Intended for healthcare professionals


Patient's sex does not affect use of thrombolysis

BMJ 1998; 316 doi: (Published 31 January 1998) Cite this as: BMJ 1998;316:391
  1. Rosalind Raine, MRC/North Thames clinical research fellowa,
  2. Tim Crayford, Senior clinical fellowb,
  3. John Chambers, Senior lecturerc
  1. a Health Services Research Unit, London School of Hygiene and Tropical Medicine, London WC1E 7HT
  2. b Department of Public Health and Epidemiology, King's College School of Medicine and Dentistry, London SE5 9PJ
  3. c Department of Cardiology, Guy's Hospital, London SE1 9RT

    Editor—Wenger's review of the importance of coronary heart disease in American women is equally applicable to women in Britain.1 Wenger points out that the prognosis is influenced by access to clinical interventions, and she suggests that underuse of thrombolysis in women may have a cascade effect on risk stratification.

    Her assertion that patients who have had thrombolytic treatment seem more likely to undergo investigation is unreferenced. In addition, the study quoted as showing lower use of thrombolysis in women than men shows no sex difference when the 95% confidence intervals are examined.2 One of us (RR) has undertaken an (as yet unpublished) critical appraisal of the international literature on the influence of patients' sex on the use of cardiac interventions, which showed that in nine of 11 studies the patient's sex did not affect use of thrombolysis. Five of these studies were conducted in Britain, of which four showed no independent effect of sex.

    The cascade effect probably does occur, but later in the clinical management pathway. We recently performed a retrospective cohort study of 715 people admitted to five hospitals in inner London with acute chest pain. After differences in age, chest pain characteristics, comorbidity, and cardiac risk factors were controlled for, the patient's sex did not influence use of thrombolysis, but men were 70% more likely than women to undergo exercise testing (adjusted odds ratio=1.72 (95% confidence interval 1.19 to 2.50)). This may be because exercise testing is less accurate in women than men.

    Women were no more likely than men, however, to receive alternative non-invasive investigations (male to female adjusted odds ratio for isotope scanning 0.93 (0.54 to 1.61)). Similar findings were reported in a previous English study.3 When women did undergo exercise testing, sex related differences in their subsequent management did not occur. This suggests that the cascade effect occurs at the level of non-invasive testing. Referral for angiography was mainly influenced by the results of the exercise test (positive result of exercise test, adjusted odds ratio=7.59, P<0.05). In turn, findings at angiography were the most important determinants of revascularisation, and sex did not exert a significant effect (male to female adjusted odds ratio for revascularisation after angiography=1.37 (0.33 to 5.56)).

    Taken together, these findings have different implications for the promotion of equitable access to cardiac services than might be drawn from the review article.


    1. 1.
    2. 2.
    3. 3.
    4. 4.