisbury SP2 8BJ
  4. d Royal United Hospital, Bath BA1 3NG
  5. e Centre for Molecular and Vascular Biology, University of Leuven, Campus Gasthuisberg, B-3000 Leuven, Belgium

    Editor—Systematic review is the best way of summarising the results of randomised trials. It is therefore disappointing that the Fortnightly Review on the prophylaxis of venous thromboembolism is so unsystematic.1 M Verstraete cites as sources of evidence, firstly, a set of guidelines whose method of preparation is not specified and, secondly, at least two review articles that were derived by consensus rather than by systematic appraisal of the literature. This prevents readers gauging the strength of evidence supporting his recommendations. Nor does he tell us whether he tried to find other reviews, how he selected reviews for inclusion, what he omitted from the reviews cited, how he reconciled differences between them, or whether he checked important sources of evidence such as the Cochrane Library and the Agency for Health Care Policy and Research. Three published relevant systematic reviews are not cited.2 3 4

    Were trials appraised before inclusion? This concern is underlined by the unequivocal affirmation of the effectiveness of low molecular weight heparins in spinal cord injury being based on an unrandomised trial. How did Verstraete use Cook et al 's grading system for evidence? That approach requires a trial's effect size to exceed the minimal clinically important benefit to attain grade A, and Verstraete does not specify that level.

    Perhaps because of the lack of use of meta-analyses, the review provides little quantitative information on the absolute or relative effectiveness of different methods of thromboprophylaxis. Low molecular …

    View Full Text

    Log in

    Log in through your institution


    * For online subscription