The Epidemiology of Cardiovascular Disease in Postmenopausal Women

CARDIOVASCULAR disease (CVD) is the leading cause of death in the United States for all women (1, 2), and in 1993 was responsible for twice as many deaths (500,387) in females as all forms of cancer combined (250,529 deaths) (1). Rates of CVD mortality increase sharply with age in both men and women, with higher rates in men than women at all ages. Although the death rates from coronary heart disease for females declined 27.6% from 1984 to 1994, heart attack remains the single largest cause of death for US females, and death rates were 34.3% higher in black than in white women in 1993 (1). An examination of age-specific graphs of CVD mortality rates shows an apparently large gender difference at younger ages, narrowing after the age of menopause. The largest apparent increase in coronary mortality in women occurs around age 50, coinciding with the time of menopause (3). This has led to speculation that declining estrogen levels associated with menopause result in an increased risk of CVD (2). However, evidence from vital statistics data does not support the theory that menopause apart from chronological aging increases risk of CVD in women (4). The narrowing of the gender gap appears to be due in part to declining incidence rates in men rather than to increases in women (3). Moreover, as several investigators have pointed out, if one plots the rates on the log scale to measure the slope, or rate of increase, there is no sharp increase, or “bump,” in the CVD mortality rate for women at the age of menopause (3–5). In contrast, rates of breast cancer do show a menopausal effect, with nonparallel rates of increase for the 50–69 and 40–49 age groups (3). Autopsy findings also suggest that changes such as atherosclerosis occur steadily with age, with no steep increase at the age of menopause (3). CVD mortality rates differ markedly by country, and the sex advantage in CVD mortality rates appears to be lower or nonexistent in populations that have a generally low level of risk, as in Japan (6). This indicates that behavioral and environmental risk factors may play a larger role in explaining CVD mortality in women than differences in endogenous estrogen (7). Studies based on population statistics are limited in that menopausal status cannot be related to CVD risk on an individual level.