Jump to comment:

• Age as a risk factor for hemorrhage in aspirin users
  John I Levitt
  Published on: 19 April 2001
• Re: Age as a risk factor for hemorrhage in aspirin users
  PS Sanmuganathan
  Published on: 19 April 2001

• Published on: 19 April 2001

  Age as a risk factor for hemorrhage in aspirin users

  • John I Levitt, Physician

  April 3, 2001

  To the Editor:

  It seems that there is a potential flaw in the analysis by Sanmuganathan, et al (1), of the relative risk benefit ratio in the use of aspirin for the primary prevention of coronary artery disease. The coronary risks are calculated using tables in which age is a major determinant, but the risks of a bleeding complication are averaged over all ages. Since bleeding complicat...

  Show More

  April 3, 2001

  To the Editor:

  It seems that there is a potential flaw in the analysis by Sanmuganathan, et al (1), of the relative risk benefit ratio in the use of aspirin for the primary prevention of coronary artery disease. The coronary risks are calculated using tables in which age is a major determinant, but the risks of a bleeding complication are averaged over all ages. Since bleeding complications almost certainly increase with age, this would have the net effect of over-estimating the dangers of aspirin use in younger patients, and under-estimating them in older ones.

  For example, the study assumed an increased risk of hemorrhagic stroke among aspirin users of 0.039%/year, a value taken from a recent meta-analysis by He et al (2). But table 4 of that meta-analysis records a risk of 0.029% in those under 64 and of 0.110% in those 64 and older. This was not statistically significant, presumably due to the small number of events, but this almost 4-fold difference would certainly appear to be of clinical significance.

  Determining the effect of age on non-cerebral major bleeds in aspirin users is difficult. None of the 4 studies cited in the article provide any information on bleeding rates at various ages and the literature in general on this subject seems rather sparse. One study which at least alluded to age differences was the Stroke Prevention in Atrial Fibrillation II Study (3) which noted that major hemorrhages occurred in 0.9% of aspirin users under the age of 75 and in 1.6% of those over 75. Unfortunately, there was no control group and for reasons that are unclear, the bleeding risk in this study was 5 to 10 times as great as the 4 studies referenced by Sanmuganathan.

  In any case, it would appear that one needs a table of bleeding risk at various ages before one can safely advise aspirin use in the elderly.

  1. Sanmuganathan PS, Ghahramani P, Jackson PR, et al. Aspirin for primary prevention of coronary heart disease: safety and absolute benefit related to coronary risk derived from meta-analysis of randomised trials. Heart 2001; 85:265-271.

  2. He J, Whelton PK, Vu B, et al. Aspirin and risk of hemorrhagic stroke. A meta-analysis of randomized controlled trials. JAMA 1998; 280:1930-1935.

  3. Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: stroke prevention in atrial fibrillation II study. Lancet 1994; 343:687-691.

  John I. Levitt, M.D.

  Park Nicollet Clinic

  2001 Blaisdell Ave. S.

  Minneapolis, MN 55404

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 19 April 2001

  Re: Age as a risk factor for hemorrhage in aspirin users

  • PS Sanmuganathan
  • Other Contributors:
    • "P Ghahramani, PR Jackson, EJ Wallis, LE Ramsay"

  Dear Editor,

  Dr Levitt has raised a valuable point that the risk benefit relationship of aspirin may vary with age as well as CHD risk and it is worth re-evaluating the conclusions made in our paper [1].

  He has however misunderstood the way in which we calculated benefit from aspirin treatment. Because haemorrhagic strokes were not always reported separately they had to be included in cardiovascular e...

  Show More

  Dear Editor,

  Dr Levitt has raised a valuable point that the risk benefit relationship of aspirin may vary with age as well as CHD risk and it is worth re-evaluating the conclusions made in our paper [1].

  He has however misunderstood the way in which we calculated benefit from aspirin treatment. Because haemorrhagic strokes were not always reported separately they had to be included in cardiovascular events along with other causes of stroke and myocardial infarction (MI) when calculating the benefit from aspirin treatment. Thus any increased risk of haemorrhagic strokes in subjects treated with aspirin would be evident as a diminution of observed relative risk reduction. If there is a marked increase in aspirin induced haemorrhagic stroke with age this should be seen as a variation in relative risk reduction with age. It is possible that an increased risk of stroke could be offset by greater prevention of MI in older people as seen in the US doctors study [2] but not the Thrombosis Prevention Trial [3]. Overall there appears to be no consistent variation in relative risk reduction with age when stroke and MI are considered together [3, 4].

  The evidence that the risk of extracranial bleeding caused by non- steroidal anti-inflammatory drugs increases with age is stronger [5]. The bulk of the patients in the studies from which we estimated the risk of bleeding were in the age range 40 - 75 years. The Framingham study excluded subjects above 75 years and risk estimation tools such as our Table, based on Framingham, cannot be used to estimate risk for this age group. Thus we are unable to make recommendations about aspirin use for patients greater than 75 years of age. Within the age range 40 - 75 there may well still be a gradient of risk of aspirin induced haemorrhage with age [5] and use of a single figure may underestimate risk in older subjects but overestimate it for younger subjects. Our analysis was extremely conservative and, even allowing for an underestimate of risk of extracranial bleeding in older subjects by 100% [5], a coronary heart disease (CHD) event risk threshold for aspirin use of 1.5% per annum (15% per 10 years) will still be above the upper 95% CI for the point at which the benefit from aspirin in preventing CHD exceeds possible harm. For younger subjects the point of intersection between the lines predicting harm and benefit from aspirin treatment may occur at a lower CHD risk than that indicated in our paper. The size of the absolute benefit achieved by aspirin treatment at these low levels of CHD risk would be very small. Regardless of the magnitude of the risk of haemorrhage, at CHD risk of 15% over 10 years the NNT to prevent 1 MI is 44 over 5 years and treatment of people with lower CHD risk may not be considered worthwhile.

  Thus although the risk-benefit relationship of aspirin may well vary with age as well as CHD risk we believe that benefit will outweigh harm when the CHD risk is greater than 15% over 10 years in people up to 75 years of age. In younger people the safety of aspirin treatment may be greater but it is still unlikely to be worthwhile until the risk of CHD exceeds 15% over 10 years.

  P R Jackson

  References (1) Sanmuganathan PS, Ghahramani P, Jackson PR et al Aspirin for primary prevention of coronary heart disease: safety and absolute benefit related to coronary risk derived from meta-analysis of randomised trials. Heart 2001;85:265-71. (2) Steering Committee of the Physicians Health Study Research Group. Final report on the aspirin component of the ongoing Physicians Health Study. New Engl J Med 1989;321:129-35. (3) Determination of who may derive most benefit from aspirin in primary prevention: subgroup results from a randomised controlled trial. BMJ 2000;321:13-17. (4) Kjeldsen SE, Kolloch RE, Leonetti G et al. Influence of gender and age on preventing cardiovascular disease by antihypertensive treatment and acetylsalicylic acid. The HOT study. J Hypertens 2000;18:629-42. (5) McDonald TM, Morant SV, Robinson GC, Shield MJ, McGilchrist MM, Murray FE, McDevitt DG. Association of upper gastrointestinal toxicity of non- steroidal anti-inflammatory drugs with continued exposure: a cohort study. BMJ 1997;315:1333-7.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top