Intended for healthcare professionals


Coronary artery disease and women

BMJ 1994; 309 doi: (Published 03 September 1994) Cite this as: BMJ 1994;309:555
  1. G Jackson

    Coronary artery disease is now the commonest cause of death in women in many countries, including Britain and the United States.1 Its incidence increases with age, rising rapidly after the menopause2: by 65 cardiovascular disease is equally common in men and women. As the proportion of older women in the population increases so the incidence of cardiovascular disease will increase. In the United States cardiovascular disease now kills proportionately more women than men.3 Before the age of 65 coronary artery disease is half as common in women as men, affecting affecting one in nine women aged 45 to 64. But coronary artery disease also affects premenopausal women - a quarter of the deaths from myo-cardial infarction in women under 65 occur in women under 45.

    The first hurdle in assessing coronary artery disease in women is to increase awareness of its existence - both among doctors and among women themselves. Women with coronary artery disease usually present differently from men.4 They present more commonly with angina than infarction. Furthermore, their symptoms are more likely to have atypical features and are more often attributed to non-cardiac causes - especially in younger women, in whom coronary artery disease is uncommon. Exercise testing is as useful in women as men when they are aged 65 or over (when the incidence of coronary artery disease is the same), but in younger women, with a lower incidence of coronary artery disease, the rate of false positive results is likely to be higher. (This may explain the finding that women with chest pain are five times more likely to have normal coronary arteries than men.5,6)

    Assessing chest pain in women, Sullivan et al confirmed the limited value of exercise testing in predicting coronary artery disease in women compared with men: they found that it had a sensitivity of 76% and a specificity of 71% (95% and 93% for men).7 Thallium scintigraphy increases the sensitivity and specificity, but breast tissue overlying the heart may erroneously suggest an area of hypoperfusion. Coronary arteriography is as accurate in men and women and remains the standard. With a complication rate of less than one in 5000, it allows the safe optimisation of treatment, especially in younger patients.

    Worse treatment for women?

    Do women receive the same treatment as men? Findlay et al identified little sex bias in the referral of patients for revascularisation4 and Lincoff et al noticed that the effectiveness of thrombolysis did not differ between men and women, though women were at greater risk of haemorrhagic stroke.8 They noted, however, that the clinical outcome in women was worse, with more reinfarctions and deaths. The women were older and had more complicating factors such as diabetes, and these factors rather than their sex may have been the reason.

    In this week's journal Clarke and colleagues report that women do not receive the same treatment as men after acute myocardial infarction (p 563).9 Women took slightly longer to reach hospital, perhaps reflecting their lack of awareness of the possibility of infarction. In common with other investigators, the authors observed that women have more severe infarcts and are older at presentation and are therefore more likely to have severe coronary artery disease, which increases their mortality.

    Clarke and colleagues' observation that patients with the highest mortality (elderly women) do not receive thrombolysis or coronary angioplasty may reflect a local policy of not giving thrombolysis outside the coronary care unit and be determined by the unit's policy concerning age rather than by sex bias. This suspicion is reinforced by the similar rates of thrombolysis for men and women once they had been admitted to the coronary care unit. Wilkinson et al report similar findings, confirming the likelihood of more severe disease in women at presentation (significantly more patients with diabetes, hypertension, and heart failure) (p 566).10 All patients in their study were admitted to a coronary care unit where the rates of thrombolysis in men and women were not significantly different. In both studies β blockers were prescribed significantly less commonly for women, but medical exclusions (for example, heart failure or diabetes) rather than sex bias may explain this. Data on the use of angiotensin converting enzyme inhibitors would be interesting: was this more common in women (reflecting larger infarct size) and did this offset the less frequent use of β blockade?

    Both studies eliminated age bias as the only explanation for the differences in management and outcome, whereas Hannaford et al suggest that increasing age, not sex, explains the discrepancies in the use of thrombolysis (p 573).11 A recent prospective study evaluating sex bias in the referral of patients for cardiac catheterisation did not identify sex as an independent predictor of further investigation and found no evidence that doctors were underestimating the risk of coronary artery disease in women - the difference in referral was related to a lower rate of exercise tests giving a positive result in women rather than to sex alone.12 Such prospective studies are helpful but reflect the care of an “aware” group of cardiologists, who may be trying hard to avoid bias.

    Different figures emerge from the survival and ventricular enlargement study of captopril after infarction: even though the incidence of angina was the same in both sexes, twice as many men than women had angiography and coronary artery bypass grafting.13 Other studies have shown that men with abnormal thallium scans were 10 times more likely than women to have angiography and for a given coronary lesion were four times more likely to undergo coronary artery bypass grafting.14

    Different risk factors

    Sexual bias may explain these discrepancies, but they could also reflect the influence of other factors: women tend to be older and more commonly have hypertension, diabetes, and cardiac failure. Some of the perceived sexual bias may therefore reflect good clinical judgment and individualised care. Do risk factors differ between women and men? Hypertension seems a stronger risk factor for coronary artery disease in women than men, and diabetes neutralises the survival advantages women have over men up to the age of 65. Hyperlipidaemia must be looked for when ischaemia is suspected, especially in older women, who lose the protective higher concentrations of high density lipoprotein cholesterol postmenopausally. It is especially important in younger women not to manage on total cholesterol concentration alone as reducing high density lipoprotein by blindly treating total cholesterol could be counterproductive.

    Smoking is increasing in (especially younger) women and may reflect social and work stresses combined with clever and carefully targeted advertising.15 Women also use cigarette smoking as an aid to control their weight: being thin is fashionable. The effect is dose dependent: premenopausal smokers have three times the rate of infarction of non-smokers; women smoking more than 40 cigarettes a day increase their risk 20-fold. The combination of smoking and diabetes is particularly hazardous. Newer oral contraceptives containing relatively low doses of oestrogen do not apparently increase the risk of coronary artery disease, but oral contraceptive users should be firmly advised not to smoke.

    Hormones could halve risk

    Epidemiological studies have suggested that oral postmenopausal oestrogen treatment halves the risk of coronary artery disease.16 Worries that hormone replacement therapy may increase the risk of breast cancer seem unfounded, though its use in women with a strong family history of breast cancer is probably unwise. The main problem with hormone replacement therapy is the lack of prospective scientific studies - no randomised intervention trials have been performed - and the absence of any effect on total mortality in the observational studies available. Thus healthy women of higher social class with lower cardiovascular risk may select themselves for hormone replacement therapy, which may distort any benefit perceived to be due to use of oestrogen.17 Randomised controlled trials are therefore needed to settle the matter (as overall mortality is unaffected). The randomised trial of the National Institutes of Health will not be reported for 14 years but we need a practical management policy now as the potential for halving the incidence of coronary artery disease is difficult to ignore.

    Women with coronary artery disease have a higher mortality after infarction and a higher operative risk at the time of coronary artery bypass grafting or angioplasty. The differences probably reflect the more advanced nature of coronary artery disease at presentation combined with naturally smaller coronary arteries. The clinical and angiographic success of angioplasty is now as good for women as men, and though women have twice the mortality from coronary artery bypass grafting men the risk remains small. After coronary artery bypass grafting men and women have the same five year and 10 year survival rates. Women presenting with angina have a better prognosis than men with angina,18 which is probably due to the increased incidence of normal coronary arteries in women.

    Once diagnosed as having coronary artery disease, women are less likely to be referred or attend for rehabilitation, and they experience more depression, anxiety, and guilt. Rehabilitation courses specifically tailored to the different needs of women are required.

    Women are different - but not that different. Although women with coronary artery disease may be more difficult to diagnose and manage than men it is a challenge that we and they must rise to. An equal opportunity killer needs equal opportunity management.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    View Abstract