Hormone Replacement Therapy (HRT) & Heart Disease

The interim report from the massive Women’s Health Initiative (WHI) study last summer shook up the world of women’s health when it was reported that HRT not only significantly increased the risk of breast and other cancers, but that HRT also increased the risk of heart disease rather than lowering it (as had been the conventional wisdom for the past five decades).  Other subsequent large-scale studies subsequently confirmed that combined HRT (with an estrogen component and a uterus-protecting progestin component) leads to a progressive increase in the risk of cardiovascular disease, including heart attacks and strokes, with increasing durations of HRT.  In this week’s New England Journal of Medicine, the WHI issues a final report on this subject. 

A total of 16,608 postmenopausal women, ages 50 to 79 years, took part in the study, and were randomly assigned to received either combination HRT or placebo pills. After an average follow-up of 5.2 years, the study was halted prematurely due to concerns about significant differences in heart disease rates between the two groups of women.  After one year of combined HRT, the women receiving the daily hormone pills had nearly twice the risk of being diagnosed with new onset coronary artery disease as did the women taking the placebo pills.  At the 5.2-year average endpoint of the study, the increased risk of heart disease had stabilized somewhat at a 24% increase in the relative risk of heart disease among women taking HRT pills.  Women who had a baseline elevation in their “bad cholesterol” (LDL) experienced an even greater risk of heart disease as a consequence of HRT use.

While the increase in the absolute individual risk of developing heart disease while on HRT is not a huge number, the implications of these finding are, nonetheless, very significant.  First, even small increases in the individual relative risk of a specific disease are very significant when the prevalence and mortality associated with that disease are already very high in the general population.  Cardiovascular disease is the number one cause of death in the United States, and so the impact of increasing or decreasing the incidence of this disease within our society, even by small percentages, can translate into very significant losses or gains in public health, respectively.  Secondly, HRT has been “sold,” by both drug manufacturers and many well-intentioned physicians, as a medication with the advantageous side effect of reducing the risk of cardiovascular disease in postmenopausal women.  Thus, these new findings, based upon huge double-blinded randomized studies, represent a paradigm shift in our understanding of the risk-to-benefit equation for HRT, and combination HRT specifically.  The WHI continues to monitor study subjects who, because they have already undergone hysterectomy, are receiving only the estrogen component of HRT.  Other studies, however, have reported less severe (but still significant) increases in the risk of breast and other cancers, and cardiovascular disease as well, for estrogen-only HRT when compared to placebo.

My position, based upon research results presented over the past two years or so:  HRT, and combination HRT in particular, carries an unacceptable risk of life-threatening complications associated with it, and should be avoided.  For women who are experiencing especially severe symptoms as they transition into menopause, HRT should be used for the least possible duration, and at the lowest effective doses: a few weeks, or perhaps a couple of months, at most.

Some proponents of HRT have argued that HRT’s recently demonstrated failure to protect women against the onset of cardiovascular disease may derive from the fact that most of the estrogen pills in the United States are manufactured from the urine of pregnant mares.  Although without scientific evidence to support their claims, some HRT advocates have suggested that HRT with the human form of estrogen may actually improve cardiovascular health based, in part, upon the empiric observation that premenopausal women have a much lower rate of cardiovascular disease than same-aged men.  A second study in the current issue of the New England Journal of Medicine takes a look at that premise.

In this study, 226 postmenopausal women with known coronary artery disease were entered into the study.  The women, with an average age of 53.5 years, were randomly assigned to standard HRT pills, the human form of estrogen (17-beta-estradiol) alone, or 17-beta-estradiol plus a sequentially administered progestin.  All study volunteers were followed with serial angiograms of their coronary arteries to monitor the rate of progression of their heart disease.  After a median 3.3 years of follow-up, 169 women had matched pairs of angiograms, taken at the beginning and end of the study, available for evaluation.  In all three groups of women, the percentage of coronary artery narrowing was essentially the same.  This study, although relatively small, would appear to dispel the notion that human estrogen is somehow superior to horse estrogen when it comes to the issue of cardiovascular health.

The bad news about HRT keeps on rolling in.  The preliminary results from a gigantic British study, the “Million Woman Study,” is being reported in the current issue of the journal Lancet.  A total of 1,084,110 women, ages 50 to 64, were followed between 1996 and 2001.  Half of these women regularly used HRT.  Among this million-plus group of women, 9,364 developed invasive breast cancers during the course of the study, and 637 of the women died of their breast cancers.  When comparing the women who currently used HRT with those who did not, the HRT users had a 66% greater relative risk of developing breast cancer, and a 22% greater relative risk of dying from breast cancer, when compared to the women who were not taking HRT.  Women who used estrogen-only HRT experienced a 30% relative increase in the risk of developing breast cancer, while those women taking combination HRT had 100%, or two-fold, increase in the relative risk of developing breast cancer.  Furthermore, this increase in breast cancer risk was minimally affected by the type of estrogen or progestin hormones prescribed, or the dosages.  The increased relative risk of breast cancer with oral estrogen-only HRT was 32%, and remained at 24% with estrogen-only skin patches, and 65% with implantable estrogen-only delivery devices.  Not surprisingly, increasing durations of HRT usage were associated with increased levels of risk for breast cancer.  Based upon statistical calculations derived from their observations, the study’s authors predicted that each decade of HRT use resulted in an additional 5 cases of breast cancer per 1,000 women using estrogen-only HRT, and 19 additional cases of invasive breast cancer per 1,000 women using combination HRT.  Based upon the number of breast cancer cases diagnosed in the UK over the past decade, it is estimated that an “extra” 20,000 cases of breast cancer occurred secondary to HRT use. 

As I have already mentioned, even relatively small increases (or decreases) in the incidence of life-threatening diseases within a large population can translate into very significant outcome differences.  To my way of thinking, simply abstaining from HRT costs society little, if nothing, but may save thousands of lives from the life-threatening diseases now linked to chronic HRT use.

FLASH:  Men and women are different!

A number of sex-related differences in disease outcomes have been identified.   For example, there is some evidence that a woman is more likely to die from a heart attack than a man, although the incidence of heart disease tends to be lower for women during most of their lives.  Diabetes, and its precursor “Metabolic Syndrome,” are rapidly reaching epidemic proportions as our society become ever fatter, and ever more sedentary.  Due to pathological changes in the body’s arterial system caused by diabetes, cardiovascular disease tends to show up early in diabetics, and progresses much more rapidly than in non-diabetics.  The current issue of the Archives of Internal Medicine features a study from the world-famous Framingham Heart Study that looks at the impact of gender on death due to coronary heart disease and diabetes.  Specifically, the study compared the relative contributions of diabetes and preexisting heart disease on the risk of heart attack death in men and women.  A total of 5,243 study volunteers were analyzed after a minimum of 20 years of follow-up.  Adjustments for cardiac mortality secondary to smoking, age, high blood pressure, serum cholesterol levels, and body mass were made. 

The men who had diabetes, but no heart disease, at the beginning of the study experienced twice the risk of dying from cardiovascular disease as did the men without diabetes.  Among men who entered the study with known cardiovascular disease but no diabetes, there was a four-fold increase in the risk of death due to coronary artery disease when compared to men with no history of heart disease or diabetes at the beginning of the study.  When the same parameters were evaluated for women, the results were the opposite.  In women with preexisting diabetes, but not heart disease, there was a nearly four-fold increase in the risk of dying of coronary artery disease when compared to women with neither a history of diabetes nor heart disease.  In the women who entered the study with a history of heart disease, but not diabetes, the risk of dying from coronary heart disease was approximately two-fold when compared to women without a preexisting history of either disease.  Thus, this study appears to show that a man’s risk of dying from heart disease is more closely linked to a previous diagnosis of coronary artery disease alone than to a history of diabetes alone.  This seems rather intuitive.  However, the reverse appears to apply to women.  Women with preexisting and stable heart disease appeared less likely to actually die from a heart attack than did the women who were diabetic (but had no history of heart disease) when they entered this study 20 years previously.  Whether this difference in the relative contributions of diabetes and heart disease to cardiovascular disease deaths relates to the smaller caliber of the arteries that nourish women’s hearts is not clear, although this difference has often been invoked to explain the higher mortality that women appear to suffer following heart attacks.  Also, the differing impacts of the male and female sex hormones on cholesterol levels may play a role in this gender-specific difference.  Whatever the explanation, this study suggests that the effects of diabetes on mortality from heart disease hit women more severely than men, and that women should be especially careful to avoid obesity and a sedentary lifestyle which are, after genetic predisposition, the two greatest risk factors for developing diabetes.

Briefly….

Archives of Internal Medicine:  In a Finnish study of 2,011 healthy men, males who were aerobically unfit were found to have a more than three-fold increase in the risk of experiencing a stroke when compared to men who were fit.  This finding remained significant even after correcting for known risk factors for stroke, including smoking, alcohol consumption, socioeconomic status, presence of coronary artery disease, diabetes, high blood pressure, or serum cholesterol levels.  Low levels of cardiorespiratory fitness produced similar increases in the relative risk of stroke as did obesity, hypertension, alcohol consumption, smoking and elevated LDL levels.

Archives of Dermatology:  A small study of university students (15 men and 7 women) looked at the role of stress on the incidence of acne outbreaks.  The volunteers were examined during periods when they were taking their exams, and when they were not.  The study found that during high-stress exam periods, acne outbreaks were more frequent and more severe than during non-exam periods.  While this correlation between stress and acne is anecdotally obvious to many people, it has not been confirmed by scientific study until now.

Archives of Surgery:  There is growing evidence that certain complex surgical procedures are more safely performed in hospitals where these operations are done more commonly.  A large study of cancer surgery outcomes between 1995 and 1997 was performed in the United States, looking at eight complex cancer operations.  Surgery for cancer of the esophagus, pancreas and lung was associated with fewer complications (including death) when performed at high-volume hospitals.  Interestingly, removal of one or more lobes of a lung was associated with increased complications at low-volume hospitals, but removal of an entire lung was less of a problem at these hospitals (probably because removing an entire lung is actually less complicated than removing a lobe).  Surgery to remove cancers of the stomach also trended towards being safer in high-volume hospitals, although this trend did not reach statistical significance.  Operations for cancers of the kidney or colon, however, did not appear to be significantly safer whether performed in high or low-volume hospitals.  As a cancer surgeon, I find these conclusions to be in line with my own experience and perspective.

Dr. Robert A. Wascher