1000
Asia Pacific J Clin Nutr (1997) 6(3): 226-228
Asia Pacific J Clin Nutr (1997) 6(3): 226-228
Is
there an effect of menopause on CHD mortality?
Asia Pacific comparisons
Iain K Robertson MMedSci, MBBCh,
FRCA and Madeleine J Ball BSc, MD, MRCP, FRCPath
School of Nutrition and Public Health, Deakin University,
Malvern, Melbourne, Victoria, and Centre for Health Program Evaluation,
Monash University, West Heidelberg, Melbourne, Victoria, Australia
This study examines the mortality rates of men and
women from coronary heart disease (CHD) in a number of Asia-Pacific
countries with very different incidences of this disorder. In all
countries, mortality rates from CHD were higher in men. In women,
no unequivocal change in mortality rate from CHD attributable to
the menopause could be detected.
Key words: Coronary heart disease,
menopause, mortality statistics, Australasian-epidemiology, Asian-epidemiology
Introduction
When the coronary heart disease (CHD) mortality of
men and women is compared, mortality in men is higher1.
It appears that the rise in mortality is delayed in women, and this
rise is said to start after the age of the menopause. The conventional
interpretation of this is that products of ovarian action protect
women from CHD2, and that after menopause that protection
is lost, with women suffering a rise in CHD mortality similar to that
of men. As post-menopausal women have higher levels of risk factors
than premenopausal women (hypertension, LDL cholesterol and central
obesity3), many people have accepted that the menopause
is the adverse event causing rises in mortality, rather than age.
However, there is a problem with this interpretation.
The increase in CHD mortality in both men and women appears roughly
exponential (see Figure 1). In cases of exponential increases there
are no "take-off points", only a smooth escalation of rate
of increase. The process of drawing graphs to illustrate the changes
produce a false appearance of a "take-off point". The position
of the artificial "take-off point" is highly dependent upon
the y-axis scale chosen for the graph. Graphic representation of exponentials
must be based on the logarithm of the mortality rate in order to eliminate
the visual effect of the exponential. The mortality rate from CHD
(semi-logarithmic presentation) in the United States has been described
almost 30 years ago1, but that description seems to have
had little impact and would benefit from being re-examined.
This paper presents an analysis of the relationship
between CHD mortality and age in men and women from a number of Asia-Pacific
countries where such data are available. It includes data from countries
with high mortality from CHD as well as countries with relatively
low mortality rates. Th 1000 e aim is to detect whether a change in
mortality rate exists that is consistent with the hypothesis that
the menopause increases the rate of rise of CHD mortality in women.
Figure 1. Relationship between mortality from
all coronary heart disease and age in women and men in Australia.
Methods
Mortality and population data were obtained from the
WHO publication, World Health Statistics Annual, 1987 to 19944.
The number of deaths from CHD (ICD 9 codes 410-414, 410 - acute myocardial
infarction, 411-414 other forms of acute, subacute and chronic ischaemic
heart disease) were transcribed to a computer spreadsheet, along with
the corresponding population figures. The information was recorded
by sex and 10 year age groups. The mortality rate per 100,000 population
was calculated for each of the countries for each year in which the
data were available for each age sex group. The numbers reported are
an average of all the year rates that were available in the publications
examined. Information was available for the following years in: Australia
1985 to 1992; Japan 1986 to 1993; Singapore 1986 to 1992; Hong Kong
1986 to 1993; New Zealand 1985 to 1992; South Korea 1985 to 1986 and
1988 to 1991; China 1987 to 1990. No population data were available
for South Korea for 1988 onwards, so population was extrapolated from
previous figures. The mortality figures for China represent results
from about 10% of the population.
Semi-logarithmic plots are used in the graphic representation
of the results.
Results
Figure 2 shows the mortality rates from CHD in different
countries in men and women of different ages. Data from the same source4
shows that total mortality is higher in men in all countries at all
ages. CHD mortality rates of different countries fall roughly into
2 groups: 1) New Zealand, Australia and Singapore; and 2) Hong Kong,
China, Japan, South Korea. CHD mortality rates differ by a factor
of about 20 between these countries. The populations of these countries
are known to differ greatly in their diet, incidence of obesity, smoking
rates and exercise patterns, and in the cultural significance of the
menopause.
Figure 2. Relationship between mortality from
all coronary heart disease and age in women and men in some Asia-Pacific
countries.
In women, no unequivocal increase in the slope of
the mortality rate from CHD occurs following the age of menopause.
In women from Hong Kong and Japan, who have the lowest early mortality
rates, there is an increase in the exponential curve after the age
group 35-44 years, whereas all the other populations show a similar
increase in the age group 10 years younger. This pattern would best
fit a sigmoid curve, and this argues against a significant effect
of the menopause.
Discussion
The ratio of CHD mortality in men compared to women
covers a wide range amongst the different Asia-Pacific countries across
the age range. A maximum difference between men and women is found
in most countries in the age range 35-44 years, following which the
difference gradually is reduced but not abolished. The assumption
that seems to be made is that male mortality rates represent some
fixed normative CHD state that is constant across the age range against
which female mortality can be compared. For this ratio of male to
female CHD mortality to have any meaning, it would be necessary to
postulate that there 1000 is some effect that tends to reduce the
rate of exponential rise in CHD mortality in older men and women,
but that this is precisely counteracted in women by an adverse effect
of the menopause. This seems quite contrived, and a better explanation
of the observed mortality rates is that the menopause has little adverse
effect on those rates.
A more satisfactory explanation for the observed pattern
of CHD mortality in these populations is that the CHD mortality increase
is governed by an exponential rise whose initial slope is greater
in men than in women, and that the rate of rise diminishes at higher
absolute mortality rates in both men and women. The data are much
more consistent with an adverse effect operating in post-pubertal
men explaining the differences in younger men and women, and that
the strength of this adverse effect diminishes in older men. Thus
the differences appear to be established early in life, either in
early adult life or in adolescence. This would be consistent with
a specific adverse effect of male sex hormones or a more general effect
of various behaviours of young men. It should be noted that the graph
presented by Furman1 in 1968 and reproduced recently5
implies a qualitative difference in men between the 25-45 and 45+
age groups. He bases one of the regression lines on two points, which
is clearly of limited validity, and seems also to have been limited
by the technology available to him (pencil, ruler and paper). Current
computer graphics allow accurate interpolation of polynomial regression
lines, and these produce much closer fit to the observations.
The putative beneficial effect on CHD mortality of
hormone replacement therapy in postmenopausal women must be understood
in relation to the natural epidemiology of CHD as demonstrated in
this study. No randomised controlled trial has been completed to established
this beneficial effect, although at least one has recently been commenced.
The benefits rest on the circumstantial evidence of prospective observational
and case-control studies6, which are unable to exclude
a healthy-subject effect. If, however, the benefit was established,
this would need to be regarded as a pharmaceutical intervention, rather
than a "natural" correction of the "pathological condition"
of menopause.
Lifestyle factors in later life will have an influence
at an individual or population level on the absolute risk of CHD3,7-9.
The menopause in women may be associated with many changes, and longitudinal
data, although sparse, have indicated changes in some established
risk factors. However, this must not be over-interpreted to blame
the menopause itself for outcomes which may be due to biological and
behavioural consequences of aging.
Acknowledgment. We thank Professor Elizabeth Barrett-Connor for her comments.
References
- Furman RH. Are gonadal hormones (estrogens and
androgens) of significance in the development of ischemic heart
disease? Annals of the New York Academy of Sciences 1968; 149: 822-33.
- Wren BG. The effect of oestrogen on the female
cardiovascular system. Medical Journal of Australia 1992; 157: 204-8.
- Kannel WB. Metabolic risk factors for coronary
heart disease in women: perspectives from the Framingham Study.
American Heart Journal 1987; 114: 413-9.
- World Health Statistics Annual 1987 to 1993.
- Barrett-Connor E. Sex differences in coronary heart
disease. Circulatio 85f n 1997; 95: 252-264.
- Psaty BM, Heckbert SR, Atkins D, Lemaitre R, Kocpsell
TD, Wahl PW, Siscovick DS, Wagner EH. The risk of myocardial infarction
associated with the combined use of estrogens and progestins in
postmenopausal women. Archives of Internal Medicine 1994; 154: 1333-9.
- van Beresteijn ECH, Korevaar JC, Huijbregts PCW,
et al. Perimenopausal increase in serum cholesterol: a 10-year longitudinal
study. American Journal of Epidemiology 1993; 137: 383-393.
- Kuhn FE, Rackley CE. Coronary artery disease in
women: risk factors, evaluation, treatment and prevention. Archives
of Internal Medicine 1993; 153: 2626-2636.
- Barrett-Connor E, Bush TL. Estrogen and coronary
heart disease in women. Journal of the American Medical Association
I 991; 265: 1861 - 1867.
Is there an effect of menopause
on CHD mortality? Asia Pacific comparisons
Iain K Robertson and Madeleine J Ball
Asia Pacific Journal of Clinical Nutrition (1997) Volume 6, Number
3: 226-228
Copyright © 1997 [Asia Pacific Journal of Clinical
Nutrition]. All rights reserved.
to the top
0
HEC
PRESS Publisher
of the 
